English

• With the increasing global energy demands and environmental pollution, water electrolysis, as a promising route for hydrogen generation, has drawn much attention in recent years due to high conversion efficiency and reaction selectivity, ease of operation as well as zero emission of pollutant. However, a considerable high overpotential is usually necessary to realize water electrolysis owing to the sluggish four-electron transfer mechanism of the anodic oxygen evolution reaction (OER) [1]. Fortunately, electrocatalysis of small molecules, such as urea and hydrazine hydrate, could overcome this drawback by replacing the OER to couple with cathodic hydrogen evolution reaction (HER), realizing hydrogen production and pollution treatment simultaneously [2]. Notably, the thermodynamic equilibrium potential can decrease from 1.23 V to 0.37 V vs. the reversible hydrogen electrode (RHE) by using urea oxidation reaction (UOR) [3]. Besides, electrocatalytic oxidation (ECO) of alcohols and aldehydes into the carbonyl materials, is also highly promising to replace the OER to facilitate hydrogen generation and achieve value-added products [4]. Moreover, hazardous oxidants or O₂ with high pressure which are usually applied in the traditional organic oxidation synthesis can be avoided [5]. However, in either UOR or ECO, highly active electrocatalysts are necessary because of the multi-electron transfer process.

  Transition metal sulfides (TMSs) based catalysts, strictly speaking, the "pre-catalysts", have been proved that usually exhibit superior catalytic activity to metal oxides because of the in-situ formed metal (oxy)-hydroxides on the surface of TMSs during electrocatalytic oxidation due to the polarized conditions, especially in a strong oxidation environment [6]. However, their catalytic behavior still cannot meet the need of industrial applications in electrocatalysis owing to the unsatisfied ability to enrich the active species and the unfulfilled electron transfer rate. Among the strategies to enhance the catalytic activity, polymetallic synergy is highly effective yet challenging route to improve the charge transfer rate and facilitate reaction kinetics [7, 8]. For instance, Ni can react with Fe ions to optimize the electronic structure of Fe-based catalysts, resulting in the enhanced electrical conductivity and electron transport ability [9]. Moreover, the partial charge-transfer between Ni and Fe can induce the generation of Ni³⁺ ions which are deemed as the active sites for electrocatalytic oxidation [10]. Besides, in order to enrich active species, construction of a hierarchical porous architecture with both 2D and 3D structures can not only shorten the diffusion channels to facilitate the mass and electron transport and enable more active species participating in the reaction, but also supply enough buffer spaces for stress release to ensure structure stability during continuous operation, further contributing to the robust catalytic activity, the rapid reaction kinetics as well as the excellent electrochemical stability [11, 12]. In view of these premises, we deduced that constructing TMSs pre-catalysts with both 2D and 3D structures based on polymetallic synergy would be an effective strategy to enrich surface sites and facilitate electron transport, contributing to the promoted surface oxidation to boost the catalytic activity.

  In this work, a multi-functional pre-electrocatalyst based on (Ni, Co)S₂ solid solution with 2D/3D porous polyhedron-on-sheet hierarchical structure was fabricated, and the synthetic route is shown in Fig. 1a. Firstly, the Ni-Co precursor with 2D nanosheet structure aligns on carbon cloths (CC) evenly via the electro-deposition method (Figs. S1a–d in Supporting information). Subsequently, the nanosheets, as the template to realize the structure reconstruction, coordinate with 2-methylimidazole (2-MeIM) to achieve the hierarchical architecture with both 2D and 3D characteristics. Scanning electron microscopy (SEM) images give strong evidence for the in-situ structural remodeling (Figs. S2a and b in Supporting information). New rhombic dodecahedrons are formed and grafted uniformly on the nanosheet layers after coordination, indicating the successful structure reformation. Transmission electron microscopy (TEM) image further identifies the solid characteristics and the intimately bound between dodecahedrons and nanosheets (Fig. S2c in Supporting information). The X-ray diffraction (XRD) pattern in Fig. S2d (Supporting information) identifies that all peaks can be indexed to the phase of ZIF-67 (NiCo-ZIF/CC). For sulfur addition, the NiCo-ZIF/CC precursors are hydrothermally vulcanized at 160 ℃ for 2 h. Figs. 1b and c show that the products keep the original structure of NiCo-ZIF/CC. However, the surface of polyhedrons appears coarse and porous, and the interior becomes hollow. Fig. S3 (Supporting information) further confirms a higher specific surface area (31.57 m²/g) of the precursor after sulfuration than NiCo-ZIF/CC (10.10 m²/g), which is attributed by the mesoporous structure of polyhedrons. As mentioned above, such porous hierarchical structure can ensure sufficient diffusion pathways for ion and mass transfer, and enrich more active species during oxidation process, resulting in the robust catalytic activity and facile reaction kinetics. Other splits with different levels of Ni incorporation display similar morphology in Fig. S4 (Supporting information).

