|Home | About | Journals | Submit | Contact Us | Français|
Metal-O2 adducts, such as metal-superoxo and -peroxo species, are key intermediates often detected in the catalytic cycles of dioxygen activation by metalloenzymes and biomimetic compounds. The synthesis and spectroscopic characterization of an end-on nickel(II)-superoxo complex with a 14-membered macrocyclic ligand was reported previously. Here we report the isolation, spectroscopic characterization, and high-resolution crystal structure of a mononuclear side-on nickel(III)-peroxo complex with a 12-membered macrocyclic ligand, [Ni(12-TMC)(O2)]+ (1) (12-TMC = 1,4,7,10-tetramethyl-1,4,7,10-tetraazacyclododecane). Different from the end-on Ni(II)-superoxo complex, the Ni(III)-peroxo complex is not reactive in electrophilic reactions, but is capable of conducting nucleophilic reactions. The Ni(III)-peroxo complex transfers the bound dioxygen to manganese(II) complexes, thus affording the corresponding nickel(II) and manganese(III)-peroxo complexes. The present results demonstrate the significance of supporting ligands in tuning the geometric and electronic structures and reactivities of metal-O2 intermediates that have been shown to have biological as well as synthetic usefulness in biomimetic reactions.
Metalloenzymes activate dioxygen to carry out a variety of biological reactions including biotransformation of naturally occurring molecules, oxidative metabolism of xenobiotics, and oxidative phosphorylation. One goal in biomimetic research is to understand the mechanistic details of dioxygen activation and oxygenation reactions and the structures of reactive intermediates occurring at the active sites of the metalloenzymes1. In the unified mechanism of dioxygen activation, dioxygen first binds to a reduced metal center that forms metal-superoxo and -peroxo intermediates, followed by O-O bond cleavage leading to the formation of high-valent metal-oxo species that are believed to carry out substrate oxidations1. Among the metal-oxygen intermediates, mononuclear metal-O2 adducts, such as metal-superoxo and -peroxo species, have attracted much attention as key intermediates in the catalytic cycles of dioxygen activation by metalloenzymes, including heme and non-heme iron and copper enzymes2-4. In biomimetic and synthetic chemistry, mononuclear metal-O2 complexes, including titanium, vanadium, chromium, manganese, iron, cobalt, nickel, copper, and the second and third row transition metals, have been synthesized and characterized with various spectroscopic techniques and X-ray crystallography, and their reactivities have been extensively investigated in nucleophilic and electrophilic oxidation reactions5-11. A notable example is the mononuclear copper-O2 species, which shows a diverse and rich chemistry in structures, spectroscopic properties, and reactivities10-16. X-ray crystal structures of side-on (η2) and end-on (η1) Cu(II)-superoxo and side-on (η2) Cu(III)-peroxo complexes were successfully obtained17-19, and the mode of O2 coordination (e.g., side-on vs end-on O2-binding) and the electronic nature of the Cu-O2 core (e.g., Cu(II)-superoxo vs Cu(III)-peroxo) were found to vary depending on the supporting ligands of copper complexes20-23.
In the case of mononuclear Ni-O2 intermediates, side-on and end-on nickel(II)-superoxo and side-on nickel(II)-peroxo complexes have been characterized by spectroscopic, X-ray crystallographic, and computational methods24-27, and the Ni(II)-superoxo complexes showed electrophilic reactivity, such as the oxidation of PPh3 to OPPh324-26. However, to the best of our knowledge, the crystal structure and reactivity of a mononuclear Ni(III)-peroxo complex have not yet been reported. In this study, we have examined the effects of supporting ligands on the mode of O2 coordination and the electronic structure of the Ni-O2 moiety in mononuclear Ni-O2 complexes, by varying the ring–size of a macrocyclic ligand coordinated to [Ni(II)(14-TMC)(O2)]+ (2) (14-TMC = 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetradecane) that was characterized as an end-on Ni(II)-superoxo complex by spectroscopic and computational methods25. We now report for the first time the synthesis, spectroscopic and electronic properties, and crystal structure of a mononuclear side-on (η2) nickel(III)-peroxo complex stabilized by a 12-membered macrocyclic ligand, [Ni(III)(12-TMC)(O2)]+ (1) (12-TMC = 1,4,7,10-tetramethyl-1,4,7,10-tetraazacyclododecane). The reactivities of the Ni(III)-peroxo complex in electrophilic and nucleophilic reactions and peroxo group transfer to other metal complexes have been discussed as well.
