A computationally designed Rh(I)-catalyzed two-component [5+2+1] cycloaddition of ene-vinylcyclopropanes and CO for the synthesis of cyclooctenones

Article abstract of DOI:10.1021/ja072505w

Through the combined use of computational (density functional theory) and experimental studies, a new [Rh(CO)₂Cl]₂ catalyzed two-component [5+2+1] cycloaddition of ene-vinylcyclopropanes and CO for the synthesis of fused bicyclic cyc

Full text of DOI:10.1021/ja072505w

Published on Web 07/27/2007
A Computationally Designed Rh(I)-Catalyzed Two-Component [5+2+1]
Cycloaddition of Ene-vinylcyclopropanes and CO for the Synthesis of
Cyclooctenones
Yuanyuan Wang,† Jingxin Wang,† Jiachun Su,† Feng Huang,† Lei Jiao,† Yong Liang,† Dazhi Yang,†
Shiwei Zhang,† Paul A. Wender,‡ and Zhi-Xiang 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,
and Departments of Chemistry and of Chemical and Systems Biology, Stanford UniVersity,
Stanford, California 94305-5080
Received April 10, 2007; E-mail: yuzx@pku.edu.cn
Scheme 1. Rationale for the Two-Component [5+2+1] Reaction
The design or discovery of new reactions for the synthesis of
eight-membered rings has been stimulated by their occurrence in
molecules of biomedical importance (e.g., taxol), ligands for
catalysis (e.g., COT, COD), and precursors for natural products
and novel materials (e.g., polyacetylenes).1 The main difficulty
encountered in the synthesis of eight-membered rings is the co-
occurrence of unfavorable entropic and enthalpic factors associated
with ring closure.1 Several transition-metal-catalyzed cycloadditions2
have been introduced in recent years to address this problem
including Ni-, Rh-, or Ru-complex catalyzed [4+4], [6+2],
[5+2+1], and [4+2+2] cycloadditions.3 Herein we report a
computationally designed and experimentally verified, new two-
component [5+2+1] cycloaddition of tethered ene-vinylcyclopro-
panes (ene-VCPs) and CO that provides for the synthesis of ring-
fused cyclooctenones, a ring system encountered in numerous na-
tural products, non-natural targets, and their precursors (Scheme 1).
Our method is based on work from the Wender laboratory leading
to the first metal-catalyzed [5+2] cycloadditions between vinyl-
cyclopropanes and 2-π components (such as alkynes, alkenes, and
allenes) which afford seven-membered ring cycloadducts.4 [Rh(CO)2-
Cl]2 was found to be an especially effective and versatile catalyst
for this and many related cycloadditions.5 Importantly, however,
the intramolecular [5+2] cycloaddition of ene-VCPs has not been
realized using [Rh(CO)2Cl]2 as catalyst,4b even though this trans-
formation can be achieved with RhCl(PPh3)3 or [(arene)Rh(COD)]-
SbF6 catalyst.4c
Scheme 1 (X ) CH2) using the B3LYP method.8 The activation
energies of the RE steps are about 25-30 kcal/mol, whereas the
CO insertion and MRE steps have activation energies of about 13-
14 and 23-24 kcal/mol, respectively, suggesting that the
[5+2+1] path would be favored over the [5+2] path and the
dominant product would be the [5+2+1] cycloadducts.8
Encouraged by the above preliminary calculations, we began our
experimental test. We observed that, in the absence of CO, the [Rh-
(CO)2Cl]2 catalyzed (10 mol %) reaction of ene-VCP substrate 1
at 110 °C did indeed give a double-bond-isomerized [5+2]
cycloadduct 3 in 59% isolated yield, together with 10% of the
Table 1. Optimization Studies of the [5+2+1] Cycloadditions
We speculated that the failure of [Rh(CO)2Cl]2 to catalyze the
[5+2] cycloaddition of ene-VCPs (relative to yne-VCPs and allenyl-
VCPs) could be attributed to the reductive elimination (RE) step
in the catalytic cycle (Scheme 1).6 The RE step in the intramolecular
[5+2] cycloaddition of ene-VCP would lead to the formation of
an (sp3)C-C(sp3) bond. However, it is known that this kind of RE
reaction is not facile when compared to the migratory reductive
elimination (MRE) of an (sp2)C-M-C(sp3) subunit for the
formation of an (sp2)C-C(sp3) bond.6,7 We hypothesized that raising
the reaction temperature or extending the reaction time of the [Rh-
(CO)2Cl]2 catalyzed intramolecular [5+2] cycloaddition of an ene-
VCP could enable this reaction to occur, if no side reactions
compete (hypothesis A). We further hypothesized that the putative
intermediate I (Scheme 1), whose RE to the [5+2] product is
disfavored, would in the presence of CO be converted to intermedi-
ate II, whose MRE to a [5+2+1] cycloadduct would be favored
by the formation of an (sp2)C-C(sp3) bond (hypothesis B).
