Near-unity electrochemical conversion of nitrate to ammonia on crystalline nickel porphyrin-based covalent organic frameworks

Abstract

Electrochemical nitrate reduction, which has attracted rapidly increasing attention over recent years, can potentially enable the indirect fixation of atmospheric N₂ as well as the efficient removal of nitrate from industrial wastewater. It is, however, limited by the lack of efficient and low-cost electrocatalysts available so far. To address this challenge, we here demonstrate a two-dimensional nickel porphyrin-based covalent organic framework (COF) as a potential candidate for the first time. The product has a highly ordered molecular structure with abundant square-shaped nanopores. In neutral solution, the reduction of nitrate ions at different concentrations from ammonia is realized with a great selectivity of ∼90% under a mild overpotential, a remarkable production rate of up to 2.5 mg h⁻¹ cm⁻², a turnover frequency of up to 3.5 s⁻¹, and an intrinsic stability that is best delivered under pulse electrolysis. This cathodic reaction can also be coupled with the oxygen evolution reaction to enable full-cell electrolysis at high efficiency. Theoretical computations indicate that nickel centers can stably adsorb nitrate, and facilitate its subsequent reduction by lowering the energy barrier of the rate-determining step.

---

Broader context

The selective electrochemical nitrate reduction reaction (NO₃RR) to ammonia may play two essential roles within the artificial nitrogen cycle. When powered using renewable electricity, it can enable the indirect fixation of atmospheric N₂ without emissions; it can also remove nitrates from industrial wastewater, and provide a much needed remedy to the global environmental and water security challenges that we face today. Unfortunately, the NO₃RR involves multistep electron transfer and can yield a spectrum of reduction products, giving rise to a limited reaction selectivity. Copper represents the most studied catalyst, but generally requires a large overpotential in order to attain satisfactory NH₃ selectivity and production rates. Here, we demonstrate transition metal macrocycles as a potential candidate system. As a proof of concept, we incorporate nickel porphyrin into two-dimensional covalent organic frameworks in order to maximize the catalytic activity and efficiency. In a neutral solution, our catalyst exhibits great selectivity that is close to unity, an excellent turnover frequency, and intrinsic stability. This work offers significant insights to extend promising electrocatalysts for nitrate reduction.

Introduction

Nitrogen is an indispensable ingredient for all living organisms. It exists in many different forms in the atmosphere, water, soils, and biota, which are naturally interconverted via the nitrogen cycle.^1,2 The development of the Haber–Bosch process more than a century ago enabled the artificial conversion of atmospheric N₂ to NH₃, which since then has made ammonia fertilizers widely available.³ Unfortunately, this industrial process is heavily energy intensive and has a massive carbon footprint. Direct NH₃ synthesis from electrochemical N₂ reduction under mild conditions is highly appealing, but is a formidable challenge and suffers from poor reaction rates (of the order of ∼1 nmol s⁻¹) owing to the chemical inertness of N₂.^4–7 By comparison, NH₃ synthesis from the nitrate/nitrite reduction reaction (NO₃RR or NO₂RR, respectively) is more kinetically favorable, and has recently attracted rapidly growing attention.^8–10 The nitrate or nitrite feed sources can be derived from N₂via plasma treatments, thereby enabling the indirect fixation of atmospheric N₂;^11,12 in addition, the reaction can be powered using renewable electricity from solar and wind sources, and is potentially emission-free. Moreover, the NO₃RR or NO₂RR offers a viable solution for removing nitrates and nitrites in eutrophic water bodies produced via over-fertilization or other anthropogenic activities and may mitigate the global water security challenge that we face today.^13–15 However, the NO₃RR or NO₂RR to NH₃ involves multistep electron transfer that generally compromises the reaction selectivity.¹⁶ There have been increasing calls for the development of efficient NO₃RR or NO₂RR electrocatalysts at low costs.

Among different candidates, Cu-based catalysts have received the most attention and have been investigated in a number of recent studies.^17–19 They do, however, generally demand a large overpotential in order to attain satisfactory NH₃ selectivity and production rates.^9,20 On the other hand, transition metal macrocycles with discrete metal centers (in particular, Fe, Co, and Ni) have found applications for a range of important electrochemical processes (e.g., the oxygen reduction reaction and the CO₂ reduction reaction),^21–24 although they have never been considered for the NO₃RR or NO₂RR to the best of our knowledge. Their diverse and well-defined molecular structures offer a unique opportunity to be used as a model catalyst and to understand structure–activity correlations. It has also been suggested that the isolated metal centers may suppress undesirable N–N coupling and inhibit the side reaction pathway.²⁵ Unfortunately, transition metal macrocycles are often subject to strong intermolecular π–π stacking, and form microsized crystals with low surface areas, which adversely impair their attainable activities.²⁶ Incorporating transition metal macrocycles within two-dimensional (2D) conjugated covalent organic frameworks (COFs) provides a promising strategy to address the above issue. The porous structure of COFs could significantly enhance the surface accessibility of active sites, and their molecular conjugation may also facilitate rapid intralayer and interlayer electron transfer to the active sites during electrochemical reactions.

