The carbon skeleton of the belt region of fullerene C84 (D2)

Article abstract of DOI:10.1002/chem.200304824

The synthesis and structural characterization of the double-stranded carbon skeleton in the belt region of a C₈₄ fullerene has been achieved. The synthetic methodology used is based on cyclic dimerization of diastereomeric AB-type monomers 18 b

Full text of DOI:10.1002/chem.200304824

FULL PAPER
The Carbon Skeleton of the Belt Region of Fullerene C (D )
84
2
[
a]
[b]
[b]
[a]
Wolf Dietrich Neudorff,* Dieter Lentz, Maribel Anibarro, and A. Dieter Schl¸ter*
Abstract: The synthesis and structural
characterization of the double-stranded
carbon skeleton in the belt region of a
C84 fullerene has been achieved. The
synthetic methodology used is based on
cyclic dimerization of diastereomeric
AB-type monomers 18 by a non-diaste-
reospecific Diels ±Alder reaction with
isobenzofuran and acenaphthylene
groups as reactive termini. Four dia-
steromeric monomer precursors 17 were
prepared for the first time by the use of
dihydropyracylene (12) in a multistep
synthesis. The synthesis of dihydropyr-
acylene itself has been optimized to the
degree that it is now available on the
10 g scale. The belt-shaped macrocycle
19, obtained from dimerization of the
monomers 17, could be partly aromat-
ized by an acid-catalyzed dehydration
reaction to give 23, which differs from
the fully unsaturated belt by two ™wa-
ter∫ molecules. Semiempirical AM1 cal-
culations of the electronic and thermo-
dynamic properties of cyclic fluoran-
thenes revealed strain energy as the
essential reason for the incomplete ar-
omatization of 19. The structures of the
macrocycles 19 and 23, one of the
monomer precursors, and two diastereo-
meric epoxybenzo[k]fluoranthenes were
elucidated by single-crystal X-ray crys-
tallography.
Keywords: arenes ¥ cyclooligomeri-
zation
¥
Diels ±Alder reaction
¥
macrocycles ¥ strain
Introduction
which cyclization becomes the main reaction. Double-strand-
ed cycles have always been appealing synthesis targets, for
example, to study orbital interactions and host ±guest phe-
nomena. Compounds with all-carbon skeletons, which are
potential precursors of [n]cyclacenes (1) and [n]beltenes (2),
The synthesis of double-stranded polymers by poly-Diels ±
[
1]
Alder (DA) addition is occasionally accompanied by the
formation of cyclic dimers and trimers.[ If linear oligomers
attain conformations, during the first growth steps, that allow
a facile ring closure, cyclization can evidently compete with
linear growth. In certain cases, conditions can be found under
2]
[
a] Prof. Dr. A. D. Schl¸ter, Dr. W. D. Neudorff
Institut f¸r Chemie, Organische Chemie,
Freie Universit‰t Berlin
Takustrasse 3, 14195 Berlin (Germany)
Fax : (‡49)30-838-53357
E-mail: adschlue@chemie.fu-berlin.de
were of special interest for a couple of years.[2, 3] All attempts
[
b] Dr. D. Lentz, M. Anibarro
Institut f¸r Chemie, Anorganische und Analytische Chemie
Freie Universit‰t Berlin
to generate [n]cyclacenes failed, presumably because of their
predicted high reactivity. According to density functional
[4]
Fabeckstrasse 34 ±36, 14195 Berlin (Germany)
Fax : (‡49)30-838-52424
calculations, for example, they should have a triplet ground
state.[ In a sense, cyclacenes resemble the infinite, hypo-
thetical polyacene.
5]
E-mail: lentz@chemie.fu-berlin.de
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/chemistry/ or from the author. It con-
tains a) a suggestion for a nomenclature system for ™fluoranthene-like∫
PAHs (SI Chart 1), b) AM1-calculated properties of [2.2.q]circocenes,
Recently attempts to obtain less reactive, yet still double-
stranded cycles were reported and the near future will show
whether they can be converted into their respective unsatu-
rated analogues.[ Though not applied in all cases, the main
synthetic tool is a electrocyclic ring-closure reaction, such as a
Diels ±Alder reaction. Attempts were made to develop
rational syntheses for representatives, the most developed
being the so-called ™substrate-directed synthesis∫ by Stoddart
and co-workers.[ This attempt is also based upon DA
chemistry; it utilizes bifunctional components with appropri-
[
2.q:2.q À 1]circocenes, [2.2.q]fluoranthenes, [2.q:2.q À 1]fluoranthenes
with q ˆ 1 ±5 and q À 1 > 0, [n]cyclacenes with n ˆ 4 ±16 (SI Tables 1 ±
); c) AM1-calculated gas-phase PAs of circocenes for different sites of
protonation (SI Chart 2); d) List of structures (with DH values) used to
calculate the MSE of A (SI Charts 3 and 4); e) Calculated MSEs and
6]
3
f
r
dehydration enthalpies for series A
r
, C
r
and a second set B
r
, D
r
(SI
1
Figure 1, SI Tables 4 and 5); f) H NMR spectra of 16 (SI Figure 2);
3a]
1
g) H NMR spectra of 17 (SI Figure 3); h) NMR spectra of the reaction
1
9 ! 27 ! 19 (SI Figure 4).
