Formal intramolecular (4 + 1)-cycloaddition of dialkoxycarbenes: Control of the stereoselectivity and a mechanistic portrait

Article abstract of DOI:10.1021/ja211927b

The stereoselective synthesis of 5-5, 6-5, and 7-5 fused O-heterocyclic compounds is reported. The key reaction is a formal intramolecular (4 + 1)-cycloaddition involving a dialkoxycarbene and an electron-deficient diene where the stereoselectivity is dependent on the length of the tether. An analysis of the stereochemical outcome of this reaction sheds light on its complex mechanistic picture. High-level calculations were used to support the proposed mechanistic portrait.

Full text of DOI:10.1021/ja211927b

Article
pubs.acs.org/JACS
Formal Intramolecular (4 + 1)-Cycloaddition of Dialkoxycarbenes:
Control of the Stereoselectivity and a Mechanistic Portrait
Francis Beaumier, Marianne Dupuis, Claude Spino,* and Claude Y. Legault*
́ ́ ́
Departement de Chimie, Universite de Sherbrooke, 2500 Boul. Universite, Sherbrooke, Quebec, Canada J1K 2R1
S
* Supporting Information
ABSTRACT: The stereoselective synthesis of 5−5, 6−5, and 7−5
fused O-heterocyclic compounds is reported. The key reaction is a
formal intramolecular (4 + 1)-cycloaddition involving a dialkox-
ycarbene and an electron-deficient diene where the stereo-
selectivity is dependent on the length of the tether. An analysis
of the stereochemical outcome of this reaction sheds light on its
complex mechanistic picture. High-level calculations were used to
support the proposed mechanistic portrait.
cyclopropanation of 1,3-diene over the (4 + 1)-cycloaddition,
Hudlicky,12 Danheiser,13 and others set the foundation for the
stereoselective synthesis of cyclopentenes.14 The incorpora-
tion of one carbon unit such as carbon monoxide,15 isocy-
anides,16 diazomethane,17 ylides,18 various carbenoids,19 or
nucleophilic carbenes6f,8,20,21 into 1,3-dienes have equally met
with some degree of success to yield (4 + 1)-cycloadducts in
good yields.
INTRODUCTION
■
Since the discovery of the Diels−Alder reaction in 1928,1 the
[4 + 2]-cycloaddition has gained tremendous popularity among
synthetic chemists because it gives rise to complex six-
membered carbo- or heterocycles from structurally simpler
starting materials by the direct formation of two C−C bonds
with control of up to four stereocenters.2 Its reliability and the
easy prediction of its structural and stereochemical outcome
have helped make this reaction an essential tool in organic
synthesis.
Using Warkentin’s 2,2-dialkoxy-5,5-dimethyl-Δ3-1,3,4-oxadiazo-
lines as a source of dialkoxycarbenes,22 our research group
reported the discovery of a (4 + 1)-annulation that demon-
strated the ability of nucleophilic carbenes to stereoselectively
add to electron-deficient dienes, either intermolecularly or intra-
molecularly.8,20 The intermolecular version gave modest
yields and worked best when the diene was activated by the
presence of two electron-withdrawing groups (Scheme 1).
Like its six-membered ring counterpart, the concerted (4 + 1)-
cycloaddition3 between a carbene and a diene is a cheletropic
reaction allowed by the frontier molecular orbitals theory.4 In
principle, this reaction enables the stereoselective synthesis of
cyclopentene derivatives with control of up to three stereocenters.
In practice, however, examples of concerted (4 + 1)-cycloadditions
are very scarce in the literature and are not general, usually
involving one pair of a particular carbene and diene.5 The high and
sometime unusual reactivity6 of free carbenes may explain why this
transformation has been little exploited in the field of organic
chemistry. Another reason is the difficulty in controlling the
chemoselectivity of carbenes for the cycloaddition reaction,
particularly given their known propensity to preferentially give
cyclopropane products instead of (4 + 1)-cycloadducts when
reacting with 1,3-dienes.7−9 Fisher carbene complexes give
mostly cyclopropanation products with electron-poor dienes.10
Given the ubiquity of five-membered rings in natural and
pharmaceutical products,11 synthetic chemists have dedicated
much attention toward developing alternative methodologies to
circumvent the important limitations of the concerted (4 + 1)-
cycloaddition, namely, the poor yields of cycloadducts obtained
and the low tolerance of the carbene for spectator functional
groups.5
Scheme 1. Examples of Stereoselective Intermolecular
(4 + 1)-Cycloadditions
Among the alternative methods, the cyclopropanation of 1,3-
dienes followed by a vinylcyclopropane rearrangement emerged as
one of the most effective formal (4 + 1)-annulation strategies. By
using to their advantage the apparent predilection of carbenes for
Received: January 6, 2012
Published: March 11, 2012
© 2012 American Chemical Society
5938
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Scheme 2. Examples of Stereoselective Intramolecular (4 + 1)-Cycloadditions
The intramolecular reaction proved far more promising,
affording good to high yields of bicyclic O-heterocycles 10a−
15a and showing good potential for the efficient synthesis
of 3-cyclopenten-1-ones 16 or 17, where the stereoselectivity
at the 1- and 4-positions is controlled by the length of the
tether (Scheme 2).
Scheme 3. Formal (4 + 1)-Cycloadditions of
Dimethoxycarbene
We report herein an in-depth study of the scope and
mechanism of this unique intramolecular (4 + 1)-cycloaddition
involving dialkoxycarbenes and electron-deficient dienes. Our
study encompasses a wide range of parameters that were varied
with the aim of acquiring knowledge on the generality,
usefulness, and stereochemical predictability of the reaction.
The observed stereoselectivities and the proposed mechanistic
picture of the (4 + 1)-cycloaddition were also supported by
computational calculations.
The most striking aspect of the reactions shown in Scheme 2
was the complete reversal of stereoselectivity observed in going
from the 5−5 (10a:11a = 5:95) to the 7−5 fused bicyclic
system (14a:15a = >99:1), with the 6−5 fused bicyclic system
(12a:13a = 70:30) being an intermediate case. Clearly, we
needed to better understand the reasons behind this behavior in
order to better predict the stereochemical outcome of the
reaction. Although we had suggested in a previous publication
that a change in mechanism could be at the origin of this
selectivity reversal,8 the present study provides mechanistic
insights that speak to the contrary.
Generation and Prior (4 + 1)-Cycloadditions of
Dialkoxycarbenes. Since the first use of dimethoxycarbene
(DMC) 22 by Hoffman and co-workers in the 1960s,23
dialkoxycarbenes have been known to act as nucleophiles and
to react with various electrophiles such as carbonylated com-
pounds (anhydrides, acyl chlorides, ketones, esters, ketenes,
and isocyanates), thiocarbonyles, and imines and with
various alkenes, alkynes, and cumulenes.22,24 Few formal
(4 + 1)-cycloadditions with DMC have been reported and
involve tropone 19 or tetraphenylcyclopentenedione 20,25
substituted bis-ketenes 24,26 and vinylisocyanates 25
(Scheme 3).27 Yet, carbene 22 gave cyclopropane derivatives
in low yield when reacted with 1-phenyl- or 1,1-diphenyl-1,3-
butadiene.25
trivial. Warkentin’s oxadiazolines, such as 1 or 7a−9a, have greatly
simplified this task for the generation of dialkoxycarbenes: they are
easily prepared from simple precursors (vide infra) and produce
only volatile and innocuous byproduct (N2 and acetone).22
Moreover, the nature of the oxadiazoline can be easily changed to
alter the thermolysis temperature.30
To illustrate how oxadiazolines are easily prepared, a typical
example is given in Scheme 4. In the presence of a catalytic
amount of acid, primary alcohols will substitute the acetate on
29 in good yield, as was the case of alcohols 27 and 28,
affording the corresponding oxadiazoline in 87% and 90%
yields, respectively. Besides being a reliable source of the
dialkoxycarbene, the oxadiazoline also proved to be a tolerant
functional group. It withstands temperatures of up to 80−85 °C
and is unreactive toward many nucleophilic or basic reagents.
For more details on the complete synthesis of all precursors,
see the Supporting Information.
RESULTS AND DISCUSSION
■
Dienes tethered to the oxadiazoline moiety with a three-carbon
chain (8) gave 6−5 bicyclic products 12 and 13 upon
thermolysis (cf. Scheme 2). They were the starting point of
our investigation on the effect of the nature and length of the
tether on the stereochemical outcome of the reaction. We first
revisited the thermolysis of diene 8a in more details because its
moderate diastereoselectivity contrasted with the high stereo-
selectivities obtained with the two- and the four-carbon
tethered dienes 7a and 9a, respectively.20 Note that in all
observed products 10a−15a the relative stereochemistry of the
ring fusion is cis.31
Fundamental studies and synthetic applications of dialkox-
ycarbenes have been constrained for a long time by the difficulty in
generating them. The thermolysis of norbornadiene derivatives28
or the thermolysis or photolysis of 3,3-dialkoxydiazirines29 proved
problematic, either because they produce large amounts of
byproduct or because they are known to be explosive.
Furthermore, the synthesis of these carbene precursors is not
5939
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Scheme 4. Typical Preparation of a Cycloadduct Precursor
Scheme 5. Attempted Formation of Oxabicyclic Compounds 32 and 35
After careful re-examination of the reaction conditions,
cycloadducts 12a and 13a were obtained with a significant
increase in yield and a slight increase in ratio using pyridine as
an additive (84% vs 61%, 14a:15a = 70:30 vs 75:25). These
reaction conditions also gave higher yields for the 5−5 (86% vs
85%, 10a:11a = 5:95) and the 7−5 oxabicyclic compounds
(79% vs 70%, 14a:15a = >99:1).
Table 1. Effect of the Nature of the Electron-Withdrawing
Group on the Intramolecular (4 + 1)-Cycloaddition
The (4 + 1)-cycloaddition could not be extended to the
formation of 4−5 or 8−5 oxabicyclic compounds (Scheme 5).
