Monomeric Octahedral Ruthenium(II) Complex Enabled meta-C–H

Jun 2, 2017 - Ruthenium-Catalyzed Remote C–H Sulfonylation of N-Aryl-2-aminopyridines with Aromatic Sulfonyl Chlorides. Balu Ramesh and Masilamani J...
44 downloads 10 Views 1MB Size
Letter pubs.acs.org/OrgLett

Monomeric Octahedral Ruthenium(II) Complex Enabled meta-C−H Nitration of Arenes with Removable Auxiliaries Zhoulong Fan,†,‡ Jie Li,§ Heng Lu,§ Dong-Yu Wang,† Chao Wang,∥,⊥ Masanobu Uchiyama,∥,⊥ and Ao Zhang*,†,‡,§ †

CAS Key Laboratory of Receptor Research and the State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica (SIMM), Chinese Academy of Sciences, Shanghai 201203, China ‡ University of Chinese Academy of Sciences, Beijing 100049, China § ShanghaiTech University, Shanghai 201210, China ∥ Graduate School of Pharmaceutical Sciences, University of Tokyo, Tokyo-to 113-0033, Japan ⊥ Advanced Elements Chemistry Research Team, RIKEN Center for Sustainable Resource Science and Elements Chemistry Laboratory, Saitama-ken 351-0198, Japan S Supporting Information *

ABSTRACT: A removable oxime-assisted meta-C−H nitration of arenes is reported. Mechanistic investigations and DFT calculations reveal a new monomeric octahedral ruthenium(II) complex is responsible for the meta-selective nitration. Dioxygen as a cooxidant is crucial for achieving high conversion and good yields. Moreover, the utility of the present reaction protocol is further showcased by the late-stage modification of the clinical CNS drugs Diazepam and Fluvoxamine.

T

ransition metal-catalyzed remote inert C−H bond activation has attracted increasing attention in recent years.1 Among many others, the meta-selective C−H bond activation of arenes continues to hold much appeal, yet it is a highly challenging research topic.2 Recently, an ortho-directing group (DG)-assisted ruthenation strategy has been successfully used for meta-selective C−H functionalization. This strategy facilitates a convenient switch of ortho to meta selectivity by altering the catalysts. In 2011, Frost described the first case of ruthenium(II)-catalyzed meta-C−H sulfonation of 2-phenylpyridines.3 Subsequently, the meta-C−H alkylation,4 bromination,5 and mono- or difluoromethylation6 were reported as well. However, to date, the workable DGs are limited to Nheterocycles. We recently reported the first case of ruthenium(0)-catalyzed meta-C−H nitration of arenes through a dimeric octahedral ruthenium(II) intermediate (A) (Figure 1a).7 Though this method is suitable for arene substrates bearing diversified Nheterocycles, major limitations are conspicuous. These include (1) the requirement for employing strong coordinating heteroaromatic DGs, which are difficult to remove or further transform, and (2) the intolerance of oximes under the acidic conditions. These drawbacks likely originate from the stability of the dimeric octahedral complex, which prompted us to speculate that a monomeric octahedral Ru(II) complex might be sufficiently flexible and overcome these deficiencies. Herein, we report our investigations on the removable oxime-mediated meta-C−H nitration of arenes under a nonacidic catalytic system that uses low loadings of AgNO3 as the nitro source and © 2017 American Chemical Society

Figure 1. Ruthenium-catalyzed meta-C−H nitration of arenes.

