Synergistic Cu-Sb oxide electrocatalyst synthesized by sol-gel method for high-efficiency nitrate reduction to ammonia with suppressed hydrogen evolution Authors: AliReza Babaei, Milad Rezaei, Pirooz Marashi, William Mark Rainforth Published: 2025 ABSTRACT AbstractAntimony is one of the hydrogen evolution poison and copper is one of the well-known catalysts for the nitrate reduction to ammonia (NRA). Therefore, to synthesize an efficient electrode that can have the highest selectivity for the reduction of nitrate to ammonia and is accompanied by the lowest hydrogen evolution, in this study, a combination of different Cu to Sb ratios was prepared on a Ti substrate using the sol-gel method in the form of crystalline oxides (CuSbyOₓ/Ti) to achieve the best catalytic activity and ammonia yield. A 2:1 ratio of Sb to Cu synthesized and calcined at 600 °C had the best catalytic properties. CuSbyOₓ/Ti electrocatalysts were evaluated for NRA in an alkaline medium. The CuSb2Ox electrocatalyst had the highest activity with the lowest cathodic Tafel constant of 78 mV dec− 1. EIS results also showed that CuSb2Ox has the lowest charge transfer resistance (Rtotal=3.75 Ω cm2). The measurements revealed that the CuSb2Ox electrocatalyst performed superlative, with an NH3 yield of 7280 µg h− 1 cm− 2 and an F.E. of 89% at a potential of -0.7 V vs. RHE. The high NH3 yield of this oxide electrocatalyst, compared to other oxide compounds, highlights the synergistic effect of Cu and Sb.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-32713-4.Keywords: Electrocatalyst, Electrocatalytic nitrate reduction to ammonia, NH3 yield, Bimetallic, Ammonia synthesisSubject terms: Chemistry, Materials science Introduction Ammonia is a widely used material for the production of a wide range of fertilizers, chemicals, and pharmaceuticals1,2. Due to its high energy density and hydrogen storage capacity, NH₃ can also be considered a renewable energy source3. Industrial production of NH₃ is of particular importance given the global demand for nitrogen-containing fertilizers for example ammonium nitrate (NH₄NO₃), urea (CO(NH2)2), and various nitrogen compounds. This sector has contributed significantly to increasing global food production by a factor of 7 to 8 since the beginning of the 20th century4. NH₃ is mainly generated by the well-known Haber-Bosch process5. However, the cleavage of the stable nitrogen-nitrogen bond (N ≡ N) in N₂, which is required for the reaction with hydrogen (H₂), takes place under harsh conditions of high temperature (between 400 and 600 °C) and pressure (⁓400 atm). This indicates the high energy consumption in the Haber-Bosch process6. Furthermore, ammonia production accounts for more than 1% of global energy consumption and contributes significantly to CO2 emissions7. The hydrogen needed for ammonia synthesis is primarily obtained through coal gasification or natural gas reforming, processes that emit about 0.5 billion tons of CO₂ each year. Since Haber–Bosch process entails harsh conditions, intricate infrastructure, high energy intensity, and notable environmental impacts, it is crucial to address its shortcomings. In light of escalating energy demand and ongoing environmental challenges, it is imperative to develop environmentally friendly ammonia synthesis technology that operates under milder process8. The electrochemical nitrogen reduction reaction (NRR) offers several advantages, including lower energy consumption, a better carbon footprint, and a simpler reactor design910, However, at high current densities, the electrochemical NRR process still exhibits a relatively low ammonia yield, significantly lower than that of the Haber-Bosch process. Therefore, replacing N₂ with nitrogen sources, especially nitrate/nitrite (NO₃⁻/NO₂⁻), is considered a promising option to reduce energy requirements, enhance ammonia production, and improve the energy efficiency of the system11. Nitrate/nitrite reduction to ammonia (NRA) offers an effective solution to address the energy crisis and control NO₃⁻/NO₂⁻ contamination in wastewater. Due to the global use of nitrogen fertilizers, fossil fuel combustion, and wastewater discharge, large amounts of NO3− have been identified in surface and groundwater12. In order to protect health, the WHO and the U.S. EPA have set maximum permissible levels of 10 mg/L (NO₃⁻) and 1 mg/L (NO₂⁻) in drinking water13. Given the high NO₃⁻ content, the electrochemical eight-electron transfer in the nitrate reduction to ammonia is considered a more efficient process to address energy and environmental problems14. Various