Iron-Carbonyl-Catalyzed Redox-Neutral [4+2] Annulation of N-H Imines and Internal Alkynes by C-H Bond Activation

Article abstract of DOI:10.1002/anie.201600365

Stoichiometric C-H bond activation of arenes mediated by iron carbonyls was reported by Pauson as early as in 1965, yet the catalytic C-H transformations have not been developed. Herein, an iron-catalyzed annulation of N-H imines and internal alkynes to furnish cis-3,4-dihydroisoquinolines is described, and represents the first iron-carbonyl-catalyzed C-H activation reaction of arenes. Remarkablely, this is also the first redox-neutral [4+2] annulation of imines and alkynes proceeding by C-H activation. The reaction also features only cis stereoselectivity and excellent atom economy as neither base, nor external ligand, nor additive is required. Experimental and theoretical studies reveal an oxidative addition mechanism for C-H bond activation to afford a dinuclear ferracycle and a synergetic diiron-promoted H-transfer to the alkyne as the turnover-determining step. Double dose of iron: The titled redox-neutral [4+2] annulations to furnish cis-3,4-dihydroisoquinolines were achieved by using iron catalysis. Mechanistic studies show the synergy of dinuclear iron in the C-H bond activation and turnover-limiting hydrogen-transfer steps. The reaction demonstrates excellent atom economy and exclusive cis stereoselectivity.

Full text of DOI:10.1002/anie.201600365

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201600365
Homogeneous Catalysis
German Edition:
DOI: 10.1002/ange.201600365
À
Iron-Carbonyl-Catalyzed Redox-Neutral [4+2] Annulation of N H
Imines and Internal Alkynes by C H Bond Activation
À
Teng Jia, Chongyang Zhao, Ruoyu He, Hui Chen,* and Congyang Wang*
À
Abstract: Stoichiometric C H bond activation of arenes
mediated by iron carbonyls was reported by Pauson as early
Low-valent iron species formed in situ were frequently
proposed to be the truly active catalysts. Regrettably, these
key intermediates have not been identified and their defined
structures remain elusive so far. In contrast, well-defined iron
carbonyl complexes such as [Fe2(CO)9] and [Fe3(CO)12] were
À
as in 1965, yet the catalytic C H transformations have not been
À
developed. Herein, an iron-catalyzed annulation of N H
imines and internal alkynes to furnish cis-3,4-dihydroisoquino-
lines is described, and represents the first iron-carbonyl-
catalyzed C H activation reaction of arenes. Remarkablely,
À
showcased by Pauson and others to promote C H bond
activation of aromatic imines as early as in 1965 [Eq. (1)].[5]
À
this is also the first redox-neutral [4+2] annulation of imines
À
and alkynes proceeding by C H activation. The reaction also
features only cis stereoselectivity and excellent atom economy
as neither base, nor external ligand, nor additive is required.
Experimental and theoretical studies reveal an oxidative
À
addition mechanism for C H bond activation to afford
a dinuclear ferracycle and a synergetic diiron-promoted
H-transfer to the alkyne as the turnover-determining step.
Notably, the reduction of the imino group occurred after the
À
C H bond activation, thus forming a dinuclear iron complex
bearing a 1-ferraisoindoline moiety. However, no iron-
T
ransition-metal-catalyzed direct transformations of inert
À
À
À
À
C H bonds into C C and C X (X = heteroatom) bonds have
demonstrated great potential in organic synthesis over the
past few decades.[1] Second- and third-row transition metals
carbonyl-catalyzed C H transformations based on this stoi-
chiometric cycloferration protocol was reported, even after
half a century.
