English

• Carbon dioxide reduction is of particular interest in terms of greenhouse alleviation and the production of carbon feedstock. Therefore, the homogeneous and heterogeneous electrocatalytic CO₂ conversion is becoming more and more attractive recently [1-6]. Various electrochemical CO₂ reduction can be involved in different pathways including two, four, six and eight-electron processes, leading to various carbon-based products such as, CO, HCOOH, CH₃OH, HCHO, CH₄, and other value-added chemicals or liquid fuels which could benefit our society sustainability [2,7,8].

  Among different families of molecular catalysts for electrocatalytic CO₂ reduction, earth-abundant transition metal complexes, particularly the manganese-based bipyridine complexes, have attracted many interests [9-12]. However, they displayed limited efficiency although extensive exploration has been performed on modifying pyridyl and bipyridyl ligands. N-Heterocyclic carbene (NHC) complexes exhibit unique selectivity and stability in catalysis [13-16], owing to the strong σ-donation and the weak π-back-donation properties of the NHC ligands [17]. Recently, manganese complexes supported by NHC ligands have been reported with potential electrocatalytic performances (Fig. 1a) [18-21].

  Figure 1

  

  Figure 1.  NHC Mn(Ⅰ) complexes for electrocatalytic reduction of CO₂.

  DownLoad: Full-Size Img PowerPoint

  In 2018, Luca et al. reported an NHC-based manganese(I) pincer complex [MnCNC^H]Br (L1-Mn) for the selectively electrocatalytic reduction of CO₂ to CO [22]. Inspired by this pincer framework, and guided by our previous theoretical understanding of the para electronic effect [23,24], herein, we developed new types of manganese pincer complexes to achieve highly active and selective electro-reduction of CO₂ by tuning the para electronic effect (Fig. 1b).

  By the introduction of pyridyl para substituents on complex L1-Mn [22,25], we designed and synthesized new complexes L2-Mn and L3-Mn (Scheme 1) incorporating an electron-donating methoxy (-OMe) group and an electron-withdrawing methyl formate (-COOMe) group, respectively. Ligands L1, L2 and L3 were synthesized from dibromopyridine, 2, 4, 6-trifluoropyridine, and methyl 2, 6-dihydroxyisonicotinate, respectively (Schemes S1 and S2 in Supporting information). Complexes L1-Mn, L2-Mn and L3-Mn were obtained with conventional Mn(CO)₅Br as starting materials by overnight reactions in the presence of excess Cs₂CO₃ in MeCN under argon atmosphere (Scheme 1). ¹H NMR, ¹³C NMR, infrared spectroscopy and single-crystal X-ray diffraction analyses (Scheme 1,Tables S1, S2, Figs. S1-S10 in Supporting information) verified the successful synthesis of this series of complexes. Single-crystal X-ray diffraction analysis confirmed the octahedral structures of L2-Mn and L3-Mn, with Mn-C_NHC bond lengths of 1.999(5), 2.009(9) Å, and Mn-N bond lengths of 2.001(0), 2.012(4) Å, respectively (Scheme 1). The structures and bond lengths of both complexes are in accordance with the previously reported L1-Mn [22] with Mn-C_NHC bond length of 1.989(3) Å and Mn-N bond length of 2.017(3) Å.

  Scheme 1

  

  Scheme 1.  Synthesis of Mn(Ⅰ) NHC complexes and X-ray structures of L2-Mn and L3-Mn.

