Rhodium-Catalyzed [4+2+1] Cycloaddition of In Situ Generated Ene/Yne-Ene-Allenes and CO

Article abstract of DOI:10.1002/anie.201805908

Reported herein is the first rhodium-catalyzed [4+2+1] cycloaddition of in situ generated ene/yne-ene-allenes and CO to synthesize challenging seven-membered carbocycles fused with five-membered rings. This reaction is designed based on the 1,3-acyloxy mi

Full text of DOI:10.1002/anie.201805908

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Rh-Catalyzed [4 + 2 + 1] Cycloaddition of in situ Generated Ene/
Yne-Ene-Allenes and CO
Authors: Zhixiang Yu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201805908
Angew. Chem. 10.1002/ange.201805908
Link to VoR: http://dx.doi.org/10.1002/anie.201805908
http://dx.doi.org/10.1002/ange.201805908

10.1002/anie.201805908
Angewandte Chemie International Edition
COMMUNICATION
Rh-Catalyzed [4+2+1] Cycloaddition of in situ Generated Ene/Yne-
Ene-Allenes and CO
Zi-You Tian, Qi Cui, Cheng-Hang Liu, Zhi-Xiang Yu*
Abstract: We report here the first Rh-catalyzed [4+2+1]
cycloaddition of in situ generated ene/yne-ene-allenes and CO to
synthesize challenging seven-membered carbocycles fused with
five-membered rings. This reaction is designed based on the 1,3-
acyloxy migration of ene/yne-ene-propargyl esters to ene/yne-
ene-allenes, followed by oxidative cyclization, CO insertion and
reductive elimination to form the final [4+2+1] cycloadducts. The
possible competing [4+1], [4+2] and [2+2+1] cycloadditions were
disfavored, making the present reaction as an efficient way to
access functionalized 5/7 rings.
In recent years, transition metal catalyzed cycloadditions
such as [5+2],[3] [6+1],[4] [4+3],[5] [3+2+2],[6] et al. have emerged as
powerful tools to synthesize various carbocycles. One easy way
to synthesize seven-membered carbocycles is to develop
transition metal-catalyzed [4+2+1] reaction of ene/yne-dienes
with one-carbon synthons (such as CO, carbenes[7]), considering
that both ene/yne-dienes and CO (or carbenes) are readily
accessible. This [4+2+1] cycloadditions, which are easily
conjectured but difficult to be realized, have only one successful
methodology (Scheme 1a). Montgomery’s group showed that
under Ni catalysis, yne-diene and trimethylsilyldiazomethane can
give seven-membered carbocycles.[8a] This Ni catalyzed [4+2+1]
reaction was proposed to occur through [3, 3] sigmatropic
rearrangement of in situ generated divinylcyclopropanes.[8b] In
2003, Wender’s group proposed a creative [4+2+1] reaction of
yne-dienes and CO, but the major products in their reaction were
Seven-membered carbocyclic rings are privileged skeletons
widely found in natural products[1] with biological and medicinal
significances (for example, colchicine, guanacastepene, ingenol).
However, the synthesis of seven-membered carbocycles is still
[2]
posing challenges
to chemists due to the facts that 1) only
[4+2] and Pauson–Khand [2+2+1] cycloadducts (Scheme 1b)[9a]
.
limited number of reactions and strategies are available; and 2)
many molecules with seven-membered carbocycles have
different substitution patterns and stereocenters, which could be
difficult or impossible by the previously reported methods and
strategies of seven-membered ring synthesis. Therefore,
developing new reactions to access seven-membered
carbocycles, which could either expand or complement previous
reactions, is an important frontier in reaction development.
They then optimized this reaction and expanded the 2C synthon
to alkenes and allenes to develop several powerful [2+2+1]
reactions.[9b-e] Herein we report the second successful reaction of
transition metal catalyzed [4+2+1] cycloaddition of in situ
generated ene/yne-ene-allenes and CO to achieve the synthesis
of bicyclic 5/7 ring system (Scheme 1c).
