English

• As one of the most attractive approaches to achieve carbon peaking and carbon neutralization, electrochemical CO₂ reduction reaction (CO₂RR) receives widespread attention [1-4]. Generally, products of CO₂RR can be divided into C₁ (CO, HCOO⁻, etc.) [5-7] and C₂₊ (C₂H₄, C₂H₅OH, etc.) [8-11]. Technical and economic analysis indicates that formic acid/formate are the most valuable product, which are important industrial raw materials and fuels for fuel cells [12-14]. Compared with other metallic materials, such as Pd, In, and Sn, bismuth (Bi)-based materials are potential candidates due to the effective inhibition of competitive hydrogen evolution reaction during CO₂RR [15].

  Recently, some modulation strategies have been applied to improve the performance of CO₂RR [16,17], such as the two-dimensional transformation [18], interface construction [19], defect engineering [20,21], and element doping [22]. For instance, catalyst reconstruction could effectively modulate the local electronic structure of active sites, which was beneficial for the adsorption and conversion of CO₂ molecules [23]. However, the reconstruction process of BiVO₄ with oxygen vacancies (O_vs) was rarely noted in electrocatalytic CO₂ reduction.

  Herein, BiVO₄ was synthesized by a relatively simple process (no strong acids and strong reductant) (Table S1 in Supporting information). As a precatalyst, the optimized sample (BV-2) reconstructed into active metal Bi during the reaction process, which presented the maximum formate selectivity of 97.45% at −0.9 V and a large partial current density of −45.82 mA/cm² at −1.2 V. In two-electrode flow cells, the formate selectivity maintained ~95% from 20 mA/cm² to 180 mA/cm², and the maximum partial current density could reach 175.23 mA/cm² at 2.8 V.

  As shown in Fig. 1a, a series of BiVO₄ (BV) were synthesized by a hydrothermal strategy. The difference among them is the amount of sodium dodecyl sulfonate (SDS) during prepared process. BV-0, BV-1, BV-2 and BV-3 were constructed via adding 0, 0.1, 0.2 and 0.3 g SDS, respectively. It is clear that there was no significant difference in the color of the series of samples as shown in Fig. S1. The X-ray diffraction (XRD) patterns in Fig. 1b showed that the phase of BiVO₄ was multiple crystal structures. BV-0, BV-1 and BV-2 were consistent with monoclinic BiVO₄ (m-BiVO₄, PDF#14–0688) and tetragonal BiVO₄ (t-BiVO₄, PDF#14–0133). When adding 0.3 g SDS, additional crystal phase ascribed to tetragonal structure (PDF#49–0198) in XRD of BV-3. The highest diffraction peak at 24.4° in XRD of BV-2 was corresponding to the BiVO₄ (200) tetragonal structure (PDF#14–0133). However, the highest diffraction peak at 28.9° of other samples was BiVO₄ (121), which was consistent with monoclinic structure (PDF#14–0688).

  Figure 1

  

  Figure 1.  (a) Schematic illustration for the synthesis of BV-2. (b) XRD patterns of different samples. (c) Raman results of BV-2 and BV-0.

  DownLoad: Full-Size Img PowerPoint

  Fourier transform infrared (FT-IR) spectra (Fig. S2 in Supporting information) were aimed to explore the functional groups in BiVO₄. The strong peak of 749 cm⁻¹ was assigned to the asymmetric stretching of VO₄³⁻, revealing successful synthesis of BiVO₄ [24]. Raman spectra (Fig. 1c) were investigated to verify the structure of BiVO₄ [25]. The bands at 211, 326, 368 and 827 cm⁻¹ were assigned to the vibrations of monoclinic BiVO₄ [26]. The additional shoulder band at 853 cm⁻¹ was ascribed to the symmetric stretching mode of V-O band for tetragonal BiVO₄ [27]. It meant that the prepared samples were the mixed crystalline structure of BiVO₄, which was in compliance with XRD results. The band at 246 cm⁻¹ of BV-2 was attributed to O_vs. We suppose that adding SDS could introduce O_vs in BiVO₄ [28].