À
À
(such as Pd, Ir, Rh, Ru) have been used successfully in C H
The application of C H activation strategies to the
activation reactions in terms of diverse reactivity and
controlled selectivity.[1] However, the high cost and rarity of
these precious metals necessitate the hunt for new catalysts
based on first-row transition metals because of their earth
abundance and environmental benign characteristics. More-
over, the unique structures and properties of the first-row
synthesis of heterocycles has recently attracted immense
attention, and proved to be very powerful. As illustrated
examples, the annulations of imine derivatives with alkynes
À
by C H activation allow expedient access to various
N-heterocycles in a highly efficient manner. Among these,
the oxidative and redox-neutral [3+2] cyclizations of imines
and alkynes to approach indenimines and indenamines,
respectively were achieved through either Rh-, Ru-, Ir-, or
Re catalysis (Scheme 1a).[6] Meanwhile, the oxidative [4+2]
annulations of imines and alkynes leading to isoquinolines
À
transition metals may enable new C H transformations with
unprecedented reactivity and selectivity.[2] In this context,
iron is arguably the ideal metal for catalyst development
because of its high abundance, low cost, and non-toxicity.[3]
Since 2008, the groups of Nakamura and Ilies, Ackermann,
Cook, Yoshikai, and others have elegantly disclosed a series
À
of iron-catalyzed C H activation reactions by using organo-
magnesium/zinc reagents as key activators, and either dipyr-
idines or diphosphines as essential ligands of iron catalysts.[4]
[*] T. Jia, Dr. R. He, Prof. Dr. C. Wang
Beijing National Laboratory for Molecular Sciences (BNLMS)
CAS Key Laboratory of Molecular Recognition and Function
Institute of Chemistry, Chinese Academy of Sciences
Beijing 100190 (China)
E-mail: wangcy@iccas.ac.cn
Homepage: http://wangcy.iccas.ac.cn/
C. Zhao, Prof. Dr. H. Chen
BNLMS, CAS Key Laboratory of Photochemistry
Institute of Chemistry, Chinese Academy of Sciences
Beijing 100190 (China)
E-mail: chenh@iccas.ac.cn
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201600365.
Scheme 1. Transition-metal-catalyzed annulations of imines and
À
alkynes by C H bond activation.
5268
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 5268 –5271

Angewandte
Communications
Chemie
were also well established by using various transition-metal
catalysts and proper oxidants (Scheme 1b).[7,8] However, the
redox-neutral [4+2] annulations of imines and alkynes to give
3,4-dihydroisoquinolines remain elusive. Herein, we describe
À
the first redox-neutral [4+2] cyclization of N H imines and
internal alkynes to access cis-3,4-dihydroisoquinolines by
employing iron carbonyls as catalysts (Scheme 1c). This
reaction also represents the first iron-carbonyl-catalyzed
À
C H transformation of arenes since its parent cycloferration
reaction was reported in 1965.[5]
Recently, we reported a manganese-catalyzed dehydro-
À
genative [4+2] annulation of N H imines and alkynes, thus
affording isoquinolines with the evolution of dihydrogen.[8]
During the evaluation of various transition-metal catalysts,
we serendipitously found that the cis-3,4-dihydroisoquinoline
3ca, rather than the isoquinoline 4ca, was selectively formed
1
À
in 78% yield ( H NMR) from the reaction of the N H imine
1c and diphenylacetylene (2a) in diethyl ether when [Fe3-
(CO)12] was used as a catalyst instead of [MnBr(CO)5]
[Eq. (2); PMP = para-methoxyphenyl].[9] Of note, no trans-
3,4-dihydroisoquinoline product was detected during the
optimization process.
Scheme 2. Scope with respect to the imines and alkynes. Reaction
conditions: 1 (0.5 mmol), 2 (0.75 mmol), [Fe3(CO)12] (5 mol%), Et2O
(1.6 mL), 1208C, 24 h. Yield are those of the isolated products. [a] Et2O
(0.25m). [b] Yield of the pure isolated isomer. [c] Determined by
1H NMR analysis of the crude reaction mixture.
We next investigated the substrate scope (Scheme 2).
