An Fe₃S₄/Ni₃S₂ heterostructure realizing highly efficient electrocatalysis of ethylene glycol and alkaline electrolyte to produce high value-added chemicals and hydrogen

Hongwei Ren^ab, Junming Zhang*^a, Tianjun Hu^a, Yongfeng Li^b, Haocheng Zhao^b, Ergui Luo^a, He Xiao^a, Man Zhao^a, Jingxiao Tang*^c and Jianfeng Jia*^a
aKey Laboratory of Magnetic Molecules and Magnetic Information Materials of Ministry of Education, School of Chemistry and Chemical Engineering, Shanxi Normal University, Taiyuan 030032, China. E-mail: zhangjunming@sxnu.edu.cn; jiajf@dns.sxnu.edu.cn
^bDepartment of Energy Chemistry and Materials Engineering, Shanxi Institute of Energy, Jinzhong 030600, China
^cSchool of Chemistry & Chemical Engineering and Environmental Engineering, Weifang University, Weifang 261061, China. E-mail: 20230094@wfu.edu.cn

First published on 31st July 2025

---

---

Green foundation 1. An Fe3S4/Ni3S2 (NiFeS) heterostructure electrocatalyst has been synthesized on a nickel foam substrate via a facile one-pot solvothermal strategy for energy-saving H2 and value-added formate production from alkaline media. 2. In 1.0 M KOH + 1.0 M ethylene glycol electrolyte, the highest Faraday efficiency (92.4%) and rapid productivity (0.652 mmol cm−2 h−1) could be achieved for formate production at an applied potential of 1.50 V vs. RHE. Notably, the coupled electrolysis system consisting of NiFeS(+)∥Pt/C(−) achieves a cell voltage of 1.55 V at 100 mA cm−2, which is 140 mV lower than that of the traditional water electrolysis system. 3. Preparing non-precious metal-based electrodes and optimizing their synthesis strategies will enhance the sustainability of “green hydrogen”. The coupling of organic small molecule oxidation and hydrogen evolution reactions can achieve the production of high-value-added chemicals while reducing the energy consumption of hydrogen production.

1. Introduction

Hydrogen, as a clean and renewable resource, is gradually becoming a novel energy carrier to replace fossil fuels in solving the energy crisis and environmental pollution.^1–4 Currently, electrolysis of water to generate “green hydrogen” is one of the most efficient and promising strategies for hydrogen production.^5–8 In contrast to the hydrogen evolution reaction (HER) on the cathode, the OER on the anode has been widely recognized as the bottleneck for electrochemical overall water splitting due to the rigid OO double bond and the multi-step proton and electron transfer process.^9–13 Even if commercial precious metal catalysts (e.g., IrO₂, RuO₂, and Pt/C) are used, the actual potential of water splitting is much higher than the theoretical potential (1.23 V vs. RHE).^14–16 To overcome this limitation, researchers have been exploring alternative anodic reactions that can perform oxidation reactions at lower potentials and produce value-added products, such as glycerol oxidation, urea oxidation, glucose oxidation, benzyl alcohol oxidation, methanol oxidation, etc.^17–21

In this regard, the coupling of the EGOR and the HER enables selective oxidation to high-value formate at the anode while generating hydrogen fuel at the cathode, as compared to conventional electrolysis of water.^22–24 For a long time, the electrocatalytic oxidation of ethylene glycol over noble metal-based catalysts has been extensively studied.^25–27 For example, Wang et al. proposed a controllable strategy to prepare defect-rich Pd@PdO nanobelts with a thickness of 1.8 nm, which exhibited excellent electrocatalytic activity and stability for the EGOR.²⁸ The trimetallic Pd₁₁Ni₁₁Pt₂ nanocatalyst with abundant three-phase surfaces exhibited significant electrocatalytic EGOR performance, in which Ni and Pt synergistically enhanced this reaction through the oxophilic effect and surface modification mechanism, respectively.²⁹ In addition, the Pd–PdSe heterostructure exhibited excellent C–C bond breaking ability for ethylene glycol, showing high EGOR selectivity (more than 44%) for C1 products.³⁰ However, high-cost noble metal materials are susceptible to poisoning by intermediates such as CO, thus hindering the industrial application of the coupled hydrogen production strategy. Therefore, it is urgent to develop non-precious-metal EGOR catalysts with high electrocatalytic activity, selectivity, and stability, so as to improve the reaction efficiency and reduce the production cost of chemicals and “green hydrogen” fuel.

Transition metal sulfides are considered potential alternatives for precious metal catalysts because of their low cost and high efficiency.^31,32 It is worth noting that nickel- and iron-based sulfides show impressive OER performance due to their abundant redox active sites.^33,34 However, the poor conductivity and the challenge of polysulfide shuttle effects leading to particle agglomeration seriously hinder further improvement of their electrochemical performance.³⁵ Engineering heterostructures is an effective strategy that combines the advantages of more than two materials to improve the electronic structure of catalysts and promote electrochemical activity.^36,37 For example, Xiong et al. constructed an atomically thick Ni₃S₂/MnO₂ nanosheet array with abundant heterogeneous interfaces, which achieved fast kinetics and excellent water splitting performance due to strong electronic coupling resulting from the interfacial interactions between the two components.³⁸ Fan et al. prepared a NiS–FeS ultrathin nanosphere heterostructure on an iron foam for alkaline seawater splitting at industrial grade density via one-step sulfuration etching.³⁹ The NiS–FeS facilitates interfacial electron transfer, optimizes active site exposure, and accelerates bubble release and mass transfer. Zhao et al. constructed a Co_0.3–Ni₃S₂/Ni₃Sn₂S₂ heterostructure, which inhibited the aggregation of active species during the electrolysis process and exhibited excellent performance for the electrochemical oxidation reaction of freshwater, seawater, and urea.⁴⁰ All these studies have demonstrated the possibility of heterostructure electrocatalysts for practical applications.