Chem. Eur. J. 2003, 9, 2745 ±2757
DOI: 10.1002/chem.200304824
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2745

A. D. Schl¸ter, W. D. Neudorff et al.
FULL PAPER
ate curvature and high endo ± exo stereoselectivity, and also
allows for a predictable cyclization to occur. None of the
double-stranded cycles with an all-carbon frame available
nowadays can be completely converted into their fully
unsaturated analogues yet. Some time ago, we showed that
a complete aromatization of DA polymers can be achieved
when this process leads to macromolecules with fullerene-like
structural elements, as in 3 and 4.[ The incorporation of peri-
condensed naphthalene units with two adjacent five-mem-
bered rings considerably stabilizes the resulting polycyclic
permeation chromatography (GPC) of the raw material.[7b]
Evidently the distance between the respective two termini is
not commensurable, and the attainment of a curved con-
formation of 6 is too costly in terms of energy for a cyclic 1:1
adduct to form. The stereoselectivity of the DA reaction
between a acenaphthylene dienophile and an isobenzofuran-
type diene is low. The formation of a cyclic 2:2 adduct is,
therefore, unlikely to occur under step-growth conditions.
7]
AB-type monomer 7, which proved very successful in
polymerizations to give derivatives of polymer 3,[
7a,c]
did not
seem to be a good candidate either. In contrast to other
cyclopentadienones, such as tetracyclone and phencyclone,[
carbonyl-containing adducts of cyclopentene-fused cyclopen-
tadienones are stable only under special conditions.[ Thus, it
is reasonable to assume that, upon self-condensation of two
molecules of 7, carbon monoxide is eliminated faster than the
second condensation step and that cyclization can take place.
After carbon monoxide elimination, the attainment of a
curved conformation again is too costly to form a cyclic dimer.
Cycles with more than two units of 7 are unlikely to form for
probability reasons.
9]
10]
We therefore developed the route to building block 18
(Scheme 3), which should give cycle 19 upon dimerization.
This route combines the following features:
1) According to model studies, dimers of 18 with the correct
stereochemistry should react to give an unstrained 1:1
cycle.
aromatic compounds (PAHs) compared with PAHs with a
polyacene structure. This was a motivation for us to develop a
synthetic strategy towards double-stranded cycles whose
carbon skeletons resembles the belt-regions of fullerenes
(
here: C84 fullerene with D symmetry) and can thus serve as
2
potential precursors for the unsaturated analogues. Other
strategies for fragments of fullerene structures have recently
appeared.[ Here we report on the synthesis of the carbon
skeleton of the equator region of fullerene C84 by cyclo-
dimerization of in-situ-generated isobenzofuran AB-type
Diels ±Alder (DA) monomers, X-ray characterization of the
cyclization product, first steps toward aromatization, and the
influence of strain energy on the outcome of the aromatiza-
tion.
2) The shape of the open-chain dimer, unlike 7, remains
unchanged until cyclization occurs.
3) The ether bridges in the oxanorbornene units of the cycle
19 can potentially be removed under aromatization,
namely, by dehydration.
Dihydropyracylene 12 (Scheme 1) was selected as the
source of the naphthalene unit and the adjacent five-
membered rings of the target cycle 19. Compound 12 is
8]
R
R
a, b, c
Results and Discussion
9
a R = -Br
8
9
b R = -CH2OH
The polymerization of the in-situ-prepared AA-type DA
monomer 5 with its BB-counterpart 6, which leads to a
substituted version of polymer 4, is not accompanied by
cyclization. Only linear structures are formed according to gel
9c R = -CH2Br
d
2
Li+
e
R1
R2
O
O
O
12
11
10
R1
R2
Scheme 1. Synthesis of dihydropyracylene 12. a) NBS/DMF; b) BuLi,
7
5
II
paraformaldehyde; c) PBr
3
; d) PhLi; e) BuLi/TMEDA, [Cu (acac)
2
].
R3
R3
R3
known;[11] however, it has not been used in synthetic
chemistry, presumably because of its limited accessibility.
Only its deprotonation has been studied so far.[ To make
larger quantities of this hydrocarbon available, a route was
R3
12]
6
R1 =- (CH2)12 - ; -COOC12H25 ; -C6H4-C12H25
R2 = -C6H13 R3 = - (CH2)12 -
[11c]
devised which is similar to the one developed by Trost et al.
2746
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2745 ±2757

Belt Region of Fullerenes
2745±2757
However, diol 9b was prepared through the intermediate
dibromide 9a, which is readily available from direct bromi-
nation of acenaphthene 8;[13] this saves two steps. The final
dehydrogenation reaction from 10 to 12 with dichlorodicyano
p-benzoquinone (DDQ) was reported to proceed with 42%
yield. Despite numerous attempts, we were unable to
reproducibly achieve yields higher than 10 ±15%. Therefore,
we decided to synthesize compound 12 via 11 by a dilithiation/
oxidation sequence.[ Dianion 11 was obtained by depro-
tonation with a slight excess of butyl lithium in the presence of
tetramethylethylene diamine (TMEDA) and oxidized with
14]
Cu(ii) acetylacetonate [Cu(acac) ]. These reactions are easy to
2
perform and gave compound 12 in 70 ±75% yield on the 10 g
scale. Evidence for the stabilization of dianion 11 as a
cyclopentadienyl ±allyl system is provided by a dinuclear
iron carbonyl complex of acenaphthylene.[ These modifi-
cations rendered this important hydrocarbon a syntheti-
cally accessible compound and its chemistry can now be ex-
plored.