Both electron-donating and electron-withdrawing groups on
À
N H imines were tolerated successfully (3aa-da). An ortho-
substituted diarylimine showed good reactivity despite its
was run (Scheme 3b). No reaction occurred with 1c and
[Fe3(CO)12], whereas the trinuclear iron complexes Fe-I and
Fe-II were selectively generated from reactions of [Fe3(CO)12]
and 2a at 60 and 1208C, respectively.[10] Although no reaction
took place with 1c and either Fe-I or Fe-II, they did catalyze
the annulation of 1c with 2a to afford 3ca in comparable
yields (Scheme 3c). Furthermore, different alkynes, 2g and
2b, were reacted with 1c under the catalysis of either Fe-I or
Fe-II and only 3cg and 3cb were formed. Therefore 3ca was
not detected, and suggests that the alkyne moieties in Fe-I and
Fe-II were not involved in the final product. Third, the
increased steric hindrance (3ea). The less sterically congested
À
C H bond was preferably annulated in the meta-substituted
imine, albeit in lower yield (3 fa). The di(naphthalenyl)me-
thanimine 1g was also a suitable substrate (3ga), and an
excellent level of regiocontrol was obtained when an unsym-
metrical diarylimine was used (3ha). Alkyl aryl imines
bearing electronically varied groups were also applicable to
this reaction (3ia–pa). A series of aromatic alkynes contain-
ing functional groups like F, Cl, Br, and CF3 in different
substitution patterns was all compatible with the reaction
conditions (3cb–k), which allowed further synthetic elabo-
rations. Unfortunately, the reactions of other alkynes such as
4-octyne did not proceed and phenylacetylene failed to give
the expected product.[9] Two regioisomers were formed when
an unsymmetrical diaryl alkyne was used, and fortunately
could be separated by simple chromatography (3cl, 3cl’).
To probe the possible reaction intermediates, a series of
mechanistic experiments was carried out. First, the intercon-
version between 3 and 4 was tested (Scheme 3a). It was
shown that no formation of 4ca was observed from 3ca under
the standard reaction conditions. Meanwhile, only 3ca was
obtained without 3cg detected when 4cg was added to the
reaction of 1c and 2a, and thus ruled out the intermediacy of
4cg for the formation of 3cg. Second, a stoichiometric
reaction of either the imine 1c or alkyne 2a with [Fe3(CO)12]
À
dinuclear ferracycle Fe-III was synthesized by C H activation
from the aldimine 1q and [Fe3(CO)12][5] and it mainly
decomposed to 1q under the reaction conditions even in the
presence of either 2a or 1c, which implies the reversibility of
À
the C H activation and H-migration steps (Scheme 3d).
Interestingly, Fe-III could also catalyze the annulation of 1c
and 2a, albeit with lower reactivity. Finally, two possible
reaction intermediates, 6aa and 7aa,[11] formally arising from
À
À
the C H and N H addition of 1a to 2a, respectively, were
synthesized and examined under the reaction conditions
(Scheme 3e). It turned out the former could deliver product
3aa with or without [Fe3(CO)12], thus indicating its possible
involvement in the reaction.
À
To shed light on the nature of the C H activation step,
several deuterium-labeling experiments were carried out. The
reaction of the deuterated imine [D10]-1a with 2a was first
Angew. Chem. Int. Ed. 2016, 55, 5268 –5271
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 5269

Angewandte
Communications
Chemie
Scheme 4. Deuterium-labeling experiments. [a] Yield of isolated prod-
1
uct. [b] Yield determined by H NMR spectroscopy.
Scheme 5. The most plausible reaction mechanism obtained from DFT
calculations.
Through comparison of reaction energetics, we discovered
that the dinuclear one is most likely the catalytically active
species.[9] The calculated reaction profile of 1a and 2a are
Scheme 3. Mechanistic experiments. 1H NMR yields are shown. The
structure of Fe-II was confirmed by X-ray crystallographic analysis.[13]
[a] Determined by 1H NMR spectroscopy. [b] Yield of isolated product.