  Figure 1

  

  Figure 1.  (a) Schematic diagram of synthetic route. (b, c) SEM images of (Ni, Co)S₂/CC 1:1. (d) XRD patterns of catalysts. High-solution XPS spectra of (e) Co, (f) Ni, (g) S.

  DownLoad: Full-Size Img PowerPoint

  XRD patterns in Fig. 1d reveal that the pristine sample without Ni incorporation after sulfurization is pure CoS₂ (JCPDS card No. 41–1471) except for the diffraction peaks of carbon (JCPDS card No. 41–1487), while the nearly identical peaks can be observed after Ni addition except for the weak shift from the (200) plane at 32.30° in CoS₂ towards the (200) plane at 31.59° in NiS₂ (JCPDS card No. 11–0099), indicating the formation of (Ni, Co)S₂ solid solution. Besides, as shown in the peak at 54.94°, the more the addition of nickel, the more obvious the shift to the plane of NiS₂ (53.65°). Valence state of the pre-catalyst ((Ni, Co)S₂/CC 1:1) was investigated by X-ray photoelectron spectroscopy (XPS). From Fig. 1e, we can see two dominated peaks can be divided into the spin-orbit doublets of Co 2p_3/2 and Co 2p_1/2, respectively. The energies at 782.0 and 798.2 eV are designated as Co²⁺ and the peaks at around 779.1 and 794.1 eV are assigned to Co³⁺ which is active for electro-oxidation [13, 14]. Besides, two broad peaks located at 786.0 and 803.2 eV are the shakeup satellites. It should be noted that the characteristic peaks of Co³⁺ in (Ni, Co)S₂/CC 1:1 display a negative shift about 0.38 eV compared with CoS₂/CC, ascribing to the partial charge transfer between Ni and Co ions (Fig. S5a in Supporting information). For the high resolution profiles of Ni, the characteristic peaks of Ni²⁺ locate at 853.7 and 871.0 eV, while the peaks of Ni³⁺ anchor at 857.3 and 875.4 eV, confirming the successful incorporation of Ni for the generation of (Ni, Co)S₂ solid solution (Fig. 1f) [12, 15]. Besides, two wide peaks at 861.8 and 881.2 eV represent the satellites. We can see that the peak intensity of Ni³⁺ increases obviously in (Ni, Co)S₂/CC, indicating the partial charge transfer between Ni and Co after Ni incorporation (Fig. S5b in Supporting information). In the sample of NiS₂/CC, the total peak area ratio of Ni³⁺ to Ni²⁺ is 0.94:1. However, this ratio in the sample of (Ni, Co)S₂/CC increases to 1.57:1, illustrating that a part of Ni²⁺ converted to Ni³⁺ due to the electron transfer from Ni to Co. Additionally, three peaks can be fitted in the high-resolution of S 2p, ascribing to the M-S bond, such as Ni-S and Co-S, at 162.3 and 163.6 eV, and S-O bond resulted by surface oxidation at 169.4 eV (Fig. 1g) [9].

  TEM was employed to further identify the detailed structure of (Ni, Co)S₂/CC 1:1. Figs. 2a and b display the structure of thin nanosheet as well as the polyhedron with porous and hollow inherent features, confirming the 2D/3D porous multistage configuration. Fig. 2c displays the clear lattice which is about 0.24 nm, corresponding to the (210) facet of CoS₂ or NiS₂. The fast fourier transform (FFT) pattern originated from Fig. 2c verifies the single crystal essence of the solid solution. Element distribution was further probed via the energy dispersive spectrometry (EDS) analysis. As shown in Fig. 2d, Ni, Co and S disperse uniformly on the substrate, indicating the homogeneous composition of (Ni, Co)S₂ without partial element aggregation. The actual molar ratio of Ni to Co was identified as 1:0.99 by inductively coupled plasma optical emission spectrometry (ICP-OES, Table S1 in Supporting information). The value is nearly the same with the addition ratio, confirming the successful incorporation of Ni in CoS₂.

  Figure 2

  

  Figure 2.  Detailed structure of (Ni, Co)S₂/CC 1:1. (a, b) TEM, (c) HRTEM and the corresponding FFT image, (d) the element mapping.