  To investigate the chemical valence states and compositions of BV species, the X-ray photoelectron spectroscopy (XPS) measurement was employed. The survey spectra were shown in Fig. S3 (Supporting information), indicating co-existences of Bi, V and O elements. As shown in Bi 4f XPS spectra (Fig. 2a), for BV-0, two peaks at 164.2 and 158.9 eV were corresponded to Bi 4f_5/2 and Bi 4f_7/2, respectively, implying the valence of Bi was +3. Compared with BV-0, characteristic peaks of BV-2 showed a slight shift to low binding energy. Moreover, two characteristic peaks of BV-2 in V 2p spectra also showed a slight shift (~0.1 eV) to low binding energy (Fig. S4 in Supporting information). As shown in Fig. 2b, the high-resolution O 1s XPS spectra were deconvoluted into three peaks with the lattice oxygen, surface hydroxyls and adsorbed oxygen, corresponding to the binding energies of 533, 530.9 and 529.4 eV, respectively. The surface hydroxyls (OH) were related to the dissociative adsorption of water with O_vs. BV-2 displayed more OH, indicating more O_vs, which were calculated based on the area of peaks (Table S2 in Supporting information) [29]. The slight shift of Bi 4f and V 2p XPS spectra, was possibly due to the different chemical environment caused by O_vs.

  Figure 2

  

  Figure 2.  (a) Bi 4f XPS results and (b) O 1s XPS results of different samples. (c) EPR spectra of BV-0 and BV-2. (d) EDS mapping images of BV-2.

  DownLoad: Full-Size Img PowerPoint

  In order to further confirm the existence of O_vs, electron paramagnetic resonance (EPR) was employed (Fig. 2c). Paramagnetic signal at g = 2.004 was characteristics of O_vs. Compared to BV-0, the peak intensity of BV-2 was enhanced, which might improve the mobility of electrons. The crystal structure was disclosed by high-resolution transmission electron microscope (HRTEM). In Fig. S5 (Supporting information), BV-2 revealed lattices fringes with interatomic distance of 0.269 nm for the (112) plane of t-BiVO₄ and 0.308 nm for the (121) plane of m-BiVO₄. The interface (labeled by pink line) between t-BiVO₄ and m-BiVO₄ was observed. Meanwhile, BV-0 exhibited the (121) plane of m-BiVO₄ with lattices fringes of 0.308 nm (Fig. S6 in Supporting information). No planes of t-BiVO₄ were found. The energy dispersive spectra (EDS) elemental mapping images of BV-2 and BV-0 were presented in Fig. 2d and Fig. S7 (Supporting information). The results revealed the uniform distribution of Bi, V and O elements.

  To evaluate the CO₂RR performance of electrocatalysts, H-type cells were applied with 0.5 mol/L KHCO₃ as electrolyte [30]. All the potential was referred to the reversible hydrogen electrode (RHE). The commercial Bi powder (C-Bi) with the size of 200 mesh was selected for comparison. Firstly, we tested the linear sweep voltammetry (LSV) curves at different saturated atmosphere (Ar and CO₂). The difference of curves indicated their CO₂ conversion capacity (Fig. S8 in Supporting information) [31]. As displayed by LSV curves in CO₂-saturated atmosphere in Fig. 3a, the current density of BV-2 sharply increased and markedly reached −49.16 mA/cm² at −1.2 V, which was larger than that of other samples. To figure out their activity and selectivity towards CO₂RR, the constant voltage electrolysis was carried out (Fig. S9 in Supporting information). HCOO⁻ was the main products from CO₂RR for BV species (Fig. S10 in Supporting information). In Fig. 3b, the formate Faradaic efficiency (FE_HCOO-) of BV-2 was greater than 90% from −0.8 V to −1.2 V, which could arrive 97.45% at −0.9 V. Fig. 3c showed the comparison of FE_HCOO- with other samples (BV-0, BV-1, BV-3 and C-Bi), founding that BV-2 displayed the highest FE_HCOO- and widest voltage window. Also, BV-2 achieved a maximum J_HCOO- of −45.82 mA/cm²@−1.2 V, while the J_HCOO- of C-Bi was −13.42 mA/cm²@−1.2 V (Fig. 3d). BV-2 at −1.0 V could reach a stability of 40 h and FE_HCOO- of 85.8% (inset of Fig. 3d). Fig. 3e and Table S3 (Supporting information) gave a summary of up-to-date Bi-based catalysts for J_HCOO- in H-type cells. BV-2 outperformed most of reported electrocatalysts for CO₂RR.

