[4+2] Cycloadditions of Seven-Membered-Ring trans-Alkenes: Decreasing Reactivity with Increasing Substitution of the Seven-Membered Ring

Article abstract of DOI:10.1002/ejoc.201600329

The reactivity of trans-oxasilacycloheptenes in [4+2] cycloadditions depends on the substitution pattern on the seven-membered ring. Unhindered trans-alkenes undergo [4+2] cycloadditions with 1,3-diphenylisobenzofuran faster than the most reactive trans-c

Full text of DOI:10.1002/ejoc.201600329

DOI: 10.1002/ejoc.201600329
Full Paper
Reactivities of trans-Cycloalkenes
[4+2] Cycloadditions of Seven-Membered-Ring trans-Alkenes:
Decreasing Reactivity with Increasing Substitution of the
Seven-Membered Ring
John Santucci III,[a] Jillian R. Sanzone,[a] and K. A. Woerpel*[a]
Abstract: The reactivity of trans-oxasilacycloheptenes in [4+2] the trans-alkene decreases reactivity with 1,3-dienes in con-
cycloadditions depends on the substitution pattern on the
seven-membered ring. Unhindered trans-alkenes undergo [4+2]
cycloadditions with 1,3-diphenylisobenzofuran faster than the
most reactive trans-cyclooctene. Increasing the substitution of
the seven-membered ring or increasing the electron density of
certed cycloaddition reactions. Although highly substituted
trans-alkenes are unreactive in concerted cycloaddition reac-
tions, these alkenes react rapidly in stepwise reactions with di-
ethyl azodicarboxylate (DEAD), an electrophilic diene.
Introduction
The enhanced reactivity of strained alkenes and alkynes com-
pared to their unstrained counterparts has prompted numerous
investigations into their reactivity.[1] Strain-accelerated reactions
have been utilized in a variety of applications in synthesis,[2–4]
including the synthesis of natural products.[5] These reactions
have also been applied in bioorthogonal reactions,[6,7] such as
site-specific protein labeling and in vivo imaging.[8,9]
trans-Cyclooctenes have been shown to be highly reactive in
Diels–Alder cycloaddition reactions. trans-Cyclooctene 1, a
highly distorted trans-alkene, underwent a rapid inverse-elec-
tron-demand Diels–Alder cycloaddition with 3,6-diphenyl-s-
tetrazine (2) to form cycloadduct 3 (Scheme 1).[10] The cyclopro-
pane ring forces the eight-membered ring to adopt a half-chair
conformation, which is more strained than the crown conforma-
tion of trans-cyclooctene, leading to enhanced reactivity of al-
kene 1.[10] trans-Cyclooctene 4 also participated in a [4+2]
cycloaddition, reacting with 1,3-diphenylisobenzofuran (5) to
provide cycloadduct 6 as a mixture of diastereomers.[11] A com-
petition experiment between trans-alkene 4 and trans-cyclooct-
ene showed that alkene 4 is four times more reactive with furan
5 than trans-cyclooctene.
In contrast to trans-cyclooctenes, few examples of seven-
membered-ring trans-alkenes undergoing [4+2] cycloaddition
reactions have been reported. The smaller trans-cycloheptenes
should be more reactive than trans-cyclooctenes because they
are more strained.[12] trans-Cycloheptene (7), generated in situ
from trans-1,2-cycloheptene thionocarbonate and trimethyl
phosphite, underwent a Diels–Alder cycloaddition with diene 5
Scheme 1. [4+2] Cycloadditions of trans-cyclooctenes.
to afford bicyclic ether 8 in 6 % yield from the thionocarbonate
[Equation (1)].[13,14] Seven-membered-ring trans-alkenes were
also reported to undergo [4+2] cycloadditions with 1,3-dipoles
such as diazomethane[15] and phenyl azide.[16]
[a] Department of Chemistry, New York University,
100 Washington Square East, New York, NY 10003 USA
E-mail: kwoerpel@nyu.edu
(1)
In this article, we demonstrate that the reactivity of seven-
membered-ring trans-alkenes in concerted [4+2] cycloadditions
is greatly influenced by the substituents on the seven-mem-
https://wp.nyu.edu/woerpelgroup/
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/ejoc.201600329.
Eur. J. Org. Chem. 2016, 2933–2943
2933
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
bered ring. Preliminary studies showed that unhindered trans-
oxasilacycloheptenes are more reactive than a trans-cyclooct-
ene.[17] Additional studies show that strain energy is not the
only factor that influences the reactivity of these strained com-
pounds.[18] Steric effects play a significant role in cycloaddition
reactivity. Highly substituted trans-alkenes react slowly or are
unreactive in cycloadditions with dienes 2 and 5. Although hin-
dered trans-alkenes are sluggish in concerted Diels–Alder
cycloadditions, they display high reactivity in stepwise reactions
with diethyl azodicarboxylate (DEAD). These results indicate
that steric effects can be more significant than ring strain in the
reactivity of strained trans-cycloalkenes.
Figure 1. Seven-membered-ring trans-alkenes synthesized for this study.
Results and Discussion
Silyl enol ether 18 and trisubstituted alkene 19 were synthe-
sized to determine how the electron density of the trans double
bond influenced reactivity. Silyl enol ether 18 should be partic-
ularly reactive because silyl enol ethers are more nucleophilic
than most alkenes.[22] trans-Alkene 19, which is isomeric to silyl
enol ether 18 and should have similar strain energy, was syn-
thesized to control for electronic effects. The electron-with-
drawing group at the allylic position of the trans-alkene should
decrease the electron density of the double bond, making the
trans double bond in 19 electron-poor relative to trans-alkenes
12–18.
Synthesis of trans-Oxasilacycloheptenes
Seven-membered-ring trans-alkenes were synthesized from di-
enes, aldehydes, and silylenes generated in situ. The synthesis
of trans-alkene 12 illustrates the protocol (Scheme 2). Silver-
catalyzed silylene transfer to 1,3-diene 9 afforded vinylsilacyclo-
propane 11 in 80 % yield. Addition of benzaldehyde to vinylsila-
cyclopropane 11 yielded trans-oxasilacycloheptene 12 in 99 %
yield.[17,19] trans-Alkene 12 was unstable to atmospheric condi-
tions and it underwent thermal rearrangement.[19] It could,
however, be characterized and stored at –20 °C for greater than
ten days, and it could be used in further transformations.
[4+2] Cycloadditions with 1,3-Diphenylisobenzofuran
As predicted, trans-alkenes 13 and 14 were highly reactive in
Diels–Alder cycloaddition reactions with furan 5 (Scheme 3).[17]
In less than five minutes, trans-alkenes 13 and 14 were con-
sumed, forming cycloadducts 20 and 21 in high yields as mix-
tures of diastereomers. Spectroscopic studies enabled the as-
signment of the relative stereochemistry about the seven-mem-
bered ring, but they did not reveal the relative stereochemistry
at the bridgehead carbon atoms of the oxabicyclic framework.
Although the cycloaddition occurred with retention of the ge-
ometry of the trans-alkene,[23] facial selectivity on the diene was
low.
Scheme 2. Synthesis of trans-oxasilacycloheptene 12.
A variety of trans-oxasilacycloheptenes were prepared with
different substituent patterns to determine the reactivity of the
trans double bond in both concerted and stepwise reactions
(Figure 1). Because previous studies indicated that increased
substitution at the allylic position decreased reactivity in Diels–
Alder cycloadditions,[17] additional substituents were placed at
the allylic position to quantify this effect (alkenes 12 and 14).
It was hypothesized that fusing a second ring to the seven-
membered ring would lead to heightened reactivity. The fused
ring system would limit the ability of the double bond to allevi-
ate strain by altering bond and torsion angles.[20,21] Bicyclic tris-
ubstituted alkenes 15–17 were synthesized to examine the ef-
fect of the fused ring and to determine how the size of the
fused ring impacts reactivity.
Scheme 3. Cycloadditions of trans-alkenes 13 and 14 with diene 5.
The high reactivity of alkenes 13 and 14 prompted compari-
son to trans-cyclooctene 1, which is reported to have unparal-
Eur. J. Org. Chem. 2016, 2933–2943
www.eurjoc.org
2934
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
leled second-order kinetics in Diels–Alder cycloadditions
(Scheme 1).[10] Because these reactions were too rapid to deter-
mine the second-order rate constants by 1H NMR spectroscopy,
competition experiments were performed (Scheme 4). Silyl-pro-
tected trans-cyclooctene 22 was used because trans-cyclo-
octene 1 was insoluble under the reaction conditions. Diene 5
was added to an excess of the two trans-alkenes. Spectroscopic
analysis was conducted immediately after mixing to determine
relative rates of the reactions. trans-Oxasilacycloheptene 13 re-
acted seven times faster with diene 5 than trans-cyclooctene
22 did. This enhanced rate is likely due to the high distortion
of the trans double bond in alkene 13,[24] and therefore more
orbital distortion,[18] compared to trans-cyclooctene 22, leading
to enhanced reactivity in the Diels–Alder cycloaddition.