15]
The synthesis of the AB-type DA-monomer precursors 17 is
shown in Scheme 2. Under standard conditions, compound 12
[16]
and the isobenzofuran precursor 13 gave the endo and exo
1
isomers of 14 in a ratio of 4:1 ( H NMR) and a total yield
>
90%. They were separated on the 10 g scale by column
chromatography and obtained as analytically pure com-
pounds. Subsequent dehydrogenation to 15 was carried out
with DDQ and proceeded reproducibly with yields of 70 ±
8
0%. Both diastereomers of 15 were now independently
[17]
subjected to furan addition to furnish the four diastereo-
meric oxanorbornenes 16 in the following ratios: endo-
1
syn:endo-anti ˆ 1.5:1 and exo-syn:exo-anti ˆ 1:1 ( H NMR).
The yields of the final tetraphenylcyclopentadienone (tetra-
cyclone) additions to the oxanorbornenes 16 to give 17 were
not quite as high as expected because of a side reaction
involving the attack of tetracyclone at the acenaphthylene
termini as well in up to 20% yield. The resulting byproduct
(
not shown) was isolated and characterized for endo-syn-17.
Nevertheless, the endo-isomers of 17 were isolated on the
scale of a few grams and the exo-isomers on 0.5 g scale. All
stereoisomers of compounds 14 ± 17 were isolated by column
1
chromatography and fully characterized on the basis of H
and 13C NMR spectroscopy, elemental analysis, and/or high-
[18]
resolution mass spectrometry. The stereochemical assign-
ment of isomers 14 rests upon common Karplus consider-
ations of the dihedral HCCH angles at the ether bridge. This
was further confirmed by the structures of the 22 isomers in
the crystal (Figure 4). The assignment of the stereoisomers of
Scheme 2. Synthesis of the monomer precursors 17. a) heat, toluene;
b) DDQ, toluene; c) BuLi, furan; d) tetracyclone, EtOH.
1
6 is based upon NOE measurements for the endo pair and on
general.[20] This was confirmed for endo-syn-17 by single-
crystal X-ray crystallographic analysis (Figure 1).
coupling pattern of the a-CH groups for the exo pair. In
2
analogy to related anthracene epoxides (not shown),[ this
group appears as a triplet for exo-syn-16 and as AB-part of an
ABCD spin system for exo-anti-16. The coupling patterns are
further complicated because of reduced symmetry by the
attached pyracene units of 16; however, the initial coupling
pattern remains visible. The stereochemistry of the isomers of
19]
Cyclization reactions: Each of the four isomers of 17 were
individually subjected to cyclization. The same macrocycle 19
was formed via the in-situ-generated AB-monomers endo-18
and exo-18 (Scheme 3). The cartoon-like representation in
Figure 2 illustrates this important point. It shows the two
different stereoisomers of 18 in the side view. Their olefinic
termini are indicated by a double line and the isobenzofuran
termini by an oxygen atom. Their configuration is indicated by
both the kinks in the structure and the position of the oxygen
1
7 follows from the fact that all isomers of 16 were
individually treated with tetracyclone. The addition of tetra-
cyclone to the oxanorbornene termini proceeded stereospe-
cifically in an exo fashion with trans-standing oxygen atoms in
Chem. Eur. J. 2003, 9, 2745 ±2757
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2747

A. D. Schl¸ter, W. D. Neudorff et al.
FULL PAPER
Figure 2. Schematic representation of the cyclodimerization of the iso-
benzofuran monomers 18 (side view) via the open chain dimers 18a. Circles
mark bonds formed during the reaction of endo-18, squares represent those
of exo-18.
Figure 1. Structure of endo-syn-17 in the crystal (ORTEP plot, 50%
probability, H atoms not drawn). For the numbering scheme: see Table 1.
the four isomers of 17 and, thus, two isomers of 18 could have
led to 32 diastereomeric open-chain dimers, such as 18a. This
would have resulted in grossly reduced cyclization yields, as
all-mixed combinations of endo- and exo-18 do not lead to
cyclizible dimers. In contrast, individual reaction of the
isomers of 18 reduces the number of statistically possible,
diastereomeric open-chain dimers to eight for each of the two
monomers; this statistically increases the cyclization proba-
bility by the factor of two. Of the 32 isomers, two can cyclize,
compared to only one of the eight isomers. Up to 25% yield of
endo-syn-17 endo-anti-17 exo-syn-17
exo-anti-17
O
O
O
R
R
R
R
H
O
18
H
H
H
1
9 was obtained from the endo pair of 17 and up to 45% from
the exo pair. These yields are higher than expected from
simple statistical considerations, because, according to mo-
lecular models, not all of the eight isomers can actually be
formed, especially for the endo case. For the exo case the
higher yield can be attributed to the endo selectivity of the DA
reaction.
endo
exo
Cycle 19 was purified both by size exclusion chromatog-
1
R
raphy (SEC) or column chromatography. Its H NMR spec-
trum showed a set of six signals for the six sets of heterotopic
hydrogen atoms of its macrocyclic unit. Depending on its
configuration, the methine protons of the oxanorbornene
group show one set of two split signals (endo) and one set of
two singlets (exo). The upfield shift to d ˆ 1.5 ppm for one of
the latter is in agreement with their position inside the cavity,
facing the opposite naphthalene unit. The C2h symmetry of the
macrocyclic unit of 19 is also reflected in the broadband-
H
O
O
O
H
H
R
R
H
H
R = -C6H13
H
H
O
H
1
3
decoupled C NMR spectra, in which nine signals for the
aromatic subunits and one set of two signals were found for
the endo and the exo linkages, respectively. EI-MS revealed a
R
19
Scheme 3. Synthesis of the macrocycle 19 by cyclodimerization of the in-
situ-generated AB-monomers 18.
