Synthesis and properties of the first Moebius annulenes

Article abstract of DOI:10.1002/chem.200600215

Heilbronner in 1964 predicted that annulenes with "... a planar perimeter of N = 4r AO's. which would yield an open shell configuration when occupied by 4r electrons, can be twisted into a closed shell Moebius strip perimeter without loss in π electron energy". We have been able to synthesize the first [4n]annulene with such a Moebius topology and now present further Moebius isomers and the details of their preparation as stable compounds. To address the question whether the twist in the π system has an effect on the properties we systematically investigate energy, geometry and magnetic parameters of a large number isomers of [16]annulenes. The Moebius twisted annulenes are consistently more aromatic than the non-twisted isomers. This is true for the parent as well as our benzoannelated systems. Our results are in contrast to those published recently by C. Castro, W. L. Karney, P. von R. Schleyer et al.

Full text of DOI:10.1002/chem.200600215

DOI: 10.1002/chem.200600215
Synthesis and Properties of the First Mçbius Annulenes
[
a]
[a]
[a]
[b]
[c]
Dariush Ajami, Kirsten Hess, Felix Kçhler, Christian Näther, Oliver Oeckler,
[
c]
[d]
[d]
[a]
Arndt Simon, Chiyo Yamamoto, Yoshio Okamoto, and Rainer Herges*
Dedicated to Professor Edgar Heilbronner on the occasion of his 85th birthday
Abstract: Heilbronner in 1964 predict-
ed that annulenes with “… a planar pe-
rimeter of N=4r AOꢀs, which would
yield an open shell configuration when
occupied by 4r electrons, can be twist-
ed into a closed shell Mçbius strip pe-
rimeter without loss in p electron
energy”. We have been able to synthe-
size the first [4n]annulene with such a
Mçbius topology and now present fur-
ther Mçbius isomers and the details of
their preparation as stable compounds.
To address the question whether the
twist in the p system has an effect on
the properties we systematically inves-
tigate energy, geometry and magnetic
parameters of a large number isomers
of [16]annulenes. The Mçbius twisted
annulenes are consistently more aro-
matic than the non-twisted isomers.
This is true for the parent as well as
our benzoannelated systems. Our re-
sults are in contrast to those published
recently by C. Castro, W. L. Karney,
P. von R. Schleyer et al.
Keywords: annulenes · aromaticity ·
metathesis · Mçbius twist · photo-
chemistry
Introduction
is named. A more general property than the “one-sided-
ness” is the non-orientablity of this topological object. In
contrast to the “sidedness”, orientability is an intrinsic prop-
erty and more generally applicable (also in higher dimen-
sions). Consider the structure of l-lactic acid drawn on a
(transparent) Mçbius band. If you move it around the
Mçbius band it will return as its mirror image (d-lactic acid)
and upside down. The Mçbius band therefore is said to be
non-orientable. The fascinating properties of the Mçbius
topology have inspired mathematicians, engineers, and ar-
tists. Heilbronner roused the interest in chemistry in 1964
when he predicted that Mçbius-like twisted annulenes with
Most surfaces in every day life are two-sided, for example, a
round sphere or a strip of paper. There is an inside and out-
side or back and front that can be stained in two different
colors. A Mçbius band that can be created by twisting a rec-
tangular strip by 1808 before affixing both ends has only one
side and one edge. The mathematical properties of Mçbius
objects were first described and published by Johann Bene-
dict Listing in 1858, two months before August Ferdinand
Mçbius after whom the two-dimensional one-sided topology
[1]
4
n electrons should be closed-shell species. By replacing
the resonance integral b by bcos(p/N) (N=number of ring
AHCTREUNG
[
a] Dr. D. Ajami, K. Hess, F. Kçhler, Prof. Dr. R. Herges
Institut für Organische Chemie, Universität Kiel
Otto-Hahn-Platz 4, 24118Kiel (Germany)
Fax : (+49)431-880-1558
atoms) to account for the twist, he obtained an analytical
solution of the secular Hückel equations that contains exclu-
sively degenerate energy eigenvalues (except the highest un-
occupied orbital in odd-membered annulenes). Coinciden-
tally, the energy levels of a Mçbius annulene correspond to
the vertical positions of the N vertices of a regular polygon
drawn in such a way that the lowest edge (bond) is horizon-
tal. If the polygon representing the [n]annulene is inscribed
in a circle with a diameter of 4b and a center at a (Coulomb
integral, origin of the energy axis), the vertical positions of
the vertices are identical with the Hückel MO energies of
the corresponding annulene. A similar procedure is known
as the Frost–Musulin mnemonic for determining the MO en-
E-mail: rherges@oc.uni-kiel.de
[
[
b] Dr. C. Näther
Institut für Anorganische Chemie, Universität Kiel
Otto-Hahn-Platz 6/7, 24118Kiel (Germany)
[
c] Dr. O. Oeckler, Prof. Dr. A. Simon
Max-Planck-Institut für Festkçrperforschung
Heisenbergstrasse 1, 70569 Stuttgart (Germany)
d] Dr. C. Yamamoto, Dr. Y. Okamoto
Department of Applied Chemistry
Graduate School of Engineering, Nagoya University
Chikusa-ku, Nagoya 464-8603 (Japan)
Supporting information for this article is available on the WWW
under http://www.chemistry.org or from the author.
[2]
ergies of Hückel annulenes. In this case the polygon has to
5434
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5434 – 5445

FULL PAPER
be drawn in such a way that a vertex (atom) is the lowest
point in the sphere (Figure 1).
to be found by trial and error. The smallest conceivable
system of this kind is trans benzene. There are a number of
calculations which all agree that this highly strained struc-
ture is only a shallow minimum on the energy hypersur-
[18,19]
À1
face.
The trans benzene structure is about 100 kcalmol
less stable than D6h benzene, and the dihedral angles of the
trans double bond with the neighboring cis bonds are close
to 908. Therefore it is rather nonaromatic. Similar arguments
[20]
hold for trans cyclooctatetraene.
Karney, and Schleyer calculated a number of isomers of
Recently, Castro,
[
n]annulenes and identified several structures as Mçbius ar-
[21]
omatic.
Although many attempts within the last 40 years have
[22]
been made,
so far no stable aromatic Mçbius annulene
could be synthesized. So one might ask what the problems
are that antagonize the experimental verification. Unfortu-
nately Heilbronnerꢀs statement that [n]annulenes of a ring
size n<20 should not be strained was not confirmed by den-
sity functional theory calculations. In Table 1 the relative en-
Figure 1. Frost–Musulin mnemonic to determine the MO energies for
Hückel and Mçbius annulenes.
Table 1. Relative energy (B3LYP/6-31G*) of the most stable Hückel and
the most stable Mçbius isomer of several [n]annulenes.
[
n]annulene
Heilbronner suggested that Mçbius twisted [n]annulenes
with n>20 should be formed without much strain or sterical
hindrance. As this is often the case in natural sciences, the
applications of Heilbronnerꢀs prediction came from an unex-
n=6
n=8
n=16
n=20
most stable
Hückel
isomer
[3–5]
pected field. Only two years later in 1966 Zimmerman
proposed an alternative version of the Woodward–Hoffman
rules based on the Hückel–Mçbius aromaticity of transition
states. From 1964 to date more than 200 papers were pub-
lished that predict the properties of hypothetical Mçbius an-
most stable
Mçbius
isomer
[6,7]
nulenes, such as Mçbius coronenes and kekulenes,
[
8–10]
[11]
Mçbius cyclacenes,
spiroaromatic systems,
heteroan-
[12]
[13]
nulenes, and cyclic allenes.
energy dif-
107.0
21.3
5.1
6.2
ference
À1
[
kcalmol
]
ergies of the most stable Hückel and the most stable
[20]
Mçbius [n]annulenes for n=6 (benzene), n=8(cycloocta-
[20]
[21]
[21]
tetraene), n=16 and n=20 are listed. In all cases the
most stable isomer has Hückel topology. Evidently, the sta-
bilization by Mçbius aromaticity cannot overcome the de-
stabilization due to the torsional strain or the reduced over-
lap of neighboring p orbitals. Even in the [20]annulene
The first proposal of a cationic Mçbius aromatic system
[14]
À1
was published by E. Yakali in her thesis in 1973. She sug-
gested that the solvolysis of 9-chlorobicyclo-
[6.1.0]nonatriene would proceed via a Mçbius [9]annulene
system, the most stable Hückel isomer is 6.2 kcalmol more
stable than the most stable Mçbius structure.
[15]
A
C
H
T
R
E
U
N
G
In principle, if the Mçbius isomers would be kinetically
stable there would still be the chance to characterize such a
compound as an intermediate. Unfortunately, however, the
[n]annulenes with n=4r (r=4, 5, 6 …) delocalized electrons
are extremely “floppy”. Barriers to cis–trans and bond shift
+
cation (CH)9 . In 1998Schleyer et al. presented theoretical
calculations providing strong evidence that this short-lived
[16]
intermediate is indeed Mçbius aromatic. The first twisted
neutral parent annulene a Mçbius [16]annulene was pro-
[17]
[23]
posed by Rzepa et al. The “trick” to enforce a twist in a
cyclic conjugated system is to introduce trans double bonds
at suitable positions. Unfortunately, however, so far the
number of transoid bonds and the “suitable positions” have
isomerization are very low. For example, [16]annulene in
solution is an equilibrium of at least 40 isomers that rapidly
[
24–28]
interconvert at temperatures as low as À1508C.
The
larger the annulene, the more isomers exist and the lower
Chem. Eur. J. 2006, 12, 5434 – 5445
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5435

