Computational and synthetic studies with tetravinylethylenes

Article abstract of DOI:10.1021/jo5021294

Computational and experimental studies offer fresh insights into the neglected tetravinylethylene class of compounds. Both the structures and the outcomes of exploratory reactions of the parent hydrocarbon are predicted and explained in detail through hig

Full text of DOI:10.1021/jo5021294

Article
pubs.acs.org/joc
Computational and Synthetic Studies with Tetravinylethylenes
Erik J. Lindeboom,† Anthony C. Willis,† Michael N. Paddon-Row,*,‡ and Michael S. Sherburn*,†
†Research School of Chemistry, The Australian National University, Canberra ACT 0200, Australia
‡School of Chemistry, The University of New South Wales, Sydney NSW 2052, Australia
S
* Supporting Information
ABSTRACT: Computational and experimental studies offer
fresh insights into the neglected tetravinylethylene class of
compounds. Both the structures and the outcomes of
exploratory reactions of the parent hydrocarbon are predicted
and explained in detail through high-level composite ab initio
MO G4(MP2) computational studies.
Skattebøl’s ingenious synthesis of TVE is a milestone in
hydrocarbon chemistry and is the only reported synthesis of
this fundamental hydrocarbon. The synthesis commences with
1,5-cyclooctadiene and generates TVE in four steps in an
overall yield of 0.1%.3
In addition to the parent TVE, only four substituted
tetravinylethylenes have been reported thus far (Figure 1).
Dodecachloro-TVE 6 was the first to be reported in 1948,4
octaphenyl-TVE 7 in the early 1960s,5,6 tetracarboethoxy-TVE
8 in 1980,7 and octamethyl-TVE 9 in 1989.8 Only the McMurry
coupling approach to octamethyl-TVE 9 can be classed as a
practical synthesis.
INTRODUCTION
■
Herein we predict and chart the physical and chemical
properties of tetravinylethylene (TVE, 1, Scheme 1) and
analogues through a combined synthetic and computational
approach. TVE is the smallest symmetrical oligo-olefinic
structure comprising both through-conjugation (an unbranched
1,3,5-hexatriene unit) and cross-conjugation (a branched 3-
alkylidene-1,4-pentadiene).
TVE was first synthesized in 1966 by Skattebøl and co-
workers.1 The synthesis commenced with the double
dibromocarbene addition to cycloocta-1,5-diene (2) to form
bis-dibromocyclopropanation product 3. A double Doering−
LaFlamme allene synthesis produced the diallene 4 in 5%
yield.2 A [3,3]-sigmatropic rearrangement, brought about by
vacuum pyrolysis, gave 2,3-divinyl-1,3-cyclohexadiene (DVC,
5) in quantitative yield. Irradiation of 5 at 254 nm resulted in
the formation of 1, by way of 6π-electrocyclic ring opening, as a
mixture with precursor 5 and polymeric material. TVE (1) was
isolated in 7% yield by gas−liquid chromatography.
Our computational and experimental studies with the related
dendralenes9 served as the foundation to a direct general
Scheme 1. Skattebøl Synthesis of TVE (1)
Figure 1. Only substituted tetravinylethylenes reported in the
literature.
Received: September 16, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
synthesis of the TVE family. Thus, we recently disclosed the
first 4-fold cross-coupling reaction involving olefinic precursors
and applied this method to the one-step synthesis of
tetravinylethylene on a multigram scale (Scheme 2).10 Whereas
many of its close structural relatives are unstable when neat at
ambient temperature, tetravinylethylene is a remarkably robust,
bench-stable compound.
Scheme 4. Substituted TVEs Prepared by 4-Fold Cross-
Coupling
Scheme 2. Synthesis of TVE (1) by Four-Fold Cross-
Coupling
Herein we disclose the results of detailed experimental and
computational investigations with TVE and related structures.
RESULTS AND DISCUSSION
■
Our efforts to bring about selectiveinstead of exhaustive
cross-couplings have met with limited success. Poor selectivity
was encountered during attempts to achieve selective single and
2-fold couplings with tetrachloroethylene 10. Gratifyingly, the
3-fold coupling product 11 can, however, be accessed (Scheme
3). A mixture of 11 and TVE 1 is isolated as the crude product;
selective decomposition of 1 on exposure to trifluoroacetic acid
leaves chlorotetraene 11 in pure form.
Scheme 3. Three-Fold Cross-Coupling Involving 10
analysis of the significance of these X-ray crystal structures is
undermined by the likelihood of contributions from both
substituents and crystal packing forces.
We performed computational studies to identify the lowest
energy conformations of the parent, unsubstituted TVE (1).
Throughout this study, the accurate composite ab initio MO
G4(MP2) method was used. Thus, a series of CCSD(T), MP2,
and HF calculations were performed on B3LYP/6-31G(2df,p)
equilibrium geometries.14 All calculations were performed using
the Gaussian 09 program15 and refer to the gas phase, and
activation parameters were generally calculated at 298.15 K.
Ten conformers of TVE were located, the relative enthalpies of
which spanned 45 kJ/mol, although the four most stable ones
spanned only 8 kJ/mol. The geometries and relative energies of
these four conformers, which differ from the conformations of
the substituted TVEs seen in the X-ray crystal structures, are
presented in Figure 3.
The four vinyl substituents of TVE (1) experience steric
congestion, which results in two or more vinyl groups being
twisted out of coplanarity with the central double bond, thereby
attenuating conjugative stabilization. The two most stable
conformers of the parent TVE (1) are predicted to be ctcta-1
and ctct-1, both of which possess a central (E)-1,3,5-hexatriene
spine and differ in the disposition of the remaining pair of vinyl
substituents with respect to the plane of the hexatriene moiety,
being on the same side of this plane in the case of ctcta-1 and
on opposite sides in the case of ctct-1. In ctcta-1, which
possesses C2 symmetry, these two vinyl groups are gauche with
respect to the central double bond, each of which makes a
dihedral angle of 47° with the central double bond. The
magnitude of this angle is larger than the G4(MP2) value of
In contrast to the limited success in attempts to perform
nonexhaustive couplings, the 4-fold Stille coupling depicted in
Scheme 1 is one of useful scope. Thus, substituted
alkenylstannanes react with tetrachloroethylene to give
substituted TVEs 9, 12, 13, 14, and 15 (Scheme 4).
Substitution at the α-carbon leads to more sluggish reactions,
which we have thus far been unable to drive to completion
without increased temperature and significant loss of material.11
Tetrol 16 was formed by deprotection of the corresponding
silyl ether.12
Three of these substituted TVEs, namely 13, 14, and 16,
gave crystals suitable for single-crystal X-ray analysis (Figure 2).
None of the molecular structures from X-ray analysis showed
fully planar TVE units. All three crystal structures do contain an
(E,E)-1,3,5-hexatriene group that is principally in plane
(dihedral angles within ca. 20°), however, presumably for
conjugative stabilization reasons. Similar conformations of the
TVE unit are seen in the crystal structures of tetraphenyl TVE
13 (namely cttt-13) and tetratrimethylsilyl TVE 14 (cttt-14), in
which one of the four 1,3-butadiene groups is cisoid whereas the
other three are transoid.13 Conversely, the crystal structure of
tetrahydroxymethyl TVE 16, tttt-16, exhibits a conformation in
which all four 1,3-butadiene groups are transoid. Any deeper
B
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 3. G4(MP2) geometries and relative enthalpies and free
energies (kJ/mol) of the four most stable conformers of TVE (1),
namely ctcta-1 (bottom), ctct-1, cttt-1, and tttt-1 (top).
central ethylene unit in cttt-1 are −156°, −147°, −165°, and
44°. The least stable of these four TVE conformers is tttt-1.
This conformer is predicted to possess D2 symmetry in which
each vinyl group makes a 151° dihedral angle with the central
ethylene group. This structure may be obtained from the planar
Figure 2. Molecular structures from single-crystal X-ray analyses of
substituted TVEs 13 (cttt-13, top), 14 (cttt-14, center), and 16 (tttt-
16, bottom). In each case, the approximated stick representation is
depicted above the molecular structures, with a plan view of the
molecular structure on the left and a side view on the right.
D
2h structure by performing 39° conrotatory operations on the
two pairs of vicinal-vinyl groups.
In the laboratory, and consistent with Skattebøl’s observa-
tions, TVE (1) undergoes 6π-electrocyclization to generate 3,4-
divinyl-1,3-cyclohexadiene (DVC, 5) under thermal and
photochemical conditions (Scheme 5). In relatively high
dilution solutions (10 mM), the thermal (120 °C) reaction
proceeds in quantitative yield, as judged by 1H NMR
spectroscopy. At higher concentrations, DVC (5) oligomerizes,
rendering the heat-promoted electrocyclization impractical. We
therefore advocate photochemical promotion16 at ambient
temperature, which allows the preparation of solutions of DVC
5 of up to 0.2 M concentration.