Driven by the above facts, an Fe₃S₄/Ni₃S₂ heterostructure has been synthesized on a nickel foam substrate via a facile one-pot solvothermal strategy. This integrated solvothermal-sulfuration approach surpasses traditional routes in both energy efficiency and cost-effectiveness. The unique physical and chemical properties of ethylene glycol make its use possible for controlling the growth of particles, and enable tuning of the composition and structure of the as-synthesized sulfide. The NiFeS electrocatalyst exhibits excellent OER performance and long-term stability in 1.0 M KOH electrolyte. In addition, it also facilitates the conversion of ethylene glycol into formate with high faradaic efficiency in 1.0 M KOH + 1.0 M ethylene glycol electrolyte. This study puts forward a novel strategy for preparing heterojunction catalysts for energy-saving H₂ and value-added formate production from alkaline media.

2. Experimental section

2.1. Chemicals and materials

Iron(III) chloride hexahydrate (FeCl₃·6H₂O), thiourea (CH₄N₂S), ethylene glycol (C₂H₆O₂), hydrochloric acid (HCl), ethanol (C₂H₅OH), acetone, and nickel foam (NF, with a thickness of 0.5 mm) were purchased from Sinopharm Chemical Reagent Co. Ltd (Shanghai, China). All the above chemicals were of analytical grade and used directly without any further purification. In addition, ruthenium oxide (RuO₂, 99.95%, Alfa Aesar), commercial Pt/C (20% Pt, Alfa Aesar), Nafion solution (5 wt%, Sigma-Aldrich), and water (18.2 MΩ cm, Millipore) were used.

2.2. Synthesis of catalysts

First, accurately weighed 0.8 mmol of FeCl₃·6H₂O and 2.4 mmol of thiourea were dispersed in 20 mL of ethylene glycol. The above mixture was mixed evenly and poured into a 25 mL Teflon-lined stainless-steel autoclave. Then, a piece of acid-treated NF (2.0 × 3.0 cm²) was completely immersed in the above solution, which served as the nickel source and self-supporting carrier. Subsequently, the autoclave was placed in a homogeneous reactor and reacted at 160 °C for 14 hours. After natural cooling, the as-prepared material was washed and dried to obtain the Fe₃S₄/Ni₃S₂/NF catalyst (denoted as NiFeS). NiS and NiFe catalysts were synthesized by similar schemes as described above without adding FeCl₃·6H₂O or thiourea, respectively. Details of the synthesis are provided in the SI.

Preparation of Pt/C and RuO₂ electrodes: 2 mg of Pt/C or 20 mg of RuO₂ catalyst was ultrasonically dispersed in 1 mL of a mixed solution (containing 500 μL of water, 450 μL of ethanol, and 50 μL of Nafion), and then 250 μL of ink was dropped on NF (1.0 × 1.0 cm²) and dried in a vacuum oven for later use.

2.3. Characterization

The crystal structure of materials was determined using a Rigaku Ultima IV-185 X-ray diffractometer (XRD, Cu Kα radiation source, λ = 0.154056 nm) at a scan rate of 5° min⁻¹. The morphology and structure of the samples were characterized by scanning electron microscopy (SEM, JSM-7500F scanning electron microscope), combined with energy-dispersive X-ray spectroscopy (EDS). Transmission electron microscopy (TEM) images, high-resolution TEM (HRTEM) images, and elemental mapping images were obtained on a TEM instrument (JEM-2100 transmission electron microscope, 200 kV). X-ray photoelectron spectroscopy (XPS, Thermo Fisher Scientific K-Alpha, Al Kα monochromatic X-ray source) was performed to confirm the elemental composition and surface chemical valence states of catalysts. In addition, Raman spectra were recorded on a Renishaw InVia Reflex Raman microscope. ¹H and ¹³C nuclear magnetic resonance (NMR) spectroscopy was performed on an AVANCE III HD 600 MHz (Bruker).

2.4. Electrochemical tests

All electrochemical performances were recorded using a CHI 760E electrochemical workstation. The electrochemical tests were carried out in a standard three-electrode system, in which the as-prepared catalysts, a graphite rod, and a Hg/HgO electrode served as the working electrode (geometric surface area of 1 × 1 cm²), counter electrode, and reference electrode, respectively. All potential values were calibrated to the reversible hydrogen electrode (RHE) according to the equation E_RHE = E_Hg/HgO + 0.098 V + 0.059 × pH. The OER and EGOR measurements were performed in 1.0 M KOH solution with and without 1.0 M EG, respectively. The linear sweep voltammetry (LSV) polarization curves were recorded at a scan rate of 5 mV s⁻¹ with 85% iR-compensation. To eliminate the effect of oxidation current densities of metal species on the OER and EGOR, LSV curves were recorded by sweeping from high potential to low potential. Tafel slopes were obtained by plotting the overpotential against log|j| from LSV data. The electrochemically active surface area (ECSA) was calculated from the double-layer capacitance (C_dl), and the C_dl value was obtained by cyclic voltammetry measurements. The electrochemical impedance spectroscopy (EIS) measurement was carried out in the frequency range from 10⁻² to 10⁵ Hz. The long-term stability was tested by chronoamperometry and the accelerated degradation test (ADT, 50 mV s⁻¹, 5000 cycles).

