English

• NH₃ holds significant importance as a vital chemical compound within the industrial sector, playing a crucial role in modern economic development [1,2]. The electrochemical reduction of N₂ (NRR) with minimal emissions of CO₂ stands as a sustainable and environmentally friendly approach for N₂ fixation under ambient conditions, facilitating the revolution for NH₃ synthesis [3-7]. However, NRR encounters obstacles attributed to the concurrent hydrogen evolution reaction (HER) and the formidable N≡N bond [8]. In contrast, electrochemical reduction of NO to NH₃ (NORR) offers a more energy-efficient and practical pathway for sustainable NH₃ synthesis [9-14]. Additionally, NO itself serves as a prevalent atmospheric pollutant generated by vehicle exhaust and incomplete combustion of fossil fuels. NORR emerges as an efficient and eco-friendly technique for simultaneous NO treatment and NH₃ electro-synthesis. Nevertheless, NORR presents a multi-step process involving intricate five-electron mechanism, resulting in slow reaction kinetics [15-17]. Consequently, the exploration of highly performant catalysts becomes imperative to enhance the NO-to-NH₃ conversion efficiency.

  Single-atom catalysts (SACs) have garnered increasing attention in heterogeneous catalysis, owing to the exceptional potential in delivering high activity and selectivity for a diverse range of chemical reactions [18-26]. To date, a variety of transition metal-based SACs are developed as NORR electrocatalysts, attributed to the ability of transition metal to augment NO adsorption and activation facilitated by the partially occupied d-orbitals [27-35]. Nevertheless, it has been noted that these d-orbitals also promote the binding of ^*H on catalyst surfaces, subsequently promoting the undesired HER and thus impeding NH₃ synthesis [36]. Consequently, the exploration of alternative elements as active sites for SACs, beyond transition metal atoms, remains a topic of ongoing research. Promisingly, main-group metals (Al, Ga, In, Sn, Sb, and Bi, etc.) with partially occupied p-orbitals exhibit robust ability in restraining adsorption of ^*H, and can overcome the shortcoming of transition metal-based SACs [37-41]. Hence, main-group metals are expected to serve as promising SACs for NORR [42-45]. For instance, Chen et al. [42] experimentally reported that Sb single atoms confined in MoO₃ were highly efficient for the NORR with the NH₃ yield rate of 273.5 µmol h⁻¹ cm⁻² and Faradaic efficiency of 91.7%. Wu et al. [45] computationally predicted that single Mg/Al/Ga atom supported on N-doped graphene can act as active centers for direct NO-to-NH₃ conversion. Despite the great potential, main-group metal-based SACs for NORR have not been explored extensively, and it is meaningful to extend the scope both from theoretical and experimental aspects.

  Recently, graphene-like boron carbide (BC₃) monolayer, a semiconductor with a bandgap spanning between 0.46 eV and 0.73 eV, garners enormous attention due to its superior mechanical strength, optical characteristics, and thermal stability [46,47]. These exceptional properties enable it to hold great potential in various fields, including hydrogen storage [48,49], sensor technology [50,51], lithium-ion batteries [52], and catalytic applications [53-57]. Particularly, the graphitic BC₃ nanosheet has been successfully synthesized in experiment via the reaction of benzene with sodium borohydride at 800 ℃ and hold great potential to host SACs for achieving superior electrocatalytic performance [58]. Herein, we systematically investigate the potential of nine main-group metals (i.e., Mg, Ca, Al, Ga, In, Ge, Sn, Sb, and Bi) embedded BC₃ monolayer as SACs for the NORR toward NH₃ synthesis. Through first-principles calculations, we screen out Al@V_B, Ga@V_B, and Ge@V_C systems with outstanding NORR activity and high selectivity for NH₃ production from a pool of 18 candidate systems. Importantly, the NORR process is found to occur spontaneously in the three screened systems. Our research results provide a valuable guidance for further developing in NORR catalysts, as well as demonstrate the great potential of main-group metals as SACs.

  All spin-polarized first-principles calculations are performed using the Vienna Ab initio Simulation Package (VASP) [59] with the projector augmented wave (PAW) approach [60] and the Perdew-Burke-Ernzerhof (PBE) exchange-correlation functional [61]. Additionally, the van der Waals weak interactions are further treated using Grimme’s semi-empirical correction scheme (DFT-D3) [62]. A plane-wave energy cutoff of 500 eV is applied in all calculations. To minimize interactions between two periodic images, a vacuum layer of 20.0 Å is set in the z-direction. For structure optimization and electronic property calculations of a 3 × 3 × 1 supercell, 2 × 2 × 1 and 4 × 4 × 1 k-point grids centered at the Γ point are used following the Monkhorst-Pack scheme [63], respectively. The convergence thresholds are set to be 10⁻⁵ eV for the total energy and 0.02 eV/Å for atomic forces, respectively. More other computational details are provided in Supporting information.

