In situ electro-generated Ni(OH)₂ synergistic with Cu cathode to promote direct ammonia oxidation to nitrogen

High-concentration ammonia in landfill leachate, farmland drainage, and industrial wastewater lead to black odor and eutrophication in water bodies, which poses a huge threat to the environment (Mai et al. 2021; Liu et al. 2023a; Zhou & Wang 2023). Compared with traditional processing technologies such as biological treatment (Zulkifli et al. 2022), membrane separation (Li & Jin 2022), breakpoint chlorination (Huang et al. 2022; Zhang et al. 2022), ozonation (Krisbiantoro et al. 2020; Zhang et al. 2021), and photocatalysis (Zhang et al. 2020b), electrochemical ammonia oxidation (EAO) method has advantages of good tolerance to toxic pollutants, small reactor scale and simple operation (Bunce & Bejan 2011; Ding et al. 2013; Zhang et al. 2020a; AlJaberi et al. 2023; Bao et al. 2023). Moreover, in the future, the development of nuclear power plants will greatly reduce the cost of electricity usage.

To address these issues, it is necessary to design an easily obtainable and efficient Ni(OH)₂-based electrocatalyst. In this study, commercial Ni foam was used as an anode to in situ produce active Ni(OH)₂ through the EAO process. The Ni(OH)₂ generated may exhibit different physical and chemical states, and it may show different EAO performance. The EAO performance of Ni(OH)₂ at different current densities, pH and initial concentrations were studied in detail. Characterization of the physical and chemical states of Ni(OH)₂ were tested using X-ray diffractometer (XRD), X-ray photoelectron spectroscopy (XPS), and scanning electron microscope (SEM). This electrochemical system had also been applied to treat goose-raising wastewater containing high concentrations of ammonia.

The EAO experiments were conducted in a single chamber reactor with a reaction solution volume of 100 mL. Foam metal materials (Suzhou Taili Foam Metal Factory (Jiangsu, China), Fe (1.0 × 20 × 30 mm) or Cu (1.6 × 20 × 30 mm) or Ni (1.0 × 20 × 30 mm)) was used as anode for EAO, and the Pt mesh (10 × 10 mm) or Cu foam (1.6 × 10 × 10 mm) was used as cathode. The reaction solution consisted of 11.11 mg/mL Na₂SO₄ and (27.78, 55.56 or 111.12 mg/L). The electrolyte pH was adjusted by a 1.0 M NaOH solution. These EAO experiments were repeated in triplicate, and the average results were reported. All reagents were analytical. Deionized water was used throughout this study. All experiments were conducted at constant ambient temperature (25 ± 0.5 °C).

The electrochemical surface area (ECSA) was evaluated in a solution of 11.11 mg/mL Na₂SO₄ at different scan rates (100, 75, 50, 25, 10, and 5 mV s⁻¹), as reported by the studies (Cerrón-Calle et al. 2023). The relevant calculation results are shown in Supplementary material, Figure S1. The electrochemical active areas of Ni foam (1.0 mm × 20 mm ×30 mm) and Cu foam (1.6 mm × 10 mm ×10 mm) used were 10 and 5 cm², respectively. In the entire text, the calculation of current density uses electrochemical active area instead of two-dimensional area. The surface morphology and elemental distribution of the Ni(OH)₂ generated were characterized by a field emission scanning electron microscope (FE-SEM, Merlin, Zeiss 152 Co., Berlin, Germany) coupled with an energy-dispersive spectrometer (EDS). The crystalline structure was determined by XRD (X'Pert PRO MRD, PANalytical, the Netherlands). The surface chemical composition and heteroatom distribution were analyzed by XPS (ESCALAB250Xi, Thermo Fisher Scientific Waltham, MA). The concentrations of Ni and Cu ions were measured by inductively coupled plasma-mass 110 spectrometry (ICP-MS, Agilent 7700ce, Santa Clara, CA). Cyclic voltammetry curve, chronopotentiometry, and linear sweep voltammetry (LSV) experiments were carried out at an electrochemical workstation (Chenhua, CHI660E, Shanghai, China). A saturated calomel electrode (SCE) and Pt wire (5 mm diameter) were used as the reference and counter electrode, respectively. The concentrations of ⁠, ⁠, and were determined with a UV-visible spectrophotometer (UV-2600, Shimadzu, Japan). The gaseous samples (NOx) were withdrawn from the reactors and subsequently detected using a 7,890A gas chromatograph (Agilent Technologies Inc., CA, USA). Considering that the concentration of nitrogen oxides detected was very low, it could be ignored. The concentrations of total organic carbon (TOC) in actual goose farming wastewater were tested by a TOC instrument (TOC-L organic carbon analyzer, Shimadzu, Japan).