strong molecular ion peak at m/z ˆ 1004 corresponding to
‡
[
M] . Additional confirmation of the structure of 19 was
provided by an X-ray structural analysis (Figure 3) of single
crystals grown from the diffusion of ethanol into a solution of
19 in chloroform. All four oxygen bridges of 19 are on the
outer rim of the molecule. The carbon atoms of the endo-
oxanorbornene groups are 11.6 ä apart from each other; the
tertiary C atoms of the exo-oxanorbonene groups are 3.6 ä
apart. Thus, the cavity is too small to incorporate other
molecules.
bridge. As can be seen, both the cyclization of two molecules
of endo-18 as well of two molecules of exo-18 pass through
different intermediates (both denoted as 18a), but never-
theless give the same cycle 19. This cartoon also makes clear
how important it was to separate the isomers of 14 and
individually convert them into their respective stereoisomers
17. If steric constraints had been disregarded, the reaction of
2748
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2745 ±2757

Belt Region of Fullerenes
2745±2757
Figure 4. Structures of 22 in the crystal (ORTEP plot, 50% probability; H
atoms not drawn). endo-22 crystallizes with two molecules in the
asymmetric unit (only one is shown). For the numbering scheme: see
Table 1.
Figure 3. Structure of 19 in the crystal (ORTEP plot, 40% probability, H
atoms not drawn). For the numbering scheme: see Table 1.
In addition to the above-mentioned higher-than-statistical
yields for 19, another remarkable result is that the cyclization
can be carried out in bulk, in which normally linear polymer-
ization is preferred. When 25 mg samples of neat monomer
precursors 17 were heated to 2008C for 1 h under high-
vacuum conditions (0.02 mbar), cycle 19 was obtained in some
cases. While cyclization could not be detected for endo-syn-17,
endo-anti-17 gave 19 in 10% yield and exo-syn-17 gave cycle
Table 1. Selected bond lengths [ä] and angles [8] for endo-syn-17, 19, 22.
endo-syn-17 endo-22a[a] endo-22b[a] exo-22
1
9
[
b]
C1ÀC2
C2ÀC3
C3ÀC4
1.518(2)
1.560(3)
1.509(3)
1.515(2)
1.547(2)
1.515(2)
1.517(2)
1.566(2)
1.513(2)
1.520(2)
1.550(2)
1.518(1)
1.559(1)
1.512(1)
1.515(1)
1.562(1)
1.513(1)
C5
ÀC6
1.513(2)
1.561(2)
1.515(2)
C6ÀC7
C7ÀC8
C1-C2-C3 116.62(17) 116.42(13)
C2-C3-C4 107.68(14) 106.62(13)
C5-C6-C7 103.88(14) 106.55(12)
C6-C7-C8 118.92(15)
117.59(7)
108.29(6)
116.62(7)
108.24(7)
1
9 in a yield of 40%. These yields are close to those obtained
[18]
in the solution experiments. The yields were determined by
integration of the H NMR spectra of the raw products and by
107.83(9)
112.54(9)
1
C2-C3-O1 100.69(16) 101.30(12)
O1-C3-C4 102.19(14) 102.40(12)
C5-C6-O2 103.01(13) 101.56(11)
O2-C6-C7 102.19(14) 101.83(11)
101.07(7)
101.31(6)
100.86(7)
101.43(7)
SEC. The different behaviors of the endo-17 pair can be
explained by the steric demand of the [2.2.1]bicyclohepten-7-
one group of endo-syn-17. This group blocks the approach of
an endo-18 molecule and forms a dimer that can cyclize. From
the modeled structure of endo-anti-17 (not shown) it becomes
apparent that the [2.2.1]bicyclohepten-7-one group hinders
such an approach far less. To our knowledge, these are the first
examples of ™bulk macrocyclizations∫ for double-stranded
DA monomers.
101.26(9)
101.17(8)
[
a] Two molecules in the asymmetric unit. [b] Atom numbering scheme
according to Figures 1, 3, and 4. This numbering scheme was chosen for
readability reasons and differs from the crystallographic numbering
scheme.
angle is decreased by 3 ±4 8. This is accompanied by a slight
curvature of the aromatic units. These small structural differ-
ences confirm that the structurally closely related cycle 19
should actually be basically strainless.
As mentioned, one reason for the success of the above
synthetic strategy was the unstrained nature of 19. To confirm
this, epoxybenzo[k]fluoranthene 22 (Scheme 4) was prepared
from acenaphthene 20 and the isobenzofuran precursor 21.[
The endo and exo isomers of 22 were separated and
individually characterized by X-ray structural analysis (Fig-
ure 4).
20]
Aromatization experiments: A number of predominately
acid-catalyzed dehydration methods have been reported for
the aromatization of oxygen-bridged compounds. None of
them proved to be generally applicable. The method that
employed p-toluene sulfonic acid monohydrate in a toluene
solution has shown its efficiency with low-molecular-weight
compounds and polymers structurally related to 19.[
Application of this method to 19 led to its complete
consumption and furnished two products, the partly aromat-
ized macrocycle 23 (Scheme 5) and an insoluble dark solid.
