Rhodium-catalyzed cyclopropanations of allylsilanes and allylstannanes: The role of the silyl/stannyl group in trans-cis stereoselection

Article abstract of DOI:10.1016/S0022-328X(01)00938-X

The stereochemistry of cyclopropanation reactions between allylsilanes and allylstannanes 6 with four representative diazoalkanes in the presence of rhodium acetate has been studied. Cyclopropanations with ethyl diazoacetate show a preference for the formation of the trans cyclopropane esters 7, whereas the corresponding reactions with (trimethylsilyl)diazomethane favor the cis stereoisomers. In both cases, overall stereoselection is low, ranging from 1.3:1 to 2.4:1. The product ratios show a small dependence on the nature of the silyl or stannyl groups, with silyl substituents giving better stereoselection than stannyl, and larger ligands on the metal center leading to lower selectivity. On the other hand, rhodium-catalyzed cyclopropanations using methyl 2-diazo-4-phenyl-3-butenoate or 1-aryldiazoacetates occur with excellent stereocontrol. The stereochemical patterns found in these reactions are in accord with an open transition state model wherein the rhodium carbenoid approaches the allylmetal π-bond from an antiperiplanar orientation with respect to the allylic carbon-ML₃ bond. While the metal center may help stabilize developing β-cationic charge in the transition states, hyperconjugation effects appear to play a minor role, if any, in directing the stereochemical course of cyclopropanation.

Full text of DOI:10.1016/S0022-328X(01)00938-X

Journal of Organometallic Chemistry 630 (2001) 57–66
www.elsevier.com/locate/jorganchem
Rhodium-catalyzed cyclopropanations of allylsilanes and
allylstannanes: the role of the silyl/stannyl group in trans–cis
stereoselection
Yi-Lun Lin, Edward Turos *
Department of Chemistry, Uni6ersity of South Florida, 4202 East Fowler A6enue, Tampa, FL 33620-5250, USA
Received 4 January 2001; received in revised form 4 May 2001; accepted 5 May 2001
This paper is dedicated with admiration to Professor Myron Rosenblum on his 75th birthday celebration, and on his many important and
original contributions to the field of organometallic chemistry
Abstract
The stereochemistry of cyclopropanation reactions between allylsilanes and allylstannanes 6 with four representative diazoalka-
nes in the presence of rhodium acetate has been studied. Cyclopropanations with ethyl diazoacetate show a preference for the
formation of the trans cyclopropane esters 7, whereas the corresponding reactions with (trimethylsilyl)diazomethane favor the cis
stereoisomers. In both cases, overall stereoselection is low, ranging from 1.3:1 to 2.4:1. The product ratios show a small
dependence on the nature of the silyl or stannyl groups, with silyl substituents giving better stereoselection than stannyl, and larger
ligands on the metal center leading to lower selectivity. On the other hand, rhodium-catalyzed cyclopropanations using methyl
2-diazo-4-phenyl-3-butenoate or 1-aryldiazoacetates occur with excellent stereocontrol. The stereochemical patterns found in these
reactions are in accord with an open transition state model wherein the rhodium carbenoid approaches the allylmetal p-bond from
an antiperiplanar orientation with respect to the allylic carbonꢀML3 bond. While the metal center may help stabilize developing
b-cationic charge in the transition states, hyperconjugation effects appear to play a minor role, if any, in directing the
stereochemical course of cyclopropanation. © 2001 Published by Elsevier Science B.V.
Keywords: Rhodium carbenoids; Allylsilanes; Allylstannanes; Cyclopropanation; Open transition state model
1. Introduction
4 to yield vinylcyclopropane esters 5 (Fig. 1). Cyclo-
propanations of monosubstituted alkenes with simple
diazoalkanes or a-diazoacetates 2 generally occur with
low diastereoselectivity, while those involving 2-diazo-
3-butenoate esters 4 give much higher stereochemical
Rhodium-catalyzed reactions of diazoalkanes with
alkenes offer an excellent method for preparing substi-
tuted cyclopropane derivatives in high yields under mild
reaction conditions [1]. Because of their ready solubility
in organic solvents and ability to avoid the formation
of undesired side products, complexes of rhodium(II)
have become popular catalysts for these types of cy-
cloadditions [2]. The most common types of cyclo-
propanations are those between alkene substrates 1 and
a-diazoacetates 2 to give cyclopropane esters [3] 3, and
the corresponding additions of 2-diazo-3-butenoates [4]
* Corresponding author. Tel.: +1-813-974-7312; fax: +1-813-974-
3203.
E-mail address: eturos@chuma.cas.usf.edu (E. Turos).
Fig. 1. Cyclopropanation reactions of alkenes.
0022-328X/01/$ - see front matter © 2001 Published by Elsevier Science B.V.
PII: S0022-328X(01)00938-X

58
Y.-L. Lin, E. Turos / Journal of Organometallic Chemistry 630 (2001) 57–66
control. Mechanistic models, which rationalize these
observations, have been developed and discussed in the
literature [3–5].
ity of these particular cyclopropanes seems to be related
to the presence of the ethoxy silyl ligands, which may
facilitate acid-promoted opening of the cyclopropyl
ring [10]. The relative stereochemistry of the purified
cyclopropane diastereomers was established through a
combination of standard one-dimensional and two-di-
mensional NMR experiments. The yields and
diastereomeric ratios of the isolated products are given
in the Table 1. In general, the reactions were clean and
yields of the cyclopropane products were uniformly
high. In no cases did we observe the formation of other
adducts such as those arising from carbene insertion
[11] into the allylic or vinylic CꢀH bonds of the allyl-
metal reagent.
The first series of diazoalkane cycloadditions we ex-
amined were those of allylmetals 6 with ethyl diazoac-
etate (entries 1–5, Table 1). The anticipated trans and
cis cyclopropane esters 7 were afforded in high yield,
but with very low diastereoselectivity. The trans:cis
ratios fall well within the ranges expected for diazoac-
etate couplings with monosubstituted alkenes (below
2.5:1), and show a small dependency on the nature of
the silyl/stannyl group. Within the silyl series, for exam-
ple, trans:cis selectivity dropped steadily from 2.1:1 to
1.4:1 to 1.2:1 as the silyl ligand L was changed from
methyl to phenyl to ethoxy (entries 1–3). From entries
4 and 5 in the table, it appears that stereoselectivity is
even lower for the corresponding allylstannane
reactions.
The results for the analogous cyclopropanations in-
volving (trimethylsilyl)diazomethane are given in entries
6–10 in Table 1. The electronic and steric properties of
the carbenoid center in this case differ markedly from
those of the species derived from ethyl diazoacetate.
This is reflected by the reversal of trans:cis stereoselec-
tivities, which now favor the cis adduct in each case.
The level of stereoselection remains equally low, how-
ever, below 1:2.5, with the allylsilanes once again offer-
ing slightly better stereocontrol than the allylstannanes.