  The pristine unit cell of BC₃ monolayer composed of two boron (B) atoms and six carbon (C) atoms shows a graphene-like planar structure with the space group of P6/mmm (No. 191), in which each B atom is coordinated to three neighboring C atoms. The optimized lattice parameters are a = b = 5.17 Å, and the lengths of C–C and C-B bond are 1.42 and 1.57 Å, respectively, which are consistent with previous reports [64-66]. To simulate the main-group metal-embedded BC₃ monolayer, we construct a 3 × 3 × 1 supercell and consider four kinds of vacancies, including B monovacancy (V_B), C monovacancy (V_C), BC divacancy (V_BC), and CC divacancy (V_CC). To evaluate the formation possibility of the four vacancies, we calculate their respective formation energies (E_f), which are 5.37 eV for V_B, 4.44 eV for V_C, 7.47 eV for V_BC, and 6.05 eV for V_CC, indicating the thermodynamically favorable formation of monovacancies (V_B and V_C) rather than divacancies (V_BC and V_CC). Compared to that of monovacancy in graphene (7.45 eV) [67-69], the lower E_f values of V_B and V_C monovacancies suggest the high feasibility for experimental synthesis. Consequently, we concentrate on V_B and V_C monovacancies for main-group metal-based SACs.

  As shown in Figs. 1a and b, nine main-group metals (i.e., Mg, Ca, Al, Ga, In, Ge, Sn, Sb, and Bi) are chosen and embedded into V_B and V_C monovacancies, forming the structures of M@V_B and M@V_C, respectively. To quantitatively evaluate the stability of these M@V_B and M@V_C systems, we calculate the binding energy (E_b) as shown in Fig. 1c, and find that all systems exhibit negative E_b values ranging from −6.13 eV to −0.67 eV, indicating the strong binding interactions between main-group metals and BC₃ monolayer. Additionally, considering the large cohesive energy (E_coh) of metal bulk, the experimental feasibility of main-group metal-based SACs needs to be further assessed. Our results shown in Fig. 1c reveal that nine candidate systems (i.e., Al@V_B, Al@V_C, Ga@V_B, Ga@V_C, In@V_B, Ge@V_B, Ge@V_C, Sn@V_B, and Sb@V_B) have smaller E_b in comparison with their corresponding E_coh, and thus satisfy the requirement E_b - E_coh < 0, indicating that the main-group metals in these systems prefer to be stably anchored on BC₃ monolayer, without aggregating into metal clusters. To further examine the thermal stabilities of the nine systems, we also perform ab initio molecular dynamics (AIMD) simulations in the NVT ensemble, with a total time scale of 5 ps at 500 K. As shown in Fig. S1 (Supporting information), the total energies of the nine systems oscillate slightly, and their corresponding geometric structures remain intact without deformation at the end of the AIMD simulation, verifying the good thermodynamic stability. Therefore, Al@V_B, Al@V_C, Ga@V_B, Ga@V_C, In@V_B, Ge@V_B, Ge@V_C, Sn@V_B, and Sb@V_B are selected as potential candidates for the subsequent NORR.

  Figure 1

  

  Figure 1.  Top view of (a) M@V_B and (b) M@V_C systems. (c) Computed binding energies (E_b) of main-group metal-embedded BC₃ monolayer and the energy difference between E_b and cohesive energies (E_coh) of main-group metals.

  DownLoad: Full-Size Img PowerPoint

  It is well-established that NO adsorption and activation, the first step of the NORR process, plays a crucial role in the following reaction pathways [70]. Therefore, we consider three initial NO adsorption configurations (i.e., N-end, O-end, and side-on) for the above screened nine systems, and calculate their adsorption free energies (ΔG_^*NO) as shown in Fig. 2a. For all candidate systems, we can find that NO molecule is favored to adsorb via the N-end configuration, while the side-on adsorption is unstable and will be transformed into the N-end configuration after fully structural relaxation. To ensure the effective activation of adsorbed NO molecule, spontaneous chemisorption (ΔG_^*NO < 0) is considered a necessary prerequisite for the NORR. Evidently, the Sb@V_B system should be filter out from the NORR candidates because of the weak physisorption of NO (ΔG_^*NO = 0.28 eV). The remaining eight candidates present sufficient negative ΔG_^*NO values, ranging from −0.96 eV to −3.37 eV for the N-end configuration, which indicates that NO molecule would spontaneously chemisorb on these systems. Meanwhile, as shown in Fig. 2b, the chemisorbed NO molecules have considerable N–O bond elongation from 1.17 Å (free NO molecule) to 1.19 - 1.21 Å, indicating the strong activation of NO molecule. To evaluate the competitive effects of NO adsorption, we further calculate the adsorption free energies of N₂ (ΔG_^*N2) and NH₃ (ΔG_^*NH3), and compare with NO (ΔG_^*NO). As shown in Fig. S2 (Supporting information), ΔG_^*NO is more negative than ΔG_^*N2 and ΔG_^*NH3 for the M@V_B and M@V_C systems, indicating the NO molecule could be preferentially adsorbed on them.

