3-Methylenetricyclo[3.1.0.02,6]hexane, a tricyclic isomer of toluene: Synthesis and addition onto tetracyanoethylene

Article abstract of DOI:10.1002/1099-0690(200204)2002:8<1374::AID-EJOC1374>3.0.CO;2-S

A multistep ring-contraction route starting from tricyclo[4.1.0.0^2,7]heptan-3-one (1) resulted in a mixture of the title compound 7 and the homofulvene 8 in a ratio of 1:2.5. In a second synthesis, a 3.5:1 mixture of 7 and three additional hydrocarbons was obtained from 4-(phenylsulfanyl)tricyclo [3.1.0.0^2,6]hexan-3-one (9) in a two-step sequence. On being heated at 175 °C in solution, 7 rearranged to toluene. Treatment of 7 with tetracyanoethylene gave rise to a 1:5 mixture of the [2+2] cycloadduct 16 and the cyclopropadicyclopentene derivative 17. Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002.

Full text of DOI:10.1002/1099-0690(200204)2002:8<1374::AID-EJOC1374>3.0.CO;2-S

FULL PAPER
3-Methylenetricyclo[3.1.0.02,6]hexane, a Tricyclic Isomer of Toluene:
Synthesis and Addition onto Tetracyanoethylene
Dieter Hasselmann,*[a] Klaus Loosen,[a] Thomas Fischer,[b] Ulrike Kunz,[b] and
Manfred Christl*[b]
Keywords: Cycloadditions / Rearrangements / Reductions / Ring-contraction / Strained molecules
A
multistep ring-contraction route starting from tricy- ing heated at 175 °C in solution, 7 rearranged to toluene.
clo[4.1.0.02,7]heptan-3-one (1) resulted in a mixture of the
title compound 7 and the homofulvene 8 in a ratio of 1:2.5. In
a second synthesis, a 3.5:1 mixture of 7 and three additional
hydrocarbons was obtained from 4-(phenylsulfanyl)tricy-
Treatment of 7 with tetracyanoethylene gave rise to a 1:5
mixture of the [2+2] cycloadduct 16 and the cyclopropadicy-
clopentene derivative 17.
(
Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany,
clo[3.1.0.02,6]hexan-3-one (9) in a two-step sequence. On be- 2002)
Introduction
addition onto TCNE was published some time ago by
Hogeveen et al.[6]
Detailed knowledge of thermal isomerisations of strained
hydrocarbons provides information about the strength of
chemical bonds, the magnitude of strain energies, and the
rules governing bond reorganisations.[1] Our previous stud-
ies have been focused on strained methylenebicycloalkenes[2]
and endo,endo-bridged bicyclo[1.1.0]butanes.[3] The pres-
ence of both an olefinic methylene group and a bicyclo-
Results and Discussion
1. Synthesis of 7 by Ring-Contraction of
Tricyclo[4.1.0.02,7]heptan-3-one (1)
Synthetic access to highly strained compounds has often
[1.1.0]butane system within a molecule results in a special been achieved by ring-contraction of higher homologues,
situation, particularly if these subunits are in conjugation which are usually much less strained and much more easily
with each other. One compound of that kind has already available.[7] In particular, the light-induced[8] Wolff
been investigated: 3-methylenetricyclo[4.1.0.02,7]heptane,[4] rearrangement[9Ϫ12] of cyclic α-diazo ketones[13] to form ke-
the higher homologue of the title compound 7. The struc- tene intermediates,[14] which can instantaneously be trapped
ture of the major product, as well as the activation para- by alcohols to give the pertinent contracted carboxylic es-
meters,[4] fit smoothly with the results of the thermal re- ters, has proved to be of special value in this respect.[15]
arrangement of tricyclo[4.1.0.02,7]heptane and a series of its Thus, in our hands, bicyclo[3.2.0]hept-6-en-2-one could suc-
derivatives, which are believed to occur in two consecutive cessfully be transformed into methyl bicyclo[2.2.0]hex-5-
concerted steps.[3b] In the case of 7, such a mechanism ene-2-carboxylate.[2d] As a feasible synthetic route to 7, we
should be impeded if not impossible, due to the reduction therefore envisaged an analogous multistep sequence start-
of the ring size by one methylene group. We therefore ing from the known tricycloheptanone 1 (Scheme 1).
undertook to investigate 7, and report here on the synthesis
The tricyclic ketone 1, prepared by photolytic oxadi-π-
of 7 and its reaction with tetracyanoethylene (TCNE), while methane rearrangement from bicyclo[3.2.0]hept-6-en-2-
the thermolysis of 7 in the gas phase is the subject of a one,[16] was formylated in diethyl ether, with the use of a
forthcoming article.[5] The preparation of a pentamethyl large excess of ethyl formate, to give 2. Without purifica-
derivative of 7 from hexamethyl Dewar benzene and its tion, the activated ketone 2 was subjected to diazo group
transfer,[17] using p-toluenesulfonyl azide[18] as the reagent.