The third group of reactions we carried out was that
of allylmetals 6 with the Davies vinyl-stabilized diazo
reagent, methyl 2-diazo-4-phenyl-3-butenoate [12] (en-
tries 11–15). In contrast to the low stereoselectivity
obtained in the cyclopropanation reactions of entries
1–10, these 6inylcyclopropanations show high selectivity
for the trans vinylcyclopropane 7. Only allyltrimethylsi-
lane and allyltriphenylsilane (entries 11 and 12) af-
forded any detectible amounts of the cis product, while
allyltriethoxysilane, allyltributyltin, and allyltri-
phenyltin provided the trans adduct exclusively. These
results are in agreement with earlier reports from
Davies who found that rhodium acetate-catalyzed addi-
tions of vinyl diazoacetates to alkenes give an unusually
high preference for the cyclopropane whose vinyl group
is trans to the alkyl substituent [3].
Despite the wide diversity of double bond com-
pounds [1,6] that have successfully served as substrates
in rhodium-promoted cyclopropanations, allylsilanes
and allystannanes have to our knowledge not been
examined. In fact, we are aware of only one example of
an allylsilane undergoing a metal-catalyzed cyclopropa-
nation with a diazo compound, specifically, that of
allyltrimethylsilane with ethyl diazoacetate [7]. The con-
ditions described for this reaction entail the dropwise
addition of ethyl diazoacetate to a boiling suspension of
copper and copper acetate in neat allyltrimethylsilane.
The silyl- and stannyl-activated cyclopropane products
derived from these reactions can serve as homoallyl
anion equivalents in ring-opening processes with elec-
trophiles and thus are useful synthetic reagents [7,8].
During the course of recent investigations [9] on the
thermal rearrangement properties of 1-silylmethyl- and
1-stannylmethyl-2-vinylcyclopropanes, we had the op-
portunity to examine the reactions of allylsilanes and
allylstannanes with rhodium carbenoids generated from
diazoalkanes. In addition to utilizing the derived ad-
ducts, we were interested in examining the possible role
that the silyl or stannyl groups might have on the
stereochemistry of rhodium(II)-catalyzed cyclopropana-
tion. Current mechanistic models suggest that bulky,
branched substituents proximal to the alkene can influ-
ence trans:cis ratios of the cyclopropane products. In
addition to potential steric perturbations, silyl and stan-
nyl substituents adjacent to an alkene center can exert
powerful b-cation stabilizing effects, which in the case
of cyclopropanation, might play a role in guiding the
stereochemical course of ring formation. In this report,
we summarize our findings and discussions on these
particular elements of the reaction.
2. Results
In this study, we examined the rhodium acetate-cata-
lyzed reactions of allylsilanes and allystannanes 6 with
four sterically and electronically different types of dia-
zoalkanes, namely, ethyl diazoacetate, (trimethyl-
silyl)diazomethane, 2-aryldiazoacetates, and methyl 2-
diazo-4-phenylbut-3-enoate (Table 1). All of the reac-
tions were done under standard cyclopropanation
conditions at room temperature by slow addition of the
diazoester to a dilute mixture of the allylmetal reagent
and 5 mol% of rhodium acetate in methylene chloride.
Upon completion of the reaction, the product mixtures
were evaporated and subjected to flash chromatogra-
phy, with the exception of the triethoxysilylmethyl cy-
clopropanes, which were unstable and rapidly
decomposed upon attempted purification. The instabil-

Y.-L. Lin, E. Turos / Journal of Organometallic Chemistry 630 (2001) 57–66
59
Table 1
Rhodium-catalyzed cyclopropanations of allysilanes and allylstannanes 6
a
Entry
L3M
R%
R%%
Trans:cis
Isolated % yields a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Me3Si
Ph3Si
CO2Et
CO2Et
CO2Et
CO2Et
CO2Et
SiMe3
SiMe3
SiMe3
SiMe3
SiMe3
CO2Me
CO2Me
CO2Me
CO2Me
CO2Me
CO2Me
CO2Me
CO2Me
H
H
H
H
H
H
H
H
2.1:1
1.4:1
1.2:1
94
95
b
(EtO)3Si
Bu3Sn
Ph3Sn
Me3Si
Ph3Si
(EtO)3Si
Bu3Sn
Ph3Sn
Me3Si
Ph3Si
(EtO)3Si
Bu3Sn
Ph3Sn
Ph3Si
Ph3Si
Ph3Sn
91 b
94
1.5:1
1.2:1
1:2.0
1:2.4
1:2.1 b
1:1.4
1:1.7
12.7:1
9.0:1
Only trans
Only trans
Only trans
Only trans
Only trans
Only trans
96
96
97
90 b
95
H
H
96
96
90
CHꢁCHPh (E)
CHꢁCHPh (E)
CHꢁCHPh (E)
CHꢁCHPh (E)
CHꢁCHPh (E)
Ph
93 b
93 c
92
b
b
96
4-NO2Ph
Ph
96
90 b
a Based on isolated material after column chromatography, except where indicated.
b Based on crude material obtained after trituration with hexane. Diastereomeric ratios were determined by integration of the signals in the
proton NMR spectrum.
c Depending on the workup and purification conditions, variable amounts of ring opened, destannylated product can also be obtained.
from the rhodium center. p-Complex A must then
undergo a rotation about the developing CꢀC bond to
bring the carbonꢀrhodium bond into proper alignment
with the alkene center to form the cyclopropane ring.
During this rotational process, the alkene becomes
polarized such that positive charge character builds up
at the more substituted alkene carbon as it swings over
the R% substituent, as shown in transition structure B.
For carbenoids whose R% and R%% groups are non-equiv-
alent two different p-complexes can form, each leading
to a different diastereomeric cyclopropane adduct.
When the carbenoid R% or R%% group is an ester (or
other suitable electron pair donor), cationic charge
The final series of reactions we studied were those of
allyltriphenylsilane and allyltriphenylstannane with 1-
aryl diazoacetates (entries 16–18 in Table 1). Like those
of the preceding vinylcyclopropanations, the stereose-
lectivity of these additions was exceptionally high, giv-
ing only the adduct having the ester and CH2ML3
substituents trans to each other on the cyclopropane
ring.
3. Discussion
Rhodium(II)-catalyzed couplings of diazoalkanes
with alkenes are thought to proceed via a rhodium-sta-
bilized carbenoid intermediate, an electrophilic species
which transfers the carbene component to the alkene
centers with retention of the original double bond
geometry [1]. The most commonly used stereochemical
model for rhodium acetate-promoted cyclopropanation
was first postulated by Doyle in 1984, and which is
reproduced in Fig. 2 [1]. In the initial step of the
reaction, a transient p-complex (A) is formed between
the rhodium carbenoid and the alkene. Doyle has ar-
gued that the rhodium carbenoid coordinates orthogo-
nally to the alkene with respect to the carbonꢀrhodium
bond, such that the alkene substituent R orients away
Fig. 2. The Doyle model for Rh(II)-catalyzed diazoalkane cyclo-
propanation of monoalkylated alkenes.

60
Y.-L. Lin, E. Turos / Journal of Organometallic Chemistry 630 (2001) 57–66
should be stabilized equally well by the b-silyl/stannyl
group, differences in steric interactions within these two
transition states could be of greater importance. Desta-
bilizing forces between R and R% should increase as the
metal center and its ligands become larger, thus de-
creasing the relative amount of the trans adduct. This
could explain the diminishing trans selectivity found for
entries 1–5 in Table 1. However, the fact that the
trans:cis product ratios do not significantly change
suggests that the steric effects of the silyl/stannyl moi-
eties must play a minimal role in stereoregulation of the
cyclopropanation process [20].
