Chemicals and reagentsAll chemical reagents and gases used in this work were purchased and used as received without additional purification. NO (99.99%), N2 (99.999%), Ar (99.999%) and 20 ppm NO (N2 as balance gas) were purchased from DONGHAE GAS IND. The following chemicals were purchased from Sigma-Aldrich: K2SO4 ( ≥ 99.0%), KHCO3 ( ≥ 99.5%), KCl (≥ 99.0%), zirconium(IV) chloride (≥ 99.9%, ZrCl4), terephthalic acid (98.0%), 2-methylimidazole (99.0%), Zn(NO3)2·6H2O ( ≥ 99.0%), Co(NO3)2·6H2O ( ≥ 99.0%), ethylene glycol (≥ 98.0%, EG), Cu(OH)2, polyethylene glycol 2050 (PEG 2050), L-ascorbic acid (≥99.0%), Cu(NO3)2·3H2O (99%) ethylenediamine (≥99.5%), hydrazine (98%) EDTA-2Na (99%), NiSO4·6H2O (99%), urea (99%), sulfamic acid (≥99.0%), phosphoric acid (85 wt%), p-aminobenzenesulfonamide (≥99.0%), and N-(1-Naphthyl) ethylenediamine dihydrochloride (≥99.0%). Methanol (99.5%), ethanol (99.5%), N,N-Dimethylformamide (99.9%, DMF), NaOH (98%), KOH (95%), concentrated HCl (35% – 37%) and H2O2 (35.5%) were purchased from SAMCHUN Chemicals. Deionized water was purified through a Millipore water purification system with a resistivity of 18.2 MΩ cm (at 25 °C). Carbon paper (CP, GDS250,) was purchased from CeTech without any further treatment.Synthesis of MOFs materialsTo prepare UIO-66, 1.06 g of zirconium(IV) chloride (ZrCl4) and 0.76 g of terephthalic acid were dissolved in 60 mL DMF with stirring until a homogenous solution was formed. Then, such solution was transferred into a 100 mL Teflon autoclave and heated at 120 °C for 24 h. After cooling down, the prepared UIO-66 was collected and washed by centrifugation several times in ethanol54. To prepare ZIF-8, 6.5 g of 2-methylimidazole and 3.0 g of Zn(NO3)2·6H2O were dissolved in 80 mL and 40 mL of methanol. Then Zn(NO3)2·6H2O solution was slowly added into 2-methylimidazole solution with vigorous stirring for 24 h. The obtained product was centrifuged with methanol for several times and then dried at 60 °C under vacuum overnight55. To prepare ZIF-67, 2.5 g of 2-methylimidazole and 2.2 g Co(NO3)2·6H2O were dissolved in 60 mL and 30 mL methanol. Then Co(NO3)2·6H2O solution was slowly dropped into 2-methylimidazole solution with vigorous stirring for 24 h. After that, the obtained product was centrifuged with methanol for several times and then dried at 60 °C under vacuum overnight59,60.Synthesis of Cu nanoparticles and Cu nanowiresCu nanoparticles were synthesized according to the reference procedure. Typically, 25 mL ethylene glycol (EG) contained with 0.98 g Cu(OH)2 and 1.0 g PEG 2050 (polyethylene glycol, MW: 2050) heated to 80 °C under stirring for 30 min. An L-ascorbic acid (6.0 g) EG solution was heated to 80 °C under the same condition. Then the L-ascorbic acid solution was poured into the former flask, and the mixture was kept at 80 °C for 5 min. Finally, the product washed several times with ethanol via centrifugation61. In a typical synthesis process of Cu nanowires, 41.5 mL 0.1 mol/L Cu(NO3)2 aqueous solution was added into 833 mL of 10 M NaOH aqueous solution under vigorous stirring. After that, 6.225 mL ethylenediamine and 1.94 mL 35 wt% hydrazine aqueous solution was added in the above solution following with stirring. Then, such solution transferred to a fan oven under 60 °C for 2 hours. The afloat flocculent copper products washed with centrifugation several times in D.I. water, and such products stored in ascorbate acid containing ethanol solution to prevent oxidation62.Synthesis of NiO nanoparticlesIn a typical process, 0.3 g EDTA-2Na, 1.2 g of NiSO4·6H2O, and 0.3 g of urea were dissolved in 15 mL of D.I. water, respectively. Then the NiSO4 solution