Communication pubs.acs.org/JACS
Distinct Reactivity of a Mononuclear Peroxocobalt(III) Species toward Activation of Nitriles Hyeonju Noh,† Donghyun Jeong,† Takehiro Ohta,‡ Takashi Ogura,‡ Joan Selverstone Valentine,§ and Jaeheung Cho*,† †
Department of Emerging Materials Science, DGIST, Daegu 42988, Korea Picobiology Institute, Graduate School of Life Science, University of Hyogo, RSC-UH LP Center, Hyogo 679-5148, Japan § Department of Chemistry and Biochemistry, UCLA, Los Angeles, California 90095-1569, United States ‡
S Supporting Information *
peroxocobalt(III) complex, [Co III (TBDAP)(O 2 )] + (1; TBDAP = N,N-di-tert-butyl-2,11-diaza[3.3](2,6)-pyridinophane), reacts with nitrile derivatives (RCN; R = Me, Et, and Ph) in a dioxygenase-like manner to afford a series of hydroximatocobalt(III) complexes, [CoIII(TBDAP)(R-C( NO)O)]+ where R = Me (2), Et (3), and Ph (4) (Scheme 1). The one-pot activation of nitriles through 1, which is readily
ABSTRACT: A mononuclear side-on peroxocobalt(III) complex with a tetradentate macrocyclic ligand, [CoIII(TBDAP)(O2)]+ (1), shows a novel and facile mode of dioxygenase-like reactivity with nitriles (RCN; R = Me, Et, and Ph) to produce the corresponding mononuclear hydroximatocobalt(III) complexes, [CoIII(TBDAP)(RC(NO)O)]+, in which the nitrile moiety is oxidized by two oxygen atoms of the peroxo group. The overall reaction proceeds in one-pot under ambient conditions (ca. 1 h, 40 °C). 18O-Labeling experiments confirm that both oxygen atoms are derived from the peroxo ligand. The structures of all products, hydroximatocobalt(III) complexes, were confirmed by X‑ray crystallography and various spectroscopic techniques. Kinetic studies including the Hammett analysis and isotope labeling experiments suggest that the mechanistic mode of 1 for activation of nitriles occurs via a concerted mechanism. This novel reaction would be significantly valuable for expanding the chemistry for nitrile activation and utilization.
Scheme 1. Synthetic Procedures for Hydroximatocobalt(III) Complexes through Reactions of a Peroxocobalt(III) Complex 1 with Nitriles
he activation of nitriles is a field of synthetic interest for the preparation of materials important in fine chemical synthesis and pharmaceutical production,1,2 but typical methods have the disadvantage that strong bases or acids and high temperatures must be required. For that reason, development of methods for functionalization of nitriles under mild conditions is a subject of considerable current interest.3,4 Biocatalysis of nitrile hydrolysis using nitrilase or the iron- or cobalt-containing nitrile hydratase enzymes5−9 is one tactic to address this challenge, and multiple biomimetic approaches using metal complexes are also giving encouraging results.10−14 Peroxometal species are key intermediates in a variety of biomimetic oxidation reactions.15 However, only one example of nitrile group activation by peroxometal complexes has been reported so far: the peroxopalladium(II) complexes that react with CH3CN (MeCN) to give peroxometallacycle products.16 Recently, hydroperoxocobalt(III) species has been proposed to be a reactive intermediate for the CH bond activation of nitriles.17 Herein, we report an entirely new type of nitrile group activation in which the mononuclear side-on
T
© 2017 American Chemical Society
accessible by reaction of a cobalt(II) precursor with H2O2 and triethylamine (TEA), could offer practical applications. For example, the hydroximatocobalt(III) complexes of the type that we report here may prove useful as prodrugs. Hydroximato ligands, which are tautomers of hydroxamato analogues, can act as inhibitors of metalloenzymes, and they have been used as therapeutic agents for cancer18,19 and Alzheimer’s disease.20 Thus, this new approach holds considerable promise for synthetic chemistry including the preparations of some interesting prodrugs.21 The new peroxocobalt(III) complex, [CoIII(TBDAP)(O2)]+ (1), was synthesized by adding 5 equiv of H2O2 to a solution containing a CoII complex, [CoII(TBDAP)(NO3)(H2O)]+ (Supporting Information (SI), Figures S1−S3 for ESI-MS and EPR spectra and X-ray crystal structure of the CoII complex; see also Tables S1 and S3 for crystallographic data of the CoII complex), in the presence of 2 equiv of TEA under Ar in CH3CN at −20 °C, in which the color of the solution changed Received: May 2, 2017 Published: July 31, 2017 10960