[a]
Formation of 3, obtained in about 43Ϫ62% yield relative to
starting ketone 1, was indicated by strong absorption bands
in the infrared spectrum at 2085 and 1635 cmϪ1, typical of
an α-diazo ketone.[19] By irradiating 3 in methanol, through
a Pyrex filter, we obtained the ring-contracted ester 4. Re-
duction of this with LiAlH4 provided the alcohol 5 in 87%
yield, and this was converted into its tosylate 6 (95%). De-
Fakultät für Chemie, Organische Chemie II, Ruhr-Universität
Bochum
44780 Bochum, Germany
Fax: (internat.) ϩ 49-(0)234/32-14109
E-mail: hasselmann@orch.ruhr-uni-bochum.de
Institut für Organische Chemie, Universität Würzburg
Am Hubland, 97074 Würzburg, Germany
Fax: (internat.) ϩ 49-(0)931/888-4606
E-mail: Christl@chemie.uni-wuerzburg.de
[b]
1374
WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 1434Ϫ193X/02/0408Ϫ1374 $ 20.00ϩ.50/0 Eur. J. Org. Chem. 2002, 1374Ϫ1379

3-Methylenetricyclo[3.1.0.02,6]hexane, a Tricyclic Isomer of Toluene
FULL PAPER
Scheme 2. a) Ph3PMeϩBrϪ, KOtBu, Et2O, 20 °C; b) Na, liquid
NH3, EtOH, Et2O, Ϫ70 to Ϫ60 °C
Scheme 1. a) HCO2Et, NaH/MeOH, Et2O, 20 °C; b) TsN3, HNEt2;
c) hν, MeOH; d) LiAlH4, Et2O; e) TsCl, pyridine; f) KOtBu,
DMSO
Through a Wittig reaction, we obtained a 56% yield of
the phenylsulfanyl derivative 10 of 7, and this was de-
sulfurised by sodium in liquid ammonia. To achieve an
acceptable yield of 7 (24%), we had to add ethanol to the
reduction mixture. In the absence of ethanol, the major
product was 1-methylcyclohexa-1,4-diene (12), in addition
to which some 7 and toluene (13) were also formed. The
presence of ethanol gave rise to a 14:2:1:1 mixture of 7,
12, 3-methylbenzvalene (11), and 13. With the aid of NMR
spectroscopic data available in the literature, the identifica-
tion of 11[23] and 12[24] was straightforward. The NMR
signals of 7 were also highly characteristic. In particular,
the 13C NMR spectrum indicated the presence of a bicyclo-
[1.1.0]butane system by a signal at δ ϭ 7.0, and that of a
1,1-dialkylated ethene moiety by absorptions at δ ϭ 99.9
and 154.2.
Since 11 completely rearranged to 13 at room temper-
ature within 1 d, we do not view the latter as a primary
product of the desulfurisation of 10, the mechanism of
which is summarised in Scheme 3. Probably, 10 is trans-
formed in several steps into the allyl anion 14 by incorpora-
tion of two electrons and expulsion of a benzenethiolate
ion. On protonation of the allyl termini of 14, mainly the
desired product 7 arises, accompanied by some 11. How-
ever, these steps are relatively slow if ammonia is the proton
source, providing 14 with the opportunity to rearrange
primarily to the benzyl anion (15). The protonation of 15
and the Birch reduction of the resulting 13 explain the
formation of 12 as main product. Obviously, addition of
ethanol to the reaction mixture causes a faster protonation
of 14, and hence increased yields of 7 and 11. The isomeris-
ation of 14 to 15, which proceeds rather rapidly even at
temperatures of about Ϫ60 °C, is considerably faster than
that of benzvalene to benzene or that of 11 to 13.[23]
hydrotosylation to the target tricyclic alkene 7 was at-
tempted by slow injection of 6, dissolved in dimethyl sulfox-
ide (DMSO), into a solution of KOtBu in the same solvent,
maintained at 29 °C and under a pressure of 10Ϫ3 mbar.
Under these reaction conditions, volatile products evapo-
rated as they were formed, and were condensed, together
with some solvent, in a receiver kept at liquid nitrogen tem-
perature. After workup and separation by preparative gas
chromatography, a hydrocarbon fraction was isolated and
shown to consist of a 2.5:1 mixture of 4-methylenebicyclo-
[3.1.0]hex-2-ene (8) and 7. Because of their very similar re-
tention times, the two hydrocarbons could not be separated.
From NMR spectroscopic data and by comparison with an
authentic sample,[5b,20] the identity of 8 was established
without doubt. The NMR signals of the second component
fully agreed with those given below for 7.
A control experiment excluded the possibility that 8 was
a later product of 7, as it demonstrated that treatment with
KOtBu in [D6]DMSO left 7 unchanged. Most probably, the
precursor of 8 was already a bicyclo[3.1.0]hex-2-ene derivat-
ive, which could have formed from 6 by a rearrangement
catalysed by a trace of an acid that might have remained
from the preparation of 6. The acid-catalysed rearrange-
ment of the parent tricyclo[3.1.0.02,6]hexane to bicyclo-
[3.1.0]hex-2-ene is well known.[3a]
2. Synthesis of 7 from 4-(Phenylsulfanyl)tricyclo[3.1.0.02,6]-
hexan-3-one (9)
3. Reactions of 7: Thermolysis in Solution and Addition
onto TCNE
By analogy with the higher homologue,[4] the preparation
of 7 should be straightforward from tricyclo[3.1.0.02,6]-
hexan-3-one by means of a Wittig reaction. However, access
In order to test the thermal stability of 7, we heated a
to this ketone is difficult.[21] On the other hand, a simple C6D6 solution of it, which had been prepared carefully to
synthesis of its 4-phenylsulfanyl derivative 9, in three steps avoid catalysed processes.[3b] At 175 °C, the half-life was
from benzvalene in an overall yield of 23 %, has recently found to be about 200 min, with toluene (13) as the only
been discovered.[22] This ketone could be converted into the product (Scheme 4). Thus, the rearrangement of 7 pro-
target 7 in two steps (Scheme 2).
ceeded considerably more rapidly than that of the parent
Eur. J. Org. Chem. 2002, 1374Ϫ1379
1375

D. Hasselmann, K. Loosen, T. Fischer, U. Kunz, M. Christl
FULL PAPER
ment in the cationic part with formation of the new zwitter-
ion 19. This species is represented by a classical and by a
nonclassical formula in Scheme 6. The ring-closure of 19,
combining the nucleophile with the electrophilic centre of
the one-carbon bridge of the nonclassical structure, gener-
ates the bicyclo[2.1.1]hexene derivative 20, which is subject
to a sigmatropic reorganisation resulting in 17. Hogeveen
et al.[6] observed the pentamethyl derivative of 20 directly
and determined kinetic data for its conversion into the fi-
nal product.