À
depicted in Scheme 5. First, the C H activation step from the
1a-bound reactant A to intermediate B proceeds reversibly
by an oxidative addition mechanism, with a small activation
energy of only 4.4 kcalmolÀ1. At this step, 2a is not likely to
examined and the ortho D atom was transferred to the
3-position of [D10]-3aa with negligible deuterium loss, and
À
bind the iron catalyst, otherwise the C H activation energy
À
suggested a possible oxidative addition pathway for the C H
becomes considerably larger (exceeding 28 kcalmolÀ1). This
implication is in line with the experimental findings that in
alkyne-bound Fe-I/Fe-II, alkyne moieties were not involved
bond activation step (Scheme 4a). Then, intramolecular,
intermolecular, and parallel-reaction kinetic isotope effect
(KIE) experiments were performed, thus giving KIE values of
3.7, 2.6, and 2.6 respectively (Scheme 4b–d). These results
À
in the final product. After C H activation, 2a binds B to form
the complex C, which requires a substantial amount of energy,
14.9 kcalmolÀ1. The iron-hydride addition to the alkyne from
C generates D with a barrier of 9.0 kcalmolÀ1, and the alkyne
moiety changes to an allene-like structure by assistance of
conjugation involving its aromatic substituent. After over-
coming a tiny barrier of 1.5 kcalmolÀ1, D easily isomerizes
into the more stable alkenyl structure E, with a substantial
À
indicate that the C H bond cleavage might be involved in the
turnover-limiting step or in a prior step with a lower activation
barrier.[12]
To explore more details of the reaction mechanism,
extensive DFT calculations were conducted on the basis of
mononuclear, dinuclear, and trinuclear iron catalytic systems.
5270
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 5268 –5271

Angewandte
Communications
Chemie
energy of 27.1 kcalmolÀ1 released. From E to F, the reductive
Fruchey, S. P. Cook, Angew. Chem. Int. Ed. 2014, 53, 11065;
Angew. Chem. 2014, 126, 11245; h) T. Matsubara, S. Asako, L.
Ilies, E. Nakamura, J. Am. Chem. Soc. 2014, 136, 646; i) L. Ilies,
T. Matsubara, S. Ichikawa, S. Asako, E. Nakamura, J. Am. Chem.
Soc. 2014, 136, 13126; j) E. R. Fruchey, B. M. Monks, S. P. Cook,
J. Am. Chem. Soc. 2014, 136, 13130; k) J. J. Sirois, R. Davis, B.
DeBoef, Org. Lett. 2014, 16, 868; l) R. Shang, L. Ilies, E.
Nakamura, J. Am. Chem. Soc. 2015, 137, 7660; m) K. Graczyk, T.
Haven, L. Ackermann, Chem. Eur. J. 2015, 21, 8812; n) M. Y.
Wong, T. Yamakawa, N. Yoshikai, Org. Lett. 2015, 17, 442; o) G.
Cera, T. Haven, L. Ackermann, Angew. Chem. Int. Ed. 2016, 55,
1484; Angew. Chem. 2016, 128, 1506.
À
elimination occurs to form the C C bond generating 6aa with
barrier of 6.2 kcalmolÀ1 (note that the barrier of this C C
À
coupling in trinuclear system is more than 33 kcalmolÀ1),[9]
which could deliver the final product 3aa by electrocycliza-
tion according to our experimental study. In total, the
turnover-determining step in the whole calculated pathway
is the insertion of the alkyne into the iron-hydride bond,
which is in agreement with the KIE experiments. Notably, the
synergy between the two iron nuclei plays an essential role in
both the C H activation and the H-transfer steps.
In summary, the first redox-neutral [4+2] annulations of
N H imines and internal alkynes was developed by employ-
À
[5] For selected examples: a) M. M. Bagga, P. L. Pauson, F. J.