R. Herges et al.
the isomerization barriers are. Therefore a parent neutral
annulene most probably will never be synthesized.
In a recent communication we reported the synthesis of
[29]
the first [4n]annulene with a Mçbius topology
and now
present further Mçbius isomers and the details of their prep-
aration as stable compounds. Our results are in contrast to
those published recently by C. Castro, W. L. Karney,
[30]
P. von R. Schleyer et al.
Strategy to stabilize the twist: To overcome these difficulties
we developed a strategy to stabilize the 1808 twist in cyclic
conjugated molecules. Mçbius annulenes exhibit two types
of conjugation: 1) “normal” conjugation with p orbitals per-
2
pendicular to the ring plane and sp carbon atoms in their
preferred trigonal planar configuration (Figure 2, top, blue);
Figure 3. Top: Metathesis strategy to combine a “normal” and an “in-
plane”-conjugated unit to form a Mçbius ring. Bottom: Synthetic realiza-
tion using an [n]annulene as the “normal” conjugated unit, and TDDA
as the “in-plane”-conjugated component.
2) “in-plane” conjugation with pyramidalized carbon atoms
(
Figure 2, top, red). Beside carbon nanotubes, there are only
midalized part, and cyclooctatetraene as the normal p
system.
The ring-enlargement metathesis product contains a bian-
thraquinodimethane unit (red) and a polyene bridge (blue).
The bianthraquinodimethane unit is kept in the pyramidal-
ized configuration by the steric hindrance of the inner
ortho-hydrogen atoms, which prevent a planar geometry and
hence correspond to the stick or thread in our paper model.
To make sure that our simple paper model holds at higher
levels of theory we performed DFT calculations. There are
Figure 2. Our strategy to stabilize the twist in a ring.
1
2 E/Z isomers of our Mçbius stabilized [16]annulene, each
of which can adopt nine s-cis/s-trans conformations. We cal-
culated all 108isomers at the semiempirical PM3 level and
reoptimized the most stable conformations of each configu-
ration at the B3LYP/6-31G* level of theory. In Table 2 the
25 most stable isomers of the parent [16]annulene (see Com-
putational Methods) are compared with the 25 most stable
isomers of our bianthraquinodimethane-modified system. To
denominate the structures we introduce the following ste-
very few examples of experimentally known, in-plane conju-
gated structures, because the pyramidalization of sp -hybri-
2
dized carbon atoms involves a considerable amount of
strain. The strain in Mçbius annulenes almost exclusively
arises from the pyramidalized (red) part in the ring.
Consequently, our strategy is based on the idea to use a
prefabricated pyramidalized building block that is kept in its
strained configuration by a rigid molecular frame and com-
bine it with a “normal” p system. The underlying principle
can be easily demonstrated using a simple paper model.
A
H
R
U
G
is rigid and thus the corresponding double bonds donꢀt have
to be included in the stereochemical denotation. The two H
atoms at the quinoid double bonds are either syn or anti, c
and t denote the s-cis or s-trans
(
Figure 2, bottom) Bend a rectangular piece of cardboard to
a half-open cylinder (pyramidalized part) and fix it in this
position using a stick or thread connecting both ends (mo-
lecular frame). Cut a disk with a large hole in the middle
relationship
between
two
double bonds, and Z and E
define the stereochemistry at
the three central double bonds
in the bridge. In Scheme 1 the
nomenclature is exemplified
with the syn-tZcZcEt isomer.
As inferred from our simple
model the most stable isomer
of our bianthraquinodimethane
(
“normal” p system) and cut once through the resulting
ring. Now try to attach both ends of the half-cylinder with
the ends of the dissected disk. There are two ways to form a
ring with both components: a 1808 twisted Mçbius band and
a “normal” Hückel band. As you immediately notice during
the experiment, the Mçbius model is less strained than the
Hückel band and therefore should be more stable. How to
translate our paper model into chemistry?
We chose the ring enlargement metathesis reaction as the
method to combine both parts of the ring (Figure 3), tetra-
dehydrodianthracene (TDDA, bottom, in red) as the pyra-
Scheme 1. Notation of the ste-
reoisomers of the bianthraqui-
nodimethane-stabilized [16]an-
modified
Mçbius
[16]annulene
topology
has
(syn-
AHCTREUNG
tZcZcEt). There are another six nulene.
5436
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5434 – 5445

Mçbius Annulenes
FULL PAPER
Table 2. B3LYP/6-31G* and KMLYP/6-31G* calculated “energy ranking list” of the most stable isomers of the parent [16]annulene and the bianthraqui-
À1
nodimethane-modified system. Relative stabilities (Erel) in kcalmol . Note that the structures of the parent [16]annulene and those of the bianthraquino-
dimethane-modified system with the same entry number are not structurally related.
Parent [16]annulene
Topology
Bianthraquinodimethane modified [16]annulene
[
a]
[b]
[a]
[c]
[d]
[e]
Entry
E
rel
Configuration
Entry
E
rel
E
rel
Topology
Configuration
1
2
3
4
5
6
7
0.00
2.25
4.68Hückel
5.14
5.15
5.41
5.41
Hückel
5.45
5.47
5.83
5.84
6.03
6.05
6.22
6.24
6.37
Hückel
Hückel
0011001100110011
0011011011001111
1
2
0.00
0.29
0.00
1.75
Mçbius
Mçbius
syn-tZcZcEt
anti-tEcZcEt
syn-cZtEcZt
syn-cEtZcZt
anti-cEtZcEt
anti-tZtEcZt
anti-tZcZcZt
syn-tEcZcEt
syn-tZcZcEc
syn-tZtZtZt
anti-tEcEtZc
syn-tEtZcZc
anti-cEtZtEc
anti-tZtZcZc
anti-cZcZxZc
syn-cEtZtEc
syn-cZtZcZc
syn-cZcZtZc
syn-cZtEcEc
syn-cEcZcZc
anti-cZcEcEc
anti-tEtEtEt
syn-tEtEcEt
anti-tEcEcEt
anti-cEcZcEt
8
9a
0001100110011011
3
0000110011000111
0000111000110011
0001100110011001
0001001100110011
0.56
0.30
Mçbius
Mçbius
Mçbius
Mçbius
Mçbius
4
5
6
7
0.56
2.47
2.76
2.77
Hückel
6.71
6.91
8.36
9.30
9.34
0.89
4.95
1.96
1.90
Mçbius
Mçbius
Mçbius
Mçbius
5
6
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
5.45
0011011011011011
84.16
0011011001101111
0011011001111011
0001100110011011
0001101100110011
0001101100110101
0001010110011011
0001001100110011
0011001111001111
0001111001100101
1812 9. 3. 58 3
0001011100011011
0001101100010111
0010100110110011
0010100110011011
0001101100011111
0001101100011011
0001011000110011
3.26
Hückel
Hückel
Hückel
Hückel
Hückel
Hückel
Mçbius
Hückel
Hückel
9
10
11
12
13
14
15
16
17
5.23
5.29
10.74
8.86
13.06
10.25
10.25
11.56
Mçbius
Hückel
Mçbius
Mçbius
Mçbius
Mçbius
Mçbius
Hückel
0
1
2
3
4
5
6
7
7
9b
11.20
11.20
11.74
12.389.53
Mçbius
86. H3 8ückel
0001010011001111
Hückel
Mçbius
19
9
0
1
2
3
4
5
6.84
6.85
6.87
6.88
7.09
7.30
7.30
14.25
16.20
23.49
23.80
27.10
28.27
30.73
13.96
15.00
24.35
27.58
33.46
33.01
34.84
Mçbius
Mçbius
Mçbius
Hückel
Hückel
Hückel
Mçbius
Hückel
Hückel
Hückel
Mçbius
Hückel
Mçbius
20
21
22
23
24
25
[
a] B3LYP/6-31G*. [b] 0=Z, 1=E. [c] KMLYP/6-31G*. [d] For notation see Scheme 1. [e] Compound number.
À1
submitted to PM3 single point calculations and those within an energy
threshold of 20 kcalmol were optimized at the B3LYP/6-31G* level of
theory and finally again checked for redundancy.
Mçbius isomers within a relative energy of 3 kcalmol . The
À1
most stable Hückel isomer (syn-tZtZtZt) has a relative
energy of 4.2 kcalmol , which should be formed in very
À1
small amounts in a thermodynamically controlled reaction.
Thus, based on our theoretical calculations the insertion of a
bianthraquinodimethane unit into a [16]annulene stabilizes
the Mçbius twist to such an extent that the synthesis of a
Mçbius annulene was expected to be within reach.
Results and Discussion
Experiments: Upon irradiation TDDA (tetradehydrodian-
thracene) reacts with alkenes, cyclic alkenes, and even ben-
[40–42]
zene in ring enlargement metathesis reactions.
Unfortu-
nately TDDA and cyclooctatetraene (COT) upon irradiation
in benzene with a 700 W high pressure mercury lamp in a
quartz apparatus did not undergo addition reactions, but fur-
nished bianthryl as the main product. We know from inde-
pendent investigations with triplet sensitizers (acetophenone
and acetone) that the triplet-excited state of TDDA imme-
diately undergoes electrocyclic ring-opening to form 9,10’-
didehydroanthracene, which abstracts hydrogen or deuteri-
Computational Methods
[
31]
All calculations were performed by using the Gaussian 03 program.
Geometries were optimized at the B3LYP/6-31G* level of DFT.
[
32, 33]
The
bianthraquinodimethane-modified [16]annulenes were additionally opti-
[
34]
mized at the KMLYP/6-31G* level of DFT. Magnetic susceptibilities
[
35]
were calculated by using the CSGT method and nucleus-independent
chemical shift (NICS) values were determined by using the gauge-invari-
[
36]
ant atomic orbital (GIAO) method. Harmonic oscillator measure of ar-
omaticity (HOMA) values were calculated by averaging of all four cyclic
[43]
um from the solvent (e.g. [D ]acetone). This is in agree-
6
[
16]annulene pathways through the bianthraquinodimethane unit. The
ment with DFT calculations. At the UB3LYP/6-31G* level
of DFT the lowest triplet state of TDDA is not even a mini-
mum, but undergoes an electrocyclic ring-opening without
ring center for calculating the NICS correspondingly was defined as the
average center of gravity of the C atoms in each path. The topology of
the rings (Hückel or Mçbius) was determined by following the normal
vectors on the planes defined by each successive triplet of carbon atoms
around the ring. The isomers of the [16]annulenes were generated by
using a Monte Carlo algorithm, through breaking the ring, varying the di-
hedral angles in steps of 308, and rejoining both ends. The structures
were submitted to an optimization using the MM+
then checked for redundancy. The generation procedure was stopped
when no new isomers were found by doubling the number of generated
structures. All structures within an energy threshold of 60 kcalmol were
[44]
barrier. COT is known to be a triplet quencher
triplet energy of 41 kcalmol , but probably still able to sen-
sitize TDDA.
with a
À1
[
37]
To circumvent this problem we used syn-tricyclooctadiene
[
38,39]
force field and
(
syn-TCOD) as a synthetic analogue for COT. Syn-TCOD is
difficult to prepare in preparative amounts. The reaction of
syn-dichlorocyclobutene with sodium amalgam in diethyl
À1
Chem. Eur. J. 2006, 12, 5434 – 5445
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5437