Figure 2. Transition state for the reaction of alkene 14 and diene 5.
alkene 12 was subjected to similar conditions with diene 5, the
reaction was over 1000 times slower than the reactions with
trans-alkenes 13 and 14 (greater than four days compared to
less than five minutes, Schemes 3 and 5). The reaction with
diene 5 was slow compared to thermal rearrangement:[19] only
28 % of cycloadduct 25 was formed as a mixture of diastereo-
mers by 1H NMR spectroscopy. The major product resulted from
thermal rearrangement.[26] When compared to the results in
Scheme 4, these results indicate that substitution at the allylic
position severely hinders reactivity with diene 5.
Scheme 5. [4+2] Cycloaddition of trans-alkene 12 with diene 5.
Highly substituted seven-membered ring trans-alkenes can
be extremely reactive. Bicyclic trans-alkenes 15 and 16 under-
went Diels–Alder cycloadditions in ten and twenty minutes, re-
spectively [Equation (2)]. Cycloadducts 27 and 28 were formed
in high yields with high diastereoselectivity. Bicyclic alkene 15
reacted faster than alkene 16, indicating that a smaller fused
ring increases reactivity. These observations suggest that the
additional ring increased strain sufficiently to permit reactivity
with diene 5 even though the trans double bond is sterically
congested.
Scheme 4. Relative rates of reactions with 1,3-diphenylisobenzofuran (5).
Because trans-alkene 14 was predicted to have similar ring
strain to alkene 13, it was expected to be comparably reactive.
trans-Cycloalkene 14 differs from alkene 13 only by the pres-
ence of a methyl group at the allylic position. Competition ex-
periments, however, revealed that the presence of the allylic
methyl group decreased the rate of cycloaddition by over one
order of magnitude (Scheme 4).[17] The decrease in rate is likely
due to the syn-pentane interaction[25] that develops between
the allylic methyl group on alkene 14 and the phenyl group on
diene 5 in the transition state (Figure 2). This unfavourable
steric interaction raises the energy of the transition state lead-
ing to cycloadduct 21 compared to cycloadduct 20. The steric
hindrance is significant enough that the less strained trans-
cyclooctene 22 reacts faster than the more distorted trans-
cycloheptene 14.
(2)
Increased substitution of the seven-membered ring further
If the fused ring were too large, however, the trans-alkene
decreased reactivity with diene 5 (Scheme 5). When trans- was no longer reactive. Bicyclic alkene 17 did not react with
Eur. J. Org. Chem. 2016, 2933–2943
www.eurjoc.org
2935 © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
diene 5 after 24 hours [Equation (3)]. The decreased reactivity group at the allylic position, which blocks the approach of iso-
of alkene 17 compared to alkenes 15 and 16 likely results be- benzofuran 5. In addition, the methyl substituent on the trisub-
cause the larger fused ring allows the trans-alkene to alleviate stituted trans double bond may also sterically destabilize the
strain by altering torsion and bond angles, thereby decreasing transition state leading to cycloadduct 19.
the distortion of the double bond.[20,21] Steric effects are also
Electron-rich trans-alkene 18 was also unreactive with diene
likely to be more significant with the additional seven-mem-
bered ring, leading to decreased reactivity.
5. After 30 d, no cycloaddition reaction was observed between
diene 5 and strained silyl enol ether 18 (Scheme 6). This lack
of reactivity is likely because both reaction partners are elec-
tron-rich. The high HOMO of silyl enol ether 18[27] is unlikely to
interact effectively with the high LUMO of diene 5.[28]
[4+2] Cycloadditions with 3,6-Diphenyl-s-tetrazine
(3)
To compare the reactivity of trans-oxasilacycloheptenes with
that of trans-cyclooctenes (Scheme 1), seven-membered-ring
trans-alkenes were subjected to inverse-electron-demand Di-
els–Alder reactions with 3,6-diphenyl-s-tetrazine (2). The high
HOMO's of the electron-rich trans-alkenes should interact well
with the low-lying LUMO+1 of the electron-deficient tetrazine
2, leading to rapid reactions.[29]
Strain is necessary for the reaction of bicyclic alkene 16 with
diene 5 to occur. To control for strain, the cis isomer of alkene
trans-16 (i.e., cis-16), was synthesized by the photoisomeriza-
tion of trans-16. cis-Alkene 16 did not undergo a reaction
with 1,3-diphenylisobenzofuran (5) even after 100 days [Equa-
tion (4)].
Similar to the cycloaddition with isobenzofuran 5, sterically
unhindered trans-alkene 14 reacted rapidly in the inverse-de-
mand Diels–Alder cycloaddition with tetrazine 2 [Equation (5)].
In less than ten minutes, trans-oxasilacycloheptene 14 was con-
sumed, forming cycloadduct 31, through a Diels–Alder cyclo-
addition followed by a reverse [4+2] cycloaddition.[30]
(4)
Although making the carbon–carbon double bond more
electron-deficient should have accelerated the reaction with
electron-rich diene 5, steric effects played a larger role than
electronic effects. In contrast to the expected results, alkene 19
was relatively unreactive: only 38 % of cycloadduct 29 had
(5)
1
formed after 50 days, as determined by H NMR spectroscopy
(Scheme 6). The formation of dimer 30, which could be formed
from small amounts of an acid, outcompeted the formation of
cycloadduct 29, consuming trans-alkene 19. The low reactivity
of trans-alkene 19 with diene 5 is likely due to the large silyloxy
Increasing substitution of the seven-membered ring resulted
in a significant decrease in reactivity. When trans-alkene 12 was
treated with tetrazine 2, only 15 % of cycloadduct 32 was
formed after four days, as determined by 1H NMR spectroscopy
(Scheme 7). The thermal rearrangement of trans-alkene 12
to oxasilacyclopentane 26 outcompeted the reaction with
tetrazine 2.[19] It is likely that increasing the substitution at the
allylic position of trans-alkene 12 relative to alkene 14 hinders
the approach of sterically hindered tetrazine 2, resulting in de-
creased reactivity.
Compared to the reaction of bicyclic alkene 16 with diene 5
[Equation (2)], the reaction of alkene 16 with tetrazine 2 was
slow and unselective [Equation (6)]. After five days, trans-alkene
16 was consumed, giving a mixture of products. Cycloadduct
33, the only identifiable product, was formed in only 21 % yield,
1
as determined by H NMR spectroscopy.
The [4+2] cycloaddition of silyl enol ether 18 and tetrazine
2 was also sluggish. After 84 d, only 80 % of the trans-alkene
was consumed. Bicyclic compound 34 was formed in 53 %
yield, as determined by 1H NMR spectroscopy [Equation (7)].
Bicyclic compound 34 decomposed into benzonitrile and a
complex mixture of products, likely through a reverse [4+2]
cycloaddition.[31]
Scheme 6. Reactions of trans-alkenes 18 and 19 with diene 5.
Eur. J. Org. Chem. 2016, 2933–2943
www.eurjoc.org
2936
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(8)
[4+2] Cycloadditions with Benzyl Azide
Previous studies of the high reactivity of strained cycloalkenes
and cycloalkynes[34] with azides prompted investigations of the
reactivity of trans-oxasilacycloheptenes with benzyl azide.
Seven-membered-ring trans-alkenes were expected to undergo
reactions with azides because alkyl azides react well with elec-
trophilic and nucleophilic compounds.[35] Additionally, trans-
cycloalkenes are more reactive than cis-cycloalkenes in 1,3-di-
polar cycloadditions with picryl azide, with rate accelerations of
greater than 104.[36]
Scheme 7. [4+2] Cycloaddition of tetrazine 2 and trans-alkene 12.
Silyl enol ether 18 reacted with benzyl azide in a [4+2] cyclo-
addition (Scheme 8) despite its sluggish reactivity with dienes
2 and 5. Triazoles 35 and 36 were formed in two days and
isolated in 62 % isolated yield over three steps from the 1,3-
diene. The regioselectivity of this cycloaddition reaction was
poor, which is typical for these reactions.[11,34]
(6)
(7)
Scheme 8. [4+2] Cycloaddition of benzyl azide and silyl enol ether 18.