The DVC structure 5 can be generated by linking the central
olefins of [4]dendralene 1817 with a −CH2CH2− tether. DVC
is, therefore, a substituted [4]dendralene. The parent [4]-
dendralene (18) is the most stable member of the
unsubstituted [n]dendralene family, showing no sign of
decomposition when stored neat at room temperature.9,17
Despite its structural similarity, DVC (5) exhibits markedly
different behavior, decomposing rapidly at ambient temperature
when neat. We find that DVC (5) is best handled as a solution,
and we prefer to generate the hydrocarbon in situ and use it
directly.
31° calculated for gauche 1,3-butadiene and is the consequence
of unfavorable steric interactions with the terminal vinyl groups
of the hexatriene unit. Indeed, such interactions account for the
finding that the hexatriene group is not planar, with the
terminal vinyl groups each making a dihedral angle of −165°
with the central double bond. In contrast to ctcta-1, the
hexatriene unit in ctct-1 is nearly planar, with the two terminal
vinyl groups of the unit making dihedral angles of 178° and
−178° with the central ethylene group, and this is due to the
fact that the two remaining vinyl groups, which point in
opposite directions from the plane of the hexatriene group,
exert equal but opposite forces on the terminal vinyl groups of
the hexatriene unit. Thus, in this case, adverse steric forces can
only be ameliorated by increasing the dihedral angles between
the 3- and 4-vinyl groups and the central double bond, relative
to those predicted (47°) for ctcta-1. In fact, these dihedral
angles for ctct-1 are calculated to be 65° and −65°. The
dihedral angle made by the four vinyl groups with respect to the
C
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
activation enthalpy, ΔH⧧, and activation free energy, ΔG⧧, for
the reaction are 112.9 and 116.4 kJ/mol, respectively. Using
G4(MP2), we calculated ΔH⧧ and ΔG⧧ for the disrotatory
electrocyclization of the parent (Z)-1,3,5-hexatriene to be 123.4
and 130.4 kJ/mol, respectively. Hence, the two spectator vinyl
groups stabilize the TS for the electrocyclization of TVE by
about 10 kJ/mol, compared to the parent hexatriene.
Scheme 5. Electrocyclization of TVE (1) into DVC (5) and
Stability Comparisons with [3]Dendralene and
[4]Dendralene
[3]Dendralene, the simplest cross-conjugated triene, is also
the least stable of the parent dendralenes, undergoing Diels−
Alder (DA) dimerization cleanly at ambient temperature with a
half-life of around 10 h at 25 °C (Scheme 5).18 It seems
reasonable to expect that TVE (1), which can be thought of as
two [3]dendralenes conjoined at the central alkene, would be
similarly unstable. This is not the case: we routinely store TVE
(1) neat at room temperature and witness no decomposition
over several weeks. Evidently, the additional two vinyl groups
serve as a strong steric impediment toward adoption of the TS
for DA dimerization. Without access to this mode of
decomposition, TVE is perfectly stable. [3]Dendralenes
substituted at the central methylene carbon exhibit a similar
enhanced stability toward DA dimerization.19 Nevertheless,
these substituted [3]dendralenes are active participants in DA
reactions with electron-poor dienophiles, and TVE is no
exception. As predicted computationally (vide infra), TVE
reacts directly in a DA cycloaddition with N-methylmaleimide
(NMM, 19, Scheme 7). A reaction temperature of less than 60
°C was maintained to minimize20 the heat-promoted TVE 6π-
electrocyclization. The single DA adduct 20 was not isolated. In
fact, no signal could be attributed to this compound upon direct
Two transition structures (TSs) for the thermal, disrotatory
6π-electrocyclic ring closure of TVE (1) were located, and they
differ only in the conformations adopted by the pair of
spectator vinyl groups (Scheme 6). Although, at 298.15 K, both
TSs ct-5TS and tt-5TS are isoenthalpic, the free energy of the
latter is 2.6 kJ/mol lower than that of the former TS. The
Scheme 6. G4(MP2) Geometries, Relative Enthalpies, Free
Energies (kJ/mol), and Entropies [J/(mol·K)] of Reactant
ctcta-1 (Top), Most Stable TSs tt-5TS and ct-5TS (Middle),
and Product ct-5 (Bottom) for the Disrotatory 6π-
Electrocyclic Reaction of TVE (1)
1
analysis of H NMR spectra of the reaction mixture. Instead,
tetracycle 21, the product of a diene-transmissive21 double DA
sequence, was formed. The stereochemistry of this compound
was consistent with the second NMM dienophile docking to
the convex face of single adduct 20 through the endo-
cycloaddition mode. Evidently, the barrier toward the first
DA reaction is significantly higher than that of the second. This
result again highlights the difference between TVE (1) and
[3]dendralene (17), the latter undergoing a much faster first
cycloaddition with dienophiles such as NMM.22
Scheme 7. Diels−Alder Reaction between TVE (1) and
NMM (19)
D
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Ten TSs were located for the DA reaction between TVE and
NMM (19). Of these, endo-mode TS N-ttt-20TS has the lowest
energy (Scheme 8). As expected from Alder’s endo rule,23 this
Scheme 9. G4(MP2) Geometries, Relative Enthalpies, Free
Energies (kJ/mol), and Entropies [J/(mol·K)] of Reactants
[3]Dendralene ct-17 and NMM (19, Top) and Lowest
Energy TS for the Diels−Alder Reaction between Them N-t-
22TS (Bottom)
Scheme 8. G4(MP2) Geometries, Relative Enthalpies, Free
Energies (kJ/mol), and Entropies [J/(mol·K)] of Reactants
TVE ctcta-1 and NMM (19, Top), Most Stable TSs for the
DA Reaction between Them N-ttt-20TS and X-ttt-20TS
(Middle), and the Product DA Adduct 20 (Bottom)
similar situation might be expected for the preferred TS for the
TVE-NMM reaction. In this case, however, a steric factor
appears to override this electronic effect. Specifically, a planar
TVE diene component would suffer significant destabilizing
steric interactions between C1 of the reacting 1,3-butadiene
and its Z-4-vinyl substituent. In N-ttt-20TS, the Z-4-vinyl group
is rotated away from C1 by ca. 37° (= value of the dihedral
angle C2−C3−C4−C5) and also from NMM, the incoming
dienophile. This twisting results in a longer developing bond to
C4 in the TS. In contrast, bond formation at C1 is sterically
unencumbered, thus resulting in a very short (2 Å) forming
bond length. Significant closed-shell biradicaloid character
should exist in such a highly asynchronous TS,24,25 which
should be further stabilized in N-ttt-20TS (and also in X-ttt-
20TS) by the presence of a trivinylmethyl group in the TVE
component. Steric factors also explain why the significantly
larger activation enthalpy calculated for the TVE-NMM TS, N-
ttt-20TS (49.5 kJ/mol), compared to that of the corresponding
[3]dendralene-NMM TS, N-t-22TS (33.5 kJ/mol), is also
explained by these steric effects.
As mentioned previously, two competing reactions take place
upon treating TVE (1) with NMM (19), namely direct DA
reaction (Schemes 7 and 8) and electrocyclization (Schemes 5
and 6). From the activation parameters given in Schemes 6 and
8, it is seen that, although the enthalpy of activation for the
electrocyclic reaction is nearly 50 kJ/mol greater than that for
the DA reaction, the free energy of activation for the former
process is 6.4 kJ/mol lower than that for the latter. This arises
because the activation entropy for electrocyclization is
substantially larger (by a factor of 18) than that for the DA
reaction. These results suggest that there exists an isokinetic
temperature of about 63 °C, below which the rate constant for
the DA reaction is greater than that for electrocyclization and
above which, the reverse is true. Of course, the rate of the DA
reaction may also be modulated by varying the concentration of
NMM. It should be noted that the calculated isokinetic
temperature is based on gas phase calculations and is, therefore,
subject to solvent effects. In summary, the yield of DA adduct is
maximized by carrying out the reaction at low temperatures
endo TS is strongly favored enthalpically over the exo TS, X-ttt-
20TS, by 13.6 kJ/mol. Both endo and exo TSs display marked
bond-forming asynchronicities. The forming bonds in N-ttt-
20TS are 2.003 and 2.663 Å (Δr = 0.66 Å), and those for X-ttt-
20TS are 1.994 and 2.723 Å (Δr = 0.73 Å). The shorter
developing bond in the TS is with the C1 carbon of TVE (1),
as depicted in Scheme 6.
It is informative to compare the activation enthalpies and
TSs geometries for the TVE−NMM DA reaction with those
involving [3]dendralene (17) as diene (Scheme 9) on the
grounds that TVE can be regarded as a disubstituted
[3]dendralene (substituent = vinyl). The most stable TS for
the [3]dendralene−NMM reaction, N-t-22TS (also an endo-
mode docking of diene and dienophile), exhibits a significantly
lower bond-forming asynchronicity (2.403 Å, 2.154 Å; Δr =
0.25 Å) than that seen for the TVE-NMM reaction, N-ttt-20TS.