3. Results and discussion

3.1. Material properties

Scheme 1 illustrates that the Fe₃S₄/Ni₃S₂ heterostructure electrocatalyst was successfully synthesized via a facile one-pot solvothermal strategy, as described using eqn (1)–(5) (see the supplementary information (SI) for the equations and experimental details). X-ray diffraction (XRD) patterns allow analysis of the crystal structures of the as-prepared NiFeS, NiS, and NiFe catalysts. In Fig. 1A, the strong diffraction peaks presented at 44.3°, 51.7°, and 76.1° for all three samples correspond to the characteristic peaks of Ni (PDF # 89-7128) in the NF substrate. For the NiFeS catalyst, diffraction peaks of two substances, Ni₃S₂ and Fe₃S₄, appeared in the XRD pattern. The diffraction peaks located at 21.75°, 31.10°, 37.78°, 38.27°, 50.12°, 55.16°, 69.27°, and 73.04° were assigned to the (101), (110), (003), (021), (211), (122), (131), and (214) crystal planes of Ni₃S₂ (PDF # 44-1418), respectively, while the diffraction peaks at 15.48°, 29.96°, 36.34°, 47.81°, and 52.36° correspond to the (111), (311), (400), (511) and (440) crystal faces of Fe₃S₄ (PDF # 16-0713), respectively. The relatively weak diffraction intensity of Fe₃S₄ can be attributed to its lower content and poorer crystallinity.⁶ In contrast, except for the diffraction peaks of NF, the diffraction peaks of the NiS catalyst only correspond to the Ni₃S₂ substance, while no obvious diffraction peaks are observed for the NiFe catalyst. This is primarily attributable to the amorphous nature of the NiFe catalyst prepared without thiourea.^7,8 The morphology and microstructure of the NiFeS catalyst were probed by SEM and TEM. The sample grew uniformly on the NF substrate in the form of nanosheets, and some melts with rugged surfaces were randomly dispersed on the nanosheets (Fig. 1B and S1), both of which were composed of Fe₃S₄ and Ni₃S₂. The nanosheets on the bottom layer will expand the electrochemically active surface area and provide abundant active sites for electrochemical reactions. Meanwhile, the melt structure dispersed on the upper layer has abundant interflake voids, which ensures that the electrolyte or reactant fills and penetrates into the active sites.²⁴ In the HRTEM image (Fig. 1C), the lattice spacing of 0.287 nm and 0.182 nm corresponds to the (110) and (012) crystal planes of the Ni₃S₂, respectively. The lattice spacing of 0.298 nm and 0.175 nm belongs to the (311) and (440) crystal planes of Fe₃S₄, which are in accordance with the above XRD results. Obviously, different crystal planes appear simultaneously on both sides of the heterojunction surface corresponding to Fe₃S₄ and Ni₃S₂, respectively (Fig. S2). The interfacial bonds within this heterostructure can act as bridges for electron transfer, thereby accelerating the charge transport rate. Elemental mapping (Fig. 1D) further demonstrates that Ni, Fe, S, and O are uniformly distributed across the surface of the NiFeS catalyst. Additionally, X-ray energy dispersive spectroscopy (EDS) was employed to quantify the elemental composition of the electrode (Fig. S3), which yielded a Ni:Fe:S:O atomic ratio of 49.9:3.6:30.4:16.1. Similarly, the NiS catalyst exhibits a rambutan-like structure (Fig. S4), whereas the NiFe is composed of irregular structures of about 500 nm (Fig. S5). The EDS spectra and elemental compositions shown in Fig. S6 and S7 confirm the successful preparation of the NiS and NiFe catalysts, respectively.

The XPS survey spectrum (Fig. S8) shows that Ni, Fe, O, and S elements coexist in the NiFeS catalyst, and the strong O element peak is mainly due to the sulfate and adsorbed oxygen on the sample surface.⁶ For the NiFeS catalyst (Fig. 2A), the three doublets of Ni 2p can be attributed to Ni³⁺ (at about 856.8 and 874.9 eV), Ni²⁺ (at about 855.4 and 873.0 eV), and Ni⁰ (at about 852.5 and 869.6 eV), accompanied by satellite peaks (indicated as sat.).⁴¹ Moreover, for the NiFeS catalyst (Fig. 2B), the Fe 2p spectrum can be deconvoluted into a peak of Fe⁰ (at about 706.0 eV), two doublets of Fe²⁺ peaks (at about 711.0 and 724.4 eV, ascribed to Fe²⁺–S species), two doublets of Fe³⁺ peaks (at about 714.2 and 728.9 eV, ascribed to Fe³⁺–S species), and a satellite component.^5,42,43 The S 2p spectra of NiFeS shown in Fig. 2C illustrate three splitting peaks at 162.1, 163.4, and 168.3 eV, which can be assigned to S²⁻, S_n²⁻, and SO₄²⁻, respectively.^11,43 Interestingly, there is a positive shift of Ni 2p and S 2p, while a significant negative shift of Fe 2p binding energies, as compared to those of NiFe and NiS materials, which is mainly caused by the S atoms with higher electronegativity and the strong electronic interactions among various components.⁴⁴ This indicates the charge redistribution of Ni, Fe, and S atoms at the surface of the Fe₃S₄/Ni₃S₂ heterojunction surface, which can also be confirmed by the higher contents of Ni⁰ and Fe²⁺/Fe³⁺ in NiFeS than in other compounds. Therefore, the formation of Fe₃S₄ and Ni₃S₂ heterojunctions can significantly improve the electronic conductivity, and the interfacial bonds can act as a bridge for electron transfer to increase the charge transfer rate.³⁶ For the O 1s spectra shown in Fig. 2D, the peaks of the metal–oxygen (M–O) and metal–hydroxyl group (M–OH) bonds are found at 529.6 and 531.1 eV for NiFeS, while the peak at 532.4 eV is associated with adsorbed H₂O. All these results suggest that the Fe₃S₄/Ni₃S₂/NF heterostructure with nanosheet arrays optimizes the electronic states of the active species, which may be conducive to enhancing their intrinsic OER and EGOR performance.