Scheme 3
hydrocarbon (tricyclo[3.1.0.02,6]hexane), which has a sim-
ilar half-life only at 240 °C.[3a] Possible mechanisms of these
isomerisations are discussed in a forthcoming article.[5a]
Scheme 6
Scheme 4
Tetracyanoethylene (TCNE) interacts in various ways
Experimental Section
with bicyclo[1.1.0]butane[25] and
a
number of its
derivatives.[6,25Ϫ28] In the cases of benzvalene[26] and homo-
benzvalene,[27] a homo DielsϪAlder reaction, considered to
occur concertedly, onto a vinylcyclopropane subunit takes
place (among other processes). Although 7 possesses a vi-
nylcyclopropane moiety, such a reaction appeared unlikely,
General: NMR: Varian A 60 D, Varian NV 14, Bruker AM 400,
Bruker AC 200, Bruker AC 250; internal standards TMS, CHCl3
1
(δ ϭ 7.26) and C6D5H (δ ϭ 7.15) for H NMR and CDCl3 (δ ϭ
77.0) and C6D6 (δ ϭ 128.0) for 13C NMR. IR: PerkinϪElmer 257
and 325 infrared spectrometers. UV: Varian Cary 17. MS: Varian
since the product would be a derivative of (E)-cycloheptene. MAT CH5 and CH7, Finnigan MAT 8200. Analytical GC (AGC):
PerkinϪElmer F-20 and F-22, FID, carrier gas N2, glass capillary
columns; integration with PerkinϪElmer Minigrator M-2 or
HewlettϪPackard model 3370/18. Preparative GC (PGC): Varian
90-P and 920, thermal conductivity detector, carrier gas He, glass
columns. Elemental analyses: LECO CHNS 932. All experiments
described below were conducted in anhydrous solvents in dry glass-
ware and under dry argon or nitrogen.
Warming of a solution of 7 and TCNE in dichloro-
methane at 40 °C gave rise to a 1:5 mixture of the [2ϩ2]
cycloadduct 16 and the cyclopropadicyclopentene derivat-
ive 17 (Scheme 5). These products were identified by their
NMR spectra. The 13C NMR signals at δ ϭ 6.2 and 6.5
were characteristic for the tricyclo[3.1.0.02,6]hexane subunit
of 16, while the structure 17 was supported by the presence
of an CHϭCHϪCH2 grouping as part of a cyclopentene
moiety, as deduced from the magnitude of the vicinal coup-
ling between the olefinic protons (J4,5 ϭ 5.8 Hz) and their
interactions with the protons of the methylene group.
4-[(Z)-Hydroxymethylene]tricyclo[4.1.0.02,7]heptan-3-one (2) (3-Oxo-
tricyclo[4.1.0.02,7]heptane-4-carbaldehyde): Sodium hydride (1.08 g
of a 55% suspension in mineral oil, 25 mmol NaH) was purified by
treatment with pentane and decantation of the liquid phase (3 ϫ
5 mL). Traces of the solvent were removed at room temperature in
vacuo, and diethyl ether (155 mL, from LiAlH4) was added. A solu-
tion of tricyclo[4.1.0.02,7]heptan-3-one (1,[16] 1.30 g, 12.0 mmol)
and ethyl formate (14.4 g, 194 mmol) in 85 mL of diethyl ether was
added with stirring to this suspension. The reaction was started by
addition of CH3OH (0.072 mL, 1.8 mmol) and the mixture was
stirred at room temperature for 20 h. Another 0.096 mL of CH3OH
was added during that period. Despite a successful experience with
the preparation of an analogous compound from bicyclo[3.2.0]-
Scheme 5
A pentamethyl derivative of 17 has been obtained from hept-6-en-2-one,[2d] it was not possible to isolate 2 in pure form
under various conditions. The above solution of 2 in diethyl ether,
containing some CH3OH and C2H5OH, was therefore used directly
for diazotization.
a pentamethyl derivative of 7 and TCNE. The mechanism
proposed by the authors[6] should be applicable here as well
(Scheme 6). Thus, addition of TCNE to the olefinic methyl-
ene group of 7 should initially furnish the zwitterion 18,
which can undergo two reactions: either a collapse to give
4-Diazotricyclo[4.1.0.02,7]heptan-3-one (3): p-Toluenesulfonyl az-
ide[18] (4.80 g, 24.3 mmol) in diethyl ether and freshly distilled di-
16, the minor product, or a WagnerϪMeerwein rearrange- ethylamine (5.2 mL, 50 mmol) were added at 0 °C to the above
1376 Eur. J. Org. Chem. 2002, 1374Ϫ1379

3-Methylenetricyclo[3.1.0.02,6]hexane, a Tricyclic Isomer of Toluene
FULL PAPER
solution of 2 in diethyl ether. After having been stirred for 4 h at was slowly evacuated to about 10Ϫ3 mbar. The reaction flask was
room temperature, the mixture was worked up by addition of immersed in a water bath (29 °C) and, with stirring, a solution
600 mL of an aqueous buffer solution (Na2HPO4/KH2PO4, pH ϭ
of 6 (0.175 g, 0.66 mmol) in dimethyl sulfoxide (1 mL) was slowly
7), followed by extraction with diethyl ether (3 ϫ 50 mL) and injected. Volatile products were condensed in the traps, which were
dichloromethane (2 ϫ 50 mL). The dried (MgSO4) organic phases maintained at liquid nitrogen temperature. Stirring was continued
were concentrated in vacuo to yield 5.20 g of crude product con- for 30 min, and the reaction flask was then warmed to about 50Ϫ60
taining, as shown by 1H NMR analysis, about 20% of 3, 20% of p-
°C, whereupon some DMSO distilled. Water was added to the
toluenesulfonyl azide, 15% of methyl p-toluenesulfonate and 45% products in the traps, and the combined mixtures were extracted
of ethyl p-toluenesulfonate. Yield: 0.70Ϫ1.00 g (43Ϫ62% relative to
1) of 3. For spectral characterisation a small portion was purified
by chromatography on Florisil, with diethyl ether/pentane (1:3) as
with pentane (3 ϫ 2 mL). The combined pentane layers were
washed with water, dried (MgSO4), and carefully concentrated. The
residue was shown by AGC (column: 83.5 m ϫ 0.28 mm, PPG ϩ
Ϫ1
˜
eluant. IR (film): ν ϭ 2960 cm (s), 2850 (m), 2085 (s, CϭN2), KOH, 80 °C) to consist of a mixture of 8 (71%, tR ϭ 15.1 min), 7
1
1635 (s, CϭO), 1260, 1035, 800. H NMR (60 MHz, CDCl3): δ ϭ
(27%, tR ϭ 15.2 min), toluene (13, 2%, tR ϭ 16.0 min), and some
2.39 (t, J1,2 ϭ J1,6 ϭ J2,7 ϭ J6,7 ϭ 2.2 Hz, 2 H, 1-H, 7-H), pentane. Separation of this mixture by PGC (column: 2.0 m ϫ 3/8
2.65Ϫ3.00 (m, 4 H, 2-H, 5-H2, 6-H).