These two TSs also exhibit asynchronicity of the opposite
orientation: in N-t-22TS, the shorter forming bond involves
C4, whereas with N-ttt-20TS, the shorter forming bond
involves C1. The short forming bond involving C4 in N-t-
22TS leads to stabilization by delocalization of charge through
the pentadienyl system derived from [3]dendralene.24,25
A
E
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(<50 °C) and using an excess of NMM (19). These predictions
were borne out experimentally (Scheme 7).
Without further purification, the solution from the 6π-
electrocyclization of TVE (1, Scheme 5) was used to study the
DA reactivity of DVC 5. Using 1.0 molar equiv of NMM (19)
led cleanly to the formation of one single DA adduct 23
(Scheme 10). The regioisomer of 23, namely 26, was not
adduct directly. In effect, the sequence commences with TVE
(1), which undergoes the 6π-electrocyclization and three
successive DA reactions in one pot to generate one
diastereomeric product carrying seven new C−C bonds and
four new rings.26
One might be surprised by the regiochemical outcome of the
first cycloaddition 5 → 23 (Scheme 10), with a conformation-
ally flexible semicyclic diene reacting in preference to a 1,3-
cyclohexadiene, which is locked in the s-cis conformation. A
selective single cycloaddition to DVC (5) is also unexpected,
since the parent [4]dendralene (18) undergoes a rather
unselective reaction on treatment with 1 molar equiv of
NMM (19), with double adducts being the major products.17,18
To shed light on these experimental observations, the
reaction between DVC (5) and NMM (19) was investigated
computationally. Endo- and exo-TSs were located for addition
to both the cyclic diene site (internal addition), N-26TS and X-
26TS, and to the semicyclic diene site (terminal addition), N-
23TS and X-23TS (Scheme 11). Bond-forming asynchronic-
ities in all four TSs are small, ca. 0.1 Å, with the shorter bond
forming at the less substituted and, hence, less sterically
congested C1 atom in the terminal addition mode.
Scheme 10. Diels−Alder Sequences Involving DVC (5) and
NMM (19)
Consistent with experimental findings, the terminal addition
endo-TS, N-23TS, was calculated as the most stable. Both
activation enthalpy and activation free energy for this addition
mode (ΔH⧧ = 20.6 and ΔG⧧ = 80.3 kJ/mol) are extraordinarily
small; calculated barriers for the DA addition of NMM to
[3]dendralene (17) are significantly higher (ΔH⧧ = 33.5 and
ΔG⧧ = 92.8 kJ/mol; Scheme 9). The origin of the enhanced
DA reactivity of ct-5 appears not to be a conformational effect
in the reactants because the C1−C2−C3−C3′ dihedral angles
within the reactive diene component in ct-5 (Scheme 11) and
ct-17 (Scheme 9) are nearly identical, 39° and 41°, respectively.
Neither is the enhanced reactivity due to the presence of an
additional double bond in ct-5 because it is connected in a
cross-conjugative manner (to give a [4]dendralene analogue),
and so its electronic influence should be marginal. This
reasoning is supported by G4(MP2) calculations on the DA
reaction between the parent [4]dendralene (18) and NMM
(19, Scheme 12).
TSs for endo addition of NMM (19) to both terminal and
internal dienes of the parent [4]dendralene (18) were
calculated. Relative to the most stable conformer of [4]-
dendralene, tct-18, the activation energies (ΔH⧧ = 44.2 and
ΔG⧧ = 104.3 kJ/mol for terminal addition) are much larger
than those for [3]dendralene (17). The preferred conformation
of [4]dendralene, tct-18, comprises two essentially in plane s-
trans 1,3-butadiene units skewed at an angle of 78° toward one
other. In contrast, a gauche conformation of the reactive diene
units exist for both [3]dendralene (17) and DVC ct-5. The
most stable conformation of [4]dendralene, tct-18, is, therefore,
not appropriate as a starting point for activation barrier
comparisons. When the more satisfactory [4]dendralene
conformation cct-18 is employed, activation parameters (ΔH⧧
= 33.1 and ΔG⧧ = 95.0 kJ/mol) are consistent with those
obtained for [3]dendralene ct-17: ΔH⧧ = 33.5 and ΔG⧧ = 92.8
kJ/mol (Scheme 8). These findings indicate that the enhanced
DA reactivity (compared to [3]dendralene) of DVC ct-5 is not
caused by its spectator vinyl group. What is the origin of the
high DA diene reactivity of DVC? Through a combination of
hyperconjugative and +I effects, the −CH2CH2− group in ct-5
elevates DVC’s HOMO. According to B3LYP/6-31G(d)
calculations, the HOMO energy of ct-5 is 0.7 eV higher than
detected. This first cycloaddition, therefore, exhibits complete
site selectivity for one of the two equivalent terminal
(semicyclic) dienes over the internal 1,3-cyclohexadiene. This
reaction also exhibits very high chemoselectivity in that only
traces of the double NMM adduct can be detected in the
reaction mixture. It is also highly stereoselective in that only the
endo-adduct is detected.
Only the semicyclic diene site of single DA adduct 23 can
adopt the s-cis conformation; hence, the site selectivity of the
second addition is assured. In the event, exposure of single DA
adduct 23 to 1 molar equiv of NMM (19) gave one
1
diastereomer (by H NMR spectroscopic analysis) of double
DA adduct 24. A single-crystal X-ray analysis of 24 confirmed
endo selectivity for the first two cycloadditions and the anti-
approach of NMM (19) in the second addition. Exposure of
DVC (5) to 2.0 molar equiv of NMM (19) gave double DA
adduct 24 directly.
Treatment of double DA adduct 24 with NMM (19)
generated a single diastereoisomer of triple DA adduct 25. This
third DA reaction also proceeded through an endo transition
state, as evidenced by single-crystal X-ray analysis of the
product. The C2-symmetrical nature of the anti-double adduct
precludes any issue of π-diastereofacial selectivity in this third
addition. Exposure of DVC (5) to NMM (19) gave the triple
F
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 11. G4(MP2) Geometries, Relative Enthalpies, Free
Energies (kJ/mol), and Entropies [J/(mol·K)] of Reactants
DVC ct-5 and NMM (19, Center) and Most Stable TSs for
the Diels−Alder Reaction between Them: to the Cyclic
Diene Site (Top) N-ct-26TS and X-ct-26TS and to the
Semicyclic Diene Site (Bottom) N-c-23TS and X-c-23TS
Scheme 12. G4(MP2) Geometries, Relative Enthalpies, Free
Energies (kJ/mol), and Entropies [J/(mol·K)] of Reactants
[4]Dendralene tct-18/cct-18 and NMM (19, Center) and the
Most Stable TSs for the Diels−Alder Reaction between
Them: N-27TS (Internal Addition) and N-28TS (Terminal
Addition)
That a controlled, 3-fold cycloaddition sequence involving
DVC is not limited to maleimide dienophiles is demonstrated
by the experimental results depicted in Scheme 13. The
dienophile in this case is dimethylacetylene dicarboxylate
(DMAD, 29). Once again, the first dienophile addition occurs
under milder conditions than do the subsequent ones, and the
[3]dendralenic monoadduct 30 is readily isolated in pure form.
The second addition in this case, unsurprisingly, proceeds with
lower π-diastereofacial selectivity than with NMM (Scheme
10). Interestingly, whereas anti-double DA adduct 31 is isolated
cleanly, syn-double DA adduct 32 (the putative major
stereoisomer) goes on to form triple cycloadduct 33 in high
selectivity.
those of ct-17 and cct-18. This HOMO energy increase leads to
a decrease in the HOMODVC−LUMONMM energy gap, thus
leading to enhanced reactivity.
In the DVC−NMM cycloaddition (Scheme 11), endo
selectivity is predicted for both terminal and internal modes
of addition, but the enthalpic preference for the endo mode is
markedly greater, by 11 kJ/mol, at the internal site than the
terminal site. This predicted stronger endo selectivity in the
internal addition reaction is probably due to destabilization of
the exo-internal TS, X-26TS, through steric interactions
between the NMM moiety and the −CH2CH2− group of the
cyclohexadiene ring. Terminal addition is predicted to be
preferred over internal addition by ca. 5 kJ/mol, presumably
due to destabilizing steric interactions in the endo, internal TS,
N-26TS.
SUMMARY AND CONCLUSIONS
■
Parenthetically, we note that the lack of site (i.e., terminal vs
internal) selectivity in the DA reaction between NMM and
[4]dendralene (Scheme 12) predicted by our calculations is
consistent with experimental findings.17
In summary, the first olefin-based 4-fold cross-coupling
reactions has led to a one-step synthesis of tetravinylethylenes.