R. Herges et al.
ether is not complete even after reaction times of more than
several weeks and the reaction mixture has to be purified by
[45,46]
preparative GC.
phoramide (HMPTA) we used an ultra sound generator
cell disruptor) to accelerate the heterogeneous reaction and
To avoid the use of hexamethylphos-
[47]
(
were able to reduce the reaction time to three days. TDDA
was irradiated with a tenfold excess of syn-TCOD in ben-
zene with a high-pressure mercury lamp for 20 min in a
quartz photoreactor. Four products were isolated and char-
acterized: the metathesis product with the solvent benzene
[
40]
(
1),
two metathesis products with syn-TCOD (2 and 3)
and a ladderane (4). Compounds 2 and 3 were crystallized
and their structures were unambiguously identified by NMR
and X-ray analysis as the C and C symmetrical isomers of
s
2
Scheme 2. Reaction of TDDA with syn-TCOD upon irradiation with a
low-pressure mercury lamp.
1,3-cyclohexadiene products. In 2, two crystallographically
independent molecules are found, of which one is disor-
dered. However, in one of the two independent molecules
the configuration could be unambiguously elucidated. The
X-ray data are also in agreement with NOE spectra. There
is a distinct NOE relationship of protons H and H in 3 and
Beside the products identified after longer wavelength ir-
radiation, we also isolated five ring-opened [16]annulene
compounds (1:1 addition products of TDDA and syn-TCOD
a
b
no correlation in 2 (Figure 4). Unfortunately we were not
1
according to MS and fully conjugated according to H and
1
3
C NMR spectroscopy). The separation was achieved after
removing the 1,3-cyclohexadiene structures 2 and 3 by
Diels–Alder reaction with tetracyanoethylene (TCNE) and
subsequent HPLC on a silica gel column. Three of these iso-
mers were crystallized and their structure was determined
by X-ray analysis.
The structures 5 (anti-tZtEcZt), 6 (anti-tZcZcZt) and 7
(
syn-tZtZtZt) are identical (Figure 5) with the entries 6, 7,
and 10 in the list of the most stable isomers (Table 2) that
had already been calculated at the B3LYP/6-31G* level of
density functional theory. Structure 5 has only approximate
Mçbius topology. One of the torsional angles in the polyene
bridge deviates with 124.48 (DFT) or 112.18 (X-ray) from
planarity. The structure has no element of symmetry and
therefore is chiral. Both enantiomers were found in the unit
cell. Compound 5 is stable in dichloromethane and can be
exposed to diffuse daylight at room temperature for several
months without decomposition.
Structure 6 has a perfect Mçbius topology. Both ends of
the bianthraquinodimethane unit are connected by a poly-
ene bridge (similar to the model in Figure 2) in such a way
that the inner side of the anthracene units is connected to
the outer surface and vice versa. The implications on the
stereochemistry are interesting. Any attempt to determine
the prochirality (Re, Si) of the central double bond in the
Figure 4. Crystal structures of compounds 2 (left) and 3 (right), isolated
after irradiation of TDDA and syn-TCOD with light of l>300 nm.
able to isolate any fully ring-opened [16]annulenes. Most
probably syn-TCOD adds to TDDA in a [2+2]cycloaddition
to form the ladderane 4, which undergoes [2+2]cyclorever-
sion and electrocyclic ring-opening to the 1,3-cyclohexa-
dienes 2 and 3. (Scheme 2)
[50]
bridge according to Prelog and Helmchen leads to an end-
less loop. Because of the C axis that bisects the double
2
bond, both faces are homotopic. As C2 is a chiral point
group, the molecule should be chiral. The unit cell indeed
includes both enantiomers. We have been able to separate
both enantiomers by HPLC on a chiral stationary phase
(Chiralcel OD) using CH OH/H O (90:10) as solvent. The
It is known that the photostationary equilibrium in the
electrocyclic reaction of 1,3-cyclohexadiene to 1,3,5-hexa-
triene is shifted towards the ring-opened product upon irra-
[48,49]
diation with shorter wavelength.
We therefore irradiat-
3
2
ed TDDA and syn-TCOD with a low-pressure mercury
lamp (254 nm) and obtained a complex mixture of isomers
in a total combined yield of 50%.
CD spectra are given for both isomers in Figure 6.
Structure 7 has C symmetry and Hückel topology. The
s
central double bond in the bridge forms a torsional angle of
5438
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5434 – 5445

Mçbius Annulenes
FULL PAPER
to 2. We therefore assign this
compound to the syn-tZcZcEt
structure 8, which has Mçbius
topology. We isolated a fifth an-
nulene structure in very small
amounts that is anti with re-
spect to the quinoid hydrogens
and C symmetric and therefore
2
must be either 9a anti-tEcZcEt
or anti-cEtZtEc 9b. Unfortu-
nately the yield is not large
enough
to
unambiguously
assign the conformation.
Does the twist make a differ-
ence? (Theoretical investiga-
tions): To analyze the proper-
ties of our Mçbius annulenes
we performed theoretical calcu-
lations at the B3LYP/6-31G*
and KMLYP/6-31G* of density
Figure 5. Crystal structures of three ring-opened [16]annulenes 5 (anti-tZtEcZt), 6 (anti-tZcZcZt), and 7 (syn-
tZtZtZt).
Figure 6. CD spectra of both enantiomers of the Mçbius structure 6.
Figure 7. UV/Vis spectra of the Mçbius annulene 6 and the Hückel annu-
lene 7.
1038 with the two neighboring double bonds. Therefore the
conjugation is almost completely interrupted and in contrast
to 6, which forms red rhombic crystals, compound 7 is color-
less. The Mçbius annulene 6 exhibits an additional band
that extends into the visible region and has a stronger ab-
sorption compared to 7 (Figure 7).
functional theory. The latter method provided more reliable
results in the case of several [4n+2]annulenes than the fre-
[51]
quently used B3LYP functional. However, applied on our
bianthraquinodimethane-modified [16]annulenes KMLYP/6-
31G* did not give more accurate geometries when com-
pared with the X-ray structures of 5 and 6. The mean devia-
tion of the B3LYP/6-31G* calculated bond lengths in 5 rela-
tive to the X-ray structure is 0.0093 Š, and 0.0165 Š for the
KMLYP-computed geometry. The corresponding mean devi-
ations for 6 are 0.010 Š (B3LYP/6-31G*) and 0.013 Š
(KMLYP/6-31G*). The mean difference of the energies of
the 25 B3LYP and KMLYP optimized structures also is
Two other isomers were isolated, but their structure could
not be unequivocally elucidated. One of these compounds
(
8), which is unsymmetrical and has a syn relationship of the
two quinoid hydrogen atoms, undergoes a clean reaction to
the ring-closed 1,3-cyclohexadiene 2 within several hours.
According to symmetry considerations and semiempirical
calculations the syn-tZcZcEt structure (the global minimum
of all isomers) is the only unsymmetrical compound within
the ten most stable isomers that is able to undergo a ther-
mochemically (Woodward–Hoffmann) allowed ring closure
À1
small (1.91 kcalmol ). Data for the calculated bond lengths
Chem. Eur. J. 2006, 12, 5434 – 5445
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5439

