Stability, reactivity, solution, and solid-state structure of halomethylzinc alkoxides

Article abstract of DOI:10.1021/ja010837+

In this paper, we report our findings regarding the development of a Lewis acid-catalyzed cyclopropanation of allylic alcohols with bis(iodomethyl)zinc. Iodomethylzinc alkoxides can be formed by treatment of an alcohol with bis(iodomethyl)zinc. These species are not prone to undergo cyclopropanation at low temperature but the addition of a Lewis acid in catalytic amounts induces the cyclopropanation reaction. Using this procedure, we demonstrated that the Lewis acid-catalyzed pathway significantly overwhelms the uncatalyzed one. This paper describes fundamental issues regarding the preparation and stability of halomethyl zinc alkoxides in solution as well as their aggregation state in solution and solid-state structures. Furthermore, the competition reaction between the inter- vs intramolecular cyclopropanation will be studied. Finally, we will discuss the possible activation pathways to explain the Lewis acid activation of halomethylzinc alkoxides. These findings provided new insights on the reactivity of ROZnCH₂I and established the groundwork for the elaboration of an enantioselective version of the reaction.

Full text of DOI:10.1021/ja010837+

12160
J. Am. Chem. Soc. 2001, 123, 12160-12167
Stability, Reactivity, Solution, and Solid-State Structure of
Halomethylzinc Alkoxides
Andre´ B. Charette,* Carmela Molinaro, and Christian Brochu
Contribution from the De´partement de Chimie, UniVersite´ de Montre´al, P.O. 6128, Station Downtown,
Montre´al, Que´bec, Canada H3C 3J7
ReceiVed March 30, 2001
Abstract: In this paper, we report our findings regarding the development of a Lewis acid-catalyzed
cyclopropanation of allylic alcohols with bis(iodomethyl)zinc. Iodomethylzinc alkoxides can be formed by
treatment of an alcohol with bis(iodomethyl)zinc. These species are not prone to undergo cyclopropanation at
low temperature but the addition of a Lewis acid in catalytic amounts induces the cyclopropanation reaction.
Using this procedure, we demonstrated that the Lewis acid-catalyzed pathway significantly overwhelms the
uncatalyzed one. This paper describes fundamental issues regarding the preparation and stability of halomethyl
zinc alkoxides in solution as well as their aggregation state in solution and solid-state structures. Furthermore,
the competition reaction between the inter- vs intramolecular cyclopropanation will be studied. Finally, we
will discuss the possible activation pathways to explain the Lewis acid activation of halomethylzinc alkoxides.
These findings provided new insights on the reactivity of ROZnCH2I and established the groundwork for the
elaboration of an enantioselective version of the reaction.
Introduction
1).4 Although these species are potential intermediates in most
cyclopropanation reactions involving allylic alcohols and Zn-
(CH2I)2, very little is known about their rate of formation, their
stabilities, and their reactivities depending upon the nature of
the alcohol used. This paper will also attempt to differentiate
between the possible modes of activation shown in Scheme 1
(B vs B′ vs B′′).
The addition of Lewis acids is a fairly well-established
strategy to accelerate several organic transformations. The rate
of many reactions, such as the Mukaiyama aldol, the Diels-
Alder, Mannich, and ene reactions, is significantly increased
by the addition of the appropriate Lewis acid.1 The effect of
Lewis acids on the rate of cyclopropanation with haloalkylzinc
reagents is a much less understood phenomenon. The first report
of the accelerating effect of Lewis acids on the cyclopropanation
of olefins dates back to the early 80’s when Friederich noticed
that the addition of as little as 4 mol % of titanium tetrachloride
to a mixture of zinc/copper couple and diiodomethane consider-
ably improved the rate of the reaction.2 The effect of Lewis
acids, namely ZnI2, has also been discussed in detail in the
cyclopropanation of ethylzinc allyloxides with Zn(CH2I)2.3 In
this paper, we report our full account on a different strategy
that relies on the Lewis acid activation of a ROZnCH2I species
as well as an in-depth study on the stability and reactivity of
iodomethylzinc alkoxides. Our strategy was based on the
observation that iodomethylzinc alkoxides are quite poor
cyclopropanating reagents that could be activated upon Lewis
acid addition. These intermediates could be prepared by treating
any alcohol with bis(iodomethyl)zinc. It was observed that these
resulting iodomethylzinc alkoxides are quite poor cyclopro-
panating reagents and our approach was based on the novel idea
that a chiral Lewis acid could trigger the cyclopropanation of
the otherwise unreactive iodomethylzinc alkoxide (A, Scheme
Results and Discussion
Preliminary Results: Iodomethylzinc Alkoxide Formation.
NMR/Proof of Principle. The synthesis, stability, and solution
state structural studies of iodomethylzinc alkoxides were initially
tackled. The first assumption in the mechanism shown in
Scheme 1 was that an alcohol (allylic or otherwise) would be
readily converted into a relatively unreactive iodomethylzinc
alkoxide upon treatment with 1 equiv of bis(iodomethyl)zinc5,6
at low temperature.4 To test this hypothesis, cinnamyl alcohol
and phenethyl alcohol were each treated with 1 equiv of a
preformed suspension of Zn(CH2I)2 in CD2Cl2 at -20 °C.7 The
corresponding iodomethylzinc alkoxide was formed quantita-
tively as evidenced by the appearance of singlets at 1.4 (1H
NMR) and -27 ppm (13C NMR). These chemical shifts are
consistent with those of previously reported ZnCH2I signals.6,8
Also, the NMR spectra indicated that the deprotonation reaction
of the alcohol was rapid and effective as suggested by the
1
quantitative formation of CH3I (2.1 and -27 ppm by H and
13C NMR, respectively). Moreover, the deprotonation reaction
was also monitored by in situ infrared experiments, which
(1) For selected reviews see: (a) Lewis Acids in Organic Synthesis;
Yamamoto, H., Ed.; Wiley-VCH Inc.; Weinheim, Germany, 2000; Vols. I
and II. (b) Dias, L. C. Curr. Org. Chem. 2000, 4, 305-342. (c) Sibi, M. P.;
Liu, M. Curr. Org. Chem. 2001, 5, 719-755. (d) Fringuelli, F.; Piermatti,
O.; Pizzo, F.; Vaccaro, L. Eur. J. Org. Chem. 2001, 439-455.
(2) Friedrich, E. C.; Lunetta, S. E.; Lewis, E. J. J. Org. Chem. 1989, 54,
2388-2390.
(3) (a) Denmark, S. E.; Christenson, B. L.; Coe, D. M.; O’Connor, S. P.
Tetrahedron Lett. 1995, 36, 2215-2218. (b) Denmark, S. E.; Christenson,
B. L.; O’Connor, S. P.; Noriaki, M. Pure Appl. Chem. 1996, 68, 23-27.
(4) (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, 12168-12175.
(5) Zn(CH2I)2 was generated from 1 equiv of diethylzinc and 2 equiv of
diiodomethane according to Furukawa’s procedure: (a) Furukawa, J.;
Kawabata, N.; Nishimura, J. Tetrahedron Lett. 1966, 3353-3354. (b)
Furukawa, J.; Kawabata, N.; Nishimura, J. Tetrahedron 1968, 24, 53-58.