Scheme 2 shows our design of the present [4+2+1]
cycloaddition. Ene/yne-ene-propargyl esters were used as the
substrates and we reasoned that under Rh catalysis, 1,3-acyloxy
migration[10] of ene/yne-ene-propargyl esters could occur to give
ene/yne-ene-allene complex I.[11] Then oxidative cyclization took
place to give five-membered rhodacycle II,[12] which was then
converted to intermediate III via alkene/alkyne insertion. After that,
CO inserted into the C–Rh bond to form intermediate IV (it is also
Scheme 1. [4+2+1] cycloadditions.
[*]
Z.-Y. Tian, Q. Cui, C.-H. Liu, Prof. Dr. Z.-X. Yu
Beijing National Laboratory for Molecular Sciences (BNLMS) Key
Laboratory of Bioorganic Chemistry and Molecular Engineering of
Ministry of Education, College of Chemistry
Peking University, Beijing, 100871 (China)
E-mail: yuzx@pku.edu.cn
Scheme 2. Proposed reaction pathway for the [4+2+1] cycloaddition.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201805908
Angewandte Chemie International Edition
COMMUNICATION
possible to form another intermediate IV’, which is not present in
this Scheme but in Scheme S1 of the Supporting Information, by
CO insertion into another C–Rh bond in intermediate III). Finally,
reductive elimination from intermediate IV generated the final
[4+2+1] cycloadduct. Several possible side reactions such as
[4+1] (from intermediate II),[13] [4+2] (from intermediate III)[11] and
[2+2+1] (from intermediate IV),[9] which had been previous
realized separately by similar intermediates, could occur in this
situation and completely make our efforts to [4+2+1] cycloaddition
fruitless.
observed (Table 1, entry 1). The triene byproduct 3a was obtained
by isomerization from intermediate I (see Scheme S1 in
Supporting Information) and the structure of product 2a was
further confirmed by X-ray analysis (see the Supporting
Information). Several other Rh(I) catalysts with different ligands
(COD, COE, and NBD) were also screened, finding that almost
the same reaction yields (78-83%) and ratio of 2a/3a (10/1) were
Table 2. Reaction scope of the [4+2+1] cycloaddition.[a]
We hypothesized that transformation from I to II is an
intramolecular process and could be favored over possible
competing intramolecular [4+1] reaction pathway of I with CO.
Intermediate III has a strong Rh–C bond with the original allene
moiety and would not be easy to undergo reductive elimination to
give [4+2] cycloadduct.[12] Instead, CO insertion to give
intermediate IV could be preferred over the competitive [4+2]
reaction. In theory, intermediate IV could undergo reductive
elimination to give [2+2+1] reaction, but we hypothesized that IV
is an eight-membered rhodacycle with strong Rh–C bond of the
allene moiety (not a Rh-allylic bond). Consequently, generation of
seven-membered product via [4+2+1] pathway should be favored
for IV. Generation of [2+2+1] product could have a six-membered
rhodacycle with a Rh-allylic bond, as hypothesized in previous
[2+2+1] reaction of yne-dienes.[9,14] Our group will investigate
these hypotheses by DFT calculations in the future. Fortunately,
we were happy to find that the present [4+2+1] cycloaddition
worked. Therefore, we report our experimental results of the
[4+2+1] cycloaddition here.
Table 1 Optimization of reaction conditions for 1a and CO.[a]
We tested our designed [4+2+1] cycloaddition using the N-
Ts tethered ene-ene-propargyl ester 1a as the standard substrate
(Table 1). Firstly, we tested our reaction using 10 mol%
[Rh(CO)2Cl]2 catalyst, which was efficient for 1,3-acyloxy
migration[10] reaction, finding that 81% of the target product could
be obtained in DCE after 12 hours, together with isomerization
product 3a with a ratio of 10/1, and no apparent [4+2] product was
This article is protected by copyright. All rights reserved.