3.2. Electrocatalytic OER performance

In a typical three-electrode system, the self-supported electrodes served as anodes to evaluate the electrocatalytic OER performance in 1.0 M KOH electrolyte. As shown in Fig. 3A, the current densities on the LSV curve increase with the applied potential. For the NiFeS, NiFe, and NiS catalysts, the reduction peaks of oxides appear in the range of about 1.25–1.35 V, which may be attributed to the oxidation or surface remodeling of the catalysts.^45,46 Furthermore, NiFeS exhibits a larger reduction peak area than NiFe and NiS catalysts, implying that the former may have generated more active sites. As expected, the NiFeS electrode exhibits excellent electrocatalytic OER activity (Fig. 3B), with the lowest overpotential (240 mV) to drive a current density of 100 mA cm⁻² relative to NiFe (302 mV), NiS (321 mV), commercial RuO₂ (339 mV), and bare NF (480 mV). Interestingly, the NiFeS catalyst is superior to NiS, while the NiS catalyst is weaker than NiFe. This may be because of the synergistic effect between Ni and Fe, which further accelerates the electrocatalytic OER process, with the heterogeneous interface between Fe₃S₄ and Ni₃S₂ playing a significant role. Of note, the overpotential at 100 mA cm⁻² (η₁₀₀) of the NiFeS catalyst is lower than that of the state-of-the-art transition metal-based OER catalysts (Table S1).^47,48 Tafel plots can be obtained from the corresponding LSV curves (Fig. 3C), and the Tafel slope of NiFeS is only 34.3 mV dec⁻¹, which is much lower than those of NiFe (52.8 mV dec⁻¹), NiS (78.7 mV dec⁻¹), commercial RuO₂ (79.8 mV dec⁻¹), and bare NF (110.7 mV dec⁻¹), illustrating that the construction of a heterojunction can lead to higher reaction rates and faster OER kinetics. The linear relationship between the capacitance current density and scan rate was obtained through the CV curves at different scan speeds from 20 to 100 mV s⁻¹ (Fig. S9). As shown in Fig. 3D, the C_dl value of NiFeS is 16.0 mF cm⁻², which is significantly larger than those of NiFe (12.7 mF cm⁻²), NiS (3.2 mF cm⁻²), and bare NF (2.4 mF cm⁻²). Fig. S10 shows that NiFeS achieves the largest ECSA (400 cm⁻²), thereby exposing more active sites to facilitate the OER process. Furthermore, the ECSA-normalized OER LSV curves in Fig. S11 show that NiFeS attains the largest current densities at the same overpotentials, confirming that it has the highest intrinsic activity. Electrochemical impedance spectroscopy (EIS) reveals that the NiFeS catalyst has a smaller R_ct (0.18 Ω) than NiFe (0.25 Ω), NiS (0.73 Ω), commercial RuO₂ (0.75 Ω), and bare NF (1.39 Ω), demonstrating an expeditious electron transport at the NiFeS catalyst/electrolyte interface and faster kinetics (Fig. 3E). The NiFeS catalyst exhibits outstanding long-term durability, with no significant performance degradation observed after continuous electrolysis at 100 mA cm⁻² for 120 h (Fig. 3F). Furthermore, only a slight decay of the polarization curve for the NiFeS catalyst was observed after 5000 CV cycles (Fig. S12).

To explore the details of the electrode reaction and the reason for the high electrocatalytic OER performance, we plotted the Bode phase plots of three catalysts at different potentials (Fig. 4A–C). The phase angle decreases sharply in the high-frequency region, which correlates with the deep oxidative reconstruction of the catalysts. Nevertheless, the overall phase angle changes in the high-frequency region are stronger for NiFeS (1.23–1.43 V) than forNiFe (1.23–1.33 V) and NiS (1.23–1.28 V) during the OER process, which is consistent with the results of oxidation potentials and peak areas in the above LSV curves. Moreover, the response in the low-frequency region reflects the nonhomogeneous charge distribution during the electrocatalytic process, which is related to the OER at the catalyst interface.^49,50 As observed, the low-frequency peak of the NiFeS phase angle significantly decreases at 1.38 V, compared to the 1.43 V of NiFe and NiS, indicating that NiFeS has the lowest onset potential for the OER, and this trend is consistent with the OER response in LSV. These results suggest that the OER occurs on the surface of the oxide layer formed by electro-oxidation of the catalyst. In addition, for the original NiFeS catalyst (Fig. 4D), the Raman peaks at 197, 223, 246, and 300 cm⁻¹ are consistent with the vibration of Ni₃S₂, and the characteristic peaks at 329 and 345 cm⁻¹ correspond to Fe–S.^51–53 However, after the electrocatalytic OER, some new characteristic peaks appear in the NiFeS catalyst. The peak near 557 cm⁻¹ corresponds to the NiOOH species, and the peak at 671 cm⁻¹ could be attributed to the FeOOH species, further confirming that the metal oxyhydroxides (NiOOH and FeOOH) are generated by the surface reconstruction of the Fe₃S₄/Ni₃S₂ heterojunction during the OER test.^54,55 This result is consistent with numerous research findings, indicating that metal oxyhydroxides are the true active centers in the OER process.^56,57 Furthermore, for the NiFeS electrode, the changes of the chemical state, morphology, and crystalline phase after the OER durability test were analyzed by XPS, SEM, and XRD. Compared to the initial NiFeS (Fig. S8), the XPS survey spectrum (Fig. 4E) exhibits a reduction in the intensity of the sulfur peak alongside an increase in that of the oxygen peak. This is primarily attributed to the partial dissolution of sulfur in the catalyst during the OER process, coupled with the formation of oxyhydroxides via surface reconstruction of the sulfides.^3,37 The Ni⁰ peak in the Ni 2p spectrum (Fig. 4F) completely disappeared after the OER test, indicating that the Ni species were oxidized during the OER process. Similarly, Fig. 4G shows that the peak of Fe⁰ in the Fe 2p spectrum also completely disappeared, and the peak position of Fe 2p_3/2 was positively shifted by 1.77 eV, suggesting that Fe species were also oxidized during the OER process. The above results suggest that Ni and Fe species undergo deep oxidation reconstruction during the OER process, which is consistent with the above conclusions. This can also be confirmed by the negative shift and increased area ratio of M–O and M–OH in the XPS spectrum of O 1s (Fig. 4H). Fig. 4I shows that the S 2p peak located at 168.2 eV corresponding to the SO₄²⁻ species still exists. This result indicates that even after a drastic OER, there is still residual SO₄²⁻ anchored on the surface of the catalyst, which is speculated to play a significant role in the OER.^58,59 In addition, the morphology of the NiFeS catalyst remained almost unchanged (Fig. S13) after the OER stability test. However, only the diffraction peaks corresponding to Ni₃S₂ were observed in the XRD pattern (Fig. S14). This phenomenon can be ascribed to the reconstruction of low-content, poorly crystalline Fe₃S₄ during the OER process, leading to the formation of amorphous FeOOH, as evidenced by the EDS spectrum in Fig. S15. Ni₃S₂ with a stable supporting structure, by synergizing with the stable oxyhydroxide layer formed after surface reconstruction, ensures the long-term stability of NiFeS in the OER.⁵⁶