(c) Nishimura, J.; Furukawa, J.; Kawabata, N.; Kitayama, M. Tetrahedron
1971, 27, 1799-1806.
10.1021/ja010837+ CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/07/2001

Halomethylzinc Alkoxides
J. Am. Chem. Soc., Vol. 123, No. 49, 2001 12161
1
Scheme 1
after 30 min a very clean H NMR of the product is observed
and sharp signals are apparent; however, as time progresses
broader signals are observed which could be a consequence of
aggregate formation in solution.7 More importantly, however,
is that the iodomethylzinc alkoxides are surprisingly stable as
evidenced by the presence of ZnCH2I signal (even if broadened),
after one week at -20 °C.7 This observation was further
confirmed by 13C NMR of the same sample taken after 24 h
and after one week. Finally we can conclude that halomethylzinc
alkoxides are relatively facile species to generate (from an
alcohol and Zn(CH2I)2) and they display a good stability as they
are still present even after one week at -20 °C as confirmed
by 1H and 13C NMR. However, if the sample was left at room
temperature for several hours many additional signals appeared
which result from a temperature-dependent decomposition of
the sample. The structures of some of the decomposition
products are shown in eq 1. These compounds were identified
by GC-MS and NMR and by comparison with authentic sample.
provided additional evidence of MeI formation within the first
5 min of the reaction.7 Ethyl iodide is also produced by the
alkyl exchange when 1 equiv of diethylzinc and 2 equiv of
diiodomethane are mixed.5 Furthermore, neither the disappear-
ance of the alkene protons nor the appearance of cyclopropyl
C-H’s were observed when the iodomethylzinc alkoxide of
cinnamyl alcohol was left for several hours at low temperature.9
Low-temperature (-20 °C) 1H NMR monitoring of the
species as a function of time for phenethyl alcohol shows that
In contrast, monitoring of (E)-PhCHdCHCH2OZnCH2I in-
dicated that the carbenoid was not stable for a long period of
time at -20 °C and signals corresponding to the cyclopropanated
adduct appeared within 24 h.7 Two equally valuable postulates
therefore can be put forward to explain this behavior. First, the
cyclopropanation may be catalyzed by ZnI2 resulting from the
slow decomposition of the iodomethylzinc alkoxide.4,7 Alter-
natively, it can result from the generation of Zn(CH2I)2 from
the Schlenk equilibration of alkylzinc alkoxides10 as shown in
eq 2.
(6) For NMR and X-ray structures of this reagent as well as theoretical
studies see: (a) Denmark, S. E.; Edwards, J. P.; Wilson, S. R. J. Am. Chem.
Soc. 1991, 113, 723-725. (b) Denmark, S. E.; Edwards, J. P.; Wilson, S.
R. J. Am. Chem. Soc. 1992, 114, 2592-2602. (c) Charette, A. B.; Marcoux,
J.-F. J. Am. Chem. Soc. 1996, 118, 4539-4549. (d) Charette, A. B.;
Marcoux, J.-F.; Be´langer-Garie´py, F. J. Am. Chem. Soc. 1996, 118, 6792-
6793. (e) Dargel, T. K.; Koch, W. J. Chem. Soc., Perkin Trans. 2 1996,
877-881. (f) Bernardi, F.; Bottoni, A.; Miscione, G. P. J. Am. Chem. Soc.
1997, 119, 12300-12305. (g) Hirai, A.; Nakamura, M.; Nakamura, E. Chem.
Lett. 1998, 927-928. (h) Nakamura, E.; Hirai, A.; Nakamura, M. J. Am.
Chem. Soc. 1998, 120, 5844-5845. (i) Charette, A. B.; Marcoux, J.-F.;
Molinaro, C.; Beauchemin, A.; Brochu, C.; Isabel, EÄ . J. Am. Chem. Soc.
2000, 122, 4508-4509. (j) Hermann, H.; Lohrenz, J. C. W.; Ku¨hn, A.;
Boche, G. Tetrahedron 2000, 56, 4109-4115. (k) Bernardi, F.; Bottoni,
A.; Miscione, G. P. Organometallics 2000, 19, 5529-5532. (l) Boche, G.;
Lohrenz, J. C. W. Chem. ReV. 2001, 101, 697-756.
Aggregation State in Solution. To determine the degree of
association of halomethylzinc alkoxides in solution, simple
molecular weight measurements were determined by the melting
point variation of benzene solutions. This approach was favored
in our case since as shown earlier, halomethylzinc alkoxides
tend to decompose when warmed. Noltes, Coates, and Allen
have used various methods successfully to determine the degree
of association of zinc alkoxides in solution.11 Several halomethyl
zinc alkoxides were prepared and the degree of association by
(7) See Supporting Information for further information.
(8) The appearance of a signal at 1.36 ppm corresponds to the formation
of small amounts of bis(iodomethyl)zinc and is consistent with the following
equilibrium reactions that have been reported in the literature: 2RZnOR′
T Zn(OR′)2 + ZnR2 and for X-rays consistent with tetrameric alkylzinc
alkoxides 2[ROZnR]4 f ZnR2 + [(RO)8Zn7R6]: (a) Coates, G. E.; Ridley,
D. J. Chem. Soc. 1965, 1870-1877. (b) Allen, G.; Bruce, J. M.; Farrren,
D. W.; Hutchinson, F. G. J. Chem. Soc. B 1966, 799-803. (c) Shearer, H.
M. M.; Spencer, C. B. Chem. Commun. 1966, 194. (d) Matsui, Y.; Kamiya,
K.; Nishikawa, M. Bull. Chem. Soc. Jpn. 1966, 39, 1828. (e) Noltes, J. G.;
Boersma, J. J. Organomet. Chem. 1968, 12, 425-431. (f) Eisenhuth, W.
H.; Van Wazer, J. R. J. Am. Chem. Soc. 1968, 90, 5397-5400. (g) Adamson,
G. W.; Shearer, H. M. M.; Spencer, C. B. Acta Crystallogr. Suppl. 1968,
21, A135. (h) Ziegler, M. L.; Weiss, J. Angew. Chem., Int. Ed. Engl. 1970,
9, 905-906. (i) Ishimori, M.; Hagiwara, T.; Tsuruta, T.; Kai, Y. Bull. Chem.
Soc. Jpn. 1976, 49, 1165-1166. (j) Shearer, H. M. M.; Spencer, C. B. Acta
Crystallogr. 1980, B36, 2046-2050. (k) Boersma, J. Zinc and Cadmium.
In ComprehensiVe Organometallic Chemistry; Wilkinson, G., Stone, F. G.
A., Abel, E. W., Eds.; Permagon Press: Oxford, 1982; Vol. II, pp 823-
862. (l) Olmstead, M. M.; Power, P. P.; Shoner, S. C. J. Am. Chem. Soc.
1991, 113, 3379-3385. (m) O’Brien, P. Cadmium and Zinc. In Compre-
hensiVe Organometallic Chemistry II; Abel, E. W., Stone, F. G. A.,
Wilkinson, G. Eds.; Elsevier Science Ltd: Oxford, 1995; Vol. III, pp 175-
206.