Thermolysis of diene 30 failed to give any of the desired
product 32 and gave ester 31 as the only identifiable product in
35% yield.32f
The formation of the unstable dimer 34 (70%33) in the case
of diene 33 (Scheme 5) is somewhat surprising in light of the
efficiency and yield with which the 7−5 bicyclic compound 14a
was formed. Entropy is probably to blame for this difference,
but we were nevertheless expecting at least some product 35 to
be formed given that even intermolecular reactions give 20−
65% of the desired cycloadducts.8,20
Remarkably, many activating groups proved to be
compatible with the high reactivity of dialkoxycarbenes as
shown by the good yields of the expected (4 + 1)-
cycloadducts obtained upon thermolysis of dienes 8a−h
(Table 1, entries 1−5 and 8). The conversion of aldehyde 8d
into products 12d and 13d was excellent (Table 1, entry 4),
though their isolated yields were low due to their instability.
Compounds 12d and 13d easily shed methanol to give
cyclopentadiene 36 (Figure 1).
a
All reactions were performed on 0.25−0.50 mmol scale at
0.01−0.04 M with 25 mol % of pyridine. Condition A: PhMe,
111 °C, 20 h. Condition B: PhCl, 132 °C, 5 h. Condition C:
PhMe, 160 °C (sealed tube) 15 h. Isolated yield. Determined
by H NMR analysis of the crude reaction mixture. Product 36 was
also isolated in 12% yield after flash chromatography Conversion
determined by H NMR spectroscopy. Product 37 was also obtained
in 17% proportion as measured from the H NMR of the crude reac-
tion mixture.
b
c
d
1
e
f
1
1
The only low-yielding (4 + 1)-cycloaddition resulting from
the thermolysis of nitrodiene 8f (Table 1, entry 6) was likely
due to the acidity of the allylic protons34 coupled with the
basicity of the carbene (estimated pKa values for dimethox-
ycarbene = 11−16).35 This gives rise to competing acid−base
reactions.36 We were unable to identify any specific insertion
products in the present case, but we had shown in the past that
dimethoxycarbene inserts into the acidic proton of nitro-
methane.37
The transient dialkoxycarbene generated from the thermol-
ysis of amide 8g led to a complex mixture of compounds, of
which 12g and 13g were isolated in moderate yield (Table 1,
entry 7). Surprisingly, product 37 (Figure 1) was observed in
5940
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Table 2. Formation of Cyclopropanes 38g−j from Dienes
8g−j
Figure 1. Side products resulting from the thermolysis of dienes 8d
and 8g.
significant amounts (17%33).38 The Z configuration of the
acyclic double bond in 37 was established from the observed
coupling constant between the olefinic protons (11.5 Hz). At
this point, it is hard to explain the difference in reactivity of
amide 8g compared to ester 8a, but it seems that amides are in
general a poor choice of activating group for this cyclo-
addition.39
a
All reactions were performed on 0.25−0.50 mmol scale at 0.01−0.04
The volatility of 12h and 13h is thought to be in part
responsible for the loss in yield of the thermolysis of
trifluoromethyl-substituted diene 8h (Table 1, entry 8).40 Its
M with 25 mol % of pyridine. Condition A: toluene, 111 °C, 20 h.
b
Condition B: PhCl, 132 °C, 5 h. Conversion determined by 1H NMR
spectroscopy. Any attempts of purification resulted in complete
degradation of the product. After flash chromatography on silica gel.
1
c
conversion was judged to be much higher by H NMR. The
high observed stereoselectivity illustrates the potential of
inductivity-based (as opposed to resonance-based) electron-
withdrawing groups to activate a diene moiety toward
cycloaddition with a nucleophilic carbene. Given the increasing
importance and impact of fluorinated compounds in the
pharmaceutical, agrochemical, and materials industries,41 the
(4 + 1)-cycloaddition may prove useful to diastereoselectively
install a trifluoromethyl group on a five-membered ring.
From all of the cases shown in Table 1, a trend emerges: the
stronger the electron-withdrawing group on the diene, the
lower the diastereoselectivity (resonance-based electron-with-
drawing ability: CF3 < SO2Ph/CN < CONEt2/CO2Me <
RCOR < RCHO < RNO2). The trifluoromethyl group is the
least capable of stabilizing a negative charge and gives the
highest diastereoselectivity, while modest selectivities are
obtained with the aldehyde and nitro functionalities.
carbonyl compounds. However, thermolysis of diene 8j
afforded only cyclopropane 38j. Note that heating either 38i
or 38j to 160 °C did not lead to any amount of the
corresponding (4 + 1)-cycloadducts.
Moving the electron-withdrawing group to the 2-position
(dienes 39 and 41) led surprisingly to the sole formation of
5−3 and 6−3 bicyclic compounds 4033 and 42 as single
diastereomers in excellent yields (Scheme 7). However,
lengthening the tether (diene 43) afforded a 6% yield of (4 + 1)-
cycloadduct 45 along with cyclopropane 44. It should be noted
that the trans-fusion between the seven- and five-membered
rings in compound 45 is observed for the first time after dozens
of cycloadditions we have carried out. The implication of the
observed stereoselectivity will be discussed in the mechanistic
section of the paper.
Cyclopropanes 38g−j, 40, 42, and 44 are relatively stable
compounds, interesting both from a structure and reactivity
standpoint.43 This intramolecular process could provide a new
efficient way to construct dialkoxyvinylcyclopropanes that
would subsequently undergo vinylcyclopropane-cyclopentene
rearrangements (to give formally (4 + 1)-cycloadducts),14l,44
[5 + 2] cycloadditions,45 radical-mediated ring-opening
reactions,46 and a host of other reactions that we will
investigate in due course.
By contrast, the stereoselectivity of reactions leading to 7−5
bicyclic compounds remained unaffected by the variation of the
nature of the electron-withdrawing group. Using optimized
reaction conditions, all carbene precursors shown in Scheme 6
nicely gave >95% of 14a−c as single diastereomers.
Scheme 6. Effect of the Nature of the Diene’s Electron-
Withdrawing Group on 7−5 Bicyclic Systems
Transposing the electron-withdrawing group at the 3-position of
the diene had a profound effect on the stereochemical results
(Scheme 8). First, cycloadducts 47 and 49 were obtained in good
yields with complete stereoselectivity starting from dienes 46 and
48, respectively. While the isolation of a single diastereomer 47
was consistent with the stereochemical outcomes shown in
Scheme 2 (1-substituted diene 7a), the isolation of 49 as the only
adduct was quite stunning in two respects: a mixture of isomers
was anticipated, and diastereomer 49 was expected to be a minor
product.
Thermolysis of diene 50 led to the formation of three
compounds, which we identified as cycloadducts 51a, 51b, and
52, in 20%, 34%, and 17% yields, respectively (Scheme 8).
Higher structural complexity is always desirable, and it is of
synthetic relevance that chiral quaternary carbon centers could
be created on the cyclopentene scaffold with stereoselectivities
superior to the analogue 8a. We were delighted to find that
thermolysis of 53 resulted in the formation of the
That at least one electron-withdrawing group is required for
the (4 + 1)-cycloaddition to proceed was also unambiguously
established. When electronically unactivated diene 8i was
thermolyzed, cyclopropane 38i was isolated in 81% yield.
Interestingly, at lower temperature, dienes 8g−h led to the
formation of isolable bicyclic cyclopropanes 38g−h (Table 2,
entries 1−2). Inspired by the work of Carboni, Carreaux, and
Hall,42 we thought that a successful cycloaddition reaction of 8j
would have led to a (4 + 1)-cycloadduct bearing an allylic
boronic ester that could further participate in allylboration of
5941
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Scheme 7. Thermolysis of 39, 41, and 43, Having an Electron-Withdrawing Group at the 2-Position
Scheme 8. Thermolysis of 46, 48, and 50, Having an Electron-Withdrawing Group at the 3-Position
corresponding cycloadducts 54 and 55 in decent yield and in a
ratio of 81:19 (Scheme 9). We were also excited, despite a
moderate 44% yield, about the highly stereoselective formation
of 57 from oxadiazoline 56. In addition, the corresponding
cyclopropane 59 was isolated in 12% yield as a 70:30 mixture of
stereoisomers at C-3. As depicted in scheme 9, the temperature
required to accomplishing the (4 + 1)-annulation depends on
the position of the methyl group.
Mechanistic Studies. Mechanistic studies of this (4 + 1)-
cycloaddition is unfortunately complicated by the fact that the
rate-determining step is the ring opening of the 2,2-dialkoxy-
5,5-dimethyl-Δ3-1,3,4-oxadiazoline 60 in the generation of
carbene 61 (Scheme 10).47 Once generated, the highly reactive
singlet48 carbene 61 reacts immediately with the diene moiety,
one way or another, to eventually lead to the formation of the
(4 + 1)-cycloadducts 62. The concentration of the free carbene
during the thermolysis is thus likely to be low.
We elected to use stereochemistry as a probe to gain
mechanistic insights on this reaction. We are well aware that the
stereochemistry of the product cannot be used to draw a complete
picture of the mechanism of the transformation of 61 into 62.
However, coupled with evidence of the presence or the absence of
key intermediates, it can help answer questions about the
concertedness of bond-forming and bond-breaking events.
5942
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Scheme 9. Formation of Cyclopentenes with Chiral Quaternary Carbons
Scheme 10. Thermolysis of 2,2-Dialkoxy-5,5-dimethyl-Δ3-1,3,4-oxadiazolines 60 and Formation of Cyclopentenes 62
At the onset of our studies on the intramolecular (4 + 1)-
annulation of electron-rich carbenes with electron-deficient
dienes, we considered all mechanistic possibilities that allowed
the formation of bicyclic cyclopentenes 62 from the
corresponding carbene 61. Although examples of concerted
(4 + 1)-cycloadditions are quite scarce in the literature, we
believed that a concerted mechanism remained possible
(Scheme 11, top). This cheletropic reaction is allowed
according to the Frontier Molecular Orbital theory49 and
would produce either the cycloadduct 62a or 63 from transition
states TS61a or TS61b, respectively.