This study demonstrates that tetravinylethylenes are not only
easily prepared employing standard laboratory equipment and
G
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 13. Diels−Alder Sequences Involving DVC (5) and
DMAD (29)
Scheme 14. Key Findings of this Study
a
For compounds 31 and 33 the lower yield was obtained through a
sequence of two separate reactions, and the higher yield was obtained
from a one-pot process from DVC (5).
methods but also require no special precautions when being
manipulated. In terms of reactivity, stability, and ease of
handling, TVEs are distinct from both unsubstituted [3]-
dendralenes and 1,3,5-hexatrienes. Importantly, the behavior of
both TVE and related structures are predictable and explainable
by employing the composite ab initio MO G4(MP2) method
(Scheme 14).
EXPERIMENTAL SECTION
General Methods. See the Supporting Information.
■
Tetravinylethylene (TVE, 1). A mixture of Pd(OAc)2 (133 mg,
0.567 mmol, 0.040 molar equiv) and XPhos (539 mg, 1.13 mmol,
0.080 molar equiv) in a two-neck 100 mL round-bottom flask
equipped with reflux condenser was purged three times under reduced
pressure and refilled with argon. Tetrachloroethylene (10) (2.90 mL,
28.4 mmol, 1.0 molar equiv) and vinyltributyltin (41.5 mL, 142 mmol,
5.0 molar equiv) were added. The reaction mixture was stirred and
heated to 60 °C overnight. A bulb-to-trap distillation apparatus was
attached to the reaction flask, and the reaction mixture was heated to
50 °C for 3 h under reduced pressure (0.47 mbar). A dry ice/acetone
cooling bath was used to trap tetravinylethylene (1) in the distillation
1
flask as a colorless oil (2.4 g, 64%): H NMR (400 MHz, CDCl3) δ
6.62 (dd, J = 17.3 Hz, 11.0 Hz, 4H), 5.56−5.08 (m, 8H) ppm; 13C
NMR (100 MHz, CDCl3) δ 136.7 (Cq), 134.7 (CH), 119.5 (CH2)
ppm; IR (thin film) νmax 3088, 2975, 2926, 2853 cm−1; MS (70 eV, EI)
m/z 264.2 ([2M]+•, 55), 131.1 (63) 117.1 (100);* HRMS (EI) calcd
for C10H12 [M]+• 132.0939, found 132.0938; calcd for C20H24 [2M]+•
264.1878, found 264.1877;* UV−vis (n-hexane) λmax = 225 nm (ε =
10700), 283.5 (ε = 14800).
Hz, 10.9 Hz, 1H), 6.02 (s, 1H), 5.32 (dd, J = 17.4 Hz, 1.7 Hz, 1H),
5.00 (dd, J = 10.9 Hz, 1.7 Hz, 1H), 2.13 (s, 2H) ppm; 13C NMR (100
MHz, CDCl3) δ 136.6 (CH), 136.1 (Cq), 125.6 (CH), 113.9 (CH2),
22.6 (CH2) ppm; IR (thin film) νmax 3098, 3020, 2928, 2872, 2854
cm−1; MS (70 eV, EI) m/z 132.2 ([M]+•, 57), 117.1 (97), 91.1 (100);
MS (70 eV, EI) m/z 264.2 ([2M]+•, 63), 249.3 ([2M − CH3], 11);
HRMS (EI) calcd for C10H12 [M]+• 132.0939, found 132.0938; calcd
for C20H24 [2M]+• 264.1878, found 264.1873; UV−vis (n-hexane) λmax
= 205 nm (ε = 14200), 229 (ε = 14500).
*We assume that the TVE undergoes 6π-electrocyclization and the
resulting DVC undergoes DA dimerization.
2,3-Divinyl-1,3-cyclohexadiene (DVC, 5) Using Microwave
Heating. A solution of TVE (1) (13 mg, 0.10 mmol, 1.0 molar equiv)
in CDCl3 (10 mL) was heated at 120 °C for 2 h and gave 2,3-divinyl-
1
1,3-cyclohexadiene (5) as a colorless solution in CDCl3. H NMR
analysis indicated complete conversion of TVE (1) to 2,3-divinyl-1,3-
cyclohexadiene (5): 1H NMR (400 MHz, CDCl3) δ 6.36 (dd, J = 17.4
H
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
2,3-Divinyl-1,3-cyclohexadiene (DVC, 5) Using Photochem-
ical Irradiation. A solution of TVE (1) (53 mg, 0.40 mmol, 1.0 molar
equiv) in benzene-d6 (2.0 mL) was degassed by bubbling N2 for 1 min
while stirring vigorously in a reaction tube (both quartz glass and
Pyrex gave the same outcome). Following the degassing procedure, the
reaction mixture was irradiated using a medium-pressure mercury lamp
at room temperature for 2 h under N2 while stirring. The yield can be
estimated by adding durene (38 mg, 0.28 mmol, 0.70 molar equiv) as
the internal standard and recording a 1H NMR spectrum, which
reveals 100% conversion of TVE (1) and a 70% yield of 2,3-divinyl-
1,3-cyclohexadiene (5). The solution was used directly in further
reactions.
40−60 °C to give (2E,6E)-4,5-bis((E)-3-methoxy-3-oxoprop-1-en-1-
yl)octa-2,4,6-trienedioate (12) as a colorless solid (110 mg, 60%), Rf =
0.20 (1:1 EtOAc/petroleum ether 40−60 °C). An analytical sample of
dimethyl (2E,6E)-4,5-bis((E)-3-methoxy-3-oxoprop-1-en-1-yl)octa-
2,4,6-trienedioate (12) was obtained by recrystallization from
1
CH2Cl2 to give colorless needles: mp 108−109 °C; H NMR (400
MHz, CDCl3) δ 7.57 (d, J = 15.7 Hz, 4H), 6.09 (d, J = 15.8 Hz, 4H),
3.80 (s, 12H) ppm; 13C NMR (100 MHz, CDCl3) δ 166.2 (Cq), 139.7
(CH), 138.5 (Cq), 127.0 (CH), 52.2 (CH3) ppm; IR (KBr tablet) νmax
= 2955, 2919, 2850, 1724, 1618 cm−1; MS (70 eV, EI) m/z 364.2
([M]+•, 6), 333.1 ([M − OCH3], 37), 273.1 ([M − (OCH3)3], 100);
HRMS (EI) calcd for C18H20O8 [M]+•: 364.1158; found 364.1158;
Anal. Calcd (%) C18H20O8: C 59.34, H 5.53; found C 59.17, H 5.88;
UV−vis (acetonitrile) λmax = 199 nm (ε = 23200), 256 nm ε = 23200),
329.5 (ε = 32200).
Chlorotrivinylethylene (11). A mixture of Pd(OAc)2 (86 mg,
0.39 mmol, 0.030 molar equiv) and XPhos (367 mg, 0.77 mmol, 0.060
molar equiv) in a two-neck 50 mL round-bottom flask equipped with
reflux condenser was purged 3× under reduced pressure and refilled
with argon. Tetrachloroethylene (10) (1.30 mL, 12.8 mmol, 1.0 molar
equiv) and vinyltributyltin (15.0 mL, 51.3 mmol, 4.0 molar equiv)
were added. The reaction mixture was stirred and heated to 60 °C for
20 h and cooled to 0 °C. Trifluoroacetic acid (1.96 mL, 25.6 mmol, 2.0
molar equiv) was added dropwise while maintaining the temperature
of the reaction mixture below 20 °C. The reaction mixture was stirred
for 2 h upon completion of addition at room temperature and cooled
back to 0 °C. DBU (3.82 mL, 25.6 mmol, 2.0 molar equiv) was added
dropwise while stirring the reaction mixture vigorously and
maintaining the temperature below 30 °C. A bulb-to-trap distillation
apparatus was attached to the reaction flask, and the reaction mixture
was heated to 60 °C for 2 h under reduced pressure (1.3 mbar). A dry
ice/acetone cooling bath was used to trap chlorotrivinylethylene 11 in
((1E,5E)-3,4-Di((E)-styryl)hexa-1,3,5-triene-1,6-diyl)-
dibenzene (13). A mixture of Pd(OAc)2 (4.5 mg, 0.020 mmol, 0.040
molar equiv) and XPhos (19 mg, 0.040 mmol, 0.080 molar equiv) in a
2-neck 5 mL round-bottom flask equipped with reflux condenser was
purged 3 times under reduced pressure and refilled with dry argon.