Compound 23 was isolated by column chromatography in a
yield of 49%, which is rather low relative to open-chain
structures. The insoluble product has not been characterized
further yet.
Ph
Ph
7b, 21]
O
O
Ph
Ph
20
21
H
H
H
O
O
H
endo
exo
22
1
The H NMR spectrum of 23 is similar to that of 19 except
Scheme 4. Synthesis of diastereomeric epoxybenzo[k]fluoranthenes (22).
for the lack of singlets for the exo-configured oxanorbornene
groups and an additional singlet of the peri-protons of the
ortho-annulated naphthalene unit. Consequently, the
Table 1 gives selected bond lengths and angles for endo-syn-
1
3
17, 19, and 22. Significant deviations of the structural
C NMR spectrum shows eleven signals for the aromatic
parameters of 19 from those for the noncyclic structures are
only found at the exo-oxanorbornene linkages. The C6-C7-C8
bond angle is increased by about 68 and the C5-C6-C7 bond
carbon atoms along with only one set of two signals for the
oxanorbornene linkages assigned to the endo configuration.
EI-MS revealed the required molecular ion peak at m/z ˆ 968.
Chem. Eur. J. 2003, 9, 2745 ±2757
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2749

A. D. Schl¸ter, W. D. Neudorff et al.
FULL PAPER
19
insoluble product is not yet clear and that it may be a follow-
up product of a completely dehydrated material.
p-TosH
The results of the dehydration of 19 may be either
attributed to a high reactivity of 24 towards its conversion
into insoluble material, as is assumed for the [n]cyclacenes, or
explained in terms of strain energy.[ The latter argument can
be qualitatively derived from the X-ray structures of 19 and
R
R
3d]
O
H
R
R
H
R
R
2
3. The dehydration of the exo-oxanorbornene groups causes
H
fewer structural changes than would be the case for the
dehydration of the endo groups. This is accompanied by a
minor change in the strain energy of the product obtained by
dehydration of the exo groups and may explain the unusual
selectivity of the dehydration reaction. Further aromatization
is associated with the transfer from the ellipsoidal shape of 23
to the round shape of 24 during which the major part of the
strain energy has to be introduced. We call this type of strain
energy MSE (macrocyclic strain energy), which originates
from the macrocyclic structure, to distinguish it from the local
strain, induced by the bicyclic oxanorbornene groups and five-
membered rings, that is also present in the noncyclic
compounds 14 ± 18. The consequence of the necessary intro-
duction of MSE during dehydration of the endo-oxanorbor-
nene groups becomes more clear by the first step of the acid-
catalyzed dehydration mechanism. Though not well inves-
tigated, it is reasonable to assume that the oxygen bridge is
opened after protonation (Scheme 6) to form the carbenium
H
O
R
23
R = -C H
R
24
6
13
isolated
Scheme 5. Dehydration of 19.
anticipated
Final confirmation of the structure of 23 was provided by
X-ray structural analysis (Figure 5) of a single crystal grown
from diffusion hexane to a solution of 23 in chloroform. It
H
H+
O
O
OH
OH
-H2O
-H+
Figure 5. Structure of 23 in the crystal (ORTEP plot, 50% probability; H
atoms not drawn).
H
26
25
Scheme 6. Probable mechanism of acid-catalyzed dehydration of oxanor-
bornenes.
shows an ellipsoidally flattened molecule whose benzo[k]-
fluoranthene moieties are slightly bent. The maximum
distance of these moieties is 4.1 ä at the center of the
molecule. The central carbon atoms of the benzo[k]fluoran-
thene unit lie 1.3 ä over the plane that is defined by its four
nearest saturated carbon atoms of the oxanorbornene link-
ages.
ion 25, which in turn gives the cyclohexadienol structure 26
through the loss of a proton. Transferred to 23, the major part
of the MSE has to be raised in the step corresponding to the
transition of 25 to 26. Therefore, the planarization of two
atoms has to take place against the pyramidalization of any of
The seemingly selective aromatization of the exo-oxanor-
bornene groups is somewhat in contrast to earlier reports for
open chain compounds in which the endo-configured groups
react faster than the exo ones.[ This was qualitatively
confirmed by control experiments subjecting a 1:1 mixture of
endo- and exo-22 to the dehydration agent at room temper-
ature. After 1 h, the endo:exo ratio had changed to 1:2.
The lack of any other dehydration product than 23,
including 24 from complete aromatization of 19, was con-
firmed by NMR tube experiments in benzene that showed no
other sharp signals than those for 19 and 23. Applying harsher
conditions to 23 by the use of a large excess of anhydrous
para-toluene sulfonic acid (p-TosH) in refluxing toluene did
not change the overall result, except for a higher reaction rate.
It should be mentioned here, however, that the origin of the
2
the sp -hybridized carbon atoms of the macrocyclic backbone
at a time when the exothermic elimination of water cannot
contribute. It is unlikely that a cyclohexadienol structure, such
as 26, is involved during the acid-catalyzed dehydration of the
endo-oxanorbornene linkages of 19 or 23, and rather destruc-
tive side reactions may take place instead.
21]
On account of this consideration, further attempts with
different methods of acid-catalyzed dehydration have not yet
been undertaken. The use of iodotrimethylsilane (TMSI)[ as
the dehydration agent did not yield 24 or any other
dehydration product. NMR spectroscopic monitoring showed
a very clean reaction course (see the Supporting Information).