Another accessible mechanism is the isomerization of a
vinylcyclopropane to the corresponding cyclopentene, the
product of a formal (4 + 1)-cycloaddition. The dialkoxycarbene
61 could react stereospecifically with the alkene,50 via transition
state TS-61c, to give vinylcyclopropane 64, which could then
rearrange via three distinct pathways to yield (4 + 1)-
cycloadducts 62a, 62b, or 63 (Scheme 11, center).14 If
intermediate 64 participates in a concerted [1,3]-sigmatropic
rearrangement with inversion of configuration at the migrating
center, adduct 63 will result. Conversely, the collapse of the
zwitterion 65 would provide cycloadducts 62a or 62b.51 Note
that the rearrangement of the intermediate resulting from a
radical ring opening of 64, albeit possible,52 is unlikely to
operate due to the donor−acceptor nature of the studied
vinylcyclopropanes.53 Finally, a complete ionic mechanism
(61 → 65 → 62) must also be considered as a possible competing
pathway for the (4 + 1)-cycloaddition (Scheme 11, bottom).
Any one or a mixture of the proposed pathways for each
substrate may be operative and thus lead to diastereomeric
cyclopentenes 62a, 62b, or 63. We will show that different
substrates follow a complex array of possible and energetically
close mechanistic pathways that are greatly influenced by the
architecture of the substrate. The following discussion on the
mechanism is divided into sections according to the position of
the electron-withdrawing group on the diene.
As mentioned in the previous section, dienes with a methyl
ester at position 1 experience a profound change in
stereoselectivity when the length of the chain tethering the
carbene to the diene is varied. Indeed, a complete reversal of
stereoselectivity was observed in going from the 5−5 (10a:11a)
to the 7−5 fused ring system (14a:15a) with the 6−5 fused
ring system (12a:13a) being an intermediate case (Scheme 12).
Each pair of diastereomeric cycloadducts are stereochemically
stable, either upon further heating or when resubmitted to the
initial reaction conditions. Therefore, we can conclude that, at
the very least, the last step in the overall mechanism is
irreversible. We can confirm that this is the case for all
cyclopentene adducts found in the present study, and this
important issue can be considered solved.
Cycloadduct 11a (corresponding to 62b in scheme 11) was
clearly not produced by a concerted (4 + 1)-cycloaddition, but
what of cycloadduct 14a (corresponding to 62a in Scheme 11)?
Could the lengthening of the tether provoke a change from an
ionic to a concerted mechanism? Or did the tether length simply
influence the selectivity via conformational biases? Lastly, are
vinylcyclopropane intermediates 64a−c involved, and if so, what
role do they play in this diastereoselective process?
We started by tackling the question of whether intermediate
cyclopropanes 64a−c were involved in the overall process or
not. For the transformation of 7a−9a to 10a−15a under the
conditions shown in Scheme 12, cyclopropanes were never
detected. A computational study of all the available reaction
pathways for 7a−9a was carried out, and the formation of a
cyclopropane intermediate was found to be favored by
calculations (Figure 2). The common reaction solvent is
5943
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Scheme 11. Mechanistic Pathways for the (4 + 1)-Cycloaddition of Carbenes 61
Scheme 12. Stereoselective Intramolecular (4 + 1)-Cycloadditions of Dienes 7a−9a
toluene. Due to its nonpolar nature, we considered that
optimization of the structures in the gas phase was adequate. A
thorough analysis of the reaction pathways did not lead to any
concerted (4 + 1)-cycloaddition transition structure. Analysis of
the process gave evidence of the initial formation of the
vinylcyclopropane intermediate. It was found that the reaction
proceeds by an initial conjugate addition of the carbene on C-4 of
the diene ester moiety (Figure 2 top).54 In all cases, this addition
was found to occur with a preference of 2 kcal/mol on the s-trans
conformation of diene 7a−9a. The formed adducts then preclude
the direct formation of (4 + 1)-cycloadducts 10a−15a, as rotation
around the C2−C3 bond is strongly disfavored due to the
delocalization of the anionic ester moiety.55
Analysis of the vibrational mode of the imaginary frequency
clearly shows a unique bond-forming event (conjugate
addition) in transition structures TS65a−c. IRC calculations
were carried out to determine if a stable zwitterionic
intermediate could be found following the addition. In all
5944
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Figure 2. B3LYP/6-311++G(d,p)//B3LYP/6-31+G(d) optimized conjugate addition TSs and PES inflection points obtained from IRC calculations
on respective TSs. Relative energies reported in kcal/mol with respect to the corresponding most stable conformation of the free carbene.
Scheme 13. Generation of Cyclopropanes 64a−c at Low Temperature
these cases, a rapid collapse to the corresponding vinyl-
cyclopropanes 64a−c was obtained. These are clear examples
of two-step no-intermediate processes.56 Inspection of the
inflection point that follows the initial transition structure
showed structures IP64a−c with a partially formed bond
between Ccarbene and C-3, leading to the cyclopropane.
Optimization of these transition structures using a solvation
model (i.e., PCM/Toluene) resulted in almost identical
structures and no stable intermediates on the calculated IRC
paths. This is an evidence of the reliability of gas phase
calculations for this study.
We thus turned to a modified version of these oxadiazolines,
namely, the 2,2-dialkoxy-5-methyl-5-(p-methoxy)phenyl-Δ3-
1,3,4-oxadiazolines 66a−c, which undergo cycloreversion at
approximately 50 °C (Scheme 13).30b−d When dienes 66a−c
were heated in toluene at temperatures ranging from 50 to 80 °C,
the unstable and moisture-sensitive 64a−c were observed
and characterized as single diastereomers. Importantly, heating
these cyclopropanes in toluene converted them into the
corresponding (4 + 1)-cycloadducts 10a−15a with almost
identical stereoselectivities as the reactions reported in Schemes
2 and 12. Cyclopropane 64a readily rearranged at 50 °C, while
64b and 64c necessitated higher rearrangement temperatures.
The reactivity trend of cyclopropanes 64a−c seems to be
directly related to ring strain of the 5, 6 and 7−3 bicyclic
systems. As a control experiment, we thermolyzed 66b directly
With this in mind, we considered that the required
temperature for the thermolysis of the 2,2-dialkoxy-5,5-
dimethyl-Δ3-1,3,4-oxadiazolines 7a−9a (>100 °C) may have
been too high to allow the observation of cyclopropanes 64a−c.
5945
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Scheme 14. Thermolysis of Dienes Z,E-7a−9a
at 111 °C and observed nearly the same yield and ratio of
products 12a and 13a as for the thermolysis of 8a or 64b. Aside
from the small difference in yields, attributable to the relative
instability of dienes 66a−c and vinylcyclopropanes 64a−c at
high temperature, these experimental results are in accordance
with computational findings and support very strongly the
implication of cyclopropanes in the mechanism of formation of
(4 + 1)-cycloadducts.
rearrangement as a possible pathway for their formation (cf.
Scheme 11, center); (3) in the case of intermediates 64a and
67a (leading to 5−5 fused bicyclic systems) and 64c and 67c
(leading to 7−5 fused bicyclic systems), the cyclopropane ring
opens to a zwitterion (65a or 65c) and free bond rotation
occurs before its stereoselective collapse to the corresponding
cycloadduct 11a and 14a, respectively (Scheme 15, krot > k1
or k2).
We surmised that the stereochemistry of the vinyl-
cyclopropane intermediates 64a−c had a direct influence on
the stereochemistry of the (4 + 1)-cycloadducts. To prove or
disprove this hypothesis, we decided to examine the effect that
the double bond geometry in the starting diene would have on
the ratio of cycloadducts. Changing the geometry at the 3,4-
position of the diene should led to the cyclopropanes epimeric
to 64a−c, namely, cyclopropanes 67a−c (Scheme 14). The
thermolysis of Z,E-7a−9a gave interesting results. While both
Z,E-7a and Z,E-9a led to very similar ratios of cycloadducts as
per dienes E,E-7a and E,E-9a, respectively (cf. Scheme 12),
only diene Z,E-8a afforded a complete reversal of stereo-
selectivity. We repeated these experiments several times, and
the reported ratios were constant and reproducible. Moreover,
the isolation of cyclopropane 67c in 19% yield as a single
stereoisomer was evidence of the stereospecificity of the (2 + 1)-
cycloaddition reaction. This was the first time a dialkoxycyclo-
propane did not rearrange readily at 111 °C. Thermolyzing Z,E-
9a at a higher temperature (PhCl, 132 °C) gave 14a (38%) as
the only isolable bicyclic adduct. When comparing cyclo-
propanes 64c and 67c, it seems that the stereochemistry of the
cyclopropane intermediate has an important effect on its rate of
rearrangement.57
Yet, the selectivity switch in going from diene 7a to 9a still
puzzled us. A further conundrum was why a selectivity switch
was observed in going from E,E-8a to Z,E-8a (leading to 6−5
fused bicyclic systems) but not in the case of E,E-7a to Z,E-7a
or E,E-9a to Z,E-9a. A free equilibration of the different
rotamers of zwitterion 65b is excluded on the basis that the
same ratios of products should have been obtained regardless of
double bonds geometries in the starting diene. The only
possibility left is that the cyclopropanes 64b and 67b open to
different rotamers of zwitterion species 65b that collapse at a
rate competitive with bond rotation (krot < k1 or k2) (each may
open to give zwitterionic species E-65 (Scheme 15), but the
trans internal double bond prevents cyclization). Although
experimental data gave us many hints on the reaction
mechanism, we were still unable at this point to explain the
discrepancy in the stereochemical outcomes for the rearrange-
ment of 64a−c and 67a−c, the implication of their
corresponding zwitterions 65a−c, and the fundamental origin
of the stereoselectivities. We thus had recourse to high level
calculations.