Tetrachloroethylene (10) (50 μL, 0.50 mmol, 1.0 molar equiv) and
(E)-tributyl(styryl)stannane29 (983 mg, 2.5 mmol, 5.0 molar equiv)
were added. The reaction mixture was stirred and heated to 60 °C
overnight. The product precipitated out of the solution during the
reaction and was filtered off upon cooling to room temperature. The
precipitate was washed subsequently with EtOH (15 mL), EtOAc (15
mL) and Et2O (15 mL) to give ((1E,5E)-3,4-di((E)-styryl)hexa-1,3,5-
triene-1,6-diyl)dibenzene (13) as a yellow solid (150 mg, 69%). An
analytical sample of ((1E,5E)-3,4-di((E)-styryl)hexa-1,3,5-triene-1,6-
diyl)dibenzene (13) was obtained by recrystallization from CH2Cl2 to
give colorless needles: mp 158−159 °C; 1H NMR (400 MHz,
CD2Cl2) δ 7.50 (d, J = 7.5 Hz, 8H), 7.42−7.20 (m, 16H), 6.82 (d, J =
16.2 Hz, 4H) ppm; 13C NMR (100 MHz, CD2Cl2) δ 138.0 (Cq),
137.4 (Cq), 134.7 (CH), 129.1 (CH), 128.1 (CH), 128.0 (CH), 127.0
(CH) ppm; IR (KBr tablet) νmax = 3078, 3055, 3026, 2960, 2924, 2851
cm−1; MS (70 eV, EI) m/z 436.2 ([M]+•, 8), 418.1 (61), 386.1 (100);
HRMS (EI) calcd for C34H28 [M]+• 436.2191, found 436.2189; UV−
vis (acetonitrile) λmax = 199.4 nm (ε = 79000), 273.6 nm (ε = 23400).
((1E,5E)-3,4-Bis((E)-2-(trimethylsilyl)vinyl)hexa-1,3,5-triene-
1,6-diyl)bis(trimethylsilane) (14). A mixture of Pd(OAc)2 (29 mg,
0.13 mmol, 0.040 molar equiv) and XPhos (122 mg, 0.256 mmol,
0.080 molar equiv) in a two-neck 10 mL round-bottom flask equipped
with reflux condenser was purged three times under reduced pressure
and refilled with dry argon. Tetrachloroethylene (10) (0.317 mL, 3.2
mmol, 1.0 molar equiv) and (E)-trimethyl(2-(tributylstannyl)vinyl)-
silane30 (6.0 g, 15.5 mmol, 5.0 molar equiv) were added. The reaction
mixture was stirred and heated to 60 °C overnight. Upon cooling to
room temperature, the crude mixture was subjected to flash column
chromatography using silica gel eluting with n-pentane to give
((1E,5E)-3,4-bis((E)-2-(trimethylsilyl)vinyl)hexa-1,3,5-triene-1,6-
diyl)bis(trimethylsilane) (14) as a colorless solid (1.3 g, 89%), Rf =
0.70 (100% n-pentane). An analytical sample of ((1E,5E)-3,4-bis((E)-
2-(trimethylsilyl)vinyl)hexa-1,3,5-triene-1,6-diyl)bis(trimethylsilane)
(14) was obtained by recrystallization from petroleum ether 40−60 °C
to give colorless needles: mp 135 °C; 1H NMR (400 MHz, CDCl3) δ
6.77 (d, J = 19.0 Hz, 4H), 5.95 (d, J = 19.0 Hz, 4H), 0.12 (s, 36H)
ppm; 13C NMR (100 MHz, CDCl3) δ 142.5 (CH), 139.8 (Cq), 136.4
(CH), −1.0 (CH3) ppm; IR (KBr tablet) νmax = 2955, 2898, 1576
cm−1; MS (70 eV, EI) m/z 420.2 ([M]+•, 20), 347.2 ([M − TMS],
27); HRMS (EI) calcd for C22H44Si4 [M]+• 420.2520, found 420.2520;
UV−vis (acetonitrile) λmax = 195 nm (ε = 15300), 246 nm (ε =
22000), 316 (ε = 25100).
1
the distillation flask as a colorless oil (668 mg, 37%): H NMR (400
MHz, CDCl3) δ 6.98 (ddd, J = 27.1 Hz, 16.9 Hz, 10.7 Hz, 2H), 6.39
(dd, J = 17.5 Hz, 11.1 Hz, 1H), 5.77 (d, J = 16.4 Hz, 1H), 5.62−5.21
(m, 5H) ppm; 13C NMR (100 MHz, CDCl3) δ 136.1(Cq), 134.2
(CH), 132.0 (CH), 131.8 (CH), 131.2 (Cq), 122.7 (CH2), 120.0
(CH2), 118.4 (CH2) ppm; IR (thin film) νmax 3092, 3058, 3021, 2978,
1833, 1784 cm−1; MS (70 eV, EI) m/z 140.0 ([M]+•, 53) 105.1 (M −
Cl, 100); HRMS (EI) calcd for C8H9Cl35 [M]+• 140.0393, found
140.0392; calcd for C8H9Cl37 [M]+• 142.0363, found 142.0360; UV−
vis (n-hexane) λmax = 225.5 nm (ε = 8110), 282 (ε = 25300).
2,7-Dimethyl-4,5-bis(2-methylprop-1-en-1-yl)octa-2,4,6-tri-
ene (9). A mixture of Pd(OAc)2 (11 mg, 0.050 mmol, 0.10 molar
equiv) and XPhos (48 mg, 0.10 mmol, 0.20 molar equiv) in a two-neck
5 mL round-bottom flask equipped with reflux condenser was purged
three times under reduced pressure and refilled with dry argon.
Tetrachloroethylene (10) (50 μL, 0.50 mmol, 1.0 molar equiv) and
tributyl(2-methylprop-1-en-1-yl)stannane27 (863 mg, 2.50 mmol, 5.0
molar equiv) were added. The reaction mixture was stirred and heated
to 80 °C for 3 days. Upon cooling to room temperature, the crude
mixture was subjected to flash column chromatography using silica gel
eluting with petroleum ether 40−60 °C to give 2,7-dimethyl-4,5-bis(2-
methylprop-1-en-1-yl)octa-2,4,6-triene (9) as a colorless oil (60 mg,
1
49%): Rf = 0.90 (100% petroleum ether 40−60 °C); H NMR (400
MHz, CDCl3) δ 5.74 (s, 4H), 1.78 (s, 12H), 1.57 (s, 12H) ppm; 13C
NMR (100 MHz, CDCl3) δ 134.3 (Cq), 134.1 (Cq), 126.1 (CH), 26.5
(CH3), 19.9 (CH3) ppm; IR (thin film) νmax 2965, 2923, 2909, 2854
cm−1; MS (EI) m/z 244.2 ([M]+•, 39), 229.2 ([M − CH3]), 34),
214.1 ([M − (CH3)2], 3), 159.1 (100); HRMS (EI) calcd for C18H28
[M]+• 244.2191, found 244.2191; UV−vis (n-hexane) λmax = 225 nm
(ε = 10900), 283.5 nm (ε = 15800).
Dimethyl (2E,6E)-4,5-Bis((E)-3-methoxy-3-oxoprop-1-en-1-
yl)octa-2,4,6-trienedioate (12). A mixture of Pd(OAc)2 (11 mg,
0.050 mmol, 0.10 molar equiv) and XPhos (48 mg, 0.10 mmol, 0.20
molar equiv) in a two-neck 5 mL round-bottom flask equipped with
reflux condenser was purged three times under reduced pressure and
refilled with dry argon. Tetrachloroethylene (10) (50 μL, 0.50 mmol,
1.0 molar equiv) and methyl (E)-3-(tributylstannyl)acrylate28 (938
mg, 2.5 mmol, 5.0 molar equiv) were added. The reaction mixture was
stirred and heated to 40 °C overnight. Upon cooling to room
temperature, the crude mixture was subjected to flash column
chromatography using silica gel eluting with 50% EtOAc in petrol
(6E,10E)-8,9-Bis((E)-3-((tert-butyldimethylsilyl)oxy)prop-1-
en-1-yl)-2,2,3,3,14,14,15,15-octamethyl-4,13-dioxa-3,14-disila-
hexadeca-6,8,10-triene (15). A mixture of Pd(OAc)2 (8.7 mg,
0.043 mmol, 0.10 molar equiv) and XPhos (41 mg, 0.086 mmol, 0.20
molar equiv) in a two-neck 5 mL round-bottom flask equipped with
reflux condenser was purged 3 times under reduced pressure and
refilled with dry argon. Tetrachloroethylene (10) (43 μL, 0.43 mmol,
1.0 molar equiv) and (E)-tert-butyldimethyl((3-(tributylstannyl)allyl)-
I
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
oxy)silane31 (1.19 g, 2.58 mmol, 6.0 molar equiv) were added. The
reaction mixture was stirred and heated to 60 °C overnight. Upon
cooling to room temperature, the crude mixture was subjected to flash
column chromatography using basic alumina eluting with 10% EtOAc
in petroleum ether 40−60 °C to give (6E,10E)-8,9-bis((E)-3-((tert-
butyldimethylsilyl)oxy)prop-1-en-1-yl)-2,2,3,3,14,14,15,15-octamethyl-
4,13-dioxa-3,14-disilahexadeca-6,8,10-triene (15) as a yellow oil (275
(64 mg, 70%): Rf = 0.30 (30:70 EtOAc/petroleum ether 40−60 °C);
1H NMR (400 MHz, CDCl3) δ 6.24 (dd, J = 17.3 Hz, 10.9 Hz, 1H),
5.99 (s, 1H), 5.82 (s, 1H), 5.26 (d, J = 17.5 Hz, 1H), 5.01 (d, J = 10.9
Hz, 1H), 3.19−3.06 (m, 2H) 2.89 (s, 3H), 2.82 (dd, J = 15.1 Hz, 1H),
2.62−2.48 (m, 1H), 2.35−2.13 (m, 3H), 2.11−1.88 (m, 2H) ppm; 13C
NMR (100 MHz, CDCl3) δ 180.0 (Cq), 178.1 (Cq), 137.5 (Cq), 135.9
(Cq), 135.1 (CH), 127.8 (CH), 119.3 (CH), 115.0 (CH), 43.6 (CH),
40.9 (CH), 35.5 (CH), 24.9 (CH3), 24.6 (CH2), 24.5 (CH2), 23.7
(CH2) ppm; IR (thin film) νmax 3031, 2943, 2879, 2846, 1766, 1694
cm−1; MS (70 eV, EI) m/z 243.1 ([M]+•, 35), 228.1 ([M − CH3], 2)
131.1 (100); HRMS (EI) calcd for C15H17NO2 [M]+• 243.1259, found
243.1260.