was added slowly to the EDTA-2Na solution with stirring, until the mixed solution color turned to deep blue. After that, the urea solution was also dropped into the above-mixed solution and stirred for 3 min. The pH value of the reaction system was adjusted to 6 by ammonia and a dilute sulfuric acid solution. The obtained solution was transferred to a 60 mL Teflon-lined autoclave and kept at 180 °C for 4 h. After the reactor cooled to room temperature, the green powder was collected by filtration and dried at 60 °C63. Subsequently, the NiO nanoparticles were obtained by annealing the above-mentioned green powders at different temperature (400, 500, and 600 °C) and different atmosphere (air, and Ar) for 2 h in a flow furnace. Those NiO HA samples came from the prepared NiO Air samples which further treated in 5% H2/Ar mixture gas flow for 10 minutes at different temperature.Preparation of MOFs-modified GDEsThe MOFs modified GDEs were prepared using a dropping method. A carbon paper (CP) (15 mm × 15 mm) was used as the substrate for MOFs-modified GDEs. Firstly, 10 mg as-synthesized UIO-66, ZIF-8 or ZIF-67 was dispersed in 500 µL of ethanol under ultrasonic to prepare MOFs ink. Then, such MOFs ink was slowly dropped on one side of carbon paper to load MOFs. Then, the catalyst ink was prepared and dropped on another side of carbon paper which mixed 5 mg Cu NWs (or Cu NPs) or 10 mg NiO NPs, 950 µL of ethanol and 50 µL of Nafion solution (5 wt% in ethanol, Sigma-Aldrich). All control samples prepared by using the same dropping method, UIO-66/CP, ZIF-8/CP and ZIF-67/CP were prepared without dropping Cu NWs ink. CP/Cu NWs, CP/Cu NPs and CP/NiO were prepared without dropping MOFs ink.Electrocatalysis tests and electrochemical measurementsAll the electrocatalysis tests and measurements were conducted using a potentiostat (Gamry Instruments Reference 600) in a flow cell reactor with a standard three-electrode set up. In all of the electrochemical measurements, iR compensation was not applied. During NO electro-reduction, the prepared MOFs-modified GDE with Cu-based catalysts was served as the working cathode. Pt wire was served as the counter electrode and Ag/AgCl (in 3 M KCl) was used as the reference electrode. The external calibration method was used to calibrate the Ag/AgCl reference, a saturated calomel electrode (SCE) is employed for calibration. The calibration was performed by measuring the potential difference between the two electrodes. Unless otherwise specified, in general electrochemical tests, the anode and cathode were separated by a proton exchange membrane (Nafion 117, 1.5 cm × 1.5 cm). Nafion membrane need to be pretreated before use. Typically, the membrane was placed in a 1 mol/L H2SO4 solution and heated at 90 °C for 1 h. After that, using deionized water washed several times. Then, put it into a 5% H2O2 solution and treated at 90 °C for 1 h follow with deionized water wash several times. Finally, the treated membrane stored in deionized water for use. In the flow cell setting, both cathode chamber volume and anode chamber volume were 1 cm3 (1 cm × 1 cm × 1 cm), 10 mL electrolytes were circulated in both cathode and anode chamber controlled by a peristaltic pump with the flow rate of 2.0 mL/min.The NO electro-reduction tests were carried out with varying the concentration of NO gas that flowed through the gas chamber with a flow rate of 250 mL/min controlled by a mass flow controller. As for the NO electro-oxidation setup, the prepared NiO-based GDE served as the working anode and Pt wire as the counter electrode for NO electro-oxidation. During the tests, 10 mL electrolyte (0.5 M K2SO4, pH = 