  DownLoad: Full-Size Img PowerPoint

  It has been well accepted that the active metal species with high valence can be in-situ generated due to the surface oxidation of the pre-catalysts and thus facilitate the electrocatalytic oxidation behavior to realize urea splitting. Therefore, the UOR performance of as-prepared samples was evaluated through linear sweep voltammetry (LSV) measurement using a three-electrode setup in the electrolyte of 1 mol/L KOH with 0.33 mol/L urea. NiS₂ was fabricated in advance for performance comparison (Fig. S6 in Supporting information). As revealed in Fig. 3a, (Ni, Co)S₂/CC samples display a much lower onset potential than NiS₂/CC and CoS₂/CC, indicating the positive effect of Ni addition on the enhancement of UOR performance. For reaching a 10 mA/cm² current density, the lowest potential is presented by (Ni, Co)S₂/CC 1:1 (1.22 V vs. RHE), which is about 40–140 mV lower than other splits (Table S2 in Supporting information), confirming that the optimal molar ratio of Ni and Co is 1:1 to achieve the highest UOR activity. Of note, this significantly improved value is also the most attractive one among a series of advanced catalysts reported previously (Table S3 in Supporting information) [16-22]. To explore the intrinsic activity, electrochemical double-layer capacitance (C_dl), as a general factor to evaluate the electrochemical surface area (ECSA), was calculated to diminish the morphology benefits of porous multistage structure (Fig. 3b and Fig. S7 in Supporting information) [12]. The structure reconstruction with 2D/3D porous configuration results in a much higher C_dl value of 42.45 mF/cm², demonstrating the important role of the 2D nanosheet morphology grafting with 3D porous polyhedrons of the (Ni, Co)S₂/CC pre-catalysts on fertilizing the reactive surface sites for the surface oxidation process to in-situ generated active species. Fig. 3c is the corresponding LSV profiles normalized by C_dl. As expected, the (Ni, Co)S₂/CC 1:1 exhibits the superior intrinsic activity to its counterparts, further confirming the synergistic effects of nickel and cobalt on the enhancement of catalytic activity. Besides, Tafel plots in Fig. S8a further confirm the facile UOR kinetics of the catalysts. (Ni, Co)S₂/CC 1:1 reveals the smallest Tafel slope about 109 mV/dec, which is much smaller than NiS₂ (138 mV/dec) and CoS₂ (183 mV/dec), possibly ascribing to the facilitated mass and charge pathways, the boosted electron transfer rate as well as the enhanced conductivity. Electrochemical impedance spectra (EIS) were measured at 1.40 V vs. RHE to further identify the above reasons (Fig. S8b and Table S2). Improved electrical conductivity and electron transport ability can be observed after Ni incorporation, revealing that the interaction between Ni and Co can optimize the electronic structure of CoS₂, contributing to the favorable charge transport behavior of (Ni, Co)S₂/CC samples for the combined surface oxidation and UOR process. Among them, (Ni, Co)S₂/CC 1:1 shows the smallest charge transfer resistance (R_ct) of 2.13 Ω.

  Figure 3

  

  Figure 3.  (a) LSV curves for UOR without IR correction at 2 mV/s, (b) current density variation at 1.05 V vs. RHE as a function of scan rates, (c) LSV curves normalized by C_dl, (d) CA tests at 1.53 V for 50 h.