electrocatalytic systems have been developed for NRA, using heterogeneous catalysts based on different metals includes platinum group, copper, etc. However, at high potentials, the hydrogen evolution reaction (HER) dominates, which is unfavorable for the NRA process. In fact, the HER competes with the NRA, consuming electrons and reducing the overall efficiency. Therefore, the development of electrocatalysts that can selectively reduce NO₃⁻/NO₂⁻ to NH₃ while simultaneously suppressing the HER is highly desirable15–21. However, the high cost and limited availability of noble metals have significantly hampered their development and widespread application. In contrast, base-metal-based electrocatalysts are more commonly used, making their development more desirable22–24. Furthermore, the development of new technologies for green ammonia synthesis (electrochemical reaction of nitrate/nitrite reduction to ammonia) is of great importance for human society25–28. The strong synergy between two components, optimizes the electron transport properties, tunes the active sites, and improves the efficiency of nitrate ion adsorption in the multicomponent oxide electrocatalyst. Recent studies have investigated multicomponent oxide electrocatalysis systems for the electrochemical synthesis of ammonia from nitrate, focusing on advanced catalysts to improve efficiency and stability29,30. Li and colleagues30 investigated the formation of a two-dimensional heterogeneous compounds of copper and iron oxide, which enhanced the electrochemical performance by increasing the catalytic activity and improving the stability of nitrate reduction. Tang and colleagues31 developed a heterogeneous CoO-CuOx structure for the electrochemical nitrate reduction to ammonia. This heterogeneous structure combined the properties of cobalt oxide (CoO) and copper oxide (CuOx), thus improving the electrochemical efficiency. The heterostructure of CoO-CuOx demonstrated high efficiency in nitrate reduction and benefited from increased surface area and improved electrical conductivity, while exhibiting good stability under various reaction conditions. Furthermore, Huang et al.32 introduced 3D flower-like zinc cobaltite (ZnCo₂O₄) as an impressive electrocatalyst for the NRA. This structure offered a high surface area, enabled facile electron/ion transfer, and significantly improving the electrochemical performance. The system demonstrated high efficiency and stability under ambient conditions, making it a promising catalyst for environmental and industrial applications. Further advances in electrocatalysis were reported by Hai et al.33. They introduced NiCoO₂ nanoarrays on copper foam, which demonstrated effective catalysis for the nitrate reduction to ammonia (NRA). The copper foam acted as a three-dimensional support increasing the active surface area and stability of the catalyst over time, thus significantly improving the electrochemical performance. It should be noted that the performance of NRA systems can be significantly improved, but their practical application still presents several challenges. Due to the competitive reaction of HER during NRA, the selectivity and Faradaic efficiency of electrocatalysts remain important challenges. In this study, we will incorporate copper, a widely used element in NRA processes, along with antimony as a hydrogen release inhibitor (since this element can prevent the hydrogen (H2) evolution34, into the electrocatalyst composition. Copper acts as the active element for NRA, while antimony suppresses hydrogen gas production and provides stability in the system. This study demonstrates that the synergy between copper oxide and antimony oxide (copper antimonates) creates a powerful catalyst for the nitrate reduction to ammonia. The electrocatalytic stability, NO₃⁻ conversion, and selectivity for different molar ratios of antimony to copper, from Sb: Cu = 0.5 to 2.5, in the NRA process were evaluated. Copper antimonate (CuSb2OX with Sb: Cu = 2) in 0.1 M NaOH with 0.1 M NaNO₃ achieves a Faradaic efficiency of 89% and an NH₃ yield of 7320 µg hr− 1 cm− 2 at optimum working potential of −0.7 V vs. RHE. Results and discussion Morphology, chemical composition, and structure of the electrocatalysts FESEM, elemental mapping, and EDS results for CuSb2.5Ox, CuSb2Ox, CuSb1.5Ox, CoSbOx, and CuSb0.5Ox are shown in Fig. 1. FESEM images showed that the morphology of each electrocatalyst, depending on the ratio of Sb to Cu, differed from the others. The FESEM images at higher magnification (Fig. S3) show that the particles that make up the electrode