English

• As the main end product and important intermediate, NH₃ has been widely used in the preparation of rubbers, plastics, dyes, fertilizers and medicines, which is essential to human life and economic development [1]. Furthermore, benefiting from the high weight hydrogen content (17.8 wt%) and large volumetric hydrogen energy density (12.7 MJ/L, liquid NH₃), as well as the advantages of easy handling, storage and transportation of liquefaction, NH₃ is also considered as a potential fuel and ideal green energy carrier in recent years [2]. Currently, industrial NH₃ synthesis still relies on the traditional Haber-Bosch process with the thermocatalytic conversion of N₂ and H₂ in the presence of Fe-based catalysts under harsh conditions (400–600 ℃ and 20–40 MPa) due to the chemical inertness of N≡N (bond energy: 941 kJ/mol) [3]. As an energy-intensive technology for NH₃ synthesis, the Haber-Bosch process accounts for more than 1% of global energy consumption each year [4]. Meanwhile, H₂ used in the Haber-Bosch process is mostly produced through the decomposition of fossil fuels, causing 3%–5% of the world's natural gas consumption and leading to 1.87 tons CO₂ byproduct for each ton of NH₃, which will further aggravate the global greenhouse effect [5]. In result, it is imperative to develop a clean, efficient and sustainable technology alternative to the Haber-Bosch process.

  Among other approaches (e.g., biocatalysis, photocatalysis, etc.) for the synthesis of NH₃ by N₂ reduction reaction (NRR) under ambient conditions, electrocatalytic NRR (N₂ + 6H⁺/e^– → 2NH₃) has attracted increasing attention because of the following merits: (1) about 20% higher energy efficiency than the Haber-Bosch process according to the thermodynamic prediction, (2) application of renewable energy resources (solar, wind, and hydro) as the driving force for electrocatalytic NRR, (3) elimination of fossil fuels as the H₂ sources via an oxidation reaction at the anode for H⁺, and (4) modularity, scalability and on-site, on-demand NH₃ generation of the electrochemical systems [6-8]. Benefiting from the merits mentioned above, electrocatalytic NRR has great potential to replace the Haber-Bosch process. However, the industrial application of electrocatalytic NRR is still plagued by some problems such as sluggish kinetics for N₂ adsorption and N≡N bond rupture which result in large overpotentials and low reaction rates, and the competition from H₂ evolution reaction (HER) which leads to a low Faraday efficiency (FE) [9]. Under these circumstances, the focus of current research is to design and develop advanced catalysts to improve the activity and selectivity of electrocatalytic NRR [10].

  Over the past decades, 2D transition metal dichalcogenides (TMDs) have been extensively studied in catalysis, energy storage, and optoelectronics due to their large surface area, economic feasibility, simple preparation and chemical stability [11, 12]. Among them, 2D Mo-based TMDs such as MoS₂ have attracted particular attention as NRR catalysts since Mo provides the pivotal active center for N₂ fixation in natural nitrogenase due to unique Mo-N₂ interaction [13, 14]. Compared to MoS₂, MoSe₂ may have better electrochemical properties due to lower bandgap and higher conductivity [15]. As such, we explored the NRR performance of MoSe₂ in our previous work; however, it was found to be unsatisfactory because the active center Mo was only exposed at the semi-open edge, and the electronegative Se-mantled surface area remained inaccessible to N₂ [16]. Hence, basal plane activation through vacancy engineering has become an important strategy to improve the NRR performance by regulating the charge transport and surface adsorption capacity [17]. For example, Liu et al. synthesized vacancy-rich ReSe₂ through high-temperature annealing in an inert atmosphere, and found that the introduction of Se vacancies could modulate the electronic structures of ReSe₂ and enhanced its NRR performance [18]. However, this process was sophisticated, and the vacancy concentration was still limited.