To this end, we here report a 2D nickel porphyrin-based COF (NiPr-TPA-COF) for the efficient electrochemical NO₃RR to NH₃ in a neutral medium. NiPr-TPA-COF features a highly crystalline framework and has abundant mesoporosity. It exhibits great electrocatalytic activity and an NH₃ selectivity of ∼90%. A remarkable NH₃ production rate of up to 2.5 mg h⁻¹ cm⁻² and a turnover frequency (TOF) of 3.5 s⁻¹ are measured. Density functional theory (DFT) calculations suggest that the NiPr moieties can stably adsorb NO₃⁻, and enable efficient electron transfer for its further reduction with a decreased energy barrier.

Results and discussion

NiPr-TPA-COF was synthesized from the Schiff base condensation reaction between 5,10,15,20-tetrakis(4-aminophenyl)porphinato nickel and terephthalaldehyde (TPA) in the presence of 6 M acetic acid as the catalyst (see ESI† for details). The slow and reversible crosslinking of the amine and aldehyde precursors catalyzed by acetic acid triggers a self-correction mechanism that is proven essential for high product crystallinity.^27–29 In addition, we also prepared metal-free H₂Pr-TPA-COF as a control by replacing the NiPr precursor with H₂Pr under otherwise identical conditions.

Fig. 1a schematically depicts the structural topology of NiPr-TPA-COF, which consists of the A–A stacking of 2D molecular layers with a square shaped porosity, which is confirmed via its powder X-ray diffraction (XRD) pattern. The sharp diffraction peaks indicate long-range structural ordering and high product crystallinity (Fig. 1b). The most intense peak located at 3.45° is assignable to (100) diffraction and corresponds to an average pore width of 2.5 nm, in good agreement with the proposed structure. The molecular structure of NiPr-TPA-COF was further interrogated using a multitude of spectroscopic analyses. Compared with the two starting monomers, both the C=O stretching vibration (1697 cm⁻¹) and the N–H stretching vibration (3300–3500 cm⁻¹) disappear in the Fourier-transform infrared (FT-IR) spectrum of NiPr-TPA-COF. A new peak at 1617 cm⁻¹ is instead observed in line with previous results,^30,31 attesting to the complete condensation and the formation of the imine linkage (Fig. 1c). Raman analysis corroborates the formation of the imine linkage via its characteristic signal at 1587 cm⁻¹ (Fig. S1, ESI†).³² Moreover, the imine functionalities are also reflected from the ¹³C solid-state nuclear magnetic resonance (NMR) spectrum of NiPr-TPA-COF, which shows a resonance peak at δ = 158.6 ppm along with other signals from the porphyrin (Fig. S2, ESI†).³³ The X-ray photoelectron spectroscopy (XPS) of NiPr-TPA-COF supports the presence of Ni centers in the divalent state (Fig. S3, ESI†).³⁴ Spectroscopic characterization of H₂Pr-TPA-COF reveals a similar molecular structure except for the transition metal centers, as summarized in Fig. S4 (ESI†).

Fig. 1 Structural characterization of NiPr-TPA-COF. (a) Schematic topology of NiPr-TPA-COF and its constituent units. (b) XRD pattern of NiPr-TPA-COF. (c) FT-IR spectra of NiPr-TPA-COF, NiPr, and TPA. (d) TEM image, (e) HR-TEM image, and (f) EDS elemental mapping of NiPr-TPA-COF.

NiPr-TPA-COF was subsequently subjected to microscopic imaging. Under low-resolution transmission electron microscopy (TEM), the product was unveiled to consist of square nanoflakes that are further stacked together (Fig. 1d). The individual nanoflakes have a lateral size of ∼100 nm, and exhibit smooth edges and sharp corners. The square shape reflects the inherent four-fold symmetry of the molecular structure. Under high-resolution TEM, evident square fringes are observed with a d-spacing of 2.5 nm. These nanoflakes are determined to be terminated with (001) planes. Periodic mesoporous channels run perpendicularly through nanoflakes and correspond to the light contrast regions in Fig. 1e. Energy dispersive spectroscopy (EDS) elemental mapping in Fig. 1f shows the uniform spatial distribution of C, N, and Ni species and hence the homogeneity of the composite. The porous nature of NiPr-TPA-COF is additionally confirmed via BET analysis, which measures a large surface area of 899 m² g⁻¹ and a pore volume of 0.49 cm³ g⁻¹ (Fig. S5, ESI†). Altogether, the above results unambiguously support that the highly crystalline 2D COF is formed via imide linkages and with the successful incorporation of NiPr moieties. Its ordered and porous structure facilitates fast mass transport and charge transport, which are essential for efficient electrocatalysis.