in, 20% Carbowax 20M ϩ KOH on Chromosorb P 60/80 mesh, 78
°C, 100 mL He/min) gave one fraction (tR ϭ 11 min) shown by
AGC still to be a mixture of 8 (65%) and of 7 (35%), which fur-
nished the following data: GC-MS (EI, 70 eV): m/z (%) ϭ 92 (100)
[Mϩ], 91 (79) [C7H7ϩ], 65 (68) [C5H5ϩ], 39 (53) [C3H3ϩ]. IR (CCl4):
Methyl Tricyclo[3.1.0.02,6]hexane-3-carboxylate (4): A solution of
4-diazotricyclo[4.1.0.02,7]heptan-3-one (3, 0.40 g, 3.0 mmol) in
50 mL of absolute CH3OH was purged with dry argon and then
irradiated (Pyrex, PhilipsϪHPK, 125 W) at room temperature for
about 90 min, resulting in evolution of nitrogen (45 mL, corres-
ponding to a yield of 62%). The solvent was carefully removed in
vacuo, and the residue was subjected to a short-path distillation at
10Ϫ3 mbar to yield 4 as a colourless oil of about 90% purity. MS
ν ϭ 3055 cmϪ1 (m), 2995 (w), 1635 (w, CϭC), 1620 (w, CϭC), 863,
˜
1
660. UV (cyclohexane): λmax ϭ 245 nm. H NMR (60 MHz, CCl4)
for 7: δ ϭ 4.55 (m), 4.75 (m), and 1.8Ϫ2.4 (m) in an intensity ratio
of about 1:1:6; further NMR spectroscopic data are given below;
for 8: δ ϭ 0.37 (q, average of J1,6endo, J5,6endo, and J6endo,6exo
ϭ
(EI, 70 eV): m/z (%) ϭ 138 (16) [Mϩ], 107 (37) [C7H7Oϩ], 78 (100)
3.5 Hz, 1 H, 6-Hendo), 0.95 (td, average of J1,6exo and J5,6exo ϭ 7.5,
J6endo,6exo ϭ 3.5 Hz, 1 H, 6-Hexo), 1.8Ϫ2.4 (complex, 2 H, 1-H, 5-
H), 4.85 (m, 2 H, ϭCH2), 5.63 (m) and 6.15 (m) (2 ϫ 1 H, 2-H, 3-
H), in agreement with the literature.[20] Independently synthesised
sample of 8:[2d][5b] 1H NMR (400 MHz, CDCl3/C6D6): δ ϭ 0.40/
0.38 (q, average of J1,6endo, J5,6endo, and J6endo,6exo ϭ 3.5 Hz, 1 H, 6-
Hendo), 1.02/0.71 (td, average of J1,6exo and J5,6exo ϭ 7.5, J6endo,6-
[C6H6ϩ], 77 (100) [C6H5 ]. IR: ν ϭ 3045 cmϪ1 (w), 2950 (m), 2875
ϩ
˜
(w), 1725 (vs, CϭO), 1435, 1195, 1170, 760. 1H NMR (60 MHz,
CCl4): δ ϭ 1.4Ϫ2.7 (complex, 7 H, 1-H, 2-H, 3-H, 4-H2, 5-H, 6-
H), 3.53 (s, 3 H, CH3).
(Tricyclo[3.1.0.02,6]hex-3-yl)methanol
(5):
Methyl
tricyclo-
[3.1.0.02,6]hexane-3-carboxylate (4, 130 mg, 0.94 mmol) in 0.5 mL of
diethyl ether was added to a suspension of LiAlH4 (38 mg,
1.0 mmol) in diethyl ether (4 mL, from LiAlH4). After having been
stirred for 18 h at room temperature, the mixture was cooled to 0
°C and then treated with 0.1 mL of H2O. Filtration of the mixture,
drying of the filtrate (Na2SO4) and evaporation of the solvent in
ϭ 3.5 Hz, 1 H, 6-Hexo), 2.12Ϫ2.21 (m, 2 H, 1-H, 5-H)/1.83 (m,
exo
1 H, 1-H) and 1.96 (m, 1 H, 5-H), 4.91/4.94 (br. s, 1 H, 7-HZ),
5.03/5.05 (br. s, 1 H, 7-HE), 5.77/5.70 (d, J2,3 ϭ 5.5 Hz, 1 H, 3-H),
6.28 (dddd, J2,3 ϭ 5.5, J ഠ 2.5, 1.5, 1.0 Hz)/6.04 (dm) (1 H, 2-H);
the assignment is based on H,H COSY spectra and NOE measure-
ments. 13C NMR (101 MHz, CDCl3/C6D6): δ ϭ 21.3/21.8 (C-5),
24.1/24.2 (C-6), 24.3/24.7 (C-1), 105.3/105.6 (ϭCH2), 129.7/130.2
(C-3), 141.4/141.3 (C-2), 154.8/155.0 (C-4); the assignment is based
on C,H COSY spectra.