  Herein, we propose a simple and effective strategy to create high-concentration Se vacancies in 2D MoSe₂ through heteroatom doping induced lattice strain. At first, MoSe₂ nanosheets (with and without heteroatom) are facilely synthesized on a conductive matrix, i.e., a carbon nanofibrous membrane (CNM) by a one-step hydrothermal method. Considering Zr⁴⁺ has the same (and unchangeable) valence state as Mo⁴⁺, and its ionic radius (0.72 Å) is somewhat smaller than that of Mo⁴⁺ (0.79 Å), we expect it can enter the hexagonal lattice of MoSe₂ and generate lattice strain substantially [19, 20]. In result, abundant Se vacancies are created, as characterized by X-Ray diffraction (XRD), electron paramagnetic resonance (EPR), and X-ray photoelectron spectroscopy (XPS). The vacancy-rich MoSe₂ exhibits enhanced Mo-N₂ interaction than neat MoSe₂ nanosheets, delivering an NH₃ yield of 3.04 × 10^–10 mol s^–1 cm^–2 and an FE of 21.61% at –0.45 V vs. RHE in 0.1 mol/L Na₂SO₄. This strategy may pave an avenue to a new type of high-performance catalysts toward NRR.

  The schematic representation of the synthesis process of MoSe₂ and Zr-MoSe₂ nanosheets@CNM is shown in Fig. 1. Firstly, a homogeneous solution with a polyacrylonitrile (PAN) content of 13 wt% was sucked into a syringe and fabricated into a PAN nanofibrous membrane (PAN NM) through an electrospinning progress. Then, the PAN NM was stabilized in air, followed by carbonization in N₂ to obtain a CNM. Finally, MoSe₂ or Zr-MoSe₂ nanosheets were grown on the surface of individual carbon nanofibers by a hydrothermal process (in the absence or presence of Zr⁴⁺). As shown in Fig. S1 (Supporting information), the PAN nanofibers have a smooth surface with an average diameter of around 306 nm, and are connected with each other to form a 3D porous network. After stabilization and carbonization, the average diameter of the resultant carbon nanofibers decreases to 235 nm because of the entropic shrinkage and chemical shrinkage [21]. As can be seen from the field-emission scanning electron microscopy (FESEM) images in Figs. 2a and b, both nanofibers have a core-shell structure with the carbon nanofiber skeleton uniformly decorated with nanosheets after the hydrothermal process. From the enlarged view of individual nanofibers in the insets of Figs. 2a and b, the nanosheets grow perpendicularly and form channels with open spaces on the surface of individual carbon nanofibers, which is helpful to promote the electrochemical reaction by storing the electrolyte and shortening the diffusion path of ions [22]. Taking Zr-MoSe₂ nanosheets@CNM as the example, the average thickness of the shell is around 125.5 nm from the distribution of fiber diameters in Fig. S1. The high-resolution transmission electron microscopy (HRTEM) image in Fig. 2c shows that the nanosheets have an interlayer distance of 0.65 nm corresponding to the (002) plane of 2H-MoSe₂ [23]. The electron dispersion spectroscopy (EDS) images of Zr-MoSe₂ nanosheets@CNM in Fig. 2d reveals the homogeneous distribution of Mo, Se and Zr elements, further proving the successful synthesis of Zr-MoSe₂ nanosheets on the carbon nanofiber.

  Figure 1

  

  Figure 1.  Schematic representation of the synthesis process of Zr-MoSe₂ nanosheets@CNM through electrospinning of a PAN NM, conversion to a CNM, and hydrothermal synthesis of MoSe₂ nanosheets in the presence of Zr⁴⁺ doping ions.

  DownLoad: Full-Size Img PowerPoint

  Figure 2

  

  Figure 2.  FESEM images of (a) MoSe₂ and (b) Zr-MoSe₂ nanosheets@CNM. (c) HRTEM and (d) EDS images of Zr-MoSe₂ nanosheets. The insets in (a) and (b) are enlarged FESEM images of individual nanofibers.

