Efficient co-production of ammonia and formic acid from nitrate and polyester via paired electrolysis

Abstract

Paired electrolysis, which integrates a productive cathodic reaction, such as the nitrate reduction reaction (NO₃⁻RR) with selective oxidation at the anode, offers an intriguing way to maximize both atomic and energy efficiency. However, in a conventional design, the NO₃⁻RR is often coupled with the anodic oxygen evolution reaction, leading to substantial energy consumption while yielding low-value oxygen. Here, we report a hybrid electrolysis system that combines cathodic reduction of nitrate to ammonia and anodic oxidation of polyethylene-terephthalate-derived ethylene glycol (EG) to formic acid (FA), utilizing oxygen-vacancy-rich (O_V) Co₃O₄ and Cu doped Ni(OH)₂ as the cathode and anode, respectively. Remarkably, this paired electrolysis system demonstrates a faradaic efficiency (FE) of 92% for cathodic ammonia production and a FE of 99% for anodic FA production, while reducing the cell voltage by 0.54 V compared to the conventional NO₃⁻RR system at the same current density of 100 mA cm⁻². Experimental investigations combined with theoretical calculations reveal that the O_V introduction effectively addresses the insufficient NO₃⁻ adsorption and hydrogenation on bare Co₃O₄. Additionally, Cu incorporation increases the Ni–O covalency, resulting in an improved EG adsorption ability. This work presents a promising way for waste management via paired electrolysis.

---

New concepts

Electrochemical ammonia (NH₃) production from nitrate-containing wastewater streams presents a sustainable route to concurrent chemical synthesis and waste management, which is thus recognized as an alternative to the conventional, energy-intensive Haber–Bosch process. However, the nitrate (NO₃⁻) reduction reaction (NO₃⁻RR) typically pairs with the oxygen evolution reaction (OER) at the anode that suffers from sluggish kinetics and high energy input while yielding low-value oxygen. To tackle this issue, we implement paired electrolysis that couples the cathodic NO₃⁻RR with anodic oxidation of plastic-derived ethylene glycol. This integrated system enables simultaneous valorization of nitrate wastewater and upcycling of plastic waste within a single electrolyzer, thereby maximizing the utilization of waste resource. This study offers an environmentally benign and energy-efficient pathway for transforming waste streams into valuable chemicals.

Introduction

Ammonia (NH₃) plays a crucial role in the production of fertilizers and is recognized as a critical precursor for the synthesis of a plethora of nitrogen-containing chemicals, including urea and amino acids.^1–3 The industrial-scale NH₃ synthesis is currently being implemented via the Haber–Bosch process, which, however, features energy-intensive nature and high carbon emission.^4–6 The electrochemical nitrate (NO₃⁻) reduction reaction (NO₃⁻RR) driven by renewable electricity offers a cost-effective and low-carbon way for sustainable NH₃ production.^7–10 The NO₃⁻RR for NH₃ production involves a sluggish eight-electron transfer kinetics, which necessitates catalysts endowed with abundant active sites.^11–14 Moreover, the byproducts (e.g., NO₂⁻, N₂, and NH₂OH) evolved in the NO₃⁻RR and the competitive hydrogen evolution reaction (HER) significantly impede the selectivity and current efficiency of the NO₃⁻RR.^15–17 Thus far, developing high-performance electrocatalysts for the NO₃⁻RR remained a huge challenge, primarily due to the limited understanding of the catalytic mechanism and the structure–activity relationship.

Spinel oxides, characterized by their tetrahedral and octahedral coordination centers, such as Co₃O₄, have garnered substantial interest in the NO₃⁻RR, potentially owing to their tunable electronic structure and rich redox chemistry.^18–20 Furthermore, the octahedral Co³⁺ sites with empty d orbitals exhibit a higher affinity for atomic hydrogen adsorption (*H), which thereby serve as hydrogenation centers in converting NO₂⁻ to NH₃.^21,22 In spite of these aforementioned advantages, bare Co₃O₄ encounters several limitations in the NO₃⁻RR, particularly poor NO₃⁻ adsorption capability, which arises from the elevated energy of the lowest unoccupied molecular π* orbital (LUMO π*) of NO₃⁻.^23,24 Defect engineering has been deemed as a powerful tool for optimizing the reaction intermediate adsorption on spinel structures.^25–27 Notably, the incorporation of an oxygen vacancy (O_V) into Co₃O₄ leads to an upshift of the d-band center of Co 3d orbitals. This contributes to charge injection into the π* orbitals of NO₃⁻, thus facilitating promoting NO₃⁻ adsorption and activation.

In this context, creation of an O_V in Co₃O₄ can be expected to play a pivotal role in transforming NO₃⁻ into NO₂⁻.

However, further enhancement of the overall performance of the NO₃⁻RR still remains essential for its widespread application. Notably, a limiting factor increasing the energy input of the conventional system is the anodic oxygen evolution reaction (OER), which requires a high oxidative potential yet produces low-cost oxygen (O₂).^28–30 Electrochemical oxidation of plastic-derived feedstocks, benefitting from the low operation potential and the production of value-added chemicals, represents a promising alternative to the OER.^31–34 Consequently, integrating the NO₃⁻RR and plastic upcycling into one electrolyzer is anticipated to yield valuable chemicals at minimal investment. Nevertheless, the operational stability and economic feasibility of this paired system still face challenges.