  DownLoad: Full-Size Img PowerPoint

  In addition, the chronoamperometry (CA) test was conducted at 1.53 V and 1.35 V to further identify the surface oxidation and assess the long-term durability of the (Ni, Co)S₂/CC 1:1 pre-catalyst (Fig. 3d and Fig. S9 in Supporting information). We can see that a sharp decrease was happened in the first 2 h, which may be ascribed to the delayed removal of CO₃²⁻ ions, thus leading to the rapid reduction of the surface active sites for the subsequent UOR. Fortunately, as the catalytic process proceeds, steady catalytic behavior can be confirmed, resulting in high current retention rates of 93.40% and 94.90% after 50 h continuous UOR at 1.53 V and 1.35 V, respectively. This can be ascribed to the surface oxidation process without structural transformation and the high porosity as the stress buffer basin. Therefore, the morphological change of the catalyst can be ignored after constant redox reaction (Fig. S10 in Supporting information). In order to identify the relationship between the catalytic activity and the in-situ formed surface active species, the LSV curves before and after the CA tests at different times were recorded. From Fig. S11a (Supporting information), we can see the LSV curves after 1 h are obviously different from the initial one, indicating the surface phase of the pre-catalyst has been changed, which can be proved by the XPS results. As depicted in Fig. S12 (Supporting information), the peak intensity of O 1s increases markedly, while the intensity of S 2p peak dramatically decreases along with the increase of the test time. Meanwhile, the gradual increment of the current density can be observed, revealing that the surface oxidation of the pre-catalyst is actually a process for activation. Besides, EIS data display the reduced R_ct value after the first two hours of testing, illustrating the accelerated reaction kinetics for the surface oxidation and UOR (Fig. S11b in Supporting information). Until 50 h, XPS spectra show that the peaks of S 2p nearly disappear accompanying with the significantly enhanced peak of O 1s, indicating the completely surface oxidation of (Ni, Co)S₂ to the metal (oxy)hydroxides (Figs. S13a and b in Supporting information). Meanwhile, the proportion of active Ni³⁺ significantly surpasses the Ni²⁺ in the high resolution profiles of Ni after 50 h. CA test, further confirming the important effect of the surface oxidation to boost active sites for promoted electrocatalytic oxidation (Figs. S13c and d in Supporting information). Generally, these in-situ generated species cannot be identified by XRD due to the amorphous nature or too thin layer on the surface of the pre-catalyst. However, the dense hot spots of O in the results of EDS mapping after CA test can also confirm the surface transformation from (Ni, Co)S₂/CC into the corresponding metal (oxy)hydroxides accompanied by S loss during the UOR to form the unique (Ni, Co)S₂@Co(Ni) based (oxy)hydroxides core-shell structure on CC (Fig. S13e in Supporting information).

  For the purpose to investigate the multi-functional catalytic activity of (Ni, Co)S₂/CC 1:1, the LSV curves in 1 mol/L KOH with different organic molecules, involving benzyl alcohol (BA), 5-hydroxymethylfurfural (HMF), furfuryl alcohol (FFA) and furfural (FF), are drawn in Fig. 4a. The catalyst demonstrates a much higher activity on ECO of these organic molecules than water, implying a high preference to electro-oxidize organic species. Fig. 4b displays the conversion and selectivity for ECO in 1.0 mol/L KOH with 100 mmol/L BA at 1.43 V vs. RHE for 30 h. The concentration of benzoic acid increases linearly with ECO time, while the amount of BA is getting less and less. Finally, a high conversion of BA about 97.50% and an excellent selectivity to benzoic acid about 93.78% can be achieved. The Faraday efficiency (FE) has been calculated to be as high as 91.86%. The inset in Fig. 4b is the photo of products during 30 h ECO operation. We can see that as the reaction time lengthens, the white powder becomes moreand more, confirming the mass production of benzoic acid.

  Figure 4

  

  Figure 4.  (a) LSV curves for ECO, (b) conversion and selectivity for ECO in 1.0 mol/L KOH with 100 mmol/L BA at 1.43 V for 30 h. The inset is the photo of products during ECO.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, a novel multi-functional pre-electrocatalyst based on (Ni, Co)S₂ solid solution with well-designed hierarchical structure and high porosity was fabricated. The interaction between Ni and Co can optimize the electronic structure, resulting in the improved conductivity and accelerated charge transfer rate. In addition, the 2D/3D architecture can enrich more active species and endow the mass and electron transport to facilitate the surface oxidation and the following electrocatalytic oxidation process. Post-structure and catalytic characterizations confirm the surface oxidation of (Ni, Co)S₂/CC during the stability test, and the in-situ formed Co(Ni) based (oxy)hydroxides exhibit superior catalytic activity and facilitated charge transfer ability, contributing to the impressive multifunctional performance and making the NiCo-based sulfide solid solution as a promising pre-catalyst for hydrogen production with the treatment of nitrogenous wastewater and the acquisition of value-added products synchronously.

  Declaration of competing 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.

  Acknowledgments

  This work was supported by National Natural Science Foundation of China (Nos. 21927811, 51602182, 21808129) and the Natural Science Foundation of Shandong Province, China (No. ZR2021ME032).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2022.107843.