  DownLoad: Full-Size Img PowerPoint

  In order to determine the presence of Se vacancies in Zr-MoSe₂ nanosheets@CNM, XRD, EPR and XPS measurements were performed. As shown in Fig. S2 (Supporting information), all the diffraction peaks correspond to 2H-MoSe₂ (JCPDF card No. 29–0941) and no secondary phases can be detected in Zr-MoSe₂ nanosheets@CNM, implying new compounds were not formed. From the enlarged view of XRD patterns at 10°–20° in Fig. 3a, slight deviation of the (002) diffraction peak can be observed in Zr-MoSe₂ nanosheets@CNM due to lattice distortion. Note that Zr⁴⁺ with a smaller radius brings about a shrinkage of the MoSe₂ lattice, leading to the higher shift of 2θ degree [24]. An obvious shift of 0.21° caused by the same valance state and suitable size between Zr⁴⁺ and Mo⁴⁺ is favorable for the replacement of Mo⁴⁺ by Zr⁴⁺ in the similar crystal structure and the production of abundant Se vacancies [25, 26]. Compared to MoSe₂ nanosheets@CNM, a stronger EPR signal (g = 2.003) caused by the trapping of electrons in the Se vacancies of Zr-MoSe₂ nanosheets@CNM can be observed in Fig. 3b, further verifying the existence of a high concentration of Se vacancies [27]. Meanwhile, when Zr⁴⁺ is incorporated in the MoSe₂ nanosheets, the XPS spectra of Mo 3d_3/2 and Mo 3d_5/2 (Fig. 3c) as well as Se 3d (Fig. 3d) shift to lower binding energies and provide another proof of the presence of Se vacancies and Mo³⁺ due to the decreased electron density around Mo (Fig. S3 in Supporting information) [28, 29]. Furthermore, from the XPS spectrum of Zr 3d in Fig. S4 (Supporting information), we can conclude that Zr⁴⁺ is indeed incorporated into the lattice of the MoSe₂ nanosheets [30]. The presence of Se vacancies can be directly observed under high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM), where more dark dots (missing anions) occur in the Zr-MoSe₂ nanosheets@CNM (Fig. S5 in Supporting information).

  Figure 3

  

  Figure 3.  (a) Enlarged view of XRD patterns in 10°–20° range, (b) EPR spectra, (c) Mo 3d XPS spectra, and (d) Se 3d XPS spectra of MoSe₂ and Zr-MoSe₂ nanosheets@CNM.

  DownLoad: Full-Size Img PowerPoint

  As shown in Fig. S6 (Supporting information), the Zr-MoSe₂ nanosheets@CNM after the hydrothermal process can still be bent attributed to its excellent mechanical properties, which can serve as a self-supporting catalyst. The NRR performance was tested in a three-electrode system in an H-type electrolysis cell with a Nafion 212 membrane as the separator. According to the linear sweep voltammetry (LSV) curves in Fig. S7 (Supporting information), the current density in N₂-saturated 0.1 mol/L Na₂SO₄ electrolyte is larger than that in Ar-saturated electrolyte between –0.8 V vs. RHE and –0.2 V vs. RHE, implying the effective N₂ reduction of Zr-MoSe₂ nanosheets@CNM [31]. The chronoamperometric curves at different potentials are shown in Fig. 4a, and no obvious fluctuation in the current can be observed, indicating good stability during the electrocatalytic process [32, 33]. The indophenol blue method was adopted to detect the concentration of NH₃ at different potentials. According to the standard curves and UV–vis absorption spectra of electrolyte stained with indophenol blue indicator after 2 h reaction in Fig. S8 (Supporting information), the NH₃ yield and FE can be calculated. As plotted in Fig. 4b, the highest NH₃ yield (3.04 × 10^–10 mol s^–1 cm^–2) and FE (21.61%) occur at –0.45 V vs. RHE for Zr-MoSe₂ nanosheets@CNM. In contrast, the highest NH₃ yield and FE of MoSe₂ nanosheets@CNM occur at –0.55 V vs. RHE, being 1.12 × 10^–10 mol s^–1 cm^–2 and 5.96%, respectively. Moreover, the NRR performance of Zr-MoSe₂ nanosheets@CNM is among the best when compared to MoSe₂ or MoS₂-based catalysts reported elsewhere [13, 16, 34-42], due to the abundant Se vacancies generated by Zr⁴⁺ doping (Table S1 in Supporting information). Considering the stability of catalysts is a critical parameter in practical applications, cyclic and long-time electrolysis tests for Zr-MoSe₂ nanosheets@CNM at –0.45 V vs. RHE were assessed. As observed in Fig. 4c, the NH₃ yield and FE do not significantly change during 5 consecutive cyclic tests, showing the excellent recyclability for N₂ reduction. Besides, no obvious fluctuation of current density (Fig. 4d) and well-preserved morphology of nanosheets (inset of Fig. 4d) after 12 h electrolysis suggest that Zr-MoSe₂ nanosheets@CNM has a good electrochemical durability. The possible byproduct N₂H₄ was not detected after 12 h electrolysis by the Watt and Chrisp method (Fig. S9 in Supporting information). Note that these values are significantly lower than those of Zr-MoSe₂ nanosheets@CNM, proving that the higher vacancy concentration is beneficial to lower the reaction overpotential and improve the performance of electrocatalytic NRR [43-45].