Fig. 3. Stereochemistry of electrophilic addition to an allymetal.
To provide a clearer picture as to why this might be
the case, we postulate an alternative depiction for the
cycloaddition process that is based on the well-known
open transition state model for allylsilane/stannane ad-
ditions to aldehyde–Lewis acid complexes [21]. Accord-
ing to this model, attack of the electrophile (E+)
occurs from the less shielded face of the allylmetal
p-system and antiperiplanar to the allylic carbonꢀmetal
bond (Fig. 3). This orientation allows developing
cationic charge in the transition state to be stabilized
through hyperconjugation with the adjacent ML3 moi-
ety. Adaptation of the open transition state model to
the rhodium-catalyzed cyclopropanations of allylsi-
lanes/stannanes, shown in Fig. 4, suggests that the
rhodium carbenoid can orient its substituents either
endo and exo to the allylmetal p-system during the
addition process. This arrangement is effectively equiv-
alent to that in the Doyle model for the alkene–
rhodium carbenoid p-complex, where the substituents
Rendo and Rexo in the open transition state model corre-
spond to R% and R%% in the Doyle model, respectively.
Substituents on the carbenoid that can stabilize cationic
charge in the transition state (e.g. CO2Et), either
through electrostatic or secondary orbital interactions,
would prefer to occupy the Rendo position and give the
trans product. Conversely, a sterically repulsive group
would favor the more open Rexo orientation, affording
the cis stereoadduct. Indeed, our model reactions of
allylmetals 6 with the sterically hindered (trimethylsi-
lyl)diazomethane (entries 6–10 in Table 1) confirm that
the favored product is the cis cyclopropane (Rendo=H,
Fig. 4. Comparison of the open transition state model (above) and
Doyle model (below) for rhodium–carbenoid additions to an ally-
metal.
developing in the transition state can be stabilized
through electrostatic interaction with the eclipsing R%
moiety in B, thus delivering the trans adduct as the
major product. Substituents that do not have this stabi-
lization capability prefer to occupy the R%% site to
minimize unfavorable steric interactions, thus giving the
cis product predominantly [13].
Application of the Doyle model to the reactions of
the allylsilanes and allystannanes with ethyl diazoac-
etate (entries 1–5 in Table 1) would suggest that the
allylic metal group (R=CH2ML3) could also play a
role in promoting electropositive charge development at
the alkene center in the transition states, and thereby
possibly reduce to some degree the through-space stabi-
lizing effect of the carbenoid ester group (R%=ester).
Silyl and stannyl substituents are effective at stabilizing
electropositive character at a b-carbon [14] through
orbital hyperconjugation [15], and this effect can be
influenced by the ligands on the silyl or stannyl center
which attenuate the electron donating ability of the
metalꢀcarbon bond [16–18]. Mayr [16] and Brook [17]
have determined that the b-cation stabilizing ability of
silyl and stannyl groups increases in the order
of (RO)3SiBAr3SiBR3SiB(RO)3SnBAr3SnBR3Sn.
This corresponds with the steady drop in trans:cis
selectivity we see in the allylsilane cyclopropanation
reactions upon changing the silyl ligands (entries 1–3).
However, given that the amount of the trans product in
these cyclopropanations diminishes even further when
the silyl center is replaced with stannyl (entries 4 and 5),
it would appear that hyperconjugation effects do not
decisively influence stereochemistry of these carbenoid
additions [19]. Since developing positive charge in the
transition states leading to the trans and cis products
Rexo=SiMe3 in Fig. 4). The overall low level of stereo-
chemical control exerted by the allylic silyl and stannyl
moieties in these additions supports the notion that the
bulky ML3 group must lie well outside the congested
regions of the transition state, and distal to the site
where CꢀC bond formation is occurring.
To account for the unusually high stereochemical
control in the rhodium-catalyzed vinylcyclopropana-
tions (entries 11–15, Table 1), we cite the earlier pro-
posal [3] by Davies that the transition state leading to
the disfavored (minor) adduct suffers serious destabiliz-

Y.-L. Lin, E. Turos / Journal of Organometallic Chemistry 630 (2001) 57–66
61
ing interactions between the alkene substituent and the
vinyl moiety, while the transition state leading to the
favored (major) cyclopropane product experiences sta-
bilizing interactions between the ester and the cationic
center (Fig. 5) [22]. For the trimethylsilyl and
triphenylsilyl systems (entries 11 and 12), the small
amount of cis product formed may be due to a still
tolerable amount of steric interactions in the transition
state leading to the disfavored cis compound, which
become prohibitive when ML3 is Si(OEt)3 or SnL3.
A similar argument can be invoked to explain the
complete stereochemical control observed in entries
16–18 of Table 1 for the reactions of allyltriphenylsi-
lane and allyltriphenylstannane with 2-aryl substituted
diazoacetates. Here, the additions yield exclusively ad-
ducts having the CH2ML3 and ester groups trans on the
ring. Application of the Doyle/Davies models would
indicate that steric repulsion between the CH2ML3
moiety and the carbenoid aryl ring in the p-complex
(and subsequent transition state) leading to the disfa-
vored cyclopropane product would be severe (Fig. 6).
In conclusion, these investigations indicate that the
factors which typically govern trans:cis stereoselection
in rhodium-catalyzed cyclopropanations of alkenes re-
main dominant for the corresponding reactions involv-
ing allylsilane and allystannane substrates. Although
the silyl/stannyl centers and their ligands do provide a
secondary influence on trans:cis product ratios, these
affects are relatively small and appear to be primarily
steric in nature. Our results can best be explained by
the combination of the Doyle and open transition state
models, which places the allylic silyl/stannyl moiety
outside the sterically congested regions and well away
from where bond formation is occurring in the transi-
tion state. This precludes the ML3 group from playing
a more prominent role in controlling the reaction stere-
ochemistry. Hyperconjugative stabilization of positive
charge by the silyl and stannyl substituents in the
transition state may be a contributing factor in the
reactions, but does not play a significant role in regulat-
ing the trans:cis stereoselection. These concepts may be
extended to related metal-catalyzed cyclopropanation
reactions involving other types of alkene substrates
bearing allylic substitution.
4. Experimental
4.1. General
All organic reagents were purchased from Aldrich
Chemical Company and used without further purifica-
tion. Solvents were obtained from Fisher Scientific
Company. CH2Cl2 was distilled from CaH2 under ni-
trogen immediately before use. All reactions were per-
formed under an argon or nitrogen atmosphere using
glassware and syringes that had been pre-dried
overnight at 120 °C. Thin layer chromatography (TLC)
was carried out using EM Reagent plates with a
fluorescence indicator (SiO2-60, F-254). Products were
purified by flash chromatography using J.T. Baker flash
1
chromatography silica gel (40 m). H- and 13C-NMR
Fig. 5. Pathways for Rh2(OAc)4-catalyzed cyclopropanation of allyl-
metals 6 with methyl 2-diazo-4-phenyl-3-butenoate.
spectra were recorded in CDCl3 solution on Varian
Inova-500, Inova-400, and VXR-300 instruments or
Bruker AMX-360 and AMX-250 spectrometers. For
1
sequential H-NMR assignments, one-dimensional nu-
clear Overhauser enhancement difference spectroscopy
(1D NOE), two-dimensional correlation spectroscopy
(2D COSY), two-dimensional rotating frame Over-
hauser enhancement spectroscopy (2D ROESY), and
two-dimensional total correlation spectroscopy (2D
TOCSY) experiments were performed. 1D NOE differ-
ence spectra were recorded on a Varian Inova-500 and
a Bruker AMX-360 using multiple irradiation with 32
scans each. 2D spectra were acquired in a phase sensi-
tive mode using time proportional phase increments
(TPPI), a phase sensitive four-quadrant transformation,
on a Bruker AMX-360. 1024 data points were acquired
over a 3.6 kHz spectral width for 128 increments. Eight
to sixty-four scans plus two to four dummy scans were
Fig. 6. Pathways for Rh2(OAc)4-catalyzed cyclopropanation of allyl-
metals 6 with methyl 2-diazo-2-arylacetates.

