Thank you for visiting Nature.com. The version of browser you are using has limited CSS support. For best results, we recommend that you use a newer version of your browser (or disable Compatibility Mode in Internet Explorer). In the meantime, to ensure ongoing support, we are showing the site without styling or JavaScript.
        Oxidative CO coupling is a simple and economical, safe synthetic route in the laboratory and industry for the synthesis of valuable α-diketone fragments containing C2 or higher carbon compounds, but has not yet been explored. In this work, a rare coplanar binuclear carbonyl complex of hydroxycobalt(III) with a macrocyclic equatorial Schiff base ligand and a c-k1(O):k1(O’)- acetate bridge connects the axial ligand. The Co(III)-COOH bond in this complex can be photocleaved to form oxalic acid. Moreover, based on dicobalt(III) complex using O2 as an oxidant at ambient temperature and pressure, photostimulated catalysis for the direct production of oxalic acid from CO and H2O with good selectivity (>95%) and atomic economy was achieved. , turnover 38.5. 13C labeling and 18O labeling experiments confirmed that CO and H2O are the sources of -COOH groups in binuclear carbonyl complexes of hydroxycobalt(III) and oxalic acid products.
        Carbon monoxide (CO) is one of the main C1 raw materials in laboratories and the chemical industry1,2,3, and its selective compound CC has long been considered an important and efficient route for the synthesis of C2 and high-carbon products4,5,6. 7. Most previously documented CO coupling strategies can be divided into two categories: oxidative CO coupling and reductive CO coupling. Research on reductive CO fixation dates back to the early 19th century, when molten potassium was reported to reduce CO to form [C2O2]n2n-8,9 anions. Since then, a number of complexes of s-, 7p-, 10, 11d-, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 and f-block22, 23, 24 elements have been shown to be in able to perform reductive coupling of CO, resulting in the formation of a number of CC bond-forming products, including acetylene glycol esters, alkenediol esters, and anoxic hydrocarbons. In contrast, research on oxidative CO binding is lagging far behind, with only limited examples reported to date25. For example, the organolithium element is known to react with CO to form 1,2-diones as CO oxidative coupling products26,27,28. In addition, it has been shown that some transition metal complexes, such as Re(I/III) and Pd(II) complexes, are capable of mediating stoichiometric and catalytic oxidative coupling of CO29,30,31,32,33,34,35 (Fig. 1a, route i). High CO pressure (65–80 atm) is critical for good yields and selectivity of known catalytic systems, and carbonate is often the main product formed at low CO pressure in the process34. From a mechanistic point of view, high gas pressure is a prerequisite for a one-point pathway in which two CO molecules are activated at the metal center, controlling the above process. Therefore, cooperative mechanisms in two or more geometrically linked reaction centers may pave the way for oxidative CO sequestration under ambient conditions36.
        The reported route for the synthesis of oxalic acid from CO. b Direct synthesis of oxalic acid by oxidative coupling of CO.
        Oxalic acid plays an indispensable role in various industrial processes such as metal processing, rare earth mining, leather processing, pharmaceuticals, etc., with an annual market volume of 350,000 tons37. CO is one of the main raw materials for the industrial production of oxalic acid, in which CO is first converted into an intermediate alkali metal formate, and then through formate coupling and acidification, oxalic acid is obtained (Fig. 1a, route ii)). However, this multi-step, labor-intensive and energy-intensive route requires harsh conditions, including high reaction temperatures and carbon dioxide pressure. In this case, the formation of carbonate by-products due to the decomposition of alkali metal formates and oxalate products is inevitable. In principle, oxalic acid can be prepared directly by the oxidative coupling of CO with H O, but to our knowledge, such a process has not been reported.
        Here we demonstrate photopromoted direct selective production from CO and H O by coplanar binuclear hydroxycobalt(III) carbonyl complexes using O as the oxidant at ambient temperature (room temperature) and atmospheric pressure with good atom economy (1 atm CO and 1 atm) . O2), has a turnover number of 38.5.