Before testing our hypotheses experimentally, we computed the
energies for the RE, the CO insertion, and MRE steps shown in
CO
[atm]
catalyst
[mol %]
T
concn
[M]
t
yield
[%]
entry
[
°
C]
solvent
[h]
1
2
3
4
5
6
7
8
0
1
4
10% [Rh(CO)2Cl]2
5% [Rh(CO)2Cl]2
5% [Rh(CO)2Cl]2
110 toluene
80 dioxane
80 dioxane
80 dioxane
60 dioxane
90 dioxane
100 dioxane
80 DCE
80 toluene
80 dioxane
80 dioxane
80 dioxane
80 dioxane
80 dioxane
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.01
0.20
0.05
0.05
0.05
24 10a
5
24
5
44b
8
0.2c 5% [Rh(CO)2Cl]2
70d
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
1
5% [Rh(CO)2Cl]2
5% [Rh(CO)2Cl]2
5% [Rh(CO)2Cl]2
5% [Rh(CO)2Cl]2
5% [Rh(CO)2Cl]2
5% [Rh(CO)2Cl]2
5% [Rh(CO)2Cl]2
10% [Rh(CO)2Cl]2
10% RhCl(PPh3)3
10% RhCl (PPh3)3 +
10% AgOTf
48 17
5
5
5
70
61
62e
9
12 14
5
5
10
11
12
13
14
68
34
72
5
17
N.R.
1
18 23f
15
1
5% [Rh(CO)2Cl]2 +
10% AgOTf
80 dioxane
0.05
13
7
a Accompanied with a [5+2] cycloadduct 3 (59%); see Supporting
Information for details. b Cis/trans ) 5:1. c Conditions: 0.2 atm CO + 0.8
atm N2. d Cis/trans > 20:1. e Cis/trans ) 4:1. f Cis/trans ) 1:1.
† Peking University.
‡ Stanford University.
9
10060
J. AM. CHEM. SOC. 2007, 129, 10060-10061
10.1021/ja072505w CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Rh(I)-Catalyzed [5+2+1] Cycloaddition Reactions of Ene-vinylcyclopropane Substrates and COa
a E ) CO2Me. Isolated yields were reported unless otherwise indicated. b GC yield. Isolated yield is 44% owing to the volatility of the product. c Confirmed
by X-ray analysis. d A [5+2] product was obtained in 11% yield. e Combined yield of diastereomers (trans/cis ) 5:1).
Scheme 2
mechanistic implications suggesting that the starting alkene geom-
etry is conserved through the multistep pathway.
In conclusion, a new Rh(I)-catalyzed [5+2+1] cycloaddition has
been designed computationally and verified experimentally. This
study provides a flexible, convenient, and efficient method for
constructing bicyclic cyclooctenones. The reaction proceeds in good
to excellent yields with a variety of tether types and substitution
patterns and allows for the preparation of 5/8- and 6/8-fused ring
systems even those containing quaternary centers. Further studies
of this reaction (scope, mechanism, and stereochemistry) and its
synthetic applications to natural products are being pursued.
Acknowledgment. We are indebted to generous financial
support from Peking University, the Natural Science Foundation
of China (9800445, 0240203 and 20672005), and the National
Science Foundation USA (Grant CHE-040638). Professor K. N.
[5+2+1] cycloadduct 2 in which the carbonyl group comes from
the CO ligand of the [Rh(CO)2Cl]2 catalyst (Table 1, entry 1). This
result supported our hypothesis A that [5+2] ene-VCP cycloaddi-
Houk of UCLA is highly appreciated for his support to Z.X.Y.
and the theoretic and synthetic organic chemistry lab at PKU.
Supporting Information Available: Computational and experi-
mental details. This material is available free of charge via the Internet
at http://pubs.acs.org.
tions can be achieved when the reaction temperature is increased.
We then turned our attention to promoting the
[5+2+1] cycloaddition. Gratifyingly, substrate 1 in the presence
of CO (balloon, 1 atm) and 5 mol % [Rh(CO)2Cl]2 catalyst in
dioxane, gave after 5 h at 80 °C the [5+2+1] cycloadduct, cyclo-
octenone 2 (cis/trans ) 5:1), in 44% isolated yield (entry 2). We
then systematically optimized the reaction conditions for this new
[5+2+1] cycloaddition using substrate 1. Comparison of the
reactions in entries 2 to 12 suggested that the optimal conditions
for the [5+2+1] cycloaddition were the use of a substrate
concentration of 0.05 M, 5 mol % [Rh(CO)2Cl]2 catalyst, CO
(delivered by balloon admixed with N2 in the ratio of 1:4 at a
pressure of 1 atm),9 dioxane as solvent, and a reaction temperature
of 80-90 °C (entries 4 and 6). Under these conditions, substrate 1
gave the cis-fused [5+2+1] cycloaddition product 2 as a single
diastereomer. Other Rh(I) catalysts such as Wilkinson’s complex
and the cationic rhodium(I) catalysts were found ineffective for
the [5+2+1] cycloaddition (entries 13, 14, and 15).