  
  
• 1. [1]

     X. Wang, L. Yu, B.Y. Guan, S. Song, X.W. Lou, Adv. Mater. 30 (2018) 1801211. doi: 10.1002/adma.201801211

  2. [2]

     L. Wang, Y. Zhu, Y. Wen, et al., Angew. Chem. Int. Ed. 60 (2021) 10577–10582. doi: 10.1002/anie.202100610

  3. [3]

     P. Hao, W. Zhu, L. Li, et al., Electrochim. Acta 338 (2020) 135883. doi: 10.1016/j.electacta.2020.135883

  4. [4]

     Z. Li, J. Zhang, X. Jing, et al., J. Mater. Chem. A 9 (2021) 6152–6159. doi: 10.1039/D0TA09421H

  5. [5]

     X. Chen, X. Zhong, B. Yuan, et al., Green Chem. 21 (2019) 578–588. doi: 10.1039/C8GC03451F

  6. [6]

     M. Zhou, Q. Weng, X. Zhang, et al., J. Mater. Chem. A 5 (2017) 4335–4342. doi: 10.1039/C6TA09366C

  7. [7]

     A. Haeussler, S. Abanades, J. Jouannaux, et al., AIMS Mater. Sci. 6 (2019) 657–684. doi: 10.3934/matersci.2019.5.657

  8. [8]

     Z. Ji, Y. Song, S. Zhao, et al., ACS Catal. 12 (2022) 569–579. doi: 10.1021/acscatal.1c05190

  9. [9]

     P. Hao, Y. Xin, J. Tian, et al., Sci. Chin. Chem. 63 (2020) 1030–1039. doi: 10.1007/s11426-020-9770-5

  10. [10]

      P. Hao, W. Zhu, L. Li, et al., Chem. Commun. 55 (2019) 10138–10141. doi: 10.1039/C9CC05115E

  11. [11]

      K. Rui, G. Zhao, Y. Chen, et al., Adv. Funct. Mater. 28 (2018) 1801554. doi: 10.1002/adfm.201801554

  12. [12]

      P. Hao, H. Wen, Q. Wang, et al., J. Mater. Chem. A 9 (2021) 8576–8585. doi: 10.1039/D1TA00043H

  13. [13]

      H. Tan, Z. Liu, D. Chao, et al., Adv. Energy Mater. 8 (2018) 1800685. doi: 10.1002/aenm.201800685

  14. [14]

      Y. Lu, D. Fan, Z. Chen, et al., Sci. Bull. 65 (2020) 460–466. doi: 10.1016/j.scib.2019.12.020

  15. [15]

      F. Li, P. Wang, X. Huang, et al., Angew. Chem. Int. Ed. 58 (2019) 7051–7056. doi: 10.1002/anie.201902588

  16. [16]

      Q. Zhang, F.M. Kazim, S. Ma, et al., Appl. Catal. B 280 (2021) 119436. doi: 10.1016/j.apcatb.2020.119436

  17. [17]

      Z. Cao, T. Zhou, X. Ma, et al., ACS Sustain. Chem. Eng. 8 (2020) 11007–11015.

  18. [18]

      H. Liu, S. Zhu, Z. Cui, et al., Nanoscale 13 (2021) 1759–1769. doi: 10.1039/D0NR08025J

  19. [19]

      Q. Li, N. Li, J. An, H. Pang, Inorg. Chem. Front. 7 (2020) 2089–2096. doi: 10.1039/D0QI00316F

  20. [20]

      W. Han, X. Li, L. Lu, et al., Chem. Commun. 56 (2020) 11038–11041. doi: 10.1039/D0CC03177A

  21. [21]

      M. Yuan, X. Guo, N. Li, H. Pang, J. Colloid Interface Sci. 589 (2021) 56–64. doi: 10.1016/j.jcis.2020.12.100

  22. [22]

      N. Chen, Y. Du, G. Zhang, W. Lu, F. Cao, Nano Energy 81 (2021) 105605. doi: 10.1016/j.nanoen.2020.105605

• 

  Figure 1  (a) Schematic diagram of synthetic route. (b, c) SEM images of (Ni, Co)S₂/CC 1:1. (d) XRD patterns of catalysts. High-solution XPS spectra of (e) Co, (f) Ni, (g) S.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Detailed structure of (Ni, Co)S₂/CC 1:1. (a, b) TEM, (c) HRTEM and the corresponding FFT image, (d) the element mapping.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) LSV curves for UOR without IR correction at 2 mV/s, (b) current density variation at 1.05 V vs. RHE as a function of scan rates, (c) LSV curves normalized by C_dl, (d) CA tests at 1.53 V for 50 h.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) LSV curves for ECO, (b) conversion and selectivity for ECO in 1.0 mol/L KOH with 100 mmol/L BA at 1.43 V for 30 h. The inset is the photo of products during ECO.

  下载: 全尺寸图片 幻灯片
•