  Figure 2

  

  Figure 2.  (a) Adsorption free energies (ΔG_^*NO) of NO on M@V_B and M@V_C systems with N-end and O-end configurations. (b) Optimized structures of NO adsorption on M@V_B and M@V_C systems through the N-end configuration, where the N—O bond lengths are labelled.

  DownLoad: Full-Size Img PowerPoint

  To elucidate the underlying mechanism of NO adsorption and activation on these main-group metal-embedded BC₃ systems, we further calculate the partial density of states (PDOS) and charge density differences to investigate the interaction between main-group metal atom and NO molecule. Take for instance the Al@V_B system illustrated in Fig. 3a. When a NO molecule chemisorbs on it, the π^* states of NO would interact with the Al-3p states, resulting in the appearance of occupied and unoccupied p-π^* states around the Fermi level (E_F). Additionally, due to the hybridization between the occupied Al-3p states and NO-σ states below the E_F, the occupied p-σ states can be observed. As displayed in Fig. 3b, we can see not only a significant charge transfer with 0.45 |e| but also bidirectional charge transfer, namely, both electron accumulation (yellow color) and depletion (cyan color) on the NO molecule and Al atom. Thus, as illustrated in Fig. 3c, the interaction nature between Al and NO can be attributed to the "acceptance-donation" of electrons analogous to that in NRR [71,72]. The Al atom can accept lone-pair electrons of NO and simultaneously donate electrons back to the empty π^* orbitals of NO Similarly, as shown in Figs. S3-S9 (Supporting information), such electrons transfer mechanism can also be verified in other main-group metal-embedded BC₃ systems.

  Figure 3

  

  Figure 3.  (a) Partial density of states (PDOS) of Al-3p orbitals for Al@V_B, NO-2p orbitals for free NO molecule, and Al-3p and NO-2p orbitals for NO adsorbed on Al@V_B. The Fermi level is set to be 0 eV. (b) Charge density difference of NO molecule adsorbed on Al@V_B through the N-end configuration, where the positive and negative charges are shown in yellow and cyan, respectively. The isosurface value is set to be 0.005 e/ų. (c) Schematic mechanism of the interaction between main-group metal and NO molecule.

  DownLoad: Full-Size Img PowerPoint

  Having confirmed the effective activation of NO on the eight main-group metal-embedded BC₃ systems, we then move on to examine their catalytic performances of NORR toward NH₃ synthesis. Depending on the different adsorption configuration of NO, three possible reaction pathways shown in Fig. 4a, named as N-end, O-end, and side-on, are taken into consideration based on previous studies [73,74]. Noted that we only consider the N-end pathway for these systems since the N-end adsorption configuration is energetically more favorable. Subsequently, we optimize the geometric configurations of all reaction intermediates during the NORR process, and calculate the Gibbs free energy changes of each elementary step. The obtained NORR free energy diagrams for the eight main-group metal-embedded BC₃ systems are illustrated in Fig. 4 and Fig. S10 (Supporting information), and the corresponding intermediate structures are given in Figs. S11-S18 (Supporting information). There are two possible intermediates (i.e., ^*HNO and ^*NOH) for the first hydrogenation elementary step. For all of the eight main-group metal-embedded BC₃ systems, ^*HNO is preferred with respect to ^*NOH. Subsequently, these systems exhibit distinct NORR paths and potential-determining steps (PDS). Taking Al@V_B, Ga@V_B, and Ge@V_C as example, the optimum NORR pathway for Al@V_B shown in Fig. 4b can be summarized as NO(g) → ^*NO → ^*HNO → ^*HNOH → ^*H₂NOH → ^*NH₂ → ^*NH₃, with the PDS being the second hydrogenation step of ^*HNO → ^*HNOH. In terms of Ga@V_B, Fig. 4c shows that the second hydrogenation step prefers to form ^*H₂NO instead of ^*HNOH, and the PDS is identified as the hydrogenation of ^*HNO to ^*H₂NO. While for Ge@V_C shown in Fig. 4d, the reaction path NO(g) → ^*NO → ^*HNO → ^*HNOH → ^*NH → ^*NH₂ → ^*NH₃ is predicted to be the preferable one with the PDS being the hydrogenation of NO to ^*HNO.

