Improved zinc-catalyzed simmons-smith reaction: Access to various 1,2,3-trisubstituted cyclopropanes

Article abstract of DOI:10.1021/ol500267w

The Simmons-Smith reaction of zinc carbenoids with alkenes is a powerful method to access cyclopropanes containing various substitution patterns. This work exploits the high reactivity of aryldiazomethanes toward zinc halides to generate aryl-substituted carbenoids catalytically. These carbenoids are able to cyclopropanate various alkenes diastereoselectively, including unfunctionalized substrates such as styrenes. The zinc catalyst can be modified to tolerate the use of free allylic alcohols.

Full text of DOI:10.1021/ol500267w

Letter
pubs.acs.org/OrgLett
Improved Zinc-Catalyzed Simmons−Smith Reaction: Access to
Various 1,2,3-Trisubstituted Cyclopropanes
́
Eric Levesque, Sebastien R. Goudreau, and Andre B. Charette*
́
́
́
Centre in Green Chemistry and Catalysis, FAS-Department of Chemistry, Universite
Downtown, Montreal, Quebec, Canada, H3C 3J7
́ ́
de Montreal, P.O. Box 6128, Station
́
́
S
* Supporting Information
ABSTRACT: The Simmons−Smith reaction of zinc carbenoids
with alkenes is a powerful method to access cyclopropanes
containing various substitution patterns. This work exploits the
high reactivity of aryldiazomethanes toward zinc halides to generate
aryl-substituted carbenoids catalytically. These carbenoids are able
to cyclopropanate various alkenes diastereoselectively, including
unfunctionalized substrates such as styrenes. The zinc catalyst can
be modified to tolerate the use of free allylic alcohols.
yclopropanes are important structural units in the field of
excess of alkene11a or are limited to sodium allyloxides.12
Herein we report the scope and applications study of the first
zinc-catalyzed system to cyclopropanate styrenes13 and allylic
ethers efficiently (Scheme 2). Also, a simple modification of the
catalyst enables direct cyclopropanation of free allylic alcohols
without deprotonation.
C
synthetic chemistry. They are present in many natural and
bioactive products1 and can also participate in useful chemical
reactions, taking advantage of their unique ring strain.2 The
zinc-mediated reaction of gem-diiodoalkanes with alkenes, first
reported by Simmons and Smith,3 is among the most widely
used approaches to access complex cyclopropanes.4 The typical
procedure5−8 involves generating an excess of the desired zinc
carbenoid9 (1) before introducing the substrate alkene
(Scheme 1). Such a procedure can suffer from solubility issues,
Scheme 2. Arylcyclopropanations Using a Catalytic Amount
of Zinc
Scheme 1. Typical Simmons−Smith Procedure
We started our investigation by treating the benzyl ether of
cinnamyl alcohol (2a) with zinc iodide14 and slowly adding
phenyldiazomethane at −20 °C (Table 1, entry 1). During the
addition, significant accumulation of the deep red diazo reagent
was noticed, only fading away when the mixture was warmed to
room temperature. As zinc carbenoids react faster with diazo
compounds than with alkenes,11b this led to poor conversions
and the undesired formation of stilbenes. Adding the diazo
reagent at room temperature increased the turnover rate
enough to keep its concentration low at all times (Table 1,
entry 2), thus favoring the desired quantitative formation of
cyclopropane. This also has the advantage of reducing reaction
times such that complete conversion was reached shortly after
the end of the diazo compound addition.
as coordinating additives are often required to solubilize the
zinc reagent.7,8,9b On a larger scale, the handling of air-sensitive
organozinc compounds5 and the generation of metal-containing
waste also become problematic. Using only a catalytic amount
of zinc would be a straightforward way to circumvent these
issues.
The key for the establishment of a viable catalytic cycle
would be to use a precursor able to regenerate the reactive
carbenoid from the zinc halide produced after the cyclo-
propanation. Alkyl-10 and aryl-11 diazo reagents have been
demonstrated to fit such a requirement. Recently, our group
successfully used this approach to generate enantioenriched
arylcyclopropanes using stoichiometric amounts of both zinc
halide and chiral tartrate-derived dioxaborolane.12 Furthermore,
a few examples of this transformation using catalytic amounts of
zinc halides have been reported, but they either require a large
To explore substrate generality, a series of protected allylic
alcohols were submitted to the ZnI2-catalyzed reaction
Received: January 28, 2014
Published: February 20, 2014
© 2014 American Chemical Society
1490
dx.doi.org/10.1021/ol500267w | Org. Lett. 2014, 16, 1490−1493