(10) Inoue, S.; Kobayashi, M.; Tozuka, T. J. Organomet. Chem. 1974,
81, 17-21.
(11) (a) Coates, G. E.; Ridley, D. J. Chem. Soc. 1965, 1870-1877. (b)
Allen, G.; Bruce, J. M.; Hutchinson, F. G. J. Chem. Soc. 1965, 5476-
5481. (c) Allen, G.; Bruce, J. M.; Hutchinson, F. G. J. Chem. Soc. 1965,
54-76. (d) Coates, G. E.; Lauder, A. J. Chem. Soc. A 1966, 264-267. (e)
Coates, G. E.; Ridley, D. J. Chem. Soc. A 1966, 1064-1069. (f) Boersma,
J.; Noltes, J. G. Tetrahedron Lett. 1966, 14, 1521-1525. (g) Bruce, J. M.;
Cutsforth, B. C.; Hutchinson, F. G.; Rabagliati, F. M.; Reed, D. R. J. Chem.
Soc. B 1966, 1020-1024. (h) Coates, G. E.; Roberts, P. D. J. Chem. Soc.
A 1967, 1233-1234. (i) Boersma, J.; Noltes, J. G. J. Organomet. Chem.
1967, 8, 551-553. (j) Boersma, J.; Noltes, J. G. J. Organomet. Chem. 1968,
13, 291-299. (k) Noltes, J. G.; Boersma, J. J. Organomet. Chem. 1968,
12, 425-431. (l) Boersma, J.; Spek, A. L.; Noltes, J. G. J. Organomet.
Chem. 1974, 81, 7-15.
(9) This last observation was also confirmed through measurement of
the amount of cyclopropanation product upon quenching the reaction mixture
at low temperature; see the Supporting Information.

12162 J. Am. Chem. Soc., Vol. 123, No. 49, 2001
Table 1. Molecular Weight Determinationa
Charette et al.
Figure 1. ORTEP drawing of iodomethylzinc alkoxide derived from
4-methoxybenzyl alcohol. Ellipsoids are drawn at the 30% probability
level.
a Samples were prepared in CH2Cl2 for 0.5 h, left under high vacuum
for 2 h, dissolved in benzene, and cryoscopy measurements were made.
In all cases the alkoxides were completely soluble in benzene.
b Concentration of alkoxide. c N, degree of association. d Measured after
48 h.
molecular weight determination are presented in Table 1. Quite
unexpectedly based on litterature assumptions,6h the molecular
weight determination of the iodo- or chloro12-methylzinc
alkoxides derived from phenethyl alcohol or 2-propanol at
different concentrations indicated these species are mostly
monomeric in benzene (entries 1-14, Table 1). This is also
the case for iodomethylzinc butoxide (entry 7, Table 1).
Although molecular weight determination of the species result-
ing from the reaction between simple alcohols and dialkyl zinc
has led to the misconception that most alkoxides are tetrameric
(cryoscopically in benzene), zinc alkoxides also exist in lower
aggregation states such as dimers or trimers in benzene solution
as a result of intermolecular Zn-O and/or intramolecular Zn-L
bridging (L being a ligand for zinc).11 In the case of halom-
ethylzinc alkoxides, it is conceivable that benzene (solvent), the
halogen (I or Cl), or the phenyl group in PhCH2CH2OZnCH2-
Hal (in CH2Cl2) could effectively solvate or act as a ligand for
the more acidic zinc centers thus allowing a stabilization to the
monomeric species. Even after 48 h (entries 6 and 12, Table 1)
these species appear to be mainly monomeric. Ethylzinc
2-phenylethoxide measurements also gave a monomeric struc-
ture where intramolecular interactions between the phenyl group
and the zinc center may lead to the stabilization of the
monomeric species.13 Isopropyl ethylzinc gave an aggregation
state averaging the tetrameric form as previously reported.11a
Solid-State Structure of ROZnCH2I and ROZnCH2Cl.
Having demonstrated that halomethylzinc species are monomeric
in benzene and stable for at least one week at low temperature,
Figure 2. ORTEP drawing of chloromethylzinc alkoxide derived from
4-methoxybenzyl alcohol. Ellipsoids are drawn at the 30% probability
level.
the establishment of their solid-state structure became manda-
tory. After numerous attempts, we successfully obtained suitable
crystals by slow diffusion of hexanes in a CH2Cl2 solution of
the halomethylzinc alkoxide at low temperature. Two X-ray
crystal structures of halomethylzinc alkoxides produced from
4-methoxybenzyl alcohol and Zn(CH2I)214 as well as from Zn-
(CH2Cl)2 were obtained (Figures 1 and 2). Selected bond lengths
and angles are given in Table 2. These structures are reminiscent
of those of methylzinc methoxide8c,g,j and methylzinc tert-
butoxide,8d which also crystallized as tetramers having Zn and
O on alternate corners of a slightly distorted cube. The distortion
angles of Zn-O-Zn are greater than 90°. The average bond
lengths (Å) are the following: C-O (1.431-1.457), O-Zn
(2.022-2.137), Zn-C (1.940-1.965), C-I (2.159-2.182),
(13) π-Chelation between Zn and a phenyl group or CdC double bond:
(a) St. Denis, J.; Oliver, J. P. J. Organomet. Chem. 1972, 44, C32-C36.
(b) St. Denis, J.; Oliver, J. P. J. Organomet. Chem. 1974, 71, 315-323. (c)
Albright, M. J.; St. Denis, J. N.; Oliver, J. P. J. Organomet. Chem. 1977,
125, 1-8. (d) Beruben, D.; Marek, I.; Normant, J. F.; Platzer, N. J. Org.
Chem. 1995, 60, 2488-2501. (e) Cossy, J.; Blanchard, N.; Meyer, C. J.
Org. Chem. 1998, 63, 5728-5729.
(14) Two different crystals were obtained for 4-MeO-PhCH2OZnCH2I.
The X-ray for the second crystal is isostructural to the ROZnCH2Cl
analogue. The X-ray crystal structure of MeOZnCH2I prepared from the
photoinduced method was also obtained but the structure will be reported
separately: Charette, A. B.; Beauchemin, A.; Marcoux, J.-F.; Francoeur,
S.; Enright, G.; Garie´py-Be´langer, F. Unpublished results.
(12) Chloromethylzinc alkoxide of phenethyl alcohol was generated from
1 equiv of Zn(CH2Cl)2 and 1 equiv of the alcohol. See Supporting
Information for 1H and 13C NMR spectra for chloromethylzinc alkoxide.
For the preparation of Zn(CH2Cl)2 see: Denmark, S. E.; Edwards, J. P. J.
Org. Chem. 1991, 56, 6974-6981.