Herein, we designed a paired electrochemical system that couples the cathodic NO₃⁻RR for NH₃ production with anodic polyethylene-terephthalate-derived (PET-derived) ethylene glycol (EG) oxidation for formic acid (FA) evolution. Through constructing an oxygen vacancy Co₃O₄ (O_V–Co₃O₄) cathode and a Cu doped Ni(OH)₂ (Cu–Ni(OH)₂) anode, we achieved a FE of NH₃ (92%) at the cathode and a FE of FA (99%) at the anode at 100 mA cm⁻². Based on experimental observations and theoretical insights into the improved ability for NO₃⁻ adsorption and activation on O_V–Co₃O₄, as well as the enhanced EG binding on Cu–Ni(OH)₂, we propose a continuous co-production of NH₄Cl and formate from liter-scale nitrate wastewater and hundred-gram-scale PET bottles. Techno-economic analysis (TEA) suggests that this paired electrolysis system remains profitable, thus attracting industrial investment.

Results and discussion

Cathode catalyst characterization studies

The preparation procedure of the O_V–Co₃O₄ cathode initiates with the growth of Co(OH)F nanorods on Ni foam (Co(OH)F/NF) through a hydrothermal method (Fig. 1a). Scanning electron microscopy (SEM) and X-ray diffraction (XRD) patterns clearly show the uniform growth of Co(OH)F nanorods throughout the entire NF substrate (Fig. S1 and S2, ESI†). The following calcination treatment in air successfully achieves the conversion of Co(OH)F/NF into Co₃O₄/NF, as evidenced by the SEM and transmission electron microscopy (TEM) observations (Fig. S3 and S4, ESI†). The O_V is subsequently introduced into Co₃O₄via reduction treatment in a H₂ atmosphere for 1 h. Unless otherwise specified, the O_V–Co₃O₄/NF catalyst refers to the sample subjected to H₂ reduction for 1 h. The crystal structure of O_V–Co₃O₄ is elucidated by using XRD analysis, which reveals all the peaks observed corresponding to the standard spinel structure (Fig. 1b). Furthermore, SEM and TEM observations collectively exhibit the retention of the nanorod morphology in O_V–Co₃O₄ (Fig. 1c and d). Selected area electron diffraction (SAED) patterns (Fig. 1d, inset) exhibit the exposure of the characteristic (220), (311), (400), and (511) planes of Co₃O₄. High-resolution TEM reveals a set of lattice fringes with a spacing of 0.24 nm (Fig. 1e), which is assigned to the (311) plane of Co₃O₄.³⁵ High-angle annular dark field scanning TEM (HAADF-STEM) imaging and electron energy loss spectroscopy (EELS) demonstrate the uniform distribution of Co and O elements throughout the O_V–Co₃O₄ nanorod (Fig. 1f). The O_V was identified by electron spin resonance (ESR) spectra (Fig. 1g), in which a weak ESR signal at g = 2.002 is observed for Co₃O₄. In contrast, O_V–Co₃O₄ exhibits a significantly enhanced O_V signal, suggesting the successful introduction of O_V. The O_V introduction inevitably leads to a change in the valence state of Co atoms adjacent with O_V. As illustrated in Co 2p X-ray photoelectron spectroscopy (XPS) (Fig. 1h), the proportion of Co²⁺ is substantially improved from 39% to 87% after O_V introduction, which is attributed to the transformation of octahedral Co³⁺ to Co²⁺ upon O_V incorporation.³⁶ Furthermore, the local electrons at the O_V sites have the potential to reduce the adsorbed water molecule to hydrogen and hydroxyl.^37,38 This phenomenon accounts for the higher content of surface hydroxyls on O_V–Co₃O₄/NF (79%) compared with bare Co₃O₄/NF (22%), as observed in the O 1s XPS spectra (Fig. 1i).

Fig. 1 Morphology and structure characterization of the cathode catalyst. (a) Schematic illustration of the O_V–Co₃O₄/NF preparation process. (b) XRD patterns of Co₃O₄ and O_V–Co₃O₄ powders scraped from the Ni substrate. (c) SEM image, (d) TEM image (the inset shows the corresponding SAED patterns), (e) HRTEM image, (f) HAADF-STEM image and its corresponding EELS maps of O_V–Co₃O₄. (g) ESR spectra of O_V–Co₃O₄ and Co₃O₄ powders scraped from the NF substrate. (h) Co 2p and (i) O 1s XPS spectra of Co₃O₄/NF and O_V–Co₃O₄/NF.

NO₃⁻RR activity and mechanism analysis

The electrochemical NO₃⁻RR activity of O_V–Co₃O₄/NF was measured in a three-electrode H-type cell under ambient conditions. Colorimetric methods, utilizing a calibration curve (Fig. S5, ESI†), were employed to quantitatively determine ammonia concentrations. All potentials are referenced to the reversible hydrogen electrode (RHE), unless otherwise specified. A slight enhancement in the HER activity can be observed for O_V–Co₃O₄/NF relative to Co₃O₄/NF in the absence of NO₃⁻ (Fig. 2a). Upon NO₃⁻ addition, O_V–Co₃O₄/NF enables a dramatically increased current density compared with Co₃O₄/NF, highlighting the key role of O_V in the NO₃⁻RR. To evaluate the NO₃⁻RR performance more fairly, specific activities were calculated by normalizing linear sweep voltammetry (LSV) curves with the electrochemical active surface area (ECSA) (Fig. S6, ESI†). The results suggest that the O_V–Co₃O₄/NF still allows better NO₃⁻RR activity than Co₃O₄/NF. Additionally, O_V–Co₃O₄/NF exhibits more favorable reaction kinetics and enhanced charge transfer than Co₃O₄/NF, as corroborated by the reduced Tafel slope and charge transfer resistance (Fig. S7 and S8, ESI†).