Optimization of transition structures leading to both
cycloadducts 10a, 12a, 14a and 11a, 13a, 15a were done to
help us understand this rather drastic selectivity difference
(Scheme 15, k1 vs k2) for the 5−5, 6−5, and 7−5 fused bicyclic
systems with an ester at C-1. The existence of diradical species
was investigated, and they were found to be either nonexistent
(singlet) or highly disfavored (triplet). It was possible to find
and characterize well-defined polar transition structures for the
formation of all (4 + 1)-cycloadducts (Figure 3). These
transition structures possess relative free energies lower than
At this time, we can put forth the following hypotheses: (1)
the thermolysis of substrates bearing an electron-withdrawing
group at position 1 of the diene, led to the corresponding
carbenes, which undergo a stereospecific cyclopropanation to
give 64a−c or 67a−c depending on alkene geometry (Scheme 15);
(2) the fact that we observed cycloadducts 10a−15a with
cis ring junctions precludes a concerted vinylcyclopropane
5946
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Scheme 15. Competitive Rates for the Rearrangements of 64a−c and 67a−c
Figure 3. B3LYP/6-311++G(d,p)//B3LYP/6-31+G(d) optimized cyclization TSs, leading to 10a, 12a, 14a (a) and 11a, 13a, 15a (b). Relative
energies reported in kcal/mol with respect to the respective most stable conformation of the free carbene.
their related conjugate addition step (cf. Figure 2, TS65a−c).
This is evidence that the initial addition of the free carbene is
irreversible. A notable feature of all transition states in Figure 3
is their “early” character with the forming bond length ranging
from 2.83 to 2.97 Å. IRC calculations were performed on these
TS structures to ensure their connection to the final products
and determine if there was the existence of stable zwitterionic
intermediates (65a−c). TS10a and TS11a both led to a
common intermediate 65a through the IRC calculations, via a
barrier-less rotation process. A more thorough geometric
optimization of 65a, however, led to a very flat potential
energy surface near the transition states and an almost barrier-
less closure to the cyclopropane 67a. IRC of TS12a also shows
a very flat surface with no stable zwitterionic intermediate.
Interestingly, IRC of TS13a directly connects 67b to 13a, with
no intermediate. The same type of behavior was found for both
TS14a and TS15a, which show paths that directly connect 64c
to 14a and 67c to 15a, respectively. These results clearly
indicate the transient nature of intermediates 65a−c and the
possible dynamic effects that could prevent equilibration
between rotamers 1 and 2 of 65a−c (Scheme 15). They
could even be perceived as concerted [1,3] sigmatropic
rearrangements with retention of configuration. Such forbidden
processes have been previously rationalized by stabilization
though the “subjacent orbital effect”.58 However, in the cases
described in the current study, the stabilization can be
explained by charge delocalization, as observed by the highly
polar nature of the transition structures. This is reminiscent of
oxyanion-accelerated vinyl cyclopropanes rearrangements re-
ported by Danheiser et al.59
To evaluate this possibility, the free energy barriers for the
isomerization of vinylcyclopropanes 64a−c to the correspond-
ing cycloadducts were compared (Table 3). The barriers of
isomerization for 64a are lower than the corresponding barriers
for 64b−c. This is in good agreement with the experimental
results obtained for the reactions of 64a−c (cf. Scheme 13), as
64a readily converts to 11a at 50 °C, whereas 64b and 64c
need a higher temperature (111 °C) to react.
The correlation between theory and experiment for
selectivities observed in the rearrangement of 64a is very
good (Table 3, entries 1 and 2). The computed value of 2.1
kcal/mol, which predicts a 6:94 ratio for 10a and 11a, is almost
5947
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
compared to E,E-8a (Scheme 12). This is well represented by
the IRC of TS13a.55
Table 3. Computed B3LYP/6-311++G(d,p)//B3LYP/6-
31+G(d) Free Energy Barriers and Differences for the
Isomerization of Vinyl Cyclopropanes 64a−c to the
We have previously established that the stereoselectivity in
the thermolysis of analogues of E,E-8b−e, where the ester
group is replaced by other electron-withdrawing groups,
decreased when the electron-withdrawing power of the group
increased (cf. Table 1). We believe that activating groups better
at stabilizing the negative charge in zwitterion 65b favor the
equilibrium between the two rotamers (cf. Scheme 15), leading
to an overall less stereoselective process.
a
Corresponding Cyclopentenes
b
⧧
entry
TS
ΔGiso ΔΔG⧧
calcd ratio
exptl ratio
1
2
3
TS10a
TS11a
TS12a
27.6
25.5
30.5
+2.1
6:94 (10a:11a) 5:95 (10a:11a)
c
d
+0.6
31:69 (12a:13a) 75:25 (28:72)
(12a:13a)
4
5
6
TS13a
TS14a
TS15a
29.9
31.0
36.8
Another way to confirm the influence of the stabilization of
zwitterion 65b was to probe the effect of solvent on the
product ratio obtained from the thermolysis of diene E,E-8a.
Although many solvents or additives are incompatible with the
intermediate dialkoxycarbene and cannot be used (nitro-
methane,35,62 alcoholic solvents,36b and other nucleophilic
solvents63,64), a reversed ratio of 28:72 was obtained in
acetonitrile for 12a and 13a. This value is nearly identical to the
calculated one (cf. Table 3, entries 3 and 4). A more detailed
study of the effect of solvents and additives can be found in the
Supporting Information. These experiments support the notion
that the dynamic effect observed in toluene is not seen in polar
media: the increased stability of the zwitterion 65b decreases k1
and k2 to the benefit of krot and allows an equilibrium between
the different rotamers of 65b, which then collapse preferentially
to 13a (cf. Scheme 15). Conversely, heating E,E-7a and E-E-9a
in acetonitrile gave the same ratios of products as per in toluene
(24% and 21% yields,65 respectively). It appears that k1 and k2,
in the case of 65a and 65c, are lower than krot in polar or apolar
solvents.
The results obtained from the thermolysis of methyl-
substituted diene 56 at lower temperature (Scheme 9 vs
Scheme 16) further demonstrate that the stereoselectivity of
the (4 + 1)-cycloaddition is dictated by the difference in
cyclization rate of a zwitterion intermediate (like 65b) and the
rate of bond rotation (similar to Scheme 15, krot vs k1 and k2).
At 111 °C, only cyclopropane 59 was isolated as a 77:23
mixture of stereoisomers.54 When diene 59 or cyclopropane 59
was heated to 160 °C, bicyclic adducts 57 and 58 (44%, 57:58
= 94:6 ratio) were formed along with recovered cyclopropane
59 (12%) as a 70:30 mixture of diastereomers (cf. Scheme 9).
Epimerization of 59 at C-3 occurs presumably because the rate
of cyclization into bicyclic adducts 57 or 58 is slow and the
zwitterionic species may revert back to cyclopropane after bond
rotation.51,66 By opposition, the stereoselectivity observed for
the thermolysis of 53 (cf. Scheme 9, 67%; 54:55 = 81:19) is
similar to the one obtained from diene E,E-8a (75:25). It is
believed that the methyl substituent in a remote position from
the reacting center has a negligible effect on the stereo-
selectivity.
−5.8
>99:1 (14a:15a) >99:1 (14a:15a)
a
Free energy barriers calculated relative to the corresponding
b
vinylcyclopropanes 64a−c. Ratio observed in toluene (cf. Scheme 3).
c
d
⧧
ΔΔEZPE = −0.1 kcal/mol. Ratio observed in acetonitrile (see
discussion on solvent effect later in this article).
identical to the experimental ratio of 5:95 obtained at 111 °C,
⧧
which corresponds to a ΔΔGiso of 2.3 kcal/mol. Here, the
stereoselectivity can be explained by a small conformational
bias in the transition structures: in TS10a, the vinylogous
ester enolate chain is pseudo equatorial on the five-membered
ring, causing a torsion (22°) of the C1−C2 bond to accom-
modate orbital alignment to form the bond; at the opposite, in
TS11a, the vinylogous ester enolate moiety can remain planar
during the bond-forming event. Scan of the C1−C2 bond on a
model system indicates that this torsion can account for as
much as 1.7 kcal/mol, in good agreement with the observed
difference.
Additionally, the energy difference (5.8 kcal/mol) computed
for the isomerization of 64c predicts a ratio of >99:1, in good
agreement with experimental observations (Table 3, entries 5
and 6). Due to the extra degree of freedom of the seven-
membered ring, it is possible in TS14a to position the
vinylogous ester enolate in a pseudoaxial fashion, enabling
cyclization without any torsion in the π-system. In contrast, a
severe interaction developing between the C-1 vinylic proton
and the ring residue can be found in TS15a. Furthermore, the
seven-membered ring needs to adopt a higher energy
conformation to position the vinylogous ester enolate in a
pseudo equatorial fashion. Summation of both factors can
explain the observed stereoselectivity.
For the rearrangement of 65b, a competition of two factors
occurs: as was the case for TS10a, there is also a destabilizing
torsion of the C1−C2 bond in the vinylogous ester enolate
π-system of TS12a; in TS13a, a severe interaction is developing
between the C-1 vinylic proton and the axial hydrogen on the
six-membered ring. Both effects compete to lead to a small
energy difference between both transition structures (0.6 kcal/
mol). The zero-point energy barrier difference (ΔΔEZPE⧧) is
almost nonexistent (−0.1 kcal/mol). IRC calculation on TS12a
shows a very flat surface, whereas IRC on TS13a shows a direct
connection between 67b and 13a.55 By taking into account the
flat nature of the TSs, PES and the small energy difference
between TS12a and TS13a, we believe that we are observing
here a dynamic effect.60 On the basis of the least motion
principle,61 64b will lead preferentially to 12a and 67b will lead
to 13a, respectively. This would account for the difference
between the experimental (75:25) and predicted (31:69) ratios
(Table 3, entries 3 and 4). The slight energetic preference for
TS13a could even explain why a better selectivity is observed
experimentally for the reaction of Z,E-8a (Scheme 14)
Computational analysis of the reaction of 56 indicates that
the transition structures TS57 and TS58, leading to 57 and 58,
respectively, possess relative free energies higher than that of
the initial conjugate addition TS59 (Figure 4). This could
explain the epimerization of 59 and the lower yield of adducts
57 and 58. Indeed, the possible reversibility between the free
carbene, the zwitterion, and cyclopropane 59 would be
accompanied by alternative degradation pathways. The high
stereoselectivity, here, stems from a kinetic preference for TS57
due to the severe interaction between the vinylic methyl group
and the ring residue and the larger C1−C2 torsion on
vinylogous ester enolate π-system present in TS58 (Figure 4).