In order to improve the N₂ selectivity of EAO progress, Cu foam with recognized good electrochemical reduction of nitrate performance among commercial electrode materials was selected as the cathode (Li et al. 2021; Zhou et al. 2022; Bi et al. 2023), and the influence of area of Cu foam cathode on EAO performance was studied, as shown in Figure 1(b). When Cu foam was used as a cathode instead of Pt mesh, the removal rate of ammonia remained at nearly 100%, and the N₂ selectivity of the system increased from 71 to 85%. This may be because nitrate, which was the oxidation product of EAO on the anode, had a faster reduction reaction at the Cu foam cathode interface than at Pt mesh interface. This viewpoint had also been validated in experimental results using nitrate as a reactant and Cu foam and Pt mesh as cathodes, respectively (Figure 1(c)) with Ni foam anode. The generated ammonia at the Cu foam cathode could be transferred to the anode interface for oxidation. Therefore, no ammonia was detected in the experiment using nitrate as the reactant. This nitrogen cycle made the EAO reaction more towards the reaction direction of generating N₂.

In addition, when the area of the Cu foam cathode increased without changing the anode current density, the removal rate of ammonia decreased. It may be that the reduction reaction rate of the generated nitrate slowed down in kinetics at the cathode interface, which led to the accumulation of ammonia oxidation products, thus blocking the EAO reaction. On the other hand, it may be due to the decrease in cathode current density leading to a slower hydrogen evolution reaction, resulting in a decrease in the pH value of the reaction solution. The changes in the pH value of the corresponding solution during the electrolysis process are shown in Figure 1(d). Works of literature have shown that pH values in the range of 7–12 are conducive to EAO reaction. As the pH value increased, the concentration of deionized ammonia increased, and more ammonia was present in the EAO reaction (Almomani et al. 2020; Almomani & Saad 2021; Liu et al. 2023b). To sum up, the EAO performance was further evaluated with Ni foam anode and Cu foam cathode.

The control experiment of NH₃ volatilization at different pH values with identical operation conditions without applying current was carried out. When the pH value was high, which was a higher value than NH₃ pKa (9.25), the volatilization of NH₃ had indeed occurred, especially within 5 min after preparing the ammonia nitrogen solution, and the small amount of ammonia volatilization afterwards can be almost ignored. The residual amount of ammonia nitrogen after volatilization is shown in Supplementary material, Figure S2. As the pH value increased, the volatilization of ammonia increased. When the initial pH value of the ammonia solution (55.56 mg/L ⁠) was 13.01, the volatilization of ammonia was 10.5%. Comparing Figure 2(b), it can be seen that the removal of ammonia nitrogen was still mainly due to the EAO process. The changes in pH values during the electrolysis process are shown in Figure 2(c). When initial pH values were 7.45 and 9.40, the pH values increased with increasing time. This was because compared to the oxidation reaction of the anode, the hydrogen evolution reaction of the cathode was more intense with a large cathode current density. When the initial pH values were 11.70, 12.62, and 13.01, the pH of the solution remained almost unchanged with the increase of electrolysis time, although the EAO reaction weakened at this time. Within these pH ranges, anodic oxidation reactions (enhanced oxygen evolution reaction and weakened EAO reaction) and the cathodic reduction reaction (hydrogen evolution reaction) may reach the balance of gain and loss of electrons.