Halomethylzinc Alkoxides
J. Am. Chem. Soc., Vol. 123, No. 49, 2001 12163
Table 2. Selected Bond Lengths (Å) and Angles (deg) for
Iodomethyl Zinc 4-Methoxybenzyloxy (ROZnCH2I) and
Chloromethyl Zinc 4-Methoxybenzyloxy (ROZnCH2Cl)
Et2AlCl, etc.) are usually much more effective at catalyzing the
reaction whereas weaker Lewis acids (Ti(Oi-Pr)4, B(OMe)3, etc.)
are less effective.18 For example, the addition of as little as 1
mol % of TiCl4 is sufficient to promote the cyclopropanation
reaction of cinnamyl alcohol providing a conversion of 65% at
0 °C. It also appears that Lewis acids having nucleophilic
counterions are more effective than those that contain nonnu-
cleophilic ones.
Lewis Acid-Catalyzed Cyclopropanation and Substrate
Generality. Under standard conditions, a minimum of 2 equiv
of Zn(CH2I)2 is required to efficiently cyclopropanate an allylic
alcohol. Typically, the first equivalent is used to generate the
halomethylzinc alkoxide (see Scheme 1) and the second will
act as the methylene transfer reagent to generate the corre-
sponding cyclopropane. Our new protocol implies the replace-
ment of the extra equiValent of Zn(CH2I)2 by a catalytic amount
of a Lewis acid rendering the overall process more efficient.
The efficiency of the cyclopropanation process with ROZnCH2I/
Lewis acid was compared to that involving 2 equiv of
Zn(CH2I)2. The relative rate of the cyclopropanation reaction
of cinnamyl alcohol was qualitatively assessed by using TiCl4.
A slightly lower reaction rate was observed in the Lewis acid-
catalyzed version, but considering that there is only one active
CH2I group compared to three in the uncatalyzed reaction, the
rate of the former version nicely competes with the rate of the
latter.
ROZnCH2I
bond lengths
ROZnCH2Cl
Zn(1)-C(1)
I(1)-C(1)
1.962(5)
2.159(5)
1.955(5)
Cl(1)-C(1)
Zn(1)-O(1)
Zn(1)-O(2)
Zn(1)-O(3)
Zn(1)-Zn(4)
C(11)-O(1)
1.809(5)
2.061(3)
2.052(3)
2.086(3)
3.0705(8)
1.450(5)
2.042(3)
2.123(3)
2.052(3)
2.9917(9)
1.450(5)
bond angles
Zn(1)-C(1)-I(1)
Zn(1)-C(1)-Cl(1)
C(1)-Zn(1)-O(1)
C(1)-Zn(1)-O(2)
C(1)-Zn(1)-O(3)
C(11)-O(1)-Zn(2)
C(11)-O(1)-Zn(1)
O(2)-Zn(1)-O(1)
O(2)-Zn(1)-O(3)
O(1)-Zn(1)-O(3)
Zn(2)-O(1)-Zn(3)
Zn(1)-O(1)-Zn(3)
Zn(2)-O(1)-Zn(1)
116.8(3)
111.0(3)
142.08(19)
115.9(2)
129.28(18)
118.4(3)
129.28(18)
133.87(19)
127.01(19)
122.4(3)
120.1(3)
120.1(3)
79.77(12)
87.71(13)
83.02(12)
93.28(12)
95.53(12)
100.14(13)
83.26(13)
82.82(12)
82.73(12)
97.53(13)
96.62(12)
96.54(13)
C-Cl (1.727-1.809), and Zn-Zn (2.9917-3.1118). The aver-
age bond angles (deg) are the following: C-O-Zn (117.0-
126.5), O-Zn-C (112.7-148.9), Zn-C-I (105.4-116.8),
Zn-C-Cl (108.0-118.3), O-Zn-O (79.77-87.71), and Zn-
O-Zn (91.15-100.14).
The difference between the solution and solid-state structure
of halomethylzinc alkoxides is quite stricking but it is clear that
in solution either the solvent (benzene) or the presence of a
basic donor group on the reagent may help in stabilizing the
monomeric form.
These reaction conditions were tested on several allylic
alcohols and the results are presented in Table 3. A variety of
substitution patterns are tolerated with this protocol as exempli-
fied by the effective cyclopropanation of trans (entries 1-3,
Table 3), cis (entries 4-6, Table 3), as well as trisubstituted
allylic alcohols (entries 7-9, Table 3). Also, interestingly a
chiral, racemic secondary allylic alcohol (entry 10, Table 3)
gave a 3:1 (syn:anti) ratio which is similar to that obtained when
the reaction is done with 2 equiv of Zn(CH2I)2.19
Intra- vs Intermolecular Methylene Transfer: Competi-
tive Experiments. In an attempt to improve our understanding
of the mechanism by which the methylene transfer of iodom-
ethylzinc species occurs, several control experiments were
designed to obtain additional information on this step. First, a
key issue was to determine whether this transfer was occurring
through an inter- or an intramolecular process (or both) in the
case of an allylic alcohol. In the first experiment, 1 equiv of
(E)-PhCHdCHCH2OZnCH2I and 1 equiv of (E)-PhCH2CH2-
CHdCHCH2OZnCD2I were mixed and 5 mol % of TiCl4 (or
15 mol % of BF3‚OEt2) was added to catalyze the cyclopro-
panation reaction (Scheme 2). Formation of cyclopropanes B
and D in Scheme 2 indicates that complete scrambling of the
“CH2” and “CD2” units of the corresponding cyclopropanes had
Lewis Acid Activation of Iodomethylzinc Alkoxides (Scheme
1).7 The second assumption in the mechanism shown in Scheme
1 was that the addition of a Lewis acid to a relatively unreactive
halomethylzinc alkoxide would generate a reactive intermediate,
sufficiently increase the electrophilic character of the carbenoid,
and lead to a cyclopropanation reaction (eq 3). This last
observation was confirmed by measuring the amount of cyclo-
propanated product from cinnamyl alcohol upon quenching the
reaction mixture at low temperature. As pointed out earlier, less
than 23% of cyclopropane adduct was observed if the iodom-
ethylzinc alkoxide was left for 6 h at 0 °C. However, we have
also noticed that the deliberate introduction of trace amounts
of oxygen triggered a non-Lewis acid-catalyzed cyclopropana-
tion reaction to a modest extent at 0 °C.15 This observation is
consistent with Miyano’s report that unambiguously showed that
the presence of oxygen could accelerate the cyclopropanation
of unfunctionalized olefins.16 In contrast, several achiral Lewis
acids, added in catalytic amounts to the preformed iodometh-
ylzinc alkoxide, triggered the cyclopropanation reaction.17 In
this system, a general trend follows the strength of the Lewis
acid. Stronger Lewis acids (such as TiCl4, BBr3, SiCl4, SnCl4,
(16) Traces of oxygen appear to increase the yield of the uncatalyzed
reaction. For a discussion of the effect of oxygen on the rate of the
cyclopropanation reaction see: (a) Miyano, S.; Hashimoto, H. J. Chem.
Soc., Chem. Commun. 1971, 1418-1419. (b) Miyano, S.; Yamashita, J.;
Hashimoto, H. Bull. Chem. Soc. Jpn. 1972, 45, 1946. (c) Miyano, S.;
Hashimoto, H. Bull. Chem. Soc. Jpn. 1973, 46, 892-897. (d) Miyano, S.;
Matsumoto, Y.; Hashimoto, H. J. Chem. Soc., Chem. Commun. 1975, 364-
365. (e) Miyano, S.; Hashimoto, H. Bull. Chem. Soc. Jpn. 1975, 48, 3665-
3668.