5948
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Scheme 16. Results from the Thermolysis of Diene 56 at Lower Temperature
Figure 5. B3LYP/6-311++G(d,p)//B3LYP/6-31+G(d) optimized (4 + 1)-
cycloaddition TSs leading to 45. Free energies reported in kcal/mol with
respect to TS45a.
Figure 4. B3LYP/6-311++G(d,p)//B3LYP/6-31+G(d) optimized
cyclization TSs, leading to 57 and 58. Relative energies (ΔGrel)
formation of diastereomers 51a, 51b and 52 in 20%, 34%, and
17% yield, respectively.
reported in kcal/mol with respect to the respective most stable
We initially assumed that the corresponding cyclopropanes
were intermediates in the mechanism of this formal (4 + 1)-
cycloaddition. However, heating 46 at 111 °C instead of
160 °C gave yet another surprising result: the cyclic orthoester
69 (70%54) was formed as the major compound (Scheme 17).
This orthoester probably resulted from a cyclization of the
corresponding zwitterionic species 72, themselves originating
either from the opening of the corresponding cyclopropanes or
from a conjugated addition of the carbene to the electron-
deficient alkene. Recourse to the aromatic oxadiazoline 6830b−d
and heating to 50 °C was necessary to investigate the formation
of the cyclic orthoester in the case of the 6−5 bicyclic system.
In this case, only orthoester 70 (74%54) was obtained, and no
trace of the corresponding cyclopropane or the (4 + 1)-
cycloadduct 49 was detected. Lastly, when diene 50 was
thermolyzed in PhCl at 132 °C, orthoester 71 was the major
isolated product (65%54) along with compound 52 (10%).
We attempted to confirm that orthoesters 69−71 were the
exclusive source of (4 + 1)-products. The conversion of
orthoester 69 into its corresponding O-heterobicycle required
temperature as high as 160 °C. Yields of products and ratios
were nearly identical as per when diene 46 was heated directly
to 160 °C. Again, the 6−5 bicyclic system behaved differently
than the other bicyclic systems and gave decomposition
products in a reproducible manner when orthoester 70 was
submitted to heat. We have no explanation for this behavior
yet. Thermolysis of diene 50 for a short period of time (PhMe,
160 °C, 1.5 h) led to a mixture of orthoester 71 and
cycloadducts 51a, 51b, and 52 in 35%, 13%, 22%, and 17%
yield, respectively. Importantly, the relative amount of product
52 (17%, Schemes 8 and 17) did not change when heating was
continued further to 160 °C (15 h), while the yields of 51a and
⧧
conformation of the free carbene. Isomerization barriers (ΔGiso
reported with respect to trans-59.
)
Unlike cyclopropanes 64a−b, compounds 40 and 42, which
possess an activating group at C-2, have no driving force to
open into a zwitterionic species and are thus thermally stable.
We rapidly dismissed the idea that cycloadduct 45 would come
from the thermal rearrangement of vinylcyclopropane 44 when
resubmission of the latter to the reaction conditions or to
higher temperatures failed to give any of compound 45
(cf. Scheme 7). This provides strong experimental evidence that 45
may be the product of a truly concerted (4 + 1)-cycloaddition.
Contrary to 1-substituted dienes, a concerted (4 + 1)-
cycloaddition could be advantaged by the stabilizing effect at
the transition state of electron-withdrawing groups at the C-2/
C-3 position.67 We have been able to optimize and characterize
in silico two transition structures that also support this assertion
(Figure 5). In the case of TS45a, a two-step no-intermediate
process was found to account for 45. The almost equi-energetic
TS45b showed a truly concerted process that was confirmed by
analysis of the IRC and the vibrational mode of the imaginary
frequency. The particular stereochemistry of the trans ring-
junction of 45, which had never been observed up to now in
this study, might result from a conformational preference where
the methoxy group of the dialkoxycarbene is directed inside the
developing tricyclic structure.
Recall that moving the activating group at C-3 on the diene
resulted in several changes in the stereoselectivity of the
annulation and revealed a yet more complex mechanistic
picture. Thermolysis of dienes 46 and 48 gave bicyclic
cyclopentenes 47 and 49 in 72% and 82% yield, respectively,
both as single diastereomers (cf. Scheme 8). The thermolysis of
diene 50 (PhMe, 160 °C, 15 h) resulted in the unselective
5949
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Scheme 17. Results from the Thermolysis of 46, 68, and 50
Figure 6. Three natural products that are targeted for synthesis.
51b increased gradually. Orthoester 71 is thus not the source of
cycloadduct 52.
to favor the concerted pathway over the others is not obvious at
this point, but it might be desirable to be able to do so in order to
improve the stereochemical predictability of the reaction.
We are currently investigating a chiral nonracemic version of
the cycloaddition and plan to use it to prepare the natural
products carotol,69 pseudolaric acid A,70 and secospatacetal B71
(Figure 6).
The global picture for the thermolysis of 3-substituted dienes
can now be summarized in two points: (1) in accord with
Davies’ findings,68 the diastereocontrolled formation of cyclo-
adducts 47, 49, 51a, and 51b would stem from a kinetic facial
selectivity of the zwitterion 72. The latter, in our case, initially
and reversibly, form the corresponding orthoesters 69−71; (2)
cyclopentene 52 (7−5 fused ring system) is formed from a
concerted (4 + 1)-cycloaddition. The trans ring junction, as in 45,
seems to be a structural characteristic of cycloadduct arising
from an intramolecular concerted (4 + 1)-cycloaddition.
ASSOCIATED CONTENT
■
S
* Supporting Information
Characterization and NMR spectra. This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
Claude.Spino@USherbrooke.ca
■
CONCLUSION
■
We have shown that dialkoxycyclocarbenes can react with many
dienes bearing various electron-withdrawing groups at two
different positions to lead to (4 + 1)-cycloadducts in good
yields and good-to-excellent stereoselectivities. We demon-
strated that the intramolecular (4 + 1)-cyclization is a useful
and valuable reaction that allows access to stereodefined fused
bicyclic ring systems. The creation of adducts bearing a
quaternary center is a definite highlight of this work.
We have shown that the mechanistic portrait of the intra-
molecular reactions of dialkoxycarbenes with electron-deficient
dienes is complex. Many of the mechanistic pathways available to
the free carbenes are close in energy, and small changes in the
substrate affect which one is followed. Nevertheless, most carbenes
react via the initial formation of a cyclopropane, which then opens
to a zwitterion that closes again to the desired cyclopentene
products with stereoselectivities that depend on the substrate. We
have provided evidence of a truly concerted intramolecular (4 + 1)-
cycloaddition in two examples (dienes 43 and 50), each case
featuring the formation of a 7−5 fused bicyclic system. It seems
that a trans ring junction is obtained in such cases as opposed to a
cis ring junction when a zwitterionic intermediate is involved. How
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Natural Sciences and Engineering Research
Council of Canada (NSERC) and the University of Sherbrooke
for financial support. Computational resources were provided
by the Res
(RQCHP).
́ ́ ́
eau Quebecois de Calcul de Haute Performance
REFERENCES
■
(1) Diels, O.; Alder, K. Justus Liebigs Ann. Chem. 1928, 460, 98−122.
(2) For books or reviews on the Diels−Alder cycloaddition, see:
(a) Fringuelli, F.; Taticchi, A. The Diels-Alder Reaction: Selected
Practical Methods; John Wiley & Sons: U.K., 2002. (b) Carruthers, W.
Cycloaddition Reactions in Organic Synthesis; Pergamon Press: Oxford,
U.K., 1990. (c) Oppolzer, W. Comprehensive Organic Synthesis; Trost,
B. M., Fleming, I., Paquette, L. A., Eds.; Pergamon Press: Oxford, U.K.,
1991; Vol. 5, p 315. (d) Nicolaou, K. C.; Snyder, S. A.; Montagnon, T.;
Vassilikogiannakis, G. Angew. Chem., Int. Ed. 2002, 41, 1668−1698.
5950
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
(3) According to the IUPAC, two different notations can be used to
describe cycloaddition reactions. Round and square brackets describe
the number of atoms or electrons, respectively, involved in the
cycloaddition. Therefore, a reaction between a carbene and a
conjugated diene should be described as a (4 + 1)-cycloaddition or
as a [4 + 2]-cycloaddition. See: Muller, P. Pure Appl. Chem. 1994, 66,
1077−1184. To avoid any confusion with the Diels-Alder cyclo-
addition, we shall use the former throughout.
(4) (a) Flemming, I. Molecular Orbitals and Organic Chemical
Reactions; John Wiley & Sons: U.K., 2010; Chapter 6. (b) Hoffmann,
R.; Woodward, R. B. Acc. Chem. Res. 1968, 1, 17−22. (c) Woodward,
R. B.; Hoffmann, R. Angew. Chem., Int. Ed. 1969, 8, 781−853.
(5) Very few examples of (4 + 1)-cycloadditions are known to be
unequivocally concerted, and in all cases, the (4 + 1)-cycloadducts are
obtained in very low yields: (a) Burger, U.; Gandillon, G. Tetrahedron
Lett. 1979, 20, 4281−4284. (b) Le, N. A.; Jones, M. Jr.; Bickelhaupt,
F.; de Wolf, W. H. J. Am. Chem. Soc. 1989, 111, 8491−8493.