contain a 2:1 ratio of Sb to Cu (CuSb2Ox electrocatalyst), is in the range of 50 to 120 nm. Figure 1 and the EDS-Mapping results also demonstrate that the elements are evenly distributed throughout the coating. Fig. S4 shows the FESEM image along with the EDS-mapping of the cross-section of the coating. As can be seen in Fig. S4, the coating thickness from the set of measurements (according to ASTM B487) was 38.0 ± 4.4 μm, indicating an acceptable uniformity of the coating on the electrode surface. To obtain a more precise chemical composition of the electrocatalysts, ICP-MS was used in addition to EDS analysis. Table 1 shows the EDS and ICP results for each electrocatalyst. According to the EDS and ICP results, the copper and antimony concentrations in the coating are consistent with the precursors during the synthesis of CuSb2Ox and CuSb1.5Ox. However, the stoichiometric ratio of antimony to copper in other compounds is not necessarily the same as its amount in the precursor solution, which may be due to the difference in the evaporation rate of chloride salts at different molar ratios of Sb to Cu. In addition, it should be noted that the amount of oxygen in these compounds, due to the weakness of EDS in detecting elements lighter than sodium, cannot be the same as its actual amount, and therefore, X-ray diffraction must be employed to detect the chemical composition of the formed mixed oxide. PXRD patterns of catalysts synthesized from different molar ratios of antimony and copper are shown in Fig. 2. According to Fig. 2, CuSb2O6 is the main phase. Figure 2 shows that, as expected, the compound with a molar ratio of 2:1 antimony to copper exhibits the highest percentage of the CuSb2O6 phase. At a ratio of 2.5:1, the excess antimony appears as the oxide phases of Sb2O4 and/or Sb2O5. At ratios of 1:1 and 1.5:1, the excess copper is observed as divalent copper oxide. However, at a ratio of 0.5:1, the target phase is not visible, and the Cu9Sb4O19 phase is displayed along with CuO as the main phases. It is very likely that the lack of formation of CuSb2O6 phase could be the reason for the lack of suitable electrocatalytic performance for this compound (see Sect. 3.2.). SEM images, elemental mapping, and EDS analysis results (atomic percentage and weight%) for a) CuSb2.5Ox, b) CuSb2Ox, c) CuSb1.5Ox, d) CoSbOx and e) CuSb0.5Ox. Table 1Comparison of EDS and ICP results for determining the molar ratio of copper and antimony in the chemical composition of electrocatalysts.ElectrocatalystICPEDSCu(ppm)Sb(ppm)Sb: Cu (at./at.)Cu (at%)Sb (at%)Sb: Cu (at./at.)CuSb2Ox388.911510.4 2.03 20.6040.26 1.95 CuSb1.5Ox440.031199.3 1.42 20.7532.17 1.55 CuSbOx503.21892.22 0.93 21.7119.22 0.89 CuSb0.5Ox712.37409.69 0.30 17.646.73 0.38 CuSb2.5Ox284.321608.1 2.95 16.3344.06 2.70 Comparison of EDS and ICP results for determining the molar ratio of copper and antimony in the chemical composition of electrocatalysts. PXRD patterns for the CuSb2Ox (purple), CuSb2.5Ox (black), CuSb1.5Ox (red), CuSbOx (blue), and CuSb0.5Ox (green) compounds calcined at 600 °C along with phase identification. On the other hand, the PXRD pattern of CuSb2O6 revealed a slightly distorted monoclinic phase (space group P21/n) with lattice constants a = b = 4.643 Å, and c = 9.299 Å. In the tri-rutile structure of CuSb2O6, the triplication of the c axis generates in a lattice of CuO6 and SbO6 octahedra split at the edges and corners. Arrays of magnetic Cu²+ are sandwiched between layers of diamagnetic Sb5+, which creates quasi-one-dimensional antiferromagnetism through an in-plane super-exchange mechanism. According to the Jahn–Teller distortion of Cu²⁺35, CuSb2O6 occurs in two polymorphs, α and β, which are distinguished primarily by the way in which the CuO₆ octahedra are compressed or stretched. The β polymorph is characterized by a distortion perpendicular to the one-dimensional magnetic chain, while α-CuSb2O6 exhibits octahedral compression along the chain. The aforementioned changes cause the CuSb2O6 compound to crystallize in both the α- and β-states, namely tetragonal (P42/mnm) and monoclinic (P21/n), respectively35–38. In this study, after calcination of all the electrocatalysts at 600 °C, CuSb2O6 transformed from the α-polymorphic tri-rutile structure (P42/mnm) to the β form, which crystallizes in the monoclinic space group P2₁/n. Electrocatalytic activity, cathodic Tafel Constant (βc), electrochemical active surface area, and charge transfer resistance of CuSbyOx The LSV curves in Fig. 3a show that the presence of nitrate ions in the solution causes a significant increase in