  DownLoad: Full-Size Img PowerPoint

  Cyclic voltammetry analyses of the series of Mn(Ⅰ) complexes were performed under argon and CO₂ atmosphere, respectively (Fig. 2). Under Ar atmosphere, the cyclic voltammetry of L1-Mn showed an irreversible redox wave at E_p1 = −2.02 V (Fig. 2a, all reduction potentials are reported vs. Fc⁺/Fc), in agreement with previous studies [22]. The redox wave in the CV of complex L2-Mn with -OMe shifted cathodically by approximately 200 mV (Fig. 2b, E_p2 = −2.16 V under Ar atmosphere) compared to the redox waves in the CV of complex L1-Mn, reflecting the strong electron-donating ability of the -OMe group. The CV of L3-Mn with-COOMe shows three pairs of redox waves at −1.64 V, −1.86 V, and −2.14 V, respectively (Fig. 2c and Fig. S23 in Supporting information). Under CO₂ atmosphere, the catalytic current in the CV of L1-Mn increased by about 40-fold with the redox wave at E_cat1 = −2.37 V. Surprisingly, the CV of L2-Mn shows a dramatic current increase (ca. 63-fold) at E_cat2 = −2.83 V, suggesting that the promising promotion of catalytic activity of complex L2-Mn for CO₂ reduction caused by the influence of the electron-donating effect. However, the CV of complex L3-Mn displayed only minor catalytic current increase (Fig. 1c, ca. 1.4-fold), which is in sharp contrast to the behavior of L1-Mn, indicating the negative effect of the electron-withdrawing group on the electro-reduction.

  Figure 2

  

  Figure 2.  Cyclic voltammetry of L1-Mn (a), L2-Mn (b) and L3-Mn (c) under Ar (black) and CO₂ (red) atmosphere at 0.1 V/s in dry MeCN with 0.1 mol/L tetrabutylammonium hexafluorophosphate (TBAPF₆) supporting electrolyte. All complexes were loaded at a concentration of 1 mmol/L. Glass carbon working electrode, platinum counter electrode, and Ag⁺/Ag reference electrode was used.

  DownLoad: Full-Size Img PowerPoint

  Cyclic voltammetry analyses of complexes L1-Mn and L2-Mn at scan rates of 8–12 V/s under both CO₂ and Ar atmosphere were performed (Figs. S11, S12, S17 and S18 in Supporting information). An ideally steady-state catalytic progress will lead to an S-shaped CV response, which is independent of the scan rate, with plateau current to determine the maximum turnover frequency (TOF_max). CVs of both complexes showed the S-shaped catalytic waves, indicating the ideally and purely kinetic CO₂ catalytic reduction progresses (Figs. S12 and S18). We operated the i_cat/i_p analyses to calculate the TOF_max for L1-Mn and L2-Mn from the reported equations [26-29]L1-Mn showed a TOF_max of 206 s⁻¹ under the CO₂ atmosphere in anhydrous MeCN (Fig. S40 in Supporting information). Notably, the calculated TOF_max of L2-Mn is 26, 127 s⁻¹, which is approximately 127 times higher than the value of L1-Mn under the same conditions (Fig. S42 in Supporting information). Reduction potentials, catalytic currents, and calculated TOF_max under different conditions are summarized in Table 1.

  Table 1

  

  Table 1.  Results of cyclic voltammetry and controlled potential electrolyses.

  DownLoad: CSV

  The addition of Brønstic acid, for example, 2, 2, 2-trifluoroethanol (TFE) as the proton source, would generally result in the enhancement of catalysis, as reported previously [11,12,22,30,31]. We performed cyclic voltammetry of complexes L1-Mn and L2-Mn at incremental TFE additions until the catalytic current no longer increased to study the optimum concentration for catalytic conditions (Figs. S13 and S19 in Supporting information). The addition of TFE promoted the catalytic activity of complex L1-Mn, as shown in Fig. S13. 240 µL TFE was added to achieve the optimal catalytic conditions. Cyclic voltammetry of L1-Mn under Ar and CO₂ atmosphere in 0.1 mol/L TBAPF₆/MeCN with 240 µL TFE added gave a 56-fold catalytic current enhancement (Fig. S14 in Supporting information). However, complex L2-Mn displayed poor catalytic property when TFE was added as the proton source. The value of i_cat/i_p for L2-Mn at the presence of TFE decreased to 50 × (Fig. S20 in Supporting information). This phenomenon can be attributed to the protonation of the -OMe group, which thereby suppresses its electron-donation ability. L1-Mn and L2-Mn with 240 µL TFE added display S-shaped CVs response at scan rates of 8–12 V/s, which represented an ideally steady-state catalysis (Figs. S15 and S21 in Supporting information). With 240 µL TFE added, the calculated TOF_max of L1-Mn is enhanced to 454 s⁻¹ (Fig. S41 in Supporting information). Owing to the possible protonation of -OMe mentioned above, the calculated TOF_max value of L2-Mn decreased to 408 s⁻¹ with 240 µL TFE addition (Fig. S43 in Supporting information). Cyclic voltammetry of L3-Mn with 240 µL TFE added showed a low value of catalytic current enhancement (i_cat/i_p = 2.3), indicating its inefficient electrocatalytic properties for CO₂ reduction. Further control experiments with tetrabutylammonium bromide (TBABr) to introduce Br⁻ anion into the catalytic system of L2-Mn were performed (Figs. S25 and S26 in Supporting information), indicating the counter anions did not affect the electrochemical properties [32,33].