We next evaluated the electrochemical performance of NiPr-TPA-COF for the NO₃RR using a two-compartment H-cell (see ESI† for details). The catalyst powder was dropcaste onto a 1 × 1 cm carbon fiber paper electrode at a mass loading of 1 mg cm⁻². Two control samples, i.e., metal-free H₂Pr-TPA-COF and the NiPr monomer were brought into the comparison side by side. We initially attempted to carry out the NO₃RR in 1 M KOH, like done in many other studies.^17,35,36 NiPr-TPA-COF delivered a large current density and an NH₃ selectivity that is superior to most competitors under similar conditions (Fig. S6, ESI†).^17,36,37 However, we surprisingly discovered that the current collector itself was electrochemically active in the alkaline medium (which will be the subject of a future study) and strongly interfered with our measurements (Fig. S7, ESI†). Therefore, all the experiments were carried out in a neutral solution where the activity of the bank current collector was minimized (Fig. S8, ESI†).

Fig. S9 (ESI†) depicts the polarization curves of NiPr-TPA-COF in Ar-saturated 0.5 M K₂SO₄ with and without 0.1 M KNO₃. In the absence of KNO₃, only the hydrogen evolution reaction (HER) takes place, which does not exhibit any appreciable current density until less than −1.5 V (versus the saturated calomel electrode or SCE, the same hereinafter). The addition of KNO₃ positively displaces the polarization curve. The current density starts to take off at −0.8 V and increases to 46 mA cm⁻² at −1.5 V, which is the first indication of the great NO₃RR activity of NiPr-TPA-COF. By contrast, both H₂Pr-TPA-COF and NiPr have negligible cathodic responses in the presence of KNO₃ within the same potential region.

In order to identify and quantify the reduction products, chronoamperometric (i–t) analysis was conducted at different working potentials. The amount of NH₃ production was measured via the indophenol blue method using ultraviolet-visible (UV-Vis) spectrophotometry (Fig. S10, ESI†).³⁸ Its corresponding faradaic efficiency was derived for different samples and plotted with respect to the working potential, as shown in Fig. 2a. It is worth noting that both NiPr-TPA-COF and NiPr have a remarkable NH₃ selectivity between −1.3 V and −1.5 V. In particular, the former exhibits a great value of >80% over the potential region examined, and delivers a maximum selectivity of ∼90% at −1.38 V, which is superior to most single atom catalysts (SACs)^25,39 and some Cu-based electrocatalysts previously reported under similar conditions.⁴⁰ The NH₃ faradaic efficiency of NiPr starts to decline at less than −1.4 V owing to competition from the HER. These two samples share the same NiPr moieties that are believed to be catalytically active sites. Metal-free H₂Pr-TPA-COF has much inferior selectivity of <30%, which clearly manifests the critical role of Ni²⁺ centers during the NO₃RR.

Fig. 2 Electrochemical NO₃RR performance of NiPr-TPA-COF. (a) NH₃ faradaic efficiency and (b) NH₃ partial current density and production rate of NiPr-TPA-COF in comparison with those of H₂Pr-TPA-COF and NiPr; the top illustration in (b) shows the diluted catholyte after the NO₃RR at different potentials and with the addition of indophenol indicator. (c) TOF values of NiPr-TPA-COF and NiPr. (d) ¹H NMR spectra of NH₃ using K¹⁴NO₃ or K¹⁵NO₃ for the NO₃RR. (e) NH₃ faradaic efficiency and (f) NH₃ partial current density and production rate of NiPr-TPA-COF in 0.5 M K₂SO₄ with the addition of different KNO₃ concentrations. (g) Cycling test of NiPr-TPA-COF at −1.38 V showing the average NH₃ faradaic efficiency and production rate within each cycle.

The potential-dependent NH₃ partial current density and production rate were then calculated. They are observed to rise continuously with the increasing overpotential (Fig. 2b). At −1.46 V, the NH₃ partial current density of NiPr-TPA-COF reaches 31.2 mA cm⁻², and its NH₃ production rate reaches 2.5 mg cm⁻² h⁻¹. These values are ∼3 times larger than those of NiPr and >60 times larger than those of H₂Pr-TPA-COF. They are also orders of magnitude larger than the reported rates for electrochemical N₂ reduction to NH₃. Moreover, we estimated the TOF by normalizing the NH₃ partial current density over the amount of surface accessible Ni sites (Fig. S11, ESI†). It was derived as 12 754 h⁻¹ (or 3.5 s⁻¹) on NiPr-TPA-COF and 1717 h⁻¹ (or 0.5 s⁻¹) on NiPr (Fig. 2c), reflecting the significantly higher site-specific activity when NiPr moieties are incorporated within the molecular frameworks. Among the very few studies that reported TOF values for the NO₃RR to NH₃, Ni nanoparticles were previously disclosed to have a TOF value of ∼6 h⁻¹ (or 0.0017 s⁻¹) at −0.5 V vs. RHE.⁴¹ The orders-of-magnitude higher TOF value observed for our NiPr-TPA-COF highlights its extraordinary activity. To confirm the chemical origin of NH₃, ¹⁵N-labeled isotope experiments were carried out on NiPr-TPA-COF (Fig. 2d). The ¹H NMR spectrum of normal ¹⁴NH₃ features a triplet with a coupling constant of 52 Hz. When K¹⁵NO₃ is used instead of K¹⁴NO₃, the ¹H NMR spectrum of its reduction product exhibits a distinct doublet with a coupling constant of 72 Hz that is characteristic of ¹⁵NH₃.⁴² These results support that KNO₃ rather than other possible nitrogen contaminants is reduced during the NO₃RR to yield NH₃ at high rates.