Fig. 2 NO₃⁻RR activity and enhanced mechanism. (a) Polarization curves of Co₃O₄/NF and O_V–Co₃O₄/NF in 1 M KOH solution with (solid line) and without (dotted line) 0.1 M NO₃⁻. (b) The NO₃⁻RR activity comparison between Co₃O₄/NF and O_V–Co₃O₄/NF. (c) ¹H NMR spectra of post-reaction electrolytes using ¹⁵NO₃⁻ and ¹⁴NO₃⁻ feedstocks. (d) Cycling experiments of O_V–Co₃O₄/NF for the NO₃⁻RR conducted at −0.3 V. (e) PDOS of the (311) planes of Co₃O₄ and O_V–Co₃O₄. (f) The band center calculations of O 2p and Co 3d in Co₃O₄ (311) and O_V–Co₃O₄ (311). (g) The Gibbs free energy for H adsorption over Co₃O₄ (311) and O_V–Co₃O₄ (311). (h) In situ DEMS analysis of nitrogen-containing intermediates during the NO₃⁻RR over O_V–Co₃O₄/NF. (i) In situ FTIR spectra of the NO₃⁻RR over O_V–Co₃O₄/NF recorded under varying potentials. (j) Free energy diagram of each intermediate state within the NO₃⁻RR on Co₃O₄ (311) and O_V–Co₃O₄ (311).

The FE and productivity of NH₃ over O_V–Co₃O₄/NF and Co₃O₄/NF were measured to reveal the role of O_V. As presented in Fig. 2b, O_V–Co₃O₄/NF exhibits a higher FE of NH₃ compared with Co₃O₄/NF across a wide potential range from −0.1 to −0.45 V. Impressively, O_V–Co₃O₄/NF achieves a maximum FE of 95.2% at −0.3 V, which outperforms most of recently reported NO₃⁻RR catalysts (Table S1, ESI†). Additionally, O_V–Co₃O₄/NF displays an NH₃ productivity of 0.31 mmol cm⁻² h⁻¹ at −0.3 V, notably higher than that of bare Co₃O₄/NF (0.17 mmol cm⁻² h⁻¹). This enhanced performance is attributed to the promotion of NO₃⁻ adsorption and activation by O_V, thus accelerating the NO₃⁻-to-NO₂⁻ conversion step and improving the NH₃ yield rate.

To identify the impact of the O_V concentration on the NO₃⁻RR, O_V–Co₃O₄/NF with varying O_V contents were synthesized by adjusting the reductive treatment time in a H₂ atmosphere (Fig. S9 and S10, ESI†). The results indicate that the O_V–Co₃O₄/NF catalyst treated with H₂ reduction for 1 h exhibits the highest FE and productivity of NH₃. To exclude the environmental contaminants in NH₃ detection, isotope-labeling tests were carried out (Fig. 2c). The results confirm that the produced NH₃ originates from the NO₃⁻RR rather than the environmental contaminants. The electrochemical NO₃⁻RR durability of O_V–Co₃O₄/NF was assessed in ten consecutive cycles (Fig. 2d). The productivity and FE of NH₃ exhibited slight fluctuations in each cycle but remained stable. Strikingly, no obvious changes in morphology, crystal structure, and chemical state can be observed for the recovered O_V–Co₃O₄/NF, as confirmed by a combination of SEM, TEM, XRD, and XPS characterization studies (Fig. S11–S13, ESI†).

The role of O_V in the NO₃⁻RR was further investigated via density functional theory (DFT) calculations (Fig. S14, ESI†). Note that the distance between the Co 3d and O 2p band centers (denoted as D_M) is recognized as a descriptor of the degree of Co–O covalency, with a smaller D_M indicating an increased Co–O covalency.^39,40 The partial density of state (PDOS) combined with the band center calcination reveals that O_V–Co₃O₄ enables a reduced D_M value (0.17 eV) compared with bare Co₃O₄ (0.19 eV), suggesting a higher overlap between Co 2p and O 1s in O_V–Co₃O₄ (Fig. 2e and f). Such improved Co–O covalency contributes to the NO₃⁻ adsorption and activation.⁴¹ Moreover, the O_V sites promote the generation of active adsorbed H (H*), as indicated by the free energy for H* evolution being closer to zero on O_V–Co₃O₄ compared to Co₃O₄ (Fig. 2g). The enhanced activity for H* generation on O_V–Co₃O₄ contributes to the hydrogenation of NO₂⁻ to NH₃.⁴²