        A macrocyclic ligand (H2L) with two coplanar metal binding sites was synthesized according to a previously described method. The reaction of H2L with 2 equivalents of Co(OAc)2 in ethanol under N2 followed by recrystallization from Et2O isolated yellow complex 1 in 86% yield (Fig. 2a). Solid-state structure 1 shows two Co(II) centers located on opposite sides of the ligand with additional coordination from the acetate ligand (Fig. 3a). Two secondary benzylamine moieties in the ligand framework are converted to imines during metalation, likely through cobalt-mediated dehydrogenation39. 1H-NMR monitoring of 1 synthesized in a J. Young NMR tube showed a singlet resonance at δ = 4.35 ppm, indicating the formation of H2 (Supplementary Figure 1). The effective magnetic moment 1 (7.67 μB) measured at room temperature indicates the presence of two high-spin (S = 3/2) Co(II) centers (Supplementary Figure 2). The spin-orbit splitting energy (15.9 eV) obtained from X-ray photoelectron spectroscopy (XPS) measurements is also consistent with the +2 oxidation state assignment of the two cobalt centers (Supplementary Figure 3). 1 is readily oxidized by O2 in methanol at room temperature to form the red mixed-valence dicobalt(II/III) complex 2, as confirmed by XPS measurements (measured Co(II) and Co(III) values ​​are 15.6 eV and 14.8 eV, respectively ). eV, Supplementary Figure 5) with an isolated yield of 81% (Figure 2b). The solid structure of compound 2 shows that the Co(III) center is well placed in the N2O2 plane and occupies an octahedral coordination geometry, while the larger Co(II) center remains above the ligand framework (Fig. 3b). Accordingly, the Co(III)-equatorial (1.881(3)/1.868(2) Å) and Co(III)-non-equatorial (1.867(3)/1.870(3) Å) bonds in 2 are significantly shorter than in 2. key . 1 (Co(II)-equator: 2.063(2)/2.144(3) Å; Co(II)-equator: 2.108(3)/2.071(3) Å). Effective magnetic moment 2 (4.12 μB) represents high-spin (S = 3/2) Co(II) centers and low-spin (S = 0) Co(III) centers (Supplementary Figure 6).
        Synthesis 1. Synthesis of b 2 . Synthesis of c 3 . d synthesizes 4 from 3. Synthesis e 5. f synthesizes 4 out of 5.
        a represents 1. b represents 2. c represents the cationic moiety of 3. d represents the cationic moiety of 4. Crystallization solvent molecules, counteranions, and hydrogen atoms (excluding oxygen atoms) have been omitted for clarity. For details of these X-ray structures, see Supplementary Data 1.
        After adding 1.1 eq. Ce(NH4)2(NO3)6, further oxidation of 2 at an elevated reaction temperature gave the brown dicobalt(III) complex 3 with an isolated yield of 86% (Fig. 2c). The measured spin-orbit splitting energy (15.1 eV) is consistent with the presence of only Co(III) centers (Supplementary Figure 7). The 1H NMR spectrum of compound 3 shows a sharp resonance in the range δ = 1–10 ppm, indicating the diamagnetic nature of compound 3 (Supplementary Figure 10). As shown in Figure 3c, both Co(III) centers in 3 are well integrated into the N2O2 plane of the ligand. Co(III)-Equator (1.914(2)/1.894(2)/1.896(2)/1.900(2) Å) and Co(III)-Equator (1.859(3)/1.874(3)/1.868(3) )/1.866(3) Å) bond length is very similar to the bond length of the Co(III) center in 2 . The two Co(III) centers coordinate in a μ-κ1(O):κ1(O’) manner, while the hydroxyl and methoxy-axial ligands coordinate with each Co(III) center on the other side.
