Improved protocol for the diastereoselective cyclopropanation of alkenes using geminal dizinc carbenoids: A study on the effect of zinc iodide

Article abstract of DOI:10.1002/ejoc.200400029

A mixture of ZnI₂, EtZnI·2OEt₂ and CHI ₃ produces a gemdizinc carbenoid that is an efficient cyclopropanating reagent. The presence of ZnI₂ allows for shorter reaction times and cleaner reactions, particularly with less reactive substrates. This modification improves the scope of the reaction and it raises important mechanistic issues about this reaction. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejoc.200400029

SHORT COMMUNICATION
Improved Protocol for the Diastereoselective Cyclopropanation of Alkenes
using Geminal Dizinc Carbenoids: A Study on the Effect of Zinc Iodide
[a]
[a]
´
Jean-Francois Fournier and Andre B. Charette*
¸
Keywords: Carbenoids / Carbocycles / Zinc / Alkenes / Cycloaddition
A mixture of ZnI2, EtZnI·2OEt2 and CHI3 produces a gem-
dizinc carbenoid that is an efficient cyclopropanating re-
agent. The presence of ZnI2 allows for shorter reaction times
and cleaner reactions, particularly with less reactive sub-
strates. This modification improves the scope of the reaction
and it raises important mechanistic issues about this reaction.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
cohols, since the competing reaction involving the (diiodo-
methyl)zinc reagent 3 becomes important, leading to iodo-
substituted cyclopropanes.
1,2,3-Substituted cyclopropanes are widely found in
natural products and other biologically active molecules.[1]
Our interest in developing enantioselective methods for
their synthesis using zinc carbenoids[2] prompted us to look
at novel organometallic species that would complement ex-
isting methods.[3] Recently, an approach was developed that
relied on the synthesis and reactivity of novel gem-dizinc
carbenoids 2, which led to a cyclopropylzinc intermediate
4. This intermediate could then be further functionalized
by treatment with a suitable electrophile (Scheme 1).[4] This
approach is highly efficient with 2-butene-1,4-diol deriva-
tives, such as 1. However, lower yields are observed with
other protected or unprotected disubstituted allylic al-
The effect of zinc halides in the Simmons-Smith reaction
with zinc carbenoids has been the subject of several investi-
gations. For example, Denmark has shown that addition of
ZnI2 increases the reaction rate, as well as the enantioselec-
tivities, when the reaction is run in the presence of a chiral
disulfonamide catalyst.[5,6] He concluded that ZnI2 un-
dergoes the Schlenk equilibrium with Zn(CH2I)2 to yield
the more reactive IZnCH2I reagent, but ZnI2 was not in-
volved in the postulated transition structure. A recent
theoretical investigation by Nakamura led to a novel
suggestion about the role of zinc halides in these reac-
tions.[7] DFT calculations led to the finding that the acti-
vation energy of the cyclopropanation of alkenes with
ClZnCH2Cl is significantly lowered in the presence of
ZnCl2. Presumably, the Lewis acid acts as a chloride shuttle
via a five-centered transition structure A.[8] Although the
positive effect of Lewis acids in the Simmons-Smith cyclo-
propanation has been sporadically observed,[9] their exact
role still remains uncertain.
Scheme 1
[a]
´
´
´
Departement de Chimie, Universite de Montreal,
Phillips and co-workers recently reported a DFT investi-
gation on the reactivity of the gem-dizinc carbenoid 2 as a
cyclopropanation reagent.[10] These calculations compared
the geometry of different gem-dizinc carbenoids
(RZnCHIZnRЈ, where R, RЈ ϭ Et, I) and their respective
´
´
P. O. Box 6128, Station Downtown, Montreal, Quebec H3C
3J7, Canada
Fax: (internat.) ϩ 514-343-5900
E-mail: andre.charette@umontreal.ca
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2004, 1401Ϫ1404
DOI: 10.1002/ejoc.200400029
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1401

J.-F. Fournier, A. B. Charette
SHORT COMMUNICATION
transition states in the cyclopropanation of alkenes. Phillips
For the purpose of our studies, we have also prepared the
found that the energy of the transition state is lowered by gem-dizinc reagent 2c from 2 equiv. of EtZnI and 1 equiv.