DOI: 10.1021/jacs.7b04479 J. Am. Chem. Soc. 2017, 139, 10960−10963
Communication
Journal of the American Chemical Society from pink to green. Complex 1 persisted for several days at −20 °C, which allowed us to isolate and use it in spectroscopic and structural characterization and reactivity studies. The UV−vis spectrum of 1 in MeCN at −20 °C presents two absorption bands at λmax = 722 (ε = 150 M−1 cm−1) and 974 nm (ε = 85 M−1 cm−1) (SI, Figure S4a). Note that the UV−vis absorption spectrum of 1 was changed depending on the temperature (SI, Figure S5a). By increasing temperature, the absorption band at 722 nm decreased with concurrently increasing the absorption intensity at 974 nm probably due to the spin crossing between S = 0 and S = 1 states of 1 (SI, Figure S5b for the data of the superconducting quantum interference device (SQUID) magnetic susceptibility measurements). The ESI-MS spectrum of 1 exhibits a prominent signal at a mass-tocharge (m/z) ratio of 443.2, whose mass and isotope distribution pattern correspond to [Co(TBDAP)(O 2 )] + (1-16O) (calculated m/z of 443.2) (SI, Figure S4b). When the reaction was performed with isotopically labeled H218O2, a mass peak assignable to [Co(TBDAP)(18O2)]+ (1-18O) appeared at m/z of 447.2 (calculated m/z of 447.2) (SI, Figure S4b). The four mass unit shifts on the substitution of 16O with 18 O indicate that 1 contains two oxygen atoms. The EPR spectrum of 1 recorded at 103 K was silent. The resonance Raman spectrum of 1 was collected using 355 nm excitation in MeCN at −30 °C. 1-16O exhibits two isotope-sensitive bands at 879 and 550 cm−1 that shift to 837 and 528 cm−1, respectively, upon 18O-labeling of 1 (1-18O; SI, Figure S4a, inset). The higher energy feature (879 cm−1) with a 16−18Δν of 42 cm−1 (calculated 16−18Δν = 50 cm−1) is assigned to the OO stretching vibration of the peroxo ligand, and the lower energy feature (550 cm−1) is ascribed to the CoO stretching vibration (observed 16−18Δν = 22 cm−1 and calculated 16−18Δν = 24 cm−1). In addition to the above spectroscopic characterization, 1 was structurally characterized via single-crystal X-ray crystallography (SI, Tables S1 and S2 for crystallographic data of 1). The X-ray crystal structure of 1-BPh4 reveals the mononuclear side-on O2−cobalt complex in a distorted octahedral geometry arising from the triangular CoO2 moiety with a small OCoO bite angle of 46.44(9)° (Figure 1a; SI, Table S2). The OO bond length of 1.456(3) Å in 1 (SI, Table S2) is comparable to those in other peroxocobalt(III) complexes reported previously, [CoIII(tmen)2(O2)]+ (1.457 Å; tmen = tetramethylethylenediamine), [CoIII(12-TMC)(O2)]+ (1.4389 Å; 12-TMC = 1,4,7,10-tetramethyl-1,4,7,10-tetraazacyclododecane), and [CoIII(13-TMC)(O2)]+ (1.438 Å; 13-TMC = 1,4,7,10tetramethyl-1,4,7,10-tetraazacyclotridecane),22,23 indicating that the OO bond length in 1 lies well within the metal binding peroxo category (∼1.4−1.5 Å) as supported by the vibrational data (vide supra).24,25 In the plot of MO versus OO bond lengths, the average CoO (1.846 Å) bond distance was also well-correlated with the OO (1.456(3) Å) bond distance (SI, Figure S6 and Table S2). Interestingly, complex 1 was found to react readily with MeCN in C6H6 at 40 °C (Figure 2a). Upon addition of MeCN (3.8 M) to a solution of 1 (2.0 mM), the absorption band at 974 nm due to 1 disappeared obeying a first-order kinetics (Figure 2a, inset) and a new species, 2, which corresponds to the electronic absorption bands at λmax = 450 (ε = 420 M−1 cm−1) and 790 nm (ε = 430 M−1 cm−1), appeared with an isosbestic point at 960 nm (Figure 2a). Pseudo-first-order rate constant enhanced linearly with an increase in the MeCN concentration (SI, Figure S7a), giving a second-order rate
Figure 1. ORTEP diagrams of (a) the starting peroxocobalt(III) complex, [Co(TBDAP)(O2)]+ (1), and the products obtained in the nitrile group activation, [CoIII(TBDAP)(RC(NO)O)]+ where R = (b) Me (2), (c) Et (3), and (d) Ph (4), with thermal ellipsoids drawn at the 30% probability level. Hydrogen atoms are omitted for clarity.