R. Herges et al.
and dihedral angles of compounds 5–8 and 9a and 9b are
given in Tables 3 and 4.
Computational Methods), and optimized the 153 most
stable structures at the B3LYP/6-31G* level of theory (the
25 most stable isomers are listed in Table 2). Interestingly,
among these structures, which have relative energies up to
One of the most interesting questions concerning the
twisted annulenes is whether there is a significant relation-
ship between topology (Hückel or Mçbius) and geometry,
energy and magnetic properties or in other words: Does the
twist change the properties? Numerous calculations on
Mçbius structures notwithstanding, a thorough and system-
atic study has not been made so far. To address the conse-
quences of a twist in the p system, we first analyzed the
data of the parent [16]annulene and then turned to the ques-
tion if this relationship is also valid for our bianthraquinodi-
methane-stabilized [16]annulenes. Purely based on statistics,
there are 2250 conceivable cis/trans isomers of [16]an-
nulenes, many of which are not minima on the energy hyper-
surface, because they are highly strained. We used a Monte
Carlo algorithm to generate the [16]annulene isomers (see
À1
19.66 kcalmol , there are more Mçbius (85) than Hückel
isomers (68). However, in agreement with experiment the
most stable isomers are Hückel structures. There are two in-
dependent X-ray structures available for the global mini-
[
52,53]
mum.
The second most stable isomer was detected by
[25,27]
NMR spectroscopy.
With a relative energy of 5.14 kcal
À1
mol , the energetically most favorable Mçbius compound
has little chance to be experimentally detected if a rapid
equilibrium of all isomers is assumed. The Mçbius structure
proposed by Rzepa et al.
7.1 kcalmol is number 24 in the ranking list of the most
stable isomers and the Mçbius isomer for which a strong ar-
[17]
with a relative energy of
À1
[21]
omaticity was predicted by Schleyer and Castro is ranked
number 137 with
energy of 15.8kcalmol . The
energy data confirm our initial
a
relative
À1
n
Table 3. B3LYP/6-31G* calculated bond lengths r of Mçbius compound 6 and its isomers 8, 9a, 5, 7, and 9b.
Note that the bonds are numbered not the atoms. Note also that because of the bianthraquinodimethane units
there are four different pathways to describe a [16]annulene periphery in each structure.
assumption that
synthesis of
16]annulene with Mçbius top-
a
successful
a
stable parent
[
ology is very unlikely and that
the twist has to be stabilized by
proper substitution.
The large set of computed
geometries provides
a
solid
r
r
1
9
r
r
2a
10
r
r
2b
r
r
3a
12
r
r
3b
r
4a
r
4b
r
5
r
6
r
7
r
8
r
16b
basis for the statistical analysis
of the properties. Most interest-
ing are those properties that
have been related to aromatici-
ty, such as bond-length equali-
11
13
r
14a
r
14b
r
15a
r
15b
r
16a
8
9
5
6
7
9
1.3656
.3609
1.3721
.3609
1.3683
.3572
1.3621
.3618
1.3649
.3423
1.3944
.3789
1.4872
1.4513
14 . 24 8 1.4942
1.4945
1.4398
1.4137
1.3596
1.4139
1.4242
1.3621
1.4271
1.36581.4773
1.4177 1.4823
1.4830
1.4808
1.4773
1.4855
1.4818
1.47981.36581.4376
1.47981.4139
1.4859
1.3643
1.4150
1.4448
1.4225
1.3599
1.3603
1.4888
1.4557
1.4595
1.4962
1
a
1
1.4557
1.4976
1.4575
1.4960
1.4561
1.4952
1.4846
1.4752
1.4295
1.4376
1.3599
1.4187
1.3608
1.4208
1.3607
1.4165
1.3490
1.4191
1.3649
1.4271
1.3542
14 . 24 8 1.4942
1.4721
1.3579 1.4454 zation
[54]
[55]
[56]
1.4948
1.4580
1.4901
1.4435
1.4952
1.4547
1.4870
1.4485
(HOMA,
Julg ),
1
1.3607 1.4831 1.4841 1.4226 1.4169 1.4959 1.4927 maximum deviation from pla-
1.4153
1.3626
1.4199
1.3575
1.4309
1.3721
1.4846
1.4846
1.4835
1.4835
1.4710
1.4710
1.4854
1.4855
1.4866
1.4866
1.4767
1.4767
1.3626
1.4207
1.3575
1.4165
1.3721
1.4191
1.4435
1.4153
1.4547
1.4199
1.4485
1.4309
1.3607
1.4960
1.3490
1.4952
1.3649
1.4752
1.4561
1.4901
1.4846
1.4952
narity, aromatic stabilization
1
[
57]
energy
(isomerization staili-
[58]
zation energy; ISE ), magnet-
1
[59–61]
b
1.4295 ic
1.4870
susceptibility,
and
[
62,63]
1
NICS.
All of these methods
have drawbacks that have been
extensively discussed in the lit-
erature. We refrain here from
following up this discussion.
The most straightforward ap-
proach to avoid the problems
inherent to each method is to
apply all methods to the same
data set and to do a principal
component analysis (PCA).
PCA is a powerful statistical
method to reduce the dimen-
sionality and redundancy to
n
Table 4. B3LYP/6-31G* calculated dihedral angles d of the single bonds in Mçbius compound 6 and its iso-
mers 8, 9a, 5, 7, and 9b. For the numbering see Table 3. The dihedral angles are defined by the atoms of the
numbered bond and the two neighboring carbon atoms.
d
d
1
9
d
d
2a
10
d
d
2b
d
d
3a
d
d
3b
d
4a
d
4b
d
5
d
6
d
7
d
8
d
16b
11
12
13
d
14a
d
14b
d
15a
d
15b
d
16a
8
9
5
6
7
9
À174.9
À3.6
115.5 À112.6
À22.0 À166.6
À0.3
170.2
11.0
À6.7 À132.4
156.5 À127.5
132.0
136.1 À173.6
168.9
À0.9
À45.5
132.8
À138.7
À1.7
16.0
3.4
À10.0
111.5 À106.8
a
À175.7 À117.8111.8 3.4
167.1 À168.0
35.2
6
.4
À163.7
75.6
35.2
167.1 À168.0 À156.5
132.0 À138.7
11.0 À117.8111.8
114.8 À115.9
5.8
À16.4
À4.7
À4.0 À137.1
À9.3 À139.0
136.3 À174.3
124.4
À7.3
167.9
À12.0
173.3
1
À38.8
150.5
109.6
167.9
141.1
3.3
2.8
118.9 À117.4
148.7 À108.7
2.4
129.8 À128.2
1.3
À34.9
À3.4
À34.9
1.3 À167.4
129.8 À128.2
À4.7
2.4 À108.7
109.6
0.0
114.0 À116.1
6.5
À1.9 À136.0
133.4
136.0 À133.4
136.9 À146.7
136.9 À146.7
2.2 À153.1
4.2
103.1 find patterns in data of high di-
116.1 mensionality. It has found ap-
0
.0 À103.1
153.1
À4.2 À176.3
À6.5
28.6
8.5
1.9 À114.0
b
À154.6 À133.2
124.0
29.4
8.5
12.3
29.4
164.3 À163.9
plication in fields such as com-
munication theory, face recog-
2
.6 À163.9
164.3 À147.0
12.3 À133.3
124.0
5440
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5434 – 5445