        Treatment with 3 1 atm CO in methanol (in the presence of 0.01% water by weight) at 50 °C resulted in the formation of a diamagnetic binuclear hydroxycobalt(III) carbonyl complex (4) within 24 h (Figure 2d). The spin-orbit splitting energy (15.0 eV) measured by XPS is almost the same as that of 3 (Supplementary Figure 12). With the exception of different axial ligands, the molecular structure of 4 is largely similar to the geometric features of 3 (Figure 3d). For other hydroxycarbonyl metal complexes, the bond lengths C = O (1.193(5)/1.207(4) Å) and C(O)-OH (1.231(5)/1.249(4) Å) are within the normal range40,41. The infrared spectrum of compound 4 shows two close absorptions at 1697 and 1670 cm-1, which is in satisfactory agreement with the simulation results based on density functional theory (DFT) calculations (1724 and 1685 cm-1). When using 13CO for the synthesis of 4, both absorptions are red-shifted to 1662 and 1651 cm-1 (Supplementary Fig. 13, calculated values ​​1685 and 1648 cm-1). 1H NMR signals of two carboxyl protons were detected at δ = 13.34 and 13.08 ppm. respectively (Supplementary Figure 14). Additionally, thermogravimetric analysis was performed on 4 and it was observed that weight loss (10.1 wt%) occurred at 280–290 °C due to the release of two -COOH groups (Supplementary Figure 17). 4 is expected to be formed by the incorporation of CO into the Co(III)-OH bond of the proposed binuclear hydroxycobalt(III) intermediate, which can be produced by reacting 3 with H2O leading to axial methoxy. The main ligand is replaced at the hydroxyl group. Although attempts to isolate this intermediate were unsuccessful, cobalt tetranuclear complex 5 was obtained in isolated 42% yield by heating an ethanol solution of 3 at 80 °C for 3 h, as shown in Figure 2e. The measured spin-orbit splitting energy (15.0 eV) of the cobalt center in 5 is almost the same as that of the Co(III) center in 3 and 4 (Supplementary Figure 18). The Co-Oμ-OH bond length in 5 (1.905(4)/1.909(4)/1.907(4)/1.895(4) Å, Figure 4a) is very close to the reported values ​​(1.888 – 1.912). Å) for binuclear µ-hydroxyl forms of cobalt(III)40,41, but shorter than the typical Co(III)-OH2 bond (1.9​45 Å)42,43 and significantly longer than Co(III)-Oµ -O bond (1.783 – 1.796 angstroms)44. An effective magnetic moment of 5 (1.14 μB) measured at room temperature indicates the presence of only one unpaired electron (Supplementary Figure 19). Unconstrained corresponding orbital analysis and calculated spin density 5 indicate that the unpaired electron density is located at the center of cobalt (Supplementary Figure 20). EPR measurements of solid-state 5 were also performed at 97 K. An anisotropic signal (g1 = 2.023, g2 = 2.222, g3 = 2.305) with high-resolution ultrafine splitting from the Co core was observed (I = 7/2, A1 = 264.00 MHz) , A2 = 64.50 MHz, A3 = 60.00 MHz) (Fig. 4b). In addition, 4 can be prepared by reacting 5 with CO and O2 (Fig. 2f). Supplementary Figure 21 depicts the likely formation pathway of 3 to 5 via the proposed dinuclear cobalt(III) hydroxyl intermediate.
        ORTEP representation of cationic moiety 5 (50% probability). Crystallization solvent molecules, counteranions, and hydrogen atoms (other than oxygen atoms) have been omitted for clarity. For details of this X-ray structure, see Supplementary Data 1. b Solid-state experimental (black) and simulated (red) X-band EPR spectra b 5 .
        The relatively small distance (3.419 Å) between the two carbon atoms of the -COOH ligand in 4 prompts us to further investigate the formation of oxalic acid in 4. Use a xenon lamp as a light source to illuminate 4 under N2. At room temperature, the formation of oxalic acid was observed with a yield of 57%. When replacing the N2 atmosphere with a CO/O2 gas mixture (1:1 v/v, 2 atm), the catalytic production of oxalic acid showed good selectivity (>95%) with a turnover number (TON) of 38.5 (Table 1). . Only trace amounts (TOH ~ 0.3) of dimethyl carbonate are formed, which is a common by-product in the production of oxalic acid31,34,35 catalyzed by Pd(II) complexes. At the same time, the dimethyl oxalate process was not detected. in this process. It is noteworthy that the formation of H2O2 (TON ~ 0.1) was also detected, which may indicate an intermediary role of binuclear hydroxycobalt(III) complexes in the catalytic cycle. It is worth noting that during the catalytic production of oxalic acid, precipitation of a yellow solid was observed from the reaction solution due to catalyst decomposition. XPS measurements of this sediment revealed the presence of an unidentified Co(II) complex (Supplementary Figure 23). It is worth noting that 3 and 5 are also capable of catalyzing the production of oxalic acid under the same conditions, but with less efficiency, with TON of 10.4 and 11.3, respectively, while 1 and 2 are not suitable for this catalysis (Table 1). . To further confirm the origin of the carbonyl and hydroxyl groups in the oxalic acid products, isotope labeling experiments were performed. In 13C labeling experiments, the resulting oxalic acid was converted to calcium oxalate by reaction with CaCl2 and then collected for infrared measurements. When using 13CO, all recorded stretching frequencies of C=O, C(O)-OH and (O)CC(O) calcium oxalate bonds are red-shifted, confirming the formation of Ca13C2O4 (Supplementary Figure 24). Similar trends were observed for 13C-labeled and unlabeled sodium and potassium oxalates45,46. In 18O labeling experiments using H218O and 16O2, only 18OH-labeled oxalic acid was observed by MS measurements (m/z- = 93.0 [MH]-, Supplementary Fig. 25b). In contrast, when H216O and 18O2 entered the catalytic system, no 18OH-labeled products were observed (Supplementary Figure 25c). These results confirm that the source of hydroxyl groups in the resulting oxalic acid is H2O and not O2.