  Figure 4

  

  Figure 4.  (a) Schematic illustration of the NORR pathways toward NH₃. Computed Gibbs free energy diagrams of the NORR on (b) Al@V_B, (c) Ga@V_B and (d) Ge@V_C systems. (e) Maximum free energy change (ΔG_max-NH3) of the remaining eight systems via the respective most favorable NORR pathway.

  DownLoad: Full-Size Img PowerPoint

  As an important index, the maximum free energy change (ΔG_max-NH3) along the most favorable reaction pathway, are usually used to estimate the NORR activity [45]. Here, ΔG_max-NH3 < 0 is selected as screening benchmark for targeting at finding potential NORR catalysts. As shown in Fig. 4e, the ΔG_max-NH3 values for Al@V_C and Sn@V_B are 0.04 and 0.10 eV, respectively. Thus, we rule out them from NORR candidates. It is quite interesting that the remaining six systems, including Al@V_B (−0.23 eV), Ga@V_B (−0.28 eV), Ga@V_C (−0.16 eV), In@V_B (−0.03 eV), Ge@V_B (−0.12 eV), and Ge@V_C (−0.25 eV), have significantly negative ΔG_max-NH3 values, indicating that the NORR process can proceed spontaneously on them without using any applied voltage. Furthermore, taking Al@V_B as an example, we investigate the impact of solvation correction on the free energy during the NORR process by using an implicit solvation model as implemented in VASPsol [75]. As shown in Fig. S19 (Supporting information), the PDS and ΔG_max-NH3 of Al@V_B are insensitive to the solvation effects. In this regard, the solvation effects are not considered in this work.

  In addition to the catalytic activity, the NH₃ selectivity is also an essential factor for achieving high catalytic performance [76,77]. At high NO coverage, N₂O and N₂ could be produced as main byproducts during the NORR process. To this end, we further investigate the electrochemical NO-to-N₂O/N₂ conversion following the reaction pathways illustrated in Fig. 5a, which starts with the 2O-side and 2N-side modes of NO dimer (labelled as ^*N₂O_{2–O and ^*N2}O_{2–N), respectively. The free energy diagrams of N2}O and N₂ formation on the remaining six systems (Al@V_B, Ga@V_B, Ga@V_C, In@V_B, Ge@V_B, and Ge@V_C) are shown in Figs. 5b–g, and the optimized atomic structures of intermediates are given in Figs. S20-S25 (Supporting information). One can see that converting ^*N₂O₂–O to ^*O + N₂O is the PDS in the NORR process toward N₂O synthesis, which demand the energy inputs of 0.83, 0.49, 0.58, 0.68, 0.47, and 0.27 eV for the Al@V_B, Ga@V_B, Ga@V_C, In@V_B, Ge@V_B, and Ge@V_C systems, respectively. On the other hand, the protonation of ^*N₂O to ^*N₂OH in the NORR process toward N₂ synthesis have considerably large ΔG values of 0.17, 0.40, and 0.12 eV for the Al@V_B, Ga@V_B, and Ge@V_C systems, respectively. Noted that the NO-to-N₂ conversion on the Ga@V_C, In@V_B, and Ge@V_B systems is an exothermic process without any extra energy, indicative of the spontaneous N₂ formation. Hence, the Al@V_B, Ga@V_B, and Ge@V_C systems can effectively prevent the byproduct reactions of N₂O and N₂ formation, indicating high selectivity of NORR toward NH₃ synthesis.

  Figure 5

  

  Figure 5.  (a) Schematic illustration of the NORR pathways toward N₂O/N₂ byproducts under high NO converge. Computed Gibbs free energy diagrams of electrochemical NO-to-N₂O/N₂ conversion on (b) Al@V_B, (c) Ga@V_B, (d) Ga@V_C, (e) Ge@V_B, (f) In@V_B, and (g) Ge@V_C systems.

