So2-extrusion of an 8-thiabicylo[3.2.1]octa-2,6-diene 8,8-dioxide and rearrangement of the resulting cycloheptatriene

Article abstract of DOI:10.1002/hc.20079

The reaction of 3,4-di-tert-butylthio-phene 1-oxide (8) with tetrachlorocyclopropene provided 6,7-di-tert-butyl-2,3,4,4-tetrachloro-8-thia- bicylo[3.2.1]-octa-2,6-diene 8-oxide (10), which was oxidized to the corresponding 8,8-dioxide 16 by m-chloroperbenzoic acid. The thermolysis of 16 in refluxing chlorobenzene, xylene, or octane gave 5-tert-butyl-1,2-dichloro-3- [(1,1-dichloro-2,2-dimethyl)propyl]benzene (18) with extrusion of SO₂ and 2-tert-butyl-4,5,6-trichloro-9,9-dimethylbicyclo[5.2.0]nona-1,3,5-triene (19) with extrusion of SO₂ and HCl in 73-78% combined yields. On the other hand, the thermolysis of 16 in the presence of triethylamine gave 19 as the sole product in 98% yield. A mechanism that involves the initial formation of 4,5-di-tert-butyl-1,2,7,7-tetrachlorocycloheptatriene (17) is proposed to explain the observed products.

Full text of DOI:10.1002/hc.20079

Heteroatom Chemistry
Volume 16, Number 2, 2005
SO -Extrusion of an
2
8-Thiabicylo[3.2.1]octa-2,6-diene 8,8-dioxide
and Rearrangement of the Resulting
Cycloheptatriene
Jun Takayama, Yoshiaki Sugihara, and Juzo Nakayama
Department of Chemistry, Faculty of Science, Saitama University, Sakura-ku, Saitama, Saitama
338-8570, Japan
Received 8 November 2004
(2) to give the corresponding adducts (3), which
ABSTRACT: The reaction of 3,4-di-tert-butylthio-
extrude SO2 spontaneously to furnish cyclohepta-
phene 1-oxide (8) with tetrachlorocyclopropene
trienes (4) as the final product [1] (Scheme 1). We
provided 6,7-di-tert-butyl-2,3,4,4-tetrachloro-8-thia-
then examined the reaction of 3,4-di-tert-butylthio-
bicylo[3.2.1]octa-2,6-diene 8-oxide (10), which was
phene 1,1-dioxide (5) [2] with tetrachlorocyclopro-
oxidized to the corresponding 8,8-dioxide 16 by
pene (6) with the expectation of obtaining a cyclo-
m-chloroperbenzoic acid. The thermolysis of 16 in
heptatriene 7, which would serve as the precursor
refluxing chlorobenzene, xylene, or octane gave 5-tert-
that leads to a range of seven-membered nonben-
butyl-1,2-dichloro-3-[(1,1-dichloro-2,2-dimethyl)pro-
zenoids possessing two tert-butyl groups at vicinal
pyl]benzene (18) with extrusion of SO2 and 2-tert-
positions. Disappointingly, however, 5 failed to re-
butyl-4,5,6-trichloro-9,9-dimethylbicyclo[5.2.0]nona-
act with 6 probably due to the steric hindrance of 5;
1,3,5-triene (19) with extrusion of SO2 and HCl in
under forcing conditions (190◦C in a sealed tube),
73–78% combined yields. On the other hand, the ther-
considerable decomposition of 6 took place. Re-
molysis of 16 in the presence of triethylamine gave 19
cent studies uncovered that thiophene 1-oxides are a
as the sole product in 98% yield. A mechanism that in-
more reactive diene than thiophene 1,1-dioxides [3].
volves the initial formation of 4,5-di-tert-butyl-1,2,7,7-
We therefore examined the Diels–Alder reaction of
tetrachlorocycloheptatriene (17) is proposed to ex-
3,4-di-tert-butylthiophene 1-oxide (8) [4] with 6. This
ꢀ
C
plain the observed products. 2005 Wiley Periodicals,
study led us to some new findings described below.
Inc. Heteroatom Chem 16:132–137, 2005; Published on-
line in Wiley InterScience (www.interscience.wiley.com).