Reaction of 2,3-Divinyl-1,3-cyclohexadiene (5) with 2.0
Molar Equiv of N-Methylmaleimide (19). A mixture of 2,3-
divinyl-1,3-divinylcyclohexadiene (5) (16 mg, 0.12 mmol, 1.0 molar
equiv) and N-methylmaleimide (19) (26 mg, 0.24 mmol, 2.0 molar
equiv) in benzene-d6 (1.4 mL) was stirred at room temperature
overnight. Solvent was removed under reduced pressure, and the
residue was subjected to flash column chromatography using silica gel
eluting with 50% EtOAc in petroleum ether 40−60 °C to give 24 as a
colorless solid (32 mg, 76%), Rf = 0.25 (50:50 EtOAc/petroleum ether
40−60 °C). An analytical sample of 24 was obtained by
recrystallization from EtOH to give colorless needles: mp 188−192
°C; 1H NMR (400 MHz, CDCl3) δ 5.96 (s, 2H), 3.14−3.00 (m, 4H),
2.85 (s, 6H), 2.72 (dd, J = 16.6 Hz, 7.5 Hz, 2H), 2.38 (s, 2H), 2.19−
1.94 (m, 6H) ppm; 13C NMR (100 MHz, CDCl3) δ 179.9 (Cq), 178.1
(Cq), 138.4 (Cq), 119.5 (CH), 43.6 (CH), 40.6 (CH), 37.3 (CH),
1
mg, 90%): Rf = 0.50 (1:10 EtOAc/petroleum ether 40−60 °C); H
NMR (400 MHz, CDCl3) δ 6.48 (d, J = 15.7 Hz, 4H), 5.80 (dt, J =
15.8 Hz, 5.1 Hz, 4H), 4.27 (dd, J = 5.1 Hz, 1.6 Hz, 8H), 0.92 (s, 36H),
0.08 (s, 24H) ppm; 13C NMR (100 MHz, CDCl3) δ 134.4 (Cq), 134.0
(CH), 127.9 (CH), 64.2 (CH2), 26.1 (CH3), 18.6 (CH3), −5.0 (CH3)
ppm; IR (thin film) νmax 2955, 2929, 2886, 2857 cm−1; MS (ESI) m/z
731.9 ([M + Na]+•, 100), 709.0 ([M]+•, 12), 577.8 ([M − OTBS],
76), 445.6 ([M − (OTBS)2], 85); HRMS (ESI) calcd for
C38H76O4NaSi4 [M]+• 731.4719, found 731.4718; UV−vis (n-hexane)
λmax = 199 nm (ε = 23200), 256 nm (ε = 23200), 330 nm (32200).
(2E,6E)-4,5-Bis((E)-3-hydroxyprop-1-en-1-yl)octa-2,4,6-tri-
ene-1,8-diol (16). A mixture of 15 (54 mg, 0.077 mmol, 1.0 molar
equiv) and TBAF (1.0 M solution in THF, 462 μL, 0.46 mmol, 6.0
molar equiv) in anhydrous THF (1.0 mL) was stirred for 2 h at room
temperature under N2. The solvent was removed under reduced
pressure, and the residue was subjected to flash column chromatog-
raphy using silica gel eluting with 10% MeOH in CH2Cl2.
Recrystallizing twice from acetone yielded pure (2E,6E)-4,5-bis((E)-
3-hydroxyprop-1-en-1-yl)octa-2,4,6-triene-1,8-diol (16) as a colorless
solid (3.0 mg, 15%), Rf = 0.20 (1:10, MeOH/CH2Cl2). An analytical
sample of (2E,6E)-4,5-bis((E)-3-hydroxyprop-1-en-1-yl)octa-2,4,6-tri-
ene-1,8-diol (16) was obtained by recrystallization from MeOH/Et2O
to give colorless needles: mp 108−110 °C; 1H NMR (400 MHz,
CD3OD) δ 6.53 (d, J = 15.7 Hz, 4H), 5.88 (dt, J = 15.8 Hz, 5.6 Hz,
4H), 4.18 (dd, J = 5.6 Hz, 1.5 Hz, 8H) ppm; 13C NMR (100 MHz,
CD3OD) δ 135.6 (Cq), 135.3 (CH), 129.5 (CH), 63.7 (CH2) ppm; IR
(KBr tablet) νmax = 3428, 3300, 2923, 2852, 1635 cm−1; MS (70 eV,
EI) m/z 252.1 ([M]+•, 12), 236.2 ([M − OH], 11) 203.1 ([M −
(OH)2], 21), 129.1 (100); HRMS (EI) calcd for C14H20O4 [M]+•
252.1362, found 252.1364; UV−vis (H2O) λmax = 233.4 nm (ε =
11500), 273.6 nm (ε = 23400).
25.1 (CH2), 25.1 (CH2), 24.9 (CH3) ppm; IR (KBr tablet) νmax
=
2954, 1772, 1696 cm−1; MS (70 eV, EI) m/z 354.1 ([M]+•, 38), 339.1
([M − CH3]) 200.1 (100); HRMS (EI) calcd for C20H22N2O4 [M]+•
354.1580, found 354.1579. Anal. Calcd for C20H22N2O4: C, 67.78; H,
6.26; N, 7.90. Found: C, 67.78; H, 6.21; N, 8.01.
Reaction of 2,3-Divinyl-1,3-cyclohexadiene (5) with 4.0
Molar Equiv of N-Methylmaleimide (19). A mixture of 2,3-
divinyl-1,3-divinylcyclohexadiene (5) (18 mg, 0.14 mmol, 1.0 molar
equiv) and N-methylmaleimide (19) (60 mg, 0.54 mmol, 4.0 molar
equiv) in benzene-d6 (1.0 mL) was stirred at room temperature for 4
days. Solvent was removed under reduced pressure, and the residue
was subjected to flash column chromatography on silica gel eluting
with EtOAc to give 25 as a colorless solid (44 mg, 70%), Rf = 0.30
(100% EtOAc). An analytical sample of 25 was obtained by
recrystallization from EtOH to give colorless needles: mp 220 °C;
1H NMR (400 MHz, C6D6) δ = 2.95−2.85 (m, 1H), 2.81 (s, 3H),
2.75−2.60 (m, 1H), 2.66 (s, 3 H) 2.58 (s, 3H), 2.58−2.38 (m, 4H),
2.27−1.94 (m, 6H), 1.90−1.77 (m, 1H), 1.77−1.55 (m, 4H), 1.55−
1.39 (m, 1H) ppm; 13C NMR (100 MHz, C6D6) δ = 179.3 (Cq), 179.0
(Cq), 178.2 (Cq), 177.4 (Cq), 177.2 (Cq), 176.8 (Cq), 133.7 (Cq),
130.6 (Cq), 44.5 (CH), 43.9 (CH), 42.1 (CH), 41.4 (CH), 40.1 (CH),
37.1 (CH), 34.4 (CH), 33.0 (CH), 32.8 (CH), 32.4 (CH), 26.5
(CH2), 25.3 (CH3), 25.1 (CH2), 24.6 (CH3), 24.5 (CH3), 23.9 (CH2),
21.3 (CH2) ppm; IR (KBr tablet) νmax = 3060, 2937, 2875, 1771, 1694
cm−1; MS (70 eV, ESI) m/z 489.4 ([M+• + Na], 100), 466.4 ([M],
72); HRMS (ESI) calcd for C25H28N3O6 [M + H]+• 466.1978, found
466.1978.