62
Y.-L. Lin, E. Turos / Journal of Organometallic Chemistry 630 (2001) 57–66
4.2.3. cis- and trans-1-Ethoxycarbonyl-2-
(triethoxysilylmethyl)cyclopropane
acquired per increment. 2D data sets were multiplied in
both dimensions by a sinebell squared function and
zero filled to 512 in both dimensions. The IR spectra
were measured in CH2Cl2 solution on a Perkin–Elmer
Paragon 1000 Series Model FT Infrared Spectrophoto-
meter. Mass spectra were run using chemical ionization
(CI) with isobutane as the ionizing gas. High-resolution
mass spectra were obtained using either CI with pe-
rfluorokerosene as an internal standard or electron
impact (EI). Elemental analyses were performed by
Atlantic Microlabs, Inc., Norcross, GA.
Crude yellow oil, 399 mg (91%), trans:cis=1.2:1,
unstable to column chromatography. 1H-NMR (360
MHz) l 4.22 (m, cis, 1H), 4.14 (m, cis, 1H), 4.08 (m,
trans, 2H), 3.80 (m, cis/trans, 6H), 1.66 (m, cis, 1H),
1.35 (m, trans, 2H), 1.32 (m, trans, 1H), 1.28 (m, cis,
2H), 1.20 (m, cis/trans, 12H), 1.40 (m, cis, 1H), 0.90 (m,
trans, 1H), 0.88 (m, cis, 1H), 0.83 (m, trans, 1H), 0.70
(m, trans, 1H), 0.58 (dd, cis, 1H, J=15.2, 7.5 Hz).
13C-NMR (90 MHz) l 174.6 (trans), 173.1 (cis), 133.8
(cis), 130.0 (trans), 60.4 (trans), 60.3 (cis), 58.6 (trans),
58.5 (cis), 22.3 (trans), 19.3(cis), 18.4 (cis/trans), 17.5
(trans), 17.4 (cis), 16.5 (trans), 15.3 (cis), 15.1 (trans),
14.5 (cis), 14.4 (trans), 8.3 (cis). HRMS; m/z: Found:
291.1611. Calc. for C17H25O2Si: 291.1620 [M+1].
4.2. Preparation of cyclopropyl esters
A solution of ethyl diazoacetate (173 mg, 1.51 mmol)
in 1 ml of CH2Cl2 was added dropwise over 5 h to a 1
ml CH2Cl2 solution of allyltriphenylsilane (500 mg, 1.66
mmol) and rhodium(II) acetate (3.3 mg, 7.6 mmol) at
room temperature (r.t.) under nitrogen. The solution
was stirred for an additional 1 h at r.t. and then
evaporated. The cis and trans isomers of 1-ethoxycar-
bonyl-2-(triphenylsilylmethyl)cyclopropane were sepa-
rated by column chromatography with a mixed solvent
gradient of petroleum ether and CH2Cl2 to give 231 mg
(40%) of cis-1-ethoxycarbonyl-2-(triphenylsilylmethyl)-
cyclopropane and 323 mg (55%) of trans-1-ethoxycar-
bonyl-2-(triphenylsilylmethyl)cyclopropane as white
solids.
4.2.4. cis-1-Ethoxycarbonyl-2-(tributylstannylmethyl)-
cyclopropane
1
Colorless oil, 234 mg (38%). H-NMR (360 MHz) l
4.08 (m, 2H), 1.65 (m, 1H), 1.56 (m, 2H), 1.43 (m, 4H),
1.27 (m, 9H), 1.51 (m, 1H), 0.97 (m, 3H), 0.83 (m,
15H). 13C-NMR (90 MHz) l 174.1, 60.0, 29.2, 27.4,
21.1, 20.8, 15.8, 14.3, 8.9, 6.4.
4.2.5. trans-1-Ethoxycarbonyl-2-(tributylstannylmethyl)-
cyclopropane
1
Colorless oil, 344 mg (56%). H-NMR (360 MHz) l
4.11 (q, 2H, J=7.1 Hz), 1.60 (m, 2H), 1.46 (m, 6H),
1.41 (m, 1H), 1.32 (m, 6H), 1.26 (m, 12H), 1.01 (dd,
1H, J=12.9, 6.1 Hz), 0.90 (t, 15H, J=7.3 Hz), 0.81
(1m, H), 0.67 (m, 2H). 13C-NMR (90 MHz) l 174.6,
60.4, 29.4, 27.6, 24.4, 22.4, 19.1, 14.5, 13.9, 13.5, 9.2.
4.2.1. cis-1-Ethoxycarbonyl-2-(triphenylsilylmethyl)-
cyclopropane
Melting point: 61–62 °C. 1H-NMR (500 MHz) l
7.55 (dd, 6H, J=7.5, 1.5 Hz), 7.41 (tt, 3H, J=7.5, 1.5
Hz), 7.36 (t, 6H, J=7.5 Hz), 4.09 (dq, 1H, J=10.5, 7.0
Hz), 3.95 (dq, 1H, J=10.5, 7.0 Hz), 1.71 (m, 2H), 1.63
(m, 1H), 1.41 (m, 1H), 1.22 (t, 3H, J=7.0 Hz), 0.95 (m,
1H), 0.84 (m, 1H). 13C-NMR (125 MHz) l 175.0, 136.5,
135.3, 130.4, 128.7, 61.0, 23.6, 19.3, 18.8, 18.7, 15.1. IR
4.2.6. cis-1-Ethoxycarbonyl-2-(triphenylstannylmethyl)-
cyclopropane
1
Colorless oil, 310 mg (43%). H-NMR (360 MHz) l
7.57 (m, 6H), 7.39 (m, 9H), 4.08 (m, 1H), 3.94 (m, 1H),
1.82 (m, 1H), 1.72 (m, 1H), 1.58 (m, 2H), 1.20 (t, 3H,
J=7.2 Hz), 1.02 (m, 1H), 0.95 (m, 1H). 13C-NMR (90
MHz) l 172.8, 138.8, 137.3, 129.0, 129.0, 128.7, 21.5,
20.2, 16.5, 14.8, 9.1. CIMS; m/z: 478.2 ([M+1]).
+
1723 cm−1. EIMS; m/z 386.2 ([M ). Anal. Found: C,
77.80; H, 6.84. Calc. for C25H26O2Si: C, 77.68; H,
6.78%.