vacuo yielded 5 (90 mg, 87%) as an oily liquid. MS (EI, 70 eV):
m/z (%) ϭ 110 (9) [Mϩ], 79 (83) [C6H7 ], 42 (100). IR (film): ν ϭ
ϩ
˜
3600Ϫ3000 cmϪ1 (s, OH), 3035 (w), 2930 (s), 2865 (s), 1110, 1035,
750. 1H NMR (60 MHz, CDCl3): δ ϭ 1.0Ϫ2.4 (complex, 8 H,
1-H, 2-H, 3-H, 4-H2, 5-H, 6-H, -OH), 3.4 (m, 2 H, CH2OH).
4-Methylenetricyclo[3.1.0.02,6]hex-3-yl Phenyl Sulfide (10): A mix-
ture of methyltriphenylphosphonium bromide (9.83 g, 27.5 mmol)
and freshly sublimed potassium tert-butoxide (3.08 g, 27.5 mmol)
in diethyl ether (550 mL) was stirred at room temperature for
30 min. The resulting solution was added dropwise to a solution of
4-(phenylsulfanyl)tricyclo[3.1.0.02,6]hexan-3-one[22] (9, 5.56 g, 27.5
mmol) in diethyl ether (440 mL). Stirring was continued for 3 h at
room temperature. The mixture was then diluted with light petro-
leum ether (b.p. 30Ϫ50 °C) (LP) and filtered through basic Al2O3
(activity IV) with LP as eluant. After concentration of the filtrate
in vacuo, the residue was subjected to flash chromatography (basic
Al2O3, activity III, LP) to give 10 (3.09 g, 56%) as a yellowish
(Tricyclo[3.1.0.02,6]hex-3-yl)methyl p-Toluenesulfonate (6): At 0 °C,
p-toluenesulfonyl chloride (137 mg, 0.72 mmol) was added with
stirring to tricyclo[3.1.0.02,6]hex-3-ylmethanol (5) (40 mg,
0.36 mmol) in 0.5 mL of pyridine. This solution was kept in the
refrigerator for 14 h and then poured into ice/water. The mixture
was extracted with diethyl ether. The ether layer was washed with
an aqueous buffer solution (pH ϭ 7, 3 ϫ 1 mL) and dried
(Na2SO4). Removal of the solvent in vacuo at 0 °C gave 6 (90 mg,
95%) as a slightly yellow oil, which was used directly for the elim-
ination reaction. IR (film): ν ϭ 3040 cmϪ1 (w), 2940 (m), 2865 (w),
˜
1598 (m, arom. CϭC), 1495 (w, arom. CϭC), 1360, 1190, 1175,
960, 662. 1H NMR (60 MHz, CCl4): δ ϭ 1.20Ϫ2.15 (complex, 7
H, 1-H, 2-H, 3-H, 4-H2, 5-H, 6-H), 2.42 (s, 3 H, CH3), 3.60Ϫ3.83
(m, 2 H, CH2O), 7.33 (2 H) and 7.73 (2 H) (AAЈBBЈ spectrum,
C6H4).
1
liquid. H NMR (200 MHz, CDCl3): δ ϭ 2.38 (dm, J2,5 ϭ 4.9 Hz,
1 H) and 2.49 (m, 1 H) (2-H, 5-H), 2.49Ϫ2.53 (m, 2 H, 1-H, 6-H),
3.85 (m, 1 H, 3-H), 4.97 (d, J ϭ 1.5 Hz, 1 H) and 5.08 (m, 1 H)
(ϭCH2), 7.21 (tt, 1 H, p-H), 7.28 (m, 2 H, m-H), 7.43 (m, 2 H, o-
H). 13C NMR (50 MHz, CDCl3): δ ϭ 7.5 and 9.8 (C-1, C-6), 38.0
and 39.4 (C-2, C-5), 50.1 (C-3), 103.7 (ϭCH2), 126.3 (p-C), 128.7
(m-C), 130.7 (o-C), 136.0 (i-C), 153.3 (C-4); the assignment is based
on a C,H COSY spectrum. C13H12S (200.3): calcd. C 77.95, H 6.04,
S 16.01; found C 77.69, H 6.62, S 15.30.
Treatment of 6 with KOtBu ؊ Formation of 3-Methylenetricy-
clo[3.1.0.02,6]hexane (7) and 4-Methylenebicyclo[3.1.0]hex-2-ene (8):
KOtBu (0.71 g, 6.3 mmol) and dimethyl sulfoxide (10 mL, from
CaH2) were placed in a three-necked flask equipped with a septum
and connected through a short-path distillation apparatus to two Reduction of 10 with Sodium in Liquid Ammonia ؊ Formation of 3-
consecutive traps. After having been purged with argon, the system
Methylenetricyclo[3.1.0.02,6]hexane (7)
Eur. J. Org. Chem. 2002, 1374Ϫ1379
1377

D. Hasselmann, K. Loosen, T. Fischer, U. Kunz, M. Christl
FULL PAPER
a) In the Absence of Ethanol: A solution of 10 (2.00 g, 9.99 mmol)
(16) and (3aS*,3bS*,6aR*)-3b,6-Dihydrocyclopropa[1,2:1,3]dicyclo-
in diethyl ether (7 mL) was added dropwise to a stirred solution of pentene-2,2,3,3(1H,3aH)-tetracarbonitrile (17): Tetracyanoethylene
sodium (949 mg, 41.2 mmol) in liquid ammonia (30 mL), kept at
(556 mg, 4.34 mmol) was added to a solution of a 4:2:1 mixture of
Ϫ70 °C. Stirring was continued for 2 h at Ϫ60 °C, whereupon
12, 7, and 13, containing ca. 50 mg (0.54 mmol) of 7, in dichloro-
NH4Cl (2.21 g, 41.3 mmol) was carefully added in small portions. methane (10 mL) at room temperature. The resulting suspension
The mixture was then allowed to warm to room temperature and was stirred at 40 °C for 18 h. The solid was then removed by filtra-
the evaporating ammonia was piped through a cold trap (Ϫ25 °C)
to collect coevaporated product. The residue in the reaction vessel remaining oil by flash chromatography on silica gel, with a gradient
and the contents of the trap were treated with pentane and cold
of 11% to 20% to 33% of ethyl acetate in light petroleum ether
water (0 °C), and the mixtures were combined. After separation of (b. p. 30Ϫ50 °C), gave 51 mg (43%) of a beige solid, shown by
tion and the filtrate was concentrated in vacuo. Purification of the
the layers, the aqueous phase was extracted with pentane (4 ϫ
50 mL). The combined pentane layers were dried with Na2SO4/
K2CO3 and concentrated by distillation at 60 °C (bath)/atmo-
NMR to consist of 16 and 17 in the ratio of 1:5. MS (EI, 70 eV):
m/z (%) ϭ 220 (0.7) [Mϩ], 142 (17), 115 (21), 92 (32), 91 (100), 51
(11), 39 (17). C13H8N4 (220.2): calcd. C 70.90, H 3.66, N 25.44;
spheric pressure through Vigreux columns of 22 cm (initially) and found C 70.73, H 3.44, N 25.12.