Preston, R. I. Reed, J. Chem. Soc. Chem. Commun. 1965, 543;
b) P. E. Baikie, O. S. Mills, Chem. Commun. 1966, 707; c) M. M.
Bagga, W. T. Flannigan, G. R. Knox, P. L. Pauson, F. J. Preston,
R. I. Reed, J. Chem. Soc. C 1968, 36; d) W. T. Flannigan, G. R.
Knox, P. L. Pauson, J. Chem. Soc. C 1969, 2077; e) N. S.
Nametkin, V. D. Tyurin, A. I. Nekhaev, Yu. P. Sobolev, M. G.
Kondratꢀeva, A. S. Batsanov, Yu. T. Struchkov, J. Organomet.
Chem. 1983, 243, 323; f) N. S. Nametkin, V. D. Tyurin, V. V.
Trusov, A. M. Krapivin, J. Organomet. Chem. 1983, 254, 243;
g) W. Imhof, J. Organomet. Chem. 1997, 533, 31; h) D.-L. Wang,
W.-S. Hwang, L.-C. Liang, L.-I. Wang, L. Lee, M. Y. Chiang,
Organometallics 1997, 16, 3109; i) W. Imhof, Organometallics
1999, 18, 4845.
À
ing iron carbonyl catalysis, which also represents the first iron-
carbonyl-catalyzed C H transformation of arenes since its
À
parent cycloferration reaction reported in 1965. The easily
available starting materials, simple catalyst system, sole cis-
stereoselectivity, and excellent atom economy add more
benefits to this protocol. Remarkably, mechanistic studies
highlight the essential synergy of dinuclear irons in the
oxidative C H addition and turnover-limiting H-transfer to
alkyne steps in the catalytic cycle, and give important hints for
future work on dinuclear iron catalysis.
À
[6] For selected examples: a) Y. Kuninobu, A. Kawata, K. Takai, J.
Am. Chem. Soc. 2005, 127, 13498; b) Z.-M. Sun, S.-P. Chen, P.
Zhao, Chem. Eur. J. 2010, 16, 2619; c) D. N. Tran, N. Cramer,
Angew. Chem. Int. Ed. 2011, 50, 11098; Angew. Chem. 2011, 123,
11294; d) P. Zhao, F. Wang, K. Han, X. Li, Org. Lett. 2012, 14,
5506; e) J. Zhang, A. Ugrinov, P. Zhao, Angew. Chem. Int. Ed.
2013, 52, 6681; Angew. Chem. 2013, 125, 6813; f) L. Dong, C.-H.
Qu, J.-R. Huang, W. Zhang, Q.-R. Zhang, J.-G. Deng, Chem.
Eur. J. 2013, 19, 16537; g) M. Nagamoto, T. Nishimura, Chem.
Commun. 2014, 50, 6274; h) X. Jin, X. Yang, Y. Yang, C. Wang,
Org. Chem. Front. 2016, 3, 268.
[7] For a review: a) R. He, Z.-T. Huang, Q.-Y. Zheng, C. Wang,
Tetrahedron Lett. 2014, 55, 5705; For selected examples: b) N.
Guimond, K. Fagnou, J. Am. Chem. Soc. 2009, 131, 12050; c) T.
Fukutani, N. Umeda, K. Hirano, T. Satoh, M. Miura, Chem.
Commun. 2009, 5141; d) X. Wei, M. Zhao, Z. Du, X. Li, Org.
Lett. 2011, 13, 4636; e) T. K. Hyster, T. Rovis, Chem. Commun.
2011, 47, 11846; f) C. Kornhaaß, J. Li, L. Ackermann, J. Org.
Chem. 2012, 77, 9190; g) H. Wang, C. Grohmann, C. Nimphius, F.
Glorius, J. Am. Chem. Soc. 2012, 134, 19592; h) L. Qiu, D.
Huang, G. Xu, Z. Dai, J. Sun, Org. Lett. 2015, 17, 1810.