10.1002/anie.201805908
Angewandte Chemie International Edition
COMMUNICATION
realized (Table 1, entries 2–4). We hypothesized that these Rh
catalysts gave the same catalytic species, through replacement
of these ligands by CO molecule. In addition, catalyst
Rh(CO)(PPh3)2Cl was found to be ineffective and the reaction
substrate was recovered (Table 1, entry 5). Solvents other than
DCE were found to be less effective for the [4+2+1] cycloaddition
using [Rh(COD)Cl]2 as catalyst (Table 1, entries 6 – 10). For
example, a complex mixture was obtained for the reaction carried
out in DCM. In additon, the [4+2+1] cycloaddition can be carried
out at higher temperature, and a yield of 82% was reached as that
from the reaction at 40 °C (Table 1, entry 11). We were happy to
find that the [4+2+1] cycloaddition can be operated by using 5
mol% of [Rh(COD)Cl]2 to give comparable yield as that from the
reaction using 10 mol% catalyst, even though the reaction had to
extend to 24 h (Table 1, entry 12). Finally, we investigated how
the CO pressure affected the reaction.[3d] Using 0.2 atm CO
pressure was less effective and the reaction yield was only 69%
(Table 1, entry 13). Based on the results in Table 1, we then
decided to study the scope of the [4+2+1] cycloaddition using the
conditions in entry 12 of Table 1 for ene-ene-allene susbtrates
(Table 2). Interestingly, the triene byproduct 3a was isolated as
the major product when using iridium catalyst instead of rhodium
catalysts (Table 1, entry 14).
We also investigated the influence of the substitution patterns
in the propargyl ester moiety of the substrates. Substrates with
two substituted groups attached to the propargyl ester moiety
showed good reactivity, and moderate to good reaction yields
were obtained (Table 2, entries 2–4). Substrate 1e with only one
substituted group in the propargyl ester moiety was slow for the
[4+2+1] reaction under the standard conditions. Consequently,
higher reaction temperature (75 °C) and higher catalyst loading
(10%) had to be applied for substrate 1e, which, under the new
reaction conditions, gave a moderate yield (44%) and Z/E-mixture
of [4+2+1] cycloadducts.
It was anticipated that the allene moiety of the ene-ene-allene
could not have two large groups, otherwise, the [4+2+1]
cycloadduct could suffer from severe repulsion because the
corresponding allene insertion transition states in the [4+2+1]
reaction could be difficult to be accessed. This was proved to be
true because terminal diphenyl substituted substrate 1m
decomposed under the standard reaction conditions (Scheme 3).
We also tested whether the two-π component in the substrates
can have substituents at its terminal position. Complex mixtures
could be achieved using substrates 1n and 1o. In addition,
substrate 1p with two methyl groups in the terminal position of the
two-π ene part were synthesized, and the corresponding triene
product was isolated in 55% yield, without the formation of [4+2+1]
cycloadduct (see the Supporting Information for details).
We also synthesized yne-ene-propargyl ester substrate 1q[12]
to study whether the 2C synthon of the [4+2+1] cycloaddition can
be alkynes. To our delight, under the optimized reaction
conditions for ene-ene-propargyl ester substrates, the [4+2+1]
cycloaddition of 1q can give product 2q in 51% yield. Further
study of [4+2+1] cycloaddition of more yne-ene-propargyl ester
susbtrates will be carried out in the future.
Herein we want to point out that, lower yields for some
substrates in Table 2 usually had some isomerization products,
as judged by TLC (for example, 1h). Since we only concentrated
on developing a method for the 5/7 ring synthesis, we neither
isolated nor characterized these side products (except 3a). Under
the optimization reaction conditions for all substrates in Table 2,
only trace amount of [4+2] products could be detected by TLC.
In summary, a novel Rh(I)-catalyzed two-component [4+2+1]
cycloaddition of CO and in situ generated ene/yne-enynes from
conjugated ene/yne-yne propargyl esters has been developed.
This reaction features easily prepared substrates and broad
scope to reach 5/7 skeleton. Quaternary carbon centers can also
be introduced at the bridgehead position of the bicyclic 5/7 system.
This methodology represents the second successful example of
transition metal catalyzed [4+2+1] cycloaddition. Further study of
the reaction scope, its application in synthesis, and understanding
the reaction mechanism is ongoing.
We were happy to observe that substrates 1f and 1g, which
did not have the substitution in the internal alkene part of ene-
ene-ynes gave good yields (70% and 75%, respectively, see
Table 2, entries 6 and 7). Substrate 1h bearing methyl
substitutions in alkene moiety of the ene-ene-ynes generated
[4+2+1] cycloadduct 2h with bridgehead quaternary carbon
center in 44% yield (Table 2, entry 8).