The reaction pathway was identified using in situ differential electrochemical mass spectrometry (DEMS). There exist two potential pathways for the conversion of NO₃⁻ to NH₃, as illustrated in Fig. S15 (ESI†). The appearance of *NO signal, coupled with the absence of *NH₂OH signal, implies that NH₃ is most likely formed via the *NOH intermediate pathway (Fig. 2h). In situ Fourier transform infrared (FTIR) spectroscopy was employed to differentiate the reaction intermediates (Fig. 2i). The peaks at 1338 and 1665 cm⁻¹ are ascribed to *NO₂⁻ and *NO intermediate species, respectively.⁴³ Note that the absence of the characteristic NH₂OH band in FTIR is consistent with the results of in situ DEMS. Based on the reaction pathway confirmed by in situ spectroscopy, we calculated the Gibbs free energies of reaction intermediates on O_V–Co₃O₄ (311) and Co₃O₄ (311). As displayed in Fig. 2j, the energy barrier for NO₃⁻ adsorption on bare Co₃O₄ reaches up to 1.28 eV, which is recognized as the rate-determining step (RDS). The introduction of O_V results in a significantly reduced energy barrier for NO₃⁻ adsorption, revealing the pivotal role of O_V in the NO₃⁻-to-NO₂⁻ step. Furthermore, O_V–Co₃O₄ enables improved hydrogenation ability in the NO₂⁻-to-NH₃ step compared with Co₃O₄, as uncovered by the reduced energy barrier for hydrogenation of *NO to *NOH. This observation agrees with the results of the calculated energy for *H adsorption (Fig. 2g).

Anode catalyst characterization studies

The Cu doped Ni(OH)₂ (Cu–Ni(OH)₂) anode catalyst was synthesized by using a two-step approach (Fig. 3a), involving hydrothermal growth of Cu–Ni(OH)₂·xH₂O nanosheets on NF (Fig. S16, ESI†), followed by annealing to obtain Cu–Ni(OH)₂/NF. The crystal structure of the as-prepared Cu–Ni(OH)₂ is indexed into the β-Ni(OH)₂ phase (PDF#73-1520), as elucidated in XRD patterns (Fig. 3b). The molar ratio of Cu to Ni was determined to be 1 : 3 using inductively coupled plasma mass spectrometry (ICP-MS). Additionally, SEM and TEM images of Cu–Ni(OH)₂/NF indicate that Cu–Ni(OH)₂ exists as a nanosheet structure on the NF substrate (Fig. 3c and d), with no significant difference compared with bare Ni(OH)₂/NF (Fig. S17, ESI†). The lattice fringe of a spacing of 0.27 nm, observed in the HRTEM image (Fig. 3e), is attributed to the (100) plane of Ni(OH)₂.⁴⁴ The elemental mapping demonstrates that Cu, Ni, and O elements are uniformly dispersed on the nanosheet (Fig. 3f). The chemical states of Cu–Ni(OH)₂/NF were investigated via XPS. The Cu 2p XPS spectrum reveals that the Cu dopant is present in Cu²⁺ (Fig. S18, ESI†). In Ni 2p XPS spectra, a positive shift of 0.3 eV in the binding energy of Ni 2p can be observed for Cu–Ni(OH)₂/NF relative to Ni(OH)₂/NF (Fig. 3g), indicative of decreased charge density on the Ni site after Cu incorporation. This change in charge density results in a reduced D_M value in Cu–Ni(OH)₂, suggesting a greater Ni–O covalency and an enhanced ability for oxygen-containing adsorption (Fig. 3h and i).

Fig. 3 Morphology and structure characterization of the anode catalyst. (a) Schematic illustration of Cu–Ni(OH)₂/NF preparation. (b) XRD patterns of Ni(OH)₂ and Cu–Ni(OH)₂ powders scraped from the NF substrate. (c) SEM image, (d) TEM image (the inset shows the corresponding SAED patterns), (e) HRTEM image, (f) HAADF-STEM image and its corresponding EELS maps of Cu–Ni(OH)₂. (g) Ni 2p XPS spectra of Ni(OH)₂/NF and Cu–Ni(OH)₂/NF. (h) PDOS of Ni(OH)₂ (001) and Cu–Ni(OH)₂/NF (001). (i) Band centers of Ni(OH)₂ (001) and Cu–Ni(OH)₂/NF (001).

EGOR activity and mechanism analysis

The EGOR performances of Cu–Ni(OH)₂/NF and Ni(OH)₂/NF were assessed in 1 M KOH containing 0.3 M EG at room temperature under atmospheric pressure. The reason for fixing the EG concentration at 0.3 M is elucidated in Fig. S19 (ESI†). The LSV curves show that Cu–Ni(OH)₂/NF requires a potential of 1.39 V to achieve a current density of 100 mA cm⁻², which is 60 mV lower than that of bare Ni(OH)₂/NF, indicating the enhanced EGOR activity due to Cu doping (Fig. 4a). Additionally, Cu–Ni(OH)₂/NF exhibits a higher specific EGOR activity (normalized by ECSA) than Ni(OH)₂/NF (Fig. S20, ESI†). Furthermore, the current density observed in LSV curves dramatically increases for EGOR on both Ni(OH)₂/NF and Cu–Ni(OH)₂/NF compared with the OER, uncovering more thermodynamically favorable conditions for EGOR over the OER. To identify the impact of Cu doping concentration on EGOR activity, the Cu–Ni(OH)₂/NF catalysts with different Cu doping levels were synthesized (Fig. S21, ESI†). The results demonstrate that the Cu–Ni(OH)₂/NF catalyst with the Cu/Ni molar ratio of 1 : 3 delivers optimal EGOR activity (Fig. S22, ESI†). Unless otherwise noted, the Cu–Ni(OH)₂/NF catalyst in this work refers to the sample with a Cu/Ni molar ratio of 1 : 3. In addition to the high EGOR activity observed from LSV curves, Cu–Ni(OH)₂/NF allows faster reaction kinetics and reduced charge transfer resistance with respect to bare Ni(OH)₂/NF, as evidenced by the Tafel slopes (Fig. 4b) and EIS spectra (Fig. S23, ESI†).