Mçbius Annulenes
FULL PAPER
nition, image compression, GC-MS, and 2D-NMR spectro-
scopy. Katritzky et al. used PCA to quantify the aromaticity
of a set of heterocyclic compounds.
Table 5. Correlation matrix for the PCA analysis of the 153 most stable
isomers of the parent [16]annulene. The parameters used are Erel (rela-
tive energy), Tmax (maximum deviation of dihedral angles from planari-
ty), HOMA (bond-length equalization index), Julg (bond-length equali-
zation index), MAGS (magnetic susceptibility), and NICS (nucleus inde-
pendent shift).
[64,65]
PCA can be used to extract significant information, leav-
ing noise behind. Noises in our case are, for example, local
effects disturbing the interpretation of NICS, the influence
of the benzene rings in determining the magnetic suscepti-
bility contribution as well as the bond-length equalization of
the [16]annulene periphery, the mutual influence of bond-
length equalization and out-of-plane distortion of the p
system, and numerous other effects interfering with or even
obscuring the effect we are interested in, which is the twist
in the p system.
E
rel
Tmax
HOMA
Julg
MAGS
NICS
E
rel
1.000
0.314
À0.075
À0.054
À0.209
À0.246
0.314
1.000
À0.075
À0.218
À0.054
À0.218
À0.209
À0.330
À0.721
À0.725
1.000
À0.246
À0.399
À0.651
À0.653
0.969
Tmax
HOMA
Julg
MAGS
NICS
À0.2181.000
À0.2180.999
0.999
1.000
À0.725
À0.653
À0.330
À0.399
À0.721
À0.651
0.969
1.000
The errors and misinterpretations that arise if one relies
on a single method and applies it on a single structure can
be illustrated with an example from a recent paper of
diagonalization of the correlation matrix. The largest eigen-
vectors (based on the size of the eigenvalues) are the most
important parameters describing the system. The eigenvec-
tors (principle components, PCs) are linear combinations of
the parameters used in the initial analysis (see Table 6).
[30]
Castro, Karney, Schleyer et al. They calculated the mag-
netic susceptibility exaltation (L) of our Mçbius compound
[62,63]
6
by using the incremental system of Dauben et al.
and
obtain a strongly negative value of À30.0 cgsppm, which is
about the value of naphthalene. Hence according to L our
Mçbius compound 6 would be very aromatic. However, L is
mainly determined by the four perfectly aromatic benzene
rings and not only by the [16]annulene periphery, which is
the key question. To deduct the strong contribution from
the four benzene rings they correct for their large negative
value (Lbenzene =À13.7 cgsppm) and end up with a very large
positive value of +19.6 cgsppm for the [16]annulene sub-
unit. Thus, our Mçbius compound would be extremely anti-
aromatic in its 16-electron periphery (compare, for example,
with the strongly antiaromatic heptalene: L=+6). This is in
disagreement with their main conclusion that the Mçbius
compound 6 is “nonaromatic”.
Table 6. Principal components of the PCA analysis for the 153 most
stable isomers of the parent [16]annulene. The following parameters
were used: Tmax (maximum deviation of dihedral angles from planarity),
HOMA (bond-length equalization index), Julg (bond-length equalization
index), MAGS (magnetic susceptibility), and NICS (nucleus independent
shift).
PC1
PC2
PC3
PC4
PC5
Tmax
HOMA
Julg
MAGS
NICS
0.063
0.495
0.495
À0.807
0.321
0.320
0.229
0.302
0.588
0.392
0.382
0.375
0.463
0.004
À0.087
À0.008
À0.736
0.671
À0.003
0.701
À0.711
À0.044
0.034
À0.513
À0.493
We avoid these problems by applying several methods on
a set of very similar structures: the 153 most stable isomers
of the parent [16]annulene and the 25 most stable isomers
of our bianthraquinodimethane-modified [16]annulenes. For
example, in the latter case the influence of the benzene
rings largely cancel and local effects are eliminated in the
PCA analysis. The first step in PCA is to calculate the corre-
lation matrix that provides information on the linear de-
pendence of the methods (relative energy, maximum devia-
tion of dihedral angles from planarity, HOMA, Julg, mag-
netic susceptibility, NICS) (Table 5).
It is interesting to note that the relative energy does not
correlate strongly with any of the aromaticity parameters.
Aromaticity, evidently, does not play an important (or at
least not a systematic) role in stabilizing the structures. With
a correlation of 0.999, the two methods representing the
bond-length equalization (HOMA and Julg) correlate per-
fectly. The same is true for the two magnetic parameters,
the magnetic susceptibility (MAGS) and NICS (correlation:
For visualization, the PCꢀs are plotted as a function of
each other. In Figure 8(top) the most important (PC1) is
plotted as a function of the second most important principle
component (PC2). Both components account for 95.7% of
the variance.
The plot is a text book example of the clustering capabili-
ty of the PCA method. Without using the information on
the topology of the structures in the analysis, the PCA plot
clearly separates two classes of structures the Hückel
(Figure 8, gray) and the Mçbius isomers (Figure 8, black;
the colors were assigned after the PCA analysis!). Following
the traditional interpretation of the aromaticity parameters,
the most aromatic structures are on the top right part of the
plot and the most antiaromatic isomers are located top left.
[24]
The global minimum synthesized by Schrçder and Sond-
[23]
heimer is clearly located on the antiaromatic side, which
1
is in agreement with the H NMR spectrum (d=10.4 ppm
for the “inner” protons and 5.4 ppm for the “outer” pro-
tons). The most aromatic isomer is the Mçbius structure cal-
[21]
0.969). Both MAGS and NICS methods provide almost ex-
culated by Castro, Karney, and Schleyer. Evidently, there
is a continuum between aromatic and antiaromatic struc-
tures.
actly the same information for our data set. Geometry and
magnetic properties, however, correlate more weakly
(
HOMA/NICS: À0.651). Our findings thus are in agreement
Having analyzed the parent [16]annulenes we turned to
the question if our bianthraquinodimethane-modified
[64]
with those of Katritzky et al. The next step in PCA is the
Chem. Eur. J. 2006, 12, 5434 – 5445
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5441

R. Herges et al.
Conclusions
Heilbronnerꢀs prediction that a planar [4n]annulene “…can
be twisted into a closed shell Mçbius strip perimeter without
loss in p electron energy” is only valid at the level of
Hückel molecular orbital theory. In the parent [4n]an-
nulenes, any Mçbius aromatic stabilization does not over-
come the strain induced by the twist. Other effects, such as
strain energy determine the topology. The three most stable
isomers of the parent [16]annulene are not twisted.
By inserting a bianthraquinodimethane unit into the ring
we were able to stabilize the twist. Structures with Mçbius
topology are now more stable than Hückel isomers and we
were able to isolate and characterize the first three Mçbius
annulenes.
By using a principle component analysis with five differ-
ent aromaticity parameters on a set of 153 parent [16]an-
nulene isomers, we were able to confirm previous assump-
tions that [4n]annulenes with a Mçbius twist exhibit proper-
ties that can be assigned to aromaticity. We applied the
same analysis also to 25 isomers of our bianthraquinodi-
A
H
R
U
G
tional relationship between the aromaticity parameters and
the topology. The Mçbius isomers exhibit enhanced aroma-
ticity relative to the Hückel structures. Thus, the benzoanne-
lation may reduce the aromaticity but not to zero as stated
Figure 8. Principle component analysis (PCA) of [16]annulenes by using
the parameters Tmax (maximum deviation of dihedral angles from pla-
narity), HOMA (bond-length equalization index), Julg (bond-length
equalization index), MAGS (magnetic susceptibility) and NICS (nucleus
independent shift) based on the B3LYP/6-31G* optimized geometries
[30]
previously.
(
see text). a) PCA of the 153 most stable isomers of parent [16]annulene.
Experimental Section
The two theoretically predicted Mçbius isomers and the two experimen-
tally known structures are labeled. b) PCA of the 25 most stable isomers
of Mçbius stabilized [16]annulenes. Compounds 5–8 have been isolated
and characterized by X-ray crystallography or NMR spectroscopy. An
unambiguous assignment of one of the isomers to 9a or 9b was not ach-
ieved.
Single-crystal structure analysis: All data were measured using an STOE
Imaging Plate Diffraction System (IPDS-1). Structure solutions were per-
formed with direct methods using SHELXS-97. Structure refinement was
2
done against F using SHELXL-97. All non-hydrogen atoms were refined
with anisotropic displacement parameters. The CÀH hydrogen atoms in
compound 2, 3, 5, and 7 were positioned with idealized geometry and
were refined with isotropic displacement parameters using a riding
model. In compound 6 the hydrogen atoms were refined with varying co-
ordinates and varying isotropic displacement parameters. In compound 2
two crystallographically independent molecules are found of which one
shows some disordering and therefore, was refined using a split model.
Details of data collection and structure refinement are given in Table 7.
[
16]annulenes would exhibit similar properties. This is
indeed the case. The PCA analysis with the two most impor-
tant PCꢀs is plotted in Figure 8(bottom). Again, the Hückel
and the Mçbius structures are clearly clustered on the basis
of the aromaticity probes. Hence, the twist also makes a dif-
ference in our bianthraquinodimethane-modified annulenes.
The experimentally isolated and characterized compounds
are indicated with their structure number. The Mçbius struc-
ture 6 that we isolated and characterized by X-ray analysis
and the global minimum 8 are located within the cluster of
aromatic Mçbius compounds. Structures 5 and 7 (character-
ized by X-ray analysis) are rather nonaromatic because they
are located very close to the dividing line of both topologies.
The most aromatic is structure 9b. Unfortunately, we have
so far not been able to elucidate its conformation, which
could be either 9a or 9b. On an absolute scale, the aroma-
ticity of our benzoannelated [16]annulenes is reduced; how-
ever, the Mçbius twisted isomers are clearly more aromatic
than the nontwisted structures, as this is the case with the
parent [16]annulenes.
Synthesis of syn-TCOD: Sodium (6.5 g) was added to mercury (650 g) in
small portions under an argon atmosphere at room temperature. The
sodium amalgam was cooled to room temperature and diethyl ether
(
(
250 mL) was added. The mixture was sonicated with an ultrasonic probe
cell disruptor 250 W, 50% output) for 30 min at 58C under argon and
3
,4-cis-dichlorocyclobutene (5 mL) was added. The mixture was sonicated
at 5–108C for 2–7 d (the reaction was monitored by GC (column temper-
ature 308C). The reaction was allowed to run until all 3,4-cis-dichlorocy-
clobutene was converted to tricyclooctadiene). The mixture was filtrated
with Celite and the diethyl ether was evaporated carefully first at 308C
(
800 mbar) and then 08C (200 mbar). A flow of argon was passed
through the yellowish liquid until no diethyl ether was detected in the
NMR spectrum. The syn-tricyclooctadiene (1.5 g, 68%) was stored in
1
benzene (10 mL). H NMR (500 MHz, CDCl
3
/TMS): d=6.03 (s, 4H),
3
.13 ppm (s, 4H).
Preparation of compounds 2–4:
0
0
A
suspension of TDDA (100 mg,
.28mmol), TCOD (300 mg, 2. 88 mmol), and naphthalene (20 mg,
.15 mmol) was irradiated with a 700 W high-pressure mercury lamp in a
quartz photoreactor and agitated with a flow of argon. After 10–20 min
5442
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5434 – 5445