Although we were unable to isolate the reaction product or
obtain mass spectrometric evidence, structure 27 was tenta-
22]
2750
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2745 ±2757

Belt Region of Fullerenes
2745±2757
tively assigned on the basis of the following argumentation
and one with two different ortho-annelated subunits named
mixed circocenes (29). These calculations were compared to
earlier reported calculations on the same level of theory for
linear acenes by Notario,[ which showed good correlation
with available experimental data, namely, ionization poten-
tials. In addition, recently reported calculations for [n]cycla-
(
Scheme 7).
24]
R
R
I
H
H
O
O
H
H
H
H
O
H
[25]
cenes (n ˆ 4 ±9) were reevaluated and extended to n ˆ 10 ±
I
H
H
16. Reevaluation was necessary because the reported values
for the [5]cyclacene do not belong to a cyclacene structure, but
rather to a Dewar benzene-type derivative and, therefore,
showed, like the values for [6]cyclacene, a poor correlation to
the values of the other compounds in this series.
H
H
R
R
H
TMSI
H2O
H
H
H I
O
O
O
H
I
R
R
As a result, the values for the circocenes in comparison with
those of linear acenes and [n]cyclacenes did not give an
indication for an exceptionally high reactivity of the circo-
cenes in general and 28 (n ˆ 3) in particular. The HOMO ±
LUMO gaps for circocenes, linear acenes, and cyclacenes are
given in Figure 6.
19
R = -C6H13
27
Scheme 7. Reaction of 27 with TMSI (inner R groups for 27 are not
drawn).
The signals of the protons in the endo-oxanorbornene
group are shifted downfield from d ˆ 5.73 and 4.70 ppm for 19
to d ˆ 6.22 and 4.83 ppm for 27, respectively. These changes
are accompanied by the loss of coupling. For the protons of
the exo group, only the signal at d ˆ 5.13 ppm for 19 is slightly
shifted downfield to d ˆ 5.22 ppm for 27, whereas the other is
shifted upfield from d ˆ 1.50 ppm for 19 to d ˆ 1.10 ppm for
1
3
2
7; both remained as singlets. The C NMR spectrum of 27
shows only one signal for carbons connected to oxygen at d ˆ
2.71 ppm, which is a typical shift for exo-oxanorbornene
8
groups. A further and astounding piece of evidence for the
formation of 27 is its almost complete backreaction to 19 in
the presence of water.
We believe that the MSE, which has to be mainly built up in
the final aromatization steps, is the dominant factor for the
not yet surmounted problems with these very steps. As
mentioned above, another reason may also be high reactivity
of the completely aromatized 24 towards attack by electro-
philes or polymerization. To obtain more insight into this
matter, semiempirical AM1[ calculations were performed.
In this context a simple nomenclature was derived for
structures such as 24 that is also applicable to various other
belt-shaped PAHs (these compounds are called circocenes) as
well as for oligomeric and polymeric open-chain PAHs (see
the Supplementary Information).
Calculations: Semiempirical AM1[ calculations on the
unsubstituted parent compound and a series of eight homo-
logues (28 and 29) were performed to estimate the reactivity
of 24. These compounds belong to two series, one with two
equal ortho-annelated subunits called simple circocenes (28)
Figure 6. AM1 calculated HOMO ±LUMO gaps of linear [ n]acenes (n ˆ
[24]
1
±12, 15), [n]cyclacenes (n ˆ 4 ±16), simple circocenes ( n ˆ 1 ±5), mixed
circocenes (m ˆ 1 ±4).
23]
The HOMO ±LUMO gap can be correlated to the Diels ±
Alder reactivity of PAHs.[ The HOMO ±LUMO gaps for all
circocenes are in the range of those of tetracene and
hexacene. The gap for 28 (n ˆ 3) is similar to that of
pentacene. The increase of the HOMO ±LUMO gaps for
the simple circocenes up to 28 (n ˆ 3) is similar to earlier
reported calculations for polyfluoranthenes by Kertesz,[ and
is predominately a consequence of increasing LUMO ener-
gies.[ In addition to these electronic properties, the gas-
phase proton affinities (PA) for the circocenes were calcu-
lated for several protonation sites (see the Supplementary
Information) according to Equation (1) and with the exper-
24]
23]
26]
27]
‡
À1 [24]
imental value DH (H ) ˆ 367.2 kcalmol .
f
‡
‡
PA ˆ DH
f
(B) ‡ DH
f
(H ) À DH
f
(BH )
(1)
n
m
2
8
29
À1
The PA of 28 (n ˆ 3) was found to be 220.4 kcalmol for
protonation at the 9-position of the anthracene unit. This
value lies between the calculated values for tetracene
À1
À1 [24]
(
215.9 kcalmol ) and pentacene (223.5 kcalmol ).
For
n
m
n = 1-5
m = 1-4
comparison, the PAs of [n]cyclacenes with odd n were
simple circocenes
mixed circocenes
À1 [28]
calculated to be in the range of 252 ±257 kcalmol
.
Chem. Eur. J. 2003, 9, 2745 ±2757
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2751

A. D. Schl¸ter, W. D. Neudorff et al.
FULL PAPER
Actually, a high reactivity of the circocenes cannot be ruled
out for the failure of complete aromatization of 19 by this
simple comparison on this level of theory, but it should be far
less important than for the [n]cyclacenes.
considered to be a part of the MSE. The change in resonance
energy for A ! A was not calculated explicitly. Therefore,
1
0
the MSE of A differs by this amount.