        Considering all the above results, a reasonable mechanism for the catalytic production of oxalic acid 4 is shown in Figure 5. Moreover, all these proposed pathways based on DFT calculations were studied computationally using the ORCA package47,48 (Figure 5). 6). The classical migratory insertion of CO into the Co(III)-OH bond in the binuclear intermediate hydroxycobalt(III) is generally a reasonable pathway for the formation of 4. The calculated penalty energy surface (PES) of the pathway based on the insertion migration is shown in Figure 6a. The coordination of CO to the Co(III) center in IN 1 is substantially endothermic, with a Gibbs free energy change of 26.7 kcal/mol. Upon subsequent migrating insertion of CO, a -COOH group is formed with a Gibbs free energy change (ΔG) of 50.9 kcal/mol and a very low calculated activation energy (Ea = 0.2 kcal/mol). When CO is coordinated with the Co(III) center in IN 3, the Gibbs free energy increases by 37.3 kcal/mol. Although subsequent migrating CO incorporation requires only a small activation energy (Ea = 3.0 kcal/mol) and is exothermic (ΔG = −62.4 kcal/mol), the entire pathway is unlikely to actively occur at ambient temperature. Adverse involvement in the reaction. CO coordination process. As an alternative, a photopromoted generation pathway 4 has been proposed, in which the Co(III)-OH bond is photocleaved before reaction with CO (Supplementary Figure 30 and Figure 6b). When CO is first introduced, the coordination of CO with the Co(II) center in IN 5 is moderately exothermic (ΔG = -10.1 kcal/mol), and the subsequent attack of the hydroxyl radical on the coordinated CO completes the first – The COOH group has a strong energetic advantage (ΔG = -61.1 kcal/mol). It is worth noting that the bond of the Co(II) center with the hydroxycarbonyl group is formed due to the possible reaction of hydroxyl radicals with CO (ΔG = -35.2 kcal/mol), which is also a possible way of introducing CO. The second CO insertion calculates a very similar PES. The calculated PES supports the formation of 4 through the light-stimulated pathway.
        Gibbs free energy curve for the formation of 4 from a binuclear hydroxycobalt(III) complex via: classical CO b migratory insertion, photostimulated pathway; 4 Gibbs free energy curve for the formation of oxalic acid: c direct combination of -COOH groups; d pathway based on hydroxycarbonyl attack; Bond lengths and distances are reported in Å, and detailed structures of intermediate and transition states are summarized in Supplementary Data 2. Counteranions have been omitted for clarity.