Figure 2. (a) UV−vis spectral changes observed in the reaction of 1 (2.0 mM) with CH3CN (3.8 M) in C6H6 at 40 °C. Inset shows the time course of the absorbance at 790 nm due to 2. (b) ESI-MS spectrum of the complete reaction solution obtained in the reaction of 1 (2.0 mM) with CH3CN (3.8 M). The peak at m/z = 484.3 corresponds to [CoIII(TBDAP)(CH3C(NO)O)]+ (2-16O) (calculated m/z of 484.2). Insets show observed isotope distribution patterns for 2-16O (lower) and 2-18O (upper) derived from 1-16O and 1-18O, respectively.
10961
DOI: 10.1021/jacs.7b04479 J. Am. Chem. Soc. 2017, 139, 10960−10963
Communication
Journal of the American Chemical Society constant of 2.4(1) × 10−4 M−1 s−1 at 40 °C. The ESI-MS spectrum of 2 exhibited a prominent signal at m/z of 484.3, corresponding to [Co(TBDAP)(CH3C(NO)O)]+ (2-16O; calculated m/z of 484.2) as shown in Figure 2b. The 18Olabeling experiment was also performed by replacing 1-16O with 1-18O to support the derivation of the oxygen atoms of 2 from the peroxo group of 1 (Figure 2b, insets). The temperature dependence of the k2 value was examined in the range of 293− 323 K, where a linear Eyring plot was obtained with activation parameters of ΔH‡ = 15(1) kcal mol−1 and ΔS‡ = −27(2) cal mol−1 K−1 (SI, Figure S7b). The observed negative entropy value and second-order kinetics suggest that the nitrile activation by 1 occurs via a bimolecular mechanism. The absolute confirmation of 2 was achieved by the X-ray crystallographic analysis (Figure 1b; SI, Tables S3 and S4). The X-ray diffraction data of 2-BPh4 reveal that the acetylhydroximato ligand, which is a doubly deprotonated tautomer of acetylhydroxamic acid (Scheme 2), coordinates to cobalt via
structural characterization for 2−4, we conclude that hydroximatocobalt(III) complexes are obtained from not only MeCN but also other nitriles, such as EtCN and PhCN. The reactivity of 1 was further investigated with parasubstituted benzonitrile derivatives (para-X-Ph-CN). The effect of para substituents on the nitrile activation by 1 was examined using para-X-Ph-CN (X = CH3O, CH3, H, and Cl). Hammett plot of the pseudo-first-order rate constants versus σp+ afforded a ρ value of 0.18(1) (SI, Figure S12). The very small ρ value suggests that the reaction is almost not dependent on electrons flowing into the ring.26 As shown in Scheme 3, the two Scheme 3. Possible Mechanisms for the Nitrile Activationa
Scheme 2. Representation of the Tautomeric Structures of the Hydroxamato and Hydroximato Forms
a
An O2 Group in 1 is inserted into nitriles through three plausible ways; stepwise process (A) vs concerted processes (B and C).