6.3 kcal/mol by adding 1 equiv. of ZnI2 (transition state C). of CHI3 (Scheme 5). Although we were unable to charac-
He also found the competing transition state B, which is 0.2 terize this species due to its instability, trapping experiments
kcal/mol higher than C.
have shown that two CϪI bonds are exchanged to generate
These calculations, combined with the observation that 2c. This reagent is particularly interesting, since ZnI2 is the
the scope of the reaction of gem-dizinc carbenoids is some- by-product of the cyclopropanation reaction and Schlenk
what limited using our published protocol (without ZnI2),[4] equilibration is not an issue with this reagent upon ZnI2 ad-
prompted us to systematically investigate the effect of added dition.
ZnI2 in these reactions.
Results and Discussion
In our previous communication, we established that the
optimal stoichiometry for generation of the reagent was a
1:1 ratio of Et2Zn and CHI3. Two substrates were chosen
Scheme 5
for the purpose of this study. The benzyl ether 6 reacted
Four protocols in which various amounts of ZnI2 were
added or formed in situ were tested with both substrates
(Figure 1). The reactions with alkene 6 are much more ef-
ficient when the reagent is prepared from ethylzinc iodide
instead of diethylzinc (Protocols A, B vs. C, D). The conver-
sions were not only incomplete with Et2Zn, but also signifi-
cant amounts of the iodocyclopropanes resulting from the
reaction involving reagent 3 were produced, in addition to
extensive decomposition. For comparison purposes, the re-
action was also run using Protocol A, but in the presence
of added diethyl ether (Protocol B), since the addition of a
complexing solvent (which is present during EtZnI forma-
tion in Protocols C and D) may influence the reaction pro-
file. Although the reaction was cleaner using Protocol B,
the conversion was still quite low. Finally, the addition of
extra ZnI2 (Protocol D) led to a very clean reaction that
compared favorably with that of Protocol C, leading to the
complete consumption of the alkene within 25 min, instead
of 45 min. However, although the reaction using Protocol
D reaches completion more quickly than using Protocol C,
a significant difference in conversion is only observed after
about 15 min. It is thus still not clear whether ZnI2 allows
under our optimal protocol to give the 1,2,3-substituted
cyclopropane 7a in only 53 % yield (Scheme 2).[11] Alcohol
8 was also included in this study, although it led to the
desired cyclopropane in good yield (98 %, Scheme 3).
Scheme 2
Scheme 3
Given the stoichiometry used in these reactions, a more
accurate representation of 2 is probably the oligomeric form
2b, in which one end of the oligomer consists of an ethyl
group and the other of a (diiodomethyl)zinc carbenoid
(Scheme 4). This reagent should not lead to the formation
of any ZnI2 upon cyclopropanation, but to an alkylzinc iod-
ide.
Figure 1. Comparison of the reaction conditions for the cyclopro-
panation of the benzyl ether 6; all the reactions were run in CH2Cl2
at 0 °C, and each data point is an average of 3 reactions; Protocol
A: Et2Zn (3 equiv.) ϩ CHI3 (3 equiv.); Protocol B: Et2Zn (3 equiv.)
ϩ CHI3 (3 equiv.) ϩ Et2O (6 equiv.); Protocol C: EtZnI (3 equiv.),
CHI3 (1.5 equiv.) ϩ Et2O (6 equiv.); Protocol D: EtZnI (3 equiv.)
ϩ ZnI2 (1.5 equiv.) ϩ CHI3 (1.5 equiv.) ϩ Et2O (9 equiv.); see
supporting information for further details
Scheme 4
1402
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1401Ϫ1404

Improved Protocol for the Diastereoselective Cyclopropanation of Alkenes
SHORT COMMUNICATION
for a lower transition state energy (as suggested by Phillips), Conclusions
or if it simply stabilizes the reagent throughout the reaction
In summary, a new and improved protocol for the cyclo-
propanation of alkenes using a gem-dizinc carbenoid has
been developed. This protocol, involving carbenoid 2c and
ZnI2, has been shown to both have a broader scope of sub-
strates, as well as achieving 100 % conversion more quickly
than our previous protocol using carbenoid 2b. However,
whether this enhancement is due to reagent activation (as
proposed by Phillips) or by reagent stabilization could not
be demonstrated unequivocally. Further studies to exploit
the full synthetic potential of these reagents will be reported
in due course.