  Figure 3

  

  Figure 3.  (a) LSV curves of different samples at CO₂-saturated atmosphere, inset: LSV curves of BV-2 under different atmosphere. (b) FEs with different products of BV-2. (c) FE_HCOO-. (d) J_HCOO-, inset: the stability test of BV-2 at −1.0 V in H-type cells. (e) The performance comparison of up-to-date Bi-based catalysts for CO₂RR in H-type cells.

  DownLoad: Full-Size Img PowerPoint

  In-situ Raman spectra were further investigated (Figs. 4a and b). As the electrolysis proceeded, Bi-Bi bands were appeared at 63 and 84 cm⁻¹ [32]. It meant that BV-2 acted as a precatalyst. During the electrolysis process, BV-2 underwent surface reconstruction to produce metal Bi, which consisted with the XRD and XPS results of BV-2 after reaction (Figs. S11 and S12 in Supporting information). Then the peak at 1540 cm⁻¹ was stronger with prolonged time, which could be assigned to asymmetric C—O stretching vibration of OCHO* intermediate [33]. The reconstruction process induced a redistribution of charges, promoting the conversion of CO₂ molecules. Then we explored the electrochemical active surface areas (ECSA), which was evaluated by the electric double-layer capacitance (C_dl) method (Fig. S13 in Supporting information) [34,35]. As shown in Fig. S14a (Supporting information), BV-2 had the larger C_dl with 1090 µF/cm² than other samples, corresponding to larger ECSA. Also, C_dl of BV-0 (497 µF/cm²), BV-1 (585 µF/cm²) and BV-3 (652 µF/cm²) were greater than that of C-Bi (304 µF/cm²). In Fig. S14b (Supporting information), the Tafel slope was ~178.1 mV/dec of BV-2, significantly lower than ~201.7 mV/dec for BV-0, indicating faster reaction kinetics [36,37]. The electrochemical impedance spectroscopy (EIS) was conducted. The EIS results and the corresponding equivalent circuit was presented in Fig. 4c. BV-2 demonstrated the smallest charge-transfer resistance than other contrast samples (Table S4 in Supporting information), enabling the largest current density of formate.

  Figure 4

  

  Figure 4.  (a, b) In-situ Raman spectra recorded at −1.0 V. (c) EIS measurement, (d) FE_HCOO- and J_HCOO- of BV-2 in flow cells.

  DownLoad: Full-Size Img PowerPoint

  Furthermore, the CO₂RR performance in flow cells was evaluated [38]. Nickle foam was used as anode and BV-2 supported on gas diffusion layer was used as cathode. When changing the saturated atmosphere from Ar to CO₂, the current density in LSV curves was increased, indicating that excellent CO₂ activity (Fig. S15 in Supporting information) [39]. The maximum current density arrived 183.94 mA/cm² at 2.8 V, which was closed to industrial-scale current density [40]. The J_HCOO- of BV-2 could reach 175.23 mA/cm² at 2.8 V as well as the FE_HCOO- remain around 95% from 20 mA/cm² to 180 mA/cm² (Fig. 4d and Fig. S16 in Supporting information).

  In summary, BiVO₄ with O_vs was constructed through a relatively simple synthesis approach, which acted as a precatalyst for CO₂RR. The BV-2 achieved high FE_HCOO-, large J_HCOO- of −45.82 mA/cm² (−1.2 V) in H-type cells and near industrial current density in flow cells.

  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 financially supported by the Fundamental Research Funds for the Central Universities of Central South University (No. 2022ZZTS0579).