  Controlled potential electrolysis (CPE) of L1-Mn, L2-Mn, and L3-Mn were performed under CO₂ atmosphere at an applied potential of E_appl = −2.37 V under the conditions of anhydrous MeCN and with the addition of TFE as a proton source to study their selectivity and durability (Figs. S27–S36 in Supporting information). Quantification of the gaseous products was conducted by an online gas chromatography thermal conductivity detector (GC-TCD) analysis of the headspace. After an electrolysis over 2 h in anhydrous MeCN, approximately 40.9% current density of the starting state was retained for L1-Mn (Fig. S27), which gave an average Faradaic efficiency (FE) of 94.2% (Fig. S28). L2-Mn displayed better selectivity and durability with approximately 86.6% current density retained over 2 h (Fig. S31) and an average FE of 104.3% was obtained (Fig. S32). These results demonstrate that the electron-donating effect also enhances the selectivity and durability of the Mn NHC pincer catalyst. With 240 µL TFE added, further promotion of selectivity and durability for L1-Mn were shown by the CPE results. 77.2% current density retained over 2 h (Fig. S29) and an average FE of 94.7% was obtained (Fig. S30). Consistent with the results of CVs, L2-Mn gave a low level of current density probably because of the protonation of the -OMe. On the other hand, L3-Mn showed poor catalytic activity whether TFE was added or not (Figs. S35 and S36). Average Faradaic efficiencies are summarized in Table 1.

  According to the studies by Musgrave and Luca et al., a binuclear proton-less reduction pathway is suggested for the [MnCNC(CO)₃]Br pincer system [34]. Therefore, a comparison of the CO₂ reduction cycle for L1-Mn, L2-Mn and L3-Mn based on the bimetallic mechanism using density functional study (DFT) has been conducted to reveal the effects of the para substituents (Fig. 3a). First, the Mn(Ⅰ) species I are reduced to species Ⅱ with the required potentials of −2.21 V, −2.25 V and −1.53 V for L1-Mn to L3-Mn, respectively, which are in agreement with the reduction potentials observed by the CV experiments (Fig. 2). These results reveal the increasing reduction potential of electron-richer species with the para electron-donating effect. After the loss of two CO molecules from Ⅱ, which was observed by Luca et al. in the case of L1-Mn, complexes Ⅲ adopting square planar geometry are generated with the calculated Gibbs free energies to be endergonic by 9.3–13.7 kcal/mol. Then, a complexation process of a CO₂ molecule sandwiched between two Ⅲ intermediates generates complexes Ⅳ, which are calculated to be 10.4 (9.3 kcal/mol to 19.7 kcal/mol), 5.6 (13.4 kcal/mol to 19.0 kcal/mol), and 30.0 kcal/mol (13.7 kcal/mol to 43.7 kcal/mol) endergonic, respectively. Such results indicate the disfavor of the complexation for L3-Mn bearing a pyridyl para EWG. Subsequently, the insertion of a second CO₂ molecule into species Ⅳ leading to species Ⅴ with effective disproportionation into a carbonate anion sandwiched between two metal centers and a CO product bounded to one of them. After the release of the CO product, complexes Ⅵ will be formed. With the calculated potentials of −2.06 V, −2.27 V and −1.71 V for L1-Mn, L2-Mn and L3-Mn, respectively. Two Ⅲ species can be regenerated with the release of the carbonate to furnish the catalytic cycle.