DOI 10.1002/hc.20079
RESULTS AND DISCUSSION
Heating equimolar amounts of the 1-oxide 8 and
the cyclopropene 6 in boiling toluene for 20 h pro-
INTRODUCTION
vided the adduct 10 as the major product in 44%
Reportedly, thiophene 1,1-dioxides (1) undergo
yield (Scheme 2). Thiophenes 13 (27%) and 14
Diels–Alder reactions with a series of cyclopropenes
(27%) were obtained as by-products. The use of two
molar amounts of 6 did not improve the yield of
10; 10 was obtained in a slightly decreased yield
Correspondence to: Juzo Nakayama; e-mail: nakaj@post.
saitama-u.ac.jp.
2005 Wiley Periodicals, Inc.
(36%) together with 13 (22%), 14 (36%), and (Z)-2,3-
dichloropropenoic acid (15). The 1H NMR spectrum
c
ꢀ
132

SO2-Extrusion of an 8-Thiabicylo[3.2.1]octa-2,6-diene 8,8-dioxide and Rearrangement of the Resulting Cycloheptatriene 133
of 10 showed two singlets of the tert-butyl groups
and two doublets of the bridgehead methine pro-
tons. The 13C NMR spectrum showed 11 peaks in-
cluding the 4 sp2 carbon peaks. 10 would be formed
by ring opening of the expected Diels–Alder adduct
9 [5] with a simultaneous chlorine migration. A sim-
ilar rearrangement was observed for the adduct of
2,5-dimethoxyfuran with 6 [6]. The configuration of
the S O group in 10 was tentatively assigned on the
basis of our recent results that Diels–Alder reactions
of 8 with a variety of dienophiles take place exclu-
sively at the syn-ꢀ-face with respect to the S O group
in an endo-mode [5]. The formation of 13 and 14 as
by-products would be explained as follows. The re-
action of 8 with 6 would produce a sulfonium ion 11,
which decomposes to cyclopropenone 12, thiophene
13, and molecular chlorine. Finally, chlorination of
13 would produce 14. The cyclopropene 6 might dis-
sociate slightly into the aromatic cyclopropenium
ion 6ꢁ under the conditions. Thus, the reaction of
8 with 6ꢁ, and not the direct reaction of 8 with 6,
might be involved in the formation of 11. Unfortu-
nately, 12 could not be isolated probably because of
the volatility. Hydrolysis of the unreacted 6, and not
of 12, would produce the carboxylic acid 15.
SCHEME 1
We then examined the oxidation of 10 with the
intention of obtaining the sulfone 16 as the precur-
sor of the cycloheptatriene 17 (Scheme 3). The oxi-
dation of 10 with m-chloroperbenzoic acid (MCPBA)
or dimethyldioxirane at room temperature furnished
16 quantitatively.
Thermolysis of 16 gave rather unexpected re-
sults. Thus heating 16 in boiling chlorobenzene gave
aromatized compound 18 and bicyclic compound
19 in 23% and 55% yields, respectively (Scheme 4).
Thermolysis of 16 in xylene or octane also gave 18
and 19 in good combined yields, whereas thermoly-
sis in benzonitrile gave a complex mixture contain-
ing 18 and 19 (Table 1). The expected simple SO2-
extrusion product 17 was not obtained in any case.
SCHEME 3
SCHEME 2

134 Takayama, Sugihara, and Nakayama
FIGURE 1 Molecular structure of 18.
of 19 showed one singlet of the tert-butyl group and
two singlets of the two methyl groups, indicating
that one of the two tert-butyl groups of 16 incor-
porated in the construction of the four-membered
ring. The vinyl proton appeared at δ 6.90 as sin-
glet. Both methylene and methine protons appeared
as multiplets; decoupling experiments revealed the
chemical shifts of δ Ha = 3.12, Hb (or Hc) = 1.65,
and Hc (or Hb) = 1.69, and the coupling constants
of JHa−Hb = 8.8, JHa−Hc = 3.8, and JHb−Hc = 14.3 Hz. In
addition, 15.7% NOE was observed between the tert-
butyl and vinyl protons, and 16.6% and 17.5% NOEs
were observed between the tert-butyl protons and the
two methyl protons of the four-membered ring. The
13C NMR spectrum showed 13 peaks, 6 of which ap-
peared in the sp2 carbon region. The methylene and
methine peaks appeared at δ 42.4 and 54.2, respec-
tively. The IR spectrum showed the C C stretching
vibrations at 1545 and 1576 cm−1, and the UV-Vis
spectrum showed an absorption maximum at 281
nm indicating the presence of the cycloheptatriene
unit [7].