The slow formation and regiochemistry of compound 34
were unexpected. Silyl enol ether 18 should react easily with
trans-Alkenes 16 and 19 also underwent 1,3-dipolar cyclo-
tetrazine 2 because the alkene should be particularly electron- additions with benzyl azide (Scheme 9). Triazoles 37, 38, and
rich.[29,30] The steric strain between the phenyl groups on 39 were formed in one day in 64 %, 34 %, and 75 % yields,
1
tetrazine 2 and the substituents on the ring of trans-alkene 18 respectively, as determined by H NMR spectroscopy. The high
leads to increased reaction time. The nucleophilic character of regioselectivity of the reaction of alkene 19 with benzyl azide
the trans double bond in 18 likely accounts for the observed to form triazole 39 is unusual because 1,3-dipolar cycloaddi-
regiochemistry of this reaction.[32] Stepwise reaction mecha- tions are typically unselective.[11,34] The cycloadducts were not
nisms have been proposed for the reactions of tetrazine 2 with stable to chromatography, so they were characterized from the
strong nucleophiles. Strained silyl enol ether 18 likely attacks unpurified reaction mixtures. Upon exposure to mild acid (SiO2),
the carbon atom of 2 forming a zwitterionic intermediate,[33] the triazoles decomposed to form new products through the
and subsequent ring closure leads to products with different loss of nitrogen gas and subsequent rearrangements (see Sup-
regiochemistry.
porting Information for details).[37]
Trisubstituted alkene 19, which is isomeric to silyl enol ether
The thermal rearrangement of sterically hindered trans-
18, did not react with tetrazine 2. After 50 days, trans-alkene alkene 12 outcompeted the [4+2] cycloaddition with benzyl
19 was consumed to yield dimer 30 in 88 % yield [Equation (8)]. azide (Scheme 10). The major product of the reaction of benzyl
No other compounds were observed, and the concentration of azide with trans-alkene 12 was oxasilacyclopentane 26, the
tetrazine 2 did not decrease during the reaction time, as deter- product of thermal rearrangement. Isomeric triazoles 40 and 41
1
mined by H NMR spectroscopy.
were formed in only 35 % and 4 % yields, respectively.[38]
Eur. J. Org. Chem. 2016, 2933–2943
www.eurjoc.org
2937
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Heterocycle 43 was formed in 76 % yield over three steps from
the starting diene. The high reactivity of trans-alkene 12 with
DEAD contrasts markedly with its low reactivity with other di-
enes (Scheme 5 and Scheme 7).
Scheme 11. Formal [4+2] cycloaddition of DEAD with trans-alkene 12.
Scheme 9. Reactions of trans-alkenes 16 and 19 with benzyl azide.
The difference in reactions with DEAD compared to other
dienes may arise because the mechanisms are different.[17] Un-
like concerted Diels–Alder reactions, the reaction between
trans-alkene 12 and DEAD is expected to proceed by a stepwise
mechanism through a zwitterionic intermediate.[39] The nucleo-
philic trans double bond attacks the electrophilic nitrogen atom
on DEAD, forming zwitterion 44 (Scheme 12). Ring closure of
the anionic oxygen atom onto the ꢀ-silyl carbocation forms
heterocycle 43.
Scheme 10. 1,3-Dipolar cycloaddition of alkene 12 and benzyl azide.
The higher reactivity of trans-alkenes with benzyl azide when
compared to the reactivity with tetrazine 2 is likely influenced
by steric and electronic effects. Benzyl azide is less sterically
hindered than tetrazine 2, which facilitates reactivity with sub-
stituted trans-alkenes. Additionally, incorporating an electron-
withdrawing group or other substituents near the dipolarophile
has been shown to decrease activation energies for cycloaddi-
tions with azides.[34]
Scheme 12. Proposed mechanism for formation of 43.
The stereochemical outcome of the addition of DEAD to
trans-alkene 12 is consistent with a stepwise reaction. Ring clo-
sure of intermediate 44 could lead to two different diastereo-
mers of product (Scheme 12). Trapping of intermediate 44 with-
out a change in conformation would preserve the stereochem-
istry of the trans double bond in the product. If bond rotation
occurred in intermediate 44, however, subsequent ring closure
would give diastereomer 45. Only diastereomer 43 was ob-
served, suggesting that bond rotation had not occurred. The
retention of configuration observed likely results because the
Stepwise Reactions with DEAD
Considering the reactivity of substituted trans-cycloalkenes
with azides, [4+2] cycloadditions involving less sterically de-
manding electrophilic dienes were explored. trans-Alkenes were
predicted to be particularly reactive with electrophilic dienes.
Previous studies indicated that the trans double bond of oxasi-
lacycloheptene 12 is nucleophilic.[17,24]
The reaction of sterically hindered trans-alkene 12 with ꢀ-silyl carbocation is configurationally stable due to the hyper-
DEAD proceeded rapidly at room temperature (Scheme 11). conjugation between the σSi–C bond and the carbocation.[24,40]
Eur. J. Org. Chem. 2016, 2933–2943
www.eurjoc.org
2938
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Bicyclic alkenes 15 and 17 reacted rapidly with DEAD to give Experimental Section
the ene-type products 40 and 41 [Equation (9)]. Even trans-
General Methods: See supporting information for full details. The
synthesis and characterization of compounds 2,[43] 9,[44] 10,[45]
14,[19] 13, 16, 18, and 20–23 were reported previously.[17]
alkene 17, which was unreactive in a Diels–Alder cycloaddition
with isobenzofuran 5, underwent a rapid reaction with DEAD.
Alkenes 46 and 47 were formed with complete control of rela-
tive stereochemistry.
Vinylsilacyclopropane 11: The synthesis of trans-oxasilacyclo-
heptene 11 was adapted from a reported procedure.[19] To a solu-
tionofdiene9(0.012g,0.10mmol),thering-fusedcyclohexane-silacyclo-
propane 10 (0.034 g, 0.15 mmol), and mesitylene (0.0020 mL,
0.014 mmol, internal standard) in C6D6 (0.64 mL) in a J. Young NMR
tube was added AgOCOCF3 (0.030 mL, 0.030
M in C6D6). Vinylsila-
1
cyclopropane 11 was formed in 10 min in 80 % yield based on H
NMR spectroscopic analysis of the area of a peak of the standard
(δ = 6.71 ppm) and the area of the alkene CH peak (δ = 5.32 ppm).
Vinylsilacyclopropane 11 could not be purified and was used with-
out further purification: 1H NMR (600 MHz, C6D6, diagnostic peaks):
δ = 5.32 (dt, J = 7.4, 1.1 Hz, 1 H), 2.32–2.30 (m, 2 H), 1.83–1.77 (m,
1 H), 1.60–1.54 (m, 3 H), 1.12 (s, 9 H), 1.03 (s, 9 H), 0.57 (dd, J = 11.1,
8.6 Hz, 1 H) ppm. 13C NMR (125 MHz, C6D6, diagnostic peaks): δ =
126.8 (CH), 30.6 (CH3), 30.2 (CH2), 30.1 (CH3), 29.5 (CH2), 27.9 (CH2),
5.6 (CH2) ppm. 29Si NMR (99.3 MHz, C6D6): δ = –49.4 ppm.
(9)
The regiochemistry of alkenes 46 and 47 was opposite of
what was expected. Because these reactions likely proceed
through stepwise mechanisms,[39] bicyclic alkenes 15 and 17
should react similar to allylic silanes, forming a ꢀ-silyl carbo-
cation in the first step (Figure 3).[41] Instead, reactions pro-
ceeded via an intermediate involving a tertiary γ-silyl carboca-
tion (Figure 3),[42] possibly because it is stabilized by the addi-
tional cycloalkyl group.
trans-Oxasilacycloheptene 12: To a solution of vinylsilacyclo-
propane 11 (0.021 g, 0.080 mmol) in C6D6 (0.67 mL) was added
benzaldehyde (0.010 mL, 0.10 mmol). trans-Oxasilacycloheptene 12
was formed in 10 min in 99 % yield from vinylsilacyclopropane 11
(79 % over two steps) based on 1H NMR spectroscopic analysis of
the area of a peak of the standard (δ = 6.71 ppm) and the area of
the alkene CH peak (δ = 6.01 ppm). trans-Oxasilacycloheptene 12
could not be purified and was used without further purification: 1H
NMR (500 MHz, C6D6, diagnostic peaks): δ = 7.28–7.27 (m, 2 H),
7.20–7.17 (m, 2 H), 7.11–7.07 (m, 1 H), 6.01 (ddd, J = 17.6, 12.9,
4.6 Hz, 1 H), 4.99 (d, J = 17.6 Hz, 1 H), 4.32 (s, 1 H), 2.45–2.39 (m, 1
H), 2.14–2.10 (m, 1 H), 1.95 (dd, J = 12.2, 4.6 Hz, 1 H), 1.65–1.58 (m,
1 H), 1.45–1.37 (m, 3 H), 1.30–1.26 (m, 1 H), 1.10 (s, 9 H), 0.95 (s, 9
H), 0.82–0.74 (m, 1 H) ppm. 13C NMR (125 MHz, C6D6, diagnostic
peaks): δ = 133.6 (CH), 133.5 (CH), 128.3 (CH), 128.0 (CH), 127.8 (CH),
84.3 (CH), 35.7 (CH2), 29.3 (CH3), 28.5 (CH3), 25.9 (CH2), 25.2 (CH2),
19.5 (CH2) ppm. 29Si NMR (99.3 MHz, C6D6): δ = 1.7 ppm.
Figure 3. Possible intermediates in the reaction of alkene 15 with DEAD.