]
4.2.7. trans-1-Ethoxycarbonyl-2-(triphenylstannyl-
methyl)cyclopropane
White solid, 382 mg (53%); m.p: 85–86 °C. 1H-
NMR (360 MHz) l 7.56 (m, 6H), 7.38 (m, 9H), 4.03
(m, 2H), 1.69 (m, 2H), 1.41 (m, 2H), 1.24 (m, 1H), 1.21
(t, 3H, J=7.1 Hz), 0.74 (m, 1H). 13C-NMR (90 MHz)
l 174.1, 138.5, 137.2, 129.1, 128.8, 24.5, 21.3, 19.1, 15.8,
14.4. CIMS; m/z: 478.2 ([M+1]).
4.2.2. trans-1-Ethoxycarbonyl-2-(triphenylsilylmethyl)-
cyclopropane
Melting point: 73–74 °C. 1H-NMR (500 MHz) l
7.54 (dt, 6H, J=7.0, 1.0 Hz), 7.42 (tt, 3H, J=7.0, 1.0
Hz), 7.37 (t, 6H, J=7.0 Hz), 4.03 (m, 2H), 1.62 (dd,
1H, J=14.5, 6.0 Hz), 1.52 (m, 1H), 1.32 (dd, 1H,
J=14.5, 7.5 Hz), 1.30 (m, 1H), 1.22 (t, 3H, J=7.0 Hz),
1.10 (m, 1H), 0.60 (m, 1H). 13C-NMR (125 MHz) l
173.6, 136.5, 135.5, 130.2, 128.6, 61.0, 20.3, 18.4, 16.4,
15.1, 11.5. Anal. Found: C, 77.53; H, 6.81. Calc. for
C25H26O2Si: C, 77.68; H, 6.78%.
4.3. Procedure for the preparation of
trimethylsilyl-substituted cyclopropanes
(Trimethylsilyl)diazomethane (2 M in hexane, 1 ml,
2.0 mmol) was added dropwise over 5 h into a 1 ml

Y.-L. Lin, E. Turos / Journal of Organometallic Chemistry 630 (2001) 57–66
63
CH2Cl2 solution of allyltrimethylsilane (215 mg, 2.2
mmol) and rhodium(II) acetate (3.3 mg, 7.6 mmol) at
r.t. under nitrogen. The solution was stirred further for
1 h at r.t. and then concentrated. The crude sample was
purified by column chromatography with pentane to
give 385 mg (96%) of 1-trimethylsilyl-2-(trimethylsilyl-
methyl)cyclopropane (trans:cis=1:2.0) as a colorless
oil. 1H-NMR (360 MHz) l 0.99 (m, trans, 1H), 0.93 (m,
trans, 1H), 0.78 (m, trans, 1H), 0.69 (dd, cis, 1H,
J=14.4, 6.1 Hz), 0.56 (m, cis, 1H), 0.43 (dd, cis, 1H,
J=14.4, 7.6 Hz), 0.36 (tt, cis, 1H, J=7.2, 3.2 Hz), 0.22
(m, cis, 1H), 0.03 (s, cis, 18H), −0.06 (s, trans, 18H),
−0.44 (tt, trans, 1H, J=9.0, 8.3 Hz), –0.68 (tt, cis, 1H,
J=9.7, 6.1 Hz). 13C-NMR (90 MHz) l 24.0 (cis), 19.8
(trans), 12.0 (trans), 11.2 (cis), 10.8 (cis), 10.7 (trans),
6.5 (cis), 3.7 (trans), 0.2 (trans), −1.1 (cis), −1.4
(trans) −2.0 (cis). Anal. Found: C, 60.02; H, 11.83.
Calc. for C10H24Si2: C, 59.91; H, 12.07%.
J=13.0, 7.6 Hz), 1.37 (m, trans, 1H), 1.20 (t, trans, 1H,
J=12.6 Hz), 1.01 (m, cis, 1H), 0.43 (m, cis, 1H), 0.40
(m, trans, 1H), 0.12 (s, trans, 9H), −0.15 (s, cis, 9H),
−0.22 (m, trans, 1H), −0.45 (m, cis, 1H). 13C-NMR
(90 MHz) l 139.5 (cis), 139.4 (trans), 137.3 (cis/trans),
129.0 (cis/trans), 128.6 (cis/trans), 18.9 (cis), 15.3
(trans), 14.7 (trans), 13.7 (cis), 12.9 (cis), 12.1 (cis), 9.9
(trans), 7.5 (cis), 0.3 (trans), −2.2 (cis). CIMS; m/z:
478.2 ([M+1]).
4.4. Procedure for the preparation of
1-(2-phenylethenyl)- and 1-aryl-substituted
cyclopropanes esters
A solution of methyl 2-diazo-4-phenyl-3-butenoate
(483 mg, 2.39 mmol) dissolved in 1 ml of CH2Cl2 was
added dropwise over 5 h into a 1 ml CH2Cl2 solution
of allyltrimethylsilane (300 mg, 2.63 mmol) and rho-
dium(II) acetate (5.3 mg, 11.9 mmol) at r.t. under
nitrogen. The solution was stirred for 1 h at r.t. and
then concentrated. The residue was triturated with 5 ml
of hexanes and filtered, and the solvent was removed
under reduced pressure. The diastereomers were sepa-
rated by column chromatography with a mixed solvent
gradient of petroleum ether and CH2Cl2 to afford 616
mg (89%) of E-(1R*,2R*)-1-methoxycarbonyl-1-(2-
phenylethenyl) - 2 - (trimethylsilylmethyl)cyclopropane
and 46 mg (7%) of E-(1S*,2R*)-1-methoxycarbonyl-1-
(2-phenylethenyl)-2-(trimethylsilylmethyl)cyclopropane
as colorless oils.
4.3.1. cis-and trans-1-(Trimethylsilyl)-2-
(triphenylsilylmethyl)cyclopropane
Colorless oil, 750 mg (97%), trans:cis=1:2.4. 1H-
NMR (360 MHz) l 7.6 (m, cis/trans, 6H), 7.42 (m,
cis/trans, 9H), 1.95 (dd, trans, 1H, J=14.4, 2.2 Hz),
1.57 (dd, cis, 1H, J=14.8, 6.1 Hz), 1.47 (dd, cis, 1H,
J=14.8, 6.5 Hz), 1.19 (m trans, 1H), 1.04 (dd, trans,
1H, J=14.4, 10.8 Hz), 0.80 (m, cis, 1H), 0.73 (m, trans,
1H), 0.37 (m, cis, 1H), 0.31 (m, cis, 1H), 0.09 (s, trans,
9H), 0.03 (m, trans, 1H), −0.17 (s, cis, 9H), −0.34 (m,
trans, 1H), −0.51 (m, cis, 1H). 13C-NMR (90 MHz) l
136.0 (trans), 135.9 (cis), 135.6 (cis), 135.5 (trans),
129.6 (cis/trans), 128.0 (cis/trans), 20.7 (cis), 16.8
(trans), 12.0 (trans), 11.9 (trans), 11.7 (cis), 11.1 (cis),
7.9 (cis), 4.9 (trans), 0.2 (trans), −2.2 (cis). Anal.
Found: C, 77.91; H, 8.02. Calc. for C25H30Si2: C, 77.65;
H, 7.82%.