0
The calculated MSEs for A are given in Table 2. The MSE
r
À1
The MSE, which has to be built up during aromatization to
of A is 12 kcalmol and thus very low. This is in good
4
24 does indeed seem to play a significant role in this context.
agreement with the result from the crystallographic analysis.
The MSE was calculated for the unsubstituted parent com-
pounds A (Figure 7) according to Equation (2).
r
Table 2. AM1-calculated macrocyclic strain energies (MSE).
A
4
A
3
A
2
A
1
A
0
MSE
12.0
16.0
16.5
85.7
122.5
À1
The MSE of A is 122.5 kcalmol ; this indicates that every
0
À1
five- and six-membered ring has to carry ꢀ9 kcalmol of the
MSE. This is not an exceptionally high value compared to
other compounds.[ The critical point is the change in MSE
on going from A ! A . In this step, the first endo-oxanor-
29]
2
1
bornene group is aromatized. Approximately 57% of the total
MSE of A has to be built up during this step. This is also
0
reflected by the heats of reaction (Table 3) calculated with the
À1 [30]
use of the experimental DH (H O(gas)) ˆ 57.8 kcalmol .
f
2
Table 3. AM1-calculated heats of reaction.
Reaction
À1
r
DH [kcalmol ]
A
A
A
A
A
4
3
2
1
4
! A
! A
! A
! A
! A
3
2
1
0
0
‡ H
‡ H
‡ H
‡ H
2
2
2
2
O
O
O
O
À 24.6
À 26.6
‡ 44.6
‡ 14.9
‡ 8.3
Figure 7. Structures of model compounds A
to calculate the development of the MSE during dehydration of 19.
r
(rˆ number of ether bridges)
‡ 4 H
2
O
MSE ˆ DH
f
(A
r
) À DH
f
(C
r
) À DHr(split) À DEres
(2)
The first two dehydration steps of the exo-oxanorbornene
In Equation (2) r denotes the number of ether bridges and
C are open-chain model compounds derived from A by a
hypothetical reductive cleavage of the macrocyclic molecules
À1
groups are exothermic with ꢀ25 kcalmol . This is a similar
value to those values calculated for all steps of the dehydra-
tion of the open-chain model compounds Cr (see the
Supporting Information). On the other hand, the dehydration
of the first endo group is the most endothermic step with
r
r
at an endo linkage (Scheme 8). These model compounds are
H
H
H
H
H
H
À1
+
2H2
45 kcalmol . The endothermic dehydration of both endo
A4
O
O
O
O
groups makes the overall aromatization slightly endothermic.
These results can be correlated with the experimental
observations:
C4
H
1) The lack of NMR spectroscopic evidence of a single
dehydration product of 19 and the formation of 23 is in
good agreement with the calculated exothermicity of these
dehydration reactions, accompanied by minor changes in
the MSE.
+
2H2
O
+
O
H
Scheme 8. Hypothetical reaction to derive noncyclic model compounds C
of A (above). Sample of a model reaction to calculate DHr(split) (below).
r
r
2
) The lack of an indication for a dehydration product of an
endo-oxanorbornene group correlates with the calculated
endothermy of these reaction steps, which is caused by a
strong increase of the MSE.
considered to be free of MSE, which was examined on a
second set of model compounds (see the Supporting Infor-
mation). DHr(split) is the enthalpy of the cleavage reaction. The
value for DHr(split) was calculated as the average from a series
of five model reactions (one example is given in Scheme 8) to
3) The failure of HI elimination of 27 and its back reaction to
19 can also be explained by the necessary increase of the
MSE.
Therefore, we consider the build-up of MSE to be an
important factor with respect to the lack of complete
aromatization of 19 under the applied conditions. Because
all the precursors prepared so far for the all-carbon, double-
À1
À1
be 34.2 Æ 1.2 kcalmol (for DH (H
) ˆ 0 kcalmol ). DE
f
2(gas)
res
is the change in resonance energy on extending the p system
in every dehydration step. Except for A , this value equals
0
zero because A1±4 and C1±4 have the same p system, and that
part of DEres which results from a bending of the p system is
2752
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2745 ±2757

Belt Region of Fullerenes
2745±2757
stranded, belt-shaped aromatics have to pass a flat inter-
mediate structure similar to 23 during complete aromatiza-
tion. The problem of building up MSE may be a general one
for the synthesis of belt-shaped aromatics to succeed. The
work by Cory also points in this direction.[
indicate the required build-up of strain energy during the
conversion of 23 to the fully unsaturated belt 24 as a main
reason for the incomplete aromatization. In this context a
simple to use so-called [f.p.q] nomenclature for a wide variety
of open-chain and cyclic PAHs was suggested.
3b]
Conclusion
Experimental Section
The belt-shaped compound 19 is the first representative of a
double-stranded macrocycle with a carbon skeleton similar to
the equatorial region of a fullerene and thus is a potential
precursor for the fully unsaturated analogue 24. Macrocycle
X-ray crystal structures: Crystals of suitable for X-ray diffraction measure-
ments were grown from ethanol (endo-syn-17, endo-22, exo-22) and by
diffusion of ethanol or hexane into a solution of the compound in
chloroform (19 and 23, respectively). Crystals of 23 contained solvent
molecules and therefore tend to disintegrate with loss of the solvent outside
saturated solutions, therefore the crystal was mounted at low temperature
on the tip of a glass fiber. The data of 19 were collected on a Enraf Nonius
Turbo CAD4 at ambient temperature with CuKa irradiation. The data
collection of endo-syn-17, endo-22, exo-22, and 23 was performed at low
temperature with a BRUKER-AXS SMART CCD diffractometer (MoKa).