(either through slower reagent formation or through re-
agent stabilization).[12]
The same beneficial effect of ZnI2 is also observed in the
cyclopropanation of the allylic alcohol 8 (Figure 2). It is
interesting to note that in this case, the reagent combination
EtZnI ϩ CHI3 (Protocol G) is not more effective than those
involving Et2Zn and CHI3 (Protocols E and F). However,
Protocol H, which uses 2 equiv. of added ZnI2, is the most
efficient; the reaction is complete within 25 min. It is im-
portant to mention that the presence of less than 1 equiv.
of ZnI2 is not sufficient to achieve a faster consumption of
starting material. This observation, and the fact that Proto-
col G is not optimal in this case, may be accounted for by
the fact that the ZnI2 formed initially is trapped by the basic
zinc alkoxide, and therefore is unavailable for the activation
or stabilization of the reagent. The true activating or stabil-
izing effect takes place only when more than 1 equiv. of
ZnI2 is present in solution.
Experimental Section
Et2Zn (291 µL, 2.838 mmol, 4.5 equiv.) was slowly added to a solu-
tion of iodine (960 mg, 3.784 mmol, 6.0 equiv.) and diethyl ether
(589 µL, 5.676 mmol, 9.0 equiv.) in dry CH2Cl2 (3 mL) at 0 °C. The
ice bath was removed and the solution stirred for 10 min before
recooling to 0 °C. The benzyl ether (0.631 mmol, 1.0 equiv.) in
CH2Cl2 (1 mL) was then added by cannula to this clear colorless
solution, followed by a solution of CHI3 (372 mg, 0.946 mmol,
1.5 equiv.) in CH2Cl2 (6 mL) and of dry air (100 µL). After the
appropriate reaction time (15Ϫ45 min), the cyclopropylzinc was
quenched with either saturated aqueous NH4Cl, 1  DCl in D2O,
or (after cooling to Ϫ78 °C) a solution of iodine (400 mg,
1.577 mmol, 2.5 equiv.) in THF (3 mL). After 1 min, saturated
aqueous Na2SO3 (5 mL) was poured in. The aqueous layer was
extracted with diethyl ether (3 ϫ 4 mL), the combined organic lay-
ers washed successively with saturated aqueous Na2SO3 (4 mL),
saturated aqueous NaHCO3 (4 mL) and brine (4 mL), dried with
MgSO4 and concentrated under reduced pressure. The residue was
purified by flash chromatography to afford the desired cyclopro-
pane.
Figure 2. Comparison of the reaction conditions for the cyclopro-
panation of the alcohol 8; all the reactions were run in CH2Cl2 at
0 °C, and each data point is an average of 3 reactions; Protocol E:
Et2Zn (4.5 equiv.) ϩ CHI3 (4 equiv.); Protocol F: Et2Zn (4.5 equiv.)
ϩ CHI3 (4 equiv.) ϩ Et2O (9 equiv.); Protocol G: EtZnI (5 equiv.),
CHI3 (2 equiv.) ϩ Et2O (10 equiv.); Protocol H: EtZnI (5 equiv.)
ϩ ZnI2 (2 equiv.) ϩ CHI3 (2 equiv.) ϩ Et2O (14 equiv.); see sup-
porting information for further details
Acknowledgments
This work was supported by the E. W. R. Steacie Fund, NSERC,
Merck Frosst Canada & Co., Boehringer Ingelheim (Canada) Ltd.,
Using this new and improved protocol, we were able to
cyclopropanate (E)-alkenes diastereoselectively with differ-
ent substitution patterns, without undesired iodocyclopro-
panation from carbenoid 3 (Table 1).
´
´
and the Universite de Montreal. J.-F. F. is grateful to NSERC (PGF
B) and F.C.A.R. (B2) for postgraduate fellowships.