both oxygen atoms in which the nitrile group of MeCN is oxidized by the peroxo moiety of 1 (Figure 1b). The monovalent charge of 2 with a single tetraphenylborate anion supports that the ligand is able to be adopted the doubly deprotonated hydroximate form rather than the hydroxamate form (Scheme 2). The CN bond length of 1.273(5) Å (SI, Table S4) also clearly indicates the double bond character of CN, supporting the hydroximate form. The average CoO bond length in 2 (1.856 Å) is comparable to that of 1 (1.846 Å), but the O···O interatomic distance in 2 (2.508(4) Å) is significantly longer than the OO bond distance of 1 (1.456(3) Å) and the OCoO angle in 2 (85.05(12)°) is much larger than that of 1 (46.44(9)°), indicating the OO bond scission in 2 (SI, Table S4). To explore the scope of this unprecedented reactivity, the nitrile group activation of 1 was performed with other nitriles, propionitrile (EtCN) and benzonitrile (PhCN). 1 reacted with both EtCN and PhCN in CHCl3 at 40 °C, resulting in the formation of hydroximatocobalt(III) complexes, [CoIII(TBDAP)(CH3CH2C(NO)O)]+ (3) and [CoIII(TBDAP)(C6H5C(NO)O)]+ (4), respectively (SI, Figures S8 and S9 for UV−vis and ESI-MS spectra). The second-order rate constants for the formation of 3 and 4 were determined to be 1.3(1) × 10−3 M−1 s−1 and 9.1(5) × 10−3 M−1 s−1, respectively (SI, Figure S10). The activation parameters for the formation of 3 and 4 were determined (SI, Figure S11), and the observed negative entropy values along with the second-order kinetics suggest that a bimolecular mechanism directs the nitrile activation by 1, which is identical to the formation of 2. Finally, the single crystals of both 3-BPh4 and 4-BPh4 were successfully obtained. As shown in Figure 1c and 1d, the X-ray crystal structures of 3-BPh4 and 4-BPh4 indicate that the ethylhydroximato and benzohydroximato ligands, respectively, coordinate in a bidentate mode to the cobalt center, affording a distorted octahedral geometry (see SI, Tables S3 and S4 for the crystallographic data). On the basis of the spectroscopic and
mechanistic pathways can be considered for this process. One possibility is a stepwise mechanism in which the oxidation of nitrile by 1 is initiated by the nucleophilic attack of the peroxo group in 1 to carbon position of nitrile group, followed by the OO bond cleavage to give hydroximato species (Scheme 3, pathway A). A similar mechanism has been proposed in the alkaline epoxidation of acrylonitrile and the nucleophilic oxidative reaction of peroxometal complexes.25,27 In this stepwise mechanism, a positive large ρ value (>1) would be expected in the Hammett analysis. Because the very small ρ value of 0.18(1) was obtained in the reaction of parasubstituted benzonitriles and 1, such a stepwise process is highly unlikely. The other possibility is a concerted mechanism, which gives pericyclic intermediates (Scheme 3, pathways B and C). In addition, there would be no exchange of oxygen atoms in both reaction pathways. Our isotope labeling experiments clearly show that no scrambling of the oxygen atoms occurs during the nitrile activation by 1 (Figure 2b; SI, Figure S13). Thus, the results of Hammett analysis and isotope labeling experiments are best interpreted in terms of the concerted mechanism. Further studies, including density functional theory calculations, into a detailed mechanism for nitrile activation by 1 are under investigation. In conclusion, we have demonstrated that a nonheme mononuclear side-on peroxocobalt(III) complex bearing the macrocyclic TBDAP ligand, [CoIII(TBDAP)(O2)]+ (1), reacted with nitriles (RCN; R = Me, Et, and Ph) to produce the corresponding mononuclear hydroximatocobalt(III) complexes, [CoIII(TBDAP)(RC(NO)O)]+ (R = Me (2), Et (3), and Ph (4)). Complex 1 and products 2−4, generated through the nitrile group activation of RCN by 1, were fully characterized with various spectroscopic techniques, such as UV−vis, ESI-MS, resonance Raman, and FT-IR, together 10962
DOI: 10.1021/jacs.7b04479 J. Am. Chem. Soc. 2017, 139, 10960−10963
Communication
Journal of the American Chemical Society
(12) Vogt, M.; Nerush, A.; Iron, M. A.; Leitus, G.; Diskin-Posner, Y.; Shimon, L. J. W.; Ben-David, Y.; Milstein, D. J. Am. Chem. Soc. 2013, 135, 17004−17018. (13) Mukherjee, A.; Srimani, D.; Chakraborty, S.; Ben-David, Y.; Milstein, D. J. Am. Chem. Soc. 2015, 137, 8888−8891. (14) Ahmed, T. J.; Knapp, S. M. M.; Tyler, D. R. Coord. Chem. Rev. 2011, 255, 949−974. (15) Wertz, D. L.; Valentine, J. S. Struct. Bonding (Berlin) 2000, 97, 37−60. (16) Miyaji, T.; Kujime, M.; Hikichi, S.; Moro-oka, Y.; Akita, M. Inorg. Chem. 2002, 41, 5286−5295. (17) Müller, J.; Würtele, C.; Walter, O.; Schindler, S. Angew. Chem., Int. Ed. 2007, 46, 7775−7777. (18) Muri, E. M. F.; Nieto, M. J.; Sindelar, R. D.; Williamson, J. S. Curr. Med. Chem. 2002, 9, 1631−1653. (19) Munster, P. N.; Troso-Sandoval, T.; Rosen, N.; Rifkind, R.; Marks, P. A.; Richon, V. M. Cancer Res. 2001, 61, 8492−8497. (20) Parvathy, S.; Hussain, I.; Karran, E. H.; Turner, A. J.; Hooper, N. M. Biochemistry 1998, 37, 1680−1685. (21) Failes, T. W.; Cullinane, C.; Diakos, C. I.; Yamamoto, N.; Lyons, J. G.; Hambley, T. W. Chem. - Eur. J. 2007, 13, 2974−2982. (22) Cho, J.; Sarangi, R.; Kang, H. Y.; Lee, J. Y.; Kubo, M.; Ogura, T.; Solomon, E. I.; Nam, W. J. Am. Chem. Soc. 2010, 132, 16977−16986. (23) Rahman, A. F. M. M.; Jackson, W. G.; Willis, A. C. Inorg. Chem. 2004, 43, 7558−7560. (24) Cramer, C. J.; Tolman, W. B.; Theopold, K. H.; Rheingold, A. L. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 3635−3640. (25) Cho, J.; Sarangi, R.; Nam, W. Acc. Chem. Res. 2012, 45, 1321− 1330. (26) Clayden, J.; Greeves, N.; Warren, S. Organic Chemistry, 2nd ed.; Oxford University Press: Oxford, 2012; pp 1041−1049. (27) Payne, G. B.; Williams, P. H. J. Org. Chem. 1961, 26, 651−659.