8 cm length and finally without a column at room temperature/
Compound 16: 1H NMR (250 MHz, CDCl3): δ ϭ 1.82 (dm, J4Ј,4Ј
ϭ
500Ϫ600 mbar. Volatile components of the residue were evaporated
at 12 mbar in two stages and were condensed in traps cooled with
liquid nitrogen. Evaporation at room temperature gave 1.04 g of a
colourless liquid consisting of pentane, 1-methylcyclohexa-1,4-di-
ene (12), 7, and toluene (13) in a ratio of 10:4:2:1 as analysed by
1H NMR spectroscopy. The signals of 12 were identified from liter-
ature data.[24] Evaporation with warming of the residue to 40 °C
furnished 186 mg of a colourless liquid shown to be an 8:2:1 mix-
ture of 12, 7, and 13.
13.0 Hz, 1 H) and 2.23 (dm, J4Ј,4Ј ϭ 13.0 Hz, 1 H) (4Ј-H2), 2.20
(m, 1 H; 5Ј-H), 2.42 (dm, J1Ј,6Ј ϭ 8.9 Hz, 1 H) and 2.50 (dm, J1Ј,6Ј
8.9 Hz, 1 H) (1Ј-H, 6Ј-H), 2.67 (dt, J2Ј,5Ј ϭ 5.2, J1Ј,2Ј ϭ J2Ј,6Ј
ϭ
ϭ
1.8 Hz, 1 H, 2Ј-H), 3.04 (s, 2 H, 4-H2). 13C NMR (63 MHz,
CDCl3): δ ϭ 6.2 and 6.5 (C-1Ј, C-6Ј), 32.5 (C-5Ј), 41.3 (C-4), 41.8
(C-2Ј), 42.4 (C-4Ј), 108.8, 110.7, and 111.0 (3 ϫ CN); the signals
of the fourth CN group, C-1 (C-3Ј), C-2, and C-3 were not ob-
served, due either to too low intensity or to being superimposed by
signals of 17; the assignment is based on a C,H COSY spectrum.
b) In the Presence of Ethanol: Liquid ammonia (100 mL) was
stirred at Ϫ70 °C, and ethanol (12.9 mL), a solution of 10 (923 mg,
4.61 mmol) in diethyl ether (50 mL) and, finally, sodium (802 mg,
34.9 mmol) were added in small portions. After this mixture had
been further stirred for 2 h at Ϫ60 °C, NH4Cl (1.89 g, 35.3 mmol)
was carefully added in small portions. The mixture was then stirred
for a further 30 min at Ϫ60 °C and afterwards worked up as in
procedure a). Only one product fraction (145 mg of a yellow liquid)
was collected by evaporation at 12 mbar and condensation in a cold
trap. It consisted mainly of 7, 12, 3-methylbenzvalene (11), and 13
Compound 17: 1H NMR (250 MHz, CDCl3): δ ϭ 1.76 (br. d,
J3a,3b ϭ 1.5 Hz, 1 H, 3a-H), 2.75Ϫ2.80 (m, 2 H, 6-H2), 2.86 (m, 1
H, 3b-H), 3.08 (d, J1,1 ϭ 14.0 Hz, 1 H) and 3.18 (d, J1,1 ϭ 14.0 Hz,
1 H) (1-H2), 5.77 (dt, J4,5 ϭ 5.8, average of 2 J4,6 ϭ 2.1 Hz, 1 H,
4-H), 5.92 (dqd, J4,5 ϭ 5.8, average of J3b,5 and 2 J5,6 ϭ 2.1, J ϭ
0.9 Hz, 1 H, 5-H). 13C NMR (63 MHz, CDCl3): δ ϭ 32.5 (C-3b),
36.4 (C-3a), 36.9 (C-6), 37.2 (C-6a), 39.2 (C-1), 45.6 (C-2), 55.1 (C-
3), 109.9, 110.2, 110.3 111.4 (4 ϫ CN), 130.4 (C-5), 132.3 (C-4);
the assignment is based on a C,H COSY spectrum.