  DownLoad: Full-Size Img PowerPoint

  Another competing reaction to the NH₃ selectivity is the H₂ production through HER, which normally affects the Faradaic efficiency of NORR. Therefore, we compare the competition between H₂ and NH₃ production in the NORR process for the remaining three candidates. As shown in Fig. S26 (Supporting information), the calculated ΔG_^*H values are −0.25 eV for Al@V_B, −0.39 eV for Ga@V_B, and −1.65 eV for Ge@V_C. Obviously, the free energy barriers for the HER are considerable larger than the PDS barriers for the NORR. This indicates that the HER process can be greatly suppressed on the three systems, ensuring the excellent NORR selectivity. According to these calculations, the Al@V_B, Ga@V_B, and Ge@V_C systems are finally screened out as potential main-group metal-based SACs for NORR toward NH₃ synthesis due to the superior catalytic activity and selectivity.

  In conclusion, by performing first-principles calculations, we comprehensively investigate main-group metal-embedded BC₃ monolayer as SACs for the NORR toward NH₃ synthesis. Our calculations reveal that nine systems (i.e., Al@V_B, Al@V_C, Ga@V_B, Ga@V_C, In@V_B, Ge@V_B, Ge@V_C, Sn@V_B, and Sb@V_B) out of 18 candidates satisfy the relationship E_b - E_coh < 0, implying the high thermodynamic stability. Besides the Sb@V_B, NO molecule can be stably adsorbed and effectively activated on the remaining eight systems via the "acceptance-donation" mechanism of electron. Further scrutiny of the catalytic activity, six candidate systems including Al@V_B, Ga@V_B, Ga@V_C, In@V_B, Ge@V_B, and Ge@V_C are considered as excellent NORR electrocatalysts with ΔG_max-NH3 < 0, since the electrochemical NO-to-NH₃ conversion can proceed spontaneously without any limiting potential. In addition, after estimating the suppression of byproducts including N₂O, N₂, and H₂, Al@V_B, Ga@V_B, and Ge@V_C are expected to possess high NH₃ selectivity, and finally screened out as promising main-group metal-based SACs for NORR. Consequently, these theoretical findings provide valuable insights for designing efficient NORR electrocatalysts and expanding SACs to main-group metal-based systems.

  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.

  CRediT authorship contribution statement

  Jiajun Wang: Writing – original draft, Project administration, Conceptualization. Chen Sun: Resources, Data curation. Li Sheng: Data curation. Zhiwen Zhuo: Investigation. Shujuan Li: Writing – review & editing, Supervision. Jiayao Wang: Validation, Investigation. Weiyi Wang: Writing – review & editing, Supervision, Funding acquisition. Jinbo Sun: Visualization. Juqian Yang: Writing – original draft. Ke Xu: Funding acquisition. Shulai Lei: Supervision, Funding acquisition.

  Acknowledgments

  This work is supported by the Postdoctoral Fellowship Program (No. GZC20232540), the General Program of China Postdoctoral Science Foundation (No. 2024M753088), Natural Science Foundation of Hubei Province (No. 2022CFC030), the Science and Technology Research Project of Hubei Provincial Department of Education (No. D20212603) and Hubei University of Arts and Science (No. 2020kypytd002). Hefei Advanced Computing Center is acknowledged for computational support.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.110974.

  
  
• 1. [1]

     T.Y. Dai, T.H. Wang, Z. Wen, et al., Adv. Funct. Mater. 34 (2024) 2400773. doi: 10.1002/adfm.202400773

  2. [2]

     Y. Yang, J. Wang, Y. Shu, et al., Phys. Chem. Chem. Phys. 24 (2022) 8591–8603. doi: 10.1039/d1cp05442b

  3. [3]

     C. He, J. Wang, L. Fu, et al., Chin. Chem. Lett. 35 (2024) 109037. doi: 10.1016/j.cclet.2023.109037

  4. [4]

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

  5. [5]

     J. Wang, M. Shi, G. Yi, et al., Mol. Catal. 511 (2021) 111726.

  6. [6]

     J. Wu, L. Yang, X. Liu, et al., Chin. Chem. Lett. 34 (2023) 107659. doi: 10.1016/j.cclet.2022.07.002

  7. [7]

     J. Wang, M. Shi, G. Yi, et al., Chin. Chem. Lett. 33 (2022) 4623–4627. doi: 10.1016/j.cclet.2021.12.040

  8. [8]

     Y. Ren, S. Li, C. Yu, et al., J. Am. Chem. Soc. 146 (2024) 6409–6421. doi: 10.1021/jacs.3c11676

  9. [9]

     D. Wang, X.F. Lu, D. Luan, et al., Adv. Mater. 36 (2024) 2312645. doi: 10.1002/adma.202312645

  10. [10]

      Y. Zang, Q. Wu, S. Wang, et al., Adv. Energy Mater. 14 (2024) 2303953. doi: 10.1002/aenm.202303953

  11. [11]