with X-ray analysis. In particular, X-ray crystal structures of 2−4 reveal that 2−4 have the five-membered ring with the hydroximato ligands produced upon the oxidation of nitrile groups by the peroxo moiety of 1. Kinetic studies and isotope labeling experiments show that 1 is capable of achieving dioxygenase-like reactivity toward the nitrile group in nitrile derivatives to afford hydroximatocobalt(III) complexes (2−4). Because the spin crossing between S = 0 and S = 1 for 1 was observed, another interesting point is that the spin state of 1 may play an important role in the nitrile group activation. Our future studies will focus on the effect of the spin state of the mononuclear peroxocobalt(III) complexes on the nitrile activation as well as density functional theory calculations into a detailed mechanism.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b04479. Experimental details, Tables S1−S4, and Figures S1−S13 (PDF) Data for C22H34CoN6O7 (CIF) Data for C46H52BCoN4O2 (CIF) Data for C53H57BCoN5O2 (CIF) Data for C49H57BCoN5O2 (CIF) Data for C48H55BCoN5O2 (CIF)
■
AUTHOR INFORMATION
Corresponding Author
*
[email protected] ORCID
Jaeheung Cho: 0000-0002-2712-4295 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The research was supported by NRF (2017R1A2B4005441 to J.C.) and the Ministry of Science, ICT and Future Planning (CGRC 2016M3D3A01913243 to J.C.) of Korea, and the Ministry of Education, Culture, Sports, Science and Technology of Japan through the “Strategic Young Researcher Overseas Visits Program for Accelerating Brain Circulation” to T.O. and Grant-in-Aid for Scientific Research (No. 15H00960 to T.O.).
■
REFERENCES
(1) Varela, J. A.; Saá, C. Chem. Rev. 2003, 103, 3787−3801. (2) Kobayashi, M.; Shimizu, S. Curr. Opin. Chem. Biol. 2000, 4, 95− 102. (3) Murakami, T.; Nojiri, M.; Nakayama, H.; Odak, M.; Yohda, M.; Dohmae, N.; Takio, K.; Nagamune, T.; Endo, I. Protein Sci. 2000, 9, 1024−1030. (4) Kovacs, J. A. Chem. Rev. 2004, 104, 825−848. (5) Bork, P.; Koonin, E. V. Protein Sci. 1994, 3, 1344−1346. (6) Brenner, C. Curr. Opin. Struct. Biol. 2002, 12, 775−782. (7) Brennan, B. A.; Alms, G.; Nelson, M. J.; Durney, L. T.; Scarrow, R. C. J. Am. Chem. Soc. 1996, 118, 9194−9195. (8) Payne, M. S.; Wu, S.; Fallon, R. D.; Tudor, G.; Stieglitz, B.; Turner, I. M., Jr.; Nelson, M. J. Biochemistry 1997, 36, 5447−5454. (9) Hopmann, K. H. Inorg. Chem. 2014, 53, 2760−2762. (10) Zinn, P. J.; Sorrell, T. N.; Powell, D. R.; Day, V. W.; Borovik, A. S. Inorg. Chem. 2007, 46, 10120−10132. (11) Swartz, R. D.; Coggins, M. K.; Kaminsky, W.; Kovacs, J. A. J. Am. Chem. Soc. 2011, 133, 3954−3963. 10963
DOI: 10.1021/jacs.7b04479 J. Am. Chem. Soc. 2017, 139, 10960−10963