The formation of 18 and 19 is explained as fol-
lows (Scheme 5). It is well documented that SO2-
extrusion of 2,5-dihydrothiophene 1,1-dioxides takes
place to give 1,3-butadienes in a thermally allowed
disrotatory manner [8]. Thus the thermolysis of 16,
which possesses a dihydrothiophene dioxide ring,
would produce the cycloheptatriene 17 initially. It
is also known that an equilibrium exists between
cycloheptatriene and norcaradiene structures and,
in addition, the equilibrium lies to the norcaradi-
ene side when electron-withdrawing substituents are
attached at the 7-position [9]. Thus 17 tautomer-
izes to a norcaradiene 21, which then rearranges
to an exomethylene compound 22. Finally 22 aro-
matizes to give 18 with a migration of the tert-butyl
group. On the other hand, elimination of HCl from
17, with simultaneous four-membered ring forma-
tion, would provide 19. The elimination of HCl from
SCHEME 4
The formation of 19 requires elimination of HCl in
addition to SO2. We therefore examined thermolysis
of 16 in the presence of a base to promote elimina-
tion of HCl. Indeed, the thermolysis in the presence
of triethylamine or N-ethyldiisopropylamine in re-
fluxing chlorobenzene furnished 19 as the sole prod-
uct in 98% or 85% yield, respectively.
The structure of 18 could not be determined un-
ambiguously by 1H and 13C NMR analyses including
NOE experiments; particularly the isomeric struc-
ture 20 was not ruled out. Therefore, the structure
was determined by X-ray crystallographic analysis
(Fig. 1). The structure of 19 was determined by NMR
analyses; single crystals suitable for X-ray crystallo-
graphic analysis could not be obtained despite many
efforts. The 1H NMR spectrum (C6D6 as the solvent)
TABLE 1 Results of the Thermolysis of 16 in Varying
Conditionsa
Yields (%)b
Solvent
Base
Reaction Time (h) 18
19
PhCl
6
8
16
6
23
22
49
55
55
24
Xylene
Octane
PhCNc
PhCl
d
d
+
+
Et3N
6
0
0
98
85
PhCl
iso-Pr2NEt
6
aAll the reactions were carried out at reflux unless otherwise stated.
bYields based on the isolated products.
cAt 130◦C.
dComplex mixture containing 18 and 19.

SO2-Extrusion of an 8-Thiabicylo[3.2.1]octa-2,6-diene 8,8-dioxide and Rearrangement of the Resulting Cycloheptatriene 135
duced pressure. The residue was chromatographed
on a column of silica gel. Elution of the column with
hexane gave 116 mg of a 1:1 mixture of 13 and 14
(each 27% yield). 1H and 13C NMR spectra of 13 and
14 agreed with those of authentic samples [2a]. Fur-
ther elution of the column with CH2Cl2 gave 172 mg
1
(44%) of 10: mp 146–147◦C; H NMR (400 MHz,
CDCl3) δ 1.34 (s, 9H), 1.40 (s, 9H), 4.34 (d, J = 2.2 Hz,
1H), 4.81 (d, J = 2.2 Hz, 1H); 13C NMR (100.6 MHz,
CDCl3) δ 31.8, 33.0, 34.2, 36.2, 68.0, 72.0, 83.3, 129.1,
132.4, 142.5, 150.6; IR (KBr) 2977, 1613, 1484, 1469,
1397, 1367, 1187, 1108 (S O), 1072, 1045, 1015, 965,
807, 725, 654, 640, 546 cm−1; MS (EI, 70 eV) m/z 388
(M+), 305, 263. Anal. Calcd for C15H20Cl4OS: C, 46.17;
H, 5.16. Found: C, 46.33; H, 5.04.
Heating a mixture of 100 mg (0.47 mmol) of 8
and 167 mg (0.94 mmol) of 6 in 5 mL of toluene
for 20 h gave 59 mg of a mixture of 13 (22%) and
14 (36%), 65 mg (36%) of 10, and 42 mg of (Z)-2,3-
dichloropropenoic acid (15) [11].