It was worth noting that Ni or Cu ions were not detected in the treated wastewater. The theoretical pH of the solution for Ni²⁺ precipitation of Ni(OH)₂ is 7.1 (Mo et al. 2021; Smoak & Schnoor 2022), while the initial pH of the reaction solution was 7.45. The anode area was higher than the area of the cathode area, which resulted in a cathode current density that was higher than the anode current density. Therefore, the pH of the solution kept rising during the reaction (Figure 1(d)), which led to the formation of Ni(OH)₂ on the Ni foam in situ, and no Ni ion was detected in the solution. On the other hand, when Cu foam was used as cathode material, under the protection of a negative electric field and the alkaline solution, it was difficult for Cu to be dissolved. Therefore, Cu ions were difficult to generate in the solution. In addition, the Pourbaix diagram of Cu and Ni also indicated that metal ions were difficult to precipitate under such reaction conditions.

The XPS analysis was performed to study the chemical nature of Ni(OH)₂ generated. As shown in Supplementary material, Figure S4, XPS peaks of Ni and O were detected in the survey scan. The high-resolution Ni2p spectra are depicted in Figure 4(c), two peaks at 854.3 and 872.0 eV corresponded to Ni²⁺, and the other two peaks at 859.7 and 878.0 eV belonged to the characteristic peak shake-up satellites. The Ni2p binding energies negatively shifted compared with Ni2p in pure Ni(OH)₂ (Song et al. 2021). This made it easier for the electrode surface to adsorb NH₃ instead of OH_ad, which can promote the dimerization of intermediate products (*NH_x) to form N₂ (Hou et al. 2023). As shown in Figure 4(d), the high-resolution O1s spectra of Ni(OH)₂ generated were fitted with two peaks. The characteristic peak at 529.06 eV was attributed to surface lattice oxygen in the metal-oxygen bond. The characteristic peak at 531.8 eV corresponding to oxygen vacancies was remarkable. The oxygen vacancies should be relevant to the hydroxyl groups and adsorbed O− or species (Xue & Feng 2021; Hao et al. 2023). The surface oxygen species alter the electron state of Ni, and this was consistent with the negative shift of binding energy in the XPS results of Ni2p.

Under chlorine free conditions, commercial Ni foam and Cu foam were used as anode and cathode, respectively, for direct EAO. Due to the coupling effect between Ni anode and Cu cathode, nitrogen cycling in the electrochemical system was promoted, thereby improving the ammonia removal rate and N₂ selectivity. When the initial concentration of ammonia nitrogen was 111.12 mg/L, after 150 min of electrolysis with 250 mA of current, the removal rate of ammonia remained at 92.5%, and the N₂ selectivity was 90%. Flaky Ni(OH)₂ with oxygen defects generated during electrolysis was confirmed to be the EAO active site. When this electrochemical system was used to treat goose-raising wastewater containing 422.5 mg/L of ⁠, it had a significant removal effect on ammonia and TOC. After treatment, this electrochemical system achieved good performance with an ammonia removal rate of 87%, N₂ selectivity of 77%, and TOC removal rate of 72%. It is worth noting that the ammonia nitrogen of the treated goose wastewater has reached the discharge standard for breeding wastewater (GB18595-2001) in China. In summary, this electrochemical system with commercial electrodes was simple to construct, and had the potential to be applied to efficiently treat high-concentration ammonia wastewater.

Y.X. contributed to funding, review, project administration and conceptualization; X.W. contributed to methodology and writing; Q.L. contributed to collection of actual wastewater and data analysis; M.F. contributed to sample testing and data analysis; Z.D. contributed to sample testing; W.Z. and J.C. contributed to data curation; N.L. and Z.L. contributed to funding and supervision.

We gratefully acknowledge financial support from the Suzhou University Startup Foundation for Doctor (2021BSK042), Projects of Colleges and Universities in Anhui Province (2022AH040211), the scientific research program of the Anhui Provincial Education Department (KJ2021A1105), the fourth batch of outstanding academic and technical backbones of Suzhou University (2020XJGG06); Excellent Youth Support Program in Colleges and Universities of Anhui Province (gxyq2022104).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.