  Figure 3

  

  Figure 3.  (a) Comparison of CO₂ reduction cycle for L1-Mn (deep green), L2-Mn (red), and L3-Mn (blue) based on the binuclear proton-less reduction pathway. The Gibbs free energies and the calculated applied electro potentials are kcal/mol and V vs. Fc⁺/Fc, respectively. (b) SOMOs and ADCH charges of the Mn center of L2-Mn-Ⅲ, L2-Mn+H-Ⅲ, and L3-Mn-Ⅲ.

  DownLoad: Full-Size Img PowerPoint

  To further reveal the electronic effect of the pyridyl para substituents, the molecular orbital analysis and the atomic dipole moment corrected Hirshfeld (ADCH) charges [35,36] analysis of the Mn center were performed for the reduced species Ⅲ of L1-Mn (−0.19), L2-Mn (−0.22) and L3-Mn (+0.18) were analyzed with Multiwfn [36,37], as shown in Fig. 3b. The ADCH charge of the Mn center increases to −0.22 in L2-Mn-Ⅲ from −0.19 in L1-Mn-Ⅲ, indicating the profound impact of the electron-donating para pyridyl substituent on improving the nucleophilicity of the Mn center to activate the CO₂ molecule. The low electron density of the Mn center (+0.18) in L3-Mn-Ⅲ and its relatively stable SOMO (−1.685 eV) can account for the low activity of L3-Mn for the electrocatalytic reduction of CO₂ due to the EWG. Notably, the protonation of the -OMe (species L2-Mn+H-Ⅲ) significantly increases the ADCH charge of the Mn center to +0.02 from −0.22 in L2-Mn-Ⅲ. And the energy of the SOMO of L2-Mn+H-Ⅲ also decreases to −1.760 eV, which is in good agreement with the experimentally observed suppression of the catalysis due to the negative effect of the TFEs as a proton source.

  We also performed Fourier-transform infrared reflectance spectroelectrochemistry (FTIR-SEC) to detect the main species involved in the catalytic reduction process of L2-Mn (Figs. S37 and S38 in Supporting information). Under Ar atmosphere, the spectral features at 2117 cm⁻¹, attributing to a CO ligand released [38], corresponding to the report by Myren et al. [34]. Under the CO₂ atmosphere, the bands at 1649 cm⁻¹ and 1680 cm⁻¹ appeared, showing the formation of CO₃²⁻/HCO₃⁻ species, which is consistent with the DFT-calculated results.

  In summary, a series of manganese N-heterocyclic carbene pincer complexes were developed. The para electron-donating strategy was achieved in a highly active complex (L2-Mn) having -OME group for the selective electrocatalytic reduction of CO₂ to CO, with a larger catalytic current enhancement (63 ×) and a TOF_max (26, 127 s⁻¹, 127 times of unsubstituted L1-Mn). CPE study of L2-Mn gave an average Faradaic efficiency of 104%. While L3-Mn with electron-withdrawing -COOMe group inhibited the electrocatalytic activity. Ambient Brønstic acid suppressed the activity of L2-Mn probably due to the protonation of the –OMe group. These findings highlight the potential of electronic tuning for the design of efficient manganese NHC catalysts for CO₂ reduction.

  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 the National Natural Science Foundation of China (No. 21973113) and the Guangdong Natural Science Funds for Distinguished Young Scholar (No. 2015A030306027), and the Fundamental Research Funds for the Central Universities.

  Appendix Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2021.06.046.