  Figure 4

  

  Figure 4.  (a) Chronoamperometric curves and (b) NH₃ yield and FE of Zr-MoSe₂ nanosheets@CNM at different potentials (V vs. RHE). (c) NH₃ yield and FE of Zr-MoSe₂ nanosheets@CNM at –0.45 V vs. RHE in 5 repeated cycles. (d) Chronoamperometric curve of Zr-MoSe₂ nanosheets@CNM at –0.45 V vs. RHE for 12 h and the corresponding FESEM image after the long-term durability test.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, this work demonstrates a simple and effective strategy to create high-concentration Se vacancies in MoSe₂ through Zr⁴⁺ doping induced lattice strain, aiming to improve the electrocatalytic NRR performance. Zr-MoSe₂ nanosheets@CNM with good self-supporting properties is synthesized by a hydrothermal process. Abundant Se vacancies are created, as characterized by XRD, EPR and XPS. The vacancy-rich MoSe₂ exhibits enhanced Mo-N₂ interaction, delivering an NH₃ yield of 3.04 × 10^–10 mol s^–1 cm^–2 and an FE of 21.61% in 0.1 mol/L Na₂SO₄ with excellent stability for electrocatalytic NRR.

  Declaration of competing interest

  The authors declare no competing interests.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China (No. 52173055), the Natural Science Foundation of Shanghai (No. 19ZR1401100), the Fundamental Research Funds for the Central Universities and DHU Distinguished Young Professor Program (No. LZA2020001).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2022.03.005.

  
  
• 1. [1]

     Y. Wang, Q. Li, W. Shi, P. Cheng, Chin. Chem. Lett. 31 (2020) 1768–1772. doi: 10.1016/j.cclet.2020.01.010

  2. [2]

     M. Aziz, A.T. Wijayanta, A.B.D. Nandiyanto, Energies 13 (2020) 3062. doi: 10.3390/en13123062

  3. [3]

     G. Qing, R. Ghazfar, S.T. Jackowski, et al., Chem. Rev. 120 (2020) 5437–5516. doi: 10.1021/acs.chemrev.9b00659

  4. [4]

     F. Lai, N. Chen, X. Ye, et al., Adv. Funct. Mater. 30 (2020) 1907376. doi: 10.1002/adfm.201907376

  5. [5]

     M. Wang, M.A. Khan, I. Mohsin, et al., Energy Environ. Sci. 14 (2021) 2535–2548. doi: 10.1039/d0ee03808c

  6. [6]

     H. Shen, C. Choi, J. Masa, et al., Chem 7 (2021) 1708–1754. doi: 10.1016/j.chempr.2021.01.009

  7. [7]

     M. Zhang, Y. Wang, Y. Zhang, et al., Angew. Chem. Int. Ed. 59 (2020) 23252–23260. doi: 10.1002/anie.202010110

  8. [8]

     Y.T. Liu, L. Tang, J. Dai, J. Yu, B. Ding, Angew. Chem. Int. Ed. 59 (2020) 13623–13627. doi: 10.1002/anie.202005579

  9. [9]

     A.R. Singh, B.A. Rohr, M.J. Statt, et al., ACS Catal. 9 (2019) 8316–8324. doi: 10.1021/acscatal.9b02245