3.3. Electrocatalytic EGOR performance

Previous reports have indicated that the EGOR is a thermodynamically more favorable reaction than the anodic OER for hydrogen production from conventional water electrolysis. Thus, the coupling of the EGOR with the HER can produce hydrogen in the cathodic region and high-added-value chemicals in the anodic region of an electrolyzer. In this study, the three-electrode system was also used to complete the performance test of all catalysts in 1.0 M KOH + 1.0 M EG electrolyte with the same test procedure as for the OER. Fig. 5A demonstrates the order of electrocatalytic EGOR performance of several catalysts: NiFeS > NiFe > NiS > RuO₂, which is consistent with the results of electrocatalytic OER performance, and the NiFeS catalyst is comparable to the recently reported catalysts (Table S2). As shown in Fig. 5B, the electrocatalytic EGOR potential of the NiFeS catalyst is only 1.37 V to reach a current density of 100 mA cm⁻², which is significantly lower than the 1.47 V in the OER process. The Tafel slope is used to study the kinetics of the electrocatalytic EGOR (Fig. 5C) with the order of magnitude NiFeS (45.1 mV dec⁻¹) < NiFe (47.2 mV dec⁻¹) < NiS (75.1 mV dec⁻¹) < RuO₂ (155.6 mV dec⁻¹), suggesting that NiFeS accelerates the reaction rate of the EGOR. Notably, NiFeS achieves the largest C_dl value of 13.1 mF cm⁻² and ECSA (327.5 cm⁻²) compared to the others (Fig. 5D, S16, and S17), implying its highest intrinsic EGOR activity. Subsequently, EIS was used to evaluate the charge transfer rate of the electrode at the same open-circuit potential. As shown in Fig. 5E, the distribution trend of the Nyquist semicircle of each catalyst in the EGOR process was basically the same as that in the OER process. Similarly, the NiFeS catalyst exhibits a smaller R_ct value (0.18 Ω) compared with other catalysts. The chronoamperometry test (Fig. 5F) was conducted at 1.55 V for up to 120 h to evaluate the durability of the NiFeS catalyst during the EGOR process. With the extension of time, the concentration of EG in the electrolyte gradually decreased, so the current density gradually decreased from 100 mA cm⁻² to about 85 mA cm⁻² after 8 h. Subsequently, the current density could be rapidly recovered to 100 mA cm⁻² by replacing the fresh electrolyte every 8 h.

Compared with the Bode plot of the OER, the phase angle variation in the high-frequency region of NiFeS's Bode phase diagram for the EGOR almost disappears (Fig. 6A), which suggests that the NiFeS catalyst has not been reconstructed. Furthermore, the phase angle in the low-frequency region varies from a lower potential of 1.28 V, which is less than that of the OER, indicating that the reaction kinetics of the EGOR on the NiFeS surface is better than that of the OER. Ethylene glycol molecules are oxidized directly on the surface of the unreconstructed NiFeS catalyst, which is also confirmed by the CV curves of the OER and EGOR shown in Fig. S18. Thus, the EGOR does occur earlier than the OER, and the lower electrooxidation potential of the EGOR implies lower electrical energy consumption, which is more suitable for coupling with the HER for electrochemical H₂ production via an energy-saving approach. The Raman spectrum (Fig. 6B) shows that the characteristic peaks after the EGOR are identical to those of the original NiFeS, suggesting that the metal sulfides serve as genuine active species for the EGOR. Moreover, the XRD pattern (Fig. S19) still exhibits distinct diffraction peaks corresponding to Ni₃S₂ and Fe₃S₄, indicating that the crystalline phases of the catalyst remained largely unchanged during the EGOR process. These results collectively imply that the EGOR proceeds on the heterojunction interface between Ni₃S₂ and Fe₃S₄. The XPS survey spectrum (Fig. S20) shows that the Ni, Fe, O, C, and S elements still coexisted in the NiFeS catalyst after the electrocatalytic EGOR. Compared with the original NiFeS, the XPS spectrum of Ni 2p (Fig. 6C) shows a weakening of the Ni⁰ peak after the EGOR. Meanwhile, Fig. 6D illustrates that the peak position of Fe 2p_3/2 is positively shifted by 2.0 eV, and the ratio of Fe³⁺/Fe²⁺ is changed from 0.31 to 1.46. These results suggest that Ni and Fe species undergo slight oxidation during the EGOR process, which can also be confirmed by a negative shift of the peak position of O 1s (Fig. 6E). Fig. 6F shows that the signal of S 2p on the surface of the NiFeS catalyst has not changed. After the EGOR stability test, the morphology of the NiFeS catalyst changed slightly but remained highly dispersed (Fig. S21).

To investigate the EGOR mechanism of the NiFeS catalyst, the oxidation products of EG were qualitatively and quantitatively analyzed using ¹H and ¹³C NMR spectra. As presented in Fig. 7A, formate is the main product with the highest Faraday efficiency and selectivity within the potential window of 1.30–1.70 V. When the applied potential is 1.50 V, the Faraday efficiency and selectivity of EG-to-formate conversion are 92.4% and 94.3%, respectively. Further higher applied potentials result in a slight decrease in Faraday efficiency and selectivity, which can be ascribed to the deep oxidation of formate or enhanced water oxidation at high potentials.⁶⁰ Furthermore, the yield of formate shows a positive correlation with the applied potential, reaching 0.793 mmol cm⁻² h⁻¹ at 1.70 V. In order to gain insights into the dynamic transformation of EG during the oxidation process, the ¹H NMR spectra of EG electrooxidation at 1.50 V with continuous charge input from 0 to 4000 C were recorded. Fig. 7B shows that EG was gradually consumed while the signal of formate gradually became stronger during the electrooxidation process. Subsequently, the consumption of EG and the concentration of the formate product were calculated using the standard curve of formic acid (Fig. S22). It was concluded that the conversion rate of EG is 39.1% and the selectivity of formate is 83.8% after inputting a charge of 4000 C (Fig. 7C). By enlarging the ¹H NMR spectrum, two weak peaks corresponding to methanol and glycolate (GA) can be observed in the chemical shift regions of 3.20–3.30 ppm and 3.80–3.82 ppm, respectively (Fig. 7D and E). In addition, the ¹³C NMR spectrum of the electrolyte after the EGOR also showed the generation of formate and a small amount of carbonate (Fig. 7F). Based on these results and previous studies, the possible pathway for the electrooxidation of EG on the NiFeS electrode can be proposed (Fig. 7G).^{24,51,61–63}