        To obtain oxalic acid from 4, three different routes can be envisaged (Fig. 5): (i) direct binding of the -COOH group by “bimetallic” reductive elimination (ii) Co(III)-COOH bond in 4; One undergoes photocleavage and the hydroxycarbonyl radical subsequently attacks the intact Co(III)-COOH bond, forming oxalic acid (iii) through the combination of the resulting free hydroxycarbonyl radical 4 and oxalic acid Co(III)-The; All COOH bonds undergo photocleavage. The first two paths are further explored computationally. It has been established that direct binding of -COOH groups has very high activation energies (Ea = 55.1 kcal/mol for the formation of sE-oxalic acid, Ea = 54.2 kcal/mol for the formation of sZ-oxalic acid), which excludes their use at ambient temperature (Fig. 6c). For pathways based on the attack of hydroxycarbonyl radicals, the calculated Gibbs free energy distributions show that they all have different multiplicities (S = 0 and 1) and hydrogen bond structures (OH…O = C-OH and OH…(OH). The radical process attacks C = O, which leads to the formation of sE-oxalic acid and sZ-oxalic acid, respectively) only a small energy barrier needs to be overcome (7–9 kcal/mol), and in the case of a change in the Gibbs free energy by 32 is strongly energetically maintained at – 34 kcal/mol (Fig. 6d and Supplementary Fig. 31). Thus, both photopromoted pathways initiated by single-photon and double photocleavage of the Co(III)-COOH bond in 4 are thought to be responsible for the formation of oxalic acid catalyzed by 4 .
        In summary, we have synthesized and characterized a series of dicobalt complexes (1-4) with planar macrocyclic ligands. Light irradiation of the rare coplanar binuclear carbonyl complex of hydroxycobalt(III) (4) leads to photocleavage of the Co(III)-COOH bond and the formation of oxalic acid. Inspired by this result, a strategy was developed to directly and selectively produce oxalic acid with good atom economy through oxidative coupling of CO under 4-mediated environmental conditions. 13C and 18O labeling experiments confirmed that CO and H2O are the sources of the -COOH groups in products 4 and oxalic acid. The presented results may provide the basis for the development of new strategies to reduce CO2 emissions and shed light on the development of bimetallic composite platforms with new reactivity. Further research to expand the reaction range of this bimetallic system is ongoing.
        All operations with air-sensitive materials were carried out under a nitrogen atmosphere using the standard Schlenk technique or in a glove box. Chemicals were purchased from Sigma-Aldrich, Alfa Aesar, J&K Scientific Ltd. or Cambridge Isotope Laboratory Inc. All chemicals were used without further processing.
        A solution of o-phenylenediamine (0 .4104 g, 3.800 mmol). added dropwise to methanol (40.0 ml) at room temperature over 30 minutes and then stirred for 6 hours. The methanol solution was then removed by filtration and the residual solid was collected, followed by removal of all volatiles in vacuo to obtain the title ligand as a yellow powder (56% yield). 1H NMR (400 MHz, CDCl3) δ (ppm): 13.55 (s, 2H, OH), 8.63 (s, 2H, NCH), 7.42 (s, 1H, ArH), 7 .42 (s, 1H, ArH), 7.35 (s, 1H, ArH), 7.34 (s, 1H, ArH), 7.07 (d, J = 7.72 Hz, 1H, ArH), 7.05 (d, J = 7.72 Hz, 1H, ArH), 6.98 (br., 1H), ArH), 6.96 (br., 1H, ArH), 6.79 (t, J = 7.56 Hz, 1H, ArH), 6.78 (t, J = 7.44 Hz, 1H, ArH), 6.32 (t, J = 5.60 Hz), 2H, ArH), 4, 46 (s, 2H, ArH), 4.45 (s, 2H, ArH), 3.50 (s, 1H, CNH), 3.49 (s, 1H, CNH), 1.33 (s, 18H, tBuH) .
        In a nitrogen-filled glovebox, H2L (0.0589 g, 0.105 mmol) was dissolved in 8.0 mL ethanol, and then Co(OAc)2·4H2O (0.0549 g, 0.220 mmol) was added. The reaction mixture was stirred at 85°C for 6 hours, during which time the solution turned from a yellow suspension to a brown solution. The resulting solution was concentrated under reduced pressure to 1.0 ml and diethyl ether (30.0 ml) was added. The precipitate was collected by filtration and dried in vacuum to obtain 1 product with a yield of 86%. Crystals suitable for X-ray diffraction analysis were grown by slow evaporation of ethanol solution 1 at 1 °C for 24 hours. UV/visible light (ethanol): 304 nm (ε = 5.53 × 103 L mol-1 cm-1), 314 nm (ε = 5.54 × 103 L mol-1 cm-1), 331 nm (ε = 3.61 × 103 L mol-1 cm-1), 412 nm (ε = 2.69 × 103 L mol-1 cm-1). Analysis (calculated, C40H40Co2N4O6 3H2O): C (56.88, 56.81), H (5.49, 5.51), N (6.63, 6.43).