Bicyclic Ether 25: To a solution of trans-oxasilacycloheptene 12
(0.066 mmol) in C6D6 (0.500 mL) was added 1,3-diphenylisobenzo-
furan 5 (0.021 g, 0.077 mmol). The NMR tube was flame-sealed
under vacuum and the reaction mixture was monitored by 1H NMR
spectroscopy until trans-oxasilacycloheptene 12 was consumed
Conclusions
trans-Oxasilacycloheptenes underwent a variety of [4+2] cyclo-
addition reactions. Unhindered trans-alkenes and strained bi-
cyclic trans-alkenes reacted rapidly with dienes. The relative re-
activity of these strained alkenes was influenced by the substi-
tution pattern of the seven-membered ring. Increasing substitu-
tion of the trans double bond or at the allylic position of the
double bond decreased the reactivity of these trans-alkenes in
concerted [4+2] cycloadditions. Reactivity was increased, how-
1
(4 d). Alkene 26 was formed in 72 % yield based on H NMR spec-
troscopic analysis of the area of a peak of the standard (δ =
6.71 ppm) and the area of one of the alkene CH2 peaks (δ =
4.82 ppm). The spectroscopic data are consistent with the data re-
ported in the Supporting Information for alkene 26. The major dia-
stereomer of bicyclic ether 25 was also formed in 4 d in 25 % yield
based on 1H NMR spectroscopic analysis of the area of a peak of
the standard (δ = 6.71 ppm) and the area of the benzylic ether
ever, by additional distortion of the trans double bond (in- peak (δ = 5.04 ppm). The minor diastereomer of bicyclic ether 25
was formed in 3 % yield based on 1H NMR spectroscopic analysis
creased strain), despite increasing substitution. Although some
of the area of a peak of the standard (δ = 6.71 ppm) and the area
of the benzylic ether peak (δ = 4.47 ppm). The NMR tube was
opened and the reaction mixture was concentrated in vacuo. Purifi-
cation by flash chromatography (2:98 EtOAc/hexanes) afforded bi-
cyclic ether 25 as a yellow crystalline solid (0.008 g, 19 % over three
steps): m.p. 158–160 °C. 1H NMR (500 MHz, C6D6): δ = 8.12–8.11 (m,
2 H), 7.83–7.77 (m, 1 H), 7.73–7.72 (m, 2 H), 7.41–7.37 (m, 3 H), 7.29–
7.21 (m, 4 H), 7.13–7.10 (m, 2 H), 7.06–6.99 (m, 5 H), 5.04 (s, 1 H),
3.30 (d, J = 5.0 Hz, 1 H), 2.99–2.95 (m, 1 H), 2.13–2.10 (m, 1 H), 1.83–
1.76 (m, 1 H), 1.60–1.57 (m, 1 H), 1.17 (s, 9 H, and m, 2 H), 1.08–1.02
highly substituted alkenes were unreactive in concerted cyclo-
additions, stepwise reactions with DEAD proceeded rapidly de-
spite low reactivity with dienes 2 and 5. These results indicate
that steric strain can drastically influence the reactivity of a
compound, overriding the impact ring strain has on reactivity.
Even though sterically hindered compounds react slowly in
concerted reactions, reactivity can be restored by stepwise reac-
tions that avoid destabilizing steric interactions in the rate-de-
termining step.
Eur. J. Org. Chem. 2016, 2933–2943
www.eurjoc.org
2939
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(m, 1 H), 0.96 (s, 9 H, and m, 3 H), 0.76–0.70 (m, 1 H), 0.42–0.34 (m,
(s, 1.4 H), 0.58 (s, 9 H), 0.31 (dd, J = 15.1, 13.3 Hz, 1 H, and m, 2.2
1 H), –0.34 to –0.40 (m, 1 H). 13C NMR (125 MHz, C6D6): δ = 151.4 H), –0.27 (s, 1.4 H), –0.49 (s, 9 H) ppm. 13C NMR (100 MHz, C6D6):
(C), 145.1 (C), 144.1 (C), 140.0 (C), 139.7 (C), 129.5 (CH), 129.13 (CH),
129.06 (CH), 129.0 (CH), 128.7 (CH), 128.3 (CH), 128.0 (CH), 127.7
(CH), 127.6 (CH), 126.7 (CH), 126.3 (CH), 124.7 (CH), 120.2 (CH), 92.8
(C), 91.9 (C), 89.6 (CH), 66.3 (CH), 45.7 (CH), 44.4 (C), 34.3 (CH2), 29.5
(CH3), 28.8 (CH3), 24.5 (CH2), 24.2 (CH2), 23.5 (CH2), 23.3 (CH2), 22.11
δ = 151.0 (C), 150.2 (C), 147.3 (C), 144.4 (C), 144.0 (C), 143.8 (C),
140.5 (C), 140.4 (C), 139.8 (C), 139.5 (C), 132.4 (CH), 130.0 (CH), 129.8
(CH), 129.1 (CH), 129.01 (CH), 128.97 (CH), 128.83 (CH), 128.78 (CH),
128.5 (CH), 128.3 (CH), 127.9 (CH), 127.8 (CH), 127.73 (CH), 127.70
(CH), 127.6 (CH), 127.4 (CH), 127.2 (CH), 127.1 (CH), 126.8 (CH), 126.7
(CH), 126.4 (CH), 123.9 (CH), 123.1 (CH), 123.0 (CH), 121.3 (CH), 117.4
(CH), 96.3 (C), 93.4 (C), 90.5 (C), 90.2 (C), 83.0 (CH), 82.7 (CH), 80.5
(CH), 80.1 (CH), 56.0 (C), 55.7 (C), 54.2 (CH), 53.9 (CH), 29.7 (CH3),
28.8 (CH3), 28.2 (CH3), 27.9 (CH3), 22.6 (C), 22.2 (C), 21.1 (C), 22.0 (C),
18.4 (CH3), 17.9 (CH3), 8.8 (CH2), 6.0 (CH2), 3.0 (CH3), 1.6 (CH3) ppm.
(C), 22.07 (C), 15.8 (CH2) ppm. IR (ATR): ν = 1310, 1069, 823 cm–1
.
˜
HRMS (ESI) m/z calcd. for C44H52NaO2Si [M + Na]+ 663.3629, found
663.3654.
Bicyclic Ether 27: To a solution of 1-vinylcyclopentene (0.0087 mL,
8.64
cyclopropane 10 (0.020 g, 0.090 mmol) in C6D6 (0.480 mL) was
added AgOCOCF3 (0.015 mL, 0.050 in C6D6). After 10 min, benzal-
M in Et2O, 0.075 mmol) and the ring-fused cyclohexane-sila-
IR (ATR): ν = 1250, 1084, 837 cm–1. HRMS (ESI) m/z calcd. for
˜
C43H54NaO3Si2 [M + Na]+ 697.3507, found 697.3505.
M
dehyde (0.008 mL, 0.08 mmol) was added followed by isobenzo-
furan 5 (0.020 g, 0.075 mmol). Ether 27 was formed in 10 min as a
mixture of diastereomers (93:7 dr) in 74 % yield from 1-vinylcyclo-
pentene based on 1H NMR spectroscopic analysis of the area of a
peak of the standard (δ = 6.71 ppm) and the area of the benzylic
CH peak (δ = 4.47 ppm). The reaction mixture was concentrated in
vacuo. Purification by flash chromatography (3:97 EtOAc/hexanes)
afforded ether 27 as a white solid (0.029 g, 62 % over three steps).
1H NMR spectroscopic analysis showed that ether 27 was isolated
Pyrazine 31: To
(0.077 mmol) in C6D6 (0.87 mL) was added 3,6-diphenyl-s-tetrazine
(0.022 g, 0.093 mmol). After 10 min, pyridazine 31 was formed in
90 % yield based on H NMR spectroscopic analysis of the area of
a solution of trans-oxasilacycloheptene 14
1
a peak of the standard (δ = 6.71 ppm) and the area of the benzylic
CH peak (δ = 4.77 ppm). All attempts to purify pyridazine 31 re-
sulted in decomposition. Spectroscopic data for 31 were collected
on the unpurified reaction mixture: 1H NMR (600 MHz, C6D6,
diagnostic peaks): δ = 7.23–7.20 (m, 2 H), 4.77 (d, J = 9.6 Hz, 1 H),
3.39–3.36 (m, 1 H), 3.17 (t, J = 5.9 Hz, 1 H), 1.00 (s, 9 H), 0.96 (s, 9
H), 0.79–0.74 (m, 1 H), 0.27 (d, J = 7.2 Hz, 3 H) ppm. 13C NMR
(150 MHz, C6D6, diagnostic peaks): δ = 162.1 (C), 158.3 (C), 130.2
(CH), 87.5 (CH), 42.9 (CH), 40.2 (CH), 29.4 (CH3), 28.6 (CH3), 17.8 (CH3),
17.5 (CH2) ppm. HRMS (APCI) m/z calcd. for C34H43N2OSi [M + H]+
523.3139, found 523.3155.