the cathodic current density values, indicating a strong reduction reaction at the surface of the electrocatalysts. For the CuSb0.5Ox electrocatalyst, the current increase at more positive potentials is due to the partial reduction of copper oxide due to the presence of excess copper oxide based on PXRD results. For all electrocatalysts, the nitrate reduction current density is observed at a potential of approximately − 0.5 V vs. RHE39. It is noteworthy that the LSV curves of the CuSb2Ox electrocatalyst, in presence and absence of nitrate ions, shows the largest change compared to the other electrocatalysts. Moreover, increasing the antimony to copper ratio to 2:1 shows the best electrocatalyst activity in term of reaction rate. The LSV results of CuOx and SbOx along with the optimal ratio of copper and antimony (CuSb2Ox) are presented in Fig. S5. As can be seen in Fig. S5, CuOx shows a broad peak indicating the reduction of Cu2+ to Cu+ and Cu0. In contrast, SbOx does not exhibit suitable electrocatalytic activity and due to the lack of reduction of antimony oxide in an aqueous environment, its composition remains constant. The combination of a 2:1 ratio of antimony to copper in CuSb2Ox resulted in a significant increase in electrocatalytic activity, and almost no trace of the copper ion reduction peak is observed in the CuSb2Ox electrocatalyst, which confirms the synergistic role of the simultaneous presence of Cu and Sb. The activity of the CuSb2Ox and CuSb1.5Ox electrocatalysts was measured at different nitrate concentrations, 0.010 M to 0.100 M , which shows the good activity33 of the CuSb2Ox electrocatalyst for the NRA process as shown in Fig. S6 a. This study (Fig. 6a and b) confirms the assignment of the cathodic peak of nitrate reduction at a potential of −0.5 vs. RHE for the CuSbyOx electrocatalysts at different nitrate concentrations. a) LSV curves in 0.1 M NaOH with and without 0.1 M NaNO3. b) Tafel diagrams. c) ECSA curves (double layer capacitance values) ​​and d) Nyquist diagrams of the EIS results along with the data fitting for the various synthesized electrocatalysts. a) Concentrations of and NH3-N after the 15 h NRA process. b) Conversion of and selectivity of NH3 during the 15 h NRA process. Figure 3b shows the Tafel diagrams obtained from the LSV curves. According to the Tafel diagrams, the CuSb2Ox electrocatalyst exhibits the best catalytic activity, differing significantly (at least 3 times) from the other electrocatalysts. Based on the copper oxide reduction peak in the LSV diagram of the CuSb0.5Ox electrocatalyst, it was not possible to calculate the cathodic Tafel constant for it. The ECSA of the copper antimonate electrocatalysts were estimated based on the electric double layer capacitance (Cdl) (Fig. 3c). Figure 3c shows that CuSb2Ox has the largest specific capacitance. This demonstrates the higher the ECSA, and as result the better the electrochemical NRA process. It is likely the large ECSA is not simply due to the high specific surface area, but rather due to the appropriate arrangement of Sb and Cu within the oxide structure, the active sites on the surface for the adsorption of electroactive ions have increased; especially since, as we know, the atomic sites of transition metals within the antimonate phase can be suitable sites for the ad-desorption of electroactive species40. The corresponding small amplitude cyclic voltammograms in the non-Faradaic domain at various scan rates are shown in Fig. S7. The Nyquist diagrams of the electrochemical impedance spectroscopy are shown in Fig. 3d. The equivalent circuit obtained by fitting the experimental data, which includes two time-constants, is shown in Fig. S8. The parameters obtained from the EIS data fitting are listed in Table 2. Considering the complex mechanism and multiple half-reactions of the NRA process, it is likely that the charge transfer resistance 1 (Rct1) appearing in the Nyquist diagrams is related to reaction (1) and the charge transfer resistance 2 (Rct2) for reaction (6). These two half-reactions are likely to be the slowest half-reactions of the NRA process41,42. Nitrate has a high-energy antibonding orbital, and electron transfer from the electrode to this orbital requires crossing a significant energy barrier. This leads to a very small exchange current from a kinetic perspective which results in, a high charge transfer resistance. Experimental evidence, including a Tafel slope of about 120 mV dec− 1, indicates that this single-electron step acts as the rate-determining step in the nitrate → nitrite branch41. In the reduction