        Complex 1 (0.0844 g, 0.100 mmol) was charged into a 25.0 mL Schlenk flask and methanol (6.0 mL) was added at room temperature. After three freeze-pump-thaw cycles, 1 atm O2 was charged into the Schlenk flask. The solution was then stirred for 30 minutes to obtain a red solution. The resulting solution was concentrated under reduced pressure to 1.0 ml and diethyl ether (10.0 ml) was added. The orange precipitate was collected by filtration and dried in vacuo to give 2 in 81% yield. X-ray quality 2 crystals can be obtained from concentrated methanol solutions within three days at room temperature. UV/visible light (ethanol): 298 nm (ε = 7.43 × 103 L mol–1 cm–1), 418 nm (ε = 2.13 × 103 L mol–1 cm–1), 474 nm (ε = 1.25 × 103 L mol-1 cm-1). Analysis (calculated, C41H44Co2N4O7·H2O): C (58.58, 58.13), H (5.52, 5.46), N (6.66, 6.39).
        Ce(NH 4 ) 2 (NO 3 ) 6 (0.0868 g, 0.158 mmol) was added to a solution of complex 2 (0.1210 g, 0.1440 mmol) in 8.0 ml of methanol. After three freeze-pump-thaw cycles, a 25.0 mL Schlenk flask was charged with 1 atm O2. The reaction mixture was stirred at 80°C for 1 hour, which resulted in a color change from red to brown. The solution was concentrated under reduced pressure to 4.0 mL and diethyl ether (10.0 mL) was added. The brown precipitate was collected by filtration and then dried in vacuo to give 3 in 86% yield. X-ray quality 3 crystals can be obtained from concentrated methanol solutions within three days at room temperature. UV/visible light (ethanol): 300 nm (ε = 7.02 × 103 L mol-1 cm-1), 331 nm (ε = 7.58 × 103 L mol-1 cm-1), 438 nm (ε = 1.60 × 103 L mol-1 cm-1). 1H NMR (400 MHz, CD3OD) δ (ppm): 9.22 (s, 2H, NCH), 9.12 (br., 2H, NCH), 8.50–8.47 (m, 4H, ArH), 8.34–8.32 ( m, 4H, ArH), 7.79–7.72 (m, 4H, ArH), 1.66 (s, 3H, CH3COO), 1.47–1.46 (br., 18H, tBuH). IR (potassium bromide disc technology): ν(OH) = 3425 cm-1. Analytical (calculated for C78H82CeCo4N14O30 H2O): C (44.84, 44.99), H (4.05, 4.13), N (9.39, 9.82).
        This complex can be synthesized by two different methods. Method A: A solution of complex 3 (0.0350 g, 0.0170 mmol) in 4.0 mL methanol was degassed by three freeze-pump-thaw cycles in a 25.0 mL Schlenk flask. Fill the Schlenk flask with 1 atm O2 and then add 1 atm CO. The solution was stirred at 50°C for 24 hours. The red precipitate was collected by filtration to obtain 4, which was an analytically pure dark red powder in 26% yield. Crystals suitable for X-ray diffraction analysis were obtained by slow evaporation of methanol solution 4 at room temperature. Method B: A solution of complex 5 (0.0320 g, 0.00800 mmol) in 4.0 mL methanol was degassed by three freeze-pump-thaw cycles in a 25.0 mL Schlenk flask. Fill the Schlenk flask with 1 atm O2 and then add 1 atm CO. The solution was stirred at 50°C for 24 hours. The red precipitate was collected by filtration to obtain 4, which was an analytically pure dark red powder in 39% yield. Crystals suitable for X-ray diffraction analysis were obtained by slow evaporation of methanol solution 4 at room temperature. UV/visible light (ethanol): 304 nm (ε = 6.73 × 103 L mol–1 cm–1), 335 nm (ε = 6.21 × 103 L mol–1 cm–1), 354 nm (ε = 5.74 × 103 l mol-1 cm-1), 369 nm (ε = 4.95 × 103 l mol-1 cm-1), 445 nm (ε = 1.76 × 103 l mol-1 cm- 1). IR (potassium bromide disc technology): ν(C = O) = 1697 and 1670 cm-1 ν(13C = O) = 1662 and 1651 cm-1; 1H NMR (400 MHz, DMSO) δ (ppm): 13.34 (s, 1H, COOH), 13.08 (s, 1H, COOH), 9.49 (s, 4H, NCH), 8 .46 (br., 8H, ArH), 7.78 (s, 4H, ArH), 1.60 (s, 3H, CH3COO), 1.45 (br., 18H, tBuH). 13C NMR (400 MHz, DMSO) delta (ppm) 4-13C: 172.12 (s, -COOH). Analysis (calculated, C40H40Co2N5O11·H2O): C (53.28, 53.49), H (4.58, 4.67), N (7.77, 7.55).