      X. Sun, Y. Dai, B. Huang, et al., J. Phys. Chem. Lett. 14 (2023) 11684–11690. doi: 10.1021/acs.jpclett.3c03105

  12. [12]

      C. He, J. Wang, L. Fu, et al., Chin. Chem. Lett. 33 (2022) 1051–1057. doi: 10.1016/j.cclet.2021.09.009

  13. [13]

      S. Zhao, Y. Li, Z. Guo, et al., J. Mater. Chem. A 10 (2022) 25201–25211. doi: 10.1039/d2ta06354a

  14. [14]

      C. He, R. Sun, L. Fu, et al., Chin. Chem. Lett. 33 (2022) 527–532. doi: 10.1016/j.cclet.2021.05.072

  15. [15]

      Z. Wu, Y. Liu, D. Wang, et al., Adv. Mater. 36 (2024) 2309470. doi: 10.1002/adma.202309470

  16. [16]

      L. Xiao, S. Mou, W. Dai, et al., Angew. Chem. Int. Ed. 63 (2024) e202319135. doi: 10.1002/anie.202319135

  17. [17]

      K. Chen, J. Xiang, Y. Guo, et al., Nano Lett. 24 (2024) 541–548. doi: 10.1021/acs.nanolett.3c02259

  18. [18]

      B. Qiao, A. Wang, X. Yang, et al., Nat. Chem. 3 (2011) 634–641. doi: 10.1038/nchem.1095

  19. [19]

      F. Kraushofer, G.S. Parkinson, Chem. Rev. 122 (2022) 14911–14939. doi: 10.1021/acs.chemrev.2c00259

  20. [20]

      S. Wang, X.T. Min, B. Qiao, et al., Chin. J. Catal. 52 (2023) 1–13.

  21. [21]

      J. Wang, G. Yi, S. Guo, et al., Chin. Chem. Lett. 35 (2024) 109050. doi: 10.1016/j.cclet.2023.109050

  22. [22]

      G. Xue, S. Ping, H. Ce, et al., Energy Mater. 3 (2023) 300016.

  23. [23]

      R. Cheng, C.N. Cui, Z.X. Luo, Rare Met. 43 (2024) 3810–3818. doi: 10.1007/s12598-024-02680-2

  24. [24]

      G. Song, R. Gao, Z. Zhao, et al., Appl. Catal. B: Environ. 301 (2022) 120809. doi: 10.1016/j.apcatb.2021.120809

  25. [25]

      W. Zhang, M. Xu, T. Wang, et al., Sep. Purif. Technol. 354 (2025) 129536. doi: 10.1016/j.seppur.2024.129536

  26. [26]

      S. Peng, Y. Rao, T. Li, et al., Chin. Chem. Lett. 35 (2024) 109219. doi: 10.1016/j.cclet.2023.109219

  27. [27]

      X. Yao, L. Huang, E. Halpren, et al., J. Am. Chem. Soc. 145 (2023) 26249–26256. doi: 10.1021/jacs.3c08936

  28. [28]

      L. Kong, X. Liang, M. Wang, et al., J. Colloid Interf. Sci. 647 (2023) 375–383. doi: 10.1016/j.jcis.2023.05.158

  29. [29]

      H. Niu, Z. Zhang, X. Wang, et al., Small 17 (2021) 2102396. doi: 10.1002/smll.202102396

  30. [30]

      L. Lin, L. Yan, L. Fu, et al., Fuel 308 (2022) 122068. doi: 10.1016/j.fuel.2021.122068

  31. [31]

      R. Zhao, Z. Ma, Y. Yu, et al., Nano Res. 16 (2023) 8533–8541. doi: 10.1007/s12274-023-5619-9

  32. [32]

      Y. Zang, Q. Wu, S. Wang, et al., Mater. Horiz. 10 (2023) 2160–2168. doi: 10.1039/d2mh01440h

  33. [33]

      Y. Wu, J. Lv, F. Xie, et al., J. Colloid Interf. Sci. 656 (2024) 155–167. doi: 10.3847/1538-4357/ad509f

  34. [34]

      L. Kong, M. Wang, C.M.L. Wu, ACS Mater. Lett. 6 (2024) 1711–1721. doi: 10.1021/acsmaterialslett.3c01604

  35. [35]

      H. Li, D. Wu, J. Wu, et al., Nanoscale 16 (2024) 7058–7067. doi: 10.1039/d4nr00028e

  36. [36]