O2 as the cooxidant (Figure 1b). Mechanistic studies reveal that a new ortho-ruthenated monomeric octahedral complex is involved. Convenient application of this approach for late-stage modification of clinically prescribed CNS drugs succeeded as well. We first investigated the model reaction of substrate 1a with AgNO3. After extensive screening of a range of catalysts, Received: May 3, 2017 Published: June 2, 2017 3199

DOI: 10.1021/acs.orglett.7b01297 Org. Lett. 2017, 19, 3199−3202

Letter

Organic Letters

Scheme 1. Substrate Scope of Oxime-Mediated meta-C−H Nitrationa

ligands, oxidants, solvents, and imine-containing DGs,8 we established the optimum reaction conditions and achieved product 2a in 69% isolated yield using Ru3(CO)12 as the catalyst, PhI(TFA)2 and O2 as the oxidants, and 1,2dichloroethane (DCE) as the solvent (Table 1, entry 1). Table 1. Optimization of the Reaction Conditionsa

entry

deviation from standard conditions

conv/yield(%)b

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

none [RuCl2(p-cymene)]2 instead of Ru3(CO)12 using 30 mol % Piv-Val-OH using 30 mol % Ph3CCOOH K2S2O8 instead of PhI(TFA)2 CuO instead of PhI(TFA)2 PhI(OAc)2 instead of PhI(TFA)2 PhCl instead of DCE MeCN instead of DCE N2 instead of O2 5 mol % Ru3(CO)12 instead of 7.5 mol % 1.5 equiv of AgNO3 instead of 1.8 equiv 24 h instead of 30 h Cu(NO3)2·3H2O instead of AgNO3 no PhI(TFA)2 no Ru3(CO)12

100/74(69)c 70/0 56/17 100/62 44/8 34/0 55/0 87/42 44/0 83/22 75/40 81/53 86/57 0/0 5/0 90/0

a

Reaction conditions: 1a (0.1 mmol), AgNO3 (1.8 equiv), Ru3(CO)12 (7.5 mol %), PhI(TFA)2 (1 equiv), DCE (1 mL), O2, 100 °C, 30 h. b Conversion and yield were determined by 1H NMR analysis using CH2Br2 as an internal standard. cIsolated yield on 0.3 mmol scale.

a

Reaction conditions: 1 (0.30 mmol), AgNO3 (0.54 mmol), Ru3(CO)12 (7.5 mol %), PhI(TFA)2 (1 equiv) in DCE (3 mL) at 100 °C for 30 h in a sealed tube filled with O2. Isolated yields. bFor 36 h. cFor 48 h.

Subsequently, fine-tuning of the standard reaction conditions was conducted to understand the role of each reaction element. We noticed that no product was formed when Ru(0) was replaced by [RuCl2(p-cymene)]2 (entry 2), indicating that a Ru(0) precatalyst was beneficial to the ensuing formation of the octahedral intermediate. To investigate the ligand effect, sterically bulky acids were employed in the catalytic system. However, lower conversions/yields were observed (entries 3− 4). Low yields were obtained in the presence of K2S2O8, and the use of CuO or PhI(OAc)2 as the oxidant failed to provide product 2a (entries 5−7). We speculate that trifluoroacetates may assist the initial C−H bond cleavage to form the cyclometalated complex. Alternative solvents were also examined, and chlorobenzene gave a 42% yield (entry 8). It is worth noting that a N2 atmosphere led to incomplete conversion and lower yield (entry 10). In addition, lower equivalents of either the Ru(0) catalyst or nitro source as well as a shortening of the reaction time afforded unsatisfactory results (entries 11−13). The use of Cu(NO3)2·3H2O as the nitro source failed to provide 2a (entry 14). Finally, control experiments in the absence of PhI(TFA)2 or Ru3(CO)12 disclosed that these parameters were crucial for the success of the reaction (entries 15−16). With the optimized conditions in hand, we explored the versatility of the meta-C−H nitration of oximes (Scheme 1). Both para and meta electron-donating substituents were welltolerated, and the corresponding products 2b−e, 2i, and 2l were obtained in good yields. Substrates with different halogen substituents generally required prolonged reaction times and