of nitrogenous intermediates to ammonia, the first steps of NO reduction are also the most critical and slowest42. The first step is the protonation/reduction of NO and is electronically challenging, because NO is a relatively stable radical with a half-filled orbital and its conversion to HNO requires electronic rearrangement and a change in the N–O bonding state. Hence, reaction (6) represents the bottleneck of the NO → branch41. Due to the nature of these two slow steps, their effect also appears as two distinct loops in the Nyquist plot in EIS. Each of the rate-limiting steps, which involve slow charge transfer, introduces a distinct time constant and a significant charge transfer resistance. Therefore, the first time-constant is usually attributed to reaction (1), which is the slowest step of the nitrate → nitrite branch, and the second time-constant is attributed to reaction (6) (or the sum of the initial steps NO → ). Since these two steps have the highest energy barrier and the smallest exchange current, the charge transfer resistance associated with them is separable in the impedance spectrum and is observed as two distinct loops. Table 2Parameters determined from the EIS data for the synthesized electrocatalysts at a potential of −0.4 V vs. RHE.ElectrocatalystRs (Ω.cm2)Rct1 (Ω.cm2)YCPE(Ω−1.cm− 2.sn) n Rct2 (Ω.cm2)YCPE(Ω−1.cm− 2.sn) n Rtotal (Ω.cm2)CoSb2.5Ox0.12.3 0.7511.9 0.7314.2CoSb2Ox0.141.5 0.822.25 0.653.75CoSb1.5Ox0.11.7 0.784.3 0.716.0CoSbOx0.11.6 0.7516.9 0.6518.5CoSb0.5Ox0.12.4 0.8417.1 0.9819.5 Parameters determined from the EIS data for the synthesized electrocatalysts at a potential of −0.4 V vs. RHE. The CuSb2Ox electrocatalyst has the lowest total resistance (Rtotal) of 3.75 Ω cm2, indicating good electrocatalytic activity for the NRA process. This is in perfect agreement with the LSV results and the Tafel diagrams. Considering that the electrocatalyst is applied as a coating on the titanium sheet and due to the porosity, the solution can access limited areas of the substrate, in accordance with the electrochemical investigations, several points indicate that the LSV and EIS signals are mainly due to the CuSbyOx coating and the contribution of the titanium substrate to the electrochemical response is negligible. First, titanium in the medium used (nitrate-containing solution) exhibits minimal electrochemical activity in the cathodic potential region, especially at potentials below − 1 V vs. RHE (Fig. S5), and in this potential range, cathodic reactions primarily occur on the Cu–Sb oxide phase. This is fully consistent with the LSV results, since all changes and enhancements in the current (including the completely distinct behavior of CuSb2Ox compared to the other two compounds) are directly related to the presence and structure of the coating. Second, the EIS data also show that the charge transfer resistance and extracted time constants are related to the NRA half-reactions (Rct1 and Rct2). The two Nyquist loops obtained are in perfect agreement with the multi-step mechanism of the NRA process, and in case of significant titanium contribution, a different spectral pattern (such as a dominant series resistance or a quasi-capacitive response of the TiO2 structure) would have been expected. However, based on Fig. 3 d and the equivalent circuit in Fig. S8, both time constants are consistent with the electrocatalytic reactions of CuSbyOx and the very low total resistivity of CuSb2Ox (3.75 Ω cm²) indicates the dominance of the active coating on the electrode behavior. Third, the pattern of ECSA changes (Fig. 3c) and double layer capacitance is completely consistent with the change in Sb/Cu ratio, and if the substrate had a significant role in the electrochemical response, such a composition-structure-activity dependence would not be systematically observed. These results indicate that the effective surface area and active sites in the CuSb2Ox coating, not the substrate, are the main factors in the electrochemical response. Performance of CuSbyOx in nitrate reduction to ammonia (NRA) The performance of copper antimonate electrocatalysts in NRA was investigated using chronoamperometry (CA) at a constant potential of −0.4 V vs. RHE (the potential before the sudden current increase in LSV diagrams without nitrate) (Fig. 4a). Experimental data show that the NH3 yield (6280 µg h− 1 cm− 2) and F.E. (87%) of the CuSb2Ox electrocatalyst are significantly higher than those of CuSb1.5Ox (NH3 yield = 5860 µg h− 1 cm− 2, FE = 81%), CuSb2.5Ox (NH3 yield = 5190 µg h− 1 cm− 2, FE = 71%), CuSbOx (NH3 yield = 4905 µg h− 1 