[1] [1a]
L. A. Wessjohann, W. Brandt, T. Thiemann, Chem. Rev.
[1b]
2003, 103, 1625Ϫ1647.
Small Ring Compounds in Organic
Table 1. Cyclopropanation of various alkenes
Synthesis VI, vol. 207 (Ed.: A. de Meijere), Springer, Berlin,
2000. [1c] Methods Org. Chem. (Houben-Weyl), vol. E 17c (Ed.:
A. de Meijere), Thieme, Stuttgart, 1997.
[2]
[3]
A. B. Charette, J. Lemay, Angew. Chem. 1997, 109, 1163Ϫ1165;
Angew. Chem. Int. Ed. Engl. 1997, 36, 1090Ϫ1092.
For a recent review see: [3a] H. Lebel, J.-F. Marcoux, C. Molin-
[3b]
aro, A. B. Charette, Chem. Rev. 2003, 103, 977Ϫ1050.
A.
B. Charette, A. Beauchemin, Org. React. 2001, 58, 1Ϫ415.
R1
R2
Eϩ
11/12[a]
Ͼ 95:5
96:4
96:4
Yield [%][b]
[4]
Entry
A. B. Charette, A. Gagnon, J.-F. Fournier, J. Am. Chem. Soc.
2002, 124, 386Ϫ387.
[5] [5a]
1
2
3
4
5
Pr
H
H
H
H
H
D2O
H2O
I2
H2O
I2
85
76
76
59
74
S. E. Denmark, B. L. Christenson, S. P. O’Connor, Tetra-
[5b]
Me
Me
H
hedron Lett. 1995, 36, 2219Ϫ2222.
O’Connor, J. Org. Chem. 1997, 62, 584Ϫ594.
S. E. Denmark, S. P.
[5c]
S. E.
90:10
90:10
Denmark, S. P. O’Connor, Angew. Chem. 1998, 110,
H
1162Ϫ1165; Angew. Chem. Int. Ed. 1998, 37, 1149Ϫ1151.
[6] [6a]
S. E. Denmark, B. L. Christenson, D. M. Coe, S. P.
[a]
1
[6b]
Ratio as determined by H NMR spectroscopy. [b] Isolated yield
O’Connor, Tetrahedron Lett. 1995, 36, 2215Ϫ2218.
Denmark, S. P. O’Connor, J. Org. Chem. 1997, 62, 3390Ϫ3401.
S. E.
of 11.
Eur. J. Org. Chem. 2004, 1401Ϫ1404
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1403

J.-F. Fournier, A. B. Charette
SHORT COMMUNICATION
[7] [7a]
[11]
[12]
E. Nakamura, A. Hirai, M. Nakamura, J. Am. Chem. Soc.
The relative stereochemistry for the electrophile insertion is
that shown in Schemes 2 and 3; it was established by deuter-
ation of the cyclopropylzinc reagent, which led to 7b and 9b ex-
clusively.
Detailed kinetic studies were inconclusive because of the het-
erogeneous nature of the reaction mixture, and the difficulty in
controlling the exact initial concentration of 2c (carbenoids of
type 2 undergo decomposition and are subject to various equi-
libria).
[7b]
1998, 120, 5844Ϫ5845.
A. Hirai, M. Nakamura, E. Naka-
[7c]
mura, Chem. Lett. 1998, 9, 927Ϫ928.
M. Nakamura, A.
Hirai, E. Nakamura, J. Am. Chem. Soc. 2003, 125, 2341Ϫ2350.
For a novel reagent based on this concept see: A. B. Charette,
A. Beauchemin, S. Francoeur, J. Am. Chem. Soc. 2001, 123,
8139Ϫ8140.
[8]
[9] [9a]
A. B. Charette, C. Molinaro, C. Brochu, J. Am. Chem. Soc.
2001, 123, 12160Ϫ12167. [9b] A. B. Charette, C. Brochu, J. Am.
Chem. Soc. 1995, 117, 11367Ϫ11368.
Received January 15, 2004
[10]
C. Zhao, D. Wang, D. L. Phillips, J. Am. Chem. Soc. 2002,
124, 12903Ϫ12914.
1404
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 1401Ϫ1404