Acknowledgements
We changed the N-Ts tether in the substrate to C and O
tethers to further investigate the reaction scope. The [4+2+1]
cycloadditions gave 94% and 85% reaction yields for 1i (with C
tether) and 1k (with O tether), respectively (Table 2, entries 9 and
11). We were excited to note that substrates 1j with C tether and
1l with O tether can reach high yields of [4+2+1] products 2j and
2l, both of which have a methyl group at the bridgehead positions
of the 5/7 skeleton (Table 2, entries 10 and 12).
We thank the National Natural Science Foundation of China
(21472005) for financial support. We also thank Prof. Dr. Wen-
Xiong Zhang and Dr. Neng-Dong Wang of Peking University for
X-ray crystal analysis.
Keywords: [4+2+1] • rhodium • ene/yne-ene-allene • 1,3-acyloxy
migration •cycloaddition
[1]
a) R. A. Craig, II., B. M. Stoltz, Chem. Rev. 2017, 117, 7878; b) M. Liu,
W.-G. Wang, H.-D. Sun, J.-X. Pu, Nat. Prod. Rep. 2017, 34, 1090.
T. V. Nguyen, J. M. Hartmann, D. Enders, Synthesis 2013, 45, 845.
For selected reviews, see: a) X.-Z. Shu, D. Shu, C. M. Schienebeck, W.
Tang, Chem. Soc. Rev. 2012, 41, 7698; b) K. E. O. Ylijoki, J. M. Stryker,
Chem. Rev. 2013, 113, 2244; c) H. Pellissier, Adv. Synth. Catal. 2011,
353, 189; d) Y. Wang, Z.-X. Yu, Acc. Chem. Res. 2015, 48, 2288. For
selected recent [5+2] reactions, see: C.-H. Liu, Z.-X. Yu, Angew. Chem.,
Int. Ed. 2017, 56, 8667; Angew. Chem. 2017, 129, 8793.
[2]
[3]
Scheme 3. Unsuccessful substrates for [4+2+1] cycloaddition.
This article is protected by copyright. All rights reserved.

10.1002/anie.201805908
Angewandte Chemie International Edition
COMMUNICATION
[4]
P. A. Wender, N. M. Deschamps, R. Sun, Angew. Chem., Int. Ed. 2006,
45, 3957.
Deschamps, Angew. Chem., Int. Ed. 2006, 45, 2459; Angew. Chem.
2006, 118, 2519. For a review, see: e) M. P. Croatt, P. A. Wender, Eur.
J. Org. Chem. 2010, 19.
[5]
For selected reviews: a) H. M. R. Hoffmann, Angew. Chem., Int. Ed. 1984,
23, 1; Angew. Chem. 1983, 95, 29; b) I. V. Hartung, H. M. R. Hoffmann,
Angew. Chem., Int. Ed. 2004, 43, 1934; Angew. Chem. 2004, 116, 1968;
c) M. Harmata, Adv. Synth. Catal. 2006, 348, 2297; (d) M. Harmata, M.
Kahraman, G. Adenu, C. L. Barnes, Heterocycle 2004, 62, 583.
B. Kuila, M. Kaur, P. Singh, G. Bhargava, Eur. J. Org. Chem. 2018, 853.
For stoichiometric [4+2+1] reactions of yne-dienes and metal-carbenes
without using transition metal catalysts, see: a) D. F. Harvey, K. P. Lund,
J. Am. Chem. Soc. 1991, 113, 5066. b) D. F. Harvey, E. M. Grenzer, P.
K. Gantzel, J. Am. Chem. Soc. 1994, 116, 6719. Another [4+2+1]
reaction of cyclopropylcarbene-tunsten complexes and alkynes was
reported, see: c) J. W. Herndon, G. Chatterjee, P. P. Patel, J. J. Matasi,
S. U. Tumer, J. J. Harp, M. D. Reid, J. Am. Chem. Soc. 1991, 113, 7808.
A formal intramolecular [4+2+1] reaction of carbyne, alkyne, diene
involving [3,3]-rearrangement of in situ generated divinylcyclopropanes
was reported: d) A. Padwa, K. E. Krumpe, Y. Gareau, U. Chiacchio, J.
