Another Role of Copper in the SimmonsSmith Reaction: Copper-catalyzed Nucleophilic Michael-type Cyclopropanation of a,β-Unsaturated Ketones

Article abstract of DOI:10.1246/cl.131193

Cyclopropanation was performed using the Furukawa procedure with CH2I2/Et2Zn and a,β-unsaturated ketones. The reaction was performed in the presence of a copper salt. The reactivity was highly dependent on the substrate structure, and cyclopropanated products were obtained in better yields than those achieved using the original SimmonsSmith conditions with a ZnCu couple in some cases. Stereospecificity was observed in a certain case; however, the synthesis of an asymmetric version with a chiral ligand was not successful.

Full text of DOI:10.1246/cl.131193

CL-131193
Received: December 19, 2013 | Accepted: January 9, 2014 | Web Released: January 16, 2014
Another Role of Copper in the Simmons-Smith Reaction: Copper-catalyzed Nucleophilic
Michael-type Cyclopropanation of α,β-Unsaturated Ketones
Kanami Fujii, Tomonori Misaki, and Takashi Sugimura*
Graduate School of Material Science, University of Hyogo, 3-2-1 Kohto, Kamigori, Ako-gun 678-1297
(E-mail: sugimura@sci.u-hyogo.ac.jp)
Cyclopropanation was performed using the Furukawa
procedure with CH2I2/Et2Zn and α,β-unsaturated ketones. The
reaction was performed in the presence of a copper salt. The
reactivity was highly dependent on the substrate structure, and
cyclopropanated products were obtained in better yields than
those achieved using the original Simmons-Smith conditions
with a Zn-Cu couple in some cases. Stereospecificity was
observed in a certain case; however, the synthesis of an
asymmetric version with a chiral ligand was not successful.
O
O
+
+
1
bicycloheptanone (2)
Zn-Cu / CH2I2 / 36h
Pure Zn / CH2I2 / 36h
23%
0%
3%
42%
Scheme 1. Simmons-Smith reaction with a mixed substrate of
2-cyclohexenone (1) and cyclohexene (50/50).
The Simmons-Smith reaction was first reported in 1958,1
and numerous synthetic approaches use this synthetically
valuable cyclopropanation reaction with electron-rich olefins.2
The sluggish reaction between the zinc metal and CH2I2 was first
performed using a zinc-copper couple (Zn-Cu) at the reflux
temperature of ether. In 1966, the Furukawa modification using
the organometallic reagent Et2Zn solved this metal-surface
activation problem to allow cyclopropanation to proceed at low
temperatures in a desired solvent,3 although the true generated
species, commonly represented by IZnCH2I or ICH2ZnCH2I,
may not be the same in the two procedures. Two decades ago,
the Zn-Cu couple was found to be unnecessary to perform the
reaction when zinc was sufficiently pure. A trace lead impurity
drastically deactivated the reactivity of the zinc surface toward
CH2I2; however, in the presence of copper (or trimethylsilyl
chloride), the reactivity increased.4 Despite this current common
understanding, we found that copper plays a role in surface
activation; however, in the case of an electron-deficient olefin,
copper plays a second important role. In fact, the general method
of using the Furukawa procedure for α,β-unsaturated ketones
involves the use of an appropriate copper salt, which may open a
new future for zinc carbenoid reactions.
The Simmons-Smith reaction with an electron-deficient
substrate is limited to one report by Limasset et al.5 In that
report, 2-cyclohexenone (1) with Zn-Cu and CH2I2 resulted in
cyclopropanation to afford 2. The authors also reported that, in
the reaction with a mixture of 1 and cyclohexene (50% each),
1 exhibited greater reactivity toward cyclohexene. Initially, we
investigated the role of the copper additive under the Simmons-
Smith conditions. The same mixture with pure zinc resulted in
bicyclo[4.1.0]heptane in 42% yield, but 2 was not obtained
(Scheme 1). Thus, copper in the Zn-Cu couple clearly changed
the electrophilic property of zinc carbenoid to nucleophilic.