3.4. Electrochemical hydrogen production

Inspired by the excellent electrocatalytic OER and EGOR performance of the NiFeS catalyst, a two-electrode electrolyzer was assembled for producing hydrogen by overall water splitting or coupled HER/EGOR, using NiFeS or commercial RuO₂ as the anode and Pt/C as the cathode (Fig. 8A). In 1.0 M KOH electrolyte, the NiFeS∥Pt/C electrolyzer exhibits excellent overall water splitting performance, and a current density of 100 mA cm⁻² could be achieved at the cell potential of 1.69 V, which is much better than that of the commercial RuO₂∥Pt/C electrolyzer (1.88 V) (Fig. 8B). Meanwhile, the cell potential of NiFeS∥Pt/C (1.55 V) in 1.0 M KOH + 1.0 M EG electrolyte at 100 mA cm⁻² is significantly lower than that of RuO₂∥Pt/C (1.77 V). Furthermore, comparing the NiFeS∥Pt/C electrolyzer at 100 mA cm⁻² in 1.0 M KOH and 1.0 M KOH + 1.0 M EG electrolytes, the cell potential in the HER–EGOR coupled system is reduced by 140 mV (Fig. S23), which indicates that this system not only synthesizes formate with high-added-value but also generates H₂ in an energy-saving manner. When the two-electrode device is operated at 100 mA cm⁻² in 1.0 M KOH electrolyte, the evolution of bubbles can be clearly observed on both electrodes. Meanwhile, NiFeS∥Pt/C is more stable than RuO₂∥Pt/C, which can run stably at 100 mA cm⁻² for at least 50 h (Fig. 8C). In addition, the stability test in 1.0 M KOH + 1.0 M EG electrolyte lasted for 16 h, during which the electrolyte was replaced once, and the results show that the stability of NiFeS∥Pt/C is much better than that of RuO₂∥Pt/C (Fig. 8D).

4. Conclusion

In summary, the NiFeS heterostructure consisting of Fe₃S₄ and Ni₃S₂ has been synthesized on NF substrate by a one-step hydrothermal method. The self-supporting NiFeS electrode exhibits an OER overpotential of only 240 mV at 100 mA cm⁻² in 1.0 M KOH electrolyte. Raman spectroscopy indicates that the metal oxyhydroxides (NiOOH and FeOOH) generated by surface reconstruction are the real active species of the OER. In addition, it also exhibits excellent electrocatalytic EGOR performance with a current density of 100 mA cm⁻² at an applied potential of 1.37 V and the highest Faraday efficiency (92.4%) and rapid productivity (0.652 mmol cm⁻² h⁻¹) at 1.50 V for the formate product. The Raman spectroscopy results indicate that metal sulfides are the true active species of the EGOR. The coupled electrolysis system of the HER (commercial Pt/C electrode) and the EGOR (NiFeS electrode) delivered a current density of 100 mA cm⁻² at 1.55 V. Therefore, this work provides a new strategy for the construction of nickel–iron based sulfide electrocatalysts for energy-saving H₂ and value-added formate production from alkaline media.

Author contributions

Hongwei Ren: writing – original draft, investigation, formal analysis, and methodology. Junming Zhang: formal analysis, data curation, supervision, funding acquisition, and writing – review & editing. Tianjun Hu: formal analysis and supervision. Yongfeng Li: investigation. Haocheng Zhao: formal analysis and funding acquisition. Ergui Luo: validation, investigation, and funding acquisition. He Xiao: investigation and data curation. Man Zhao: conceptualization. Jingxiao Tang: conceptualization, investigation, and writing – review & editing. Jianfeng Jia: funding acquisition, writing – review & editing, and supervision.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

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

Additional experimental details, product analysis method and supplementary figures and tables. See DOI: https://doi.org/10.1039/d5gc02244d.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (U24A2023, 22209102), the Natural Science Foundation of Shanxi Province (20210302124473, 202203021211283, 202203021211284), the China Postdoctoral Science Foundation (2021M691366), and the Natural Science Foundation of Shanxi Normal University (JCYJ2022001).