1
as a mixture of diastereomers in a 90:10 ratio: m.p. 137–141 °C. H
NMR (400 MHz, C6D6): δ = 8.15 (dd, J = 8.4, 1.2 Hz, 2 H), 8.06 (dd,
J = 8.4, 1.2 Hz, 2 H), 7.74–7.70 (m, 1 H), 7.65–7.61 (m, 1 H), 7.39–
7.35 (m, 2 H), 7.28–7.23 (m, 3 H), 7.20–7.17 (m, 2 H), 7.10–7.04 (m,
5 H), 6.99–6.94 (m, 1 H), 4.47 (d, J = 10.4 Hz, 1 H), 3.68 (d, J =
12.8 Hz, 1 H), 2.58 (dd, J = 10.0, 6.4 Hz, 1 H), 1.86–1.70 (m, 2 H),
1.65–1.58 (m, 1 H), 1.49 (d, J = 15.2 Hz, 1 H), 1.25–1.19 (m, 1 H),
0.97 (s, 9 H), 0.96–0.94 (m, 1 H), 0.75 (s, 9 H), 0.68–0.59 (m, 1 H),
0.38 (dd, J = 15.2, 13.2 Hz, 1 H) ppm. 13C NMR (125 MHz, C6D6): δ =
149.4 (C), 146.1 (C), 144.9 (C), 140.5 (C), 140.4 (C), 128.9 (CH), 128.7
(CH), 128.3 (CH), 128.1 (CH), 128.00 (CH), 127.98 (CH), 127.7 (CH),
127.4 (CH), 126.74 (CH), 126.70 (CH), 126.6 (CH), 122.1 (CH), 121.9
(CH), 94.1 (C), 89.5 (C), 81.0 (CH), 65.6 (C), 54.5 (CH), 51.7 (CH), 29.9
(CH3), 29.7 (CH2), 28.4 (CH3), 25.1 (CH2), 22.0 (C), 21.7 (C), 21.2 (CH2),
Pyrazine 32: To
a solution of trans-oxasilacycloheptene 12
(0.068 mmol) in C6D6 (0.47 mL) was added 3,6-diphenyl-s-tetrazine
(0.017 g, 0.073 mmol). After 4 d, pyridazine 32 was formed in 15 %
yield based on 1H NMR spectroscopic analysis of the area of a peak
of the standard (δ = 6.71 ppm) and the area of the benzylic ether
peak (δ = 5.11 ppm). Alkene 26 was also formed in 85 % yield
based on 1H NMR spectroscopic analysis of the area of a peak of
the standard (δ = 6.71 ppm) and the area of one of the alkene CH2
peaks (δ = 4.82 ppm). The spectroscopic data are consistent with
the data reported in the Supporting Information for alkene 26. All
attempts to purify pyridazine 32 resulted in decomposition. Spec-
troscopic data for 32 was collected on the unpurified reaction mix-
ture: 1H NMR (600 MHz, C6D6, diagnostic peaks): δ = 8.26 (dd, J =
8.4, 1.2 Hz, 1 H), 8.00–7.98 (m, 2 H), 5.11 (s, 1 H), 3.88–3.85 (m, 1 H),
3.29 (d, J = 4.3 Hz, 1 H), 1.04 (s, 9 H), 0.96 (s, 9 H), 0.50–0.44 (m, 1
H) ppm. 13C NMR (100 MHz, C6D6, diagnostic peaks): δ = 163.7 (C),
159.4 (C), 129.4 (CH), 128.9 (CH), 91.4 (CH), 46.0 (CH), 30.7 (CH),
29.5 (CH3), 28.8 (CH3), 26.2 (CH2) ppm. HRMS (ESI) m/z calcd. for
C38H49N2OSi [M + H]+ 577.3609, found 577.3606.
6.3 (CH2) ppm. IR (ATR): ν = 1063, 822 cm–1. HRMS (APCI) m/z calcd.
˜
for C42H49O2Si [M + H]+ 613.3496, found 613.3497.
Bicyclic Ether 29: To a solution of trans-oxasilacycloheptene 19
(0.051 mmol) in C6D6 (0.50 mL) was added 1,3-diphenylisobenzo-
furan (0.021 g, 0.078 mmol). After 50 d, the major diastereomer
of bicyclic ether 29 was formed in 32 % yield based on 1H NMR
spectroscopic analysis of the area of a peak of the standard (δ =
6.71 ppm) and the area of the benzylic ether peak (δ = 5.04 ppm).
The minor diastereomer of bicyclic ether 29 was formed in 6 %
yield based on 1H NMR spectroscopic analysis of the area of a peak
of the standard (δ = 6.71 ppm) and the area of the benzylic ether
peak (δ = 4.88 ppm). Alkene 30 was also formed in 31 % yield
based on 1H NMR spectroscopic analysis of the area of a peak of
the standard (δ = 6.71 ppm) and the area of one of the alkene CH2
peaks (δ = 5.33 ppm). The spectroscopic data are consistent with
the data reported in the Supporting Information for alkene 30. The
NMR tube was opened and the reaction mixture was concentrated
Pyrazine 33: To
a solution of trans-oxasilacycloheptene 16
(0.064 mmol) in C6D6 (0.50 mL) was added 3,6-diphenyl-s-tetrazine
(0.018 g, 0.076 mmol). After 5 d, pyridazine 33 was formed in 21 %
yield based on 1H NMR spectroscopic analysis of the area of a peak
of the standard (δ = 6.71 ppm) and the area of the benzyl ether
in vacuo. Purification by flash chromatography (2:98 EtOAc/hex- peak (δ = 4.98 ppm). The NMR tube was opened and the reaction
anes) afforded bicyclic ether 29 as a yellow solid (0.010 g, 29 %
over three steps): m.p. 118–121 °C. 1H NMR (500 MHz, C6D6): δ =
8.32–8.31 (m, 0.3 H), 8.11–8.09 (m, 2 H), 8.01–8.00 (m, 2 H), 7.97–
mixture was purified by flash chromatography (1:99 MeOH/CH2Cl2)
to afford pyridazine 33 as a yellow solid. All purification attempts
resulted in minor decomposition of pyridazine 33. Spectroscopic
7.95 (m, 0.3 H), 7.54–7.52 (m, 1 H), 7.45–7.20 (m, 11 H), 7.11–7.05 data were collected on the product that was not analytically pure:
(m, 1.4 H), 7.05–6.97 (m, 4 H), 5.11 (d, J = 7.7 Hz, 1 H), 5.04 (d, J = 1H NMR (500 MHz, C6D6): δ = 7.63–7.62 (d, J = 7.3, 2 H), 7.47–7.46
7.7 Hz, 1 H), 4.88 (d, J = 8.9 Hz, 0.2 H), 3.74 (d, J = 8.9 Hz, 0.2 H),
2.95 (d, J = 13.1 Hz, 0.2 H, and d, J = 13.0 Hz, 1 H), 1.79 (s, 0.5 H),
1.37 (s, 3 H), 1.25 (s, 1.4 H), 1.02–1.01 (m, 0.2 H), 0.97 (s, 9 H), 0.86
(m, 2 H), 7.23–7.22 (m, 2 H), 7.15–7.12 (m, 5 H), 7.09–7.03 (m, 3 H),
7.00–6.97 (m, 1 H), 4.99–4.97 (d, J = 8.3 Hz, 1 H), 3.76 (dd, J = 9.8,
5.5 Hz, 1 H), 2.90–2.88 (m, 1 H), 2.60–2.58 (m, 1 H), 1.56–1.40 (m, 4
Eur. J. Org. Chem. 2016, 2933–2943
www.eurjoc.org
2940
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
H), 1.15–1.13 (m, 3 H), 1.09 (s, 9 H), 0.95–0.91 (m, 2 H), 0.77 (s, 9 H)
ppm. 13C NMR (100 MHz, C6D6): δ = 172.0 (C), 168.0 (C), 144.6 (C),
142.8 (C), 140.1 (C), 128.8 (CH), 128.6 (2 CH), 128.5 (2 CH), 128.3 (2
standard (δ = 6.71 ppm) and the area of the benzyl ether peak (δ =
5.23 ppm). Triazole 38 was formed in 33 % yield based on H NMR
spectroscopic analysis of the area of a peak of the standard (δ =
1
CH), 127.8 (CH), 127.4 (CH), 76.6 (CH), 48.4 (CH), 42.4 (CH), 41.2 (C), 6.71 ppm) and the area of the benzyl ether peak (δ = 5.05 ppm).
29.7 (CH3), 28.7 (CH3), 27.3 (CH2), 25.2 (CH2), 23.4 (CH2), 21.8 (C),
All purification attempts resulted in decomposition of triazoles 37
21.7 (CH2), 21.2 (C), 9.3 (CH2) ppm. IR (ATR): ν = 1689, 1050, 823 cm–
and 38. Spectroscopic data were collected on the unpurified reac-
˜
1. HRMS (ESI) m/z calcd. for C37H47N2OSi [M + H]+ 563.3452, found tion mixture: H NMR (500 MHz, C6D6, diagnostic peaks): δ = 7.31–
1
563.3451.