  10. [10]

      L. Shi, Y. Yin, S. Wang, H. Sun, ACS Catal. 10 (2020) 6870–6899. doi: 10.1021/acscatal.0c01081

  11. [11]

      H. Wang, C. Li, P. Fang, Z. Zhang, J.Z. Zhang, Chem. Soc. Rev. 47 (2018) 6101–6127. doi: 10.1039/C8CS00314A

  12. [12]

      X. Zhu, L. Mo, Y. Wu, et al., Compos. Commun. 9 (2018) 86–91. doi: 10.1016/j.coco.2018.06.010

  13. [13]

      L. Zhang, X. Ji, X. Ren, et al., Adv. Mater. 30 (2018) 1800191. doi: 10.1002/adma.201800191

  14. [14]

      J. Guo, T.T. Tsega, I.U. Islam, et al., Chin. Chem. Lett. 31 (2020) 2487–2490. doi: 10.1016/j.cclet.2020.02.019

  15. [15]

      P. Yang, Z. Wu, Y. Jiang, et al., Adv. Energy Mater. 8 (2018) 1870116. doi: 10.1002/aenm.201870116

  16. [16]

      M. Zhang, L. Zhang, S. Huang, et al., Appl. Catal. B: Environ. 292 (2021) 120175. doi: 10.1016/j.apcatb.2021.120175

  17. [17]

      F. Yuan, R. Sun, L. Fu, G. Zhao, Chin. Chem. Lett. 33 (2022) 2188–2194. doi: 10.1016/j.cclet.2021.09.003

  18. [18]

      F. Lai, W. Zong, G. He, et al., Angew. Chem. Int. Ed. 59 (2020) 13320–13327. doi: 10.1002/anie.202003129

  19. [19]

      A.A. Tedstone, D.J. Lewis, P. O'Brien, Chem. Mater. 28 (2016) 1965–1974. doi: 10.1021/acs.chemmater.6b00430

  20. [20]

      M.S.L. Aparicio, M.A. Ocsachoque, D. Gazzoli, I.L. Botto, I.D. Lick, Catalysts 7 (2017) 293. doi: 10.3390/catal7100293

  21. [21]

      I.M. Alarifi, A. Alharbi, W.S. Khan, R. Asmatulu, J. Appl. Polym. Sci. 133 (2016) 43235.

  22. [22]

      D. Ji, S. Peng, L. Fan, et al., J. Mater. Chem. A 5 (2017) 23898–23908. doi: 10.1039/C7TA08166A

  23. [23]

      S. Huang, M. Zhang, Y.T. Liu, J. Phys. : Conf. Ser. 1948 (2021) 012222. doi: 10.1088/1742-6596/1948/1/012222

  24. [24]

      Y.S. Jeon, H.W. Seo, S.H. Kim, Y.K. Kim, J. Nanosci. Nanotechnol. 16 (2016) 4814–4819. doi: 10.1166/jnn.2016.12264

  25. [25]

      N. Cao, Z. Chen, K. Zang, et al., Nat. Commun. 10 (2019) 2877. doi: 10.1038/s41467-019-10888-5

  26. [26]

      M.J. Mleczko, C. Zhang, H.R. Lee, et al., Sci. Adv. 3 (2017) e1700481. doi: 10.1126/sciadv.1700481

  27. [27]

      H. He, D. Huang, Q. Gan, et al., ACS Nano 13 (2019) 11843–11852. doi: 10.1021/acsnano.9b05865

  28. [28]

      Z. Luo, Y. Ouyang, H. Zhang, et al., Nat. Commun. 9 (2018) 2120. doi: 10.1038/s41467-018-04501-4

  29. [29]

      D. Xiao, Q. Ruan, D.L. Bao, et al., Small 16 (2020) 2001470. doi: 10.1002/smll.202001470

  30. [30]

      J. Song, J. Dai, P. Zhang, et al., Nano Res. 14 (2021) 1479–1487. doi: 10.1007/s12274-020-3206-x

  31. [31]