(17) Similar results, within experimental error, were observed when
chloromethylzinc alkoxide of cinnamyl alcohol was submitted to the same
reaction conditions with use of 5% TiCl4.
(18) For an excellent discussion of the relative strength of Lewis acids
see: (a) Farcasiu, D.; Stan, M. J. Chem. Soc., Perkin Trans 2 1998, 1219-
1222. (b) Beste, A.; Kramer, O.; Gerhard, A.; Frenking, G. Eur. J. Inorg.
Chem. 1999, 2037-2045. (c) Farcasiu, D.; Lukinskas, P.; Ghenciu, A.;
Martin, R. J. Mol. Catal. A-Chem. 1999, 137, 213-221.
(15) All these reactions were performed under an oxygen-free atmosphere
of nitrogen or argon with rigid exclusion of moisture from reagents and
glassware under standard conditions.
(19) Charette, A. B.; Lebel, H. J. Org. Chem. 1995, 60, 2966-2967 and
references therein.

12164 J. Am. Chem. Soc., Vol. 123, No. 49, 2001
Table 3. TiCl4-Catalyzed Cyclopropanation Reactiona
Charette et al.
Scheme 3
The next experiments were designed to determine whether
the species obtained by mixing ROZnCH2I and a Lewis acid
could lead to an intermolecular cyclopropanation. In a first set
of experiments, (E)-PhCHdCHCH2OZnCH2I was mixed with
its corresponding benzyl ether and a Lewis acid was added
(Scheme 3). When TiCl4 was used, cyclopropanation occurred
on both the allylic ether (23%) and the allylic alkoxide (28%),
indicating that an intermolecular methylene transfer is possible.
This was also observed, but to a lesser extent when BF3‚OEt2
was used as Lewis acid.
Intermolecular Cyclopropanation Reactions with Iodo-
and Chloromethylzinc Alkoxides as Reagents. An interesting
concept in this area would be to test whether iodomethylzinc
alkoxides can be used as carbenoids to cyclopropanate alkenes
in an intermolecular fashion. This hypothesis, if successful,
could eventually lead to enantioselective systems since it is quite
conceivable that one could start from a chiral alcohol and
generate a chiral reagent that could be used to cyclopropanate
olefins.20
a Typical procedure: To a mixture of Zn(CH2I)2 (prepared from 1
equiv of Et2Zn and 2 equiv of CH2I2) in CH2Cl2 at -78 °C was added
the allylic alcohol (1 equiv) in CH2Cl2. After 15 min of stirring at -20
°C, TiCl4 (0.15 equiv) was added. The mixture was stirred at -20 °C
for 2-7 h (see Supporting Information). b Isolated yields of cyclopro-
panes. c 0.10 equiv of TiCl4 was used. d Ratio 3:1 syn:anti of the
corresponding cyclopropylmethanol.
Scheme 2
Iodomethylzinc alkoxides prepared as previously described
can effectively cyclopropanate an allylic ether in the presence
of a Lewis acid (Table 4). Comparisons of the effectiveness of
the reaction as a function of the amount of Lewis acid (entry 1
vs 2 and 6 vs 7, Table 4) clearly indicate the importance of a
Lewis acid for the activation. Similar conversions were observed
when 2 equiv of halomethylzinc alkoxide was used and the
amount of Lewis acid was left unchanged (entry 2 vs 3, Table
4). However, when 0.3 equiv of Lewis acid was used higher
conversions were obtained (entry 4, Table 4). The results in
Table 4 also show that halomethylzinc alkoxides derived from
secondary alcohols can effectively cyclopropanate allylic ethers
in the presence of a Lewis acid. However, tertiary alcohols such
as tert-butyl alcohol are not effective since the yield of
deprotonation is only about 30%. The lower yield is probably
the result of the lower acidity and the steric bulkiness of the
alcohol.7
Similar results were observed with unfunctionalized olefins;
however, higher temperatures (room temperature) were needed
since the double bond is not as reactive as that of one containing
a proximal basic group such as that of an allylic ether (Table
5). Finally, the intermolecular cyclopropanation of unfunction-
occurred. This observation indicates that the formation of both
products occurs through an intermolecular methylene transfer
and/or is the result of a rapid equilibrium between the two
different zinc alkoxides. Rapid exchange between zinc alkoxides
through an aggregation process is well-known and is likely to
occur with the more complex ROZnCH2I. However, this result
does not rule out the possibility for an exclusive intramolecular
cyclopropanation.
(20) While this work was in progress, Shi has reported a similar approach
using Et2AlCl as the Lewis acid but no details were provided regarding the
number of equivalents: Yang, Z.; Lorenz, J. C.; Shi, Y. Tetrahedron Lett.
1998, 39, 8621-8624.

Halomethylzinc Alkoxides
J. Am. Chem. Soc., Vol. 123, No. 49, 2001 12165
Table 4. Intermolecular Cyclopropanation of an Allylic Ether with
Halomethylzinc Alkoxides
a Iodomethylzinc alkoxides were generated from 1 equiv of the
alcohol and 1 equiv of Zn(CH2I)2 at -20 °C. Chloromethylzinc
alkoxides were generated from 1 equiv of the alcohol and 1 equiv of
Zn(CH2Cl)2 at -20 °C. b Conversions determined by 400 MHz 1H NMR
or 100 MNz 13C NMR. c 0.15 equiv of TiCl4 was used instead of
BF3‚OEt2.
Figure 3. 11B NMR (96 MHz) of iodomethylzinc alkoxide in the
presence of BF3‚OEt2 (x equiv): (A) BF3‚OEt2 without iodomethylzinc
alkoxide; (B) iodomethylzinc alkoxide + 0.15 equiv of BF3‚OEt2; and
(C) iodomethylzinc alkoxide + 1 equiv of BF3‚OEt2.