Fig. 4 EGOR activity and enhanced mechanism. (a) Polarization curves of Ni(OH)₂/NF and Cu–Ni(OH)₂/NF measured in 1 M KOH with and without 0.3 M EG. (b) Tafel slopes of Ni(OH)₂/NF and Cu–Ni(OH)₂/NF. (c) The FE and productivity of FA on Ni(OH)₂/NF and Cu–Ni(OH)₂/NF under various potentials. (d) Concentrations of FA, GA, EG, and FE of FA as a function of consumed charges. (e) Stability tests using chronopotentiometry measurements at the current density of 100 mA cm⁻², and the electrolyte was refreshed every 5 h. (f) and (g) The computed pCOHP diagrams of Ni–O_ads bond on Cu–Ni(OH)₂ (f) and Ni(OH)₂ (g). (h) The calculated energy for EG adsorption on Cu–Ni(OH)₂ (001) and Ni(OH)₂ (001). (i) and (j) Bode phase plots of the operando EIS on Cu–Ni(OH)₂/NF in 1 M KOH (i) and 1 M KOH + 0.3 EG (j). (k) and (l) In situ Raman tests on Cu–Ni(OH)₂ under varying potentials in 1 M KOH (k) and 1 M KOH + 0.3 EG (l).

The EGOR products during 2-h potentiostatic tests on Cu–Ni(OH)₂/NF and Ni(OH)₂/NF were identified using HPLC (Fig. S24, ESI†). Compared with Ni(OH)₂/NF, Cu–Ni(OH)₂/NF exhibits a higher FE of FA (FE_FA) across a wide potential range of 1.4 to 1.6 V (Fig. 4c). When the potential is fixed at 1.5 V, Cu–Ni(OH)₂/NF achieves a maximum FE_FA of 98%, along with a FA productivity of 5.92 mmol cm⁻² h⁻¹, surpassing the performance of most recently reported EGOR catalysts (Table S2, ESI†). Intriguingly, when a total electric charge of 2000 C is transferred through the entire cell, Cu–Ni(OH)₂/NF achieves an EG conversion of 97% and a FA yield of 89%, with the production of a small amount of glycolic acid (GA) byproduct (Fig. 4d and Fig. S25, ESI†), indicating the highly selective conversion of EG to FA. The long-term stability of Cu–Ni(OH)₂/NF for EGOR was performed at a current density of 100 mA cm⁻² (Fig. 4e). Impressively, the FE_FA value is consistently maintained at around 90% throughout the 100-h operation period. The recovered Cu–Ni(OH)₂/NF catalyst was characterized by using various techniques, including SEM, TEM, XRD, and XPS (Fig. S26–S28, ESI†). These results show that the used catalyst retains the original morphology and structure, indicating the stability of Cu–Ni(OH)₂/NF.

To understand the underlying mechanism responsible for the enhanced EGOR performance on Cu–Ni(OH)₂/NF, the projected crystal orbital Hamilton populations (pCOHP) were performed (Fig. S29, ESI†), which is used to elucidate the Ni–O_ads bonding state (O_ads is from the adsorbed EG species).⁴⁵ It is well-accepted that a negative value of integrated pCOHP (IpCOHP) indicates a stable Ni–O_ads bonding state.⁴¹ As depicted in Fig. 4f and g, the IpCOHP value relative to the Fermi level is calculated to be −0.69 for Ni–O_ads on the Cu–Ni(OH)₂ (001) surface, lower than that (−0.60) for Ni–O_ads on the Ni(OH)₂ (001) surface, suggesting an enhanced bonding interaction between Ni and EG upon Cu doping. Furthermore, the EG adsorption energy calculations confirm a more thermodynamically favorable EG adsorption on Cu–Ni(OH)₂ (Fig. 4h). To experimentally verify the enhanced EG adsorption on Cu–Ni(OH)₂, open-circuit potential (OCP) measurements were performed. As shown in Fig. S30 (ESI†), the OCP change over Cu–Ni(OH)₂ is determined to be 0.32 V, significantly higher than bare Ni(OH)₂ (0.06 V), indicating the favorable EG adsorption on Cu–Ni(OH)₂.⁴⁶ The interfacial behaviors were investigated using Operando EIS at various potentials. When the EG feedstock is added, the Nyquist semicircle in the low-frequency region at 1.4 V becomes apparent (Fig. 4i and j), indicating the appearance of EGOR.⁴⁷ This observation agrees with the oxidation potential observed from the LSV curves (Fig. 4a). Upon further increasing the potentials, the EGOR peak gradually shifts to lower phase degrees. These results suggest that EGOR is thermodynamically more favorable than OER in the presence of EG.⁴⁸ The EGOR mechanism over Cu–Ni(OH)₂ was investigated using in situ Raman measurements. Under OER conditions, a pair of characteristic bands at 470 and 557 cm⁻¹, corresponding to NiOOH,⁴⁹ are observed at 1.4 V (Fig. 4k). The band intensities of NiOOH increase with increasing potential, indicating surface reconstruction on Cu–Ni(OH)₂. Upon EG addition, the NiOOH signals completely disappeared and the catalyst reverts to its initial Cu–Ni(OH)₂ (Fig. 4l), indicating that EGOR proceeds via an indirect oxidation way on Cu–Ni(OH)₂.⁵⁰