Mçbius Annulenes
FULL PAPER
Table 7. Selected crystal data and results of the structure refinements for compounds 2, 3, 5, 6 and 7.
129.12 (CH, arom), 128.78 (CH,
arom), 128.45 (CH, arom), 128.31
2
3
5
6
7
(
CH, arom), 128.08 (CH, arom),
formula
C
456.55
monoclinic
P2
8.6822(7)
29.245(3)
19.284(2)
101.43(1)
4799.4(7)
170
36
H
24
C
456.55
monoclinic
P2
10.1574(8)
8.7109(6)
28.100(2)
100.07(1)
2448.0(3)
170
36
H
24
C
456.55
monoclinic
P2
9.6883(7)
16.5449(12)
15.3416(9)
100.90(1)
2414.7(3)
150
36
H
24
C
36
H
24
C
456.55
monoclinic
1
P2 /c
8.9779(5)
29.007(2)
9.3982(5)
94.09(1)
2441.3(3)
150
36 24
H
1
1
26.90 (CH, arom), 126.7 (CH, arom),
25.37 (CH, arom), 125.12 (CH,
À1
M
r
[gmol
]
456.55
monoclinic
C2/c
16.280(2)
10.959(1)
14.373(2)
110.41(2)
2403.5(2)
173
crystal system
space group
a [Š]
b [Š]
c [Š]
arom), 124.12 (CH, arom), 121.60
CH, arom), 46.76 (CH, aliphatic),
9.10 (CH, aliphatic), 29.71 ppm (CH,
aliphatic); MS(70 eV): m/z (%): 456
1
/c
1
/c
1
/c
(
3
AHCTREUNG
+
(
100) [M ].
b [8]
V [Š ]
Preparation of compounds 5–9: A sol-
ution of TDDA (100 mg, 0.28mmol),
TCOD (300 mg, 2.88 mmol) and naph-
thalene (20 mg, 0.15 mmol) was irradi-
ated with a 15 W low-pressure mercu-
ry lamp in a quartz photoreactor and
agitated with an argon gas flow. After
3
T [K]
Z
8
4
4
4
4
À3
1
calcd [gcm
]
1.264
1.239
1.256
1.262
1.242
2
m
q range [8]8
3–46
0.07
3–483–52
0.07
2–55
3–45
À1
A
C
H
T
R
E
U
N
G
(MoKa) [mm
]
0.07
0.07
0.07
measured reflns
25066
0.0953
6826
4850
723
14086
0.0535
3651
3064
326
9917
0.0771
4658
2757
326
28913
0.0813
2766
2356
212
7739
0.0759
3150
1773
326
2
4 h of irradiation, the solvent was
R
int.
evaporated and the solid residue was
dissolved in dichloromethane and
independent reflns
observed reflns [I>2s(I)]
parameters
passed through
a short column of
silica gel. The combined yield of iso-
mers was 50%. For separation of the
R
1
[I>2s(I)]
0.0872
0.2212
1.105
0.49/À0.39
0.0495
0.1341
1.070
0.26/À0.19
0.0542
0.1397
0.959
0.23/À0.20
0.0400
0.1132
1.046
0.25/À0.22
0.0599
0.1488
0.951
0.34/À0.19
wR (all data)
Goof
residual electron density [eŠ
2
products
a semipreparative HPLC
À3
column and a mixture of 95:5 of hep-
tane and dichloromethane was used.
For a copy of the HPLC chromato-
gram (UV detector, 254 nm) see Sup-
porting Information. The cyclohexa-
]
irradiation, the solvent was evaporated and the residue was dissolved in
dichloromethane and passed through a short column of silica gel. By
using preparative HPLC (90:10 heptane/dichloromethane) three isomers
diene compound 2 (C
and Mçbius compound 8 (C
s
symmetry), Mçbius compound 6 (C
2
symmetry),
1
symmetry) have very close retention times.
Isolation of 6 was only possible after removal of 2 by titration with
TCNE. The cyclohexadiene isomer reacts faster and forms a polar [4+2]
adduct, which can be easily separated. Compound 8 is thermolabile and
has to be separated at low temperatures. Single crystals of 5, 6, and 7
were obtained by the diffusion method from heptane/dichloromethane.
were separated and characterized: 1,3-cyclohexadiene structure 2 (C
symmetry; 12 mg, 9.4%), 1,3-cyclohexadiene structure 3 (C symmetry;
mg, 6.3%) and ladderane 4 (2 mg, 1.6%). For a copy of the HPLC
s
2
8
chromatogram (UV detector, 254 nm) see the Supporting Information.
Single crystals of 2 and 3 were obtained from a heptane/dichloromethane
1
Data for compound 5 (Mçbius C ): Yield: 8mg; H NMR (500 MHz,
1
mixture.
CDCl /TMS): d=7.83 (m, 4H), 7.44 (dd, J=5.5, 3.4 Hz, 2H), 7.20 (dd,
J=7.5, 7.5 Hz, 4H), 7.09 (m, 4H), 6.72 (dd, J=9.4, 6.9 Hz, 2H; olefinic),
3
1
Data for compound 2 (C
s
symmetry): H NMR (500 MHz, CDCl
3
/TMS):
d=7.61 (d, J=7.0 Hz, 2H), 7.26 (d, J=7.0 Hz, 2H), 7.21 (dd, J=7.0,
6.41 (dd, J=6.9, 1.7 Hz, 2H; olefinic), 6.22 ppm (d, 4H; olefinic);
1
3
7
7
5
2
.0 Hz, 2H), 7.16 (m, 2H), 7.13 (dd, J=7.0, 7.0 Hz, 2H), 7.01 (d, J=
.0 Hz, 2H), 6.91 (dd, J=7.5, 7.5 Hz, 2H), 6.86 (dd, J=7.5, 7.5 Hz, 2H),
.92 (d, 10.5 Hz, 2H), 5.67 (d, J=10.5 Hz, 2H), 5.14 ppm (d J=10.5 Hz,
C NMR (125.8MHz, CDCl
3
/TMS): d=139.81 (C
q
, C=C bridgehead),
), 135.5 (C ),
138.09 (C , CR =CHR), 136.98(C
q
2
q
), 136.52 (C ), 135.90 (C
q
q
q
132.97 (CH, olefinic), 132.72 (CH, olefinic), 128.72 (CH, arom), 128.65
(CH, arom), 127.75 (CH, olefinic), 126.78(CH, arom), 126.14 (CH,
arom), 125.01 (CH, arom), 124.81 (CH, arom), 124.77 (CH, arom), 123.97
1
3
H), 3.40 ppm (m, 2H); C NMR (125.8MHz, CDCl
, C=C bridgehead), 141.08(C , CR =CHR), 140.27 (C
37.6 (C ), 134.47 (C ), 131.13 (CH, olefinic), 127.42 (CH, arom), 127.33
3
/TMS): d=141.40
(
C
q
q
2
q
), 138.58 (C ),
q
1
q
q
(CH, arom), 123.84 ppm (CH, olefinic); UV/Vis (CHCl ): lmax (e)=250
2
+
(
(
CH, cyclohexadiene), 127.28(CH, arom), 125.21 (CH, arom), 125.16
CH, arom), 125.12 (CH, arom), 124.90 (CH, arom), 124.86 (CH, arom),
(19528), 300 nm (15199); MS (70 eV): m/z (%): 456 (100) [M ].
1
Data for compound 6 (Mçbius C ): Yield: 6 mg; H NMR (500 MHz,
2
3
4
ACHTREUNG
1
24.68(CH, arom), 123.15 (CH, cyclohexadiene), 40.75 ppm (CH, ali-
CDCl /TMS): d=7.709 (ddd,
J(1-H,2-H)=7.5 Hz, J(1-H,3-H)=1.3 Hz,
A
H
R
U
G
3
+
5
ACHTREUNG
phatic); MS (70 eV): m/z (%): 456 (100) [M ].
J(1-H,4-H)=0.5 Hz, 2H; 1-H), 7.516* (m, 2H; 8-H), 7.209* (m, 2H; 4-
1
3
4
ACHTREUNG
Data for compound 3 (C
2
symmetry): H NMR (500 MHz, CDCl
3
/TMS):
H), 7.201* (m, 2H; 5-H), 7.184 (ddd, J
A
H
R
U
G
3
3
ACHTREUNG
d=7.53 (m, 4H), 7.26 (d, J=7.0 Hz, 2H), 7.15 (d, J=7.0 Hz, 2H), 7.11
d, J=7.0 Hz, 2H), 7.06 (m, 2H), 7.02 (m, 2H), 6.99 (dd, J=7.0, 1.3 Hz,
1.3 Hz, 2H; 2-H), 7.098(ddd,
J
A
H
R
U
G
4
3
ACHTREUNG
(
J
A
H
R
U
G
J(7-H,6-H)=7.4 Hz, J(7-
2
H), 5.92 (d, J=10.0 Hz, 2H), 5.66 (dd, J=10.0, 1.5 Hz, 2H), 5.54 (dd,
1
3
J=10.0, 2.8Hz, 2H), 3.67 ppm (dd, J=7.0 Hz, 2H);
C NMR
(
125.8MHz, CDCl
3
/TMS): d=140.68(C
q
, C=C bridgehead), 140.48(C
q
,
CR =CHR), 140.40 (C
2
q
), 140.05 (C ), 137.17 (C
q
q
), 134.88 (C ), 129.28
q
(
(
CH, olefinic), 127.96 (CH, cyclohexadiene), 127.18(CH, arom), 126.52
CH, arom), 126.36 (CH, arom), 125.89 (CH, arom), 125.50 (CH, arom),
1
25.03 (CH, arom), 124.48(CH, arom), 122.38(CH, arom), 121. 86 (CH,
cyclohexadiene), 36.99 ppm (CH, aliphatic); MS(70 eV): m/z (%): 456
AHCTREUNG
+
(
100) [M ].
1
Data for compound 4 (ladderane): H NMR (500 MHz, CDCl
3
/TMS): d=
.85 (dd, 2H), 7.45 (dd, 2H), 7.40 (dd, 2H), 7.33 (dd, 2H), 7.22 (dd, 2H),
.10 (dd, 2H), 7.02 (dd, 2H), 6.67 (dd, 2H), 6.50 (s, 2H; olefinic), 4.31
7
6
(
1
3
dd, 2H), 3.27 (dd, 2H), 2.86 ppm (dd, 2H); C NMR (125.8MHz,
CDCl /TMS): d=144.20 (C , C=C bridgehead), 143.38(C , CR =CHR),
38.11 (C ), 137.78(C ), 137.56 (C ), 134.47 (C ), 129.27 (CH, olefinic),
3
q
q
2
1
q
q
q
q
Chem. Eur. J. 2006, 12, 5434 – 5445
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5443