Org. Chem. 1991, 56, 2523.
[10] a) D. Shu, X. Li, M. Zhang, P. J. Robichaux, I. A. Guzei, W. Tang, J. Org.
Chem. 2012, 77, 6463; D. b) Shu, X. Li, M. Zhang, P. J. Robichaux, W.
Tang, Angew. Chem., Int. Ed. 2011, 50, 1346; Angew. Chem. 2011, 123,
1382; c) X. Li, M. Zhang, D. Shu, P. J. Robichaux, S. Huang, W. Tang,
Angew. Chem., Int. Ed. 2011, 50, 10421; Angew. Chem. 2011, 123,
10605.
[6]
[7]
[11] a) M. Murakami, M. Ubukata, K. Itami, Y. Ito, Angew. Chem., Int. Ed.
1998, 37, 2248; Angew. Chem. 1998, 110, 2362; b) M. Murakami, R.
Minamida, K. Itami, M. Sawamura, Y. Ito, Chem. Commun. 2000, 23,
2293; c) M. Murakami, K. Itami, Y. Ito, J. Am. Chem. Soc. 1997, 119,
7163.
[12] S. Huang, X. Li, C. L. Lin, I. A. Guzei, W. Tang, Chem. Commun. 2012,
48, 2204.
[13] a) X. Li, S. Huang, C. M. Schienebeck, D. Shu, W. Tang, Org. Lett. 2012,
14, 1584; b) W. Chen, J.-H. Tay, X.-Q. Yu, L. Pu, J. Org. Chem. 2012,
77, 6215; c) Y. Zhang, G. Zhao, L. Pu, Eur. J. Org. Chem. 2017, 7026;
d) M. Murakami, K. Itami, Y. Ito, Angew. Chem., Int. Ed. 1995, 34, 2691;
Angew. Chem. 1995, 107, 2943; e) M. Murakami, K. Itami, Y. Ito, J. Am.
Chem. Soc. 1997, 119, 2950; f) M. Murakami, K. Itami, Y. Ito, J. Am.
Chem. Soc. 1999, 121, 4130; g) M. Murakami, K. Itami, Y. Ito, J. Am.
Chem. Soc. 1996, 118, 11672.
[8]
[9]
a) Y. Ni, J. Montgomery, J. Am. Chem. Soc. 2004, 126, 11162; b) Y. Ni,
J. Montgomery, J. Am. Chem. Soc. 2006, 128, 2609.
a) P. A. Wender, N. M. Deschamps, G. G. Gamber, Angew. Chem., Int.
Ed. 2003, 42, 1853, Angew. Chem. 2003, 115, 1897. The reactions of
ene-dienes and allene-dienes with CO also gave [2+2+1] products, see:
b) P. A. Wender, M. P. Croatt, N. M. Deschamps, J. Am. Chem. Soc.
2004, 126, 5948; c) W. H. Pitcock, Jr., R. L. Lord, M.-H. Baik, J. Am.
Chem. Soc. 2008, 130, 5821; d) P. A. Wender, M. P. Croatt, N. M.
[14] Q. Li, Z.-X. Yu, Organometallics 2012, 31, 5185.
This article is protected by copyright. All rights reserved.

10.1002/anie.201805908
Angewandte Chemie International Edition
COMMUNICATION
COMMUNICATION
Zi-You Tian, Qi Cui, Cheng-Hang Liu,
Zhi-Xiang Yu*
Page No. – Page No.
Rh-Catalyzed [4+2+1] Cycloaddition
of in situ Generated Ene/Yne-Ene-
Allenes and CO
4+2+1=7, mathematically simple, but chemically difficult: Rh-catalyzed [4+2+1]
cycloaddtion of in situ generated ene/yne-ene-allenes and CO to synthesize
challenging seven-membered carbocycles fused with five-membered rings is
reported, where ene-allene, alkene/alkyne and CO act as 4C, 2C, and 1C synthons,
respectively.
This article is protected by copyright. All rights reserved.