      G.F. Chen, S. Ren, L. Zhang, et al., Small Methods 3 (2019) 1800337. doi: 10.1002/smtd.201800337

  37. [37]

      S. Liu, Z. Li, C. Wang, et al., Nat. Commun. 11 (2020) 938. doi: 10.1038/s41467-020-14565-w

  38. [38]

      T. Liu, X. Qu, Y. Zhang, et al., Chem. Eng. J. 457 (2023) 141187. doi: 10.1016/j.cej.2022.141187

  39. [39]

      F. Luo, A. Roy, L. Silvioli, et al., Nat. Mater. 19 (2020) 1215–1223. doi: 10.1038/s41563-020-0717-5

  40. [40]

      Q. Wang, M. Dai, H. Li, et al., Adv. Mater. 35 (2023) 2300695. doi: 10.1002/adma.202300695

  41. [41]

      L. Zhang, H. Zhou, X. Yang, et al., Angew. Chem. Int. Ed. 62 (2023) e202217473. doi: 10.1002/anie.202217473

  42. [42]

      K. Chen, Y. Zhang, J. Xiang, et al., ACS Energy Lett. 8 (2023) 1281–1288. doi: 10.1021/acsenergylett.2c02882

  43. [43]

      X. Li, G. Zhang, P. Shen, et al., Inorg. Chem. Front. 10 (2023) 280–287. doi: 10.1039/d2qi02118h

  44. [44]

      K. Chen, N. Zhang, F. Wang, et al., J. Mater. Chem. A 11 (2023) 6814–6819. doi: 10.1039/d3ta00606a

  45. [45]

      Q. Wu, B. Huang, Y. Dai, et al., npj 2D Mater. Appl. 6 (2022) 52. doi: 10.1038/s41699-022-00326-4

  46. [46]

      L. Zhao, Y. Li, G. Zhou, et al., Chin. Chem. Lett. 32 (2021) 900–905. doi: 10.1016/j.cclet.2020.07.016

  47. [47]

      H. Yanagisawa, T. Tanaka, Y. Ishida, et al., Phys. Rev. Lett. 93 (2004) 177003. doi: 10.1103/PhysRevLett.93.177003

  48. [48]

      Z. Yang, J. Ni, Appl. Phys. Lett. 100 (2012) 183109. doi: 10.1063/1.4711038

  49. [49]

      J. Li, Y. Guo, S. Ma, et al., Int. J. Hydrogen Energy 47 (2022) 24004–24013. doi: 10.1016/j.ijhydene.2022.05.154

  50. [50]

      H. Yang, C. He, L. Fu, et al., Chin. Chem. Lett. 32 (2021) 3202–3206. doi: 10.1016/j.cclet.2021.03.038

  51. [51]

      Y. Guo, X. Kang, S. Gao, et al., Phys. Chem. Chem. Phys. 25 (2023) 12420–12425. doi: 10.1039/d3cp00653k

  52. [52]

      R.P. Joshi, B. Ozdemir, V. Barone, et al., J. Phys. Chem. Lett. 6 (2015) 2728–2732. doi: 10.1021/acs.jpclett.5b01110

  53. [53]

      J. Ou, X. Kang, X. Duan, Nanoscale 14 (2022) 12823–12829. doi: 10.1039/d2nr02796h

  54. [54]

      W. Guo, S. Wang, H. Wang, et al., J. Energy Chem. 96 (2024) 336–344. doi: 10.1016/j.jechem.2024.05.002

  55. [55]

      C. Chen, B. Xiao, Z. Qin, et al., ACS Appl. Mater. Interfaces 15 (2023) 40538–40548. doi: 10.1021/acsami.3c07790

  56. [56]

      Y. Wang, N. Zhou, Y. Li, Chem. Eng. J. 425 (2021) 130631. doi: 10.1016/j.cej.2021.130631

  57. [57]

      Y. Zhou, G. Gao, W. Chu, et al., Nanoscale 13 (2021) 1331–1339. doi: 10.1039/d0nr07580a

  58. [58]

      D. McGlamery, C. McDaniel, D.M. Ladd, et al., Chem. Sci. 15 (2024) 4358–4363. doi: 10.1039/d3sc06837d

  59. [59]

      G. Kresse, J. Furthmuller, Phys. Rev. B 54 (1996) 11169. doi: 10.1103/PhysRevB.54.11169

  60. [60]

      P.E. Blöchl, Phys. Rev. B 50 (1994) 17953. doi: 10.1103/PhysRevB.50.17953

  61. [61]

      J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865. doi: 10.1103/PhysRevLett.77.3865

  62. [62]