  
  
• 1. [1]

     C.A. Carpenter, P. Brogdon, L.E. McNamara, et al., Inorganics 6(2018) 22. doi: 10.3390/inorganics6010022

  2. [2]

     R. Francke, B. Schille, M. Roemelt, Chem. Rev. 118(2018) 4631-4701. doi: 10.1021/acs.chemrev.7b00459

  3. [3]

     K. Elouarzaki, V. Kannan, V. Jose, et al., Adv. Energy Mater. 9(2019) 1900090. doi: 10.1002/aenm.201900090

  4. [4]

     S. Fukuzumi, Y.M. Lee, H.S. Ahn, W. Nam, Chem. Sci. 9(2018) 6017-6034. doi: 10.1039/C8SC02220H

  5. [5]

     A. Sinopoli, N.T. La Porte, J.F. Martinez, et al., Coord. Chem. Rev. 365(2018) 60-74. doi: 10.1016/j.ccr.2018.03.011

  6. [6]

     H. Takeda, C. Cometto, O. Ishitani, M. Robert, ACS Catal. 7(2016) 70-88.

  7. [7]

     E.E. Benson, C.P. Kubiak, A.J. Sathrum, J.M. Smieja, Chem. Soc. Rev. 38(2009) 89-99. doi: 10.1039/B804323J

  8. [8]

     J. Qiao, Y. Liu, F. Hong, J. Zhang, Chem. Soc. Rev. 43(2014) 631-675. doi: 10.1039/C3CS60323G

  9. [9]

     M. Bourrez, F. Molton, S. Chardon-Noblat, A. Deronzier, Angew. Chem. Int. Ed. 50(2011) 9903-9906. doi: 10.1002/anie.201103616

  10. [10]

      J.M. Smieja, M.D. Sampson, K.A. Grice, et al., Inorg. Chem. 52(2013) 2484-2491. doi: 10.1021/ic302391u

  11. [11]

      M.D. Sampson, C.P. Kubiak, J. Am. Chem. Soc. 138(2016) 1386-1393. doi: 10.1021/jacs.5b12215

  12. [12]

      K.T. Ngo, M. McKinnon, B. Mahanti, et al., J. Am. Chem. Soc. 139(2017) 2604-2618. doi: 10.1021/jacs.6b08776

  13. [13]

      D.J. Nelson, S.P. Nolan, Chem. Soc. Rev. 42(2013) 6723-6753. doi: 10.1039/c3cs60146c

  14. [14]

      M.N. Hopkinson, C. Richter, M. Schedler, F. Glorius, Nature 510(2014) 485-496. doi: 10.1038/nature13384

  15. [15]

      C. Johnson, M. Albrecht, Coord. Chem. Rev. 352(2017) 1-14. doi: 10.1016/j.ccr.2017.08.027

  16. [16]

      E. Peris, Chem. Rev. 118(2017) 9988-10031. doi: 10.1021/acs.chemrev.7b00037

  17. [17]

      C. Huang, J. Liu, H.H. Huang, Z. Ke, Polyhedron 203(2021) 115147. doi: 10.1016/j.poly.2021.115147

  18. [18]

      J. Agarwal, T.W. Shaw, C.J. Stanton, et al., Angew. Chem. Int. Ed. 53(2014) 5152-5155. doi: 10.1002/anie.201311099

  19. [19]

      J. Agarwal, C.J. Stanton Iii, T.W. Shaw, et al., Dalton Trans. 44(2015) 2122-2131. doi: 10.1039/C4DT03079F

  20. [20]

      F. Franco, M.F. Pinto, B. Royo, J. Lloret-Fillol, Angew. Chem. Int. Ed. 57(2018) 4603-4606. doi: 10.1002/anie.201800705

  21. [21]

      H.A. Petersen, T.H.T. Myren, O.R. Luca, Inorganics 8(2020) 62. doi: 10.3390/inorganics8110062

  22. [22]

      T.H.T. Myren, A.M. Lilio, C.G. Huntzinger, et al., Organometallics 38(2018) 1248-1253. doi: 10.1021/acs.organomet.1c00036

  23. [23]

      J.A. Therrien, M.O. Wolf, Inorg. Chem. 56(2017) 1161-1172. doi: 10.1021/acs.inorgchem.6b02213

  24. [24]