6.8) was circulated into both cathode and anode chamber with the flow rate of 2.0 mL/min. Different concentration of NO gas or Ar gas flowed through the gas chamber with a flow rate of 250 mL/min. The settings for integrated NO electrocatalysis were similar to those of the single GDE system, except that both Cu-based GDEs and NiO-based GDEs were utilized in a single cell. This configuration allowed the simultaneous occurrence of both NORR and NOOR. Two gas chambers were installed on both side of the cathode and anode, and NO-containing gas passed through both gas chambers with the flow rate of 250 mL/min. A specific amount of 0.5 M K2SO4 electrolyte was circulated in the flow cell at the flow rate of 2.0 mL/min. Chronoamperometry (CA) method was used to test the performance for NORR and NOOR. The scanning rate of linear sweep voltammetry (LSV) measurements were 10 mV/s. The applied potential on the working electrode was rescaled to the reversible hydrogen electrode (RHE) reference using the following Eq. (11):$${E}_{{RHE}}\,=\,{E}_{{Ag}/{AgCl}}+0.197+0.0592\,\times {{{\rm{pH}}}}$$
(11)
Products determination (NH4
+, NO2
− and NO3
−)We employed the colorimetric method for analysis and detection of NH4+, NO3−, and NO2− as the high background concentration of electrolytes in the samples interfere with the ion chromatographic analysis of product ions. An ultraviolet-visible (UV–vis) spectrophotometer was utilized to detect the NORR and NOOR products of NH4+, NO2− and NO3− by recording the absorption intensity of characteristic peak, and calibration curves were used to calculate the product concentration (see Supplementary Fig. 35)64. The Nessler’s reagent as the color reagent was used to determine the concentration of NH4+65. Typically, a certain amount of cathode electrolyte for NO electro-reduction was sampled from the flow cell and diluted to 3 mL with 0.5 mol/L KOH aqueous solution, then 60 μL Nessler’s reagent was added into the diluted sample with stirring for 20 min. After that the absorption intensity at a wavelength of 420 nm was recorded for NH4+ concentration calculation. 1 M HCl and 0.8 wt% sulfamic acid solution were used for NO3− determination66. Briefly, a certain amount of anode electrolyte for NO electro-oxidation was sampled and diluted to the detection range. Then, 60 µL of 1 M HCl and 6 µL of 0.8 wt% sulfamic acid were added into 3 mL diluted samples. After 5 minutes, the absorption intensity at a wavelength of 220 nm and 275 nm were recorded. And the final absorption value was calculated by the following Eq. (12):$${{{\rm{A}}}}={A}_{220{nm}}-{2A}_{275{nm}}$$
(12)
For the determination of NO2−, a mixture solution contained 4.0 g p-aminobenzenesulfonamide, 0.2 g N-(1-Naphthyl) ethylenediamine dihydrochloride, 50 mL D.I. water and 10 mL concentrated phosphoric acid was utilized as a color reagent67. Similar to determination NO3−, a certain amount of anode electrolyte was sampled and diluted to the detection range, firstly. Then, 60 µL of color reagent was added into the diluted sample with stirring for 20 min. After that, the absorption intensity was recorded at a wavelength of 540 nm. The calculation of these products were related to the concentration-absorbance plots. The Faradaic efficiency calculation for these products used the following Eq. (13):$${FE}\left(\%\right)=\frac{n\,\times \,F\,\times \,C\,\times \,V}{Q}\times 100(\%)$$
(13)