      S. Grimme, J. Comput. Chem. 27 (2006) 1787–1799. doi: 10.1002/jcc.20495

  63. [63]

      H.J. Monkhorst, J.D. Pack, Phys. Rev. B 13 (1976) 5188. doi: 10.1103/PhysRevB.13.5188

  64. [64]

      L. Pan, X. Kang, S. Gao, et al., Phys. Chem. Chem. Phys. 26 (2024) 1011–1016. doi: 10.1039/d3cp04660e

  65. [65]

      G. Xing, S. Liu, J.Y. Liu, Mol. Catal. 553 (2024) 113771.

  66. [66]

      S. Li, M. Shi, J. Yu, et al., Chin. Chem. Lett. 32 (2021) 1977–1982. doi: 10.1016/j.cclet.2020.09.056

  67. [67]

      Y. Zhou, W. Chu, F. Jing, et al., Appl. Surf. Sci. 410 (2017) 166–176. doi: 10.1089/scd.2016.0162

  68. [68]

      D.H. Lim, J. Wilcox, J. Phys. Chem. C 115 (2011) 22742–22747. doi: 10.1021/jp205244m

  69. [69]

      L.J. Ma, J. Wang, J. Jia, et al., Int. J. Hydrogen Energy 47 (2022) 28423–28433. doi: 10.1016/j.ijhydene.2022.06.152

  70. [70]

      S.J. Qian, H. Cao, Y.G. Wang, et al., J. Am. Chem. Soc. 146 (2024) 12530–12537. doi: 10.1021/jacs.4c00827

  71. [71]

      M. He, X. Chen, Y. Zhou, et al., J. Phys. Chem. Lett. 14 (2023) 7100–7107. doi: 10.1021/acs.jpclett.3c01576

  72. [72]

      C. Ling, X. Niu, Q. Li, et al., J. Am. Chem. Soc. 140 (2018) 14161–14168. doi: 10.1021/jacs.8b07472

  73. [73]

      Y. Zang, Q. Wu, S. Wang, et al., J. Phys. Chem. Lett. 13 (2022) 527–535. doi: 10.1021/acs.jpclett.1c03938

  74. [74]

      Y. Sun, Z. Wang, Y. Liu, et al., Inorg. Chem. Front. 10 (2023) 2677–2688. doi: 10.1039/d3qi00225j

  75. [75]

      K. Mathew, R. Sundararaman, K. Letchworth-Weaver, et al., J. Chem. Phys. 140 (2014) 084106. doi: 10.1063/1.4865107

  76. [76]

      S. Liu, G. Xing, J.Y. Liu, Appl. Surf. Sci. 611 (2023) 155764. doi: 10.1016/j.apsusc.2022.155764

  77. [77]

      M. Tursun, C. Wu, Inorg. Chem. 61 (2022) 17448–17458. doi: 10.1021/acs.inorgchem.2c02247

• 

  Figure 1  Top view of (a) M@V_B and (b) M@V_C systems. (c) Computed binding energies (E_b) of main-group metal-embedded BC₃ monolayer and the energy difference between E_b and cohesive energies (E_coh) of main-group metals.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Adsorption free energies (ΔG_^*NO) of NO on M@V_B and M@V_C systems with N-end and O-end configurations. (b) Optimized structures of NO adsorption on M@V_B and M@V_C systems through the N-end configuration, where the N—O bond lengths are labelled.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Partial density of states (PDOS) of Al-3p orbitals for Al@V_B, NO-2p orbitals for free NO molecule, and Al-3p and NO-2p orbitals for NO adsorbed on Al@V_B. The Fermi level is set to be 0 eV. (b) Charge density difference of NO molecule adsorbed on Al@V_B through the N-end configuration, where the positive and negative charges are shown in yellow and cyan, respectively. The isosurface value is set to be 0.005 e/ų. (c) Schematic mechanism of the interaction between main-group metal and NO molecule.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Schematic illustration of the NORR pathways toward NH₃. Computed Gibbs free energy diagrams of the NORR on (b) Al@V_B, (c) Ga@V_B and (d) Ge@V_C systems. (e) Maximum free energy change (ΔG_max-NH3) of the remaining eight systems via the respective most favorable NORR pathway.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) Schematic illustration of the NORR pathways toward N₂O/N₂ byproducts under high NO converge. Computed Gibbs free energy diagrams of electrochemical NO-to-N₂O/N₂ conversion on (b) Al@V_B, (c) Ga@V_B, (d) Ga@V_C, (e) Ge@V_B, (f) In@V_B, and (g) Ge@V_C systems.

  下载: 全尺寸图片 幻灯片
•