Co-production of NH₃ and FA

Encouraged by the high activities for half-reaction NO₃⁻RR and EGOR, we attempted to integrate the NO₃⁻RR and EGOR in a two-electrode membrane–electrode assembly (MEA) system separated by a proton exchange membrane (PEM) (Fig. 5a). The as-prepared Cu–Ni(OH)₂/NF and O_V–Co₃O₄/NF served as the anode and cathode, respectively. The LSV curves show that, to obtain a current density of 100 mA cm⁻², the paired NO₃⁻RR//EGOR system only requires a cell voltage of 1.66 V, markedly lower than that (2.20 V) for the conventional NO₃⁻RR//OER system (Fig. 5b). This disparity suggests that the replacement of the OER by EGOR contributes to reducing the overall energy input. To assess the coupled performance in the two-electrode system, the FEs of FA and NH₃ were measured under different current densities during 2-h electrolysis. It can be observed that Cu–Ni(OH)₂/NF enables stable and high FE for FA production (>95%) across a wide current density range from 25 to 100 mA cm⁻². In contrast, NO₃⁻RR activity exhibits a volcano-type trend, with a current density of 100 mA cm⁻² being recognized as optimal for NH₃ production. At higher current densities (>100 mA cm⁻²), the observed reduction in the FE of NH₃ is attributed to the competitive hydrogen evolution reaction.

Fig. 5 Integration of the NO₃⁻RR and EGOR. (a) Schematic diagram showing the two-electrode MEA reactor for NO₃⁻RR coupled with EGOR. (b) Comparison of the LSV curves for NO₃⁻RR//OER and NO₃⁻RR//EGOR system. (c) The FEs of NH₃ and FA measured under different current densities. (d) Schematic illustration of formate and NH₄Cl production from PET bottle and nitrate wastewater. (e) and (f) XRD patterns of synthesized formate (e) and NH₄Cl (f). (g) TEA of NO₃⁻RR//OER and NO₃⁻RR//EGOR system.

The practical application of this paired system was subsequently carried out using real-world PET bottles and simulated nitrate wastewater as the initial feedstock (Fig. 5d and Fig. S31, ESI†). Prior to co-electrolysis, 100 g of post-consumer plastic bottles were collected and hydrolyzed into their constituent monomers using KOH (Fig. S32, ESI†). The resulting hydrolysate was directly fed into the anodic chamber. As for the cathodic feedstock, a simulated high-concentration nitrate solution was prepared via electrodialysis, following established protocols from previous literature.¹⁵ Of note, KOH was added as electrolyte prior to electrochemical valorization. After 210-h electrolysis (Fig. S33, ESI†), FA was identified as the primary product in the anolyte, evidenced by ¹H nuclear magnetic resonance (NMR) (Fig. S34, ESI†). Following separation and purification, 76.67 g of formate were obtained (FA exists as formate under alkaline conditions), with an overall yield of 83%. In parallel, NH₃ was generated at the cathode with an 80% yield, resulting in 13.76 g of NH₄Cl crystals after acidification and evaporation. The high purity of formate and NH₄Cl was confirmed via XRD patterns, FTIR spectra, and ¹H NMR (Fig. 5e and f and Fig. S35 and S36, ESI†). To assess the economic viability of our constructed NO₃⁻RR//EGOR system, a TEA was conducted (Fig. 5g and Note S1, ESI†). The TEA results indicate that our coupled system achieves a higher gross profit (79665.9 $ d⁻¹), corresponding to a 15-fold increase than the conventional NO₃⁻RR//OER system (5173.4 $ d⁻¹), which indicates the potential market competitiveness of our built electrosynthesis system.

Conclusions

In this work, we constructed a paired electrolysis system with low energy input comprising cathodic reduction of NO₃⁻ to NH₃ and anodic oxidation of PET-derived EG to FA, which is made possible by the discovery of the O_V–Co₃O₄ cathode and Cu–Ni(OH)₂ anode. The O_V incorporation compensates for the deficiencies of bare Co₃O₄ in terms of NO₃⁻ adsorption and activation. Cu doping enlarges the Ni–O covalency, thus resulting in an enhanced capability for EG adsorption. Under optimal conditions, this paired electrochemical system achieves 13.76 g of high-purity NH₄Cl crystals with an 80% yield from simulated waster and 76.67 g of formate with an 83% yield from post-consumer PET bottles. Overall, this work provides a cost-effective and low-carbon route to mitigate wastes while producing valuable chemicals.

Author contributions

B. Qiu and Y. Zhang supervised the project and conceived the idea. B. Qiu, N. Zhang, M. Du, and M. Han designed all experiments and co-wrote the manuscript. M. Du, T. Sun, and X. Guo performed the materials synthesis and characterization studies. M. Du, W. Chen, M. Han, and J. Ma carried out electrocatalytic measurements. J. Shang performed the DFT simulations. V. Nicolosi and C. Zhou provided suggestions for the characterization studies. All authors provided comments on the manuscript.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (22302095, 22378208, and 22305243), the Fundamental Research Funds for the Central Universities (KJYQ2025012), the Shenzhen Science and Technology Planning Project (JCYJ20220531100613030), and the Natural Science Foundation of Jiangsu Province (BK20210382 and BK20230272). V. N. and X. G. wish to thank the support of the SFI-funded AMBER Research Centre and the SFI Frontiers for the Future award (Grant No. 12/RC/2278_P2 and 20/FFP-A/8950 respectively). Furthermore, V. N. and X. G. wish to thank the Advanced Microscopy Laboratory in CRANN for the provision of their facilities.