References

1. E. Jiang, C. Guo, X. Zhao, Y. Chao, D. Ma, P. Huo, Y. Yan, P. Zhou and Y. Yan, ACS Nano, 2023, 17, 10774–10782 CrossRef PubMed .
2. Y. Liang, X. Zhao, P. Yan, L. Xue, H. Li, L. Gu, Y. Long, T. Yu and Y. Yang, J. Colloid Interface Sci., 2024, 655, 565–575 CrossRef PubMed .
3. J. Zhang, Y. Fang, Y. Chen, X. Zhang, H. Xiao, M. Zhao, C. Zhao, X. Ma, T. Hu, E. Luo, J. Jia and H. Wu, J. Colloid Interface Sci., 2024, 653, 821–832 CrossRef PubMed .
4. C. Huang, Q. Zhou, L. Yu, D. Duan, T. Cao, S. Qiu, Z. Wang, J. Guo, Y. Xie, L. Li and Y. Yu, Adv. Energy Mater., 2023, 13, 2301475 CrossRef .
5. M. Ning, F. Zhang, L. Wu, X. Xing, D. Wang, S. Song, Q. Zhou, L. Yu, J. Bao, S. Chen and Z. Ren, Energy Environ. Sci., 2022, 15, 3945–3957 RSC .
6. X. Luo, P. Ji, P. Wang, X. Tan, L. Chen and S. Mu, Adv. Sci., 2022, 9, 2104846 CrossRef PubMed .
7. T. Gao, S. Wu, X. Li, C. Lin, Q. Yue, X. Tang, S. Yu and D. Xiao, Inorg. Chem. Front., 2022, 9, 3598–3608 RSC .
8. X. Hou, C. Yu, T. Ni, S. Zhang, J. Zhou, S. Dai, L. Chu and M. Huang, Chin. J. Catal., 2024, 61, 192–204 CrossRef .
9. X. Ren, T. Wu, Y. Sun, Y. Li, G. Xian, X. Liu, C. Shen, J. Gracia, H. Gao, H. Yang and Z. J. Xu, Nat. Commun., 2021, 12, 2608 CrossRef PubMed .
10. J. Jiang, C. Yan, X. Zhao, H. Luo, Z. Xue and T. Mu, Green Chem., 2017, 19, 3023–3031 RSC .
11. C. Wang, W. Zhang, J. Wang, P. Xia, X. Duan, Q. He, I. Sirés and Z. Ye, Appl. Catal., B, 2024, 342, 123457 CrossRef .
12. L. Yu, L. Wu, B. McElhenny, S. Song, D. Luo, F. Zhang, Y. Yu, S. Chen and Z. Ren, Energy Environ. Sci., 2020, 13, 3439–3446 RSC .
13. B. Guo, H. Huo, Q. Zhuang, X. Ren, X. Wen, B. Yang, X. Huang, Q. Chang and S. Li, Adv. Funct. Mater., 2023, 33, 2300557 CrossRef CAS .
14. Y. Wu, M. Chen, H. Sun, T. Zhou, X. Chen, G. Na, G. Qiu, D. Li, N. Yang, H. Zheng, Y. Chen, B. Wang, J. Zhao, Y. Zhang, J. Zhang, F. Liu, H. Cui, T. He and Q. Liu, Appl. Catal., B, 2025, 360, 124548 CrossRef CAS .
15. X. Guo, J. Li, F. Meng, D. Qin, X. Wu, Y. Lv and J. Guo, J. Colloid Interface Sci., 2024, 663, 847–855 CrossRef CAS .
16. B. Guo, X. Zhang, X. Ma, T. Chen, Y. Chen, M. Wen, J. Qin, J. Nan, Y. Chai and B. Dong, Int. J. Hydrogen Energy, 2020, 45, 9575–9582 CrossRef CAS .
17. Y. Song, X. Wan, Y. Miao, J. Li, Z. Ren, B. Jin, H. Zhou, Z. Li and M. Shao, Appl. Catal., B, 2023, 333, 122808 CrossRef CAS .
18. P. Zhao, Q. Liu, X. Yang, S. Yang, L. Chen, J. Zhu and Q. Zhang, J. Colloid Interface Sci., 2024, 673, 49–59 CrossRef CAS PubMed .
19. T. Shi, B. Gao, H. Meng, Y. Fu, D. Kong, P. Ren, H. Fu and Z. Feng, Green Chem., 2024, 26, 4209–4220 RSC .
20. C. Lv, L. Chen, J. Bai, H. Ruo, Y. Pan, S. Xu, J. Chen, D. Zhang and C. Guo, Chem. Commun., 2024, 60, 5952–5955 RSC .
21. Y. Zuo, W. Sheng, W. Tao and Z. Li, J. Mater. Sci. Technol., 2022, 114, 29–41 CrossRef CAS .
22. X. Jiang, Z. Dong, Q. Zhang, G. Xu, J. Lai, Z. Li and L. Wang, J. Mater. Chem. A, 2022, 10, 20571–20579 RSC .
23. L. Liu, Y. He, D. Ma, X. Wu and Q. Zhu, J. Colloid Interface Sci., 2023, 640, 423–433 CrossRef CAS PubMed .
24. J. Chang, L. Wang, D. Wu, F. Xu, K. Jiang, Y. Guo and Z. Gao, J. Colloid Interface Sci., 2024, 655, 555–564 CrossRef CAS PubMed .
25. G. Xu, Z. Dong, Y. Zhao, W. Zhang, Q. Sun, D. Ju and L. Wang, Small, 2024, 20, 2306341 CrossRef CAS PubMed .
26. D. Zhang, C. Zhang, H. Gao, J. Sui, N. Sui, L. Wang, M. Liu and W. W. Yu, Plasmonics, 2021, 16, 777–786 CrossRef CAS .
27. C. Zhai, J. Hu, L. Zeng, N. Fu, Y. Du and M. Zhu, J. Colloid Interface Sci., 2019, 535, 392–399 CrossRef CAS PubMed .
28. X. Wang, X. Wei, R. Zhang, M. Yan, R. Wei, X. Zhang, Z. Zhu, Y. Wang, X. Zhao and S. Yin, Inorg. Chem. Front., 2024, 11, 2562–2569 RSC .
29. J. Zhang, Y. Chen, X. Zhang, X. Xu, X. Ma, Y. Fan, C. Zhao, H. Xiao, M. Zhao, T. Hu, B. Lv, E. Luo and J. Jia, Mater. Chem. Front., 2024, 8, 2578–2591 RSC .
30. Y. Qin, W. Zhang, F. Wang, J. Li, J. Ye, X. Sheng, C. Li, X. Liang, P. Liu, X. Wang, X. Zheng, Y. Ren, C. Xu and Z. Zhang, Angew. Chem., Int. Ed., 2022, 61, e202200899 CrossRef CAS PubMed .