6,7-Di-tert-butyl-2,3,4,4-tetrachloro-8-
thiabicylo[3.2.1]octa-2,6-diene 8,8-dioxide (16)
Oxidation with MCPBA. A mixture of 100 mg
(0.26 mmol) of 10 and 46 mg (0.27 mmol) of MCPBA
in 5 mL of CH2Cl2 was stirred for 5.5 h at room
temperature. The mixture was washed with aqueous
NaHSO3, aqueous NaHCO3, and water successively,
dried over MgSO4, and evaporated to give 104 mg
SCHEME 5
(100%) of practically pure 16: mp 170–171◦C (dec)
1
the cycloheptatrienyl (tropylium) cation 23 (ionized
aromatic isomer of 17) would be least probable be-
cause 19 is formed even for the thermolysis in oc-
tane, a typical nonpolar solvent, where the forma-
tion of 23 is least possible. Decreased yield of 19
(increased yield of 18) in octane indicates that the
four-membered ring formation takes place through
a polarized transition state. The sole formation of
19 in the presence of the base suggests that the base
promotes the elimination of HCl from 17 (see 24 in
Scheme 5). Incidentally, attempted conversion of 19
to the cycloheptatrienyl salt 25 by treatment with
trityl tetrafluoroborate was unsuccessful because of
electronic and/or steric effects.
(from hexane/CH2Cl2); H NMR (400 MHz, CDCl3)
δ 1.37 (s, 9H), 1.47 (s, 9H), 4.41 (d, J = 3.3 Hz,
1H), 4.99 (d, J = 3.3 Hz, 1H); 13C NMR (100.6 MHz,
CDCl3) δ 31.3, 32.7, 35.0, 36.9, 71.7, 77.7, 85.3, 130.5,
131.6, 144.8, 152.5; IR (KBr) 3027, 1597, 1481, 1367,
1334 (SO2), 1203, 1180, 1129 (SO2), 1063, 1009, 954,
862, 814, 759, 639, 578, 458 cm−1; MS (EI, 70 eV)
m/z 404 (M+), 305. Anal. Calcd for C15H20Cl4O2S: C,
46.17; H, 5.16. Found: C, 46.11; H, 4.97.
Oxidation with Dimethyldioxirane. Oxidation of
50 mg (0.13 mmol) of 10 dissolved in 1 mL of ether
was carried out by adding a 0.54 mM solution of
dimethyldioxirane in acetone at 0◦C. The oxidation
was sluggish and the oxidant was added several times
at 10 h-intervals until 10 is completely consumed. A
total of ca. 1 mmol of the oxidant was used for com-
pletion of the oxidation to provide 52 mg (100%) of
practically pure 16.
EXPERIMENTAL
Reaction of 3,4-Di-tert-butylthiophene 1-oxide
(8) with Tetrachlorocyclopropene(6);
6,7-Di-tert-butyl-2,3,4,4-tetrachloro-8-
thiabicylo[3.2.1]octa-2,6-diene 8-oxide (10)
Thermolysis of 16
A mixture of 212 mg (1 mmol) of 8 and 174 mg
(1 mmol) of 6 [10] in 5 mL of toluene was heated at
reflux for 20 h. The mixture was evaporated under re-
Thermolysis in Refluxing Chlorobenzene. A solu-
tion of 50 mg (0.12 mmol) of 16 in 5 mL of chloroben-
zene was heated at reflux for 6 h. The reaction

136 Takayama, Sugihara, and Nakayama
mixture was evaporated under reduced pressure, and
the residue was chromatographed on a column of
silica gel. Elution of the column with hexane gave
21 mg (55%) of 19 and 9.5 mg (23%) of 18 in this
order.
X-Ray Crystal Structure Determination of 18
The crystal data for 18 were recorded on a Bruker
SMART APEX CCD area detector by using 0.30◦-wide
ω scans and graphite-monochromated Mo Kꢁ radia-
◦
˚
tion (λ = 0.71073 A). Frame data (20 s, 0.30 -wide
ω scans) were collected using the Bruker SMART
software package [12]. Peak integration was per-
formed by the Bruker SAINT-Plus software package
[13]. Absorption correction was made by the soft-
ware SADABS [14]. Space group determination was
done by the software XPREP [15]. All calculations
were performed by the Bruker SHELXTL 5.1 soft-
ware package [16]. The structure was solved by direct
methods and refined with full-matrix least squares by
all independent reflections. The non-hydrogen atoms
were refined anisotropically, and hydrogen atoms
were placed at calculated positions.