      L. Yang, T. Wu, R. Zhang, et al., Nanoscale 11 (2019) 1555–1562. doi: 10.1039/c8nr09564g

  32. [32]

      Y.T. Liu, X. Chen, J. Yu, B. Ding, Angew. Chem. Int. Ed. 58 (2019) 18903–18907. doi: 10.1002/anie.201912733

  33. [33]

      W. Xu, M. Zhang, C. Ma, S. Wu, Y.T. Liu, Appl. Catal. B: Environ. 308 (2022) 121225. doi: 10.1016/j.apcatb.2022.121225

  34. [34]

      L. Yang, H. Wang, X. Wang, et al., Inorg. Chem. 59 (2020) 12941–12946. doi: 10.1021/acs.inorgchem.0c02058

  35. [35]

      Z. Wu, R. Zhang, H. Fei, et al., Appl. Surf. Sci. 532 (2020) 147372. doi: 10.1016/j.apsusc.2020.147372

  36. [36]

      X. Li, T. Li, Y. Ma, et al., Adv. Energy Mater. 8 (2018) 1801357. doi: 10.1002/aenm.201801357

  37. [37]

      J. Zhang, X. Tian, M. Liu, et al., J. Am. Chem. Soc. 141 (2019) 19269–19275. doi: 10.1021/jacs.9b02501

  38. [38]

      H. Su, L. Chen, Y. Chen, et al., Angew. Chem. Int. Ed. 59 (2020) 20411–20416. doi: 10.1002/anie.202009217

  39. [39]

      J. Liang, S. Ma, J. Li, et al., J. Mater. Chem. A 8 (2020) 10426–10432. doi: 10.1039/d0ta03622f

  40. [40]

      X. Xu, X. Tian, B. Sun, et al., Appl. Catal. B: Environ. 272 (2020) 118984. doi: 10.1016/j.apcatb.2020.118984

  41. [41]

      J. Chen, C. Zhang, M. Huang, et al., Appl. Catal. B: Environ. 285 (2021) 119810. doi: 10.1016/j.apcatb.2020.119810

  42. [42]

      X. Zi, J. Wan, X. Yang, et al., Appl. Catal. B: Environ. 286 (2021) 119870. doi: 10.1016/j.apcatb.2020.119870

  43. [43]

      F. Lai, J. Feng, X. Ye, et al., J. Mater. Chem. A 8 (2020) 1652–1659. doi: 10.1039/c9ta11408d

  44. [44]

      Y.T. Liu, D. Li, J. Yu, B. Ding, Angew. Chem. Int. Ed. 58 (2019) 16439–16444. doi: 10.1002/anie.201908415

  45. [45]

      L. Tang, J. Dai, Y.T. Liu, et al., Compos. Commun. 23 (2021) 100551. doi: 10.1016/j.coco.2020.100551

• 

  Figure 1  Schematic representation of the synthesis process of Zr-MoSe₂ nanosheets@CNM through electrospinning of a PAN NM, conversion to a CNM, and hydrothermal synthesis of MoSe₂ nanosheets in the presence of Zr⁴⁺ doping ions.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  FESEM images of (a) MoSe₂ and (b) Zr-MoSe₂ nanosheets@CNM. (c) HRTEM and (d) EDS images of Zr-MoSe₂ nanosheets. The insets in (a) and (b) are enlarged FESEM images of individual nanofibers.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Enlarged view of XRD patterns in 10°–20° range, (b) EPR spectra, (c) Mo 3d XPS spectra, and (d) Se 3d XPS spectra of MoSe₂ and Zr-MoSe₂ nanosheets@CNM.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Chronoamperometric curves and (b) NH₃ yield and FE of Zr-MoSe₂ nanosheets@CNM at different potentials (V vs. RHE). (c) NH₃ yield and FE of Zr-MoSe₂ nanosheets@CNM at –0.45 V vs. RHE in 5 repeated cycles. (d) Chronoamperometric curve of Zr-MoSe₂ nanosheets@CNM at –0.45 V vs. RHE for 12 h and the corresponding FESEM image after the long-term durability test.

  下载: 全尺寸图片 幻灯片
•