1
in a ratio of 14:2:1:1 and some pentane as analysed by H NMR
spectroscopy immediately after preparation and workup. When the
NMR sample (CDCl3) was kept at room temperature for 1 d, 11
converted completely into 13. The known data for 11[23] allowed
its straightforward identification. Through the use of mesitylene as
internal standard, the yield of 7 was determined to be 103 mg
(24%). The purification of 7 by gas chromatography is described in
a forthcoming paper.[5] 1H NMR (250 MHz, CDCl3): δ ϭ 1.95 (m,
2 H, 4-H2), 2.10 (dtq, J2,5 ϭ 4.9, J1,5 ϭ J5,6 ϭ 1.8, J ϭ 0.9 Hz, 1
H, 5-H), 2.27 (‘‘quint’’, line distance 1.6 Hz, 2 H, 1-H, 6-H), 2.40
(br. dt, J2,5 ϭ 4.9, J1,2 ϭ J2,6 ϭ 1.8 Hz, 1 H, 2-H), 4.69 (m, 1 H)
and 4.91 (m, 1 H) (ϭCH2). 13C NMR (63 MHz, CDCl3): δ ϭ 7.0
(C-1, C-6), 31.9 (C-5), 32.6 (C-4), 40.5 (C-2), 99.9 (ϭCH2), 154.2
(C-3); the assignment is based on a C,H COSY spectrum.
Acknowledgments
We are grateful to CHEMETALL GmbH for gifts of chemicals,
and to the Fonds der Chemischen Industrie and to the Deutsche
Forschungsgemeinschaft for financial support. In addition, we
thank Dipl.-Chem. Edith Herberth for the performance of a con-
trol experiment.
[1] [1a]
A. Greenberg, J. F. Liebman, Strained Organic Molecules,
Academic Press, New York, 1978. [1b] J. J. Gajewski, Hydrocar-
bon Thermal Isomerizations, Academic Press, New York, 1981.
[1c]
H. Hopf, Classics in Hydrocarbon Chemistry, Wiley-VCH,
Weinheim, 2000.
[2] [2a]
[2b]
Thermolysis of 7 in C6D6: LiAlH4 (ca. 20 mg) was added to a solu-
tion of one drop of tetramethylethylenediamine, 7 (ca.10 mg, con-
taminated by some 12 and 13), and mesitylene (8 mg) in C6D6
(0.6 mL). The mixture was kept at 20 °C for 1 h. The volatile com-
ponents were then evaporated at 2·10Ϫ5 mbar and condensed in an
NMR tube, cooled with liquid nitrogen. (Prior to that operation,
the evacuated NMR tube had been flame-dried.) While the sample
in the NMR tube was still cooled with liquid nitrogen, the NMR
tube was evacuated and sealed. As shown by NMR analysis, 7 re-
arranged cleanly to 13 on heating at 175 °C. From the integrated
intensities of 7, 13, and mesitylene, the half-life was determined to
be ca. 200 min.
D. Hasselmann, Tetrahedron Lett. 1972, 3465Ϫ3468.
D.
D. Has-
selmann, Angew. Chem. 1975, 87, 252Ϫ254; Angew. Chem. Int.
[2c]
Hasselmann, Tetrahedron Lett. 1973, 3739Ϫ3742.
[2d]
Ed. Engl. 1975, 14, 257Ϫ258.
Angew. Chem. 1978, 90, 641Ϫ642; Angew. Chem. Int. Ed. Engl.
D. Hasselmann, K. Loosen,
[2e]
1978, 17, 606Ϫ607.
D. Hasselmann, P.-J. Rissing, Tetrahed-
ron Lett. 1979, 1745Ϫ1748.
[3] [3a]
M. Christl, U. Heinemann, W. Kristof, J. Am. Chem. Soc.
[3b]
1975, 97, 2299Ϫ2301.
M. Christl, R. Stangl, H. Jelinek-
Fink, Chem. Ber. 1992, 125, 485Ϫ497.
[4]
D. Hasselmann, K. Loosen, Angew. Chem. 1980, 92, 651Ϫ652;
Angew. Chem. Int. Ed. Engl. 1980, 19, 634Ϫ636.
[5] [5a]
D. Hasselmann, D. Baumann, M. Christl, manuscript in
[5b]
preparation.
Bochum, 2001.
D. Baumann, Dissertation, Universität
Reaction between 7 and Tetracyanoethylene ؊ Formation of Spiro-
[cyclobutane-1,3Ј-tricyclo[3.1.0.02,6]hexane]-2,2,3,3-tetracarbonitrile
[6] [6a]
R. F. Heldeweg, H. Hogeveen, L. Zwart, Tetrahedron Lett.
1378
Eur. J. Org. Chem. 2002, 1374Ϫ1379

3-Methylenetricyclo[3.1.0.02,6]hexane, a Tricyclic Isomer of Toluene
FULL PAPER
[6b]
1977, 2535Ϫ2538.
H. Hogeveen, L. Zwart, J. Org. Chem.
gew. Chem. 1987, 99, 588Ϫ590; Angew. Chem. Int. Ed. Engl.
[15h]
1979, 44, 1365Ϫ1369.
1987, 26, 554.
P. E. Eaton, P. G. Jobe, I. D. Reingold, J.
[7]
[15i]
D. Redmore, C. D. Gutsche, Adv. Alicycl. Chem. 1971, 3,
125Ϫ136.
Am. Chem. Soc. 1984, 106, 6437Ϫ6439.
W.-D. Fessner, H.
Prinzbach, G. Rihs, Tetrahedron Lett. 1983, 24, 5857Ϫ5860.
[8] [8a]
[15j]
L. Horner, E. Spietschka, Chem. Ber. 1952, 85, 225Ϫ229.
W.-D. Fessner, G. Sedelmeier, P. R. Spurr, G. Rihs, H.
[8b]
For reviews of the photolysis method, see:
M. Regitz, G.
Prinzbach, J. Am. Chem. Soc. 1987, 109, 4626Ϫ4642.
J. Ipaktschi, Chem. Ber. 1972, 105, 1996Ϫ2003.
[16]
[17] [17a]
Maas, Diazo Compounds, Academic Press, New York, 1986, pp.
[8c]
185Ϫ195.