provided moderate yields. Several ortho-substituted substrates resulted in nitration at the sterically more congested meta-C−H bonds (2m, 2o-p), except for the ortho-methoxyl-substituted substrate, which gave low regioselectivity (2n). Bicyclic and naphthyl substrates were found to be suitable as well, providing the corresponding products 2q−s in moderate to good yields. For the cyclic oxime substrates, the contrary regio-preference in meta-products 2t and 2u was obtained, likely due to the different electronic effects.4a In addition, the oximes bearing phenyl, n-butyl, and ester substituents underwent the nitration smoothly, leading to corresponding product 2v−x. The reactive oxime derived from aldehyde was suitable for this transformation (2y). We also investigated several heteroaromatic oximes, including 1-(pyridin-4-yl)ethanone O-methyl oxime and 1-(pyridin-3-yl)ethanone O-methyl oxime, and no meta products were obtained. We supposed that the generation of meta products was probably suppressed by the detrimental coordination of the pyridine N-atom to the metal. Next, we tried to demonstrate the practicality of the present meta-C−H nitration protocol. As shown in Scheme 2, the oxime group was conveniently removed by treating 2a with concentrated HCl, affording ketone 3 in nearly quantitative yield. Alternatively, 2a could also be reduced by treating with ZrCl4/NaBH4 to furnish amine 4 (Scheme 2a).9 Further, Diazepam (5)10 and Fluvoxamine (7) are clinically prescribed CNS disorder drugs, and the meta-selective nitrations are difficult to realize under KNO3/H2SO4 conditions or our 3200

DOI: 10.1021/acs.orglett.7b01297 Org. Lett. 2017, 19, 3199−3202

Letter

Organic Letters Scheme 2. Synthetic Application of meta-C−H Nitration

Scheme 4. Investigation of Ruthenium(II) Intermediate

treating II with AgNO3 in the presence of PhI(TFA)2 (Scheme 4c). In addition, a DFT calculation12 indicated that the HOMO of II is mainly localized on the meta position, prompting us to hypothesize that the subsequent electrophilic nitration would likely be preferred at the meta position (Figure 2).13 These results indicate that a similar octahedral cycloruthenated complex such as II might be involved in the catalytic cycle as the active intermediate.

previous meta-C−H nitration protocol. We found that latestage meta-C−H nitration of these drugs under our current protocol proceeded with the corresponding derivatives 6 and 8 being obtained in 62 and 27% yields, respectively (Scheme 2b). To gain insights into the details of the catalytic cycle, we first conducted the meta nitration with the addition of the radical scavenger TEMPO. The reaction was suppressed, suggesting that a radical pathway might be involved (Scheme 3a). Scheme 3. Mechanistic Studies

Figure 2. HOMO of intermediate II.

On the basis of the studies described above, a plausible mechanism for the present meta-C−H nitration process is depicted in Scheme 5. Initially, the Ru(0) precatalyst is oxidized to the active Ru(II) catalyst by PhI(TFA)2 and O2, followed by ortho-C−H bond cleavage of 1a to yield Ru(II) complex I. The Scheme 5. Proposed Mechanism

Subsequently, the reaction was conducted in the presence of D2O, and significant ortho-H/D exchange was observed, indicating that the initial ortho-C−H activation/ruthenation is reversible (Scheme 3b). Finally, we carefully measured the competitive and parallel kinetic isotope effects,11 and the values of PH/PD and kH/kD were 1.5 and 1.9, respectively. The results confirmed that the meta-C−H bond cleavage likely occurs during the rate-determining step (Scheme 3c). To gain more insight into the present transformation, we isolated a new octahedral intermediate II, and its structure was confirmed via X-ray diffraction (Scheme 4a). We found that complex II could catalyze the meta-nitration reaction (Scheme 4b). The meta product 2a was also obtained in 21% yield by 3201