7.28 (m, 3 H), 7.21–7.19 (m, 2 H), 7.14–7.11 (m, 5 H), 7.09–7.00 (m,
5 H), 5.27 (d, J = 15.8 Hz, 0.5 H), 5.23 (d, J = 10.4 Hz, 1 H), 5.05 (d,
J = 10.2 Hz, 0.5 H), 4.86 (d, J = 15.4 Hz, 1.5 H), 4.44 (d, J = 15.4 Hz,
1 H), 4.37 (dd, J = 14.3, 1.4 Hz, 0.5 H), 3.28 (dd, J = 13.8, 2.6 Hz, 1
H), 2.82 (tt, J = 13.5, 4.8 Hz, 1 H), 2.15 (m, J = 16.2, 1.7 Hz, 1 H, and
dd, 1 H), 2.08–2.01 (m, 1 H), 1.89–1.81 (m, 1.5 H), 1.68 (dd, J = 13.5,
4.4 Hz, 1 H, and m, 1 H), 1.57–1.40 (m, 5.5 H), 1.02 (s, 4.5 H), 0.96
(s, 4.5 H), 0.93 (s, 9 H), 0.77 (s, 9 H) ppm. 13C NMR (125 MHz, C6D6,
diagnostic peaks): δ = 144.6 (C), 144.4 (C), 141.2 (C), 137.9 (C), 129.1
(CH), 129.0 (CH), 128.8 (CH), 128.7 (CH), 127.9 (CH), 127.7 (CH), 127.0
(CH), 126.9 (CH), 82.4 (C), 81.7 (C), 76.5 (CH), 71.9 (CH), 66.1 (CH),
56.0 (CH2), 53.0 (CH2), 50.0 (CH), 47.8 (CH), 29.3 (CH3), 28.6 (CH3),
28.4 (CH3), 26.1 (CH2), 24.7 (CH2), 24.6 (CH2), 22.8 (CH2) ppm. HRMS
(ESI) m/z calcd. for C30H44N3OSi [M + H]+ 490.3248, found 490.3263.
Bicyclic Compound 34: To a solution of trans-oxasilacycloheptene
18 (0.058 mmol) in C6D6 (0.50 mL) was added 3,6-diphenyl-1,2,4,5-
tetrazine (0.018 g, 0.076 mmol). After 84 d, 80 % of trans-oxasila-
cycloheptene 18 had converted to product. Bicyclic compound 34
was formed in 84 d in 53 % yield based on H NMR spectroscopic
analysis of the area of a peak of the standard (δ = 6.71 ppm) and
1
1
the area of the CH peak next to nitrogen (δ = 3.52 ppm). H NMR
spectroscopic analysis showed bicyclic compound 34 was a mixture
of diastereomers in a 95:5 ratio. This ratio was persevered from the
starting diene. It was characterized in situ: 1H NMR (600 MHz, C6D6,
diagnostic peaks): δ = 7.66–7.65 (m, 2 H), 7.63–7.61 (m, 2 H), 7.51–
7.50 (m, 1 H), 5.08 (d, J = 7.6 Hz, 1 H), 3.52 (dd, J = 13.5, 1.5 Hz, 1
H), 2.94–2.90 (m, 1 H), 1.09 (s, 9 H), 0.83 (s, 9 H), 0.32 (s, 9 H) ppm.
13C NMR (150 MHz, C6D6, diagnostic peaks): δ = 167.3 (C), 166.8 (C),
76.0 (CH), 49.0 (CH), 45.5 (CH), 29.4 (CH3), 28.7 (CH3), 3.3 (CH3) ppm.
Triazole 39: To
a solution of trans-oxasilacycloheptene 19
(0.0087 mmol) in C6D6 (0.450 mL) was added benzyl azide
(0.004 mL, 0.03 mmol). Triazole 39 was formed in 1 d in 75 % yield
based on 1H NMR spectroscopic analysis of the area of a peak of
the standard (δ = 6.74 ppm) and the area of the benzyl ether peak
(δ = 4.86 ppm). All purification attempts resulted in decomposition
of triazole 39. Spectroscopic data were collected on the unpurified
reaction mixture: 1H NMR (500 MHz, C6D6, diagnostic peaks): δ =
7.36–7.35 (m, 2 H), 7.24–7.23 (m, 2 H), 7.15–7.01 (m, 6 H), 4.86 (d,
J = 9.0 Hz, 1 H), 4.81 (d, J = 15.4 Hz, 1 H), 4.34 (d, J = 15.4 Hz, 1 H),
3.88 (d, J = 9.0 Hz, 1 H), 3.25 (dd, J = 13.4, 3.0 Hz, 1 H), 1.29 (s, 3
H), 0.88 (s, 9 H), 0.79 (s, 9 H), 0.03 (s, 9 H) ppm. 13C NMR (125 MHz,
C6D6, diagnostic peaks): δ = 143.9 (C), 137.6 (C), 129.3 (CH), 129.0
(CH), 128.2 (CH), 128.1 (CH), 85.1 (CH), 81.5 (CH), 77.7 (CH), 62.8 (CH),
52.9 (CH2), 29.0 (CH3), 28.3 (CH3), 22.0 (C), 21.3 (C), 10.5 (CH3), 5.2
(CH2), 0.8 (CH3) ppm. HRMS (APCI) m/z calcd. for C30H47NO2Si2 [M –
N2]+ 510.3218, found 510.3221.
Triazoles 35 and 36: To a solution of (Z)-trimethyl(penta-1,3-dien-
3-yloxy)silane[17] (0.017 g, 0.11 mmol) and the ring-fused cyclohex-
ane-silacyclopropane 10 (0.027 g, 0.12 mmol) in C6D6 (0.70 mL)
was added AgOCOCF3 (0.035 mL, 0.030
M in C6D6). After 10 min,
benzaldehyde (0.011 mL, 0.11 mmol) was added, followed by
benzyl azide (0.014 mL, 0.11 mmol). After 2 d, the NMR tube was
opened and the reaction mixture was concentrated in vacuo. 1H
NMR spectroscopic analysis of the unpurified reaction mixture
showed a mixture of regioisomers in a 59:41 (35/36) ratio. The re-
gioisomers were assigned by an HMBC experiment. 1H NMR spec-
troscopic analysis of the unpurified reaction mixture showed tri-
azoles 35 and 36 were mixtures of diastereomers in a 95:5 ratio.
This ratio was persevered from the starting diene. Purification by
flash chromatography (3:97 EtOAc/hexanes) afforded triazoles 35
and 36 as a colorless oil (0.035 g, 62 % over three steps): 1H NMR
(500 MHz, C6D6): δ = 7.24–7.17 (m, 4.2 H), 7.14–6.99 (m, 12.5 H),
5.32 (d, J = 16.8 Hz, 1 H), 5.03 (d, J = 9.5 Hz, 0.7 H, and d, J =
15.4 Hz, 0.7 H), 4.94 (d, J = 8.9 Hz, 1 H), 4.53 (d, J = 16.8 Hz, 1 H,
and dd, J = 13.5, 1.1 Hz, 1 H), 4.21 (d, J = 15.4 Hz, 0.7 H), 3.20 (dd,
J = 13.5, 2.9 Hz, 0.7 H), 2.09–2.00 (m, 2.7 H), 1.61 (dd, J = 15.5,
13.5 Hz, 1 H), 1.15 (d, J = 6.9 Hz, 2.1 H), 1.09 (s, 9 H), 1.03 (s, 6.3 H),
0.98 (s, 9 H, and m, 0.7 H), 0.92–0.90 (m, 0.7 H), 0.80 (s, 6.3 H), 0.57
(d, J = 6.9 Hz, 3 H), 0.35 (s, 6.3 H), 0.26 (s, 9 H) ppm. 13C NMR
(125 MHz, C6D6, major regioisomer 35): δ = 144.51 (C), 140.4 (C),
Triazoles 40 and 41: To a solution of trans-oxasilacycloheptene 12
(0.070 mmol) in C6D6 (0.50 mL) was added benzyl azide (0.009 mL,
0.08 mmol). After 3 d, triazole 40 was formed in 35 % yield based
on 1H NMR spectroscopic analysis of the area of a peak of the
standard (δ = 6.71 ppm) and the area of the one of the benzyl CH2
peaks (δ = 4.31 ppm). Triazole 41 was formed in 4 % yield based
on 1H NMR spectroscopic analysis of the area of a peak of the
standard (δ = 6.71 ppm) and the area of the one of the benzyl CH2
peaks (δ = 4.09 ppm). Alkene 26 was formed in 59 % yield based
129.3 (CH), 128.81 (CH), 127.8 (CH), 127.7 (CH), 127.4 (CH), 127.1 on 1H NMR spectroscopic analysis of the area of a peak of the
(CH), 96.9 (C), 83.8 (CH), 78.6 (CH), 52.4 (CH2), 50.6 (CH), 29.0 (CH3),
28.5 (CH3), 22.0 (C), 21.6 (C), 14.4 (CH3), 7.3 (CH2), 2.7 (CH3) ppm.