W. Ando, in: The Chemistry of Diazonium and
M. Regitz, Angew. Chem. 1967, 79, 786Ϫ801; Angew.
[17b]
Diazo Groups (Ed.: S. Patai), Wiley, New York, 1978, part 1,
Chem. Int. Ed. Engl. 1967, 6, 733.
1972, 351Ϫ373.
[18] [18a]
M. Regitz, Synthesis
pp. 458Ϫ475.
[9] [9a]
L. Wolff, Justus Liebigs Ann. Chem. 1902, 325, 129Ϫ195.
M. Regitz, J. Hocker, A. Liedhegener, Org. Syn. Coll. Vol.
[9b]
[18b]
L. Wolff, Justus Liebigs Ann. Chem. 1912, 394, 23Ϫ59.
V 1968, 179Ϫ183.
W. v. E. Doering, C. H. DePuy, J. Am.
[10]
[10a]
[18c]
For reviews, see:
H. Meier, K.-P. Zeller, Angew. Chem.
Chem. Soc. 1953, 75, 5955Ϫ5957.
J. Prakt. Chem. 1930, 125, 303Ϫ340.
See, e.g.: refs.[15c,15d,15h,15i]
T. Curtius, G. Kraemer,
1975, 87, 52Ϫ63; Angew. Chem. Int. Ed. Engl. 1975, 14, 32Ϫ43.
[10b]
[19]
[20]
L. L. Rodina, I. K. Korobitsyna, Russ. Chem. Rev. 1967,
36, 260Ϫ272. [10c] W. Kirmse, Carbene Chemistry, 2nd ed., Aca-
M. Rey, U. A. Huber, A. S. Dreiding, Tetrahedron Lett. 1968,
demic Press, New York, 1971, pp. 475Ϫ493.
3583Ϫ3588.
[11]
[12]
[11a]
[21]
[22]
For recent preparative applications, see:
H. Yang, K. Fos-
A. Kraft, Dissertation, Universität Würzburg, 1989.
U. Kunz, S. Krimm, T. Fischer, T. Kottke, D. Stalke, M.
Christl, Eur. J. Org. Chem. 1998, 2171Ϫ2176.
ter, C. R. J. Stephenson, W. Brown, E. Roberts, Org. Lett. 2000,
[11b]
2, 2177Ϫ2179.
J. Podlech, M. R. Linder, J. Org. Chem.
[23] [23a]
1997, 62, 5873Ϫ5883.
U. Burger, G. Gandillon, Tetrahedron Lett. 1979,
[23b]
´
For a recent theoretical study, see: S. Calvo-Losada, D. Suarez,
T. L. Sordo, J. J. Quirante, J. Phys. Chem. B 1999, 103,
7145Ϫ7150.
4281Ϫ4284.
U. Burger, G. Gandillon, J. Mareda, Helv.
Chim. Acta 1981, 64, 844Ϫ853.
[24]
[25]
[26]
J. Beger, B. Thomas, T. Vogel, K. Kirmse, R. Lang, J. Prakt.
Chem. 1991, 333, 481Ϫ488.
K. B. Wiberg, S. T. Waddell, J. Am. Chem. Soc. 1990, 112,
2194Ϫ2216.
[13]
[14]
T. Ye, M. A. McKervey, Chem. Rev. 1994, 94, 1091Ϫ1160.
For ketene intermediates, see: [14a] R. J. McMahon, O. L. Chap-
man, R. A. Hayes, T. C. Hess, H.-P. Krimmer, J. Am. Chem.
[14b]
Soc. 1985, 107, 7597Ϫ7606.
T. Lippert, A. Koskelo, P. O.
M. Christl, E. Brunn, F. Lanzendörfer, J. Am. Chem. Soc. 1984,
[14c]
Stoutland, J. Am. Chem. Soc. 1996, 118, 1551Ϫ1552.
Visser, R. Zuhse, M. W. Wong, C. Wentrup, J. Am. Chem. Soc.
P.
106, 373 Ϫ382.
[27] [27a]
M. Christl, R. Lang, C. Herzog, R. Stangl, K. Peters, E.-
1996, 118, 12598Ϫ12602.
M. Peters, H. G. von Schnering, Angew. Chem. 1985, 97,
595Ϫ596, Angew. Chem. Int. Ed. Engl. 1985, 24, 611Ϫ613. [27b]
R. Lang, C. Herzog, R. Stangl, E. Brunn, M. Braun, M.
[15]
[15a]
See, e. g.:
J. Meinwald, P. G. Gassman, J. Am. Chem. Soc.
1960, 82, 2857Ϫ2863. [15b] K. B. Wiberg, B. A. Hess, Jr., J. Org.
[15c]
Chem. 1966, 31, 2250Ϫ2254.
K. B. Wiberg, B. L. Furtek,
Christl, E.-M. Peters, K. Peters, H. G. von Schnering, Chem.
[15d]
[27c]
L. K. Olli, J. Am. Chem. Soc. 1979, 101, 7675Ϫ7679.
K.
Ber. 1990, 123, 1193Ϫ1207.
E. Kim, M. Christl, J. K. Ko-
B. Wiberg, L. K. Olli, N. Golembeski, R. D. Adams, J. Am.
chi, Chem. Ber. 1990, 123, 1209Ϫ1218.
L. A. Paquette, C. J. Lau, R. D. Rogers, J. Am. Chem. Soc.
1988, 110, 2592Ϫ2600.
[15e]
[28]
Chem. Soc. 1980, 102, 7467Ϫ7475.
V. B. Rao, S. Wolff, W.
C. Agosta, J. Chem. Soc., Chem. Commun. 1984, 293Ϫ294. [15f]
V. B. Rao, C. F. George, S. Wolff, W. C. Agosta, J. Am. Chem.
Soc. 1985, 107, 5732Ϫ5739.
Received July 4, 2001
[15g]
A. Otterbach, H. Musso, An-
[O01327]
Eur. J. Org. Chem. 2002, 1374Ϫ1379
1379