R. Herges et al.
4
3
ACHTREUNG
H,8-H)=7.4 Hz,
J
A
C
H
T
R
E
U
N
G
(7-H,5-H)=2.0 Hz, 2H; 7-H
A
), 7.003 (ddd,
J
Acknowledgement
3
4
ACHTREUNG
H)=7.4 Hz, J
A
C
H
T
R
E
U
N
G
(6-H,7-H)=7.4 Hz, J
dd, J(11-H,12-H)=12.5 Hz, J(11-H,13-H)=1.5 Hz, 2H; 11-H), 6.118
dd, J(12-H,11-H)=12.5 Hz, J(12-H, 13-H)=11.9 Hz, 2H; 12-H), 5.940
J(14-H,13-H)=4.7 Hz,
.690 ppm (dddd, J(13-H,12-H)=11.9 Hz, J(13-H,14-H)=4.7 Hz, J(13-
H,14’-H)=1.9 Hz, J=0.9 Hz, 2H; 13-H); *chemical shifts were assigned
from HSQC spectra; C NMR (125.8MHz, CDCl
C), 139.92 (9-C), 139.36 (10b-C), 138.93 (9a-C), 138.34 (9b-C), 135.91
10a-C), 128.09 (1-C), 127.79 (12-C), 127.04 (14-C), 126.42 (6-C), 126.27
8-C), 126.15 (4-C), 125.97 (3-C), 125.41 (2-C), 125.24 (7-C), 124.89 (11-
C), 122.05 ppm (5-C); UV/Vis (CH
17251), 400 nm (br, 2946); MS (70 eV): m/z (%): 456 (100) [M ].
(6-H,8-H)=2.0 Hz, 2H; 6-H
B
3
4
(
(
(
D.A. was supported by a scholarship from the federal state of Nieder-
sachsen. The work was supported by the Deutsche Forschungsgemein-
schaft and the Fonds der Chemischen Industrie.
3
3
3
4
dd,
J(14-H,14’-H)=1.9 Hz, 2H; 14-H),
3
3
4
5
n
1
3
3
/TMS): d=140.67 (10-
[
[
[
[
1] E. Heilbronner, Tetrahedron Lett. 1964, 5, 1923–1928.
2] A. A. Frost, B. Musulin, J. Chem. Phys. 1953, 21, 572–573.
3] H. E. Zimmerman, J. Am. Chem. Soc. 1966, 88, 1564–1567.
4] H. E. Zimmerman, Science 1966, 153, 83–144.
[5] H. E. Zimmerman, Accts. Chem. Res. 1971, 4, 272–280.
[6] R. W. Zoellner, J. F. Krebs, D. M. Browne, J. Chem. Inf. Comput.
Sci. 1994, 34, 252–258.
[7] J. Cz. Dobrowolski, J. Chem. Inf. Comput. Sci. 2002, 42, 490–499.
[8] J.-M. Andrꢂ, B. Champagne, E. A. Perpꢃte, M. Guillaume, Int. J.
Quantum Chem. 2001, 84, 607–616.
(
(
2
Cl
2
): lmax (e)=253 (26231), 312
+
(
1
Data for compound 7 (Hückel C
s
): H NMR (500 MHz, CDCl
3
/TMS):
(8-H,5-H)=
(1-H,3-H)=1.5 Hz,
3
4
5
ACHTREUNG
d=7.733 (ddd,
J
A
C
H
T
R
E
U
N
G
(8-H,7-H)=7.5 Hz,
J
A
H
R
U
G
J
3
4
ACHTREUNG
0
.6 Hz, 2H; 8-H), 7.609 (ddd, J
A
H
R
U
G
[
9] M. Guillaume, B. Champagne, E. A. Perpꢃte, J.-M. Andrꢂ, Theor.
Chem. Acc. 2001, 105, 431–436.
[
[
[
[
[
10] S. Martin-Santamaria, H. S. Rzepa, J. Chem. Soc. Perkin Trans. 2
000, 2378–2381.
11] H. S. Rzepa, K. R. Taylor, J. Chem. Soc. Perkin Trans. 2 2002, 1499–
501.
12] C. J. Kastrup, S. P. Oldfield, H. S. Rzepa, Chem. Commun. 2002,
42–643.
13] S. Martin-Santamaria, B. Lavan, H. S. Rzepa, Chem. Commun. 2001,
089–1090.
2
1
6
1
14] “Genesis and Bond Relocation of the Cyclononatetraenyl Cation
and Related Compounds”: E. Yakali, Ph.D. Dissertation, Syracuse
University, 1973, pp. 39–40.
5
3
4
ACHTREUNG
J
A
C
H
T
R
E
U
N
G
(1-H,4-H)=0.6 Hz, 2H; 1-H), 7.332 (ddd,
J
A
T
E
N
(5-H,6-H)=7.3 Hz,
J
3
ACHTREUNG
[15] J. C. Barborak, T. M. Su, P. von R. Schleyer, J. Am. Chem. Soc. 1971,
93, 279–281.
[16] M. Mauksch, V. Gogonea, H. Jiao, P. von R. Schleyer, Angew.
Chem. 1998, 110, 2515–2517; Angew. Chem. Int. Ed. 1998, 37, 2395–
2398.
H,7-H)=1.5 Hz, 2H; 5-H), 7.114 (m, 2H; 4-H), 7.087 (ddd,
H)=7.5 Hz,
J
3
4
ACHTREUNG
J
A
C
H
T
R
E
U
N
G
(7-H,6-H)=7.3 Hz,
J
3
3
4
(
2
ddd,
H; 6-H), 7.005 (ddd,
H,4-H)=1.5 Hz, 2H; 2-H), 6.978(ddd,
J
A
C
H
T
R
E
U
N
G
(6-H,7-H)=7.3 Hz,
J
A
H
R
U
G
J(6-H,8-H)=1.3 Hz,
A
H
R
U
G
3
3
4
ACHTREUNG
J
A
H
R
U
G
J
A
H
R
U
G
J
3
3
ACHTREUNG
J
A
H
R
U
G
J
4
3
[17] S. Martin-Santamaria, B. Lavan, H. S. Rzepa, J. Chem. Soc. Perkin
Trans. 2 2000, 1415–1417.
[18] R. P. Johnson, K. J. Daoust, J. Am. Chem. Soc. 1996, 118, 7381–
7385.
[19] J. Verbeek, J. H. van Lenthe, J. J. A. Timmermans, A. Mackor,
P. H. M. Budzelaar, J. Org. Chem. 1987, 52, 2955.
3
3
1
1
1
5
1.0 Hz, J(12-H,13-H)=10.8Hz, 2H; 12-H), 6.260 (dd, J(11-H,12-H)=
4
3
1.0 Hz, J(11-H,13-H)=1.3 Hz, 2H; 11-H), 6.238(ddd, J(14-H,14’-H)=
3
4
1.0 Hz, J(14-H,13-H)=2.4 Hz, J(14-H,13’-H)=1.1 Hz, 2H; 14-H),
.875 ppm (dddd, J(13-H,12-H)=10.8Hz, J(13-H,14-H)=2.4 Hz, J(13-
3
3
4
4
13
H,14’-H)=1.1 Hz,
J(13-H,11-H)=0.9 Hz, 2H; 13-H);
/TMS): d=139.76 (9-C)*, 138.19 (10-C)*, 137.93 (10a-
C)*, 137.65 (9b-C)*, 137.47 (9a-C)*, 136.31 (10b-C)*, 131.52 (13-C),
C NMR
[
20] R. W. A. Havenith, J. H. van Lenthe, L. Jenneskens, Int. J. Quantum
Chem. 2001, 85, 52–60.
(
125.8MHz, CDCl
3
[
21] C. Castro, C. M. Isborn, W. L. Karney, M. Mauksch, P. von R.
Schleyer, Org. Lett. 2002, 4, 3431–3434.
1
1
1
29.83 (14-C), 128.45 (8-C), 128.02 (12-C), 127.94 (1-C), 126.44 (3-C),
26.06 (6-C), 125.41 (5-C), 125.23 (7-C), 124.60 (2-C), 123.29 (4-C),
22.00 ppm (11); * assignement according to a CSGT B3LYP/6–31G* cal-
culation; UV/Vis (CH
70 eV): m/z (%): 456 (100) [M ].
Data for compound 8 (Mçbius C , global minimum): H NMR (500 MHz,
CDCl /TMS): d=7.75 (dd, 2H), 7.37 (dd, 2H), 7.20 (dd, 2H), 7.15 (m,
H), 7.12 (m, 2H), 6.94 (dd, 2H), 6.80 (t, 2H), 6.72 (dd, 1H; olefinic),
.25 (d, 1H; olefinic), 6.15 (t, 1H; olefinic), 5.97 (d, 1H; olefinic), 5.90
dd, 1H; olefinic), 5.85 (dd, 1H; olefinic), 5.78 (dd, 1H; olefinic),
.72 ppm (dd, 1H; olefinic); MS(70 eV): m/z (%): 456 (100) [M ]. The
compound isomerizes at room temperature to give the C
clohexadiene structure 2.
[
[
22] K. C. Nicolaou, private communication.
23] F. Sondheimer, Y. Gaoni, J. Am. Chem. Soc. 1961, 83, 4863–4864.
2 2
Cl ): lmax (e)=259 (17027), 299 nm (16334); MS
+
[24] G. Schrçder, J. F. M. Oth, Tetrahedron Lett. 1966, 7, 4083–4088.
(
[
[
25] J. F. M. Oth, J.-M. Gilles, Tetrahedron Lett. 1968, 9, 6259–6264.
26] J. F. M. Oth, G. Anthoine, J.-M. Gilles, Tetrahedron Lett. 1968, 9,
1
1
3
6
265–6270.
27] C. D. Stevenson, T. L. Kurth, J. Am. Chem. Soc. 1999, 121, 1623–
624.
2
6
(
[
1
[
[
28] C. D. Stevenson, T. L. Kurth, J. Am. Chem. Soc. 2000, 122, 722–723.
29] D. Ajami, O. Oeckler, A. Simon, R. Herges, Nature 2003, 426, 819–
+
5
A
H
R
U
G
s
symmetric cy-
8
21.
[
[
30] C. Castro, Z. Chen, C. S. Wannere, H. Jiao, W. L. Karney, M.
Mauksch, R. Puchta, N. J. R. van Eikema Hommes, P. von R. Schley-
er, J. Am. Chem. Soc. 2005, 127, 2425–2432.
31] Gaussian 03, Revision C.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
1
Data for compound 9 (Mçbius C
2
): H NMR (500 MHz, CDCl
3
/TMS):
d=7.75 (dd, 2H), 7.37 (dd, 2H), 7.20 (dd, 2H), 7.15 (m, 2H), 7.12 (m,
H), 6.94 (dd, 2H), 6.80 (t, 2H), 6.72 (dd, 1H; olefinic), 6.25 (d, 1H; ole-
finic), 6.15 (t, 1H; olefinic), 5.97 (d, 1H; olefinic), 5.90 (dd, 1H; olefinic),
.85 (dd, 1H; olefinic), 5.78 (dd, 1H; olefinic), 5.72 pm (dd, 1H; olefin-
ic); MS (70 eV): m/z (%): 456 (100) [M ]. The data are not sufficient to
distinguish between the two possible conformations 9a and 9b.
2
5
A
H
R
U
G
+
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
5444
www.chemeurj.org
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 5434 – 5445