  
  
• 1. [1]

     Y.C. Wang, Q.C. Wang, J. Wu, et al., Nano Energy 103 (2022) 107815. doi: 10.1016/j.nanoen.2022.107815

  2. [2]

     Y.C. Wang, Y. Liu, W. Liu, et al., Energy Environ. Sci. 13 (2020) 4609. doi: 10.1039/D0EE02833A

  3. [3]

     Y.F. Jia, F. Li, K. Fan, L.C. Sun, Adv. Powder Mater. 1 (2022) 100012. doi: 10.1016/j.apmate.2021.10.003

  4. [4]

     M.J. Liu, L.S. Zhan, Y.C. Wang, et al., J. Mater. Sci. Technol. 165 (2023) 235. doi: 10.1201/9781003349563-35

  5. [5]

     Y. Chen, F. Hu, Y.N. Hao, et al., Nano Res. 15 (2022) 3283. doi: 10.1007/s12274-021-3978-7

  6. [6]

     Y.N. Hao, Y.J. Sun, H. Wang, et al., Electrochim. Acta 449 (2023) 142213. doi: 10.1016/j.electacta.2023.142213

  7. [7]

     M.J. Liu, Y.C. Wang, T.T. Yu, et al., Sci. Bull. 68 (2023) 1238. doi: 10.11834/jrs.20222008

  8. [8]

     H.Y. Jing, P. Zhu, X.B. Zheng, et al., Adv. Powder Mater. 1 (2022) 100013. doi: 10.1016/j.apmate.2021.10.004

  9. [9]

     B.B. Wang, S.H. Chen, Z.D. Zhang, D.S. Wang, Smart Mat 3 (2022) 84. doi: 10.1002/smm2.1101

  10. [10]

      H.H. Ou, G.S. Li, W. Ren, et al., J. Am. Chem. Soc. 144 (2022) 22075. doi: 10.1021/jacs.2c09424

  11. [11]

      J.X. Wang, D.N. Deng, Q.M. Wu, et al., ACS Nano 17 (2023) 18653. doi: 10.1021/acsnano.3c09099

  12. [12]

      X.J. She, Y.F. Wang, H. Xu, S.C.E. Tsang, S.P. Lau, Angew. Chem. Int. Ed. 61 (2022) e202211396. doi: 10.1002/anie.202211396

  13. [13]

      L.S. Zhan, Y.C. Wang, M.J. Liu, et al., Rare Met. 42 (2023) 806. doi: 10.1007/s12598-022-02212-w

  14. [14]

      X.H. Sun, L. Sun, G.N. Li, et al., Angew. Chem. Int. Ed. 61 (2022) e202207677. doi: 10.1002/anie.202207677

  15. [15]

      Y.C. Wang, L. Xu, L.S. Zhan, et al., Nano Energy 92 (2022) 106780. doi: 10.1016/j.nanoen.2021.106780

  16. [16]

      Y. Wang, X.B. Zheng, D.S. Wang, Nano Res. 15 (2022) 1730. doi: 10.1007/s12274-021-3794-0

  17. [17]

      L.G. Wang, J.B. Wu, S.W. Wang, et al., Nano Res. 17 (2024) 3261–3301. doi: 10.1007/s12274-023-6037-8

  18. [18]

      Q.C. Wang, Y.P. Lei, Y.C. Wang, et al., Energy Environ. Sci. 13 (2020) 1593. doi: 10.1039/d0ee00450b

  19. [19]

      X. Zhao, M.J. Liu, Y.C. Wang, et al., ACS Nano 16 (2022) 19959. doi: 10.1021/acsnano.2c09888

  20. [20]

      Y.C. Wang, Z.S. Huang, Y.P. Lei, et al., Chem. Commun. 58 (2022) 3621. doi: 10.1039/d2cc00114d

  21. [21]

      H.C. Peng, J. Ren, Y.C. Wang, et al., Nano Energy 88 (2021) 106307. doi: 10.1016/j.nanoen.2021.106307

  22. [22]

      X. Zhao, Y.C. Wang, L.S. Zhan, et al., Chem. Commun. 58 (2022) 12716. doi: 10.1039/d2cc05042k

  23. [23]

      H.B. Wang, C.Y. Tang, B. Sun, et al., Int. J. Extreme Manuf. 4 (2022) 035002. doi: 10.1088/2631-7990/ac7a6e

  24. [24]