cm− 2, FE = 67%) and CuSb0.5Ox (NH3 yield = 1450 µg h− 1 cm− 2, FE = 33%). a) NH3 yield and Fes of CuSbyOx at −0.4 V vs. RHE. b) NH3 yield and Fes of CuSb2Ox at different given potentials. Figure 4b shows the electrochemical nitrate reduction to ammonia for the CuSb2Ox electrocatalyst at different applied potentials, thus highlighting the positive performance of CuSb2Ox in nitrate conversion. Figure 4b shows that the increase in potential correlates well with the increase in nitrate reduction (conversion). At a potential of −0.7 V vs. RHE the ammonia yield increases to 7320 µg h− 1 cm− 2 and the FE increases to 89%. The reason for the FE reduction at a potential of −0.8 V vs. RHE is the relative dominance of the HER over NRA as seen in the LSV curves (Fig. 3a). In addition to the catalytic performance of CuSb2Ox, its electrocatalytic stability was also investigated (Fig. 5). Continuous cycling experiments were conducted at a potential of −0.7 V vs. RHE (the optimum potential from Fig. 4b). During 10 NRA cycles, the ammonia and FE yields for the electrocatalyst fluctuated within a narrow range, indicating that CuSb2Ox possesses favorable electrocatalytic stability. In addition, the changes in the concentrations of and NH3 in the electrolyte over time were investigated (Fig. 6a). At a potential of −0.7 V vs. RHE, after 15 h of reaction, the concentrations of nitrate and ammonia in the solution were 26 mM and 73.5 mM, respectively. These results demonstrate that after 15 h of continuous catalytic reaction in 70 mL of electrolyte, a significant fraction of is efficiently converted to NH3-N. This study also demonstrates the electrocatalytic stability of CuSb2Ox. Figure 6b shows the calculation results of the conversion rate and NH3 selectivity during the 15-hour NRA process. The results demonstrate a continuous increase in the conversion rate during NRA and a continuous increase in NH3 selectivity. The reason for the increase in selectivity over time is likely related to the removal of unstable oxides on the electrode surface and the presence of CuSb2O6 as the main single-phase facilitator of the nitrate reduction reaction. After 15 h of NRA, the NH3 selectivity reached 98.8%, confirming the highly efficient and continuous conversion of to NH3 on the surface of the CuSb2Ox electrocatalyst. According to the FESEM images (Fig. 7), the original morphology of the electrocatalyst is approximately preserved after 15 h of NRA. The morphology of CuSb2Ox remains intact, further demonstrating its stability. This is consistent with the XRD results before and after NRA (Fig. 8). According to Fig. 8, the main phase CuSb2O6 remained stable after chronoamperometry, however, the CuO phase was not observed in the electrocatalyst composition. This was due to the dissolution of copper oxide, for example as form of ions and water molecules, during chronoamperometry. Durability of CuSb2Ox electrocatalyst during 10 cycles at −0.7 potential vs. RHE. FESEM images for the CuSb2Ox electrocatalyst a) before and b) after 15 h of NRA process. XRD patterns for the CuSb2Ox electrocatalyst before and after 15 h of NRA process. According to Figs. 7 and 8 results, the gradual increase in NH₃ selectivity in Fig. 6b can be attributed to the gradual elimination of unstable CuOx phases and the remaining stable and active CuSb₂O₆ phase. This phase plays a major role in the nitrate reduction reaction and its dominant presence over time leads to a continuous increase in NH₃ selectivity. The preservation of the surface morphology and structural stability in FESEM images and the preservation of the CuSb₂O₆ phase in XRD supports this interpretation. Therefore, the results show that the correlation between the improvement of selectivity performance and the structural stability of the catalyst during the reaction was due to the stabilization of the active CuSb₂O₆ phase and the removal of unstable surface species (such as CuO). Table 3 compares the ammonia yield and Faradaic efficiency for nitrate reduction with some recent multicomponent electrocatalysts. It can be seen that the Faradaic efficiency for CuSb2Ox electrocatalysts is higher than that of other electrocatalysts. The ammonia yield of CuSb2Ox electrocatalysts is much higher (several times higher) than that of other electrocatalysts, even though it is synthesized on a plate substrate, as well. This comparison shows that the CuSb2Ox multicomponent electrocatalyst has well enhanced the ammonia production rate with the synergy of copper and antimony. Table 3Comparison of ammonia yield and Faradaic efficiency in nitrate