[8] R. He, Z.-T. Huang, Q.-Y. Zheng, C. Wang, Angew. Chem. Int.
Ed. 2014, 53, 4950; Angew. Chem. 2014, 126, 5050.
Acknowledgments
Financial support from the National Basic Research Program
of China (973 Program, No. 2012CB821600) and the National
Natural Science Foundation of China (21322203, 21521002,
21521062, 21290194) is gratefully acknowledged. We also
thank the Alexander von Humboldt Foundation for the
Equipment Subsidy.
À
Keywords: alkynes · annulations · C H activation ·
heterocycles · iron
How to cite: Angew. Chem. Int. Ed. 2016, 55, 5268–5271
Angew. Chem. 2016, 128, 5354–5357
À
[1] J. Q. Yu, Z.-J. Shi, C H Activation, Springer, Berlin, 2010.
[2] For reviews: a) A. A. Kulkarni, O. Daugulis, Synthesis 2009,
4087; b) C.-J. Li, Acc. Chem. Res. 2009, 42, 335; c) E. Nakamura,
N. Yoshikai, J. Org. Chem. 2010, 75, 6061; d) C.-L. Sun, B.-J. Li,
Z.-J. Shi, Chem. Rev. 2011, 111, 1293; e) C. Zhang, C. Tang, N.
Jiao, Chem. Soc. Rev. 2012, 41, 3464; f) M. C. White, Science
2012, 335, 807; g) C. Wang, Synlett 2013, 1606; h) F. Jia, Z. Li,
Org. Chem. Front. 2014, 1, 194; i) K. Gao, N. Yoshikai, Acc.
Chem. Res. 2014, 47, 1208; j) L. Ackermann, J. Org. Chem. 2014,
79, 8948.
[9] For more details, see the Supporting Information.
[10] a) R. P. Dodge, V. Schomaker, J. Organomet. Chem. 1965, 3, 274;
b) J. F. Blount, L. F. Dahl, C. Hoogzand, W. Hübel, J. Am. Chem.
Soc. 1966, 88, 292; c) D. Osella, R. Gobetto, P. Montangero,
Organometallics 1986, 5, 1247.
[11] R. S. Manan, P. Kilaru, P. Zhao, J. Am. Chem. Soc. 2015, 137,
6136.
[3] I. Bauer, H.-J. Knçlker, Chem. Rev. 2015, 115, 3170.
[4] For representative examples: a) J. Norinder, A. Matsumoto, N.
Yoshikai, E. Nakamura, J. Am. Chem. Soc. 2008, 130, 5858; b) N.
Yoshikai, A. Matsumoto, J. Norinder, E. Nakamura, Angew.
Chem. Int. Ed. 2009, 48, 2925; Angew. Chem. 2009, 121, 2969;
c) L. Ilies, S. Asako, E. Nakamura, J. Am. Chem. Soc. 2011, 133,
7672; d) R. Shang, L. Ilies, A. Matsumoto, E. Nakamura, J. Am.
Chem. Soc. 2013, 135, 6030; e) S. Asako, L. Ilies, E. Nakamura, J.
Am. Chem. Soc. 2013, 135, 17755; f) Q. Gu, H. H. Al Mamari, K.
Graczyk, E. Diers, L. Ackermann, Angew. Chem. Int. Ed. 2014,
53, 3868; Angew. Chem. 2014, 126, 3949; g) B. M. Monks, E. R.
[12] E. M. Simmons, J. F. Hartwig, Angew. Chem. Int. Ed. 2012, 51,
3066; Angew. Chem. 2012, 124, 3120.
[13] CCDC 1445220 contains the supplementary crystallographic
data for this paper. These data are provided free of charge by
The Cambridge Crystallographic Data Centre.
Received: January 14, 2016
Revised: February 17, 2016
Published online: March 22, 2016
Angew. Chem. Int. Ed. 2016, 55, 5268 –5271
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5271