13C NMR (125 MHz, C6D6, minor regioisomer 36): δ = 144.54 (C),
136.8 (C), 129.4 (CH), 128.9 (CH), 128.78 (CH), 128.5 (CH), 127.9 (CH),
standard (δ = 6.71 ppm) and the area of one of the alkene CH2
peaks (δ = 4.82 ppm). The spectroscopic data are consistent with
the data reported in the Supporting Information for alkene 26. All
purification attempts resulted in decomposition of triazoles 40 and
127.5 (CH), 106.1 (C), 78.6 (CH, overlapping with major regioisomer), 41. Spectroscopic data were collected on the unpurified product:
63.8 (CH), 52.1 (CH2), 51.1 (CH), 29.2 (CH3), 28.1 (CH3), 21.8 (C), 21.2
1H NMR (600 MHz, C6D6, diagnostic peaks): δ = 5.44 (d, J = 14.9 Hz,
0.1 H), 4.95–4.91 (m, 0.1 H), 4.31 (d, J = 15.1 Hz, 1 H), 4.09 (d, J =
(C), 14.0 (CH3), 4.3 (CH2), 3.0 (CH3) ppm. IR (ATR): ν = 1470, 1061,
˜
822 cm–1. HRMS (APCI) m/z calcd. for C30H48N3O2Si2 [M + H]+ 14.9 Hz, 0.1 H), 3.61 (d, J = 15.1 Hz, 1 H), 3.39 (ddd, J = 15.1, 12.2,
538.3280, found 538.3295. C30H47N3O2Si2 (537.89): calcd. C 66.99, H
8.81; found C 67.39, H 8.70.
2.9 Hz, 1 H), 0.91 (s, 9 H), 0.80 (s, 9 H) ppm. 13C NMR (100 MHz,
C6D6, diagnostic peaks): δ = 96.6 (CH), 80.8 (CH), 71.9 (CH), 57.8
(CH), 57.1 (CH2), 53.3 (CH2), 29.0 (CH3), 28.5 (CH3) ppm. HRMS (ESI)
m/z calcd. for C31H49N4OSi [M + NH4]+ 521.3670, found 521.3649.
Triazoles 37 and 38: To a solution of trans-oxasilacycloheptene 16
(0.043 mmol) in C6D6 (0.50 mL) was added benzyl azide (0.010 mL,
0.080 mmol). After 1 d, triazole 37 was formed in 64 % yield based
on 1H NMR spectroscopic analysis of the area of a peak of the
Heterocycle 43: To a solution of diene 9 (0.017 g, 0.14 mmol) and
the ring-fused cyclohexane-silacyclopropane 10 (0.039 g,
Eur. J. Org. Chem. 2016, 2933–2943
www.eurjoc.org
2941
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
0.18 mmol) in C6H6 (0.88 mL) was added AgOCOCF3 (0.027 mL,
0.050 in C6H6). After 10 min, benzaldehyde (0.014 mL, 0.14 mmol)
(m, 6 H), 1.02 (s, 9 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 157.8
(C), 156.6 (C), 152.1 (C), 144.3 (C), 128.9 (CH), 128.1 (CH), 127.5 (CH),
125.7 (CH), 80.6 (CH), 65.6 (CH), 62.5 (CH2), 62.2 (CH2), 55.6 (CH),
29.2 (CH2), 28.8 (CH3), 28.3 (CH2), 28.2 (CH3), 27.9 (CH2), 25.9 (CH2),
22.3 (C), 21.9 (C), 16.3 (CH2), 15.0 (CH3), 14.9 (CH3) ppm. IR (ATR):
M
was added followed by diethyl azodicarboxylate (0.021 mL,
0.14 mmol). After 20 min, the reaction mixture was concentrated in
vacuo. Purification by flash chromatography (10:90 EtOAc/hexanes)
afforded heterocycle 43 as a white solid (0.060 g, 81 % over three
ν = 3283, 1708, 1229, 1011 cm–1. HRMS (ESI) m/z calcd. for
˜
steps): m.p. 157–159 °C. 1H NMR (500 MHz, CDCl3): δ = 7.32–7.23 C30H49N2O5Si [M + H]+ 545.3405, found 545.3409. C30H48N2O5Si
(m, 5 H), 4.84 (s, 1 H, and m, 1 H), 4.75–4.71 (m, 1 H), 4.38–4.27 (m,
2 H), 4.24–4.19 (m, 2 H), 1.85–1.83 (m, 1 H), 1.72–1.66 (m, 1 H), 1.59–
1.56 (m, 2 H), 1.38 (t, J = 7.1 Hz, 3 H, and m, 2 H), 1.32 (t, J = 7.1 Hz,
3 H, and m, 1 H), 1.28–1.23 (m, 2 H), 1.08 (s, 9 H, and m, 1 H), 1.00
(s, 9 H, and m, 1 H), 0.88–0.79 (m, 1 H) ppm. 13C NMR (125 MHz,
CDCl3): δ = 162.0 (C), 155.3 (C), 141.8 (C), 128.9 (CH), 127.4 (2 CH),
85.9 (CH), 82.2 (CH), 68.5 (CH), 65.0 (CH2), 62.3 (CH2), 47.8 (C), 30.1
(CH2), 28.6 (CH3), 27.8 (CH3), 25.6 (CH2), 24.7 (CH2), 21.6 (C), 21.5
(CH2), 21.4 (CH2), 21.2 (C), 19.7 (CH2), 14.8 (CH3), 14.3 (CH3) ppm. IR
(544.81): calcd. C 66.14, H 8.88; found C 65.86, H 8.91.
Acknowledgments
This research was supported by the National Science Founda-
tion (NSF), USA (CHE-1362709) and the NYU Dean's Undergrad-
uate Research Fund (grant to J. S. III). The authors thank Dr.
Chin Lin (NYU) for his assistance with NMR spectroscopy and
mass spectrometry. Dr. Chunhua Hu (NYU) is thanked for X-ray
analysis and the NYU Molecular Design Institute for purchasing
a single-crystal diffractometer.
(ATR): ν = 1690, 1645, 1095, 825 cm–1. HRMS (ESI) m/z calcd. for
˜
C
30H49N2O5Si [M + H]+ 545.3405, found 545.3456. C30H48N2O5Si
(544.81): calcd. C 66.17, H 8.88; found C 66.17, H 8.85.
Alkene 46: To a solution of 1-vinylcyclopentene (0.017 mL, 8.64
in Et2O, 0.15 mmol) and the ring-fused cyclohexane-silacyclo-
M
Keywords: Strained molecules · Steric hindrance ·
propane 10 (0.040 g, 0.18 mmol) in C6H6 (0.952 mL) was added
Cycloalkenes · Cycloaddition
AgOCOCF3 (0.030 mL, 0.050
M in C6H6). After 10 min, benzaldehyde
(0.016 mL, 0.15 mmol) was added followed by diethyl azodicarbox-
ylate (0.023 mL, 0.15 mmol). After 10 min, the reaction mixture was
concentrated in vacuo. Purification by flash chromatography (25:75
EtOAc/hexanes) afforded alkene 46 as a white solid (0.032 g, 42 %
over three steps): m.p. 162–166 °C. 1H NMR (500 MHz, C6D6): δ =
7.30 (d, J = 7.0 Hz, 2 H), 7.19–7.17 (m, 2 H), 7.10–7.07 (m, 1 H), 6.98–
6.34 (b, 1 H), 5.72–5.33 (b, 2 H), 4.64 (d, J = 10.0 Hz, 1 H), 4.06–3.96
(m, 4 H), 3.04–2.97 (b, 1 H), 2.19–2.16 (b, 1 H), 1.99–1.96 (b, 1 H),
1.85–1.71 (b, 1 H), 1.59–1.49 (m, 2 H), 1.38–1.33 (m, 1 H), 1.20 (s, 9
H), 1.01–0.97 (m, 6 H), 0.95 (s, 9 H) ppm. 13C NMR (150 MHz, C6D6):
δ = 157.9 (C), 156.2 (C), 151.4 (C), 145.2 (C), 128.9 (CH), 128.0 (CH),
127.5 (CH), 126.6 (CH), 81.4 (CH), 62.7 (CH2), 62.3 (CH2), 57.9 (CH),
57.5 (CH), 30.9 (CH2), 29.1 (CH2), 28.9 (CH3), 28.7 (CH3), 22.1 (C), 21.6
[1] S. Vázquez, P. Camps, Tetrahedron 2005, 61, 5147–5208.
[2] A. Duda, A. Kowalski, S. Penczek, H. Uyama, S. Kobayashi, Macromolecules
2002, 35, 4266–4270.
[3] F. R. Fischer, C. Nuckolls, Angew. Chem. Int. Ed. 2010, 49, 7257–7260;
Angew. Chem. 2010, 122, 7415–7418.
[4] O. Nuyken, S. D. Pask, Polymers 2013, 5, 361–403.
[5] M. R. Wilson, R. E. Taylor, Angew. Chem. Int. Ed. 2013, 52, 4078–4087;
Angew. Chem. 2013, 125, 4170–4180.
[6] M. F. Debets, S. S. van Berkel, J. Dommerholt, A. J. Dirks, F. P. J. T. Rutjes,
F. L. van Delft, Acc. Chem. Res. 2011, 44, 805–815.
[7] R. Selvaraj, J. M. Fox, Curr. Opin. Chem. Biol. 2013, 17, 753–760.