5-tert-Butyl-1,2-dichloro-3-[(1,1-dichloro-2,2-
dimethyl)propyl]benzene (18): mp 89–90◦C (from
1
MeOH); H NMR (400 MHz, CDCl3) δ 1.26 (s, 9H),
1.31 (s, 9H), 7.51 (d, J = 2.3 Hz, 1H), 8.02 (d, J = 2.3
Hz, 1H); 13C NMR (100.6 MHz, CDCl3) δ 27.0, 30.9,
34.8, 47.8, 103.3, 128.5, 130.3, 135.4, 137.5, 148.4;
IR (KBr) 2964, 1479, 1465, 1396, 1385, 1364, 1279,
1205, 1170, 1050, 877, 804, 692 cm−1; MS (EI, 70 eV)
m/z 340 (M+), 283. Anal. Calcd for C15H20Cl4: C,
52.66; H, 5.89. Found: C, 52.70; H, 5.90.
2-tert-Butyl-4,5,6-trichloro-9,9-dimethylbicyclo-
[5.2.0]nona-1,3,5-triene (19): colorless oil; solidified
when kept in a refrigerator for a long time of period;
1
mp 106–107◦C (from MeOH); H NMR (400 MHz,
CDCl3) δ 0.96 (s, 9H), 1.13 (s, 3H), 1.41 (s, 3H),
2.01–2.10 (m, 2H), 3.31 (dd, J = 7.8, 4.7 Hz, 1H),
X-Ray Crystallographic Data of 18. C15H20Cl4,
Mw = 342.13, Monoclinic, P21/c, a = 9.5912(6)
1
˚
˚
˚
7.09 (s, 1H); H NMR (400 MHz, C6D6) δ 0.78 (s,
A, b = 17.5121(10) A, c = 10.5039(6) A, β =
◦
3
˚
3H), 0.84 (s, 9H), 1.07 (s, 3H), 1.63–1.73 (m, 2H),
3.13 (dd, J = 8.8, 3.8 Hz, 1H), 6.91 (s, 1H); 1H NMR
(400 MHz, CD3CN) δ 0.95 (s, 9H), 1.12 (s, 3H), 1.42
(s, 3H), 2.04–2.12 (m, 2H), 3.33 (dd, J = 7.0, 5.6 Hz,
1H), 7.29 (s, 1H); 13C NMR (100.6 MHz, CDCl3) δ
28.5, 29.7, 35.4, 36.9, 42.4, 43.5, 54.2, 122.6, 129.2,
131.4, 132.4, 142.7, 154.6; 13C NMR (100.6 MHz,
C6D6) δ 28.3, 29.8, 35.0, 36.8, 42.5, 43.4, 54.4, 122.9,
130.0, 132.0, 133.2, 143.0, 154.9; IR (neat) 2958,
2871, 1735, 1576, 1545, 1479, 1466, 1418, 1396,
1385, 1363, 1310, 1286, 1234, 1218, 1167, 1111, 868,
821, 765, 759, 701 cm−1. Anal. Calcd for C15H19Cl3:
C, 58.94; H, 6.27. Found: C, 59.11; H, 6.28.
107.8340(10) , V = 1679.48(17) A , Z = 4, Dc = 1.353
g/cm3, number of measured reflections 8453,
number of independent reflections 3116, number
of reflections with I > 2ꢂ(I) 1904, parameters
178, R = 0.0519(I > 2ꢂ(I)), wR = 0.1478 (all),
1
2
S = 0.998, T = 298 K.
REFERENCES
[1] (a) Reinhoudt, D. N.; Smael, P.; van Tilborg, W. J. M.;
Visser, J. P. Tetrahedron Lett 1973, 14, 3755; (b)
van Tilborg, W. J. M.; Smael, P.; Visser, J. P.;
Kouwenhoven, C. G.; Reinhoudt, D. N. Rec Trav Chim
Pays-Bas 1975, 94, 85; (c) Takeuchi, K.; Yokomichi, Y.;
Kurosaki, T.; Kiumra, Y.; Okamoto, K. Tetrahedron
1979, 35, 949; (d) Mu¨ ller, P.; Schaller, J.-P. Helv Chim
Acta 1989, 72, 1608.
[2] (a) Nakayama, J.; Yamaoka, S.; Hoshino, M. Tetrahe-
dron Lett 1988, 29, 1161; (b) Nakayama, J.; Hasemi,
R.; Yoshimura, K.; Sugihara, Y.; Yamaoka, S.;
Nakamura, N. J Org Chem 1998, 63, 4912.