Our catalyst can operate under a range of KNO₃ concentrations. Running the reaction in 0.05 M KNO₃ slightly compromises the NH₃ selectivity, and reduces the NH₃ production rate of NiPr-TPA-COF as expected. But even so, it still exhibits an NH₃ selectivity of ∼80% within −1.3 and −1.4 V and a production rate of ∼1.3 mg cm⁻² h⁻¹ at −1.46 V (Fig. 2e and f). This capability of working with low-concentration KNO₃ indicates the great potential of our catalyst for converting wastewater nitrate (typically 0.01–0.1 M) into valuable NH₃. On the other hand, increasing the KNO₃ concentration to 0.3 M in the electrolyte boosts both the selectivity (close to unity) and activity (6.6 mg cm⁻² h⁻¹ at −1.42 V) of NiPr-TPA-COF. There is no sign of catalyst poisoning at high substrate concentrations, which is a concern when using Cu-based materials.^13,43,44 Our NiPr-TPA-COF can also enable an efficient NO₂RR to NH₃ in 0.5 M K₂SO₄ with 0.1 M KNO₂ (Fig. S13, ESI†). Its high NH₃ selectivity (∼100% at −1.33 V) and great activity (5.3 mg cm⁻² h⁻¹ or 67 mA cm⁻² at −1.44 V) also render it one of the best NO₂RR electrocatalysts in neutral solution. Moreover, the effect of the catalyst loading was explored and is summarized in Fig. S12 (ESI†). Replacing the Ni centers with Co centers in CoPr-TPA-COF results in a lower NH₃ selectivity and production rate (Fig. S14 and S15, ESI†). In addition to a high activity and selectivity, operation stability is an important metric for the evaluation of catalysts. We find that when NiPr-TPA-COF is biased at a relatively negative working potential (e.g., −1.38 V), its cathodic current density is subjected to a rapid increase in the first 100–200 s and then gradually declines afterward (Fig. S16, ESI†). Such a dynamic evolution of the current density reflects an activation-to-deactivation process that appears common to NO₃RR electrolysis.⁴³ The accumulation of side reduction products near the catalyst is believed to block its active sites for a selective NO₃RR, which might explain the gradual but considerable deactivation.⁴⁵ Nevertheless, this deactivation causes no damage to the catalyst since its original activity can be fully recovered when the electrolyte is refreshed. We attempted to eliminate the deactivation process using pulse electrolysis, which has been demonstrated as an efficient tool for surface poison removal or catalyst regeneration.^46–51 The running program was set to consist of alternating reduction steps at −1.38 V for 1000 s and oxidation steps at +0.4 V for 1 s. During the cathodic step, the NO₃RR takes place at the working electrode; during the anodic step, the side reduction products accumulated on the catalyst surface are quickly oxidized. In this way, our catalyst can be periodically regenerated without sacrificing current density or NH₃ selectivity. Its current density goes back to the initial level after the anodic step. As a result, long-term electrolysis can be carried out without interruption for ∼20 000 s (Fig. S17, ESI†). During our measurements, the cathodic current density at the reduction steps stays at ∼20 mA cm⁻², and the average NH₃ faradaic efficiency is ∼73% (Fig. 2g). XRD, FT-IR and TEM characterization of the catalyst after long-term electrolysis revealed no discernible structural change on different scales, further attesting to the great stability of the catalyst (Fig. S18, ESI†).

In order to gain further insight into the great performance of NiPr-TPA-COF, DFT calculations were carried out to reveal the distinct electronic structures in comparison with metal-free H₂Pr-TPA-COF (Fig. S19, ESI†). The projected partial density of states (PDOSs) analysis shows that its Ni-3d orbitals are located close to the Fermi level (E_F), providing evidence for the high catalytic activity (Fig. 3a). PDOS analysis also exhibits a peak at E_v −1.30 eV (where E_v denotes 0 eV) that overlaps considerably with both N-2p and C-2p orbitals, which may enable fast site-to-site electron transfer from the surrounding C and N sites to the Ni center. When comparing the N-2p orbitals in NiPr-TPA-COF and H₂Pr-TPA-COF, it comes to our attention that the Ni center activates nearby N sites (N–CN₃), as reflected by their evidently increased electron density near E_F (Fig. 3b). The N sites far away from the metal center and with a coordination number (CN) of two (N–CN₂) are not affected. Similarly, the Ni center also influences nearby C sites in the porphyrin, and increases their electron density near E_F (Fig. 3c). Even at the benzene linker, carbon atoms adjacent to the imine group show upshifted C-2p orbitals relative to those connecting to H. These results reveal a gradual downshifting trend of the C-2p orbitals in NiPr-TPA-COF that forms an efficient electron path for electrolysis. Moreover, the PDOS data for NO₃ adsorption are analyzed in order to disclose the interaction between the catalyst and the reactant (Fig. 3d). The 2p orbitals of free NO₃⁻ ions show a sharp peak at E_v −1.70 eV, which is close to the Ni-3d peaks. This can enable efficient electron transfer from NiPr-TPA-COF to NO₃⁻ with a small energy barrier. Once NO₃⁻ ions are adsorbed, we observe that the strong interaction noticeably downshifts the 2p orbitals of NO₃*, supporting the electron transfer from the Ni site. Accordingly, the Ni-3d orbitals are also modulated. The good orbital overlap between them ensures the stable adsorption of NO₃* for further reduction. This also supports that the Ni center is the catalytic site for the NO₃RR.