With the above optimal conditions in hand, we studied the Rh-
(I)-catalyzed [5+2+1] cycloadditions of diverse substrates (Table
2). The results indicate that the cycloaddition reactions are tolerant
of tethers incorporating geminal diester, sulfonamide, and ether
functionalities. Comparison of reaction yields of substrates with
different tethers shows that heteroatom substitution in the tether
results in higher yields of [5+2+1] cycloadducts. This [5+2+1]
cycloaddition also tolerates methyl substitution on the alkene
(substrates 8, 10, and 12) and the VCP (substrates 18, 22, 24, 25,
and 27). Phenyl substitution of the cyclopropane leads efficiently
to the [5+2+1] cycloadducts 13, 15, 17, and 19. In addition to the
5/8-ring system, the trans-fused 6/8-ring system can also be
efficiently established in the cycloaddition reaction from 20 to 21,
albeit with decreased diastereoselectivity. It is noteworthy as well
that the E/Z geometry of the CdC bonds in the VCP moieties of
ene-VCPs affects the cis/trans stereochemistry of the bicyclic
products (Scheme 2). In addition to its synthetic merit, this has
References
(1) (a) Yet, L. Chem. ReV. 2000, 100, 2963. (b) Michaut, A.; Rodriguez, J.
Angew. Chem., Int. Ed. 2006, 45, 5740. (c) Sieburth, S. M.; Cunard, N.
T. Tetrahedron 1996, 52, 6251. (d) Mehta, G.; Singh, V. Chem. ReV.
1999, 99, 881. (e) Wender, P. A.; Croatt, M. P.; Deschamps, N. M. in
ComprehensiVe Organometallic Chemistry III; Ojima, I., Ed.; Elsevier
Ltd: 2007; Vol. 10, pp 603-648. (f) Korshak, Y. V.; Korshak, V. V.;
Kanischka, G.; Hocker, H. Makromol. Chem., Rapid Commun. 1985, 6,
685. (g) Scherman, O. A.; Rutenberg, I. M.; Grubbs, R. H. J. Am. Chem.
Soc. 2003, 125, 8515.
(2) Recent reviews: (a) Lautens, M.; Klute, W.; Tam, W. Chem. ReV. 1996,
96, 49. (b) Saito, S.; Yamamoto, Y. Chem. ReV. 2000, 100, 2901. (c)
Murakami, M. Angew. Chem., Int. Ed. 2003, 42, 718.
(3) (a) Wender, P. A.; Ihle, N. C. J. Am. Chem. Soc. 1986, 108, 4678. (b)
Wender, P. A.; Christy, J. P. J. Am. Chem. Soc. 2006, 128, 5354. (c)
Wender, P. A.; Correa, A. G.; Sato, Y.; Sun, R. J. Am. Chem. Soc. 2000,
122, 7815. (d) Wender, P. A.; Gamber, G. G.; Hubbard, R. D.; Zhang, L.
J. Am. Chem. Soc. 2002, 124, 2876. (e) Wegner, H. A.; de Meijere, A.;
Wender, P. A. J. Am. Chem. Soc. 2005, 127, 6530. (f) Murakami, M.;
Ashida, S. R.; Matsuda, S. J. Am. Chem. Soc. 2006, 128, 2166. (g)
Gilbertson, S.; DeBoef, B. J. Am. Chem. Soc. 2002, 124, 8784. (h) Evans,
P. A.; Baum, E. W.; Fazal, A. N.; Pink, M. Chem. Commun. 2005, 63.
(4) (a) Wender, P. A.; Takahashi, H.; Witulski, B. J. Am. Chem. Soc. 1995,
117, 4720. (b) Wender, P. A.; Sperandio, D. J. Org. Chem. 1998, 63,
4164. (c) Wender, P. A.; Williams, T. J. Angew. Chem., Int. Ed. 2002,
41, 4550.
(5) (a) Wender, P. A.; Pedersen, T. M.; Scanio, M. J. C. J. Am. Chem. Soc.
2002, 124, 15154. (b) Jeong, N.; Sung, B. K.; Choi, Y. K. J. Am. Chem.
Soc. 2000, 122, 6771. (c) Brummond, K. M.; Chen, H.; Sill, P.; You, L.
J. Am. Chem. Soc. 2002, 124, 15186. (d) Wender, P. A.; Paxton, T. J.;
Williams, T. J. J. Am. Chem. Soc. 2006, 128, 14814.
(6) For a theoretical study on the Wender [5+2] cycloaddition, see: Yu, Z.-
X.; Wender, P. A.; Houk, K. N. J. Am. Chem. Soc. 2004, 126, 9154.
(7) Ozawa, F.; Mori, T. Organometallics 2003, 22, 3593 and references
therein.
(8) Computational details, references, and results are given in the Supporting
Information.
(9) It was found that the Pauson-Khand reaction catalyzed by [Rh(CO)2Cl]2
was very effective under a low CO pressure, see: Kobayashi, T.; Koga,
Y.; Narasaka, K. J. Organomet. Chem. 2001, 624, 73.
JA072505W
9
J. AM. CHEM. SOC. VOL. 129, NO. 33, 2007 10061