triggers the cyclopropane formation, which follows generation
of the halozinc alkoxide (D, Scheme 1), and regeneration of
the Lewis acid completes the catalytic cycle. A second possible
mode of activation has been proposed by Nakamura6h in which
activation of the Lewis acid occurs by complexation of the
iodide group of the iodomethylzinc species (B′, Scheme 1).6h
Finally, a third possibility in the cases where the Lewis acid
contains a nucleophilic counterion would be the in situ genera-
tion of XZnCH2I, followed by an alkoxy-directed cyclopropa-
nation (B′′). The spectroscopic monitoring of the reaction
between an iodomethylzinc alkoxide and BF3‚OEt2 was studied
to gain further insight on the possible mode of activation and
perhaps to differentiate between B, B′, and B′′ (Figure 3). The
addition of 0.15 equiv of BF3‚OEt2 to 1.0 equiv of iodometh-
ylzinc 2-phenylethoxide led to the appearance of several other
Table 5. Intermolecular Cyclopropanation of an Unfunctionalized
Olefin with Iodomethylzinc Alkoxides
1
signals by H and 13C NMR which could correspond to the
partial formation of an alkoxyboron intermediate and a presum-
ably more reactive FZnCH2I species. This was also evident by
1H NMR where a second signal next to the ROZnCH2I singlet
at ca. 1.35 ppm (FZnCH2I) appeared along with multiplets at
2.95 and 3.90 ppm (corresponding to an alkoxyboron).7 Further
evidence was provided by 11B NMR, which clearly confirmed
the formation of a trialkylborate ((RO)3B) species at 17.9 ppm
(B, Figure 3), which is consistent with the 11B chemical shift
of other trialkylborates such as B(OMe)3.21 Conversely, the
addition of 1 equiv of BF3‚OEt2 to 1 equiv of iodomethylzinc
2-phenylethoxide led to the formation of a predominant ROBF2
species (0.05 ppm by 11B NMR) and two minor boron species
at 17.9 and 15.9 ppm which are consistent with (RO)3B and
(RO)2BF (C, Figure 3).21 These observations indicate that the
last hypothesis involving complexation of the Lewis acid that
a Iodomethylzinc alkoxides were generated from 1 equiv of the
alcohol and 1 equiv of Zn(CH2I)2 at -20 °C. b Conversions determined
1
by 400 MHz H NMR or 100 MHz 13C NMR.
alized double bonds is also possible with our halomethylzinc
alkoxides in the presence of a Lewis acid.
This series of experiments clearly indicates that the Lewis
acid-catalyzed cyclopropanation of iodomethylzinc alkoxides
can occur via either an inter- or intramolecular methylene
transfer reaction and both processes seem to be competitive.
The intermolecular process appears to be more facile when a
nucleophilic counterion is used in these reactions (such as Cl,
Br, etc.).
Discussion of the Mechanism of Activation. A possible
mechanistic scheme to account for the acceleration is presented
in Scheme 1. It is believed that the Lewis acid would complex
the iodomethylzinc alkoxide (A, Scheme 1) thus increasing its
electrophilicity. This species after activation (B, Scheme 1)
(21) 11B NMR: B(OMe)3 (18.3 ppm); (MeO)2BF (15.6 ppm); MeOBF2
(0.7 ppm). See: Wrackmeyer, B. Nuclear Magnetic Resonance Spectroscopy
of Boron Compounds Containing Two-, Three- and Four-Coordinate Boron.
In Annual Reports on NMR Spectroscopy; Webb, G. A., Ed.; Academic
Press Inc.: San Diego, CA., 1988; Vol. 20, pp 61-204.

12166 J. Am. Chem. Soc., Vol. 123, No. 49, 2001
Charette et al.
J ) 7 Hz, 3H), 0.90-0.80 (m, 1H), 0.68-0.58 (m, 1H), 0.40-0.29
(m, 2H); 13C NMR (100 MHz, CDCl3) δ 66.9, 35.8, 22.5, 19.9, 16.7,
13.8, 9.6.
induced the liberation of a reactive zinc reagent (XZnCH2I)
appears to be the most plausible one when a Lewis acid
containing a nucleophilic counterion is used in the activation
process (B′′, Scheme 1).
cis-2-(Ethyl)cyclopropylmethanol (entry 4, Table 3). The cyclo-
propanation of (Z)-2-pentenol (95 mg, 1.10 mmol) was performed
according to the previously described procedure (reaction time 2.5 h).
The residue was purified by flash chromatography on silica gel (25%
EtOAc/Hexanes) to produce the desired cyclopropylmethanol (99 mg,
94%): Rf 0.18 (20% EtOAc/Hexanes); 1H NMR (400 MHz, CDCl3) δ
3.67 (dd, J ) 11, 7 Hz, 1H), 3.59 (dd, J ) 11, 7 Hz, 1H), 1.50-1.24
(m, 3H), 1.15-1.07 (m, 1H), 1.02 (t, J ) 7 Hz, 3H), 0.90-0.80 (m,
1H), 0.71 (td, J ) 8, 5 Hz, 1H), -0.03 (dd, J ) 10, 5 Hz, 1H); 13C
NMR (75 MHz, CDCl3) δ 62.8, 21.6, 18.0, 17.8, 14.2, 9.03.
cis-2-(tert-Butyldiphenylsilyloxymethyl)cyclopropylmethanol (en-
try 5, Table 3). The cyclopropanation of (Z)-4-tert-butyldiphenylsi-
lyloxy-2-butenol (402.6 mg, 1.23 mmol) was performed according to
the previously described procedure (reaction time 6 h). The residue
was purified by flash chromatography on silica gel (25% EtOAc/
Hexanes) to produce the desired cyclopropylmethanol (340 mg, 85%):
Rf 0.45 (20% EtOAc/Hexanes); 1H NMR (300 MHz, CDCl3) δ 7.77-
7.64 (m, 4H), 7.49-7.35 (m, 6H), 4.10 (dd, J ) 12, 5 Hz, 1H), 4.00
(dd, J ) 12, 5 Hz, 1H), 3.35 (t, J ) 12 Hz, 2H), 1.51-1.37 (m, 1H),
1.32-1.17 (m, 2H), 1.06 (s, 9H), 0.76-0.65 (m, 1H), 0.13 (dd, J )
11, 5 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 135.5, 132.9, 129.8,
127.7, 64.9, 63.2, 26.7, 19.0, 18.3, 17.1, 8.21.
Conclusion. In conclusion, we have reported our findings
on the Lewis acid-catalyzed cyclopropanation of halomethylzinc
alkoxides, generated from an alcohol and bis(halomethyl)zinc.
The species have been shown to be stable for a long period of
time and monomeric in solution; however, they have been
isolated as tetrameric crystals. These species can be used as
inter- or intramolecular cyclopropanating reagents. A variety
of achiral Lewis acids are shown to be effective in this process.
This methodology competes favorably with the common
method, which usually requires a minimum of 2 equiv of Zn-
(CH2I)2 to obtain high yields of the cyclopropanes. Moreover,
we have shown that halomethylzinc alkoxides derived from
primary and secondary alkoxides in the presence of a Lewis
acid can intermolecularly cyclopropanate an allylic ether. These
findings provide the groundwork for further developments in
this area. The conclusions have already been extended to the
elaboration of an asymmetric version of this reaction which is
presented in the following paper.
Experimental Section7
cis-2-(Benzyloxymethyl)cyclopropylmethanol (entry 6, Table 3).
The cyclopropanation of (Z)-4-benzyloxy-2-butenol (139.5 mg, 0.78
mmol) was performed according to the previously described procedure
(reaction time 7 h). The residue was purified by flash chromatography
on silica gel (25% EtOAc/Hexanes) to produce the desired cyclopro-
pylmethanol (88.8 mg, 60%): Rf 0.30 (40% EtOAc/Hexanes); 1H NMR
(300 MHz, CDCl3) δ 7.39-7.26 (m, 5H), 4.60 (d, J ) 12 Hz, 1H),
4.53 (d, J ) 12 Hz, 1H), 3.99-3.91 (m, 2H), 3.20 (t, J ) 10 Hz, 1H),
3.16 (t, J ) 10 Hz, 1H), 2.75 (s(br), 1H), 1.44-1.24 (m, 2H), 0.82 (td,
J ) 8, 5 Hz, 1H), 0.22 (dd, J ) 10, 5 Hz, 1H); 13C NMR (100 MHz,
CDCl3) δ 137.4, 128.4, 127.84, 127.81, 73.0, 70.7, 62.9, 18.3, 14.6,
8.55; HRMS calcd for C12H17O2 (M + H) 193.12285, found 193.12200.
2,2-Dimethylcyclopropylmethanol (entry 7, Table 3). The cyclo-
propanation of 3-methyl-2-butenol (98.3 mg, 1.14 mmol) was performed
according to the previously described procedure (reaction time 2 h).
The residue was purified by flash chromatography on silica gel (25%
EtOAc/Hexanes) to produce the desired cyclopropylmethanol (98 mg,
90%): Rf 0.20 (20% EtOAc/Hexanes); 1H NMR (300 MHz, CDCl3) δ
3.70 (dd, J ) 11, 7 Hz, 1H), 3.53 (dd, J ) 11, 8 Hz, 1H), 1.12 (s, 3H),
1.08 (s, 3H), 0.96-0.87 (m, 1H), 0.49 (dd, J ) 8, 4 Hz, 1H), 0.12 (t,
J ) 8 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 63.8, 27.1, 26.5, 19.6,
18.1, 15.9.
1-Methyl-2-phenylcyclopropylmethanol (entry 8, Table 3). The
cyclopropanation of (E)-2-methyl-3-phenyl-2-propenol (115.4 mg, 0.78
mmol) was performed according to the previously described procedure
(reaction time 7 h). The residue was purified by flash chromatography
on silica gel (30% EtOAc/Hexanes) to produce the desired cyclopro-
pylmethanol (88 mg, 70%): Rf 0.14 (20% EtOAc/Hexanes); 1H NMR
(400 MHz, CDCl3) δ 7.31-7.26 (m, 2H), 7.22-7.17 (m, 3H), 3.57 (d,
J ) 11 Hz, 1H), 3.53 (d, J ) 11 Hz, 1H), 2.06 (dd, J ) 9, 6 Hz, 1H),
1.57 (s (br), 1H), 0.94 (dd, J ) 9, 5 Hz, 1H), 0.89 (s, 3H), 0.87 (t, J
) 5 Hz, 2H); 13C NMR (100 MHz, CDCl3) δ 138.7, 129.0, 127.9,
125.8, 71.6, 26.6, 25.0, 15.6, 15.0; HRMS calcd for C11H14O1 (M)
162.1045, found 162.1047.
General Procedure for the Lewis Acid-Catalyzed Cyclopropa-
nation of Allylic Alcohols: trans-(3-Phenylcyclopropyl)methanol
(entry 1, Table 3). To a stirred solution of CH2I2 (160 mL, 2 mmol)
in anhydrous CH2Cl2 (8 mL) at 0 °C was added dropwise diethylzinc
(100 µL, 1 mmol). The resulting solution was stirred at that temperature
for 15 min and a white precipitate was formed. The solution was cooled
at -78 °C and a solution of cinnamyl alcohol (140 mg, 1.05 mmol) in
anhydrous CH2Cl2 (5 mL) was added. The resulting heterogeneous
solution was stirred at -20 °C for 15 min and TiCl4 (16 mL, 0.15
mmol) was then added. After 3 h of stirring at -20 °C, the resulting
solution was cooled at -40 °C and poured into an aqueous solution of
saturated NH4Cl. The layer was washed with EtOAc (3×). The
combined organic layers were washed with saturated aqueous NH4Cl
and brine, dried over MgSO4, filtered, and concentrated under reduced
pressure. The crude residue was osmylated22 (OsO4 (catalyst), NMO
(2 equiv), acetone/water (4:1)) and purified by flash chromatography
on silica gel (20% EtOAc/Hexanes) to produce (133 mg, 90%) the
desired cyclopropylmethanol: Rf 0.22 (20% EtOAc/Hexanes); 1H NMR
(400 MHz, CDCl3) δ 7.30-7.25 (m, 2H), 7.20-7.15 (m, 1H), 7.10-
7.07 (m, 2H), 3.67-3.59 (m, 2H), 1.86-1.82 (m, 1H), 1.75 (s (br),
1H), 1.51-1.43 (m, 1H), 1.01-0.92 (m, 1H); 13C NMR (100 MHz,
CDCl3) δ 142.2, 128.5, 125.3, 124.6, 66.0, 24.2, 21.4, 13.5; HRMS
calcd for C10H12O1 (M) 148.0888, found 148.0880.
trans-2-(2-Phenylethyl)cyclopropylmethanol (entry 2, Table 3).
The cyclopropanation of (E)-5-phenyl-2-pentenol (191 mg, 1.17 mmol)
was performed according to the previously described procedure (reaction
time 7 h). The residue was purified by flash chromatography on silica
gel (25% EtOAc/Hexanes) to produce the desired cyclopropylmethanol
1
(174 mg, 88%): Rf 0.28 (20% EtOAc/Hexanes) H NMR (400 MHz,
CDCl3) δ 7.31-7.17 (m, 5H), 3.47-30.32 (m, 2H), 2.77-2.64 (m,
2H), 1.70-1.48 (m, 2H), 1.16 (s (br), 1H), 0.88-0.80 (m, 1H), 0.67-
0.59 (m, 1H), 0.41-0.32 (m, 2H); 13C NMR (100 MHz, CDCl3) δ
142.1, 128.4, 128.2, 125.7, 66.9, 35.8, 35.3, 21.3, 16.8, 9.7; HRMS
calcd for C12H15 (M - OH) 159.1174, found 159.1169.
trans-2-(Propyl)cyclopropylmethanol (entry 3, Table 3). The
cyclopropanation of (E)-2-hexenol (120 mg, 1.20 mmol) was performed
according to the previously described procedure (reaction time 7 h).
The residue was purified by flash chromatography on silica gel (25%
EtOAc/Hexanes) to produce the desired cyclopropylmethanol (111 mg,
85%): Rf 0.18 (20% EtOAc/Hexanes); 1H NMR (300 MHz, CDCl3) δ
3.50-3.39 (m, 2H), 1.44-1.35 (m, 2H), 1.30-1.20 (m, 3H), 0.92 (t,
1-Hydroxymethylbicyclo[4.1.0]heptane (entry 9, Table 3). The
cyclopropanation of 1-cyclohexenemethanol23 (87.6 mg, 0.78 mmol)
was performed according to the previously described procedure (reaction
time 7 h). The residue was purified by flash chromatography on silica
gel (25% EtOAc/Hexanes) to produce the desired cyclopropylmethanol
(72.1 mg, 73%): Rf 0.30 (30% EtOAc/Hexanes); 1H NMR (400 MHz,
CDCl3) δ 3.36 (d, J ) 11 Hz, 1H), 3.31 (d, J ) 11 Hz, 1H), 1.90-
1.83 (m, 2H), 1.76-1.69 (m, 1H), 1.64-1.57 (m, 1H), 1.35-1.16 (m,
5H), 0.85-0.79 (m, 1H), 0.47 (dd, J ) 9, 5 Hz, 1H), 0.26 (t, J ) 5
(22) When a quantitative conversion to the cyclopropane was not
achieved, the crude product was treated with osmium tetroxide, O3, or
KMnO4 to destroy any residual alkene and to facilitate the purification.
(23) Charette, A. B.; Marcoux, J.-F. Tetrahedron Lett. 1993, 34, 7157-
7160.

Halomethylzinc Alkoxides
J. Am. Chem. Soc., Vol. 123, No. 49, 2001 12167
Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 72.4, 26.2, 23.5, 21.7, 21.5,
21.2, 15.5, 14.8; HRMS calcd for C8H14O1 (M) 126.1045, found
126.1066.
1.90-1.85 (m, 1H), 1.58-1.49 (m, 1H), 1.09-0.98 (m, 2H); 13C NMR
(100 MHz, CDCl3) δ 142.6, 138.4, 128.3, 128.2, 127.6, 127.5, 125.8,
125.5, 73.4, 72.4, 22.6, 21.4, 14.1; HRMS calcd for C17H18O (M)
238.1357, found 238.1350. Anal. Calcd for C17H18O (238.33): C, 85.67;
H, 7.61. Found: C, 85.30; H, 7.82.
trans-1-(2-Phenylcyclopropyl)ethan-1-ol (entry 10, Table 3). The
cyclopropanation of (E)-4-phenylbut-3-en-2-ol (155 mg, 1.05 mmol)
was performed according to the previously described procedure (reaction
time 5 h). The residue was purified by flash chromatography on silica
gel (25% EtOAc/Hexanes) to produce the desired cyclopropylmethanols
(syn + anti) (97 mg, 60%). Anti-diastereomer: Rf 0.55 (25% EtOAc/
Cyclopropanation of 2-Methyl-5-phenylpent-2-ene with Halom-
ethylzinc Alkoxide in the Presence of a Lewis Acid (Table 5). The
cyclopropanation of 2-methyl-5-phenylpent-2-ene (125.0 mg, 0.78
mmol) was performed according to the previously described procedure
with CH2I2 (125 µL, 1.55 mmol), diethylzinc (80 µL, 0.78 mmol),
phenylethyl alcohol (93 µL, 0.778 mmol), and BF3‚OEt2 (0.38 M in
dichloromethane, 0.31 mL, 0.118 mmol) in dichloromethane (6.5 mL)
(reaction time 6 h at room temperature). 2-(2,2-Dimethylcyclopropyl)-
1-phenylethane: Rf 0.36 (100% Hexanes); 1H NMR (400 MHz, CDCl3)
δ 7.30-7.26 (m, 2H), 7.20-7.15 (m, 3H), 2.74-2.62 (m, 2H), 1.68-
1.54 (m, 2H), 1.02 (s, 3H), 1.00 (s, 3H), 0.55-0.47 (m, 1H), 0.38 (dd,
J ) 8.5 Hz, 4.1 Hz, 1H), -0.10 (t, J ) 4.7 Hz, 1H); 13C NMR (100
MHz, CDCl3) δ 142.7, 128.4, 128.1, 125.4, 36.5, 31.9, 27.4, 24.3, 19.7,
19.5, 15.4; HRMS calcd for C13H18 (M) 174.1408, found 174.1409.
1
Hexanes); H NMR (300 MHz, CDCl3) δ 7.30-7.08 (m, 5H), 3.40
(qn, J ) 6 Hz, 1H), 1.95-1.89 (m, 1H), 1.68 (s, 1H), 1.34 (d, J ) 6
Hz, 3H), 1.34-1.25 (m, 1H), 0.98-0.90 (m, 2H); 13C NMR (100 MHz,
CDCl3) δ 142.6, 128.2, 125.7, 125.5, 71.7, 30.7, 22.3, 21.2, 13.2. Syn-
1
diastereomer: Rf 0.32 (25% EtOAc/Hexanes); H NMR (300 MHz,
CDCl3) δ 7.30-7.06 (m, 5H), 3.39 (dq, J ) 8, 6 Hz, 1H), 1.85-1.79
(m, 1H), 1.67 (s, 1H), 1.36 (d, J ) 6 Hz, 3H), 1.33-1.24 (m, 1H),
1.05-0.93 (m, 2H); 13C NMR (100 MHz, CDCl3) δ 142.6, 128.4, 126.0,
125.6, 71.3, 30.7, 22.8, 20.7, 13.9; HRMS calcd for C11H14O1 (M)
162.10446, found 162.10464. Anal. Calcd for C11H14O (162.23): C,
81.44; H, 8.70. Found: C, 81.10; H, 9.34.
General Procedure for the Intermolecular Cyclopropanation:
Cyclopropanation of 1-Benzyloxy-3-phenylprop-2-ene with Halom-
ethylzinc Alkoxide in the Presence of a Lewis Acid (Table 4). To a
solution of CH2I2 (125 µL, 1.55 mmol) in dichloromethane (5 mL) at
-20 °C was added diethylzinc (80 µL, 0.78 mmol) with stirring for 20
min (a milky white solution was formed). Phenylethyl alcohol (93 µL,
0.78 mmol) was then added and stirred for an additional 30 min after
which time a solution of BF3‚OEt2 (0.38 M in dichloromethane, 0.31
mL, 0.118 mmol) was added. Finally, after 5 min a solution of
1-benzyloxy-3-phenylprop-2-ene (175.6 mg, 0.78 mmol) in dichlo-
romethane (1.5 mL) was added and the reaction mixture was stirred at
-20 °C. After 6 h the reaction mixture was quenched by adding an
aqueous solution of saturated NaHCO3. The organic layer is extracted
and washed with saturated aqueous Na2SO3 and brine, dried over
MgSO4, filtered, and concentrated under reduced pressure. The conver-
Acknowledgment. This work was supported by grants from
the National Science and Engineering Research Council (NSERC)
of Canada, E. W. R. Steacie Fund, Merck Frosst, Boehringer
Ingelheim, and F.C.A.R. (Que´bec). C.M. and C.B. would both
like to thank Boehringer Ingelheim and F.C.A.R. for postgradu-
ate fellowships. The authors also thank Ms. Francine Be´langer-
Garie´py for the elucidation of the solid-state structures.
Supporting Information Available: Experimental proce-
dures and characterization data for all the tabulated examples
and experimental procedures for the synthesis of starting
materials, description of the various complexes including atomic
coordinates, anisotropic thermal parameters, and fixed atom
coordinates, and a listing of atomic coordinates, distances, and
angles (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
1
sions were measured by quantitative 400 MHz H NMR or 75 MHz
13C NMR. 1-Benzyloxymethyl-2-phenylcyclopropane: Rf 0.40 (100%
toluene); 1H NMR (400 MHz, CDCl3) δ 7.44-7.13 (m, 10H), 4.63 (s,
2H), 3.61 (dd, J ) 10.3, 6.4 Hz, 1H), 3.51 (dd, J ) 10.3, 6.8 Hz, 1H),
JA010837+