Reaction of TVE (1) with N-Methylmaleimide (19). A mixture
of TVE (1) (50 mg, 0.38 mmol, 1.0 molar equiv) and N-
methylmaleimide (19) (126 mg, 1.14 mmol, 3.0 molar equiv) in
CDCl3 (1 mL) was heated to 50 °C for 6 days in an NMR tube sealed
with a Young’s tap. Solvent was removed by rotary evaporation, and
the residue was subjected to flash column chromatography using silica
gel eluting with 20% EtOAc in petroleum ether 40−60 °C to give 21
as a colorless solid (44 mg, 33%), Rf = 0.25 (20:80, EtOAc/petroleum
ether 40−60 °C); An analytical sample of 21 was obtained by
recrystallization from EtOH to give colorless needles: mp 225 °C dec;
1H NMR (400 MHz, CDCl3) δ 6.30 (dd, J = 17.0 Hz, 10.8 Hz, 1H),
5.90−5.70 (m, 1H), 5.54 (dd, J = 17.5 Hz, 10.5 Hz, 1H), 5.44−5.17
(m, 3H), 5.01 (dd, J = 17.3 Hz, 1.4 Hz, 1H), 4.75 (d, J = 17.4 Hz, 1H),
3.53−3.21 (m, 1H), 3.15−3.03 (m, 3H), 2.99 (d, J = 10.7 Hz, 2H),
2.96 (s, 4H), 2.89 (s, 4H), 2.88−2.72 (m, 3H), 2.35−2.17 (m, 3H),
2.16−1.99 (m, 1H) ppm; 13C NMR δ 179.9 (Cq), 179.4 (Cq), 177.4
(Cq), 177.1 (Cq), 141.0 (Cq), 140.6 (CH), 138.8 (CH), 124.9 (CH),
118.1 (CH2), 117.4 (CH2), 50.9 (Cq), 48.1 (CH), 44.2 (CH), 39.9
(CH), 39.0 (CH), 33.0 (CH), 25.1 (CH3), 25.0 (CH3), 24.4 (CH2),
23.4 (CH2) ppm; IR (KBr tablet) νmax = 3082, 2932, 2882, 2831, 1772,
1684 cm−1; MS (70 eV, EI) m/z 354 ([M]+•, 100), 339 ([M − CH3],
85); HRMS (EI) calcd for C20H22N2O4 [M]+• 354.1580, found
354.1582. Anal. Calcd for C20H22N2O4: C, 67.78; H, 6.26; N, 7.90.
Found: C, 67.75; H, 6.50; N, 7.87.
Double DA Adduct 24 from Mono DA Adduct 23. A mixture
of 23 (12.6 mg, 0.0518 mmol, 1.0 molar equiv) and N-
methylmaleimide (19) (7.0 mg, 0.062 mmol, 1.2 molar equiv) in
CDCl3 (1.0 mL) was stirred at room temperature for 2 days. Solvent
was removed under reduced pressure, and the residue was subjected to
flash column chromatography using silica gel eluting with 50% EtOAc
in petroleum ether 40−60 °C to give 24 as a colorless solid (17 mg,
1
92%). H NMR data were consistent with those reported above.
Triple DA Adduct 25 from Double DA Adduct 24. A mixture
24 (26 mg, 0.073 mmol, 1.0 molar equiv) and N-methylmaleimide
(19) (24 mg, 0.22 mmol, 3.0 molar equiv) in CDCl3 (1.0 mL) was
stirred at reflux overnight. Solvent was removed under reduced
pressure, and the residue was subjected to flash column chromatog-
raphy using silica gel eluting with EtOAc to give 25 as a colorless solid
(31 mg, 91%). 1H NMR data were consistent with that reported
above.
Reaction of 2,3-Divinyl-1,3-cyclohexadiene (5) with 1.0
Molar Equiv of N-Methylmaleimide (19). A mixture of 2,3-
divinyl-1,3-divinylcyclohexadiene (5) (35 mg, 0.26 mmol, 1.0 molar
equiv) and N-methylmaleimide (19) (29 mg, 0.26 mmol, 1.0 molar
equiv) in benzene-d6 (2.44 mL) was stirred at room temperature for 1
h. Solvent was removed under reduced pressure, and the residue was
subjected to flash column chromatography using silica gel eluting with
20% EtOAc in petroleum ether 40−60 °C to give 23 as a colorless oil
J
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Reaction of 2,3-Divinyl-1,3-cyclohexadiene (5) with 2.0
Molar Equiv of DMAD (29). A mixture of 2,3-divinyl-1,3-
divinylcyclohexadiene (5) (35 mg, 0.26 mmol, 1.0 molar equiv) and
DMAD (29) (65 μL, 0.54 mmol, 2.0 molar equiv) in benzene-d6 (1.0
mL) were stirred at room temperature overnight. Solvent was removed
under reduced pressure, and the residue was subjected to flash column
chromatography on silica gel eluting with 10% EtOAc in petroleum
ether 40−60 °C to give dimethyl (S)-5-vinyl-3,7,8,8a-tetrahydronaph-
thalene-1,2-dicarboxylate (30) as a colorless oil (46 mg, 63%): Rf =
0.50 (10:90 EtOAc/petroleum ether 40−60 °C); 1H NMR (400 MHz,
CDCl3) δ 6.34 (dd, J = 17.2 Hz, 10.8 Hz, 1H), 5.85 (s, 1H), 5.71 (s,
1H), 5.36 (d, J = 17.2 Hz, 1H), 5.07 (d, J = 10.6 Hz, 1H), 3.81 (s, 3H),
3.75 (s, 3H), 3.30−2.90 (m, 3H), 2.48−2.10 (m, 2H), 2.02−1.82 (m,
1H), 1.53 (dq, J = 18.0 Hz, 6.2 Hz, 1H) ppm; 13C NMR (100 MHz,
CDCl3) δ 169.6 (Cq), 167.2 (Cq), 140.8 (Cq), 136.2 (Cq), 135.8
(CH), 132.8 (Cq), 127.9 (Cq), 126.4 (CH), 117.0 (CH), 115.4 (CH2),
52.3 (CH3), 52.2 (CH3), 37.1 (CH), 28.4 (CH2), 27.7 (CH2), 26.5
(CH2) ppm; IR (thin film) νmax 3083, 3001, 2951, 2830, 1726 cm−1;
MS (70 eV, EI) m/z 274.1 ([M]+•, 10), 259.1 ([M − CH3], 1) 155.1
([M − (OCH3)2], 100); HRMS (EI) calcd for C16H18O4 [M]+•
274.1205, found 274.1204; UV−vis (acetonitrile) λmax = 209.5 nm (ε
= 18600), 228 nm (ε = 12400).
double DA adduct 31 (17 mg, 34%); fraction 2 contains triple DA
adduct 33 (29 mg, 43%). H NMR data were consistent with that
reported above.
1
ASSOCIATED CONTENT
■
S
* Supporting Information
1H and 13C NMR spectra for all new compounds; computa-
tional details including coordinates for all structures; cifs and
anisotropic displacement ellipsoid plots for compounds cttt-13,
cttt-14, tttt-16, 21, 24, 25, 31, and 33 (CCDC nos. 985422−
985427, 1023991, and 1023992, respectively). This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: m.paddonrow@unsw.edu.au (computational).
*E-mail: michael.sherburn@anu.edu.au (synthetic).
Notes
The authors declare no competing financial interest.
Reaction of 2,3-Divinyl-1,3-cyclohexadiene (5) with 4.0
Molar Equiv of DMAD (29). A mixture of 2,3-divinyl-1,3-
divinylcyclohexadiene (5) (35 mg, 0.26 mmol, 1.0 molar equiv),
DMAD (29) (65 μL, 0.54 mmol, 2.0 molar equiv), and a crystal of
BHT in benzene-d6 (2.0 mL) were stirred at room temperature
overnight. More DMAD was added (65 μL, 0.54 mmol, 2.0 molar
equiv), and the reaction was heated to reflux and stirred overnight.
Upon cooling to room temperature, the solvent was removed under
reduced pressure, and the residue was subjected to flash column
chromatography on silica gel eluting with 30% EtOAc in petroleum
ether 40−60 °C. Fraction 1 contains double DA adduct 31 (42 mg,
38%); fraction 2 contains triple DA adduct 33 (60 mg, 51%).
Trimethyl (8aS,10aS)-7-((Dimethyl-λ3-oxidanyl)carbonyl)-
3,6,8a,9,10,10a-hexahydrophenanthrene-1,2,8-tricarboxylate
(31). Double DA adduct 31 was isolated as a colorless solid, Rf = 0.35;
25:75 EtOAc/petroleum ether 40−60 °C. An analytical sample of
double DA adduct 31 was obtained by recrystallization from ethanol to
give colorless needles: mp 127−128 °C; 1H NMR (400 MHz, CDCl3)
δ 5.58 (s, 2H), 3.78 (s, 6H), 3.74 (s, 6H), 3.28−3.00 (m, 4H), 2.91
(dd, J = 22.2 Hz, 8.3 Hz, 2H) 2.23−1.87 (m, 2H), 1.60−1.19 (m, 2H)
ppm; 13C NMR (100 MHz, CDCl3) δ 168.3 (Cq), 167.3 (Cq), 139.7
(Cq), 137.5 (Cq), 128.2 (Cq), 116.4 (CH), 52.4 (CH3), 52.3 (CH3),
ACKNOWLEDGMENTS
■
This work was supported by the Australian Research Council.