References

1. G.-F. Han, F. Li, Z.-W. Chen, C. Coppex, S.-J. Kim, H.-J. Noh, Z. Fu, Y. Lu, C. V. Singh, S. Siahrostami, Q. Jiang and J.-B. Baek, Nat. Nanotechnol., 2021, 16, 325–330.
2. S. Li, Y. Zhou, X. Fu, J. B. Pedersen, M. Saccoccio, S. Z. Andersen, K. Enemark-Rasmussen, P. J. Kempen, C. D. Damsgaard, A. Xu, R. Sažinas, J. B. V. Mygind, N. H. Deissler, J. Kibsgaard, P. C. K. Vesborg, J. K. Nørskov and I. Chorkendorff, Nature, 2024, 629, 92–97.
3. S. Li, X. Fu, J. K. Nørskov and I. Chorkendorff, Nat. Energy, 2024, 9, 1344–1349.
4. M. Wang, M. A. Khan, I. Mohsin, J. Wicks, A. H. Ip, K. Z. Sumon, C.-T. Dinh, E. H. Sargent, I. D. Gates and M. G. Kibria, Energy Environ. Sci., 2021, 14, 2535–2548.
5. D. Ye and S. C. E. Tsang, Nat. Synth., 2023, 2, 612–623.
6. K. Chu, Y. Luo, P. Shen, X. Li, Q. Li and Y. Guo, Adv. Energy Mater., 2022, 12, 2103022.
7. S. Han, H. Li, T. Li, F. Chen, R. Yang, Y. Yu and B. Zhang, Nat. Catal., 2023, 6, 402–414.
8. J. Hao, T. Wang, R. Yu, J. Cai, G. Gao, Z. Zhuang, Q. Kang, S. Lu, Z. Liu and J. Wu, Nat. Commun., 2024, 15, 9020.
9. K. Kim, A. Zagalskaya, J. L. Ng, J. Hong, V. Alexandrov, T. A. Pham and X. Su, Nat. Commun., 2023, 14, 823.
10. M. Du and B. Qiu, Nat. Water, 2024, 2, 1053–1054.
11. P. H. van Langevelde, I. Katsounaros and M. T. Koper, Joule, 2021, 5, 290–294.
12. M. Duca and M. T. Koper, Energy Environ. Sci., 2012, 5, 9726–9742.
13. G. Qing, R. Ghazfar, S. T. Jackowski, F. Habibzadeh, M. M. Ashtiani, C.-P. Chen, M. R. Smith, III and T. W. Hamann, Chem. Rev., 2020, 120, 5437–5516.
14. M. Li, X. Liu, L. Wang, F. Hou, S. X. Dou and J. Liang, EcoMat, 2021, 3, e12096.
15. F.-Y. Chen, A. Elgazzar, S. Pecaut, C. Qiu, Y. Feng, S. Ashokkumar, Z. Yu, C. Sellers, S. Hao, P. Zhu and H. Wang, Nat. Catal., 2024, 7, 1032–1043.
16. Y. Liu, Z. Zhuang, Y. Liu, N. Liu, Y. Li, Y. Cheng, J. Yu, R. Yu, D. Wang and H. Li, Angew. Chem., Int. Ed., 2024, 63, e202411396.
17. C. Li, H. Li, B. Zhang, H. Li, Y. Wang, X. Wang, P. Das, Y. Li, X. Wu, Y. Li, Y. Cui, J. Xiao and Z.-S. Wu, Angew. Chem., Int. Ed., 2024, 63, e202411542.
18. Q. Hu, S. Qi, Q. Huo, Y. Zhao, J. Sun, X. Chen, M. Lv, W. Zhou, C. Feng, X. Chai, H. Yang and C. He, J. Am. Chem. Soc., 2024, 146, 2967–2976.
19. S. Qi, Z. Lei, Q. Huo, J. Zhao, T. Huang, N. Meng, J. Liao, J. Yi, C. Shang, X. Zhang, H. Yang, Q. Hu and C. He, Adv. Mater., 2024, 36, 2403958.
20. J. Yan, P. Liu, J. Li, H. Huang and W. Song, Small, 2024, 20, 2308617.
21. Z. Fan, C. Cao, X. Yang, W. Yuan, F. Qin, Y. Hu, X. Sun, G. Liu, Y. Tian and L. Xu, Angew. Chem., Int. Ed., 2024, 63, e202410356.
22. J.-Y. Fang, Q.-Z. Zheng, Y.-Y. Lou, K.-M. Zhao, S.-N. Hu, G. Li, O. Akdim, X.-Y. Huang and S.-G. Sun, Nat. Commun., 2022, 13, 7899.
23. K. Zhang, P. Sun, Y. Huang, M. Tang, X. Zou, Z. Pan, X. Huo, J. Wu, C. Lin, Z. Sun, Y. Wan, X. Zhang and L. An, Adv. Funct. Mater., 2024, 34, 2405179.
24. M. Zhang, Z. Ma, S. Zhou, C. Han, V. Kundi, P. V. Kumar, L. Thomsen, B. Johannessen, L. Peng, Y. Shan, C. Tsounis, Y. Yang, J. Pan and R. Amal, ACS Catal., 2024, 14, 11231–11242.