The starting nickel complex, [Ni(12-TMC)(CH3CN)]2+ (3), was synthesized and characterized with UV-vis absorption spectroscopy, electrospray ionization mass spectrometry (ESI MS), and X-ray crystallography (see Supplementary text, Figs. S1 and S2, Tables S1 and S2). The reaction of 3 with 5 equiv H2O2 in the presence of 2 equiv triethylamine (Et3N) in CH3CN at 0 °C produces a green intermediate, 1, that exhibits distinct absorption features that are different from those of 2 (Fig. 1)25. The intermediate persisted for several days at 25 °C, and the greater thermal stability of 1 allowed us to isolate crystals that were used in spectroscopic and structural analyses and reactivity studies. The ESI MS of 1 exhibits a prominent ion peak at a mass-to-charge (m/z) ratio of 318.0 (see Supplementary Fig. S3A), whose mass and isotope distribution pattern correspond to [Ni(12-TMC)(O2)]+ (calculated m/z of 318.2) (see Supplementary Fig. S3A, inset). When the reaction was carried out with isotopically labeled H218O2, a mass peak corresponding to [Ni(12-TMC)(18O2)]+ appeared at m/z of 322.0 (calculated m/z of 322.2) (see Supplementary Fig. S3A, inset). The 4-mass unit upshift upon the substitution of 16O with 18O indicates that 1 contains an O2 unit. The EPR spectrum of a frozen acetonitrile solution of 1 measured at 4.3 K exhibits a rhombic signal with g values of 2.22, 2.17 and 2.06 (see Supplementary Fig. S3B), which is indicative of a (dz2)1 electron configuration typically observed for Ni(III) species28, Ni(II)-superoxo complexes24-26, and a six-coordinated bis(μ-superoxo)Ni2(II) complex29. The room temperature magnetic moment of 2.13 μB, determined using the 1H NMR Evans method30, is consistent with an S = 1/2 ground state of 1.
The resonance Raman spectrum of 1 was collected using 442-nm excitation at −20 °C. 1 prepared with H216O2 exhibits an isotopically sensitive band at 1002 cm−1 that shifts to 945 cm−1 when H218O2 is used, consistent with its assignment as an O-O stretching vibration on the basis of its 16Δ−18Δ value of 57 cm−1 (16Δ−18Δ (calcd) = 57 cm−1) (Fig. 1, inset). Interestingly, the observed O-O stretching frequency (1002 cm−1) of 1 is significantly lower than that (1131 cm−1) of 225, but is between the superoxo (i.e., νOO of ~1050–1200 cm-1) and peroxo (i.e., νOO of ~800–930 cm-1) categories31. It is worth noting that the O-O stretching frequency (1002 cm−1) of 1 is quite close to those observed in Cu(III)-peroxo species (~970 cm-1)18,22. Thus, the resonance Raman spectroscopic data suggests that the O2 in 1 has significant peroxo character, O22−.
The X-ray crystal structure of 1-(ClO4)·CH3CN revealed the mononuclear side-on 1:1 nickel complex of O2 in a distorted octahedral geometry arising from the triangular NiO2 moiety with a small bite angle of 43.04(11)° (Fig. 2). Notably, the O-O bond length (1.386(4) Å) of 1 is longer than those of Ni(II)-superoxo complexes, such as 2 (1.301 Å, from DFT calculations)25 and Ni(II)(O2) with a β-diketiminato ligand (1.347 Å)26; the O-O bond length of 1 is between those of metal-superoxo compounds (~1.2–1.3 Å) and metal-peroxo compounds (~1.4–1.5 Å)31, but closer to the metal-peroxo category. For comparison, the O-O bond length of 1.392 Å was reported in a Cu(III)-peroxo complex22. In addition, the Ni-O average bond of 1 (1.889 Å) is shorter than that of 2 (1.984 Å, from DFT calculations)25, supporting a Ni(III) formulation. In conjunction with the low O-O stretching frequency measured by Raman spectroscopy, the structural data, such as O-O and Ni-O bond distances, suggest that 1 can be formulated as a Ni(III)-peroxo species, Ni(III)-(O22−).
Ni K-edge X-ray absorption spectroscopy was then performed to directly probe the oxidation state of nickel and the ligand field (LF) of the nickel center. The normalized Ni K-edge X-ray absorption spectra of 1 and 2 are presented in Fig. 3. The inset shows the expanded pre-edge region. The pre-edge features are due to an electric dipole-forbidden quadrupole-allowed 1s→3d transition32. The energy position of the pre-edge transition is dominantly affected by changes in LF at the absorbing Ni center and increases with increase in LF23. The pre-edge transitions in 1 and 2 occur at 8332.3 eV and 8331.6 eV, respectively, indicating an increase in LF on going from 2 to 1 (Fig. 3). The Ni K-edge energy position increases with increase in QNi, the charge on the absorbing Ni center in the complex. Typically, for Ni complexes, the K-edge first-maxima do not show a large change with change in QNi33; however, as seen in Fig. 3 the first-maximum is shifted ~1.8 eV in going from 2 to 1, clearly indicating an increase in QNi in 1 compared to 2. Ni K-edge EXAFS data and their analyses show that the first shell coordination has increased from 5 in 2 to 6 in 1 (see Supplementary Figs. S4–S6, Tables S3 and S4), yet the Ni-O bond distance is ~0.06 Å shorter in 1 than in 2. Together, the Ni K-edge and EXAFS data support a unique Ni(III)-(O22−) description for 1. These data combined with the crystal structure indicate that the O2 binds side-on to the Ni center in 1, whereas the O2 is end-on bound in 2 as previously reported25. This binding mode difference is accompanied by an electronic structure change from Ni(II)-(O2−) in 2 to Ni(III)-(O22−) in 1.
Density functional theory (DFT) calculations were performed to understand the role of the macrocyclic ring in determining the geometric and electronic structures of 1 and 2 (see Supplementary Fig. S7 and Table S5 for structural details). The geometry optimized structures of 1 and 2 show that the smaller 12-membered ring in 1 leads to contraction of the trans Ni-N bond angles. This displaces the Ni out of the N1N2N3N4 plane and allows for its facile side-on overlap with the O2. The side-on coordination of the O2 in 1 allows for a stronger overlap of its π* orbital with the Ni 3dx2-y2 orbital. This stronger σ overlap leads to the relative destabilization of the Ni 3dx2-y2 orbital, transfers charge from the Ni 3dx2-y2 orbital to the O2, and results in a stable Ni(III)-(O22−) species. We therefore conclude, based on the spectroscopic results, X-ray crystallography, and DFT calculations, that 1 is best described as a mononuclear Ni(III)-peroxo complex. The role of the supporting ligand, 12-TMC, is to sterically allow O2 to bind in a side-on fashion, resulting in more sigma anti bonding with 3dx2-y2 orbital as indicated above. It also stabilizes the high oxidation state of Ni3+, by the strong interaction of Ni-Nequatorial bond (2.059 Å) in 1 compared to Ni-Nequatorial bond (2.10 Å from DFT calculations) in 2 (see Supplementary Table S5). Thus, as discussed in the mononuclear Cu-O2 chemistry10-17,20-23, the geometric and electronic structures of Ni-O2 species are modulated by the nature of supporting ligands (Fig. 4).
The activity of 2 was investigated in electrophilic and nucleophilic reactions. First, the electrophilic character of 1 was tested in the oxidation of PPh3, thioanisole, and xanthene. Upon addition of the substrates to the reaction solution of 2 in CH3CN at 25 °C, the intermediate remained intact without showing any absorption spectral changes, and product analysis of the reaction solutions revealed that no oxygenated products were formed in the reactions. These results demonstrate that 2 is not capable of conducting electrophilic oxidation under the reaction conditions. In contrast, Ni(II)-superoxo complexes have shown reactivities in electrophilic reactions, such as the oxidation of PPh325,26 and xanthene34. The nucleophilic character of 1 was investigated in aldehyde deformylation, with precedents that metal(III)-peroxo complexes with heme and non-heme ligands react with aldehydes to give the corresponding deformylated products6,35,36. Upon addition of 2-phenylpropionaldehyde (2-PPA) to 1 in CH3CN at 25 °C, the characteristic UV-vis absorption bands of 1 disappeared with a first-order decay profile (see Supplementary Fig. S8A), and pseudo-first-order rate constants increased proportionally with the aldehyde concentration (k2 = 4.0 × 10−2 M−1 s−1 at 25 °C) (see Supplementary Fig. S8B). Similar results were obtained in the reactions of cyclohexanecarboxaldehyde (CCA) but with a faster rate (k2 = 2.0 × 10-1 M−1 s−1 at −10 °C) (see Supplementary Fig. S9). Product analysis of the resulting solutions revealed the formation of acetophenone (92 % based on 1) and cyclohexene (85 % based on 1) in the reactions of 2-PPA and CCA, respectively. The reactivity of 1 was further investigated using substituted benzaldehydes with a series of electron-donating and -withdrawing substituents at the para-position of the phenyl group (para-Y-Ph-CHO; Y = Me, F, H, Br, Cl) (see Supplementary Fig. S10). A positive ρ+ value of 6.1 in the Hammett plot was obtained that is consistent with the process involving nucleophilic character of the Ni(III)-O2 unit in the oxidation of aldehydes.
More interestingly, we have observed a complete intermolecular O2-transfer from 1 to different transition metal complexes, such as [Mn(II)(14-TMC)]2+ (4) (Fig. 5). Addition of 4 to a solution of 1 afforded changes in the absorption spectrum which are consistent with transfer of the O2 from 1 to 4, thereby producing 3 and a manganese(III)-peroxo complex, [Mn(III)(14-TMC)(O2)]+ (5) (Fig. 6)36. Well-defined isosbestic points were observed at 416 and 687 nm in the titration reaction (Fig. 6). The intermolecular O2-transfer from 1 to 4 was further confirmed by ESI MS of the reaction solution, in which the mass peak corresponding to 1 disappeared with a concomitant appearance of mass peaks corresponding to 3 and 5 (see Supplementary Fig. S11). When the O2-transfer reaction was carried out under an 18O2 atmosphere, the product 5 did not contain the isotopically labeled 18O2 group, demonstrating that molecular oxygen was not involved in the peroxo-transfer reaction. Since the peroxo-transfer reaction was fast at 25 °C in CH3CN, kinetic studies were performed in acetone at −50 °C. Upon addition of 10 equiv of 4 to the solution of 1, 1 disappeared with a first-order decay profile (see Supplementary Fig. S12A). Pseudo-first-order fitting of the kinetic data allowed us to determine the kobs value to be 1.7(2) × 10−3 s−1 at −50 °C (see Supplementary Fig. S12A, inset). The first-order rate constants increased proportionally with the concentration of 4, giving a second-order rate constant of 0.2 M−1 s−1 at −50 °C (see Supplementary Fig. S12B). The rates were dependent on reaction temperatures, from which a linear Eyring plot was obtained between −60 and −30 °C to give the activation parameters of ΔH‡ = 49 kJ mol−1 and ΔS‡ = −76 J mol−1 K−1 (see Supplementary Fig. S12C). The observed second order kinetics and significant negative entropy value support that a bimolecular mechanism is operating in the O2-transfer reaction, where the formation of the [(12-TMC)Ni–O2–Mn(14-TMC)]3+ intermediate is presumed to be the rate-determining step (Fig. 5). Finally, we found that the reverse reaction, which is the peroxo-transfer from 5 to 3, does not occur. In this section, we have shown the first example of the complete O2-transfer between metal complexes. The observation of the complete O2-transfer from a Ni(III)-peroxo complex to a Mn(II) complex is different from the behavior of other systems where the formation of homo- or hetero-dinuclear complexes comprising [M2(μ-O)2]n+, [M2(O2)]n+, or [MM′(μ-O)2]n+ cores in the reactions of mononuclear M-O2 adducts (M = Cu and Ni) and a second metal complex, M or M′, is observed18,24,37,38. The formation of (Porp)FeIII-(O22−)-CuII(L) complexes has also been observed in the reactions of Fe(III)-peroxo porphyrins with Cu(II)(L) complexes as chemical models of cytochrome c oxidase39. Detailed mechanistic investigations are underway in this laboratory to elucidate the difference between the complete O2-transfer and the O2-bridged dinuclear formation occurring in the reactions of mononuclear M-O2 and second metal complexes.
In conclusion, a mononuclear side-on Ni(III)-peroxo complex was successfully synthesized by varying the supporting ligand of the previously reported Ni(II)-superoxo complex (i.e., the ring-size of the macrocyclic ligand). Combined with the precedents of mononuclear Cu-O2 intermediates14,17,18, the present results highlight the importance of supporting ligands in determining geometric and electronic structures of mononuclear Ni-O2 complexes (i.e., Ni(III)-peroxo vs Ni(II)-superoxo species). Whether other factors, such as solvents, influence the geometric and electronic structures of Ni-O2 complexes will be the subject of future studies. The reactivities of Ni(III)-peroxo and Ni(II)-superoxo complexes were compared in electrophilic and nucleophilic reactions. In contrast to Ni(II)-superoxo complexes, which show reactivities in oxidative electrophilic reactions24, the Ni(III)-peroxo complex is not reactive in electrophilic reactions but is capable of deformylating aldehydes through nucleophilic reactions. The observation of complete O2-transfer between metal complexes is unprecedented; whether this is a general feature in other metal-O2 adducts requires further experimental and computational study.
See experimental section in supplementary information for detailed experimental conditions and procedures, spectroscopic and kinetics analyses, and computational calculations.
The research was supported by KOSEF/MEST of Korea through the CRI and WCU (R31-2008-000-10010-0) Programs (W.N.), the Ministry of Education, Culture, Sports, Science and Technology of Japan through the Global COE program and Priority Area (No. 20050029) (T.O.), and NIH grant DK-31450 (E.I.S.). SSRL operations are funded by the Department of Energy, Office of Basic Energy Sciences. The SSRL Structural Molecular Biology program is supported by the National Institutes of Health, National Center for Research Resources, Biomedical Technology Program, and the Department of Energy, Office of Biological and Environmental Research.
Author contributions: J.C., E.I.S., and W.N. conceived and designed the experiments; J.C., R.S., J.A., S.Y.K., and M.K. performed the experiments; J.C., R.S., J.A., M.K., and T.O. analyzed the data; J.C., R.S., E.I.S., and W.N. co-wrote the paper.
Additional information: Supplementary information and chemical compound information accompany this paper at www.nature.com/naturechemistry. Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/