4.4.1. E-(1R*,2R*)-1-Methoxycarbonyl-1-
(2-phenylethenyl)-2-(trimethylsilylmethyl)cyclopropane
1H-NMR (500 MHz) l 7.43 (d, 2H, J=7.5 Hz), 7.33
(t, 2H, J=7.5 Hz), 7.24 (t, 1H, J=7.5 Hz), 6.66 (d,
1H, J=16.0 Hz), 6.31 (d, 1H, J=16.0 Hz), 3.71 (s,
3H), 1.69 (m, 2H), 1.04 (m, 1H), 0.76 (dd, 1H, J=16.0,
4.0 Hz), 0.33 (m, 1H), 0.03 (s, 9H). 13C-NMR (90
MHz) l d 175.2, 137.3, 132.0, 128.7, 127.6, 126.5,
4.3.2. cis- and trans-1-(Tributylstannylmethyl)-2-
(trimethylsilyl)cyclopropane
Colorless oil, 793 mg (95%), trans:cis=1:1.4. 1H-
NMR (360 MHz) l 1.50 (m, cis/trans, 6H), 1.31 (m,
cis/trans, 6H), 1.19 (m, cis, 1H), 0.90 (m, cis/trans,
15H), 0.73 (m, trans, 1H), 0.40 (m, trans, 2H), 0.23 (m,
cis, 1H), 0.05 (s, trans, 9H), −0.03 (m, cis, 2H), −0.37
(m, trans, 1H), −0.69 (m, cis, 1H). 13C-NMR (90
MHz) l 29.5 (cis/trans), 27.9 (cis/trans), 16.6 (cis), 15.0
(trans), 14.2 (cis), 14.0 (cis/trans), 12.8 (trans), 12.6
(cis), 11.7 (trans), 9.2 (cis/trans), 9.1 (cis), 6.8 (trans),
0.3 (trans), −1.9 (cis). CIMS; m/z: 418.2 ([M+1]).
125.2, 52.3, 31.0, 28.8, 21.1, 15.7, −1.3. IR 1749 cm−1
.
CIMS (isobutene); m/z: 289.2 ([M+1]). HRMS (CI,
isobutane) Found: 289.1599. Calc. for C17H25O2Si
289.1625 [M+1].
4.4.2. E-(1S*,2R*)-1-Methoxycarbonyl-1-
(2-phenylethenyl)-2-(trimethylsilylmethyl)cyclopropane
1H-NMR (500 MHz) l 7.36 (d, 2H, J=7.0 Hz), 7.30
(t, 2H, J=7.0 Hz), 7.21 (tt, 1H, J=7.5, 1.5 Hz), 6.90
(d, 1H, J=16.5 Hz), 6.19 (d, 1H, J=16.5 Hz), 3.75 (s,
3H), 1.50 (m, 2H), 1.33 (m, 1H), 0.91 (dd, 1H, J=15.0,
5.0 Hz), 0.73 (dd, 1H, J=15.0, 10.5 Hz), 0.05 (s, 9H).
13C-NMR (90 MHz) l 172.6, 137.5, 130.0, 128.7, 127.3,
126.5, 126.3, 52.16, 32.1, 32.0, 22.5, 15.4, −1.3.
4.3.3. cis- and trans-1-(Trimethylsilyl)-2-
(triphenylstannylmethyl)cyclopropane
Colorless oil, 916 mg (96%), trans:cis=1:1.7. 1H-
NMR (360 MHz) l 7.61 (m, cis/trans, 6H), 7.41 (m,
cis/trans, 9H), 2.04 (dd, trans, 1H, J=12.6, 2.9 Hz),
1.73 (dd, cis, 1H, J=13.0, 6.5 Hz), 1.58 (dd, cis, 1H,

64
Y.-L. Lin, E. Turos / Journal of Organometallic Chemistry 630 (2001) 57–66
4.4.7. E-(1R*,2R*)-1-(Methoxycarbonyl)-1-
(2-phenylethenyl)-2-(triphenylstannylmethyl)-
cyclopropane
4.4.3. E-(1R*,2R*)-1-Methoxycarbonyl-1-
(2-phenylethenyl)-2-(triphenylsilylmethyl)cyclopropane
1
Colorless oil, 924 mg (81%). H-NMR (500 MHz) l
1
Colorless oil, 463 mg (92%). H-NMR (360 MHz) l
7.52 (dd, 6H, J=8.0, 1.0 Hz), 7.44 (tt, 3H, J=8.0, 1.5
Hz), 7.37 (m, 11H), 6.28 (d, 1H, J=16.0 Hz), 6.24 (d,
1H, J=16.0 Hz), 3.64 (s, 3H), 1.92 (m, 1H), 1.55 (dd,
1H, J=9.0, 5.0 Hz), 1.52 (dd, 1H, J=15.0, 6.0 Hz),
1.37 (dd, 1H, J=15.0, 8.0 Hz), 0.99 (dd, 1H, J=6.5,
5.0 Hz). 13C-NMR (125 MHz) l 175.5, 137.6, 136.4,
135.1, 132.8, 130.3, 129.2, 128.6, 128.2, 127.1, 125.4,
52.8, 32.3, 28.3, 21.9, 13.0. IR 1723 cm−1. Anal.
Found: C, 81.23; H, 6.01. Calc. for C32H30O2Si: C,
80.97; H, 6.37%.
7.50 (m, 6H), 7.31 (m, 14H), 6.49 (d, 1H, J=16.2 Hz),
6.29 (d, 1H, J=16.2 Hz), 3.59 (s, 3H), 2.04 (m, 1H),
1.59 (m, 2H), 1.51 (dd, 1H, J=8.3, 6.2 Hz), 1.09 (dd,
1H, J=6.2, 4.8 Hz). 13C-NMR (90 MHz) l 174.5,
138.4, 137.2, 132.4, 129.2, 129.0, 128.8, 127.6, 126.6,
124.4, 52.2, 33.0, 30.3, 22.2, 10.6. IR 1723 cm−1
.
CIMS; m/z: 566.3 ([M+1]).
4.4.8. (1R*,2R*)-1-Methoxycarbonyl-1-phenyl-
2-(triphenylsilylmethyl)cyclopropane
1
Colorless oil, 450 mg (96%). H-NMR (360 MHz) l
7.47–7.18 (m, 20H), 3.53 (s, 3H), 2.05 (m, 1H), 1.64
(ABm, 2H), 0.89 (dd, 1H, J= 6.5, 4.3 Hz), 0.42 (dd,
1H, J=14.8, 11.2 Hz). Anal. Found: C, 80.54; H, 6.49.
Calc. for C30H28O2Si: C, 80.32; H, 6.29%.
4.4.4. E-(1S*,2R*)-1-Methoxycarbonyl-1-
(2-phenylethenyl)-2-(triphenylsilylmethyl)cyclopropane
1
Colorless oil, 97 mg (9%). H-NMR (360 MHz) l
7.55 (m, 6H), 7.38 (m, 9H), 7.18 (m, 1H), 6.72 (d, 1H,
J=16.1 Hz), 6.04 (d, 1H, J=16.1 Hz), 3.65 (s, 3H),
1.82 (dd, 1H, J=14.8, 4.8 Hz), 1.66 (dd, 1H, J=14.8,
8.9 Hz), 1.49 (m, 1H), 1.39 (m, 2H). 13C-NMR (90
MHz) l 172.5, 137.4, 136.0, 134.7, 129.8, 129.6, 128.7,
128.1, 127.4, 126.9, 126.3, 52.3, 32.6, 30.9, 23.2, 12.1.