Fig. 3 DFT simulations of the catalyst electronic structures and NO₃RR pathway. (a) PDOSs for NiPr-TPA-COF. (b) Site-dependent PDOSs of N-2p orbitals in NiPr-TPA-COF and H₂Pr-TPA-COF. (c) Site-dependent PDOSs for C-2p orbitals in NiPr-TPA-COF. (d) PDOSs for free NO₃ and NO₃ adsorbed on NiPr-TPA-COF. (e) Adsorption energies of NO₃ and H₂O on NiPr-TPA-COF and H₂Pr-TPA-COF. (f) Energetic reaction pathways of the NO₃RR to NH₃ on NiPr-TPA-COF and H₂Pr-TPA-COF.

Furthermore, we calculated the adsorption energies of NO₃ and H₂O on the two catalysts (Fig. 3e). It is worth noting that the adsorption of NO₃ on NiPr-TPA-COF (−0.23 eV) is spontaneous owing to its strong binding to the Ni center as explained above, while the adsorption of NO₃ on metal-free H₂Pr-TPA-COF (0.15 eV) is endothermic, which may hinder its subsequent reduction. This confirms the enhanced binding strength of N-containing intermediates on Ni sites, which are significant for guaranteeing the stable fixation of intermediates for ammonia formation. The adsorption of H₂O is thermodynamically favorable on both NiPr-TPA-COF (−0.11 eV) and H₂Pr-TPA-COF (−0.27 eV), but its stronger binding on the latter may enhance the competing HER process and lower the NO₃RR efficiency. The mechanism of nitrate reduction is complicated, and involves 9 electron transfers in the process with valence state changes.^52,53Fig. 3f depicts the simulated full reduction pathway from nitrate to ammonia.⁵⁴ Nitrate is first adsorbed on the catalyst surface, and dissociates to NO₂* and then NO*. Subsequently, NO* undergoes three proton-coupled electron transfers and is converted to NH₂OH* before further reduction to NH₃ and final desorption from the catalyst surface. We find that the overall reaction trend is downhill on NiPr-TPA-COF with a total energy of −4.1 eV for NO₃RR.

Compared with H₂Pr-TPA-COF, it has a stronger affinity towards all the intermediates except for NO₂*. The reduction step from NH₂OH* to NH₂* is identified to be the rate-determining step for both catalysts, and its energy barrier is lowered from 1.01 eV on H₂Pr-TPA-COF to 0.28 eV on NiPr-TPA-COF. In addition to its favorable electronic structure, we believe that the highly ordered and porous structure of NiPr-TPA-COF may facilitate mass transport, i.e., the diffusion in of reactants (NO₃⁻ and H₂O) and the diffusion out of products (NH₃ and OH⁻). Its conjugated molecular framework may also form conductive pathways for efficient electron transfer to the active centers. All these factors combined lead to the great NO₃RR activity and selectivity observed experimentally on NiPr-TPA-COF.

Having established NiPr-TPA-COF as a promising NO₃RR electrocatalyst, we then explored its potential in the coupled electrolysis of the NO₃RR and the oxygen evolution reaction (OER). A two-electrode configuration was assembled by pairing the NiPr-TPA-COF cathode with a piece of IrO₂-coated Ti mesh (IrO₂@Ti) as the anode (Fig. 4a). Fig. 4b presents the polarization curve without iR compensation. It starts to take off at 2.4 V and delivers 34 mA cm⁻² at 3.5 V. A lithium-ion battery was then used to power the two-electrode cell. Under the voltage of ∼3.4 V (Fig. 4c), the full reaction can be driven at a current density of 23 mA cm⁻² (Fig. 4d). The average NH₃ faradaic efficiency and energy efficiency were measured to be 91% and 14.5%, respectively.

Fig. 4 Full-cell NO₃RR-OER electrolysis. (a) Photograph showing the setup for battery-powered NO₃RR–OER electrolysis. (b) Full-cell polarization curve. (c) Voltage change and (d) current evolution for full-cell electrolysis.