Organic Letters
Letter
Table 1. Screening of the Cyclopropanation Conditions
b
b
entry
initial R
temp (°C)
Zn (mol %)
X anion
time (h)
yield 5a (%)
yield 3a (%)
1
2
3
4
5
6
7
Bn
Bn
H
−20 to rt
10
10
10
10
10
10
I−
I−
16
2.5
2.5
2.5
2.5
2.5
2.5
−
−
60
65
65
70
81
48
99
23
8
rt
rt
rt
rt
rt
rt
I−
−
H
(BuO)2PO2
PhO−
2,6-t-Bu2C6H3O−
2,6-t-Bu2C6H3O−
H
5
H
5
H
13
8
a
b
0.75 M solution in CH2Cl2, added dropwise over 1.5 h. Yields determined on the crude reaction mixture by 1H NMR analysis using Ph3CH as an
internal standard.
a
Scheme 3. Synthesis of Phenylcyclopropanes from Protected Allylic Alcohols and Styrenes
a
Yields measured by 1H NMR of crude mixture using Ph3CH or DMAP as an internal standard. b 0.75 M solution in CH2Cl2, added dropwise over
1.5 h. c Isolated yield, major diastereomer. d Isolated yield, both diastereomers combined. e 15 mol % of ZnI2 and 2.5 equiv of phenyldiazomethane
were used. 13 mol % of 2,6-t-Bu2C6H3OZnI were used instead of ZnI2.
f
(Scheme 3).15 It is noteworthy that aryl-substituted allylic
ethers (3a−c) reacted with higher yields and higher
diastereoselectivity than alkyl-substituted allylic ethers (3d−
g). This led us to believe that styrenes may be sufficiently
reactive to undergo this reaction without a Lewis basic directing
group. Indeed, various styrene derivatives reacted smoothly
(3h−n), only needing a slight increase in catalyst loading to
improve the conversion (Scheme 3). To our knowledge, this is
the first report of a zinc-catalyzed cyclopropanation of styrene
derivatives. While similar transformations can be performed
with other metals such as Rh,16 Fe,17 Ru,18 or Os,19 the diazo is
always the limiting reagent, and large excesses (5−10 equiv) of
the substrate are often required. Here the alkene is the limiting
reagent and the required excess of diazo (2.5 equiv) is relatively
small. This is probably due to the reactivity difference between
zinc carbenoids and the carbenes usually formed with other
metals.
infra), a reaction usually considered fast and quantitative.20,21
The SN2 reaction between the resulting benzyl iodide and zinc
alkoxide yields the benzyl ether 3a. As only one Zn−I bond is
necessary to generate the carbenoid, we envisioned modulating
its reactivity by replacing the catalyst’s second iodide with
another anion.22 A screening of common anionic ligands23
showed that zinc iodophosphates (entry 4) and iodophenoxides
(entries 5−7) were suitable catalysts for this reaction. The
hindered zinc phenoxide obtained from the deprotonation of
2,6-di-tert-butylphenol (2,6-t-Bu2C6H3OZnI) gave the best
results (entries 6−7).24
This catalyst enabled us to cyclopropanate allylic alcohols of
various substitution patterns (Scheme 4) with a reduced
amount of benzylated byproducts. Furthermore, some sub-
strates (4c−e) only required 1.6 equiv of diazo reagent to
achieve complete conversion, and no significant amount of
benzyl iodide or alcohol was observed in the crude mixtures.
These results suggest that the decreased Lewis acidity of the
zinc phenoxide catalyst leads to less acidification of the
substrate’s OH when it is complexed to the carbenoid, making
cyclopropanation more prevalent than protonolysis. A similar
When free allylic alcohol 4a was treated with similar
conditions, the undesired O-benzylated cyclopropane 3a was
observed (Table 1, entry 3). This product putatively arises from
protonolysis of the carbenoid’s C−Zn bond by the alcohol (vide
1491
dx.doi.org/10.1021/ol500267w | Org. Lett. 2014, 16, 1490−1493