Where n is the electron transfer number (NO to NH4+, NO to NO3− and NO to NO2−, the value of n = 5, 3 and 1), F is the Faraday constant (96500 C/mole e−), C is the concentration of N-containing products (mol/L), V is the electrolyte volume (L), Q is the total charge (C).Measurements of NO transfer flux, transfer efficiency and NO utilization efficiencieshTo measure the NO transfer flux and transfer efficiency on MOFs modified GDE, NO molecular should be converted to NO3−, initially. This conversion process allowed for the quantification of the NO3− production rate, which served as an indicator of the transfer flux of NO. Then the amount of NO3− in electrolyte was analyzed by using color reagent as the same method as the determination of NO3− product for NO electro-oxidation. Hydrogen peroxide (H2O2) was used to oxidize NO molecular to NO3− according to the following chemical reaction Eq. (14):$$2{{{\rm{NO}}}}+{3{{{\rm{H}}}}}_{2}{{{{\rm{O}}}}}_{2}\to 2{{{\rm{HN}}}}{{{{\rm{O}}}}}_{3}+{2{{{\rm{H}}}}}_{2}{{{\rm{O}}}}$$
(14)
In a flow cell setting as electrocatalysis tests without applied potential, the prepared MOFs modified GDEs (without catalyst loading) were used to separate the gas chamber and liquid chamber. In this configuration, NO containing gas flowed through the gas chamber and was adsorbed by MOFs modified GDEs (see Supplementary Fig. 14). Subsequently, the adsorbed NO diffused to liquid phase and convert to NO3− by H2O2. This configuration allowed for the investigation of NO adsorption and capture capacity from gas phase on the prepared GDEs, thereby evaluating their effectiveness as gas separation membranes. During the measurements, 10 mL mixture solution contained 0.5 M K2SO4 and 0.5 M H2O2 was circulated in the liquid chamber with a flow rate of 2.0 mL/min, and a series of NO containing gas continuously passed through the gas chamber with the flow rate of 250 mL/min. After running for 20 minutes, a certain amount of liquid were taken out and added the color reagent for NO3− determination following with the analysis by ultraviolet-visible (UV–vis) spectrophotometer. The calculations of NO3− amount were related to the concentration-absorbance plots. In this method, the concentration of NO3− equaled to the concentration of NO transfer amount. The calculation of NO transfer flux and transfer efficiency were according to the following Eqs. (15, 16):$${{{\rm{NO}}}} \; {{{\rm{transfer}}}} \; {{{\rm{flux}}}}=\frac{{C}_{{{NO}}_{3}^{-}}\times \,{V}_{{electrolyte}}}{{T}_{{time}}}$$
(15)
$${{{\rm{NO}}}} \; {{{\rm{transfer}}}} \; {{{\rm{efficiency}}}}=\frac{{n}_{{{NO}}_{3}^{-}}}{{n}_{{NO\; in\; gas\; flow}}}\times 100\%$$
(16)
For the integrated NO electrocatalysis section, the calculation of NO utilization efficiencies was according to the following Eq. (17):$${{{\rm{NO}}}} \; {{{\rm{transfer}}}} \; {{{\rm{flux}}}}=\frac{{n}_{{{NH}}_{4}^{+}}+{n}_{{{NO}}_{3}^{-}}\,}{{n}_{{NO}({total})}}\times 100\%$$
(17)
Where the total n (NO) in double gas line system was combine both cathode gas and anode gas, and the total n (NO) in single gas line system was only one gas flow.Characterization for GDEThe morphology of such prepared samples was characterized using a field emission scanning electron microscope (FESEM, JSM 7800 F). The X-ray diffraction (XRD) patterns for all the samples were obtained using an X-ray diffractometer (Ultima IV) with Cu-Kα radiation (λ = 1.54178 Å). The surface area was investigated by N2 adsorption measurements on Micromeritics TriStar II analyzer by the Brunauer-Emmett-Teller (BET) method. The NO-TPD measurements were performed by chemisorption analyzer (PCA-1200). Transmission electron microscopy (TEM) images were obtained using a JEM-1400 (JEOL Ltd., Japan) at 120 kV. The X-rays photo spectroscopy (XPS) of NiO samples were measured by Nexsa (ThermoFisherScientific) at a base pressure of 2.0 × 10−8 mBar with monochromated Al Kα (1486.6 