        A solution of complex 3 (0.0299 g, 0.0330 mmol) in 8.0 mL ethanol was degassed by freeze-pump-thaw cycles in a 25.0 mL Schlenk flask. Oxygen was loaded into a Schlenk flask under a pressure of 1 atmosphere. The solution was stirred at 80°C for 6 hours. The precipitate was filtered off and the filtrate was dried in vacuum. The residue was washed with cold ether (-20°C) to obtain 5 in the form of a brown powder with a yield of 42%. Crystals suitable for X-ray diffraction analysis were obtained by slow evaporation of an ethanol solution 5 at room temperature. UV/visible light (ethanol): 300 nm (ε = 6.47 × 103 L mol–1 cm–1), 324 nm (ε = 6.26 × 103 L mol–1 cm–1), 442 nm (ε = 1.41 × 103 L mol-1 cm-1). IR (potassium bromide disc technology): ν(OH) = 3379 and 3209 cm-1. Analytical (calculated for C152H152CeCo8N26O50·H2O): C (48.39, 47.98), H (4.09, 4.26), N (9.66, 9.58).
        Transfer a total of 4.0 mL of methanol solution containing 0.0170 mmol of catalyst (complex 1-5) into a 25.0 mL Schlenk flask. After three freeze-pump-thaw cycles, the Schlenk flask was charged with 1 atm O2, and then 1 atm CO was added. Place the Schlenk flask 20.0 cm from a 500 W xenon lamp and leave it at 30 °C for 28 hours. HOUR. After the reaction, the precipitate was filtered off and the filtrate was analyzed by LCMS.
        All calculations were performed in the ORCA quantum chemistry package (version 5.0.3) using the B97-3c calculation setup. The setup is based on the functionality of the B97 GGA, including D3 with three-body contributions and short-range coupling length correction. This setup used a modified trimmed triple z base def2-mTZVP50. For the larger tetracobalt complex 5, the crystal structure with optimized hydrogen atom positions was used to analyze the corresponding orbital and spin densities without constraints.
        Data related to this study can be found in the Supporting Information and Supplementary Data article. 1 (CIF of the X-ray structure) and 2 (coordinates of the optimized structure used in the computational study). Crystallographic data for the structures described in this paper have been deposited at the Cambridge Structural Data Center under CCDC numbers 2209326, 2209329, 2209330, 2209332 and 2209333. Copies of the data are available at https://www.ccdc.cam.ac.uk/structs/. All data are available from the corresponding author.
        Braunschweig, H. et al. Metal-free coupling and carbon monoxide coupling at the boron-boron triple bond. Nat. Chemical. 5, 1025–1028 (2013).
        Giuni, M. et al. Formation of carbon-nitrogen bonds during the electrolysis of carbon monoxide. Nat. Chemical. 11, 846–851 (2019).
        Sharp, HR, et al. Selective reduction and homologation of carbon monoxide by organometallic iron complexes. Nat. communicate. 9, 3757 (2018).
        Evans, W.J., Great, J.W., Hughes, L.A., Zhang, H. and Atwood, J.L. Reduction of CO to ketene carboxylates via low-valent lanthanide organocomplexes: synthesis and synthesis of [(C5Me5)4Sm2(O2CCCO) x-ray crystal structure (tetrahydrofuran)]2. J. Chem. Sociological Society 107, 3728–3730 (1985).