Organic Letters
Letter
a
Scheme 4. Synthesis of Phenylcyclopropanes from Allylic Alcohols
a
1
b
c
Yields measured by H NMR of crude mixture using Ph3CH as an internal standard. 0.75 M solution in CH2Cl2, added dropwise over 1.5 h.
d
e
Isolated yield, major diastereomer. Isolated yield, both diastereomers combined. Only 1.6 equiv of diazo used.
effect has already been reported for zinc phosphate derived
carbenoids.23b
Scheme 5. Proposed Mechanism for the Zinc-Catalyzed
Cyclopropanation and Side Reactions
Throughout the reported products, the relative configuration
of the introduced phenyl group was found to be dependent on
the alkene’s substitution pattern. For monosubstituted alkenes,
the results show a trans-directing effect from an allylic oxygen
(3d, 5h), while aryl substituents are cis-directing (3h−k). These
effects are reversed or diminished when an alkyl group is trans
to an aryl (3l−m) or cis to an allylic oxygen (3e−g, 5l−n),
suggesting alkyls also exhibit a cis-directing effect. The inversion
of diastereoselectivity between 5l and 5n and the sharp
decrease in dr between 5k and 5l also highlight the cis-directing
effect of the primary or secondary alkyl groups. These
observations are correlated to earlier results using alkyl-
substituted zinc carbenoids.6a,25
As previously reported,12 the proposed mechanism for this
reaction involves the formation of an iodo(phenyl)methylzinc
carbenoid (6, Scheme 5) via double nucleophilic displacement5
between phenyldiazomethane and the iodozinc catalyst XZnI.
This carbenoid can react with the substrate alkene to yield the
desired cyclopropane (pathway A), with another equivalent of
diazo reagent to yield a stilbene molecule (pathway B), or with
a free hydroxyl group (if present) to yield a benzyl ether
(pathway C). All of these reactions regenerate the starting
XZnI catalyst. The two undesired pathways were minimized
respectively by slowly adding the diazo at room temperature
and by reducing the carbenoid’s Lewis acidity. The cyclo-
propanation is believed to proceed through a transition state
similar to the classic Simmons−Smith reaction of halomethyl-
zinc species with alkenes.21 In the cases where the alkene
possesses a vicinal Lewis basic group, precomplexation of the
zinc salt to the substrate is probably involved.
In summary, we have developed a catalytic Simmons−Smith
reaction able to convert allylic ethers and styrenes into the
corresponding phenylcyclopropanes. A modified catalyst even
tolerates free allylic alcohols. Furthermore, this system
theoretically allows the introduction of a chiral anionic ligand
on the catalyst, enabling the potential development of the first
enantioselective Simmons−Smith reaction using catalytic
amounts of both zinc and a chiral mediator. The search for
efficient chiral ligands is being pursued and will be reported in
due course.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures, NMR spectra, and compound
characterization data. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: andre.charette@umontreal.ca.
Notes
The authors declare no competing financial interest.
1492
dx.doi.org/10.1021/ol500267w | Org. Lett. 2014, 16, 1490−1493