Thermolysis in Other Solvents. The thermolysis
of 50 mg of 16 in refluxing xylene (5 mL) gave 9 mg
(22%) of 18 and 21 mg (55%) of 19, that in octane
(2 mL) gave 21 mg (49%) of 18 and 9 mg (24%) of
19, and that in benzonitrile at 130◦C gave a complex
mixture containing 18 and 19.
[3] For reviews on thiophene 1-oxides, see (a) Nakayama,
J.; Sugihara, Y. Sulfur Rep 1997, 19, 349; (b)
Nakayama, J. Sulfur Rep 2000, 22, 123.
Thermolysis of 16 in the Presence of a Base. A
solution of 100 mg (0.25 mmol) of 16 and 51 mg
(0.50 mmol) of triethylamine in 5 mL of chloroben-
zene was heated at reflux for 6 h. The mixture was
evaporated, the residue was chromatographed on
a column of silica gel, and the column was eluted
with hexane to give 74 mg (98%) of 19. Similarly
heating a solution of 100 mg (0.25 mmol) of 16
and 159 mg (1.2 mmol) of N-ethyldiisopropylamine
in 5 mL of chlorobenzene gave 64 mg (85%)
of 19.
[4] Nakayama, J.; Yu, T.; Sugihara, Y.; Ishii, A. Chem Lett
1997, 499.
[5] For Diels–Alder reactions of 8 and the stereochem-
istry: (a) Otani, T.; Takayama, J.; Sugihara, Y.; Ishii,
A.; Nakayama, J. J Am Chem Soc 2003, 125, 8255;
(b) Takayama, J.; Fukuda, S.; Sugihara, Y.; Ishii,
A.; Nakayama, J. Tetrahedron Lett 2003, 44, 5159;
(c) Nakayama, J. J Synth Org Chem Jpn 2003, 61,
1106.
[6] Seitz G.; Van Gemmern, R. Chemiker-Zeitung 1987,
111, 209. See also Jefford, C. W.; Gunsher, J.; Hill,
D. T.; Brun P.; Gras J.-L.; Waegell B. Org Synth 1988,
Coll Vol 6, 142 and references cited therein.

SO2-Extrusion of an 8-Thiabicylo[3.2.1]octa-2,6-diene 8,8-dioxide and Rearrangement of the Resulting Cycloheptatriene 137
[7] λmax 260 nm for the parent cycloheptatriene: Dryden,
H. L., Jr., J Am Chem Soc 1954, 76, 2841. λmax
265 nm for a ca. 1:4 mixture of 2-bromo- and 3-
bromocycloheptatrienes and λmax 287 nm for 1-
methoxycycloheptariene: Lindsay, D. G.; Reese, C. B.
Tetrahedron 1965, 21, 1673.
[8] Formation of 1,3-dienes by extrusion of SO2 of
2,5-dihydrothiophene 1,1-dioxides and related com-
pounds, see Nakayama, J.; Machida, H.; Saito, R.;
Akimoto, K.; Hoshino, M. Chem Lett 1985, 1173 and
references cited therein.
Roberts, J. D. J Am Chem Soc 1971, 93, 2203; (b) Du¨rr,
H.; Kober, H. Angew Chem, Int Ed 1971, 10, 342.
[10] Tetrachlorocyclopropene was purchased from
Aldrich and used without further purification.
[11] (a) Tobey, S. W.; West, R. J Am Chem Soc 1964, 86,
56; (b) Kurtz, A. N.; Billups, W. E.; Greenlee, R. B.;
Hamil, H. F.; Pace, W. T. J Org Chem 1965, 30, 3141.
[12] Bruker, SMART version 5.630, Bruker AXS Inc.,
Madison, WI, USA (2003).
[13] Bruker, SAINT-Plus version 6.45, Bruker AXS Inc.,
Madison, WI, USA (2003).
[9] For the cycloheptatriene-norcaradiene equilibrium,
see le Noble, W. J. Highlights of Organic Chem-
istry; Marcel Dekker: New York, 1974, pp. 402–406
and references cited therein. Reportedly, the electron-
withdrawing substituents at the C7 shift the equilib-
rium toward the norcaradiene isomer: (a) Hall, G. E.;
[14] Sheldrick, G. M. SADABS version 2.05, Bruker AXS
Inc., Madison, WI, USA (2002).
[15] Sheldrick, G. M. XPREP version 6.12, Bruker AXS
Inc., Madison, WI, USA (2003).
[16] Sheldrick, G. M. SHELXTL version 6.14, Bruker AXS
Inc., Madison, WI, USA (2003).