Achieving a high current density for electrochemical catalysis has been a significant challenge for the practical application of electrocatalysts. Although there have been great efforts for the NO₃RR, most electrocatalysts still cannot break through a current density of 200 mA cm⁻², which strongly hinders their further practical applications for NH₃. For the NO₃RR, previous research has proposed that the adsorption-energy scaling relations are the key factor that causes the tradeoff between the partial current density and the faradaic efficiency.⁵⁵ In the more negative potential, the competitive hydrogen evolution reaction (HER) process causes shifting of the faradaic efficiency, which limits the current density of the NO₃RR and the production rate of NH₃.⁵⁶ For transition metal alloys, the surface roughness of the catalyst materials also limits the current density due to a lower mass transfer.⁵⁷ These studies indicate that further optimization studies are still needed in the future to achieve an even higher current density.

The experiment here demonstrates the practical viability of the battery-driven NO₃RR using our catalyst, even though both the current density and the energy efficiency are not fully optimized. Future optimization can be approached by carrying out the coupled reaction in an alkaline solution since both the NO₃RR and OER are intrinsically more active in alkaline solution than in neutral solution. We may also replace IrO_x/C with more active NiFeO_x catalysts for the OER in the alkaline solution. To further improve the energy efficiency, the sluggish OER half-reaction can be substituted with the electrochemical oxidation of alcohols such as benzyl alcohol and glycerol. These alternative reactions are thermodynamically more favorable and kinetically faster and can yield oxidation products with higher economic values and in more condensed forms. Therefore, in the future, this work will inspire more related research to achieve practical applications of the NO₃RR.

Conclusions

In summary, here we reported the first demonstration of COF-based NO₃RR electrocatalysis to NH₃. NiPr-TPA-COF was prepared via the reversible Schiff base condensation reaction between amine-terminated NiPr and TPA. The product featured great structural crystallinity and abundant mesoporosity, as characterized using a multitude of spectroscopic and microscopic techniques. When assessed as the electrocatalyst for the NO₃RR in neutral solution, NiPr-TPA-COF exhibited a near-unity NH₃ selectivity, a great NH₃ production rate of up to 2.5 mg h⁻¹ cm⁻², and a TOF of 3.5 s⁻¹ at −1.46 V. These values were superior to single atom catalysts and Cu-based materials under similar conditions. Moreover, NiPr-TPA-COF showed great stability that could be periodically recovered and was further improved under pulse electrolysis. DFT simulations revealed that NiPr-TPA-COF had unique electronic structures that enhanced its interaction with NO₃, and facilitated electron transfer during electrocatalysis. The presence of the Ni center lowered the activation energy barrier of the rate-determining step considerably. Finally, we demonstrated the coupled electrolysis of the NO₃RR and the OER by combining NiPr-TPA-COF as the cathode catalyst and IrO₂@Ti as the anode catalyst. When powered using a lithium-ion battery, the two-electrode cell stably delivered a current density of 23 mA cm⁻² and a great NH₃ selectivity of 91%. Our study here showcases the great potential of transition metal macrocycles for selective the NO₃RR to NH₃. The catalyst reported here consists of Earth-abundant elements (C, H, and N) and a small fraction of Ni, and is potentially more cost-effective than other candidate catalysts.

Author contributions

Y. Li proposed the concept and designed all experimental studies. Y. Hu, W. Huang, and F. Lv carried out materials synthesis, characterization, and electrochemical NO₃RR measurements. J. Xu performed the TEM analysis. B. Huang and M. Sun designed and performed DFT calculations. All authors contributes to the preparation of the manuscript. F. Lv, M. Sun and Y. Hu contributed equally in this work.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors acknowledge support from the National Natural Science Foundation of China (U2002213, 52161160331 and 21902114), the Science and Technology Development Fund Macau SAR (0077/2021/A2), the Collaborative Innovation Center of Suzhou Nano Science and Technology, the 111 Project, Joint International Research Laboratory of Carbon-Based Functional Materials and Devices, the National Natural Science Foundation of China/Research Grants Council (RGC) of Hong Kong Joint Research Scheme Project (N_PolyU502/21), the funding for Projects of Strategic Importance of The Hong Kong Polytechnic University (Project Code: 1-ZE2V), Research Institute for Smart Energy (RISE) and Research Institute for Intelligent Wearable Systems (RI-IWEAR) of the Hong Kong Polytechnic University.