M.N.P.-R. acknowledges that the computational component of
this research was undertaken with the assistance of resources
provided at the NCI National Facility through the National
Computational Merit Allocation Scheme supported by the
Australian Government.
REFERENCES
■
(1) The synthesis of TVE (1) was reported over three publications:
(a) Skattebøl, L. Acta Chem. Scand. 1963, 17, 1683−1693.
(b) Skattebøl, L.; Solomon, S. J. Am. Chem. Soc. 1965, 87, 4506−
4513. (c) Skattebøl, L.; Charlton, J. L.; deMayo, P. Tetrahedron Lett.
1966, 7, 2257−2260.
(2) Two diastereomers of 4 are possible, and no comment was made
on the stereochemical identity of the product from this reaction. For a
discussion, see: Dehmlow, E. V.; Stiehm, T. Tetrahedron Lett. 1990, 31,
1841−1844.
(3) Characterization data for TVE were reported, but no
experimental details have been published for the conversion of DVC
(5) into TVE (1). See ref 1c.
40.0 (CH), 31.4 (CH2), 27.7 (CH2) ppm; IR (KBr tablet) νmax
=
2996, 2951, 2847, 1722, 1648 cm−1; MS (70 eV, EI) m/z 558.2
([M]+•, 4), 527.2 ([M − OCH3], 60), 467.1 (56); HRMS (EI) calcd
for C28H30O12 [M]+• 558.1737, found 558.1728.
(4) Bost, R. W.; Krynitsky, J. A. J. Am. Chem. Soc. 1948, 70, 1027−
1029.
Hexamethyl (3aR,5aS,8aS,10aR)-3,3a,5a,6,8a,9,10,10a-Octa-
hydropyrene-1,2,4,5,7,8-hexacarboxylate (33). Triple DA adduct
33 was isolated as a colorless solid, Rf = 0.50; 50:50 EtOAc/petroleum
ether 40−60 °C. An analytical sample of triple DA adduct 33 was
obtained by recrystallization from ethanol to give colorless needles:
mp 146−148 °C; 1H NMR (400 MHz, CDCl3) δ 3.80 (s, 6H), 3.78 (s,
6H), 3.72 (s, 6H), 3.40−3.12 (m, 4H), 3.07−2.85 (m, 2H), 2.30−2.06
(m, 2H), 1.91−1.71 (m, 2H), 1.59−1.46 (m, 2H) ppm; 13C NMR
(100 MHz, CDCl3) δ 168.6 (Cq), 167.5 (Cq), 166.6 (Cq), 143.4 (Cq),
135.7 (Cq), 129.5 (Cq), 127.2 (Cq), 52.7 (CH3), 52.4 (CH3), 37.8
(CH), 37.0 (CH), 34.8 (CH2), 24.7 (CH2) ppm; IR (KBr tablet) νmax
= 2996, 2951, 2859, 2821, 1722, 1681, 1643 cm−1; MS (70 eV, EI)
m/z 416.1 ([M]+•, 7), 385.1 ([M − OCH3], 52), 357.1 ([M −
CO2Me], 74), 230.1 (100); HRMS (EI) calcd for C22H24O8 [M]+•
416.1471, found 416.1474.
Double and Triple DA Adduct (31 and 33) from Mono DA
Adduct (30). A mixture of mono DA adduct 30 (33 mg, 0.12 mmol,
1.0 molar equiv), DMAD (29) (59 μL, 0.48 mmol, 4.0 molar equiv),
and 1 crystal of BHT in benzene-d6 (1.0 mL) was stirred at reflux for 2
days. Solvent was removed under reduced pressure, and the residue
was subjected to flash column chromatography on silica gel eluting
with 30% EtOAc in petroleum ether 40−60 °C. Fraction 1 contains
(5) (a) Kuhn, R.; Schulz, B. Chem. Ber. 1963, 96, 3200−3208.
(b) Kuhn, R.; Schulz, B. Angew. Chem., Int. Ed. Engl. 1963, 2, 395−396.
(6) Kuhn and Schulz (ref 5) corrected the structure of compound 7,
which was previously assigned as tetraphenyl[3]cumulene in the
following two publications: (a) Bohlmann, F.; Kieslich, K. Abhandl.
Braunschweig. Wiss. Ges. 1957, 9, 147−166. (b) Cadiot, P.;
Chodkiewicz, W.; Pauss-Godineau, J. Bull. Soc. Chim. Fr. 1961,
2176−2193.
(7) Lombardo, L.; Sondheimer, F. Synthesis 1980, 950−952.
(8) Pons, J. M.; Santelli, M. J. Org. Chem. 1989, 54, 877−884.
(9) For recent reviews, see: (a) Mackay, E. G.; Sherburn, M. S. Pure
Appl. Chem. 2013, 85, 1227−1239. (b) Hopf, H.; Sherburn, M. S.
Angew. Chem., Int. Ed. 2012, 51, 2298−2338.
(10) Lindeboom, E. J.; Willis, A. C.; Paddon-Row, M. N.; Sherburn,
M. S. Angew. Chem., Int. Ed. 2014, 53, 5440−5443.
(11) We ascribe these results to 6π-electrocyclizations and
accompanying decomposition. Related transformations involving
DVC (5) are described later.
(12) The low yield reflects a problem of isolation: H NMR analysis
of the crude product with an internal standard gave a 70% yield of the
tetraol.
1
K
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(13) We use the terms cisoid and transoid to identify 1,3-butadiene
conformations in which the values of the dihedral angles are less than
90° and greater than 90°, respectively.
(14) Curtiss, L. A.; Redfern, P. C.; Raghavachari, K. J. Chem. Phys.
2007, 126, 084108/1−12.
(15) Frisch, M. J. et al. Gaussian 09, revision E.1; Gaussian Inc.:
Wallingford, CT, 2009. See the Supporting Information for the full list
of authors.
(16) Irradiation at wavelengths of 254, 300, and 350 nm were
investigated for the 6π-electrocyclization in a variety of solvents
(benzene, diethyl ether, petroleum ether, acetonitrile). Irradiation with
a lamp source providing an output at 350 nm gave superior results,
irrespective of the solvent. DVC (5) was the major product formed,
but small amounts of unidentified byproducts were also present in all
experiments.
(17) Payne, A. D.; Willis, A. C.; Sherburn, M. S. J. Am. Chem. Soc.
2005, 127, 12188−12189.
(18) Payne, A. D.; Bojase, G.; Paddon-Row, M. N.; Sherburn, M. S.
Angew. Chem., Int. Ed. 2009, 48, 4836−4839.
(19) See, for example: Miller, N. A.; Willis, A. C.; Paddon-Row, M.
N.; Sherburn, M. S. Angew. Chem., Int. Ed. 2007, 46, 937−940.
(20) Nevertheless, a significant amount of the electrocyclization
product was still isolated from this reaction.
(21) For a review of diene-transmissive Diels−Alder sequences, see
ref 9b. This term was introduced by Tsuge: Tsuge, O.; Wada, E.;
Kanemasa, S. Chem. Lett. 1983, 12, 239−242.
(22) Bradford, T. A.; Payne, A. D.; Willis, A. C.; Paddon-Row, M. N.;
Sherburn, M. S. J. Org. Chem. 2010, 75, 491−494.
(23) Alder, K.; Stein, G. Angew. Chem. 1937, 50, 510−519.
(24) Paddon-Row, M. N.; Sherburn, M. S. Chem. Commun. 2012, 48,
832−834.
(25) Toombs-Ruane, H.; Pearson, E. L.; Paddon-Row, M. N.;
Sherburn, M. S. Chem. Commun. 2012, 48, 6639−6641.
(26) Skattebøl (ref 1b) reports the generation of a mixture of double
and triple cycloadducts on exposure of DVC (5) to maleic anhydride.
These adducts were separated by sublimation and no yields were
reported.
(27) Mans, D. J.; Cox, G. A.; RajanBabu, T. V. J. Am. Chem. Soc.
2011, 133, 5776−5779.
(28) Oda, H.; Kobayashi, T.; Kosugi, M.; Migita, T. Tetrahedron
1995, 51, 695−702.
(29) Liu, Y.; Di, C.-a.; Du, C.; Liu, Y.; Lu, K.; Qiu, W.; Yu, G.
Chem.Eur. J. 2010, 16, 2231−2239.
(30) Stulgies, B.; Prinz, P.; Magull, J.; Rauch, K.; Meindl, K.; Ruhl, S.;
̈
de Meijere, A. Chem.Eur. J. 2005, 11, 308−320.
(31) Darwish, A.; Lang, A.; Kim, T.; Chong, J. M. Org. Lett. 2008, 10,
861−864.
L
dx.doi.org/10.1021/jo5021294 | J. Org. Chem. XXXX, XXX, XXX−XXX