DOI: 10.1021/acs.orglett.7b01297 Org. Lett. 2017, 19, 3199−3202

Letter

Organic Letters

De Sarkar, S.; Ackermann, L. Top. Organomet. Chem. 2015, 55, 217. (e) Sharma, R.; Thakur, K.; Kumar, R.; Kumar, I.; Sharma, U. Catal. Rev.: Sci. Eng. 2015, 57, 345. (f) Yang, J. Org. Biomol. Chem. 2015, 13, 1930. (g) Dey, A.; Agasti, S.; Maiti, D. Org. Biomol. Chem. 2016, 14, 5440. (3) (a) Saidi, O.; Marafie, J.; Ledger, A. E.; Liu, P. M.; Mahon, M. F.; Kociok-Kohn, G.; Whittlesey, M. K.; Frost, C. G. J. Am. Chem. Soc. 2011, 133, 19298. (b) Marcé, P.; Paterson, A. J.; Mahon, M. F.; Frost, C. G. Catal. Sci. Technol. 2016, 6, 7068. (4) (a) Hofmann, N.; Ackermann, L. J. Am. Chem. Soc. 2013, 135, 5877. (b) Li, J.; Warratz, S.; Zell, D.; De Sarkar, S.; Ishikawa, E. E.; Ackermann, L. J. Am. Chem. Soc. 2015, 137, 13894. (c) Paterson, A. J.; St. John-Campbell, S.; Mahon, M. F.; Press, N. J.; Frost, C. G. Chem. Commun. 2015, 51, 12807. (d) Li, G.; Ma, X.; Jia, C.; Han, Q.; Wang, Y.; Wang, J.; Yu, L.; Yang, S. Chem. Commun. 2017, 53, 1261. (5) (a) Teskey, C. J.; Lui, A. Y.; Greaney, M. F. Angew. Chem., Int. Ed. 2015, 54, 11677. (b) Yu, Q.; Hu, L.; Wang, Y.; Zheng, S.; Huang, J. Angew. Chem., Int. Ed. 2015, 54, 15284. (c) Warratz, S.; Burns, D. J.; Zhu, C.; Korvorapun, K.; Rogge, T.; Scholz, J.; Jooss, C.; Gelman, D.; Ackermann, L. Angew. Chem., Int. Ed. 2017, 56, 1557. (6) (a) Ruan, Z.; Zhang, S.-K.; Zhu, C.; Ruth, P. N.; Stalke, D.; Ackermann, L. Angew. Chem., Int. Ed. 2017, 56, 2045. (b) Li, Z.-Y.; Li, L.; Li, Q.-L.; Jing, K.; Xu, H.; Wang, G.-W. Chem. - Eur. J. 2017, 23, 3285. (7) Fan, Z.; Ni, J.; Zhang, A. J. Am. Chem. Soc. 2016, 138, 8470. (8) For detailed information, see the Supporting Information. (9) Liang, Y.-F.; Wang, X.; Yuan, Y.; Liang, Y.; Li, X.; Jiao, N. ACS Catal. 2015, 5, 6148. (10) For selected examples of ortho-C−H activation of Diazepam, see: (a) Khan, R.; Felix, R.; Kemmitt, P. D.; Coles, S. J.; Day, I. J.; Tizzard, G. J.; Spencer, J. Adv. Synth. Catal. 2016, 358, 98. (b) Min, M.; Kang, D.; Jung, S.; Hong, S. Adv. Synth. Catal. 2016, 358, 1296. (c) Shu, S.; Fan, Z.; Yao, Q.; Zhang, A. J. Org. Chem. 2016, 81, 5263. (11) Simmons, E. M.; Hartwig, J. F. Angew. Chem., Int. Ed. 2012, 51, 3066. (12) Calculation was performed with Gaussian 09, revision D.01, based on the crystal structure of intermediate II at the level of B3LYP/ 6-31g*(H, C, N, O, F) and LANL2DZ (Ru). (13) For selected examples of predicting the regioselectivity of electrophilic aromatic substitution, see: (a) Hirao, H.; Ohwada, T. J. Phys. Chem. A 2003, 107, 2875. (b) Kruszyk, M.; Jessing, M.; Kristensen, J. L.; Jorgensen, M. J. Org. Chem. 2016, 81, 5128. (c) Hisamatsu, Y.; Kumar, S.; Aoki, S. Inorg. Chem. 2017, 56, 886. (d) Tamura, Y.; Hisamatsu, Y.; Kumar, S.; Itoh, T.; Sato, K.; Kuroda, R.; Aoki, S. Inorg. Chem. 2017, 56, 812. (14) (a) Liu, Y.-K.; Lou, S.-J.; Xu, D.-Q.; Xu, Z.-Y. Chem. - Eur. J. 2010, 16, 13590. (b) Pawar, G. G.; Brahmanandan, A.; Kapur, M. Org. Lett. 2016, 18, 448. (15) (a) Sabbasani, V. S.; Lee, D. Org. Lett. 2013, 15, 3954. (b) Sun, J.; Qiu, J.-K.; Wu, Y.-N.; Hao, W.-J.; Guo, C.; Li, G.; Tu, S.-J.; Jiang, B. Org. Lett. 2017, 19, 754.

isolated complex II might be produced through a water-ligand exchange of I during our workup process. Meanwhile, AgNO3 is oxidized to a bivalent silver salt in the presence of PhI(TFA)2,14 and a nitrogen dioxide radical (•NO2) is released15 to attack the para position of the C−Ru bond, affording Ru(II) species III.7 Subsequently, the reductive deprotonation of III delivers intermediate IV. Finally, ligand exchange of IV with CF3COOH releases the meta-nitrated product 2a and regenerates the active Ru(II) catalyst for the next catalytic cycle. In summary, a versatile, removable oxime-mediated meta-C− H nitration of arenes has been developed. Dioxygen as cooxidant is crucial for achieving high conversion and yields. Mechanistic investigations and DFT calculations have revealed that an octahedral cycloruthenated complex may be responsible for the radical electrophilic meta-C−H nitration of arenes. A plethora of functional groups were well-tolerated. In addition to the easy-cleavage and convenience for further transformations of the oxime as the DG, the present reaction protocol has been further applied for the late-stage modification of clinically prescribed drugs and nucleosides.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.orglett.7b01297. Experimental procedures, characterization data and 1H and 13C NMR spectra for all new compounds (PDF) Crystallographic data for II (CCDC 1534973) (CIF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Ao Zhang: 0000-0001-7205-9202 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by grants from Chinese NSF (81373277, 81430080). Support from the National Program on Key Basic Research Project of China (2015CB910603), the International Cooperative Program (GJHZ1622) and Key Program of the Frontier Science (160621) of the Chinese Academy of Sciences, the Shanghai Commission of Science and Technology (16XD1404600, 14431905300, 14431900400), as well as grant from CAS Key Laboratory of Receptor Research of SIMM (SIMM1606YKF-08, SIMM1606YZZ-06) are also highly appreciated.



REFERENCES

(1) For selected recent reviews on remote C−H activation, see: (a) Schranck, J.; Tlili, A.; Beller, M. Angew. Chem., Int. Ed. 2014, 53, 9426. (b) Chiba, S.; Chen, H. Org. Biomol. Chem. 2014, 12, 4051. (c) Qiu, G.; Wu, J. Org. Chem. Front. 2015, 2, 169. (d) De Sarkar, S. Angew. Chem., Int. Ed. 2016, 55, 10558. (2) For recent reviews on meta-selective C−H activation, see: (a) Juliá-Hernández, F.; Simonetti, M.; Larrosa, I. Angew. Chem., Int. Ed. 2013, 52, 11458. (b) Ackermann, L.; Li, J. Nat. Chem. 2015, 7, 686. (c) Frost, C. G.; Paterson, A. J. ACS Cent. Sci. 2015, 1, 418. (d) Li, J.; 3202

DOI: 10.1021/acs.orglett.7b01297 Org. Lett. 2017, 19, 3199−3202