Cyclopropanation of 1 with an electron-deficient olefin was
performed using the Furukawa procedure with CH2I2/Et2Zn. In
the absence of a copper additive, the reaction did not proceed;
however, in the presence of a copper salt, cyclopropanated
product 2 was obtained, as shown in Table 1. For this Michael-
type Furukawa cyclopropanation, both Cu(I) and Cu(II) were
effective.6
Table 1. Isolated yield of 2 from the reaction of 1 with Et2Zn/
CH2I2 in the presence of various copper saltsa
Yield of 2/%
Entry Additive
Additive/equiv: 0.1 0.2 0.5 1.0
1
2
3
4
5
6
CuBr
CuCN
CuBr¢SMe2
CuI¢PBu3
CuOTf
¯2 36
¯2 20
¯2 37 <1
14
8
38
34
13
<1 ¯2 <1 <1
®
®
¯2
39
33
20
®
®
Cu(OTf)2
aThe reaction of 1 was performed at r.t. for 3 days with Et2Zn
(5 equiv) and CH2I2 (10 equiv) in diethyl ether in the presence
of 0.1, 0.2, 0.5, or 1.0 equivalents of copper salts.
Notably, 1 did not react under the conventional Furukawa
cyclopropanation conditions (5 equiv of Et2Zn and 10 equiv of
CH2I2 in diethyl ether at r.t.), which are applicable for most
simple and electron-rich olefins. In the presence of 0.2 or
0.5 equiv of copper salt, 1 was consumed and 2 was obtained in
low to moderate yield. The reaction yield was not improved
through the use of other solvents: 28% in cyclopentyl methyl
ether and 2% in THF (both with 0.2 equiv of CuBr¢SMe2). The
yield dependency on the copper salt amounts is irregular and
the reaction conditions were not completely optimized, but
the subsequent reactions were performed in diethyl ether with
0.2 equiv of Cu(OTf)2, which is the best catalyst reported to
date.7
The copper-catalyzed Furukawa cyclopropanation was
performed with various unsaturated ketones to demonstrate the
synthetic limitations of the reaction and to provide mechanistic
information. In Table 2, the isolated product yields with cyclic
enones achieved using 5 equiv of Et2Zn, 10 equiv of CH2I2, and
0.2 equiv of Cu(OTf)2 in diethyl ether are summarized to show a
comparison of the classic Simmons-Smith reaction with Zn-Cu/
CH2I2 (5 equiv each) in diethyl ether. The ring size of the endo-
cyclic substrates was a dominant factor with respect to product
yield, which increased with increasing ring size (Entries 1-4).
The exo-cyclic substrate in a fixed conformation resulted in a
634 | Chem. Lett. 2014, 43, 634–636 | doi:10.1246/cl.131193
© 2014 The Chemical Society of Japan

Table 2. The product yieldsa with Zn-Cu vs. Et2Zn + Cu
catalyst
Table 3. Cyclopropanation of acyclic substrates
O
O
R2
Et2Zn / CH2I2 / Cu(OTf)2
dry Et2O
R2
R1
R1
Yield/%
Entry
1
Substrate
R3
R1
R3
S–S reaction
Zn–Cu
Furukawa method
Et2Zn + Cu(OTf)2
Entry
R2
R3
Yield/% (time)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Pent
Pent
Pent
Cy
Cy
Bu
Ph
Ph
H
H
Me
H
H
Me
H
H
Me
H
H
H
Ph
H
H
H
H
Me
H
84 (3 h)a
61 (21 h)
49 (21 h)
82 (1 h)a
54 (24 h)
31 (18 h)b
54 (3 h)a
51 (20 h)a
75 (7 h)
O
0
<2
H
O
Me
Me
H
Me
H
H
H
H
H
2b
3
45
39
1
Ph
O
50
p-Anis
p-NO2C6H4
p-CF3C6H4
Ph
2-Fulyl
OC12H25
H
48 (4 h)
13 (9 h)
41
60
23 (Cu salt 0.4 equiv)
<1 (3 h)
20c (3 days)
<1 (4 h)
<1 (3 days)
<1 (3 days)
O
H
H
Ph
4
61
aWith a longer reaction time, the reaction yields produced by
the Simmons-Smith reaction with Zn-Cu were 51% (24 h),
44% (5 h), 21% (4 h), and 44% (23 h) for the substrates shown
in Entries 1, 4, 7, and 8, respectively. bThe product decomposed
gradually (<1%, 3 days). cIntramolecular O-alkylation fol-
lowed by the second cyclopropanation afforded 2,4-diphenyl-
4,5-dihydrofuran, which was an abnormal product, in 19%
yield in addition to the 20% yield of the expected product.