25. S. Kang, X. Guo, D. Xing, W. Yuan, J. Shang, V. Nicolosi, N. Zhang and B. Qiu, Small, 2024, 20, 2406068.
26. K. Chu, W. Zong, G. Xue, H. Guo, J. Qin, H. Zhu, N. Zhang, Z. Tian, H. Dong, Y.-E. Miao, M. B. J. Roeffaers, J. Hofkens, F. Lai and T. Liu, J. Am. Chem. Soc., 2023, 145, 21387–21396.
27. Y. Li, J. Ma, Z. Wu and Z. Wang, Environ. Sci. Technol., 2022, 56, 8673–8681.
28. J. Li, H. Li, K. Fan, J. Y. Lee, W. Xie and M. Shao, Chem Catal., 2023, 3, 100638.
29. Y. Yang, J. Liu, M. Ahmad, X.-Q. Sun, C. Liu, S. Chen, J. Zhang, J.-L. Luo and X.-Z. Fu, Chem. Eng. J., 2025, 505, 159653.
30. S.-Q. Liu, S. Xie, S. Wu, Y. Yang, P.-X. Lei, S. Luo, R. Feng, X.-Z. Fu and J.-L. Luo, Adv. Funct. Mater., 2024, 34, 2404105.
31. M. Du, R. Xue, W. Yuan, Y. Cheng, Z. Cui, W. Dong and B. Qiu, Nano Lett., 2024, 24, 9768–9775.
32. T. Ren, Z. Duan, H. Wang, H. Yu, K. Deng, Z. Wang, H. Wang, L. Wang and Y. Xu, ACS Catal., 2023, 13, 10394–10404.
33. M. Du, M. Xing, S. Kang, Y. Ma, B. Qiu and Y. Chai, EcoMat, 2022, 4, e12259.
34. Y. Ma, X. Guo, M. Du, S. Kang, W. Dong, V. Nicolosi, Z. Cui, Y. Zhang and B. Qiu, Green Chem., 2024, 26, 3995–4004.
35. Y. Ma, Y. Zhang, M. Xing, S. Kang, M. Du, B. Qiu and Y. Chai, Chem. Commun., 2022, 58, 6642–6645.
36. Z. Xiao, Y.-C. Huang, C.-L. Dong, C. Xie, Z. Liu, S. Du, W. Chen, D. Yan, L. Tao and Z. Shu, J. Am. Chem. Soc., 2020, 142, 12087–12095.
37. H. Idriss, Surf. Sci., 2021, 712, 121894.
38. T. J. Frankcombe and Y. Liu, Chem. Mater., 2023, 35, 5468–5474.
39. Y. Ma, H. Tao, X. Guo, P. Yang, D. Xing, V. Nicolosi, Y. Zhang, C. Lian and B. Qiu, J. Mater. Chem. A, 2024, 12, 27220–27228.
40. Y. Sun, H. Liao, J. Wang, B. Chen, S. Sun, S. J. H. Ong, S. Xi, C. Diao, Y. Du and J.-O. Wang, Nat. Catal., 2020, 3, 554–563.
41. N. Zhang, J. Shang, X. Deng, L. Cai, R. Long, Y. Xiong and Y. Chai, ACS Nano, 2022, 16, 4795–4804.
42. C. Zhou, Y. Zhang, C. Xie, J. Bai, J. Li, H. Zhang, H. Zhu, M. Long, B. Zhou and G. Zheng, Environ. Sci. Technol., 2024, 58, 14940–14948.
43. J. Ni, J. Yan, F. Li, H. Qi, Q. Xu, C. Su, L. Sun, H. Sun, J. Ding and B. Liu, Adv. Energy Mater., 2024, 14, 2400065.
44. F. Ma, Y. Wen, P. Fu, J. Zhang, Q. Tang, T. Chen, W. Luo, Y. Zhou and J. Wang, Small, 2024, 20, 2305767.
45. Y.-C. Hao, Y. Guo, L.-W. Chen, M. Shu, X.-Y. Wang, T.-A. Bu, W.-Y. Gao, N. Zhang, X. Su and X. Feng, Nat. Catal., 2019, 2, 448–456.
46. M. Du, Y. Zhang, S. Kang, C. Xu, Y. Ma, L. Cai, Y. Zhu, Y. Chai and B. Qiu, Small, 2023, 19, 2303693.
47. M. Song, Y. Wu, Z. Zhao, M. Zheng, C. Wang and J. Lu, Adv. Mater., 2024, 36, 2403234.
48. F. Ma, Z. Li, R. Hu, Z. Wang, J. Wang, J. Li, Y. Nie, Z. Zheng and X. Jiang, ACS Catal., 2023, 13, 14163–14172.
49. H. Zhong, Q. Zhang, J. Yu, X. Zhang, C. Wu, Y. Ma, H. An, H. Wang, J. Zhang and X. Wang, Adv. Energy Mater., 2023, 13, 2301391.
50. W. Luo, H. Tian, Q. Li, G. Meng, Z. Chang, C. Chen, R. Shen, X. Yu, L. Zhu and F. Kong, Adv. Funct. Mater., 2024, 34, 2306995.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5mh00130g

‡ These authors contributed equally to this work.

---