Mçbius Annulenes
FULL PAPER
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
[47] L. A. Paquette, M. J. Carmody, J. Am. Chem. Soc. 1976, 98, 8175–
8181.
[48] W. G. Dauben, E. G. Olsen, J. Org. Chem. 1980, 45, 3377–3382.
[49] B. Matuszewski, A. W. Burgstrahler, R. S. Givens, J. Am. Chem. Soc.
1982, 104, 6874–6876.
[50] V. Prelog, G. Helmchen, Angew. Chem. 1982, 94, 614–631; Angew.
Chem. Int. Ed. Engl. 1982, 21, 567–583.
[
51] C. S. Wannere, K. W. Sattelmeyer, H. F. Schaefer III, P. von R.
Schleyer, Angew. Chem. 2004, 116, 4296–4302; Angew. Chem. Int.
Ed. 2004, 43, 4200–4206.
[
[
[
[
[
32] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5852.
33] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1998, 37, 785–789.
34] J. K. Kang, C. B. Musgrave, J. Chem. Phys. 2001, 115, 11040–11051.
35] T. A. Keith, R. F. W. Bader, Chem. Phys. Lett. 1993, 210, 223–231.
36] K. Wolinski, J. F. Hilton, P. Pulay, J. Am. Chem. Soc. 1990, 112,
[
[
52] S. M. Johnson, I. C. Paul, J. Am. Chem. Soc. 1968, 90, 6555–6556.
53] S. M. Johnson, I. C. Paul, S. D. Geoffrey, J. Chem. Soc. B 1970, 4,
6
43–649.
[
[
[
[
[
54] T. M. Krygowski, M. K. Cyranski, Chem. Rev. 2001, 101, 1385–1420.
55] T. M. Krygowski, M. Cyranski, Tetrahedron 1996, 52, 1713–1722.
56] A. Julg, P. Francois, Theor. Chim. Acta 1967, 7, 249–259.
57] L. J. Schaad, B. A. Hess, Jr., Chem. Rev. 2001, 101, 1465–1476.
58] C. S. Wannere, D. Moran, N. L. Allinger, B. A. Hess, Jr., L. J.
Schaad, P. von R. Schleyer, Org. Lett. 2003, 5, 2983–2986.
8
251–8260.
37] G. Chang, W. C. Guida, W. C. Still, J. Am. Chem. Soc. 1989, 111,
379–4386.
[
4
[
[
38] N. L. Allinger, J. Am. Chem. Soc. 1977, 99, 8127–8134.
39] J. Lii, S. Gallion, C. Bender, H. Wikstrom, N. L. Allinger, K. M.
Flurchick, M. M. Teeter, J. Comput. Chem. 1989, 10, 503–513.
40] S. Kammermeier, R. Herges, Angew. Chem. 1996, 108, 470–472;
Angew. Chem. Int. Ed. Engl. 1996, 35, 417–419.
[
59] H. J. Dauben, Jr., J. D. Wilson, J. L. Laity, J. Am. Chem. Soc. 1968,
[
[
[
9
0, 811–812.
60] H. J. Dauben, Jr., J. D. Wilson, J. L. Laity, J. Am. Chem. Soc. 1969,
1, 1991–1998.
[
41] S. Kammermeier, P. G. Jones, R. Herges, Angew. Chem. 1996, 108,
9
2
834–2836; Angew. Chem. Int. Ed. Engl. 1996, 35, 2669–2671.
[
[
61] W. H. Flygare, Chem. Rev. 1974, 74, 653–687.
42] S. Kammermeier, R. Herges, Angew. Chem. 1997, 109, 2317–2319;
Angew. Chem. Int. Ed. Engl. 1997, 36, 2200–2202.
43] S. Kammermeier, R. Herges, unpublished results.
62] P. von R. Schleyer, C. Maerker, A. Dransfeld, H. Jiao, N. J. R. van
Eikema-Hommes, J. Am. Chem. Soc. 1996, 118, 6317–6318.
63] P. von R. Schleyer, H. Jiao, N. J. R. van Eikema-Hommes, V. G.
Malkin, O. L. Malkina, J. Am. Chem. Soc. 1997, 119, 12669–12670.
64] A. R. Katritzky, P. Barczynski, G. Musumarra, D. Pisano, M. Sza-
fran, J. Am. Chem. Soc. 1989, 111, 7–15.
[
[
[
[
44] T. N. Das, K. I. Priyadarsini, J. Chem. Soc. Faraday Trans. 1994,
9
63–968.
[
[
45] C. D. Nenitzescu, M. Avram, E. Marica, I. G. Dinulescu, P. G. Ma-
teescu, Angew. Chem. 1963, 75, 8 8 A; ngew. Chem. Int. Ed. Engl.
[65] A. R. Katritzky, K. Jug, D. C. Oniciu, Chem. Rev. 2001, 101, 1421–
1
963, 1, 659.
1449.
46] M. Avram, I. G. Dinulescu, E. Marica, G. Mateesu, E. Sliam, C. D.
Received: February 15, 2006
Published online: June 13, 2006
Nenitzescu, Chem. Ber. 1964, 97, 382–389.
Chem. Eur. J. 2006, 12, 5434 – 5445
ꢁ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5445