reduction by some newer multicomponent electrocatalysts.electrocatalystPotential(V vs. RHE)NH3 yieldFE (%)ElectrolyteRef.Co-Fe@Fe2O3−0.745880.5 µg h− 1 cm− 2850.1 M Na2SO4 + 500 ppm NO3– 45 Cu0.65Pd0.35Ox−0.21141 µg h− 1 cm− 2740.1 M KNO2 + 0.1 M KNO3 46 ZnCo2O4−0.62100 µg h− 1 cm− 2800.1 M KOH + 0.1 KNO3 32 LaxFeO3−δ−0.81025 µg h− 1 cm− 2780.1 M Na2SO4 + 0.1 M NaNO3 45 Cu-PTCDA0.4521 µg h− 1 cm− 2860.1 M PBS + 500 ppm NO3– 46 10Cu/TiO2 − x−0.751943 µg h− 1 mgcat−1810.5 M Na2SO4 + 200 ppm NO3− 47 B2.6C@TiO2−0.93679 µg h− 1 cm− 2880.1 M Na2SO4 + 0.5 mM Fe2+_EDTA 48 Pd/Co3O4−0.6453400 µg h− 1 cm− 2890.5 M K2SO4 + 200 ppm NO3− 49 Bi2S3/MoS2−0.82550 µg h− 1 cm− 2880.1 M Na2SO4 + 0.1 M NaNO3 50 CuSb2Ox−0.77320 µg h− 1 cm− 2890.1 M NaOH + 0.1 M NaNO3This work Comparison of ammonia yield and Faradaic efficiency in nitrate reduction by some newer multicomponent electrocatalysts. Experimental Synthesis of electrocatalysts Due to good dimensional stability, a grade 2 titanium (Ti-G2) plate was chosen as the substrate. The titanium plate was first sandblasted and then placed in a 10 wt% oxalic acid solution at 100℃ for complete etching. This was intended to create a rough surface and improve coating adhesion51. The starting solution was prepared using the Pechini method52. Ethylene glycol and citric acid were used as solvents. Once the solution reached 60℃, copper chloride and antimony chloride salts were added to the solution. Five solutions were prepared with antimony to copper molar ratios of 0.5, 1, 1.5, 2 and 2.5. The solutions were stirred for 1 h until completely homogenous. The solutions were applied to etched titanium with a brush and then placed in a 200 °C oven to completely evaporate the ethylene glycol solution. To induce the crystallization of copper antimony, a furnace heat treatment (600 °C, 7 min) was applied to the electrodes. Brushing was repeated 10 to 15 times to completely cover the electrode surface with an average loading of 3 mg cm− 2. The coated electrode was left at 600 °C for 2 h to fully calcine. To better understand the synergistic effect of Cu and Sb, samples were also synthesized either with copper oxide only (without antimony) or with antimony oxide only (without copper). Physical characterizations The chemical composition and elemental distribution map were determined using energy dispersive X-ray spectroscopy (EDS). The surface of the electrocatalysts was observed and imaged using a Mira3 TESCAN field emission scanning electron microscope (FESEM). To measure the thickness of the coating, the cross-section of the electrode was examined using FESEM imaging and an EDS-mapping was used to accurately distinguish the coating from the substrate. Then, the layer thickness was measured and analyzed using ImageJ software at several points on the cross-section (according to ASTM B487). In addition, powder X-ray diffraction (PXRD) was performed to determine the electrocatalyst constituent phases. For this purpose, the copper- and antimony-containing solutions were placed in a furnace and the furnace temperature was increased to 600℃ at a heating rate of 1℃/s. Finally, they were held at 600℃ for 2 h to form the target compound powder. The diffraction pattern of the synthesized powder was then obtained using filtered Cu-Kα X-rays (λ = 0.15418 nm) at the angles of 20 to 80 degree. The stoichiometry of the electrocatalysts was determined by dissolving the catalyst powders in hydrochloric acid and measuring the concentration of dissolved ions using inductively coupled plasma mass spectrometry (ICP-MS). Absorption data for determining and NH3 concentrations were obtained using a UV-vis spectrophotometer (HACH DR6000). Electrochemical measurements Electrochemical tests were performed in a custom H-cell (Fig. S2) in 0.1 M sodium hydroxide (NaOH) medium with and without 100 mM sodium nitrate (NaNO3) using a standard three-electrode arrangement using an Autolab® PGSTAT 30. In the standard three-electrode arrangement, a platinum electrode as the auxiliary electrode was separated by a Nafion 115 membrane from a saturated Ag|AgCl electrode as the reference electrode and the synthesized electrocatalyst as the working electrode. The geometric area of ​​the working electrode was 1 cm2. The potential used ​​in this study were reported as potential versus the reversible hydrogen electrode (RHE) using the following equation: Linear sweep voltammetry (LSV) was performed to determine the activity and cathodic Tafel constant (βc) of the electrocatalysts. The LSV was performed from the open circuit potential (OCP) to the potential of −1 