O
5
6
71
0c
41
3
O
depends on the substrate structure; however, in some cases, the
yields were better than that of the reported Zn-Cu cyclo-
propanation.
O
O
Decomp.
(14)
Decomp.
(<21)
7
8
In contrast to the conformationally regulated cyclic enones,
acyclic enones were more reactive, as reported in Table 3. The
terminal olefin in aliphatic α,β-unsaturated ketones afforded high
yields of 82%-84% (Entries 1 and 4). The aromatic ketone
represented by Entry 7 showed poorer reactivity (54%), but the
yield increased to 75% with the α-methyl substitution (Entry 9).
Low reactivity with the aromatic enone (Entry 7) was likely due
to the electron-withdrawing nature of the phenyl substituent;
in fact, electron-withdrawing p-substitutions further reduced
the product yields (Entries 11 and 12). An unsaturated ester
(Entry 15) and aldehyde (Entry 16) were inert under the
investigated conditions.
One of the advantages of the reaction used in the present
study is its ability to provide a motif for an asymmetric reaction
that forms two C-C bonds and two chiralities. In the case of the
Zn-Cu reaction, the introduction of a chiral ligand or some other
chiral source can be difficult. To establish this new asymmetric
reaction by further study, we performed two additional series of
preliminary experiments.
0c
0c
O
9
25
0c
aThe reaction was performed until all of the substrate was
consumed unless otherwise noted. Product decompositions and
reactant side reactions were difficult to distinguish. bThe
reaction was performed at r.t. for 3 days with Et2Zn (5 equiv),
CH2I2 (10 equiv), and Cu(OTf)2 (0.2 equiv) in diethyl ether.
cNo reaction.
moderate yield (Entry 5). The reactivity became negligible with
the introduction of a β-methyl substituent (Entry 6). The low
product yields achieved with 5,5-disubstituted 1a (Entry 7) and
exo-carbonyl (Entry 9) are noteworthy for their implications
related to the reaction mechanism. In summary, the product yield
When a diastereomeric mixture of an aromatic enone
(R1 = Ph) in a ratio of E/Z = 2.4/1 was reacted under the
copper-catalyzed Furukawa procedure, the ratio of the remaining
substrate was unchanged, whereas the ratio of the product
decreased to 1.2/1 (Scheme 2). In contrast, an aliphatic ketone
Chem. Lett. 2014, 43, 634–636 | doi:10.1246/cl.131193
© 2014 The Chemical Society of Japan | 635

O
O
2
3
For reviews, see: a) H. E. Simmons, T. L. Cairns, S. A.
Vladuchick, C. M. Hoiness, Organic Reactions, Wiley,
1973, Vol. 20, p. 1. doi:10.1002/0471264180.or020.01.
b) W. B. Motherwell, C. J. Nutley, Contemp. Org. Synth.
1994, 1, 219. c) A. B. Charrette, A. Beacjemin, Organic
Reactions, Wiley, 2001, Vol. 58, p. 1. doi:10.1002/
0471264180.or058.01. d) M. Nakamura, A. Hirai, E.
Nakamura, J. Am. Chem. Soc. 2003, 125, 2341.
a) J. Furukawa, N. Kawabata, J. Nishimura, Tetrahedron
Lett. 1966, 7, 3353. b) J. Furukawa, N. Kawabata, J.
Nishimura, Tetrahedron 1968, 24, 53.
K. Takai, T. Kakiuchi, K. Utimoto, J. Org. Chem. 1994, 59,
2671.
Et2Zn / CH2I2 / Cu(OTf)2
dry Et2O
R1
R1
R1 = Ph or Pent
Scheme 2. Stereospecificity of the present reaction.
Ph
O
O
O
O
Ph
Ph
Ph
OH
N
P
N
N
N
Ph
HO
OH
4
5
6
7
ligand 3
ligand 4
ligand 5
J.-C. Limasset, P. Amice, J.-M. Conia, Bull. Soc. Chim. Fr.