eV) radiation, the C 1 s (284.8 eV) was used as the internal standard for charging correction.Density functional theory calculationThe density functional theory (DFT) computations were performed using the Vienna ab initio simulation package (VASP)68,69. The interaction between the ionic core and valence electrons was described by the projector augmented wave method (PAW)70,71. The total energy convergence and the forces on each atom were set to be lower than 10−6 eV and 0.02 eV A−1. Energy cutoff of 400 eV for the plane wave basis set was used for the structure optimization. The construction of the MOFs models (UIO-66, ZIF-8 and ZIF-67) and carbon model utilized data from the crystallographic database. The 11 × 11 × 11 Monkhosrt-Pack generated k point mesh was used to sample the Brillouin zone for the primitive cells of bulk Cu, Ni and NiO72. A 4-atom primitive cell was built and optimized to define lattice parameters, the optimized lattice parameters of fcc Ni and NiO are 3.48 and 4.17 Å, respectively. For Cu and Ni slab, a 3 × 3 × 4 supercell was constructed from the primitive cell along the (111) direction. For NiO, a 4 × 4 × 4 supercell was built in the (100) direction. A 15 Å vacuum layer was added along the Z-direction to avoid interactions between periodic image replicas. During simulations, the bottom two layers of atoms were fixed, with the rest fully relaxed. For the slab models, 3 × 3 × 1 Monkhosrt-Pack generated k-points mesh was sampled from the Brillouin zone. We used the Gaussian Methfessel-Paxton smearing of 0.173. For ZIF-8, UIO-66, ZIF-67, gamma point in the Brillouin zone was sampled due to the large size of the cell. The Predew-Burke-Ernzerhor (PBE) functional with generalized gradient approximation was employed to describe the electron exchange and correlation energy69,74. To accurately describe systems containing Ni atoms, spin-polarized calculations were enabled for all Ni atoms. For the Ni slab model, the spin configuration was set to a ferromagnetic state; for the NiO slab model, the thermodynamically more stable Type-II antiferromagnetic phase spin arrangement was adopted75,76. Grimme’s DFT-D2 functional, was used to correct the dispersion forces. For the calculation of Ni and NiO system, we used the DFT + U correction where the U value of 6.3 eV was used for d electrons of Ni based on the previous studies75,77,78,79. For the heterojunction calculation, we considered the two layer Ni island on the NiO surface (100), as observed from the TEM images. We considered the Ni island instead of Ni slab connected across the periodic image in order to eliminate the undesired strain arising from lattice mismatch. All of the structures data for the optimized models are provided in Supplementary Data 1.We modeled the oxidation reaction of NO(g) on Ni, NiO, and Ni/NiO with the Eqs. (4–10)80, the Eqs. (18–24) to calculate the Gibbs free energy of reaction, \(\Delta {G}_{i}\), of the above NO oxidation reaction in order are:$$\Delta {G}_{1}={G}_{{{{\rm{NO}}}}*}-{G}_{{{{{\rm{NO}}}}}_{(g)}}-{G}_{*}$$
(18)
$$\Delta {G}_{2}={G}_{{{{{\rm{HNO}}}}}_{2}*}+{\mu }_{{{{{\rm{H}}}}}_{({{{\rm{aq}}}})}^{+}}+{\mu }_{{e}^{-}}-{G}_{{{{\rm{NO}}}}*}-{G}_{{{{{\rm{H}}}}}_{2}{{{{\rm{O}}}}}_{({{{\rm{l}}}})}}$$
(19)
$$\Delta {G}_{3}={G}_{{{{{\rm{NO}}}}}_{2}*}+{\mu }_{{{{{\rm{H}}}}}_{({{{\rm{aq}}}})}^{+}}+{\mu }_{{e}^{-}}-{G}_{{{{{\rm{HNO}}}}}_{2}*}$$
(20)
$$\Delta {G}_{4}={G}_{{{{{\rm{HNO}}}}}_{3}*}+{\mu }_{{{{{\rm{H}}}}}_{({{{\rm{aq}}}})}^{+}}+{\mu }_{{e}^{-}}-{G}_{{{{{\rm{NO}}}}}_{2}*}-{G}_{{{{{\rm{H}}}}}_{2}{{{{\rm{O}}}}}_{({{{\rm{l}}}})}}$$
(21)
$$\Delta {G}_{5}={G}_{{{{{\rm{NO}}}}}_{3}*}+{\mu }_{{{{{\rm{H}}}}}_{({{{\rm{aq}}}})}^{+}}+{\mu }_{{e}^{-}}-{G}_{{{{{\rm{HNO}}}}}_{3}*}$$