      B.Q. Gao, Y. Pan, H. Yang, Appl. Catal. B: Environ. 315 (2022) 121580. doi: 10.1016/j.apcatb.2022.121580

  25. [25]

      L. Chen, C.Z. He, R. Wang, et al., Chin. Chem. Lett. 32 (2021) 53–56. doi: 10.1016/j.cclet.2020.11.013

  26. [26]

      J. Yu, A. Kudo, Adv. Funct. Mater. 16 (2006) 2163. doi: 10.1002/adfm.200500799

  27. [27]

      L. Wang, J.X. Liu, W.Y. Song, et al., Chem. Eng. J. 366 (2019) 504. doi: 10.1016/j.cej.2019.02.038

  28. [28]

      F.Q. Ma, J.W. Yao, Y.F. Zhang, Y. Wei, Chin. Chem. Lett. 29 (2018) 1689–1691. doi: 10.1016/j.cclet.2017.12.016

  29. [29]

      J.B. Pan, B.H. Wang, J.B. Wang, et al., Angew. Chem. Int. Ed. 60 (2021) 1433. doi: 10.1002/anie.202012550

  30. [30]

      Y.T. Wang, Y.H. Li, J.Z. Liu, et al., Angew. Chem. Int. Ed. 60 (2021) 7681. doi: 10.1002/anie.202014341

  31. [31]

      W.B. Wang, Z.T. Wang, R.O. Yang, et al., Angew. Chem. Int. Ed. 60 (2021) 22940. doi: 10.1002/anie.202110000

  32. [32]

      S. Liu, Y.P. Fan, Y. Wang, et al., Nano Lett. 22 (2022) 9107. doi: 10.1021/acs.nanolett.2c03573

  33. [33]

      J.Y. Duan, T.Y. Liu, Y.H. Zhao, et al., Nat. Commun. 13 (2022) 2039. doi: 10.1038/s41467-022-29699-2

  34. [34]

      Q.C. Wang, Q.G. Feng, Y.P. Lei, et al., Nat. Commun. 13 (2022) 3689. doi: 10.1038/s41467-022-31383-4

  35. [35]

      Y. Wang, J. Wu, C.L. Ye, et al., Angew. Chem. Int. Ed. 62 (2023) e202219191. doi: 10.1002/anie.202219191

  36. [36]

      T.T. Cui, Y.P. Wang, T. Ye, et al., Angew. Chem. Int. Ed. 61 (2022) e202115219. doi: 10.1002/anie.202115219

  37. [37]

      Q. Li, Y.C. Wang, J. Zeng, et al., Chin. Chem. Lett. 32 (2021) 3355–3358. doi: 10.1016/j.cclet.2021.03.063

  38. [38]

      K. Ye, G.R. Zhang, X.Y. Ma, et al., Energy Environ. Sci. 15 (2022) 749. doi: 10.1039/d1ee02966e

  39. [39]

      Y.N. Hao, F. Hu, S.Q. Zhu, et al., Angew. Chem. Int. Ed. 62 (2023) e202304179. doi: 10.1002/anie.202304179

  40. [40]

      J.D. Yi, X.P. Gao, H. Zhou, W. Chen, Y.E. Wu, Angew. Chem. Int. Ed. 61 (2022) e202212329. doi: 10.1002/anie.202212329

• 

  Figure 1  (a) Schematic illustration for the synthesis of BV-2. (b) XRD patterns of different samples. (c) Raman results of BV-2 and BV-0.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Bi 4f XPS results and (b) O 1s XPS results of different samples. (c) EPR spectra of BV-0 and BV-2. (d) EDS mapping images of BV-2.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) LSV curves of different samples at CO₂-saturated atmosphere, inset: LSV curves of BV-2 under different atmosphere. (b) FEs with different products of BV-2. (c) FE_HCOO-. (d) J_HCOO-, inset: the stability test of BV-2 at −1.0 V in H-type cells. (e) The performance comparison of up-to-date Bi-based catalysts for CO₂RR in H-type cells.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a, b) In-situ Raman spectra recorded at −1.0 V. (c) EIS measurement, (d) FE_HCOO- and J_HCOO- of BV-2 in flow cells.

  下载: 全尺寸图片 幻灯片
•