Figure 3. (a) Schematic illustration of the formation of NC-CNTs (Ni). (b) Free energy diagram of Ni@N3(pyrrolic) and Ni@N3(pyridinic) for electrochemical CO2 reduction to CO. Reproduced with permission.[65] Copyright 2019, Wiley-VCH. (c) Scheme for the synthesis of C-TpDt-Ni (carbon atoms are shown in grey, nitrogen in blue, oxygen in red, Ni in green). (d) High-resolution HAADF-STEM image of C-TpDt-Ni-900. (e) The absolute values of Fourier-transformed (FT)-EXAFS spectra of specimens. (f) The FECO of various COF-derived Ni-N-C and reference catalysts. Reproduced with permission.[67] Copyright 2022, Wiley-VCH.
3.3 Coexistence of Ni NPs and Ni SA
Although single-atom catalysts contribute the greatest atomic utilization and higher unit activity, metal loading concentrations are usually very low, resulting in the insufficient apparent activity.[73,74] In contrast, metal NPs with easily adjustable electronic properties due to their variable crystal structure and surface are endowed with appreciable apparent electrocatalytic activity, but low atomic utilization reduces the turnover frequency (TOF).[75] The coexistence of SA-substrate interactions and metal-substrate interactions confers higher freedom degrees of hybrid systems in electron modulation, creating more positive sites with diverging electron properties.[76,77] Therefore, the combination of metal NPs and SA in the system is conducive to develop advantages and eliminate disadvantages for metal NPs and single atom. Based on this result, relevant research has been carried out in sixpenny Ni-N-C materials.
For instance, Deng and Ye have jointly reported a hybrid material composed of metallic Ni cores and Ni-N-C layers (Ni-NC@Ni) by low temperature vapor deposition strategy (Figure 4 a).[78] The lamellar open nanostructures have abundant isolated Ni-N in the carbon layer (about 4.23 at%) (Figure 4b,4c), more open space to promote mass transfer and excellent conductivity, showing a high FECO (87%) and a current density of 14.8 mA cm−2. This provides an effective way to improve the mass transport, electronic conductivity and exposure of the catalytic active center. However, whether the dominant active site is Ni single atoms or Ni NPs or both has not been conclusively demonstrated.
Based on the above arguments, Liu’s team synthesized several representative catalysts (Ni-NG), including Ni single atoms, nitrogen-doped carbon-encapsulated Ni NPs, and both. Surprisingly, the activity of Ni-NG catalyst, which was treated with acid to remove the embedded Ni NPs, was not decreased for the ECR performance (Figure 4d).[79] The reason is that the N-doped carbon-encapsulated Ni NPs catalyst contained the unexpressed Ni single atoms, which played a dominant role in ECR. Theoretical calculations show that when three or more layers of N-doped graphene carbon were coated by Ni NPs (3NGR@Ni), the charge of surface carbon atoms can hardly be regulated (Figure 4e). Therefore, the free energy of intermediates in CO2 reduction on carbon layer before and after encapsulating Ni NPs did not change significantly. And the higher ΔGCOOH* of NGR@Ni and 3NGR@Ni showed the insufficient CO2 conversion ability. Moreover, the thermodynamic limiting potentials for ECR and HER (UL(CO2) and UL(H2)) revealed that N-doped carbon-encapsulated Ni NPs (NGR@Ni) possessed high catalytic selectivity for HER (Figure 4f), again indicating that Ni NPs are not the ECR active site. Lin et al. successfully achieved the ratio tunability of single-atom Ni and Ni NPs (1:0 ~ 1:5) on N-doped carbon nanorods via controlling the HCl etching process, to adjust the ratio of CO and H2 in syngas product.[80]The high H2 precipitation ratio of 90% and CO selectivity of as low as 10% of Ni NPs further indicated that Ni NPs favors HER. By contrast, the sample containing only single atom Ni-N site displayed the highest CO selectivity, more than 95%.
Although the above research results provide sufficient evidence for determining Ni single atoms as the active species in ECR, it does not deny that the existence of Ni NPs can still have a positive effect on the conductivity of material and the modulation of active site/carrier from the edge/surface metal Ni. Recently, Hu and coworkers reported a cooperative Ni single-atom-on-NPs (NiSA/NP) though the simple direct solid-state pyrolysis of melamine.[81] The NiSA/NP provided an industrial current density of 131 mA cm−2at −1.0 V in H-cell over CO (Figure 4g). Based on the characterization and experimental results, Ni-N3@Ni model was constructed and corresponding theoretical calculation was performed. The results suggest that Ni NPs can afford the electrons for Ni−N−C on the surface to achieve electronic modulation of Ni-N3. The analysis of Bader charge further shows that the introduction of Ni (111) (Δq=0.013 e) causes the Ni single atom to gather more electrons (Figure 4h), again indicating the microscopic modulation effect of Ni NPs on the single atom. Afterwards, the calculated free energy value (0.53eV) of COOH* at NiN3@Ni, is lower than that of NiN3 (0.70 eV), suggesting that COOH* possesses stronger binding strength with NiN3@Ni. The above results demonstrate that the presence of Ni NPs optimizes the charge distribution of Ni SA sites and enhances the ECR activity. In addition, Hou et al. proposed a possible ECR mechanism on the atomically dispersed Ni-Nx site in carbon materials coated at the Ni NPs (Figure 4i), and the Ni−N−C catalyst (Ni@NiN4CM) was synthesized by the carbonization of nickel nitrate and o-phenylenediamine (Figure 4j), which reveals the role of active components in the protonation process from the perspective of kinetics.[82] Also, the experiments and theoretical calculations have confirmed that Ni NPs can promote the ionization and capture of protons, accelerate the capture of protons between the active center of Ni-N4 and the adsorption intermediates, thus optimizing the ECR reaction kinetics.
Undoubtedly, single-atom Ni-Nx (x = 1, 2, 3, 4) is the main active site for ECR reaction. However, the coexistence of Ni NPs with SA offering the positive/negative impact on the activity needs to be analyzed in combination with specific structures in ECR. The current structural analysis is incomplete. The differences of experimental and theoretical results also appear in different preparation methods, such as the coordination mode of Ni NPs and Ni single atom, the electron transport capacity of the material itself, the apparent active site density and the theoretical model.