4.4.9. (1R*,2R*)-1-Methoxycarbonyl-1-
(4-nitrophenyl)-2-(triphenylsilylmethyl)cyclopropane
1
Colorless oil, 380 mg (96%). H-NMR (360 MHz) l
8.15 (d, 2H, J=10.8 Hz), 7.42–7.21 (m, 17H), 3.51 (s,
3H), 2.09 (m, 1H), 1.68 (m, 1H), 1.52 (m, 1H), 0.87 (dd,
1H, J=6.5, 4.3 Hz), 0.30 (dd, 1H, J=14.8, 11.0 Hz).
Anal. Found: C, 72.69; H, 5.73. Calc. for C30H27O4NSi:
C, 73.00; H, 5.51%.
4.4.5. E-(1R*,2R*)-1-Methoxycarbonyl-1-
(2-phenylethenyl)-2-(triethoxysilylmethyl)cyclopropane
Crude yellow oil, 313 mg (93%), unstable to column
chromatography. H-NMR (400 MHz) l 7.39 (d, 2H,
4.4.10. (1R*,2R*)-1-Methoxycarbonyl-1-phenyl-2-
(triphenylstannylmethyl)cyclopropane
Colorless oil, 240 mg (90%). H-NMR (360 MHz) l
7.48–7.16 (m, 20H), 3.52 (s, 3H), 2.11 (m, 1H), 1.70 (m,
1H), 1.55 (m, 1H), 1.00 (dd, 1H, J=6.3, 4.5 Hz), 0.55
(dd, 1H, J=14.5, 10.8 Hz). CIMS; m/z: 540.3 ([M+
1]).
1
1
J=7.6 Hz), 7.30 (t, 2H, J=7.6 Hz), 7.22 (t, 1H,
J=7.6 Hz), 6.66 (d, 1H, J=16.0 Hz), 6.30 (d, 1H,
J=16.0 Hz), 3.81 (q, 6H, J=7.2 Hz), 3.68 (s, 3H), 1.77
(m, 1H), 1.67 (dd, 1H, J=9.2, 4.8 Hz), 1.21 (t, 3H,
J=7.2 Hz), 1.11 (dd, 1H, J=6.8, 4.4 Hz), 0.83 (dd,
1H, J=15.2, 5.6 Hz), 0.49 (dd, 1H, J=15.2, 10.0 Hz).
13C-NMR (90 MHz) l 174.7, 137.0, 132.0, 128.4, 127.3,
126.2, 124.7, 58.4, 52.0, 31.1, 26.5, 20.6, 18.2, 9.7. IR
1718 cm−1. HRMS (CI, isobutane). Found: 379.1939.
Calc. for C20H31O5Si: 379.1932 [M+1].
Acknowledgements
E.T. thanks the Spanish Ministerio de Educacion y
Cultura for a grant (SAB1999-0001), and Professor
Jose Barluenga (University of Oviedo, Spain) and mem-
bers of his research group for their hospitality and
helpful discussions during the preparation of this
manuscript. We also thank the Department of Chem-
istry at University of South Florida for funds to carry
out this research.
4.4.6. E-(1R*,2R*)-1-(Methoxycarbonyl)-1-
(2-phenylethenyl)-2-(tributylstannylmethyl)cyclopropane
1
Colorless oil, 418 mg (93%). H-NMR (360 MHz) l
7.42 (d, 2H, J=7.6 Hz), 7.31 (t, 2H, J=7.6 Hz), 7.23
(d, 1H, J=7.6 Hz), 6.67 (d, 1H, J=16.0 Hz), 6.31 (d,
1H, J=16.0 Hz), 3.69 (s, 3H), 1.84 (m, 1H), 1.70 (dd,
1H, J=9.0, 4.3 Hz), 1.46 (m, 6H), 1.28 (m, 9H), 1.03
(dd, 1H, J=6.8, 4.7 Hz), 1.92 (m, 1H), 0.87 (m, 12H),
0.69 (dd, 1H, J=12.6, 10.8 Hz). 13C-NMR (90 MHz) l
174.6, 131.8, 128.6, 127.4, 126.3, 126.1, 124.8, 52.5,
33.1, 32.0, 29.2, 27.1, 22.0, 13.8, 9.2, 7.8. IR 1723
cm−1. CIMS; m/z: 506.3 ([M+1]).
References
[1] Reviews on rhodium carbenoid cyclopropanations: (a) M.P.
Doyle, D.C. Forbes, Chem. Rev. 98 (1998) 911;
(b) M.P. Doyle, Chem. Rev. 86 (1986) 919;
(c) M.P. Doyle, Acc. Chem. Res. 19 (1986) 348;
(d) G. Maas, Top. Curr. Chem. 137 (1988) 35.

Y.-L. Lin, E. Turos / Journal of Organometallic Chemistry 630 (2001) 57–66
65
(b) H.M.L. Davies, D.M. Clark, T.K. Smith, J. Org. Chem. 56
(1991) 3817;
[2] M.P. Doyle, D. van Leusen, W.H. Tamblyn, Synthesis (1981)
787.
(c) T.R. Hoye, W.S. Richardson, J. Org. Chem. 54 (1989) 688.
[13] M.P. Doyle, V. Bagheri, T.J. Wandless, N.K. Harn, D.A.
Brinker, C.T. Eagle, K.-L. Loh, J. Am. Chem. Soc. 112 (1990)
1906.
[14] (a) S.G. Wierschke, J. Chandrasekhar, W.L. Jorgensen, J. Am.
Chem. Soc. 107 (1985) 1496;
[3] (a) M.P. Doyle, J.H. Griffin, V. Bagheri, R.L. Dorow,
Organometallics 3 (1984) 53;
(b) M.P. Doyle, R.L. Dorow, W.E. Buhro, J.H. Griffin, W.H.
Tamblyn, M.L. Trudell, Organometallics 3 (1984) 44.
[4] (a) H.M.L. Davies, D.M. Clark, T.K. Smith, Tetrahedron Lett.
26 (1985) 5659;
(b) R. Calas, J. Organomet. Chem. 200 (1980) 11;
(c) R. Calas, J. Dunogues, J. Organomet. Chem. Libr. 2 (1976)
277.
(b) H.M.L. Davies, T.J. Clark, L.A. Church, Tetrahedron Lett.
30 (1989) 5057;
(c) W.R. Cantrell Jr., H.M.L. Davies, J. Org. Chem. 56 (1991)
723;
(d) H.M.L. Davies, T.J. Clark, H.D. Smith, J. Org. Chem. 56
(1991) 3817;
(e) H.M.L. Davies, E. Saikali, W.B. Young, J. Org. Chem. 56
(1991) 5696;
[15] T.G. Traylor, W. Hanstein, H.J. Berwin, N.A. Clinton, R.S.
Brown, J. Am. Chem. Soc. 93 (1971) 5715.