References

1. N. Gruber and J. N. Galloway, Nature, 2008, 451, 293–296.
2. J. N. Galloway, A. R. Townsend, J. W. Erisman, M. Bekunda, Z. Cai, J. R. Freney, L. A. Martinelli, S. P. Seitzinger and M. A. Sutton, Science, 2008, 320, 889–892.
3. H. Liu, Chin. J. Catal., 2014, 35, 1619–1640.
4. B. H. R. Suryanto, H. L. Du, D. Wang, J. Chen, A. N. Simonov and D. R. MacFarlane, Nat. Catal., 2019, 2, 290–296.
5. J. Choi, B. H. Suryanto, D. Wang, H. L. Du, R. Y. Hodgetts, F. M. Ferrero Vallana, D. R. MacFarlane and A. N. Simonov, Nat. Commun., 2020, 11, 5546.
6. G. Soloveichik, Nat. Catal., 2019, 2, 377–380.
7. F. Lv, S. Zhao, R. Guo, J. He, X. Peng, H. Bao, J. Fu, L. Han, G. Qi, J. Luo, X. Tang and X. Liu, Nano Energy, 2019, 61, 420–427.
8. P. H. van Langevelde, I. Katsounaros and M. T. M. Koper, Joule, 2021, 5, 290–294.
9. Y. Wang, C. Wang, M. Li, Y. Yu and B. Zhang, Chem. Soc. Rev., 2021, 50, 6720–6733.
10. Y. Zeng, C. Priest, G. Wang and G. Wu, Small Methods, 2020, 4, 2000672.
11. Y. Ren, C. Yu, L. Wang, X. Tan, Z. Wang, Q. Wei, Y. Zhang and J. Qiu, J. Am. Chem. Soc., 2022, 144, 10193–10200.
12. J. Sun, D. Alam, R. Daiyan, H. Masood, T. Zhang, R. Zhou, P. J. Cullen, E. C. Lovell, A. R. Jalili and R. Amal, Energy Environ. Sci., 2021, 14, 865–872.
13. S. Garcia-Segura, M. Lanzarini-Lopes, K. Hristovski and P. Westerhoff, Appl. Catal., B, 2018, 236, 546–568.
14. A. Bhatnagar and M. Sillanpää, Chem. Eng. J., 2011, 168, 493–504.
15. 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, Nat. Nanotechnol., 2022, 17, 759–767.
16. V. Rosca, M. Duca, M. T. de Groot and M. T. M. Koper, Chem. Rev., 2009, 109, 2209–2244.
17. Y. Wang, A. Xu, Z. Wang, L. Huang, J. Li, F. Li, J. Wicks, M. Luo, D. H. Nam, C. S. Tan, Y. Ding, J. Wu, Y. Lum, C. T. Dinh, D. Sinton, G. Zheng and E. H. Sargent, J. Am. Chem. Soc., 2020, 142, 5702–5708.
18. Y. Wang, W. Zhou, R. Jia, Y. Yu and B. Zhang, Angew. Chem., Int. Ed., 2020, 59, 5350–5354.
19. G. F. Chen, Y. Yuan, H. Jiang, S. Y. Ren, L. X. Ding, L. Ma, T. Wu, J. Lu and H. Wang, Nat. Energy, 2020, 5, 605–613.
20. W. Jung and Y. J. Hwang, Mater. Chem. Front., 2021, 5, 6803–6823.
21. S. Yang, Y. Yu, X. Gao, Z. Zhang and F. Wang, Chem. Soc. Rev., 2021, 50, 12985–13011.
22. S. Dey, B. Mondal, S. Chatterjee, A. Rana, S. Amanullah and A. Dey, Nat. Rev. Chem., 2017, 1, 0098.
23. F. Lv, N. Han, Y. Qiu, X. Liu, J. Luo and Y. Li, Coord. Chem. Rev., 2020, 422, 213435.
24. N. Han, Y. Wang, L. Ma, J. Wen, J. Li, H. Zheng, K. Nie, X. Wang, F. Zhao, Y. Li, J. Fan, J. Zhong, T. Wu, D. J. Miller, J. Lu, S. T. Lee and Y. Li, Chem, 2017, 3, 652–664.
25. Z.-Y. Wu, M. Karamad, X. Yong, Q. Huang, D. A. Cullen, P. Zhu, C. Xia, Q. Xiao, M. Shakouri, F. Y. Chen, J. Y. T. Kim, Y. Xia, K. Heck, Y. Hu, M. S. Wong, Q. Li, I. Gates, S. Siahrostami and H. Wang, Nat. Commun., 2021, 12, 2870.
26. L. Sun, V. Reddu, A. C. Fisher and X. Wang, Energy Environ. Sci., 2020, 13, 374–403.
27. J. L. Segura, M. J. Mancheño and F. Zamora, Chem. Soc. Rev., 2016, 45, 5635–5671.
28. L. Cusin, H. Peng, A. Ciesielski and P. Samorì, Angew. Chem., Int. Ed., 2021, 133, 14356–14370.
29. W. Huang, W. Luo and Y. Li, Mater. Today, 2020, 40, 160–172.
30. H. L. Nguyen, C. Gropp and O. M. Yaghi, J. Am. Chem. Soc., 2020, 142, 2771–2776.