        Knobloch, D., Lobkowski, E. and Chirik, P. Hafnium complexes promote dinitrog cleavage and carbon monoxide functionalization. Nat. Chemical. 2, 30–35 (2010).
        Bass, J. A. et al. Chemistry of CO reaction with terminal molybdenum carbide: proton, hydride and CO added sequentially to release ketene. J. Chem. Bitch. 141, 15664–15674 (2019).
        Liu, H.-Y. and others. Na/Mg(I) diamide performs reductive dimerization of CO. J. Chem. Bitch. 143, 17851–17856 (2021).
        Wayland, B. and Fu, X. Molecule design using carbon monoxide reductive bonding. Science 311, 790–791 (2006).
        Dobrovetsky R. and Stefan D.V. Stoichiometric metal-free reduction of CO in synthesis gas. J. Chem. Bitch. 135, 4974–4977 (2013).
        Prochenko, A.V. et al. Reduced carbon oxides of acyclic silicene: reductive combination of CO. Angew. Chemical. internationality. Ed. 58, 1808–1812 (2019).
        Hu, S., Shima, T., and Hou, Z. Hydrodeoxygenation, cyclotetramerization of carbon monoxide by trinuclear polytitanium hydride complexes. J. Chem. Bitch. 142, 19889–19894 (2020).
        Bass, J. A. and Agapi, T. Mechanism of molybdenum-mediated deoxygenation and carbon monoxide sequestration: mono- and dicarbon alkyne complexes precede CO bond cleavage and CC bond formation. J. Chem. Bitch. 138, 16466–16477 (2016).
        Sazama, G.T. and Betley, T.A. Reductive CO sequestration using iron bound to a tris(pyrrole)ethane framework as a template. Metalorganics 30, 4315–4319 (2011).
        Coffin, W.L., Brennen, W., and Wayland, B.B. Thermodynamic study of competitive adduct formation: single and double insertion reactions of carbon monoxide with rhodium octaethylporphyrin dimer. J. Chem. Bitch. 110, 6063–6069 (1988).
        Bianconi, P.A., Williams, I.D., Engeler, M.P. and Lippard, S.J. Reductive combination of two carbon monoxide ligands to form coordinated alkynes. J. Chem. Bitch. 108, 311–313 (1986).
        Miller, A.J.M., Labinger, J.A. and Bercow, J.E. Homogeneous hydrogenation of CO: Effect of ligands on reductive coupling of carbon monoxide by Lewis acid. Metalorganics 29, 4499–4516 (2010).
        Miller, A.J.M., Labinger, J.A. and Bercow, J.E. Homogeneous CO hydrogenation: Dihydrogen activation involves quenching Lewis pairs rather than platinum complexes. J. Chem. Bitch. 132, 3301-3303 (2010).
        Miller, A.J.M., Labinger, J.A. and Bercow, J.E. Reductive binding of carbon monoxide to side chain Lewis acids in rhenium carbonyl complexes. J. Chem. Bitch. 130, 11874–11875 (2008).
        Wayland B.B., Sherry A.E. and Coffin V.L. Double-stranded RNA controls the phase separation of SARS-CoV-2 nucleocapsid proteins at specific temperatures. J.Chemistry. Chemical Society. communicate. 662–663 (1989).
        Chatani N., Shinohara M., Ikeda S.-I. and Murai, S. Reductive oligomerization by rhodium-catalyzed reaction of carbon monoxide with hydrosilanes. J. Chem. Sociological Society 119, 4303–4304 (1997).
        Tsureas N., Summerscales O.T., Klock F.G.N. and Rowe S.M. Steric effects of mixed uranium(III) sandwich complexes during the reductive interaction of CO. Metalorganics 32, 1353–1362 (2013).
        Arnold P.L., Turner Z.R., Bellabarba R.M. and Tuz R.P. Carbon monoxide coupling and functionalization of simple uranium coordination complexes. Chemical. Science 2, 77–79 (2011).
        Shima, T. and Hou, Z. Hydrogenation of carbon monoxide by rare earth tetranuclear polyhydrogen complexes. Selective ethylene formation and separation of well-defined polyoxyrare earth metal clusters. J. Chem. Sociological Society 128, 8124–8125 (2006).