(22)
$$\Delta {G}_{6}=-\left({G}_{{{{\rm{N}}}}{{{{\rm{O}}}}}_{2}*}+{\mu }_{{{{{\rm{H}}}}}_{({{{\rm{aq}}}})}^{+}}+{\mu }_{{e}^{-}}-{G}_{{{{\rm{H}}}}{{{{\rm{N}}}}{{{{\rm{O}}}}}_{2}}_{\left({{{\rm{a}}}}{{{\rm{q}}}}\right)}}-{G}_{*}\right)$$
(23)
$$\Delta {G}_{7}=-\left({G}_{{{{\rm{N}}}}{{{{\rm{O}}}}}_{3}*}+{\mu }_{{{{{\rm{H}}}}}_{({{{\rm{aq}}}})}^{+}}+{\mu }_{{e}^{-}}-{G}_{{{{\rm{H}}}}{{{{\rm{N}}}}{{{{\rm{O}}}}}_{3}}_{\left({{{\rm{aq}}}}\right)}}\,-{G}_{*}\right)$$
(24)
Where Gi indicates the Gibbs free energy of the species i, \({\mu }_{{{{{\rm{H}}}}}_{({{{\rm{aq}}}})}^{+}}\) and \({\mu }_{{e}^{-}}\) are the chemical potential of proton and electron, respectively, and \({G}_{{{{\rm{sol}}}},i}\) is the solvation energy of the species i. For the desorption reaction of anion species, NO2−(aq) and NO3−(aq), we adopted the formula from the Kamat et al80. The adsorption energies of NO(g) on carbon paper, ZIF-8, UIO-66, ZIF-67, and Cu (111) were calculated using \(\Delta {G}_{1}\). As the desorption reaction occur at the infinite dilution, we did not consider the dilution energy as it was considered in Kamat et al.80The Gibbs free energy of species i, \({G}_{i}\), is calculated by using the following Eq. (25):$${G}_{i}={E}_{{{{\rm{DFT}}}},i}+{H}_{T,i}-T{S}_{T,i}+{G}_{{{{\rm{sol}}}},i}$$
(25)
where \({E}_{{{{\rm{DFT}}}},i}\) is the DFT calculated energy, \({H}_{T,i}\) is the enthalpy of the species i from the vibrational contribution for surface species, as well as the rotational and translational contribution for the aqueous and gaseous species, T is the temperature, \({S}_{T,i}\) is the entropy of the species i from the vibrational contribution for surface species, as well as the rotational and translational contribution for the aqueous and gaseous species, and \({G}_{{{{\rm{sol}}}},i}\) is the Gibbs free energy of solvation. \({H}_{T,i}\) contains the zero-point energy (ZPE) corrections from the vibrational contribution calculated using the harmonic oscillator model. In the case of surface, or the substrate, we did not consider \({H}_{T,i}\), and \(T{S}_{T,i}\). The solvation energy was not considered for NO(g), H2(g) or, H2O(l). For the Gibbs free energy of H2O(l), we calculated the Gibbs free energy of the gaseous H2O(g) at the vapor pressure of 0.0613 atm81.The solvation free energy of the solute i, \(\Delta {G}_{{{{\rm{sol}}}},i}\), at infinite dilution was calculated using implicit solvent method22,82,83,84. The solvation free energy is calculated as the difference between energy in the solvated environment, \({E}_{{{{\rm{DFT}}}},{{{\rm{s}}}}{{{\rm{ol}}}},i}\), and energy in the vacuum, \({E}_{{{{\rm{DFT}}}},i}\), as following Eq. (26):$${G}_{{{{\rm{sol}}}},i}={E}_{{{{\rm{DFT}}}},{{{\rm{s}}}}{{{\rm{ol}}}},i}-{E}_{{{{\rm{DFT}}}},i}$$
(26)
The computational hydrogen electrode (CHE) was used to calculate the electrochemical proton transfer reaction. The chemical potential of proton and the electron is calculated as Eq. (27)81,85:$${\mu }_{{{{{\rm{H}}}}}_{({{{\rm{aq}}}})}^{+}}+{\mu }_{{e}^{-}}=\frac{1}{2}{G}_{{{{{\rm{H}}}}}_{2({{{\rm{g}}}})}}-{{{\rm{e}}}}{U}_{{{{\rm{RHE}}}}}$$
(27)
Where e is the elementary charge, and \({U}_{{{{\rm{RHE}}}}}\) is the applied potential with reversible hydrogen electrode.The calculation energies, zero-point energies, entropy corrections and solvation energy corrections of each species in the free energy calculations were listed in Supplementary Table 4 and Supplementary Table 5.
Integrated electrocatalytic synthesis of ammonium nitrate from dilute NO gas on metal organic frameworks-modified gas diffusion electrodes