      H.H. Huang, M. Dai, L. Liu, et al., Catal. Sci. Technol. 11(2021) 874-885. doi: 10.1039/D0CY01832E

  25. [25]

      X.B. Lan, Z. Ye, J. Liu, et al., ChemSusChem 13(2020) 2557-2563. doi: 10.1002/cssc.202000576

  26. [26]

      F. Franco, M.F. Pinto, B. Royo, J. Lloret-Fillol, Angew. Chem. Int. Ed. 57(2018) 4603-4606. doi: 10.1002/anie.201800705

  27. [27]

      C. Costentin, S. Drouet, M. Robert, J.M. Saveant, J. Am. Chem. Soc. 134(2012) 11235-11242. doi: 10.1021/ja303560c

  28. [28]

      E.S. Rountree, B.D. McCarthy, T.T. Eisenhart, J.L. Dempsey, Inorg. Chem. 53(2014) 9983-10002. doi: 10.1021/ic500658x

  29. [29]

      C. Costentin, J.M. Savéant, ChemElectroChem 1(2014) 1226-1236. doi: 10.1002/celc.201300263

  30. [30]

      K.S. Rawat, A. Mahata, I. Choudhuri, B. Pathak, J. Phys. Chem. C 120(2016) 8821-8831. doi: 10.1021/acs.jpcc.6b02209

  31. [31]

      G.K. Rao, W. Pell, I. Korobkov, D. Richeson, Chem. Comm. 52(2016) 8010-8013. doi: 10.1039/C6CC03827A

  32. [32]

      M.L. Clark, P.L. Cheung, M. Lessio, et al., ACS Catal. 8(2018) 2021-2029. doi: 10.1021/acscatal.7b03971

  33. [33]

      H.Y. Kuo, S.E. Tignor, T.S. Lee, et al., Dalton Trans. 49(2020) 891-900. doi: 10.1039/C9DT04150H

  34. [34]

      T.H.T. Myren, A. Alherz, J.R. Thurston, et al., ACS Catal. 10(2020) 1961-1968. doi: 10.1021/acscatal.9b04773

  35. [35]

      T. Lu, F. Chen, J. Theor. Comput. Chem. 11(2012) 163-183. doi: 10.1142/S0219633612500113

  36. [36]

      T. Lu, F. Chen, Acta Phsy. Chim. Sin. 28(2012) 1-18. doi: 10.3866/PKU.WHXB2012281

  37. [37]

      T. Lu, F. Chen, J. Comput. Chem. 33(2012) 580-592. doi: 10.1002/jcc.22885

  38. [38]

      W. Luo, M. Sankar, A.M. Beale, et al., Nat. Commun. 6(2015) 6540. doi: 10.1038/ncomms7540

• 

  Figure 1  NHC Mn(Ⅰ) complexes for electrocatalytic reduction of CO₂.

  下载: 全尺寸图片 幻灯片
  

  Scheme 1  Synthesis of Mn(Ⅰ) NHC complexes and X-ray structures of L2-Mn and L3-Mn.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Cyclic voltammetry of L1-Mn (a), L2-Mn (b) and L3-Mn (c) under Ar (black) and CO₂ (red) atmosphere at 0.1 V/s in dry MeCN with 0.1 mol/L tetrabutylammonium hexafluorophosphate (TBAPF₆) supporting electrolyte. All complexes were loaded at a concentration of 1 mmol/L. Glass carbon working electrode, platinum counter electrode, and Ag⁺/Ag reference electrode was used.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Comparison of CO₂ reduction cycle for L1-Mn (deep green), L2-Mn (red), and L3-Mn (blue) based on the binuclear proton-less reduction pathway. The Gibbs free energies and the calculated applied electro potentials are kcal/mol and V vs. Fc⁺/Fc, respectively. (b) SOMOs and ADCH charges of the Mn center of L2-Mn-Ⅲ, L2-Mn+H-Ⅲ, and L3-Mn-Ⅲ.

  下载: 全尺寸图片 幻灯片

  Table 1.  Results of cyclic voltammetry and controlled potential electrolyses.

  下载: 导出CSV
•