Organic Letters
Letter
(20) (a) Charette, A. B.; Brochu, C. J. Am. Chem. Soc. 1995, 117,
11367−11368. (b) Charette, A. B.; Molinaro, C.; Brochu, C. J. Am.
Chem. Soc. 2001, 123, 12160.
(21) Nakamura, M.; Hirai, A.; Nakamura, E. J. Am. Chem. Soc. 2003,
125, 2341.
ACKNOWLEDGMENTS
■
This work was supported by the Natural Science and
Engineering Research Council of Canada (NSERC), the
Canada Research Chair Program, the FRQNT Centre in
(22) The previously reported preforming of the substrate’s sodium
alkoxide with NaH should also be a suitable way to avoid
benzylation.12 However, this procedure suffers from solubility and
reproducibility problems when a calatytic amount of zinc is used.
(23) (a) Charette, A. B.; Francoeur, S.; Martel, J.; Wilb, N. Angew.
Chem., Int. Ed. 2000, 39, 4539. (b) Lacasse, M.; Poulard, C.; Charette,
A. B. J. Am. Chem. Soc. 2005, 127, 12440.
Green Chemistry, and Catalysis and Universite
́
de Montrea
́
l.
E. L. is grateful to NSERC and Universite de Montrea
́
́
l for
postgraduate scholarships. The authors would like to thank
former group members Dr. L.-P. B. Beaulieu, Dr. J. F.
́ ́
Schneider, and Dr. V. N. G. Lindsay of Universite de Montreal
for the synthesis of some allylic alcohols and styrenes.
(24) The anion of butylated hydroxytoluene (BHT) is also a suitable
ligand, giving 80% yield in the conditions depicted in Table 1, entry 7.
(25) Nishimura, J.; Kawabata, N.; Furukawa, J. Tetrahedron 1969, 25,
2647.
REFERENCES
■
(1) Wessjohann, L.; Brandt, W.; Thiemann, T. Chem. Rev. 2003, 103,
1625.
(2) Wong, H. N. C.; Hon, M. Y.; Tse, C. W.; Yip, Y. C.; Tanko, J.;
Hudlicky, T. Chem. Rev. 1989, 89, 165.
(3) Simmons, H. E.; Smith, R. D. J. Am. Chem. Soc. 1958, 5323.
(4) (a) Charette, A. B.; Beauchemin, A. Org. React. 2001, 58, 1−395.
Caution! Diazo compounds are presumed to be highly toxic and
potentially explosive. All manipulations should be carried out in a
hood. Although in numerous preparations we have never observed an
explosion, all pyrolyses and distillations should routinely be carried out
behind a safety shield. Acetic acid should be used to neutralize any
spilled or unwanted material. (b) Furukawa, J.; Kawabata, N. Adv.
Organomet. Chem. 1974, 12, 83. (c) Charette, A. B.; Juteau, H. J. Am.
Chem. Soc. 1994, 116, 2651.
(5) Lebel, H.; Marcoux, J.-F.; Molinaro, C.; Charette, A. B. Chem.
Rev. 2003, 103, 977.
(6) (a) Charette, A. B.; Lemay, J. Angew. Chem., Int. Ed. 1997, 36,
1090. (b) Bull, J.; Charette, A. B. J. Am. Chem. Soc. 2010, 132, 1895.
(c) Wilb, N.; Charette, A. B. Synlett 2002, 1, 176−178.
(7) (a) Beaulieu, L.-P. B.; Zimmer, L. E.; Gagnon, A.; Charette, A. B.
Chem.Eur. J. 2012, 18, 14784. (b) Beaulieu, L.-P. B.; Schneider, J. F.;
Charette, A. B. J. Am. Chem. Soc. 2013, 135, 7819.
(8) (a) Charette, A. B.; Gagnon, A.; Fournier, J.-F. J. Am. Chem. Soc.
2002, 124, 386. (b) Zimmer, L. E.; Charette, A. B. J. Am. Chem. Soc.
2009, 131, 15624.
(9) (a) Emschwiller, G. Compt. Rend. 1929, 188, 1555. (b) Charette,
A. B.; Marcoux, J.-F. J. Am. Chem. Soc. 1996, 118, 4539.
(10) Wittig, G.; Schwarzenbach, K. Angew. Chem. 1959, 652.
(11) (a) Goh, S. H.; Closs, L. E.; Closs, G. L. J. Org. Chem. 1969, 34,
25. (b) Crumrine, D. S.; Haberkamp, T. J.; Sutherl, D. J. J. Org. Chem.
1975, 40, 2274. (c) Marvell, E. N.; Lin, C. J. Am. Chem. Soc. 1978, 100,
877. (d) Altman, L. J.; Kowerski, R. C.; Laungani, D. R. J. Am. Chem.
Soc. 1978, 100, 6174.
(12) Goudreau, S. R.; Charette, A. B. J. Am. Chem. Soc. 2009, 131,
15633.
(13) A similar, metal-free transformation has already been reported,
but it requires high temperatures and long reaction times: Barluenga,
J.; Quinones, N.; Tomas
2012, 2312.
́
-Gamasa, M.; Cabal, M.-P. Eur. J. Org. Chem.
̃
(14) ZnI2 was generated in situ from Et2Zn and I2 to ensure
reproducibility. Commercial ZnI2, which is hygroscopic, can also be
used, but results are dependent on batch quality and storage
conditions.
(15) After the cyclopropanation, the MOM group could be cleaved
using aqueous HCl in refluxing MeOH for 2 h. The yield of isolated
alcohol was 90%. This reaction was carried out on a 0.10 mmol scale.
(16) Verdecchia, M.; Tubaro, C.; Biffis, A. Tetrahedron Lett. 2011, 52,
1136.
(17) Hamaker, C. G.; Mirafzal, G. A.; Woo, L. K. Organometallics
2001, 20, 5171.
(18) Maas, G.; Seitz, J. Tetrahedron. Lett. 2001, 42, 6137.
(19) Hamaker, C. G.; Djukic, J.; Smith, D. A.; Woo, L. K.
Organometallics 2001, 20, 5189.
1493
dx.doi.org/10.1021/ol500267w | Org. Lett. 2014, 16, 1490−1493