vs. RHE at a scan rate of 1 mV s− 1. Cyclic voltammetry in the non-Faradaic current domain (± 0.2 V vs. OCP) was used to determine the double layer capacitance (Cdl) and to estimate the electrochemical active surface area (ECSA). Chronoamperometry was performed to measure the NH3 yield and Faradaic Efficiency (F.E.). Electrochemical impedance spectroscopy (EIS) was performed to measure the charge transfer resistance (Rct) of the electrocatalysts at a constant potential of −0.4 V vs. RHE with a voltage amplitude of 10 mV and a frequency range of 100 kHz to 10 mHz using a ZIVE SP1 potentiostat. Conclusion Low-cost copper antimonate electrocatalysts were synthesized using a sol-gel method for the NRA process. The influence of different molar ratios ​​of antimony to copper on their performance in the NRA process was investigated. The results are as follows: XRD results showed that even in non-stoichiometric ratios of Sb to Cu in precursor solution, the main and, of course, stable phase is CuSb2O6. The XRD after the catalytic conversion reaction of nitrate to ammonium also confirmed that phases other than CuSb2O6 decompose and disappear over time. XRD results showed that even in non-stoichiometric ratios of Sb to Cu in precursor solution, the main and, of course, stable phase is CuSb2O6. The XRD after the catalytic conversion reaction of nitrate to ammonium also confirmed that phases other than CuSb2O6 decompose and disappear over time. LSV curves showed that CuSb2Ox exhibits the best catalytic activity and, according to Tafel diagrams, CuSb2Ox has the lowest cathodic Tafel constant. It means that the CuSb2O6 phase, with a 2:1 ratio of antimony to copper, will present the least barrier to the nitrate reduction reaction. LSV curves showed that CuSb2Ox exhibits the best catalytic activity and, according to Tafel diagrams, CuSb2Ox has the lowest cathodic Tafel constant. It means that the CuSb2O6 phase, with a 2:1 ratio of antimony to copper, will present the least barrier to the nitrate reduction reaction. According to SACV and EIS results, CuSb2Ox exhibits the highest active sites and lowest charge transfer resistance in the NRA process at only Rct = 3.75 Ω cm2. According to SACV and EIS results, CuSb2Ox exhibits the highest active sites and lowest charge transfer resistance in the NRA process at only Rct = 3.75 Ω cm2. Analysis of the CA results showed that the synthesized CuSb2Ox provided a high NH3 yield of 7320 µg h− 1 cm− 2 and a high F.E. of 89% (−0.7 V vs. RHE) in alkaline medium with pH = 13. Analysis of the CA results showed that the synthesized CuSb2Ox provided a high NH3 yield of 7320 µg h− 1 cm− 2 and a high F.E. of 89% (−0.7 V vs. RHE) in alkaline medium with pH = 13. The selectivity of CuSb2Ox was increased during the NRA process, reaching 98.8% after 15 h of electrochemical conversion of NO3− to NH3. The selectivity of CuSb2Ox was increased during the NRA process, reaching 98.8% after 15 h of electrochemical conversion of NO3− to NH3. According to the CA, XRD, and FESEM results, CuSb2Ox is an electrocatalyst with excellent stability for the NRA process. According to the CA, XRD, and FESEM results, CuSb2Ox is an electrocatalyst with excellent stability for the NRA process. This study presents a high-performance and non-precious electrocatalyst for the electrochemical conversion of NO3-N to NH3-N. Future studies should focus on further improving the selectivity of electrocatalysts for electrochemical NH3 production from low-concentration NO3− sources. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements The authors of this article are grateful to the Iran National Science Foundation (INSF) on the grant number 4032066. Author contributions **AliReza Babaei: ** Methodology, Investigation, Data curation, Visualization, Writing – original draft. **Milad Rezaei: ** Supervision, Project Administration, Validation, Writing – review & editing. **Pirooz Marashi: ** Supervision, Validation, Writing – review & editing. **William Mark Rainforth** : Formal Analysis, Writing – review & editing. Data availability Data will be made available on request. If you need to receive data from this study, please send your request via email to the Corresponding Author (Milad Rezaei). All relevant data will be made available to applicants upon reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement Data will be made available on request. If you need to receive data from this study, please send your request via email to the Corresponding Author (Milad Rezaei). All relevant data will be made available to applicants upon reasonable request.