31. Y. Li, L. Guo, Y. Lv, Z. Zhao, Y. Ma, W. Chen, G. Xing, D. Jiang and L. Chen, Angew. Chem., Int. Ed., 2021, 133, 5423–5429.
32. W. Dai, F. Shao, J. Szczerbiński, R. McCaffrey, R. Zenobi, Y. Jin, A. D. Schlüter and W. Zhang, Angew. Chem., Int. Ed., 2016, 128, 221–225.
33. S. Lin, C. S. Diercks, Y. B. Zhang, N. Kornienko, E. M. Nichols, Y. Zhao, A. R. Paris, D. Kim, P. Yang and O. M. Yaghi, Science, 2015, 349, 1208–1213.
34. L. Sun, M. Lu, Z. Yang, Z. Yu, X. Su, Y. Q. Lan and L. Chen, Angew. Chem., Int. Ed., 2022, 134, e202204326.
35. X. Deng, Y. Yang, L. Wang, X. Z. Fu and J. L. Luo, Adv. Sci., 2021, 8, 2004523.
36. J. Li, G. Zhan, J. Yang, F. Quan, C. Mao, Y. Liu, B. Wang, F. Lei, L. Li, A. W. M. Chan, L. Xu, Y. Shi, Y. Du, W. Hao, P. K. Wong, D. Wang, J. Shi-Xue, L. Zhang and J. C. Yu, J. Am. Chem. Soc., 2020, 142, 7036–7046.
37. J. Yang, H. Qi, A. Li, X. Liu, X. Yang, S. Zhang, Q. Zhao, Q. Jiang, Y. Su, L. Zhang, J. Li, Z. Tian, W. Liu, A. Wang and T. Zhang, J. Am. Chem. Soc., 2022, 144, 12062–12071.
38. P. L. Searle, Analyst, 1984, 109, 549–568.
39. W. D. Zhang, H. Dong, L. Zhou, H. Xu, H. R. Wang, X. Yan, Y. Jiang, J. Zhang and Z.-G. L. Gu, Appl. Catal., B, 2022, 317, 121750.
40. J. Wang, T. Feng, J. Chen, V. Ramalingam, Z. Li, D. M. Kabtamu, J. H. He and X. Fang, Nano Energy, 2021, 86, 106088.
41. P. Gao, Z. H. Xue, S. N. Zhang, D. Xu, G. Y. Zhai, Q. Y. Li, J. S. Chen and X. H. Li, Angew. Chem., Int. Ed., 2021, 133, 20879–20884.
42. A. C. Nielander, J. M. McEnaney, J. A. Schwalbe, J. G. Baker, S. J. Blair, L. Wang, J. G. Pelton, S. Z. Andersen, K. Enemark-Rasmussen, V. Čolić, S. Yang, S. F. Bent, M. Cargnello, J. Kibsgaard, P. C. K. Vesborg, I. Chorkendorff and T. F. Jaramillo, ACS Catal., 2019, 9, 5797–5802.
43. K. S. Rajmohan and R. Chetty, ECS Trans., 2014, 59, 397–407.
44. H. Xu, Y. Ma, J. Chen, W. X. Zhang and J. Yang, Chem. Soc. Rev., 2022, 51, 2710–2758.
45. D. Reyter, D. Bélanger and L. Roué, Electrochim. Acta, 2008, 53, 5977–5984.
46. C. W. Lee, N. H. Cho, K. T. Nam, Y. J. Hwang and B. K. Min, Nat. Commun., 2019, 10, 3919.
47. R. Casebolt, K. Levine, J. Suntivich and T. Hanrath, Joule, 2021, 5, 1987–2026.
48. H. S. Jeon, J. Timoshenko, C. Rettenmaier, A. Herzog, A. Yoon, S. W. Chee, S. Oener, U. Hejral, F. T. Haase and B. Roldan Cuenya, J. Am. Chem. Soc., 2021, 143, 7578–7587.
49. L. Gao, L. Ding and L. Fan, Electrochim. Acta, 2013, 106, 159–164.
50. T. Liu, J. Wang, X. Yang and M. Gong, J. Energy Chem., 2021, 59, 69–82.
51. C. Kim, L. C. Weng and A. T. Bell, ACS Catal., 2020, 10, 12403–12413.
52. Y. Wang, C. Wang, M. Li, Y. Yu and B. Zhang, Chem. Soc. Rev., 2021, 50, 6720–6733.
53. M. Duca and M. T. M. Koper, Energy Environ. Sci., 2012, 5, 9726–9742.
54. P. Li, Z. Jin, Z. Fang and G. Yu, Energy Environ. Sci., 2021, 14, 3522–3531.
55. Q. Gao, H. S. Pillai, Y. Huang, S. Liu, Q. Mu, X. Han, Z. Yan, H. Zhou, Q. He, H. Xin and H. Zhu, Nat. Commun., 2022, 13, 2338.
56. X. Deng, Y. Yang, L. Wang, X.-Z. Fu and J.-L. Luo, Adv. Sci., 2021, 8, 2004523.
57. O. Q. Carvalho, R. Marks, H. K. K. Nguyen, M. E. Vitale-Sullivan, S. C. Martinez, L. Árnadóttir and K. A. Stoerzinger, J. Am. Chem. Soc., 2022, 144, 14809–14818.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d2ee02647c

‡ These three authors contributed equally.

---