1969, 11, 3981.
M. Kitamura, T. Miki, K. Nakano, R. Noyori, Tetrahedron
Lett. 1996, 37, 5141.
Figure 1. Chiral ligands employed for the present reaction
with cycloheptenone.
(R1 = Pent, E/Z = 2.2/1) afforded a product with an E/Z ratio
of 2.0/1, which is identical to the starting ratio within the
experimental error range (1H NMR). Hence, the stereospecificity
of the new reaction is currently limited to aliphatic enones or
cyclic enones.
Finally, some representative chiral ligands of Cu(OTf)2 for
the dialkylzinc reagents were tested with cyloheptenone. With
the addition of ligand 38 shown in Figure 1 (left, 0.2 equiv), the
reaction rate increased, but the product was determined to be
racemic by chiral GLC analysis (64% yield). Another represen-
tative chiral ligand 49 also failed to yield the optically active
product (90% yield). The zinc species itself can be a chiral
complex; however, ligand 510 did not lead to a stereoselective
reaction (76% yield).
In this investigation, the Simmons-Smith reaction was
determined to be not only effective for electron-rich double
bonds but also for electron-deficient double bonds through
the use of the Furukawa procedure with a copper salt. The
developed method would be useful in organometallic chemistry,
and our results indicate that more extensive study is necessary to
understand all the reaction mechanism.
a) V. Wendisch, N. Sewald, Tetrahedron: Asymmetry 1997,
8, 1253; N. Krause, Angew. Chem., Int. Ed. 1998, 37, 283.
b) Y. Hu, J. Yu, S. Yang, J.-X. Wang, Y. Yin, Synth.
Commun. 1998, 28, 2793. c) B. L. Feringa, Acc. Chem. Res.
2000, 33, 346. d) A. Alexakis, J. Burton, J. Vastra, C.
Benhaim, X. Fournioux, A. van den Heuvel, J.-M. Levêque,
F. Mazé, S. Rosset, Eur. J. Org. Chem. 2000, 4011. e) B. L.
Feringa, R. Naasz, R. Imbos, L. A. Arnold, in Modern
Organocopper Chemistry, ed. by N. Krause, Wiley-WCH,
Verlag, 2002, Chap. 7. doi:10.1002/3527600086.ch7. f) N.
Shibata, M. Yoshimura, H. Yamada, R. Arakawa, S.
Sakaguchi, J. Org. Chem. 2012, 77, 4079.
a) E. Gallo, F. Ragaini, L. Bilello, S. Cenini, C. Gennari, U.
Piarulli, J. Organomet. Chem. 2004, 689, 2169. b) L. Palais,
L. Babel, A. Quintard, S. Belot, A. Alexakis, Org. Lett.
2010, 12, 1988. c) A. Rimkus, N. Sewald, Synthesis 2004,
135. d) A. Alexakis, J. E. Bäckvall, N. Krause, O. Pàmies,
M. Diéguez, Chem. Rev. 2008, 108, 2796.
8
9
a) M. Schinnerl, M. Seitz, A. Kaiser, O. Reiser, Org. Lett.
2001, 3, 4259. b) M. Schinnerl, C. Böhm, M. Seitz, O.
Reiser, Tetrahedron: Asymmetry 2003, 14, 765. c) V. K.
Aggarwal, L. Bell, M. P. Coogan, P. Jubault, J. Chem. Soc.,
Perkin Trans. 1 1998, 2037.
References and Notes
1
a) H. E. Simmons, R. D. Smith, J. Am. Chem. Soc. 1958, 80,
5323. b) H. E. Simmons, R. D. Smith, J. Am. Chem. Soc.
1959, 81, 4256.
10 a) K. Soai, S. Yokoyama, T. Hayasaka, J. Org. Chem. 1991,
56, 4264. b) K. Soai, S. Yokoyama, K. Ebihara, T. Hayasaka,
J. Chem. Soc., Chem. Commun. 1987, 1690.
636 | Chem. Lett. 2014, 43, 634–636 | doi:10.1246/cl.131193
© 2014 The Chemical Society of Japan