(c) Kraakman, P. A.; de Wolf, W. H.; Bickelhaupt, F. J. Am. Chem. Soc.
1989, 111, 8534−8535. (d) Lambert, J. B.; Ziemnicka-Merchant, B. T.
J. Org. Chem. 1990, 55, 3460−3464 and references therein.
(e) Turkenburg, L. A. M.; de Wolf, W. H.; Bickelhaupt, F. Tetrahedron
Lett. 1982, 23, 769−770. (f) Mayr, H.; Heigl, U. W. Angew. Chem., Int.
Ed. 1985, 24, 579−580. (g) Jenneskens, L. W.; de Wolf, W. H.;
Bickelhaupt, F. Angew. Chem., Int. Ed. 1985, 24, 585−586.
555−561. For other recent examples on the rearrangement of
vinylcyclopropanes to cyclopentenes, see: (e) Willcott, M. R. III;
Cargle, V. H. J. Am. Chem. Soc. 1969, 91, 4310−4311. (f) Willcott,
M. R. III; Cargle, V. H. J. Am. Chem. Soc. 1967, 89, 723−724.
(g) Baldwin, J. E.; Leber, P. A. Org. Biomol. Chem. 2008, 6, 36−47 and
references therein. (h) Patel, P. R.; Boger, D. L. J. Am. Chem. Soc.
2010, 132, 8527−8529. (i) Qiao, Y.; Chu, T.-S. J. Org. Chem. 2011, 76,
3086−3095. (j) Coscia, R. W.; Lambert, T. H. J. Am. Chem. Soc. 2009,
131, 2496−2498. (k) Widenhoefer, R. A. Angew. Chem., Int. Ed. 2009,
48, 6950−6952. (l) Wang, S. C.; Tantillo, D. J. J. Organomet. Chem.
2006, 691, 4386−4392. (m) Baldwin, J. E. Chem. Rev. 2003, 103,
1197−1212. (n) Davies, H. M. L.; Kong, N.; Churchill, M. R. J. Org.
Chem. 1998, 63, 6586−6589. For a discussion on the isomerization of
various cyclopropanes substituted with carbonylated groups into their
corresponding five-membered heterocycles, see: (o) Schneider, T. F.;
Werz, D. B. Org. Lett. 2011, 13, 1848−1851.
(15) For metal-catalyzed carbonylative (4 + 1)-cycloadditions, see:
(a) Meng, Q.; Li, M.; Zhang, J. J. Mol. Struct.: Theochem. 2005, 726,
47−54 and references therein. (b) Lautens, M.; Klute, W.; Tam, W.
Chem. Rev. 1996, 96, 49−92. (c) Fruhauf, H.-W. Chem. Rev. 1997, 97,
̈
523−596.
(16) For leading references, see: (a) Li, Z.; Moser, W. H.; Deng, R.;
Sun, L. J. Org. Chem. 2007, 72, 10254−10257. (b) Oshita, M.;
Yamashita, K.; Tobisu, M.; Chatani, N. J. Am. Chem. Soc. 2005, 127,
761−766. (c) Rigby, J. H.; Qabar, M. J. Am. Chem. Soc. 1991, 113,
8975−8976.
(6) For selected books and reviews on carbenes, see: (a) Brinker, U. H.
In Advances in Carbene Chemistry; Elsevier Science B.V.: Amsterdam,
2001. (b) Bertrand, G. In Carbene Chemistry: From Fleeting Intermediates
to Powerful Reagents; Marcel Dekker: New York, 2002. (c) Moss, R. A.,
Platz, M., Jones, M. Jr. In Reactive Intermediate Chemistry; John Wiley &
(17) For examples of formal (4 + 1)-cyclizations involving diazo
reagents, see: (a) Dalton, A. M.; Zhang, Y.; Davie, C. P.; Danheiser, R.
L. Org. Lett. 2002, 4, 2465−2468. (b) Moser, W. H.; Feltes, L. A.; Sun,
L.; Giese, M. W.; Farrell, R. W. J. Org. Chem. 2006, 71, 6542−6546.
(c) Zhao, L.-B.; Guan, Z.-H.; Han, Y.; Xie, Y.-X.; He, S.; Liang, Y.-M.
J. Org. Chem. 2007, 72, 10276−10278. (d) Giese, M. W.; Moser, W. H.
Org. Lett. 2008, 10, 4215−4218.
(18) For a recent example, see: Lu, L.-Q.; Zhang, J.-J.; Li, F.; Chen,
Y.; An, J.; Chen, J.-R.; Xiao, W.-J. Angew. Chem., Int. Ed. 2010, 49,
4495−4498 and references therein.
(19) For examples of formal (4 + 1)-annulations using copper
carbenoids, see: (a) Son, S.; Fu, G. C. J. Am. Chem. Soc. 2007, 129,
1046−1047. For rhodium carbenoids, see: (b) Mizuno, A.; Kusama,
H.; Iwasawa, N. Angew. Chem., Int. Ed. 2009, 48, 8318−8320.
(c) Schnaubelt, J.; Marks, E.; Reissig, H. U. Chem. Ber. 1996, 129, 73−
75. For chromium carbenoids, see: (d) Barluenga, J.; Fanlo, H.;
Sons: New York, 2004. (d) Vignolle, J.; Cattoen, X.; Bourissou, D. Chem.
̈
́
Rev. 2009, 109, 3333−3384. (e) de Fremont, P.; Marion, N.; Nolan, S. P.
Coord. Chem. Rev. 2009, 253, 862−892. (f) Cheng, Y.; Meth-Cohn, O.
Chem. Rev. 2004, 104, 2507−2530.
(7) For exceptions where the (4 + 1)-cycloadduct is predominant,
see: (a) Davies, H. M. L.; Clark, T. J. Tetrahedron 1994, 50, 9883−
9892. (b) Sugita, H.; Mizuno, K.; Mori, T.; Isagawa, K.; Otsuji, Y.
Angew. Chem., Int. Ed. Engl. 1991, 30, 984−986.
(8) Boisvert, L.; Beaumier, F.; Spino, C. Org. Lett. 2007, 9, 5361−
5363.
(9) For theoretical investigations on the concerted (4 + 1)-
cycloaddition and its competitive cyclopropanation reaction, see:
(a) Fujimoto, H.; Hoffmann, R. J. Phys. Chem. 1974, 78, 1167−1173.
(b) Schoeller, W. W.; Yurtsever, E. J. Am. Chem. Soc. 1978, 100, 7548−
7550. (c) Bauld, N. L.; Wirth, D. J. Comput. Chem. 1981, 2, 1−6.
(d) Schoeller, W. W.; Aktekin, N. J. Chem. Soc., Chem. Commun. 1982,
20−22. (e) Evanseck, J. D.; Mareda, J.; Houk, K. N. J. Am. Chem. Soc.
1990, 112, 73−80 and references therein.
(10) (a) Buchert, M.; Hoffmann, M.; Reißig, H.-U. Chem. Ber. 1995,
128, 605−614. (b) Buchert, M.; Reißig, H.-U. Chem. Ber. 1992, 125,
2723−2729. (c) Buchert, M.; Reißig, H.-U. Tetrahedron Lett. 1988, 29,
2319−2320.
(11) See for example: (a) Kurteva, V. B.; Afonso, C. A. M. Chem. Rev.
2009, 109, 6809−6857 and references therein. (b) Das, S.; Chandrasekhar,
S.; Yadav, J. S.; Gree, R Chem. Rev. 2007, 107, 3286−3337.
́
(12) (a) Hudlicky, T.; Koszyk, F. J.; Kutchan, T. M.; Sheth, J. P.
J. Org. Chem. 1980, 45, 5020−5027. (b) Hudlicky, T.; Koszyk, F. J.;
Dochwat, D. M.; Cantrell, G. L. J. Org. Chem. 1981, 46, 2911−2915.
(c) Short, R. P.; Revol, J.-M.; Ranu, B. C.; Hudlicky, T. J. Org. Chem.
1983, 48, 4453−4461.
(13) (a) Danheiser, R. L.; Bronson, J. J.; Okano, K. J. Am. Chem. Soc.
1985, 107, 4579−4582. (b) Danheiser, R. L.; Martinez-Davilla, C.;
Auchus, R. J.; Kadonaga, J. T. J. Am. Chem. Soc. 1981, 103, 2443−2446.
(c) Danheiser, R. L.; Martinez-Davilla, C.; Morin, J. M. Jr J. Org.
Chem. 1980, 45, 1340−1341.
(14) For selected reviews, see: (a) Hudlicky, T.; Reed, J. W. Angew.
Chem., Int. Ed. 2010, 49, 4864−4876 and references therein. (b) Frey,
H. M.; Walsh, R. Chem. Rev. 1969, 69, 103−124. (c) Sarel, S.; Yovell,
J.; Sarel-Imber, M. Angew. Chem., Int. Ed. Engl. 1968, 7, 577−588. See
also: (d) Mazzocchi, P. H.; Tamburin, H. J. J. Am. Chem. Soc. 1975, 97,
Lop
́
ez, S.; Flor
́
ez, J. Angew. Chem., Int. Ed. 2007, 46, 4136−4140.
(e) Barluenga, J.; Lop
́
ez, S.; Florez, J. Angew. Chem., Int. Ed. 2003, 42,
́
231−233. For α-benzotriazolyl organolithium compounds, see:
(f) Davie, C. P.; Danheiser, R. L. Angew. Chem., Int. Ed. 2005, 44,
5867−5870.
(20) Spino, C.; Rezaei, H.; Dupont-Gaudet, K.; Bel
Chem. Soc. 2004, 126, 9926−9927.
́
anger, F. J. Am.
(21) For more examples of (4 + 1)-cycloadditions involving
nucleophilic carbenes and various dienes, see: (a) Rigby, J. H.;
Sidique, S. Org. Lett. 2007, 9, 1219−1221 and references therein.