[8] K. Lang, L. Davis, S. Wallace, M. Mahesh, D. J. Cox, M. L. Blackman, J. M.
Fox, J. W. Chin, J. Am. Chem. Soc. 2012, 134, 10317–10320.
[9] R. Rossin, P. R. Verkerk, S. M. van der Bosch, R. C. M. Vulders, I. Verel, J.
Lub, M. S. Robillard, Angew. Chem. Int. Ed. 2010, 49, 3375–3378; Angew.
Chem. 2010, 122, 3447–3450.
[10] M. T. Taylor, M. L. Blackman, O. Dmitrenko, J. M. Fox, J. Am. Chem. Soc.
2011, 133, 9646–9649.
[11] K. Tomooka, S. Miyasaka, S. Motomura, K. Igawa, Chem. Eur. J. 2014, 20,
7598–7602.
(C), 14.91 (CH3), 14.86 (CH3), 14.6 (CH2) ppm. IR (ATR): ν = 3283,
˜
1711, 1230, 1011 cm–1. HRMS (APCI) m/z calcd. for C28H45N2O5Si [M
+ H]+ 516.3092, found 517.3091.
Alkene 47: To
a solution of 1-vinylcycloheptene (0.018 g,
0.15 mmol) and the ring-fused cyclohexane-silacyclopropane 10
(0.037 g, 0.17 mmol) in C6H6 (0.970 mL) was added AgOCOCF3
[12] K. B. Wiberg, Angew. Chem. Int. Ed. Engl. 1986, 25, 312–322; Angew. Chem.
1986, 98, 312–322.
(0.030 mL, 0.050
M in C6H6). After 10 min, benzaldehyde (0.016 mL,
0.15 mmol) was added followed by diethyl azodicarboxylate
(0.024 mL, 0.15 mmol). After 20 min, the reaction mixture was con-
centrated in vacuo. Purification by flash chromatography (15:85
EtOAc/hexanes) afforded alkene 47 as a white solid (0.044 g, 54 %
over three steps). Purification gave a mixture of alkene 47 and a
minor isomer (98:2 alkene 47:minor isomer). The minor isomer
could not be characterized because of an insufficient yield. Alkene
47 was a mixture of rotamers. The characterization data reported
[13] E. J. Corey, F. A. Carey, R. A. E. Winter, J. Am. Chem. Soc. 1965, 87, 934–
935.
[14] R. Hoffmann, Y. Inoue, J. Am. Chem. Soc. 1999, 121, 10702–10710.
[15] Y. Inoue, T. Ueoka, T. Kuroda, T. Hakushi, J. Chem. Soc. Perkin Trans. 2
1983, 983–988.
[16] A. Krebs, K.-I. Pforr, W. Raffay, B. Thölke, W. A. König, I. Hardt, R. Boese,
Angew. Chem. Int. Ed. Engl. 1997, 36, 159–160; Angew. Chem. 1997, 109,
159–161.
[17] J. R. Sanzone, K. A. Woerpel, Angew. Chem. Int. Ed. 2016, 55, 790–793;
Angew. Chem. 2016, 128, 800–803.
1
below were recorded at 22 °C. A variable temperature H NMR ex-
periment was performed at 70 °C and the rotamer peaks sharpened:
m.p. 137–141 °C. 1H NMR (500 MHz, C6D6, 22 °C): δ = 7.32 (d, J =
7.2 Hz, 2 H), 7.19–7.17 (m, 2 H), 7.11–7.08 (m, 1 H), 6.98–6.40 (b, 1
H), 5.88–5.74 (b, 1 H), 5.05 (d, J = 9.6 Hz, 1 H), 4.98–4.83 (m, 1 H),
4.07–3.95 (m, 4 H), 3.03–2.95 (b, 1 H), 2.21–2.10 (b, 2 H), 1.66–1.44
(m, 5 H), 1.36–1.28 (m, 3 H), 1.18 (s, 9 H), 1.00 (s, 9 H), 0.98–0.95 (m,
[18] A. Sella, H. Basch, S. Hoz, J. Am. Chem. Soc. 1996, 118, 416–420.
[19] M. A. Greene, M. Prévost, J. Tolopilo, K. A. Woerpel, J. Am. Chem. Soc.
2012, 134, 12482–12484.
[20] J. A. Marshall, Acc. Chem. Res. 1980, 13, 213–218.
[21] G. W. Wijsman, G. A. Iglesias, M. C. Beekman, W. H. de Wolf, F. Bickelhaupt,
H. Kooijman, A. L. Spek, J. Org. Chem. 2001, 66, 1216–1227.
[22] H. Mayr, B. Kempf, A. R. Ofial, Acc. Chem. Res. 2003, 36, 66–77.
[23] X-ray crystallography and nuclear Overhauser enhancement measure-
ments confirmed the relative configuration for all products. Details are
provided as Supporting Information.
1
6 H) ppm. H NMR (500 MHz, C6D6, 70 °C): δ = 7.36–7.34 (m, 2 H),
7.20–7.17 (m, 2 H), 7.11–7.07 (m,1 H), 6.42–6.37 (b, 1 H), 6.07–5.84
(b, 1 H), 5.06–5.04 (d, J = 10.0 Hz, 1 H), 4.89–4.87 (m, 1 H), 4.10–
3.97 (m, 4 H), 2.99–2.95 (m, 1 H), 2.23–2.09 (m, 2 H), 1.66–1.57 (m,
2 H), 1.52–1.44 (m, 3 H), 1.41–1.26 (m, 3 H), 1.10 (s, 9 H), 1.06–1.03
[24] B. Hurlocker, C. Hu, K. A. Woerpel, Angew. Chem. Int. Ed. 2015, 54, 4295–
4298; Angew. Chem. 2015, 127, 4369–4372.
Eur. J. Org. Chem. 2016, 2933–2943
www.eurjoc.org
2942
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[25] D. Crich, K. Ranganathan, J. Am. Chem. Soc. 2005, 127, 9924–9929.
[26] The rearrangement of trans-alkene 12 to oxasilacyclopentane 26 is rela-
tively slow, requiring 21 h to consume 50 % of trans-alkene 12.
[27] S. Deuri, P. Phukan, J. Mol. Struct. THEOCHEM 2010, 945, 64–70.
[28] H. Zhang, A. Wakamiya, S. Yamaguchi, Org. Lett. 2008, 10, 3591–3594.
[29] F. Liu, Y. Liang, K. N. Houk, J. Am. Chem. Soc. 2014, 136, 11483–11493.
[30] M. L. Blackman, M. Royzen, J. M. Fox, J. Am. Chem. Soc. 2008, 130, 13518–
13519.
[31] The high strain energy of bicyclic compound 34 likely contributed to
the facile decomposition.
[32] Diels–Alder cycloadditions with tetrazine 2 are likely concerted, but a
stepwise mechanism cannot be ruled out for the reaction of silyl enol
ether 18 and diene 2.
[36]
[37]
[38]
K. J. Shea, J.-S. Kim, J. Am. Chem. Soc. 1992, 114, 4846–4855.
D. S. Reddy, W. R. Judd, J. Aubé, Org. Lett. 2003, 5, 3899–3902.
Triazoles 40 and 41 were also unstable to purification by column chro-
matography, but the decomposition products could not be identified.
T. Kanzian, H. Mayr, Chem. Eur. J. 2010, 16, 11670–11677.
T. Müller, M. Juhasz, C. A. Reed, Angew. Chem. Int. Ed. 2004, 43, 1543–
1546; Angew. Chem. 2004, 116, 1569–1572.
[39]
[40]
[41]
J. B. Lambert, Tetrahedron 1990, 46, 2677–2689.
[42] J. Coope, V. J. Shiner Jr., M. W. Ensinger, J. Am. Chem. Soc. 1990, 112,
2835–2837.
[43] R. Selvaraj, J. M. Fox, Tetrahedron Lett. 2014, 55, 4795–4797.
[44] L. T. Kliman, S. N. Mlynarski, G. E. Ferris, J. P. Morken, Angew. Chem. Int.
Ed. 2012, 51, 521–524; Angew. Chem. 2012, 124, 536–539.
[45] T. G. Driver, K. A. Woerpel, J. Am. Chem. Soc. 2003, 125, 10659–10663.
[33] M. J. Haddadin, E. H. Ghazvini Zadeh, Tetrahedron Lett. 2010, 51, 1654–
1656.
[34] F. Schoenebeck, D. H. Ess, G. O. Jones, K. N. Houk, J. Am. Chem. Soc.
2009, 131, 8121–8133.
[35] S. Bräse, C. Gil, K. Knepper, V. Zimmerman, Angew. Chem. Int. Ed. 2005,
44, 5188–5240; Angew. Chem. 2005, 117, 5320–5374.
Received: March 17, 2016
Published Online: May 30, 2016
Eur. J. Org. Chem. 2016, 2933–2943
www.eurjoc.org
2943
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim