1-phenyl-1,2-cyclohexadiene: Generation, interception by activated olefins, dimerisation and trimerisation

Article abstract of DOI:10.1002/chem.200900717

Four possible precursors of 1-phenyl-1,2-cyclohexadiene (2) were examined, namely, 6,6-dibromo-1-phenylbicyclo[3.1.0]hexane, (1α,5α,6α)- 6-bromo-6-fluoro-1-phenylbicyclo-[3.1.0]hexane, 1-bromo-2-phenylcyclohexene and 1-bromo-6-phenylcyclohexene. All four

Full text of DOI:10.1002/chem.200900717

DOI: 10.1002/chem.200900717
1-Phenyl-1,2-cyclohexadiene: Generation, Interception by Activated Olefins,
Dimerisation and Trimerisation**
Manfred Christl,*[a] Michael Schreck,[a] Thomas Fischer,[a] Marcus Rudolph,[a]
Damien Moigno,[a] Hartmut Fischer,[a] Stephan Deuerlein,[b] and Dietmar Stalke*[b]
Dedicated to Professor Waldemar Adam on the occasion of his 72nd birthday
Abstract: Four possible precursors of
1-phenyl-1,2-cyclohexadiene (2) were
[4+2] cycloadduct each. To create the
[2+2] cycloadducts, the p bond of 2
that is more remote from the phenyl
group reacted, whereas the p bond of 2
conjugated with the phenyl group ex-
clusively produced the [4+2] cycload-
ducts. The generation of 2 in the ab-
sence of a trapping reagent brought
about relatively good yields of a dimer
or a trimer of 2 depending on the
mode of the liberation of 2. Being de-
rivatives of triphenylene, the dimer as
well as the trimer have unusual struc-
tures, thereby indicating that a phenyl
group is participating in the formation
of these compounds. The most surpris-
ing structure of the trimer was elucidat-
ed by X-ray crystal diffraction. As to
the mechanisms, diradical intermedi-
ates are proposed both for the cycload-
ditions and for the dimerisation. The
initial steps of the latter seem to pro-
ceed also in the trimerisation.
examined,
namely,
6,6-dibromo-1-
phenylbicycloACHTUNGTRENNUNG[3.1.0]hexane, (1a,5a,6a)-
6-bromo-6-fluoro-1-phenylbicyclo-
ACHTUNGTRENNUNG[3.1.0]hexane, 1-bromo-2-phenylcyclo-
hexene and 1-bromo-6-phenylcyclohex-
ene. All four compounds could be con-
verted into 2, as demonstrated by the
products of the interception of 2 with
activated olefins. Styrene, 1,1-diphenyl-
ACHTUNGTRENNUNG
Keywords: cyclic allenes · cycload-
dition · diradicals · polycycles ·
regioselectivity
ACHTUNGTRENNUNG
as the first three gave [2+2] cycload-
ducts of 2, the last two provided one
Introduction
agents in [2+2] and [4+2] cycloadditions, thus indicating a
significant synthetic potential.[1] This was impressively con-
firmed by the preparation of dozens of new cephalosporin
derivatives by means of the dihydro-5d2-1,3-thiazine inter-
mediate 1.[1,3]
Six-membered cyclic allenes, in which the ring contains
second-row elements exclusively or even a sulfur atom, are
extremely short-lived intermediates, of which only very few
could be observed directly.[1,2] Nevertheless, many of them
have been efficiently intercepted by appropriate trapping re-
[a] Prof. Dr. M. Christl, Dr. M. Schreck, Dr. T. Fischer, Dr. M. Rudolph,
Dr. D. Moigno, Dr. H. Fischer
Institut fꢀr Organische Chemie
Universitꢁt Wꢀrzburg
Am Hubland, 79074 Wꢀrzburg (Germany)
Fax : (+49)931-8884606
E-mail: christl@chemie.uni-wuerzburg.de
Compared with the unsubstituted 1,2-cyclohexadiene,
which has been investigated extensively, there are ten
simply substituted derivatives, the reactivity of which has
been only fragmentarily studied.[1] One of them is 1-phenyl-
1,2-cyclohexadiene (2), although it was the target of three
research groups. In addition to the Wꢀrzburg team, some re-
sults of which were reported in a review,[1] Tolbert, Johnson
et al.[4] as well as Ceylan and Budak[5] examined the inter-
[b] Dr. S. Deuerlein, Prof. Dr. D. Stalke
Institut fꢀr Anorganische Chemie
Universitꢁt Gçttingen
Tammannstrasse 4, 37077 Gçttingen (Germany)
Fax : (+49)551-393373
E-mail: dstalke@chemie.uni-goettingen.de
[**] Cycloallenes, Part 21. For Part 20, see reference [13].
11256
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11256 – 11265

FULL PAPER
mediacy of 2. The American group[4] proved the existence of
2 by trapping it with furan and 1,3-diphenylisobenzofuran,
resulting in the formation of the [4+2] cycloadducts 3 and 4.
As the mode for the generation of 2, the photolysis or the
thermolysis of the potassium salt of 1-chloro-2-phenylcyclo-
hexene was employed. Compound 4 also emerged from the
Doering–Moore–Skattebøl reaction of 6,6-dibromo-1-
Scheme 1. Preparation of the precursors 5 and 7–9 for 2. TEBA=benzyl-
triethylammonium chloride.
phenylbicyclo
ACHTUNGTRENNUNG[3.1.0]hexane (5) in the presence of 1,3-
diphen
G
quantitatively to the allyl dibromide 6 at room temperature
within 11 h and three days, respectively. Ceylan and Budak
claimed to have achieved this rearrangement by heating at
1508C, whereas heating at 1808C gave rise to biphenyl.[5] In
our hands, biphenyl was formed already at much lower tem-
peratures from 5,[9] and 6 could only be obtained by keeping
a solution of 5 at room temperature. Previously, 6 was de-
scribed as being a colourless liquid,[5,9] but we have now iso-
lated colourless crystals with m.p. 558C. The published
13C NMR spectroscopic data[5] deviate substantially from
ours.
bromo-2-phenylcyclohexene (8) with potassium tert-butoxide
(KOtBu), but obtained only rather indirect evidence in
favour of the intermediate 2 (see below).
Including our previous results, which have been described
only briefly to date,[1] we herein report on the generation of
2 from 5 and from the bromofluorophenylbicyclohexane 7
by the Doering–Moore–Skattebøl reaction as well as from 8
and its isomer 9 by b eliminations. For the trapping of 2, ac-
tivated olefins were utilised that are typical reaction part-
ners of six-membered cyclic allenes.[1] In the absence of such
a reagent, 2 underwent a dimerisation or trimerisation de-
pending on the mode of the liberation, with the dimer and
the trimer having unusual structures. The ultimate goal of
our studies was the enantioselective generation of 2, which
permitted the examination of the stereochemical course of
the cycloadditions onto 2.[6]
As a precursor of 2 in terms of a b elimination, 8 was pre-
pared. It had formerly been obtained as the single product
by the reduction of 6 with LiAlH4 in tetrahydrofuran.[5,9]
Performing this reaction in diethyl ether, we arrived at a
1.7:1.0 mixture of 8 and its isomer 9, which we isolated in 32
1
and 8% yield, respectively. Ceylan and Budak[5] reported H
and 13C NMR spectroscopic data of 8. Whereas the former
agree with ours, the latter differ from ours.
Results and Discussion
For the enantioselective generation of 2 by employing the
Doering–Moore–Skattebøl reaction,[6] compound 5 is a can-
didate since it is chiral. However, we expected problems in
the resolution of the racemate by chromatography on a
chiral and enantiopure phase due to its thermolability.
Therefore, we envisaged the synthesis of the bromofluoro-
carbene adduct 7 of 1-phenylcyclopentene. Our experience
Preparation of the precursors for 2: The straightforward
precursors for the title allene 2 are the dibromcarbene
adduct 5 of 1-phenylcyclopentene and 1-bromo-2-phenylcy-
clohexene (8), which should release 2 by an a and a b elimi-
nation, respectively. Thus, we treated bromoform with
KOtBu in the presence of 1-phenylcyclopentene and ob-
tained the dibromophenylbicyclohexene 5 in 54% yield
(Scheme 1). Tolbert, Johnson et al.[4] arrived at merely 20%
yield and characterised 5, which served as precursor of 2 en
route to 4, only as being “thermally unstable”. Ceylan and
Budak[5] described 5 as a colourless liquid, collected in 67%
yield, whereas we isolated colourless crystals with m.p. 44.5–
458C. In addition, our 13C NMR spectroscopic data are se-
verely at variance with the published ones.[5]
with exo-6-bromo-endo-6-fluorobicycloACTHNUTRGENUG[N 3.1.0]hexane deriva-
tives[10] made us confident that 7 should rearrange much
ꢀ
slower than 5. The significantly higher energy of a C F rela-
tive to a C Br bond[11] is the basis for this anticipation.
ꢀ
Performed under the conditions of the phase-transfer ca-
AHCTUNGTREGtNNUN alysis, the addition of bromofluorocarbene onto 1-phenylcy-
clopentene afforded 7 in 16% yield (Scheme 1). We did not
observe any indication of the thermal rearrangement of 7,
and it could be subjected to chromatography without de-
composition. Because of the rather small yield of 7, we
assume that its diastereomer was also formed. Since this
compound should be similarly reactive as 5, it most probably
underwent an SN1 process with rearrangement in the polar
reaction medium, the product of which could well have
been lost during the workup.
On the basis of the readily occurring rearrangement of
the unsubstituted 6,6-dibromobicycloACTHNUTRGNEUNG
[3.1.0]hexane[7] and its
1-methyl derivative,[8] it could not be expected that 5 is
highly persistent at room temperature. In fact, it can be
stored over extended periods only at ꢀ308C. Dissolved in
dichloromethane or chloroform, it transformed virtually
Chem. Eur. J. 2009, 15, 11256 – 11265
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11257

M. Christl, D. Stalke et al.
Cycloadditions of 2 with activated olefins: The interception
of a six-membered cyclic allene by an activated olefin with
formation of a [2+2] or [4+2] cycloadduct is adequate proof
for the existence of such an intermediate.[1] Because of its
rather rapid trapping of the parent 1,2-cyclohexadiene,[1,12]
at first styrene was chosen as reagent, and 2 was generated
from 5 by methyllithium. The diastereomeric [2+2] cycload-
ducts endo- and exo-10 were obtained as a 1:1 mixture in
51% yield (Scheme 2). Dissolved in light petroleum ether,
such a mixture furnished crystals of pure endo-10 on storage
at ꢀ308C.
ducts. In some cases, only the exo isomer was observed; in
others, the endo compound emerged as the minor compo-
nent of the isomeric mixture,[1] and 1 and two of its deriva-
tives[1,3] as well as 1-oxa-2,3-cyclohexadiene[1,10a] furnished
both diastereomers in the same amount. The latter outcome
is an unambiguous indication of the kinetic control of the
product formation, since in the examples examined so far,
namely, the styrene adducts of 1,2-cyclohexadiene, its 1-
methyl derivative[8] and 1-oxa-2,3-cyclohexadiene,[10a] the exo
diastereomer is the thermodynamically more stable one, as
demonstrated by establishing the equilibrium by thermolysis.
Accordingly, we now heated a solution of pure endo-10 in
C6D6 at 150–1608C and noticed the generation of exo-10.
After six hours, the ratio of endo-/exo-10 was determined to
be about 1:1, after 26 h about 1:10. Because of the slow de-
composition of the sample, occurring under these conditions,
the equilibrium ratio could be smaller than 1:10.
The comparison of the temperature necessary for the
above isomerisation with that required in the case of the sty-
rene adducts of 1,2-cyclohexadiene, the equilibrium ratio
endo/exo of which amounts to 1:13,[8] proves that the phenyl
group at the six-membered ring of 10 does not accelerate
the interconversion of the diastereomers.
We assume that the reaction between 2 and styrene pro-
ceeds in two steps via the diradicals 13a and 13b
(Scheme 3) and its enantiomers. At ꢀ308C, they probably
Scheme 2. Generation of 2 from 5 by methyllithium and interception of 2
by styrene, 1,1-diphenylethene and indene.
Scheme 3. Mechanisms proposed for the formation of endo- and exo-10
from 2 and styrene and the conversion of endo- into exo-10.
The same products 10 in the same ratio could be prepared
in a one-pot procedure from 1-phenylcyclopentene without
isolation of 5, but the yield was only 24%. To that end, a so-
lution of 1-phenylcyclopentene and tetrabromomethane in
diethyl ether was treated with methyllithium at ꢀ608C. The
resulting solution of 5 was admixed with styrene, and then a
second batch of methyllithium was added at ꢀ308C, giving
rise to 2, which underwent the addition onto styrene.
Dissolved in pentane, the 1:1 mixture of endo- and exo-10
turned out not to be persistent for long periods at room
temperature. After two months, it had completely and clean-
ly converted into a product that could be a 1:1 mixture of
isomers according to the NMR spectra, but the structure of
which was not elucidated.
emerge in the ratio of 1:1 and then collapse to give endo-
and exo-10, respectively, without equilibration, that is, with-
out rotations around the newly formed single bond and the
one that results from the double bond of styrene. More de-
tails about the stereochemical course of this cycloaddition
are provided by the utilisation of the enantioselectively gen-
erated 2.[6] At about 1508C, the ring closure becomes rever-
sible, thereby leading to the equilibration of 13a and 13b
and eventually to the thermodynamically more stable com-
pound exo-10. The instability of endo-10 is caused by the
steric crowding between the endo-oriented phenyl group
and the cyclohexene moiety. However, no clear-cut reason
can be given for the regioselectivity of the cycloaddition,
which comes down to the regioselective collapse of the di-
Styrene as trapping reagent of six-membered cyclic al-
lenes leads to various ratios of endo- and exo-[2+2] cycload-
11258
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11256 – 11265

1-Phenyl-1,2-cyclohexadiene
FULL PAPER
A
located at the ethene subunit. This regioisomer was ob-
served neither by the American group[4] nor by us.
The constitution of the products 3 and 14 can easily be de-
1
also gave rise to [2+2] cycloadducts, and with the same re-
gioselectivity as with styrene (Scheme 2). Whereas the
adduct 11 of 1,1-diphenylethene was obtained by a one-pot
procedure from 1-phenylcyclopentene through 5 in 25%
yield, the indene adduct 12 was prepared from pure 5, but
could be isolated in only 7% yield. The configuration indi-
cated by formula 12 was not proven but assumed in analogy
to the indene adducts of 1[3] and 3d2-1H-naphthalene.[13] The
desired products of both experiments were accompanied by
significant amounts of the trimer 17 of 2, described below.
This finding characterises these activated olefins as sluggish-
ly reacting allenophiles. Although benzofuran gave an
adduct with 1 in 63% yield,[3] no reaction of 2 with benzo-
furan occurred and only 17 was observed.
Furan and substituted furans were frequently employed to
intercept six-membered cyclic allenes and, as a rule, gave
rise to [4+2] cycloadducts of these.[1,13] As demonstrated by
the products 3 and 4,[4] this rule is valid also for 2. We now
trapped 2 by 2,5-dimethylfuran and arrived at the [4+2] cy-
cloadduct 14 (Scheme 4). The intermediate 2 was liberated
rived from their H NMR spectra, which contain three sig-
nals of olefinic protons and thus determine the position se-
lectivity of the furan cycloaddition onto 2. Moreover, Tol-
bert, Johnson et al.[4] proved the structure of 3 by X-ray dif-
fraction. Given these facts, the trapping reactions of 2 pro-
ceed with different regiospecifities, as the allene double
bond remote from the phenyl group is active in the [2+2]
cycloadditions, whereas the double bond carrying the phenyl
group reacts in the [4+2] cycloadditions. This phenomenon
is not unusual with six-membered cyclic allenes.[1] However,
the concerted process proposed for the [4+2] cycloadditions
of 1-oxa-2,3-cyclohexadiene[10a] will most probably not meet
the reality, since quantum chemical calculations[14] and the
experimental result of the addition of (Z,Z)-1,4-dideuterio-
1,3-butadiene onto 1,2-cyclohexadiene[1] clearly favour two-
step reactions via diradicals.
In consequence, we assume that diradicals of type 15
emerge as intermediates en route to 3 and 14 (Scheme 5).
But why the 1!4 ring closure of the diradicals 13
Scheme 5. Mechanism proposed for the formation of the [4+2] cycload-
ducts 3 and 14 of 2.
(Scheme 3) takes place at the less substituted terminus of
the phenylallyl-radical moiety and the 1!6 collapse of the
diradicals 15 at the phenyl-substituted one, cannot be ex-
plained satisfactorily up to now. A theoretical study of these
reactions would be highly desirable.
Scheme 4. Generation of 2 from 7, 8 and 9 by methyllithium and potassi-
um tert-butoxide, respectively, and its interception by 2,5-dimethylfuran.
Dimerisation and trimerisation of 2: Short-lived six-mem-
bered cyclic allenes oligomerise if they are generated in the
absence of a trapping reagent. Presumably, some do so un-
specifically, because a homogeneous product could not be
identified. Others, particularly 1,2-cyclohexadiene and its
simply substituted derivatives, furnish dimers, which contain
a 1,2-bismethylenecyclobutane subunit[1] and are thus analo-
gous to the dimers of non-cyclic allenes.[15–17] The informa-
tion given by Tolbert, Johnson et al.[4] as well as by Ceylan
and Budak[5] as to the liberation of 2 in the absence of a
trapping reagent appears to indicate a behaviour of 2 that
deviates from the above rules, since only biphenyl was ob-
served, which should emerge from 2 by base-catalysed rear-
rangement and subsequent dehydrogenation.
from three precursors, namely, from 7 by methyllithium in a
Doering–Moore–Skattebøl reaction, with the yield of 14
amounting to 22%, and from 8 as well as 9 by KOtBu in b
eliminations.
Tolbert, Johnson et al.[4] had prepared the furan adduct 3
of 2 in 50% yield, having generated 2 by photolysis or ther-
molysis of the potassium salt of 1-chloro-2-phenylcyclohex-
ene, and noticed that the liberation of 2 from 5 in a Doer-
ing–Moore–Skattebøl reaction failed due to furan lithiation.
On treatment of a mixture of 8 and 9 with KOtBu in the
presence of furan, we now obtained 3 as well. Different
from these findings, Ceylan and Budak[5] reported the for-
mation of 1-phenylnaphthalene on reaction of 8 with KOtBu
in the presence of furan. Their explanation of this result
cannot be correct, since they did not assume 3 as product of
the trapping of 2 but its regioisomer with the phenyl group
Different from these findings, we now observed that, de-
pendent on the mode of the generation of 2, the triphenyl-
AHCTUNGTREGeNNNU ne derivatives 16 or 17 (Scheme 6) are formed in significant
Chem. Eur. J. 2009, 15, 11256 – 11265
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11259

M. Christl, D. Stalke et al.
illustrates the structure of 17 in the solid state, in which the
methylene group of position 17 (position 3 in Scheme 6) is
disordered. The most abundant fragment of the MS spec-
trum (m/z 311) also supports this structure, since it indicates
the loss of the phenylcyclohexenyl group from the molecular
ion.
As is manifest by the triphenylene skeleton of 16 and 17,
a phenyl group of 2 actively participated in the product for-
mation. In the synthesis of rubrene from chlorotriphenylal-
lene, such a step has been employed since 1926,[18] even
though the mechanism was only recognised in 1979.[19]
Other examples are provided by the thermolysis of the 1,2-
bismethylenecyclobutane derivatives produced at low tem-
peratures upon dimerisation of triphenylallene[20] and 1-
(2,2’-biphenylylen)-3-phenylallene.[17] Heretofore, only one
case has been documented that describes the isolation of a
primary product resulting from the attack of a radical centre
of a phenyl-substituted tetramethyleneethane diradical at
the phenyl group of the second molecule half, that is, the
dimer 20 of racemic 1-phenyl-1,2-cyclooctadiene (19), which
was generated from racemic 8,8-dibromo-1-phenylbicyclo-
Scheme 6. Formation of the dimer 16 and the trimer 17 of 2.
amounts. These products are a dimer and a trimer of 2, re-
spectively. Interestingly, the dimer 16 only emerged on the b
elimination route to 2, that is, on treatment of 8 with
KOtBu, whereas the trimer 17 exclusively resulted as prod-
uct of the Doering–Moore–Skattebøl reaction of 5. We did
not observe biphenyl as a consecutive product of 2, but we
did not search for this compound in detail, either.
The identity of 17 was established by single-crystal X-ray
diffraction and was a great surprise, which is why our previ-
ous proposal for the structure of the trimer, being based on
NMR spectroscopic data only,[1,9] had to be revised. Figure 1
[5.1.0]octane (18; Scheme 7).[16,21]
ACHTUNGTRENNUNG
Scheme 7. Dimerisation of racemic 1-phenyl-1,2-cyclooctadiene (19).[16,21]
The first two steps from 2 en route to 16 and 17 should be
analogous to the pathway proposed for the conversion of 19
into 20.[16,21] Thus, two molecules of 2 should combine to
give the tetramethyleneethane diradical 21, which then does
not undergo the standard reaction of such species with for-
mation of a 1,2-bismethylenecyclobutane derivative,[1] but
collapses by bond formation of a phenyl-bearing allyl-radical
terminus with the phenyl group of the second molecule half
furnishing the 1-methylene-2,4-cyclohexadiene derivative 22
(Scheme 8). Corresponding exactly to 20, derivative 22 is
not isolable, since it is transformed to 16 or 17 under the re-
action conditions.
The presence of the strong base KOtBu should lead to the
deprotonation of the methylenecyclohexadiene subunit of
22 with aromatisation, that is, the generation of a 1-phenyl-
allyl anion moiety, which is then reprotonated in position 3
with the formation of 16. This pathway has a precise parallel
in the dimerisation of 1-phenyl-1,2-cycloheptadiene liberat-
ed from appropriate precursors by b elimination with
KOtBu.[1,9]
Figure 1. Molecular structure of 17 as determined by single-crystal X-ray
diffraction. The anisotropic displacement parameters are depicted at the
50% probability level. The methylene group of position 17 (position 3 in
Scheme 6) is disordered.
11260
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11256 – 11265

1-Phenyl-1,2-cyclohexadiene
FULL PAPER
two steps, with the first being a hydrogen-atom transfer
from 22 to 2, which would give rise to the phenylallyl-radical
derivatives 23 and 24. This radical pair could collapse to 17,
but there would be several other possibilities for product
formation, with the routes to the three diastereomers of 17
being the most prominent ones. Indeed, 17 was the major
component of the product mixture and the only compound
identified, yet one side product showed signals of three ole-
1
finic protons in the H NMR spectrum that are very similar
to those of 17 and thus could originate from a diastereomer
of 17.
Why does the reaction stop at the dimer 20 in the case of
racemic 1-phenyl-1,2-cyclooctadiene (19), whereas the anal-
ogous dimer 22 of 2 binds an additional monomer to give 17
under very similar conditions? This variation is certainly
caused by different strain energies of 2 and 19, which are es-
timated to be 32 and 5 kcalmolꢀ1, respectively, if the calcu-
lated values of the parent hydrocarbons 1,2-cyclohexadiene
and 1,2-cyclooctadiene[23] are applied. Clearly, the strain-
energy difference of 27 kcalmolꢀ1 brings about a much
higher reactivity of 2 relative to that of 19, which is why 2
readily attacks 22 to afford 17 as proposed in Scheme 9,
whereas 19 does not interact with 20. Definitely supported
Scheme 8. Mechanism proposed for the dimerisation of 2 with formation
of 22, which should be converted into 16 or 17 depending on the reaction
conditions.
When 2 is generated from 5, the intermediate 22 is ex-
posed to methyllithium, which is also a rather strong base,
and yet a deprotonation does not occur. Instead, 22 seems
to react with another molecule of 2 and thus produces the
trimer 17 (Scheme 8). Apparently, methyllithium is subject
to a fast bromine–lithium exchange with 5, and the resulting
carbenoid rapidly eliminates lithium bromide to give 2.
Hence, 22 never faces a large concentration of the base,
when methyllithium is slowly added to a solution of 5.
There may be a second reason for the failure of methyllithi-
um and the carbenoid generated by it to deprotonate 22,
namely, the well-known phenomenon that, at a given pK
value, carbon atom bases abstract protons considerably
more slowly than hetero atom bases such as KOtBu.[22]
ꢀ
by the weakness of the respective C H bond of the methyl-
enecyclohexadiene subunit of 22, the ability of 2 to abstract
a hydrogen atom or to perform an ene reaction is unprece-
dented for six-membered cyclic allenes.
Conclusion
Scheme 9 illustrates two possible mechanisms for the con-
version of 22 into 17. One is a one-step process, that is, an
ene reaction. As the shape of the molecule 17 (Figure 1)
demonstrates, the transition state could suffer from steric
overcrowding, thereby reducing the likelihood for this con-
certed reaction. The alternative pathway would proceed in
Both the Doering–Moore–Skattebøl reaction and the b-
elimination route were shown to be suitable for the genera-
tion of 1-phenyl-1,2-cyclohexadiene (2). This intermediate
behaves towards activated olefins in the same way as 1,2-cy-
clohexadiene and a number of its derivatives studied previ-
ously.[1] Accordingly, it is intercepted with the formation of
[2+2] or [4+2] cycloadducts. Up to now, a satisfactory ex-
planation has not been advanced for the astounding phe-
nomenon that the p bond of 2 that is more remote of the
phenyl group is employed in the [2+2] cycloadditions,
whereas the p bond conjugated with the phenyl group is ex-
clusively active in the [4+2] cycloadditions. The liberation of
2 in the absence of a trapping reagent leads to the formation
of the dimer 16 or the trimer 17 of 2, depending on the
mode of the generation of 2. Both products exhibit unusual
structures, since the phenyl group of one molecule of 2 ac-
tively participates in the assembly of the skeletons. In addi-
tion, the association of the third molecule of 2 en route to
17 is unprecedented in the reactivity of six-membered cyclic
allenes.
The preparation of a precursor of 2 that would allow the
enantioselective generation of 2 was the most important
goal of this investigation, which was fully reached by the
synthesis of the racemic bromofluorocarbene adduct 7 of 1-
phenylcyclopentene and the demonstration of the trapping
of 2, generated from 7, by 2,5-dimethylfuran, which gave
Scheme 9. Mechanisms proposed for the conversion of the assumed
dimer 22 of 2 into the trimer 17 by 2.
Chem. Eur. J. 2009, 15, 11256 – 11265
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11261

M. Christl, D. Stalke et al.
formed was filtered off and washed with DEE (2ꢃ10 mL). The filtrate
and the washings were combined, extracted with water (20 mL) and
dried with MgSO4. After the evaporation of the solvent in vacuo, a
yellow oil (6.47 g, 78%) remained, which essentially consisted of 8 and 9
in the ratio of 1.7:1.0 according to the 1H NMR spectrum. A 2.35 g por-
tion of this material was subjected to flash chromatography (silica gel,
PE), which furnished 8 (970 mg, 32%) and 9 (228 mg, 8%) as colourless
oils.
rise to the [4+2] cycloadduct 14. In the following paper in
this issue, we report the results obtained on liberation of 2
from enantiopure 7 in the presence of activated olefins.
Experimental Section
Compound 8: 1H NMR (400 MHz, CDCl3): d=1.76–1.85 (m, 4H), 2.39
(m, 2H), 2.64 (m, 2H), 7.22 (o-H), 7.27 (p-H), 7.34 ppm (m-H);
13C NMR (50 MHz, CDCl3): d=22.7, 24.6 (C4,C5), 33.8, 36.7 (C3, C6),
120.0 (C1), 126.9 (p-C), 127.9 (o-C), 128.0 (m-C), 137.6, 143.3 ppm (C2, i-
C); as far as specified, the assignment is supported by a C,H COSY spec-
trum; MS (70 eV, EI): m/z (%): 238, 236 (32, 32) [M]+; 157 (50), 142
(12), 129 (62), 128 (29), 127 (11), 115 (31), 91 (100), 79 (20), 77 (25), 76
(13), 64 (13), 63 (12), 51 (18); elemental analysis calcd (%) for C12H13Br:
C 60.78, H 5.53; found: C 61.10, H 5.29.
Compound 9: characterised by NMR spectroscopy only. 1H NMR
(400 MHz, CDCl3): d=1.51–1.67 (m, 2H), 1.80 (m, 1H), 2.10–2.22 (m,
3H), 3.70 (m, 1H), 6.35 (td, J=4.0, 1.2 Hz, 1H), 7.21–7.28 (o-H, p-H),
7.33 ppm (m-H); 13C NMR (100 MHz, CDCl3): d=17.8 (C4), 27.7 (C5),
33.8 (C3), 49.7 (C6), 124.3 (C1), 126.6 (p-C), 128.2, 128.3 (m-C, o-C),
131.9 (C2), 143.3 ppm (i-C); the assignment is supported by a DEPT
spectrum.
General: NMR: Bruker AC 200, AC 250, Avance 400, and DMX 600
spectrometers; chemical shifts (d) in ppm relative to Me4Si (d=
0.00 ppm) by using solvent signals as internal reference [CDCl3: d=
7.26 ppm (1H NMR of CHCl3) and 77.0 ppm (13C NMR); C6D6: d=
7.16 ppm (1H NMR of C6D5H) and 128.0 ppm (13C NMR)]. IR: Perkin–
Elmer 1420 ratio recording infrared spectrophotometer. UV: Perkin–
Elmer UV/Vis spectrophotometer 330, JASCO V-570 UV/Vis/NIR spec-
trophotometer. MS: Varian MAT CH 7, Finnigan MAT 8200 and MAT
90. Elemental analysis: LECO CHNS 932. Frequently used solvents:
DEE=diethyl ether, EA=ethyl acetate, PE=light petroleum ether (b.p.
30–50 or 40–658C), THF=tetrahydrofuran.
6,6-Dibromo-1-phenylbicyclo
G
(5):
KOtBu
(13.1 g,
117 mmol) was suspended in
a
(12.1 g, 83.9 mmol) in PE (120 mL). The mixture was vigorously stirred
and treated dropwise with a solution of bromoform (23.3 g, 92.2 mmol) in
PE (30 mL) at ꢀ158C. After the cooling bath had been removed, the re-
action mixture was allowed to warm to 208C and was cautiously hydro-
lysed (50 mL). Some solid material was filtered off, and the layers were
separated. The organic layer was dried with MgSO4, and the solvent was
evaporated in vacuo without warming. The remaining orange-red oil
(21.8 g) was dissolved in methanol (150 mL) at 208C, and the solution
was kept at ꢀ308C overnight, which gave rise to colourless crystals of 5
(14.3 g, 54%). M.p. 43–458C; m.p. 44.5–458C after one recrystallisation
from methanol; 1H NMR (250 MHz, CDCl3): d=1.76–1.96 (m, 2H), 2.12
(m, 1H), 2.18–2.41 (m, 2H), 2.43–2.58 (m, 2H), 7.14–7.40 ppm (m, 5H);
13C NMR (63 Hz, CDCl3): d=26.0 (C3), 30.3 (C4), 37.9 (C2), 42.8 (C5),
46.3 (C1), 49.8 (C6), 127.1 (p-C), 128.1, 128.2 (m-C, o-C), 140.8 ppm (i-
C); the assignment is based on the comparison of the values with those
AHCTUNGTERNN(GUN 1a,5a,6a)-6-Bromo-6-fluoro-1-phenylbicycloAHCTUNGTREGUN[NN 3.1.0]hexane (7): A solu-
tion of 1-phenylcyclopentene[24] (10.0 g, 69.3 mmol), benzyltriethylammo-
nium chloride (250 mg, 1.10 mmol) and dibromofluoromethane[25] (21.4 g,
111.6 mmol) in CH2Cl2 (20 mL) was cooled in an ice bath. After addition
of aqueous sodium hydroxide (15 g, 375 mmol in 15 mL of water), which
was pre-cooled to 58C, the mixture was vigorously stirred for 3 d. During
that period, the ice of the cooling bath was not replaced as it melted.
Thus, the mixture reached room temperature after several hours. After
3 d, water (50 mL) was admixed and the layers were separated. The
aqueous phase was extracted with CH2Cl2 (3ꢃ30 mL), the combined or-
ganic layers were dried with MgSO4 and the solvent was evaporated in
vacuo. The remaining brown oil was purified by flash chromatography
(silica gel; PE/EA, 50:1) to give pure 7 (2.84 g, 16%) as a colourless
liquid. 1H NMR (600 MHz, CDCl3): d=1.72 (d quint d, J3a,3b =13.0 Hz,
(average of J2a,3b, J2b,3b, J3b,4a and J3b,4b)=9.3 Hz, J3b,F =6.0 Hz, 1H; H3b),
1.92 (qm, (average of J2a,3a, J3a,3b and J3a,4a)ꢁ10 Hz, 1H; H3a), 2.23 (dt,
J2a,2b =13.5 Hz, J2a,3a =J2a,3b =9.7 Hz, 1H; H2a), 2.25–2.33 (m, 3H; H4a,
H4b, H5), 2.59 (dddt, J2a,2b =13.5 Hz, J2b,3b =9.0 Hz, J=1.1, 0.8 Hz, 1H;
H2b), 7.25 (m, 2H; o-H), 7.29 (tt, 1H; p-H), 7.35 ppm (m, 2H; m-H); the
assignment is based on an HSQC spectrum; 13C NMR (151 Hz, CDCl3):
of the unsubstituted 6,6-dibromobicycloACTHUNRGTNE[NUG 3.1.0]hexane and its 1-methyl de-
rivative;[8] MS (70 eV, EI): m/z (%): 318, 316, 314 (0.6, 1.2, 0.6) [M]+;
237, 235 (26, 27) [MꢀBr]+; 156 (45) [MꢀBr2]+, 155 (100), 153 (12), 141
(14), 129 (15), 128 (41), 127 (15), 117 (14), 115 (37), 102 (13), 91 (57), 78
(19), 77 (38), 76 (21), 64 (12), 63 (15), 51 (27), 39 (17); elemental analysis
calcd (%) for C12H12Br2: C 45.61, H 3.83; found: C 45.73, H 3.68.
1,6-Dibromo-2-phenylcyclohex-1-ene (6):
A solution of 5 (15.0 g,
47.5 mmol) in CH2Cl2 (150 mL) was kept at room temperature for about
11 h. Then the solvent was evaporated in vacuo at 208C. The remaining
crystals (14.6 g, 97%) were identified to be virtually pure 6.
d=24.9 (d, JC,F =10.3 Hz; C3), 28.0 (d, JC,F =1.2 Hz; C4), 35.1 (d, JC,F
2.6 Hz; C2), 39.0 (d, JC,F =11.2 Hz; C5), 46.7 (d, JC,F =10.9 Hz; C1), 90.8
(d, JC,F =321.8 Hz; C6), 127.1 (p-C), 128.186 (m-C), 128.195 (d, JC,F
=
=
In a previous experiment, a solution of 5 (1.00 g, 3.16 mmol) in chloro-
form (10 mL) was stirred at room temperature. As shown by monitoring
the progress of the rearrangement by NMR spectroscopy, the reaction
was almost complete after 3 d. The evaporation of the solvent in vacuo at
208C gave rise to an orange-yellow oil, which was dissolved in PE
(5 mL). Storage of this solution at ꢀ308C overnight afforded colourless
2.9 Hz; o-C), 140.0 ppm (d, JC,F =1.7 Hz; i-C); the assignment is based on
an HMBC spectrum; UV (hexane): lmax (loge)=260 (2.29), 254 (2.43),
248 (2.40), 2.33 (sh, 3.71), 218 (4.11), 208 nm (4.13); MS (70 eV, EI): m/z
(%): 236 (25), 234 (36) [MꢀHF]+, 175 (77), 155 (100), 154 (27), 153 (40),
152 (32), 129 (24), 115 (32), 109 (25), 91 (59), 77 (37), 76 (46), 51 (21);
HRMS (70 eV, EI): m/z: calcd for C12H12F [MꢀBr]+: 175.0918; found
175.0917.
1
crystals of 6 (900 mg, 90%). M.p. 558C; H NMR (400 MHz, CDCl3): d=
1.91 (m, 1H), 2.25 (m, 2H), 2.39 (m, 1H), 2.56 (m, 2H), 5.05 (m, 1H),
7.22 (m, o-H), 7.27–7.40 ppm (m, m-H, p-H); 13C NMR (63 MHz,
CDCl3): d=18.0 (C4), 33.8, 34.0 (C3, C5), 56.6 (C6), 120.3 (C1), 127.3 (o-
C), 127.6 (p-C), 128.1 (m-C), 142.0, 142.5 ppm (C2, i-C); as far as speci-
fied, the assignment is based on a C,H COSY spectrum; MS (70 eV, EI):
m/z (%): 318, 316, 314 (3, 5, 3) [M]+; 237 (83), 236 (22), 235 (84), 234
(32), 232 (22), 156 (54), 155 (100), 154 (21), 153 (32), 152 (34), 151 (36),
149 (35), 137 (40), 135 (40), 128 (27), 115 (24), 91 (56), 77 (42), 76 (40),
55 (24), 41 (57); elemental analysis calcd (%) for C12H12Br2: C 45.61, H
3.83; found: C 45.52, H 3.83.
Preparation of (6a,7b)- (endo-10) and (6a,7a)-2,7-diphenylbicyclo-
AHCTUNGERTG[NNUN 4.2.0]oct-1-ene (exo-10) from 5: Under nitrogen, a stirred solution of 5
(3.07 g, 9.71 mmol) in styrene (10 mL) was cooled to ꢀ258C and was
treated dropwise with methyllithium (12.0 mmol, 15 mL of 0.8m in DEE)
in a manner so that the temperature remained at ꢀ258C. After removal
of the cooling bath, the temperature was allowed to rise to 208C, and the
mixture was then cautiously hydrolysed (10 mL). The layers were sepa-
rated, the aqueous layer was extracted with DEE (2ꢃ10 mL) and the
combined organic phases were dried with MgSO4 and concentrated in
vacuo (15 mbar). From the remaining oil, the excess of styrene was dis-
tilled off in a kugelrohr (308C/0.04 mbar). The residue, a light yellow oil,
was purified by flash chromatography (silica gel, pentane) to give a 1:1
mixture of endo- and exo-10 (1.29 g, 51%) as a colourless oil. Attempts
to separate the diastereomers by chromatography on silica gel were un-
1-Bromo-2-phenyl- (8) and 1-bromo-6-phenylcyclohex-1-ene (9): A solu-
tion of 6 (11.1 g, 35.1 mmol) in DEE (20 mL) was added dropwise to a
stirred suspension of LiAlH4 (2.52 g, 66.4 mmol) in DEE (100 mL) at
208C. The mixture was then heated at reflux for 18 h and thereafter hy-
drolysed with a saturated solution of Na2SO4 (50 mL). The precipitate
11262
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11256 – 11265

1-Phenyl-1,2-cyclohexadiene
FULL PAPER
successful. However, dissolution of the mixture in a small amount of PE
and storage of the solution at ꢀ308C led to the crystallisation of pure
endo-10.
Compound endo-10: M.p. 758C; 1H NMR (600 MHz, CDCl3; values in
the mixture was kept at ꢀ308C. Stirring was continued for 1.5 h at
ꢀ258C, and the mixture was then cautiously hydrolysed (20 mL) at 08C.
The workup as in the preceding preparation afforded a brown oil (5.5 g),
which gave rise to a 1:1 mixture of endo- and exo-10 (2.20 g, 24%) after
flash chromatography.
square brackets refer to C6D6 as solvent): d=0.73 [0.92] (dddd, J5b,6
=
13.4 Hz, J5a,5b =12.2 Hz, J4a,5b =11.0 Hz, J4b,5b =3.0 Hz, 1H; H5b), 1.38
[1.37] (ꢁddt, J5a,5b =12.2 Hz, J4b,5a =6.4 Hz, J5a,6 =3.7 Hz, J4a,5a =3.1 Hz,
2,7,7-Triphenylbicyclo
lution of 1-phenylcyclopentene (2.00 g, 13.9 mmol) and tetrabromometh-
ane (5.00 g, 15.1 mmol) in anhydrous DEE (15 mL) was cooled to ꢀ608C
ACHTUNGTREN[NGNU 4.2.0]oct-1-ene (11): Under nitrogen, a stirred so-
1H; H5a), 1.63 [1.58] (ꢁtddd, J3b,4a =13.9 Hz, J4a,4b =13.1 Hz, J4a,5b
=
ACHTUNGTRENNUNG
11.0 Hz, J3a,4a =5.8 Hz, J4a,5a =3.1 Hz, 1H; H4a), 1.88 [1.77] (ꢁddtd,
and was treated dropwise with methyllithium (13.9 mmol, 10.7 mL 1.3m
in DEE) in a manner so that the temperature remained at ꢀ608C
(30 min). Thereafter, the mixture was stirred for 1 h at ꢀ608C and was
then allowed to warm to ꢀ308C. After 1,1-diphenylethene (7.00 g,
38.8 mmol) had been admixed at once, methyllithium (13.9 mmol,
10.7 mL 1.3m in DEE) was added dropwise, while the temperature was
kept at ꢀ308C. Stirring was continued for 2 h at ꢀ308C, and the mixture
was then cautiously hydrolysed (30 mL) at 08C. The layers were separat-
ed, and the aqueous layer was extracted with DEE (2ꢃ20 mL). The com-
bined organic layers were washed with a saturated aqueous solution of
NaCl (20 mL), dried with MgSO4 and concentrated in vacuo. Purification
of the residue by flash chromatography (silica gel; at first PE and then
PE/CH2Cl2, 2:1) furnished, in the order of elution, 1-phenylcyclopentene,
1,1-diphenylethene, a fraction (1.61 g) consisting mainly of 11 (about
1.2 g, 25%), and the almost pure trimer 17 (542 mg, 25%). The fraction
with 11 afforded crystals after standing for several months at ꢀ208C,
which were recrystallised from PE to give almost pure 11. M.p. 81–838C;
J4a,4b =13.1 Hz, J4b,5a =6.4 Hz, J3b,4b =4.2 Hz, J4b,5b =3.0 Hz, J3a,4b =1.7 Hz,
1H; H4b), 2.26 [2.24] (ꢁddtt, J3a,3b =16.8 Hz, J3a,4a =5.8 Hz, J3a,8a
=
3.1 Hz, J3a,8b =2.4 Hz, J3a,4b =1.7 Hz, J3a,6 =1.2 Hz, 1H; H3a), 2.55 [2.48]
(ꢁddqd, J3a,3b =16.8 Hz, J3b,4a =13.9 Hz, J3b,4b =4.2 Hz,
J3b,6 =3.2 Hz,
J3b,8a =3.1 Hz, J3b,8b =1.5 Hz, 1H; H3b), 2.96 [3.08] (dtt, J8a,8b =14.3 Hz,
J3a,8b =J7,8b =2.4 Hz, J3b,8b =J6,8b =1.5 Hz, 1H; H8b), 3.52 [3.38] (ꢁddtq,
J
5b,6 =13.4 Hz, J6,7 =9.1 Hz, J5a,6 =3.7 Hz, J3b,6 =3.2 Hz, J6,8a =1.6 Hz, J6,8b
=
=
1.5 Hz, J3a,6 =1.2 Hz, 1H; H6), 3.65 [3.49] (ddtd, J8a,8b =14.3 Hz, J7,8a
9.1 Hz, J3a,8a =J3b,8a =3.1 Hz, J6,8a =1.6 Hz Hz, 1H; H8a), 3.77 [3.62] (td,
J
6,7 =J7,8a =9.1 Hz, J7,8b =2.4 Hz, 1H; H7), 7.08 [7.36] (m, 2H; o-H of 7-
Ph), 7.16 [7.12] (tt, 1H; p-H of 7-Ph), 7.23 [7.26] (m, 2H; m-H of 7-Ph),
7.32–7.42 [6.98–7.10] ppm (m, 5H; 2-Ph); the assignment is based on H,H
COSY and NOESY spectra; 13C NMR (151 MHz, CDCl3): d=22.2 (C4),
23.1 (C5), 26.8 (C3), 38.8 (C8), 42.5 (C7), 47.5 (C6), 125.3 (C2), 125.8 (p-
C of 7-Ph), 125.9 (o-C of 2-Ph), 126.1 (p-C of 2-Ph), 127.7 (o-C of 7-Ph),
128.05 (m-C of 7-Ph), 128.15 (m-C of 2-Ph), 138.9 (C1), 139.5 (i-C of 2-
Ph), 141.8 ppm (i-C of 7-Ph); the assignment is based on HSQC and
HMBC spectra; elemental analysis calcd (%) for C20H20: C 92.26, H 7.74;
found: C 91.57, H 8.13.
Compound exo-10: 1H NMR (600 MHz, CDCl3): d=1.40 (m, 1H; H5b),
1.66 (m, 1H; H4a), 2.03–2.14 (m, 2H; H4b, H5a), 2.33 (m, 1H; H3a), 2.70
(m, 1H; H3b), 3.15–3.26 (m, 3H; H6, H7, H8a), 3.31 (m, 1H; H8b), 7.20–
7.40 ppm (m, 10H; 2-Ph, 7-Ph); the assignment is based on H,H COSY
and NOESY spectra; 13C NMR (151 MHz, CDCl3): d=22.8 (C4), 27.0
(C3), 28.3 (C5), 40.8 (C8), 46.4 (C7), 51.0 (C6), 124.6 (C2), 125.9 (o-C of
2-Ph), 126.0 and 126.1 (p-C of 2-Ph and 7-Ph), 126.5 (o-C of 7-Ph), 128.1
and 128.3 (m-C of 2-Ph and 7-Ph), 136.3 (C1), 139.7 (i-C of 2-Ph),
144.5 ppm (i-C of 7-Ph); the assignment is based on HSQC and HMBC
spectra.
1H NMR (600 MHz, CDCl3): d=0.72 (tdd, J5b,6 =J5a,5b =13.0 Hz, J4a,5b
10.5 Hz, J4b,5b =3.0 Hz, 1H; H5b), 1.66–1.75 (m, 2H), 1.90 (m, 1H), 2.26
=
(brd, J3a,3b =16.2 Hz, 1H; H3a), 2.51 (m, 1H; H3b), 3.63 (dtd, J8a,8b
=
14.7 Hz, J=3.1, 1.7 Hz, 1H), 3.72 (dm, J8a,8b =14.7 Hz, 1H; H8a, H8b),
3.86 (m, 1H; H6), 6.95 (m, 2H; o-H of 7b-Ph), 7.10 (tt, 1H; p-H of 7b-
Ph), 7.14 (m, 2H; m-H of 7b-Ph), 7.21–7.27 (m, 4H), 7.32–7.40 ppm (m,
6H); the assignment is based on the comparison with the spectra of
endo- and exo-10; 13C NMR (151 MHz, CDCl3): d=22.3 (t, JC,H
=
127.8 Hz; C4), 24.5 (t, JC,H =128.9 Hz; C5), 27.0 (t, JC,H =125.8 Hz; C3),
46.3 (t, JC,H =137.2 Hz; C8), 52.9 (d, JC,H =135.9 Hz; C6), 54.5 (s; C7),
125.62 (s; C2), 125.66, 125.8, 126.2 (3dt; 3p-C), 125.9, 127.0, 127.5 (3dt;
3o-C), 127.8, 128.17, 128.20 (3dd; 3m-C), 135.1, 139.5, 143.7, 150.4 ppm
(4s; C1, 3i-C); as far as specified, the assignment is based on the compar-
ison with the spectra of endo- and exo-10; IR (KBr): n˜ =3070, 3045, 3020,
2915, 2855, 2835, 1595, 1490, 1440, 1030, 770, 755, 695 cmꢀ1; elemental
analysis calcd (%) for C26H24: C 92.81, H 7.19; found: C 93.32, H 7.53.
Mixture of endo- and exo-10 in the ratio of 1:1: UV (hexane): lmax
(loge)=275 (sh, 3.97), 258 (4.24), 222 (sh, 4.11), 216 (sh, 4.24), 208 (sh,
4.30), 202 nm (sh, 4.34); the UV spectra of the isolated diastereomers,
measured with the pure enantiomers,[6] are virtually the same; MS
(70 eV, EI): m/z (%): 260 (16) [M]+, 183 (28), 182 (100), 167 (26), 154
(19), 142 (16), 141 (39), 129 (17), 128 (18), 115 (35), 105 (36), 104 (32), 91
(55), 77 (24).
(4aa,4bb,9ab)-3,4,4a,4b,9,9a-Hexahydro-1-phenyl-2H-benzo-
ACHUTNGRENU[NG 3,4]cyclobutaACHTUNTGREN[NUGN 1,2-a]indene (12): Under nitrogen, a stirred solution of 5
(1.97 g, 6.23 mmol) in indene (10 mL) was cooled to ꢀ38C and was treat-
ed dropwise with methyllithium (9.8 mmol, 10 mL of 0.98m in DEE) in a
manner so that the temperature remained at ꢀ38C. After removal of the
cooling bath, the temperature was allowed to rise to 208C, and the mix-
ture was then cautiously hydrolysed (20 mL). The layers were separated,
the aqueous layer was extracted with CH2Cl2 (2ꢃ5 mL), and the com-
bined organic phases were dried with MgSO4 and concentrated in vacuo
(15 mbar). From the remaining oil, the excess of indene was distilled off
in a kugelrohr (458C/0.03 mbar). The residue, a light brown oil, was puri-
fied by flash chromatography (silica gel; PE/EA, 100:1) to give almost
pure 12 (124 mg, 7%) as a colourless oil. Another fraction (433 mg,
44%) consisted mainly of the trimer 17 of 2 and products that may be
Thermolysis of endo-10: A solution of endo-10 (74 mg) in C6D6 (1.2 mL)
was sealed in an NMR spectroscopy tube and heated at 150–1608C.
After 6 h, about one half of the amount of endo-10 had rearranged to
exo-10. After 26 h, the ratio of endo-/exo-10 was about 1:10. Prolonged
heating led to the decomposition of the sample.
Transformation of endo- and exo-10 at room temperature: After standing
at room temperature for 2 months, endo- and exo-10 (1 g of the 1:1 mix-
ture), dissolved in pentane (10 mL), had completely and cleanly convert-
ed into a product that could be a 1:1 mixture of two isomers. The struc-
ture was not elucidated. 13C NMR (63 MHz, CDCl3): d=18.2, 18.7, 21.8,
27.9, 28.5, 30.0, 31.8, 34.4 (8CH2), 36.3, 40.9, 43.8, 47.3 (4CH), 62.1
(double intensity), 65.5, 67.8 (4Cquart.), 126.0, 126.20, 127.2, 127.3 (4p-C),
126.23, 126.25, 126.5, 127.9, 128.16 (double intensity), 128.21, 128.4 (4o-
C, 4m-C), 139.1, 139.5, 141.3, 143.9 (4i-C).
other oligomers of 2. 1H NMR (600 MHz, C6D6): d=1.23 (dtd, J3a,4b
=
13.5 Hz, J4a,4b =11.3 Hz,
J
4b,4a =10.7 Hz, J3b,4b =2.4 Hz, 1H; H4b), 1.33
(tddd, J3a,4b =13.5 Hz, J3a,3b =13.3 Hz, J2b,3a =10.8 Hz, J2a,3a =5.9 Hz,
J3a,4a =2.6 Hz, 1H; H3a), 1.78 (ddddd, J3a,3b =13.3 Hz, J2b,3b =6.5 Hz,
J3b,4a =3.4 Hz, J3b,4b =2.4 Hz, J2a,3b =1.4 Hz, 1H; H3b), 1.82 (dddd, J4a,4b
=
One-pot preparation of endo- and exo-10 from 1-phenylcyclopentene:
11.3 Hz, J4a,4a =6.2 Hz, J3b4a =3.4 Hz, J3a,4a =2.6 Hz, 1H; H4a), 2.17 (dddt,
J2a,2b =16.9 Hz, J2a,3a =5.9 Hz, 2a,4a =2.5 Hz, J2a,3b =J2a,9a =1.4 Hz, 1H;
Under nitrogen,
a
stirred solution of 1-phenylcyclopentene (5.00 g,
J
34.7 mmol) and tetrabromomethane (12.6 g, 38.0 mmol) in anhydrous
DEE (50 mL) was cooled to ꢀ608C and was treated dropwise with meth-
yllithium (38.0 mmol, 29.2 mL of 1.3m in DEE) in a manner so that the
temperature remained at ꢀ608C. Thereafter, the mixture was stirred for
1 h at ꢀ608C and was then allowed to warm to ꢀ308C. After styrene
(20.8 g, 200 mmol) had been admixed at once, methyllithium (38.0 mmol,
29.2 mL of 1.3m in DEE) was added dropwise, while the temperature of
H2a), 2.53 (ddddd, J2a,2b =16.9 Hz, J2b,3a =10.8 Hz, J2b,3b =6.5 Hz, J2b,4a
=
3.6 Hz, J2b,9a =1.4 Hz, 1H; H2b), 2.80 (ꢁdtq, J4b,4a =10.7 Hz, J4a,4a =6.2 Hz,
J
4a,4b =5.7 Hz, J2b,4a =3.6 Hz, J4a,9a =2.9 Hz, J2a,4a =2.5 Hz, 1H; H4a), 3.20
(dd, J9a,9b =16.7 Hz, J9b,9a =10.7 Hz, 1H; H9b), 3.31 (dd, J9a,9b =16.7 Hz,
J
9a,9a =5.3 Hz, 1H; H9a), 3.41 (ꢁt,
J4b,9a =6.8 Hz, J4a,4b =5.7 Hz, 1H;
H4b), 3.67 (ddddt, J9b,9a =10.7 Hz, J4b,9a =6.8 Hz, J9a,9a =5.3 Hz, J4a,9a
=
2.9 Hz, J2a,9a =J2b,9a =1.4 Hz, 1H; H9a), 7.06–7.17 (m, 4H; H5, H6, H7,
Chem. Eur. J. 2009, 15, 11256 – 11265
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11263

M. Christl, D. Stalke et al.
H8), 7.15 (tt, 1H; p-H), 7.30 (m, 2H; m-H), 7.39 ppm (m, 2H; o-H); the
assignment is based on H,H COSY and NOESY spectra; 13C NMR
(151 MHz, C6D6): d=23.0 (C3), 27.0 (C2), 27.8 (C4), 38.3 (C9), 49.4
(C9a), 51.8 (C4a), 53.1 (C4b), 123.7 (C5), 125.4 (C8), 125.5 (C1), 126.4
(o-C of Ph), 126.6 (p-C of Ph), 126.77, 126.83 (C6, C7), 128.5 (m-C of
Ph), 140.1 (i-C of Ph), 141.9 (C9b), 145.6 (C8a), 147.4 ppm (C4c); the as-
signment is based on HSQC and HMBC spectra; UV (hexane): lmax
(loge)=275 (3.89), 266 (sh, 4.09), 260 (4.15), 255 (4.15), 250 (4.07), 233
(sh, 4.56), 218 (4.95), 209 nm (sh, 4.91); MS (70 eV, EI): m/z (%): 272
(100) [M]+, 257 (22), 244 (20), 229 (64), 228 (22), 216 (23), 215 (58), 181
(69), 173 (23), 168 (52), 167 (33), 165 (22), 157 (32), 155 (22), 153 (26),
144 (26), 142 (26), 141 (37), 129 (32), 128 (49), 116 (28), 115 (72), 91 (56),
77 (26); HRMS (70 eV, EI): m/z: calcd for C21H20 [M]+: 272.1560; found
272.1557.
tent of 3 was about 70 mg (15%). In view of the previous characterisa-
tion of 3,[4] we did not carry out its isolation.
1,2,3,4,6,7,8,8a-Octahydro-8a-phenyltriphenylene (16): KOtBu (325 mg,
2.90 mmol) was added to a solution of 8 (596 mg, 2.51 mmol) in anhy-
drous THF (10 mL) at 208C. The mixture was vigorously stirred at 608C
for 7 h and then cooled to 208C. Water (30 mL) and CH2Cl2 (30 mL)
were admixed and the layers were separated. The aqueous layer was ex-
tracted with CH2Cl2 (30 mL). The combined organic layers were dried
with Na2SO4 and concentrated in vacuo to give an oily residue. Purifica-
tion by chromatography (silica gel, pentane) afforded pure 16 (122 mg,
31%) as colourless crystals. M.p. 141–1428C; 1H NMR (600 MHz, C6D6):
d=1.25 (m, 1H; H2), 1.36–1.46 (m, 3H; 2H7, H3), 1.49 (m, 1H; H2),
1.62 (m, 1H; H3), 1.85 (brdt, 2J=16.9 Hz, average of 23J=4.5 Hz, 1H;
H1), 2.04 (apparent dt, line distances=8.1, 4.4 Hz, 2H; 2H6), 2.12 (td,
(average of 2J and 3J)=12.5 Hz, 3J=3.4 Hz, 1H; H8), 2.12–2.24 (m, 2H;
H1, H4), 2.33 (ddd, 2J=12.5 Hz, 3J=4.5, 2.8 Hz, 1H; H8), 2.53 (brdt,
2J=16.9 Hz, average of 23J=5.0 Hz, 1H; H4), 5.99 (t, line distance
Preparation of (1a,4a,4aa)-1,4-dimethyl-4a-phenyl-1,4,4a,5,6,7-hexahy-
dro-1,4-epoxynaphthalene (14)
From 7: Under nitrogen, a stirred solution of 7 (1.10 g, 4.31 mmol) in 2,5-
dimethylfuran (6 mL) was cooled to ꢀ308C and was treated dropwise
with methyllithium (8.0 mmol, 10 mL of 0.8m in DEE) in a manner so
that the temperature remained at ꢀ308C (20 min). After removal of the
cooling bath, the temperature was allowed to rise to 08C, and the mix-
ture was then cautiously hydrolysed (10 mL). The layers were separated,
the aqueous layer was extracted with DEE (3ꢃ10 mL), and the com-
bined organic phases were dried with MgSO4 and concentrated in vacuo.
The residue was purified by flash chromatography (silica gel; PE/EA,
20:1) to give pure 14 (237 mg, 22%) as a colourless oil. 1H NMR
(600 MHz, C6D6): d=0.70 (ꢁtd, J5a,5b =J5b,6a =12.2 Hz, J5b,6b =6.0 Hz, 1H;
H5b), 1.20 (s, 3H; 4-Me), 1.35–1.46 (m, 2H; H6a, H6b), 1.63 (s, 3H; 1-
Me), 1.72–1.83 (m, 2H; H7a, H7b), 2.24 (ddd, J5a,5b =12.2 Hz, J5a,6a and
J5a,6b =4.2, 3.0 Hz, 1H; H5a), 5.39 (dd, J7a,8 and J7b,8 =4.6, 3.1 Hz, 1H;
H8), 5.81 (d, J2,3 =5.3 Hz, 1H; H3), 5.94 (d, J2,3 =5.3 Hz, 1H; H2), 7.10
(tt, 1H; p-H), 7.21 (m, 2H; m-H), 7.30 ppm (m, 2H; o-H); the assign-
ment is based on NOESY and HMQC spectra; 13C NMR (151 MHz,
D6D6): d=15.0 (1-Me), 16.5 (4-Me), 19.9 (C6), 23.6 (C7), 32.3 (C5), 54.1
(C4a), 86.6 (C1), 90.8 (C4), 117.6 (C8), 126.0 (p-C), 127.6 (broad, m-C),
130.0 (very broad, o-C), 136.1 (C3), 140.2 (C2), 142.9 (i-C), 151.0 ppm
(C8a); the assignment is based on HMQC and HMBC spectra; UV
(methanol): lmax (loge)=270 (2.18), 266 (2.31), 263 (sh, 2.31), 259 (2.41),
253 (2.39), 248 (2.36), 215 nm (3.88); MS (70 eV, EI): m/z (%): 252 (7)
[M]+, 210 (21), 209 (100), 181 (27), 179 (12), 178 (11), 167 (18), 165 (15),
129 (11), 117 (10), 115 (18), 105 (18), 91 (30), 77 (13), 43 (37); HRMS
(70 eV, EI): m/z: calcd for C18H20O [M]+: 252.1509; found 252.1507.
4.2 Hz, 1H; H5), 6.94 (p-H of Ph), 7.08 (m-H of Ph), 7.11 (dd, J11,12
7.7 Hz, J10,12 =1.5 Hz, 1H; H12), 7.16 (td, (average of J10,11 and J11,12)=
7.5 Hz, J9,11 =1.3 Hz, 1H; H11), 7.22 (td, (average of J9,10 and J10,11)=
=
7.5 Hz, J10,12 =1.5 Hz, 1H; H10), 7.28 (o-H of Ph), 7.59 ppm (dd, J9,10
=
7.7 Hz, J9,11 =1.3 Hz, 1H; H9); the assignment is based on a H,H COSY
spectrum and NOE measurements; 13C NMR (151 MHz, CDCl3): d=
18.3, 22.5, 22.9 (C2, C3, C7), 25.5, 25.9 (C1, C6), 27.0 (C4), 36.4 (C8),
45.9 (C8a), 122.4 (C11), 125.1 (C9), 125.4 (p-C of Ph), 125.5 (C5), 125.7
(C10), 126.3 (C12), 127.2 (m-C of Ph), 127.6 (o-C of Ph), 128.2, 131.8,
135.7, 139.0, 142.2, 147.3 ppm (C4a, C4b, C8b, C12a, C12b, i-C of Ph); as
far as specified, the assignment is based on a C,H COSY spectrum; UV
(CH3CN): lmax (loge)=304 nm (3.98); MS (70 eV, EI): m/z (%): 312
(100) [M]+, 284 (19), 269 (19), 241 (17), 236 (21), 235 (85), 234 (17), 208
(19), 195 (19), 193 (26), 191 (20), 179 (24), 178 (29), 167 (22), 165 (26),
157 (45), 141 (15), 129 (18), 128 (16), 115 (19), 91 (50); elemental analysis
calcd (%) for C24H24: C 92.26, H 7.74; found: C 92.03, H 7.71.
(4aR*,8bR*,1’R*)-2,3,4,4a,8b,9,10,11-Octahydro-8b-phenyl-4a-(3-phenyl-
cyclohex-2-en-1-yl)triphenylene (17): Under nitrogen, a stirred solution
of 5 (2.00 g, 6.33 mmol) in anhydrous tert-butyl methyl ether (20 mL),
cooled to ꢀ308C, was treated dropwise with methyllithium (6.3 mmol,
4.2 mL 1.5m in DEE) within 15 min. The mixture was then allowed to
warm to 08C and was cautiously hydrolysed. The layers were separated,
the organic layer was extracted with water (20 mL), dried with MgSO4
and concentrated in vacuo. The residue was purified by chromatography
(silica gel; PE/EA, 13:1) to give a colourless material (390 mg), which
was dissolved in pentane at room temperature. This solution was kept at
ꢀ308C. After 14 d, a solid (m.p. 121–1228C, 180 mg, 18%) had precipi-
tated, which turned out to be almost pure 17. 1H NMR (600 MHz,
CDCl3): d=1.41–1.53 (m, 2H), 1.56 (m, 1H), 1.72 (m, 1H), 1.75–1.88 (m,
3H), 1.89–1.95 (m, 2H), 1.98 (td, J=12.7, 4.7 Hz, 1H), 2.02–2.19 (m,
4H), 2.20–2.29 (m, 2H), 2.35 (brd, J=16.5 Hz, 1H), 2.53 (m, 1H, H1’),
2.69 (ꢁdt, J=12.5, 3.9 Hz, 1H), 5.70 (t, J=4.0 Hz, 1H), 5.80 (t, J=
4.0 Hz, 1H), 6.19 (brs, 1H), 7.06 (1H; p-H of Ph), 7.11–7.20 (m, 5H),
7.22–7.28 (m, 4H), 7.31 (m, 1H), 7.36 (2H; o-H of Ph), 7.38 ppm (m,
1H); 13C NMR (151 MHz, CDCl3): d=18.8, 19.8, 23.6, 24.3, 25.0, 25.2,
27.5, 31.3, 36.1 (C2, C3, C4, C9, C10, C11, C4’, C5’, C6’), 44.9. 47.2 (C4a,
C8b), 48.3 (C1’), 124.0, 124.6, 125.2, 125.3, 125.9, 126.4, 126.5, 127.6, 129.2
(C1, C5, C6, C7, C8, C12, C2’, 2p-C of Ph), 125.1, 127.5, 127.7, 128.1
(2m-C, 2o-C of Ph), 137.0, 142.1, 142.9, 143.4 (double intensity), 143.7,
148.3 (C4b, C8a, C12a, C12b, C3’, 2i-C of Ph); as far as specified, the as-
signment is based on a C,H COSY spectrum; MS (70 eV, EI): m/z (%):
468 (1) [M]+, 312 (34), 311 (100) [MꢀC12H13]+, 233 (25), 193 (10), 157
From 8: A stirred mixture of 8 (197 mg, 0.83 mmol), 2,5-dimethylfuran
(2.26 g, 23.5 mmol), THF (2.5 mL) and KOtBu (110 mg, 0.98 mmol) was
heated at 708C for 5 h. It was hydrolysed (5 mL) after it had adopted
room temperature and the layers were separated. The aqueous layer was
extracted with DEE (2ꢃ3 mL). The combined organic layers were dried
with MgSO4 and concentrated in vacuo to give a yellow oil (46 mg), the
1H NMR spectrum of which proved it to be essentially a 1.0:1.5 mixture
of 14 (9%) and 8.
From 9: A stirred mixture of 9 (228 mg, 0.96 mmol), 2,5-dimethylfuran
(4.50 g, 46.8 mmol), THF (2.5 mL), dimethyl sulfoxide (2.5 mL) and
KOtBu (124 mg, 1.10 mmol) was heated at 708C for 5 h. It was hydro-
lysed (5 mL) after it had adopted room temperature and the layers were
separated. The aqueous layer was extracted with DEE (2ꢃ3 mL). The
combined organic layers were dried with MgSO4 and concentrated in
vacuo to give a yellow oil (58 mg), the 1H NMR spectrum of which
proved that 14 was the major component.
(1a,4a,4aa)-4a-Phenyl-1,4,4a,5,6,7-hexahydro-1,4-epoxynaphthalene (3):
A stirred mixture of 8 and 9 (1.7:1.0, 500 mg, 2.11 mmol), furan (2.81 g,
41.3 mmol), dimethyl sulfoxide (3 mL) and KOtBu (272 mg, 2.42 mmol)
was heated at 608C for 5 h. It was hydrolysed (4 mL) after it had adopted
room temperature and the layers were separated. The aqueous layer was
extracted with DEE (2ꢃ3 mL). The combined organic layers were dried
with MgSO4 and concentrated in vacuo to give a yellow oil (227 mg), the
1H NMR spectrum for which, in comparison with the published data,[4]
indicated the presence of 3 as component of a mixture with dimethyl sulf-
oxide and impurities. As estimated on the basis of the integrals, the con-
(44) [C12H13]+, 129 (11), 91 (30); elemental analysis calcd (%) for C36H36
C 92.26, H 7.74; found: C 92.08, H 7.82.
:
The 1H NMR spectrum of the crude product (before submission to chro-
matography) showed that 50% of the aromatic proton signals originated
from 17, as determined by comparison of the integral with that of an ole-
finic proton signal of 17. Crystals for the X-ray structure analysis were
obtained by dissolving the almost pure 17 (see above) in hexane heated
to reflux, which took several hours, and allowing only a very slow cooling
of the solution to room temperature. Yield: colourless prisms, m.p. 159–
11264
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11256 – 11265

1-Phenyl-1,2-cyclohexadiene
FULL PAPER
1608C. As evidenced by 1H NMR signals at d=5.81, 5.83 and 6.28 ppm,
for which the shapes are the same as those of the signals of 17 at d=
5.70, 5.80 and 6.19 ppm, even this crystalline solid held an impurity of
about 5%, which may well be a diastereomer of 17.
[5] M. Ceylan, Y. Budak, J. Chem. Res. Synop. 2002, 416–419; M.
Ceylan, Y. Budak, J. Chem. Res. Miniprint 2002, 0937–0946.
[6] M. Christl, H. Fischer, M. Arnone, B. Engels, Chem. Eur. J. 2009,
DOI: 10.1002/chem.200900718, following paper in this issue.
[7] J. Sonnenberg, S. Winstein, J. Org. Chem. 1962, 27, 748–751.
[8] M. Christl, M. Schreck, Chem. Ber. 1987, 120, 915–920.
[9] M. Rudolph, Dissertation, Universitꢁt Wꢀrzburg (Germany), 1996.
[10] a) M. Christl, M. Braun, Chem. Ber. 1989, 122, 1939–1946; b) M.
Christl, M. Braun, G. Mꢀller, Angew. Chem. 1992, 104, 471–473;
Angew. Chem. Int. Ed. Engl. 1992, 31, 473–476.
[11] S. W. Benson, Thermochemical Kinetics, 2nd ed., Wiley, New York,
1976, p. 309.
[12] A. T. Bottini, L. L. Hilton, J. Plott, Tetrahedron 1975, 31, 1997–2001.
[13] M. Christl, M. Braun, H. Fischer, S. Groetsch, G. Mꢀller, D. Leusser,
S. Deuerlein, D. Stalke, M. Arnone, B. Engels, Eur. J. Org. Chem.
2006, 5045–5058.
X-ray crystal-structure determination of 17: The data were collected from
a shock-cooled crystal using a Bruker Smart APEX II diffractometer
with a D8 goniometer (graphite-monochromated MoKa radiation, l=
71.073 pm) equipped with
a low-temperature device operating at
100(2) K.[26] An empirical absorption correction with the program
SADABS 2004/1 was employed.[27] The structure was solved by direct
methods (SHELXS-97[28]) and refined by full-matrix least squares meth-
ods against F2 (SHELXL-97[29]).
R
values: R1 =SjjFo jꢀjFc jj/SjFo j,
wR2 =[Sw(F2oꢀFc2)2/Sw(F2o)2]0.5
,
w=[s2(F2o)+(g1P)2 g2P]ꢀ1
,
P=
1= [max(F2,0)+2F2]; C36H36; Mr =468.65; monoclinic; space group P21/n;
3
o
c
a=1665.41(8), b=971.16(5), c=1726.45(8) pm; b=113.6120(10)8; V=
2.5585(2) nm3; Z=4; 1calcd =1.217 Mgmꢀ3; 28736 reflections measured,
6004 unique; R1 [I>2s(I)]=0.0465; wR2 (all data)=0.1516; g1 =0.0959,
g2 =0.3554 for 335 parameters and 12 restraints; GOF=1.124; residual
[14] M. Nendel L. M. Tolbert, L. E. Herring, Md. N. Islam, K. N. Houk,
J. Org. Chem. 1999, 64, 976–983.
density (max/min)=0.468/ꢀ0.389 eꢄꢀ3
.
[15] a) H. Hopf in The Chemistry of Allenes, Vol. 2 (Ed.: S. R. Landor),
Academic Press, London, 1982, pp. 525–562; b) M. Murakami, T.
Matsuda in Modern Allene Chemistry, Vol. 2 (Eds.: N. Krause,
A. S. K. Hashmi), Wiley-VCH, Weinheim, 2004, pp. 727–815.
[16] M. Christl, S. Groetsch, K. Gꢀnther, Angew. Chem. 2000, 112, 3395–
3397; Angew. Chem. Int. Ed. 2000, 39, 3261–3263.
[17] E. V. Banide, Y. Ortin, C. M. Seward, L. E. Harrington, H. Mꢀller-
Bunz, M. J. McGlinchey, Chem. Eur. J. 2006, 12, 3275–3286.
[18] C. Moureu, C. Dufraisse, P. M. Dean, C. R. Acad. Sci. 1926, 182,
1440–1443.
[19] J. Rigaudy, P. Capdevielle, Tetrahedron 1977, 33, 767–773.
[20] P. Capdevielle, J. Rigaudy, Tetrahedron 1979, 35, 2093–2100.
[21] M. Christl, M. Rudolph, E.-M. Peters, K. Peters, H. G. von Schner-
ing, Angew. Chem. 1995, 107, 2931–2934; Angew. Chem. Int. Ed.
Engl. 1995, 34, 2730–2732.
[22] a) M.-L. Ahrens, M. Eigen, W. Kruse, G. Maass, Ber. Bunsen-Ges.
1970, 74, 380–385; b) F. Strohbusch, Chem. Unserer Zeit 1982, 16,
103–110.
All non-hydrogen atoms were refined with anisotropic displacement pa-
rameters. Hydrogen atoms were assigned ideal positions and refined iso-
tropically by using a riding model with Uiso constrained to 1.2 times the
Ueq of the parent atom. The disorder of the methylene group of position
17 was refined using similarity restraints. Site occupation factors of 0.845
and 0.155 have been obtained.
CCDC-724091 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgements
The present work was supported by the Deutsche Forschungsgemein-
schaft and the Fonds der Chemischen Industrie. We are grateful to
CHEMETALL GmbH for gifts of methyllithium and to the group of
Prof. Dr. Christoph Lambert for help with the measurement of the UV
spectra.
[23] K. J. Daoust, S. M. Hernandez, K. M. Konrad, I. D. Mackie, J. Win-
stanley, Jr., R. P. Johnson, J. Org. Chem. 2006, 71, 5708–5714.
[24] Y. Amiel, A. Lçffler, D. Ginsburg, J. Am. Chem. Soc. 1954, 76,
3625–3628.
[25] M. Schlosser, G. Heinz, Chem. Ber. 1971, 104, 1934–1941.
[26] a) T. Kottke, D. Stalke, J. Appl. Crystallogr. 1993, 26, 615–619; b) T.
Kottke, R. J. Lagow, D. Stalke, J. Appl. Crystallogr. 1996, 29, 465–
468; c) D. Stalke, Chem. Soc. Rev. 1998, 27, 171–178.
[27] G. M. Sheldrick, SADABS 2004/1, Program for Area Detector Ab-
sorption Correction, Universitꢁt Gçttingen, Gçttingen, 2004.
[28] G. M. Sheldrick, Acta Crystallogr. Sect. A 1990, 46, 467–473.
[29] G. M. Sheldrick, SHELXL-97, Program for Structure Refinement,
Universitꢁt Gçttingen, Gçttingen, 1997.
[1] M. Christl in Modern Allene Chemistry, Vol. 1 (Eds.: N. Krause,
A. S. K. Hashmi), Wiley-VCH, Weinheim, 2004, pp. 243–357.
[2] a) T. Khasanova, R. S. Sheridan, J. Am. Chem. Soc. 2000, 122, 8585–
8586; b) A. F. Nikitina, R. S. Sheridan, Org. Lett. 2005, 7, 4467–
4470.
[3] R. L. Elliott, N. H. Nicholson, F. E. Peaker, A. K. Takle, C. M. Ri-
chardson, J. W. Tyler, J. White, M. J. Pearson, D. S. Eggleston, R. C.
Haltiwanger, J. Org. Chem. 1997, 62, 4998–5016.
[4] L. M. Tolbert, M. N. Islam, R. P. Johnson, P. M. Loiselle, W. C.
Shakespeare, J. Am. Chem. Soc. 1990, 112, 6416–6417.
Received: March 19, 2009
Published online: September 11, 2009
Chem. Eur. J. 2009, 15, 11256 – 11265
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11265