(b) Er, H.-T.; Pole, D. L.; Warkentin, J. Can. J. Chem. 1996, 74, 1480−
1489. (c) Rigby, J. H.; Laurent, S. J. Org. Chem. 1999, 64, 1766−1767.
(d) Rigby, J. H.; Cavezza, A.; Heeg, M. J. Tetrahedron Lett. 1999, 40,
2473−2476. (e) Rigby, J. H.; Laurent, S.; Dong, W.; Danca, M. D.
Tetrahedron 2000, 56, 10101−10111. (f) Rigby, J. H.; Neale, N. A.;
Schlegel, H. B. Heterocycles 2002, 58, 105−113. (g) Rigby, J. H.; Wang,
Z. Org. Lett. 2002, 4, 4289−4291. (h) Rigby, J. H.; Aasmul, M.
Heterocycles 2004, 62, 143−147. (i) Rigby, J. H.; Burke, P. J.
Heterocycles 2006, 67, 643−653. (j) Cheng, Y.; Wang, B.; Cheng, L.-Q.
J. Org. Chem. 2006, 71, 4418−4427.
(22) For a review on 2,2-dialkoxy-Δ3-1,3,4-oxadiazolines and the
reactivity of bis(heteroatom-substituted)carbenes, see: Warkentin, J.
Acc. Chem. Res. 2009, 42, 205−212 and references therein.
(23) (a) Hoffmann, R. W.; Hauser, H. Tetrahedron 1965, 21, 891−
̈
902. (b) Hoffmann, R. W. Angew. Chem., Int. Ed. Engl. 1971, 10, 529−
537. (c) Lemal, D. M.; Gosselink, E. P.; McGregor, S. D. J. Am. Chem.
Soc. 1966, 88, 582−600.
5951
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
(24) In the absence of sufficiently reactive partner reagents,
dialkoxycarbenes prefer to dimerize to form the corresponding tetra-
alkoxyethylenes.
(25) Lilienblum, W.; Hoffmann, R. W. Chem. Ber. 1977, 110, 3405−
3409.
(d) Parziale, P. A.; Berson, J. A. J. Am. Chem. Soc. 1990, 112, 1650−
1652.
(39) Dupont-Gaudet, K., M.Sc. Thesis, Universite de Sherbrooke,
́
2006.
(40) For selected examples of Diels−Alder reaction involving
trifluoromethyl dienes, see: (a) Ignatowska, J.; Dmowski, W.
J. Fluorine Chem. 2006, 127, 720−729. (b) Nagai, T.; Nasu, Y.;
Shimada, T.; Shoda, H.; Koyama, M.; Ando, A.; Miki, T.; Kumadaki, I.
J. Fluorine Chem. 1992, 57, 245−249.
(26) Colomvakos, J. D.; Egle, I.; Ma, J.; Pole, D. L.; Tidwell, T. T.;
Warkentin, J. J. Org. Chem. 1996, 61, 9522−9527.
(27) Rigby, J. H.; Casseva, A.; Gulzar, A. J. Am. Chem. Soc. 1996, 118,
12848−12849.
(28) Hoffmann, R. W. Acc. Chem. Res. 1985, 18, 248−253.
(29) Moss, R. A. Acc. Chem. Res. 2006, 39, 267−272 and references
therein.
(41) Shimizu, M.; Hiyama, T. Angew. Chem., Int. Ed. 2005, 44, 214−
231 and references therein.
(42) For selected examples of three-component [4 + 2]-cyclo-
addition−allylboration sequence, see:(a) Gao, X.; Hall, D. G.; Deligny,
M.; Favre, A.; Carreaux, F.; Carboni, B. Chem.Eur. J. 2006, 13,
3132−3142. (b) Gao, X.; Hall, D. G. J. Am. Chem. Soc. 2003, 125,
9308−9309.
(43) For examples, see: (a) Fuerst, D. E.; Stoltz, B. M.; Wood, J. L.
Org. Lett. 2000, 2, 3521−3523. (b) Fouque, E.; Rousseau, G.; Seyden-
Penne, J. J. Org. Chem. 1990, 55, 4807−4817.
(30) For the thermolysis of dimethyl oxadiazolines (100 °C), see:
(a) El-Saidi, M.; Kassam, K.; Pole, D. L.; Tadey, T.; Warkentin, J.
J. Am. Chem. Soc. 1992, 114, 8751−8752. For the thermolysis of
arylmethyl oxadiazolines (50 °C), see: (b) Sokol, W.; Warkentin, J.
Can. J. Chem. 2006, 84, 927−933. (c) Zhou, H.; Mloston, G.;
Warkentin, J. Org. Lett. 2005, 7, 487−489. (d) Zhou, H.; Mloston, G.;
Warkentin, J. Org. Lett. 2008, 10, 1323. For the thermolysis of strained
oxadiazolines (≤27 °C), see: (e) Czardybon, W.; Sokol, W.;
Warkentin, J.; Werstiuk, N. H. J. Phys. Org. Chem. 2008, 21, 41−46.
(31) The relative stereochemistries of 12b, 12c, 12e, 13f, 14b, 44, 45,
57, and 49 (after derivatization) were confirmed by single-crystal X-ray
diffraction analysis while relative stereochemistries of 12g, 12h, 13g,
14a, 38h, 38i, 40, 47, 51a, 51b, 52, 54, and 55 were confirmed by
NOESY correlations (see Supporting Information). The stereo-
chemistries of 11a, 12a, and 13a were assigned from a single crystal
X-ray analysis of derivatives, and those of aldehydes 11d and 12d were
confirmed by their reduction to the corresponding known alcohols
(see ref 20).
(44) For palladium-catalyzed reactions, see:(a) Fugami, K.; Morizawa,
Y.; Oshima, K.; Nozaki, H. Tetrahedron Lett. 1985, 26, 857−860.
(b) Burgess, K. J. Org. Chem. 1987, 52, 2046−2051. (c) Burgess, K.
Tetrahedron Lett. 1985, 26, 3049−3052. (d) Harvey, D. F.; Brown, M.
F. Tetrahedron Lett. 1991, 32, 2871−2874. For rhodium-catalyzed
reactions, see: (e) Hayashi, M.; Ohmatsu, T.; Meng, Y.-P.; Saigo, K.
Angew. Chem., Int. Ed. 1998, 37, 837−839. (f) Jun, C.-H.; Kang, J.-B.;
Lim, Y.-G. Tetrahedron Lett. 1995, 36, 277−280. (g) Alcock, N. W.;
Brown, J. M.; Conneely, J. A.; Williamsom, D. H. J. Chem. Soc., Perkin
Trans. 2 1979, 962−971. (h) Salomon, R. G.; Salomon, M. F.;
Kachinski, J. L. C. J. Am. Chem. Soc. 1977, 99, 1043−1054. (i) Aris, V.;
Brown, J. M.; Conneely, J. A.; Golding, B. R.; Williamson, D. H. J.
Chem. Soc., Perkin Trans. 2 1975, 4−10. (j) Alcock, N. W.; Brown, J.
M . ; C o n n e e l y , J . A . ; W i l l i a m s o m , D . H .
J. Chem. Soc., Chem. Commun. 1975, 792−793. (k) Grigg, R.; Hayes,
R.; Sweeney, A. J. Chem. Soc., Chem. Commun. 1971, 1248−1249.
(l) Hudlicky, T.; Koszyk, F. J.; Kutchan, T. M.; Sheth, J. P J. Org.
Chem. 1980, 45, 5020−5027. For nickel-catalyzed reactions, see:
(m) Wang, S. C.; Troast, D. M.; Conda-Sheridan, M.; Zuo, G.;
LaGarde, D.; Louie, J.; Tantillo, D. J. J. Org. Chem. 2009, 74, 7822−
7833. (n) Zuo, G.; Louie, J. Angew. Chem., Int. Ed. 2004, 43, 2277−
2279. (o) Ryu, I.; Ikura, K.; Tamura, Y.; Maenaka, J.; Ogawa, A.;
Sonoda, N. Synlett 1994, 941−942. (p) Murakami, M.; Nishida, S.
Chem. Lett. 1979, 927−930.
(32) For the reactivity of the related transient allyloxymethox-
ycarbenes, benzyloxymethoxycarbenes and dibenzyloxycarbenes, see:
(a) Venneri, P. C.; Warkentin, J. J. Am. Chem. Soc. 1998, 120, 11182−
11183. (b) Reid, D. L.; Warkentin, J. J. Chem. Soc., Perkin Trans. 2
́
2000, 1980−1983. (c) Reid, D. L.; Hernandez-Trujillo, J.; Warkentin, J.
J. Phys. Chem. A 2000, 104, 3398−3405. (d) Merkley, N.; El-Saidi, M.;
Warkentin, J. Can. J. Chem. 2000, 78, 356−361. (e) Merkley, N.;
Warkentin, J. Can. J. Chem. 2000, 78, 942−949. (f) Merkley, N.;
Venneri, P. C.; Warkentin, J. Can. J. Chem. 2001, 79, 312−318.
(g) Merkley, N.; Warkentin, J. Can. J. Chem. 2002, 80, 1187−1195.
(h) Merkley, N.; Reid, D. L.; Warkentin, J. Tetrahedron Lett. 2002, 43,
1927−1929. (i) Plazuk, D.; Warkentin, J.; Werstiuk, N. H. Tetrahedron
2005, 61, 5788−5796.
1
(33) Conversion was determined by H NMR spectroscopy. Any
(45) For leading references, see: (a) Liu, P.; Sirois, L. E.; Cheong, P.
H.-Y.; Yu, Z.-X.; Hartung, I. V.; Rieck, H.; Wender, P. A.; Houk, K. N.
J. Am. Chem. Soc. 2010, 132, 10127−10135 and references therein.
(b) Wender, P. A.; Stemmler, R. T.; Sirois, L. E. J. Am. Chem. Soc.
2010, 132, 2532−2533.
attempts at purification resulted in complete degradation of the
product.