[16] (a) J. Bartl, S. Steenken, H. Mayr, J. Am. Chem. Soc. 113 (1991)
7710;
(b) G. Hagen, H. Mayr, J. Am. Chem. Soc. 113 (1991) 4954;
(c) H. Mayr, G. Hagen, J. Chem. Soc. Chem. Commun. (1989)
91;
(f) H.M.L. Davies, T.J. Clark, G.F. Kimmer, J. Org. Chem. 56
(1991) 6440.
(d) H. Mayr, R. Pock, Tetrahedron 42 (1986) 4211.
[17] (a) M.A. Brook, C. Henry, E. Jefferson, R. Juschke, T. Sebas-
tian, M. Tomaszewski, S. Wenzel, Bull. Soc. Chim. Fr. 132
(1995) 559;
[5] For related discussions on the mechanism of rhodium-catalyzed
cyclopropanation, see: (a) J.L. Maxwell, K.C. Brown, D.W.
Bartley, T. Kodadek, Science 256 (1992) 1544;
(b) S. O’Malley, T. Kodadek, Tetrahedron Lett. 32 (1991) 2445;
(c) K.C. Brown, T. Kodadek, J. Am. Chem. Soc. 114 (1992)
8336;
(d) S. O’Bannon, W.P. Dailey, J. Org. Chem. 54 (1989) 3096.
[6] For additional articles on various alkene substrates that have
been used in rhodium carbenoid cyclopropanations, see: (a) A.
de Meijere, T.-J. Schulz, R.R. Kostikov, F. Graupner, T. Murr,
T. Biefeldt, Synthesis (1991) 547;
(b) M.A. Brook, C. Henry, R. Jueschke, P. Modi, Synlett (1993)
97;
(c) M.A. Brook, M.A. Hadi, A. Neuy, J. Chem. Soc. Chem.
Commun. (1989) 957;
(d) See also Ref. [10].
[18] For studies on silyl/stannyl ligand effects in related types of
reactions, see: (a) J.S. Panek, A. Prock, K. Eriks, W.P. Giering,
Organometallics, 9 (1990) 2175;
(b) E. Piers, N. Moss, Tetrahedron Lett. 26 (1985) 2735;
(c) A.J. Anciaux, A.J. Hubert, A.F. Noels, N. Petinot, P.
Teyssie´, J. Org. Chem. 45 (1980) 695;
(d) A.J. Hubert, A.F. Noels, A.J. Anciaux, P. Teyssie´, Synth.
Commun. (1976) 600.
(b) S. Sato, M. Kira, H. Sakurai, J. Am. Chem. Soc. 111 (1989)
6429;
(c) Z.Y. Wei, J.S. Li, D. D. Wang, Tetrahedron Lett. 28 (1987)
3441;
(d) A. Hosomi, S. Kohra, Y. Tominga, J. Chem. Soc., Chem.
Commun., (1987) 1517.
[7] S.R. Wilson, P.A. Zucker, J. Org. Chem. 53 (1988) 4682.
[8] (a) A.J. Lucke, D.J. Young, Tetrahedron Lett. 32 (1991) 807;
(b) I. Ryu, A. Hirai, H. Suzuki, N. Sonoda, S. Murai, J. Org.
Chem. 55 (1990) 1409;
[19] An alternative explanation for the poor stereochemical control in
these reactions may be that there is formation of a stable
zwitterionic intermediate ii, whose lifetime may be sufficiently
long to permit CꢀC bond rotation to iii to lead to the other
cyclopropane diastereomer. This possibility seems unlikely, how-
ever, since reactions carried out in polar versus non-polar sol-
vents, or in the presence of nucleophilic trapping agents, failed to
alter product ratios or afford any demetalated side products
derived from such zwitterionic structures.
(c) I. Ryu, H. Suzuki, S. Murai, N. Sonoda, Organometallics 6
(1987) 212;
(d) M. Ochiai, K. Sumi, E. Fujita, Chem. Pharm. Bull. 31 (1983)
3931;
(e) M. Grignon-Dubois, J. Dunogues, R. Calas, Can. J. Chem.
59 (1981) 802;
(f) M. Grignon-Dubois, J. Dunogues, R. Calas, J. Chem. Res.
(S) (1979) 6;
(g) T.H. Chan, I. Fleming, Synthesis (1979) 761.
[9] Y.-L. Lin, E. Turos, J. Am. Chem. Soc. 121 (1999) 856.
[10] Brook and Nuey have suggested that alkoxy ligands enhance the
b-cation stabilizing effect of silicon by a through-space stabiliz-
ing interaction between the oxygen n-orbital with the empty
p-orbital, as shown for structure i (M.A. Brook, A. Neuy, J.
Org. Chem., 55 (1990) 3609). This may lead to an increased
susceptibility of the cyclopropane ring toward protonation and
opening.
[11] M.E. Alonso, M. Gomez, Tetrahedron Lett. 20 (1979) 2763.
[12] (a) H.M.L. Davies, D.K. Hutcheson, Tetrahedron Lett. 34
(1993) 7243;

66
Y.-L. Lin, E. Turos / Journal of Organometallic Chemistry 630 (2001) 57–66
[20] Previous studies on the rhodium acetate-catalyzed cyclopropana-
tion of monosubstituted alkenes with diazoacetates indicate that
trans:cis selectivity can be increased by enlarging the steric
dimensions of the alkene and ester substituents. Generally, these
steric effects are limited to cases where the two groups are highly
branched. See: M.P. Doyle, V. Bagheri, T.J. Wandless, N.K.
Harn, D.A. Brinker, C.T. Eagle, K.L. Loh, J. Am. Chem. Soc.
112 (1990) 1906.
[21] For representative examples and discussions of the open transi-
tion state model applied to allylsilane and allylstannane addi-
tions to carbonyl compounds, see: (a) S.E. Denmark, E.J.
Weber, J. Am. Chem. Soc., 106 (1984) 7970;
(d) Y. Yamamoto, H. Yatagai, Y. Ishikawa, N. Maeda, K.
Maruyama, Tetrahedron 40 (1984) 2239;
(e) Y. Yamamoto, H. Yatagai, Y. Naruta, K. Maruyama, J. Am.
Chem. Soc. 102 (1980) 7107;
(f) H. Sakurai, Pure Appl. Chem. 54 (1982) 1;
(g) R.W. Hoffmann, Angew. Chem. Int. Ed. Engl. 21 (1982) 555;
(h) A. Hosomi, H. Sakurai, Tetrahedron Lett. 17 (1976) 1295;
(i) G.E. Keck, D.E. Abbott, E.P. Boden, E.J. Ehnolm, Tetrahe-
dron Lett. 25 (1984) 3927;
(j) W.R. Roush, K. Ando, D.B. Powers, A.D. Palkowitz, R.L.
Halterman, J. Am. Chem. Soc. 112 (1990) 6339.
[22] This effect can apparently be reduced when highly electron-rich
alkenes (e.g. enol ethers) are employed as substrates, giving
lower trans:cis selectivities than those of unactivated alkenes (e.g.
1-hexene). Refer: H.M.L. Davies, B. Hu, J. Org. Chem. 57
(1992) 3186.
(b) S.E. Denmark, E.J. Weber, J. Helv. Chim. Acta. 66 (1983)
1655;
(c) T. Hayashi, K. Kabeta, I. Hamaachi, M. Kumada, Tetrahe-
dron Lett. 24 (1983) 2865;
.