31. T. Zhou, J. Bai, Y. Gao, L. Zhao, X. Jing and Y. Gong, J. Colloid Interface Sci., 2022, 615, 256–264 CrossRef CAS PubMed .
32. M. He, S. Hu, C. Feng, H. Wu, H. Liu and H. Mei, Int. J. Hydrogen Energy, 2020, 45, 23–35 CrossRef CAS .
33. Y. Yang, H. Meng, C. Kong, S. Yan, W. Ma, H. Zhu, F. Ma, C. Wang and Z. Hu, J. Colloid Interface Sci., 2021, 599, 300–312 CrossRef CAS PubMed .
34. Y. Zhao, J. You, L. Wang, W. Bao and R. Yao, Int. J. Hydrogen Energy, 2021, 46, 39146–39182 CrossRef CAS .
35. A. Morozan, H. Johnson, C. Roiron, G. Genay, D. Aldakov, A. Ghedjatti, C. T. Nguyen, P. D. Tran, S. Kinge and V. Artero, ACS Catal., 2020, 10, 14336–14348 CrossRef CAS .
36. D. Zheng, L. Yu, W. Liu, X. Dai, X. Niu, W. Fu, W. Shi, F. Wu and X. Cao, Cell Rep. Phys. Sci., 2021, 2, 100443 CrossRef CAS .
37. Q. Wen, K. Yang, D. Huang, G. Cheng, X. Ai, Y. Liu, J. Fang, H. Li, L. Yu and T. Zhai, Adv. Energy Mater., 2021, 11, 2102353 CrossRef CAS .
38. Y. Xiong, L. Xu, C. Jin and Q. Sun, Appl. Catal., B, 2019, 254, 329–338 CrossRef CAS .
39. J. Fan, X. Ma, J. Xia, L. Zhang, Q. Bi and W. Hao, J. Colloid Interface Sci., 2024, 657, 393–401 CrossRef CAS PubMed .
40. H. Zhao, M. Liu, X. Du and X. Zhang, Int. J. Hydrogen Energy, 2024, 58, 117–127 CrossRef CAS .
41. Z. Dai, X. Du and X. Zhang, J. Alloys Compd., 2023, 946, 169451 CrossRef CAS .
42. J. Li, Q. Liu, G. Gou, S. Kang, X. Tan, B. Tan, L. Li, N. Li, C. Liu and B. Lai, Sep. Purif. Technol., 2022, 286, 120471 CrossRef CAS .
43. X. Lin, K. Shih, J. Chen, X. Xie, Y. Zhang, Y. Chen, Z. Chen and Y. Li, Chem. Eng. J., 2020, 391, 123558 CrossRef CAS .
44. Z. Zou, X. Wang, J. Huang, Z. Wu and F. Gao, J. Mater. Chem. A, 2019, 7, 2233–2241 RSC .
45. R. Xu, R. Wu, Y. Shi, J. Zhang and B. Zhang, Nano Energy, 2016, 24, 103–110 CrossRef CAS .
46. L. Yu, J. F. Yang, B. Y. Guan, Y. Lu and X. W. Lou, Angew. Chem., Int. Ed., 2018, 57, 1–1 CrossRef CAS .
47. M. Chen, Y. Zhang, J. Chen, R. Wang, B. Zhang, B. Song and P. Xu, Small, 2024, 20, 2309371 CrossRef CAS PubMed .
48. D. Li, W. Wan, Z. Wang, H. Wu, S. Wu, T. Jiang, G. Cai, C. Jiang and F. Ren, Adv. Energy Mater., 2022, 12, 2201913 CrossRef CAS .
49. Y. Lu, T. Liu, C. Dong, C. Yang, L. Zhou, Y. Huang, Y. Li, B. Zhou, Y. Zou and S. Wang, Adv. Mater., 2022, 34, 2107185 CrossRef CAS PubMed .
50. Y. Qi, Y. Zhang, L. Yang, Y. Zhao, Y. Zhu, H. Jiang and C. Li, Nat. Commun., 2022, 13, 4602 CrossRef CAS PubMed .
51. Z. Chen, R. Zheng, T. Bao, T. Ma, W. Wei, Y. Shen and B. Ni, Nano-Micro Lett., 2023, 15, 210 CrossRef CAS PubMed .
52. X. Ji, C. Cheng, Z. Zang, L. Li, X. Li, Y. Cheng, X. Yang, X. Yu, Z. Lu, X. Zhang and H. Liu, J. Mater. Chem. A, 2020, 8, 21199–21207 RSC .
53. W. Liu, X. Wang, F. Wang, X. Liu, Y. Zhang, W. Li, Y. Guo, H. Yin and D. Wang, Chem. Eng. J., 2023, 454, 140030 CrossRef CAS .
54. J. Xu, B.-X. Wang, D. Lyu, T. Wang and Z. Wang, Int. J. Hydrogen Energy, 2023, 48, 10724–10736 CrossRef CAS .
55. C. Huang, Q. Zhou, D. Duan, L. Yu, W. Zhang, Z. Wang, J. Liu, B. Peng, P. An, J. Zhang, L. Li, J. Yu and Y. Yu, Energy Environ. Sci., 2022, 15, 4647–4658 RSC .
56. M. Chen, W. Li, Y. Lu, P. Qi, H. Wu, K. Hao and Y. Tang, Appl. Catal., B, 2024, 358, 124415 CrossRef CAS .
57. S. Hu, Y. Li, D. Kim, M. Liu, L. Y. S. Lee and K. Wong, EcoMat, 2022, 4, e12256 CrossRef CAS .
58. X. Wu, X. Song, H. Tan, Y. Kang, Z. Zhao, S. Jin and X. Chang, Mater. Today Energy, 2022, 26, 101008 CrossRef CAS .
59. J. Na, H. Yu, S. Jia, J. Chi, K. Lv, T. Li, Y. Zhao, Y. Zhao, H. Zhang and Z. Shao, J. Energy Chem., 2024, 91, 370–382 CrossRef CAS .
60. J. Wang, X. Li, T. Zhang, Y. Chen, T. Wang and Y. Zhao, J. Phys. Chem. Lett., 2022, 13, 622–627 CrossRef CAS PubMed .
61. Z. Chen, W. Wei, Y. Shen and B. Ni, Green Chem., 2023, 25, 5979–5988 RSC .
62. K. Liu, Y. Wang, F. Liu, C. Liu, R. Shi and Y. Chen, Chem. Eng. J., 2023, 473, 145292 CrossRef CAS .
63. X. Liu, Z. Fang, X. Teng, Y. Niu, S. Gong, W. Chen, T. J. Meyer and Z. Chen, J. Energy Chem., 2022, 72, 432–441 CrossRef CAS .

This journal is © The Royal Society of Chemistry 2025