(34) The reported pKa values of CH3NO2 in H2O and DMSO are
10.2 and 17.2, respectively. H2O: Matthews, W. S.; Bares, J. E.;
Bartmess, J. E.; Bordwell, F. G.; Cornforth, F. J.; Drucker, G. E.;
Margolin, Z.; McCallum, R. J.; McCollum, G. J.; Vanier, N. R. J. Am.
Chem. Soc. 1975, 97, 7006−7014. DMSO: Turnbull, D.; Maron, S. H.
J. Am. Chem. Soc. 1943, 65, 212−218.
(46) For examples of radical-mediated ring opening of vinyl-
cyclopropanes, see: (a) Miura, K.; Fujisawa, N.; Hosomi, A. J. Org.
Chem. 2004, 69, 2427−2430. (b) Miura, K.; Fujisawa, N.; Saito, H.;
Nishikori, H.; Hosomi, A. Chem. Lett. 2002, 32−33. (c) Booker-
Milburn, K. I.; Barker, A.; Brailsford, W.; Cox, B.; Mansley, T. E.
Tetrahedron 1998, 54, 15321−15344.
(35) For the estimated pKa values for protonation of dimethox-
ycarbene, see: (a) Keeffe, J. R.; More O’Ferrall, R. A. ARKIVOC 2008,
10, 183−204. (b) Guthrie, J. P.; More O’Ferrall, R. A.; O’Donoghue,
A. C.; Waghorne, W. E.; Zrinski, I. J. Phys. Org. Chem. 2003, 16, 582−
587. (c) Pezacki, J. P. Can. J. Chem. 1999, 77, 1230−1240.
(36) For examples of insertion reactions between dimethoxycarbene
and compounds with acidic protons, see: (a) Couture, P.; Pole, D. L.;
Warkentin, J. J. Chem. Soc., Perkin Trans. 2 1997, 1565−1570. (b) Du,
X.-M.; Fan, H.; Goodman, J. L.; Kesselmayer, M. A.; Krogh-Jespersen,
K.; LaVilla, J. A.; Moss, R. A.; Shen, S.; Sheridan, R. S. J. Am. Chem.
Soc. 1990, 112, 1920−1926.
(47) The thermolysis of 2,2-dimethoxy-5,5-dimethyl-Δ3-1,3,4-oxadia-
zoline at 100 °C follows a first-order kinetic with k = 1.2 × 10−5 s−1.
See ref 30a.
(48) The possibility of the carbene reacting as a diradical is excluded
since the singlet state is stabilized by −53 kcal/mol relative to the
triplet state. See: Mendez, F.; Garcia-Garibay, M. A. J. Org. Chem.
1999, 64, 7061−7066.
(49) Flemming, I. Molecular Orbitals and Organic Chemical Reactions;
John Wiley & Sons: U.K., 2010; Chapter 6.
(50) For the stereospecific cyclopropanation of dimethoxycarbene to
β-deuteriostyrenes, see: Moss, R. A.; Huselton, J. K. J. Chem. Soc.,
Chem. Commun. 1976, 950−951.
(51) For examples of zwitterionic rearrangement of vinylcyclopro-
panes, see: (a) Barluenga, J.; Tomas, M.; Suarez-Sobrino, A. L.
́ ́
́
(37) Boisvert, Luc, Ph.D. Thesis, Universite de Sherbrooke, 2006.
(38) Similar adducts have been observed from a homo-[1,5]-
sigmatropic hydrogen migration on various vinylcyclopropanes. See:
(a) Ellis, R. J.; Frey, H. M. J. Chem. Soc. 1964, 5578−5583.
(b) Loncharich, R. J.; Houk, K. N. J. Am. Chem. Soc. 1988, 110, 2089−
2092. (c) Piers, E.; Maxwell, A. R. Can. J. Chem. 1984, 62, 2392−2395.
5952
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953

Journal of the American Chemical Society
Article
Synthesis 2000, 935−940. (b) Buchert, M.; Reissig, H.-U. Liebigs Ann.
1996, 2007−2013.
(52) For the involvement of diradicals in the isomerization of
vinylcyclopropanes into their corresponding cyclopentenes, see:
Baldwin, J. E. J. Comput. Chem. 1998, 19, 222−231. (b) See also
refs 14b and 14h.
(53) For reviews on the reactivity of donor−acceptor cyclopropanes,
see: (a) Reissig, H.-U.; Zimmer, R. Chem. Rev. 2003, 103, 1151−1196.
(b) Yu, M.; Pagenkopf, B. L. Tetrahedron 2005, 61, 321−347. (c) De
Simone, F.; Waser, J. Synthesis 2009, 3353−3374.
(54) Figures created with CYLview, 1.0b; Legault, C. Y.; Universite
de Sherbrooke, 2009 (http://www.cylview.org).
́
(55) See Supporting Information for more details.
(56) Singleton, D. A.; Hang, C.; Szymanski, M. J.; Meyer, M. P.;
Leach, A. G.; Kuwata, K. T.; Chen, J. S.; Greer, A.; Foote, C. S.; Houk,
K. N. J. Am. Chem. Soc. 2003, 125, 1319−1328.
(57) This kind of behavior has already been illustrated by
calculations. See ref 14j.
(58) (a) Berson, J. A. Acc. Chem. Res. 1972, 5, 406−414. (b) Berson,
J. A.; Salem, L. J. Am. Chem. Soc. 1972, 94, 8917−8918.
(59) Danheiser, R. L.; Martinez-Davila, C.; Auchus, R. J.; Kadonaga,
J. T. J. Am. Chem. Soc. 1981, 103, 2443−2446.
(60) (a) Oyola, Y.; Singleton, D. A. J. Am. Chem. Soc. 2009, 131,
3130−3131. (b) Vayner, G.; Addepalli, S. V.; Song, K.; Hase, W. L.
J. Chem. Phys. 2006, 125, 014317.
(61) Hine, J. Adv. Phys. Org. Chem. 1978, 15, 1−61.
(62) It has been shown that dimethoxycarbene inserts into acidic
protons of nitromethane to give 1,1-dimethoxy-2-nitroethane. See ref 37.
(63) For reviews on the preparation and properties of nitrile ylides,
see: (a) Kirmse, W. Eur. J. Org. Chem. 2005, 237−260. (b) Escolano,
́
C.; Duque, M. D.; Vazquez, S. Curr. Org. Chem. 2007, 11, 741−772.
(64) For the rate constants of pyridinium ylide formation from
oxycarbenes and pyridine, see: (a) Bonneau, R.; Liu, M. T. H.;
Lapouyade, R. J. Chem. Soc., Perkin Trans. 1 1989, 1547−1548.
(b) Jones, M. B.; Platz, M. S. J. Org. Chem. 1991, 56, 1694−1695.
(c) Ge, C. S.; Jang, E. G.; Jefferson, E. A.; Liu, W.; Moss, R. A.;
Wlostowska, J.; Xue, S. J. Chem. Soc., Chem. Commun. 1994, 1479−
1480. For related dipolar cycloadditions of pyridinium ylides derived
from the reaction of carbenes and pyridine, see: (d) Shevchenko, V. V.;
Zhegalova, N. G.; Borzenko, A. O.; Nikolaev, V. A. Helv. Chim. Acta
2008, 91, 501−509. (e) Jackson, J. E.; Soundararajan, N.; Platz, M. S.
J. Am. Chem. Soc. 1988, 110, 5595−5596.
(65) Large quantities of formates and carbonates accompanied the
cycloadducts, which explains the lower yields obtained.
(66) For ring opening and reclosing to isomeric cyclopropanes
through zwitterionic intermediates, see: (a) Howe, N. E.; Yankee, E. W.;
Cram, D. J. J. Am. Chem. Soc. 1973, 95, 4230−4237. (b) Chmurny,
A. B.; Cram, D. J. J. Am. Chem. Soc. 1973, 95, 4237−4244.
(67) For examples and an explanation of this effect in the Diels−
Alder cycloaddition, see: (a) Spino, C.; Pesant, M.; Dory, Y. L. Angew.
Chem., Int. Ed. 1998, 37, 3262−3265. (b) Spino, C.; Rezaei, H.; Dory,
Y. L. J. Org. Chem. 2004, 69, 757−764.
(68) For stereoselective Et2AlCl-induced ring expansion of donor−
acceptor substituted vinylcyclopropanes on relatively similar systems,
see ref 14i.
(69) (a) Sykora, V.; Novotny, L.; Holub, M.; Herout, V.; Sorm, F.
Collect. Czech. Chem. Commun. 1961, 26, 788−802. (b) De Broissia,
H.; Levisalles, J.; Rudler, H. Bull. Soc. Chim. Fr. 1972, 4314−4318.
(c) Miski, M.; Mabry, T. J. Phytochemistry 1986, 25, 1673−1675.
(d) Williams, C. A.; Harborne, J. E. Phytochemistry 1972, 11, 1981−
1987. (e) Benecke, R.; Reinhold, K.; Kessel, M.; Schmidt, W.
Pharmazie 1987, 42, 256−259.
(70) (a) Li, Z. L.; Pan, D. J.; Hu, C. Q.; Wu, Q. L.; Yang, S. S.; Xu,
G. Y. Acta Chim. Sin. 1982, 40, 447. (b) Li, Z. L.; Pan, D. J.; Wu, Q. L.;
Xu, G. Y. Acta Chim. Sin. 1982, 40, 757. (c) Trost, B. M.; Waser, J.;
Meyer, A. J. Am. Chem. Soc. 2008, 130, 16424−16434. (d) Geng, Z.;
Chen, B.; Chiu, P. Angew. Chem., Int. Ed. 2006, 45, 6197−6201.
(71) Yamase, H.; Umemoto, K.; Ooi, T.; Kusumi, T. Chem. Pharm.
Bull. 1999, 47, 813−818.
5953
dx.doi.org/10.1021/ja211927b | J. Am. Chem. Soc. 2012, 134, 5938−5953