A total of 600 frames (Dw ˆ 0.38) for each run were collected for three f
19 was constructed from bifunctional, in-situ-generated
Diels ±Alder building blocks endo-18 and exo-18; this afford-
ed 19 in yields of up to 45%. Insights into stereochemical
details of the sequence and its implications regarding the
course of cyclization are discussed. In order to make these
building blocks conveniently available, the synthesis of the
known dihydropyracylene (12), which is a sequence key
compound, was optimized so that it is now available on the
positions (08, 908, and 2408) resulting in 1800 frames for each data set. The
2
data were reduced to F
o
and corrected for absorption effects by using
SAINT[ and SADABS, respectively. The structures were solved with
31]
[32]
10 g scale in a few simple steps. This improvement not only
2
direct methods and refined with full-matrix least-squares on
SHELXL97[33]). The quality of the structure determination of 23 was
strongly influenced by the fact that the crystal contained disordered
chloroform solvate molecules and that one side chain also showed some
disorder. The electron density of the solvate molecules was taken into
account by means of the squeeze option of the program package
F
had considerable impact on the sequence described here, but
also makes 12 a valuable starting material for the future
synthesis of fullerene parts and related compounds, a
presently heavily pursued research area. Attempts to com-
pletely aromatize 19 yielded a partly aromatized macrocycle
(
[
34]
PLATON. Details on the data collection and structure refinement are
23, which, like 19, was characterized by single-crystal X-ray
listed in Table 4. ORTEP[ for Windows was used for to prepare the
35]
diffraction. It also furnished an insoluble material, which will
be investigated further. Semiempirical AM1 calculations
[36]
graphical representations. WINGX
computing.
was used for the crystallographic
Table 4. Crystal data and structure refinement for 19, endo-syn-17, endo-22, exo-22, and 23.
Compound
19
endo-syn-17
endo-22
exo-22
23
formula
C
72
H
76
O
4
C
67
H
60
O
3
C
20
H
14
O
C
20
H
14
O
72 72 2
C H O
M
T [K]
r
1005.32
293(2)
913.15
153(2)
270.13
133(2)
270.13
133(2)
969.30
163(2)
l [ä]
1.54060
orthorhombic
Pbca
10.176(5)
16.040(5)
33.421(5)
90
90
90
5455(3)
4
1.224
0.567
2160
0.8 Â 0.7 Â 0.3
2.64 ±59.91
À 11 ꢁ h ꢁ 11
À 18 ꢁ k ꢁ 0
À 37 ꢁ l ꢁ 0
7653
0.71073
monoclinic
P2 /c
20.059(2)
15.356(3)
17.470(3)
90
112.745(12)
90
4962.6(12)
4
1.222
0.073
1944
1.0 Â 0.82 Â 0.07
1.83 ±26.36
À 25 ꢁ h ꢁ 23
À 19 ꢁ k ꢁ 19
À 21 ꢁ l ꢁ 21
45808
0.71073
triclinic
P1
0.71073
orthorhombic
0.71073
monoclinic
P2 /c
14.8104(6)
13.7155(5)
14.4770(5)
90
95.2100(10)
90
2928.59(19)
2
1.099
0.064
1040
0.3 Â 0.24 Â 0.08
1.38 ±23.25
À 16 ꢁ h ꢁ 16
À 15 ꢁ k ꢁ 15
À 16 ꢁ l ꢁ 16
22082
crystal system
space group
a [ä]
b [ä]
c [ä]
≈
1
Pna2
1
1
9.1071(6)
12.3014(9)
12.9394(9)
73.1020(10)
76.9600(10)
87.076(2)
1351.08(16)
4
1.329
0.080
568
1.0 Â 0.6 Â 0.32
1.69 ±33.17
À 13 ꢁ h ꢁ 13
À 18 ꢁ k ꢁ 18
À 19 ꢁ l ꢁ 19
36325
11.7977(9)
21.8171(16)
5.2400(4)
90
90
90
1348.73(18)
4
1.331
0.080
568
1.6 Â 0.32 Â 0.2
1.87 ±33.10
À 17 ꢁ h ꢁ 17
À 33 ꢁ k ꢁ 32
À 7 ꢁ l ꢁ 7
17520
a [8]
b [8]
g [8]
V [ä ]
3
Z
À3
1
calcd [mgm
]
À1
m [mm
F(000)
crystal size [mm ]
q range [8]
]
3
index ranges
reflections collected
independent reflections [R(int)]
completeness to q
data/restraints/parameters
quality-of-fit on F 2
4039 [0.0856]
59.918, 99.9%
4039/460/374
1.067
10134 [0.0296]
26.368, 99.9%
10134/0/633
1.013
9714 [0.0282]
33.178, 94.4%
9714/0/379
1.006
4788 [0.0538]
33.108, 95.8%
4788/1/190
1.075
4209 [0.060]
23.258, 100%
4209/0/336
1.541
final R indices [I > 2s(I)] R
R indices (all data) R /wR
largest diff. peak/hole [eä
1
/wR
2
0.0486/0.1261
0.0526/0.1298
0.194/ À 0.193
0.0434/0.1150
0.0605/0.1292
0.444/ À 0.357
0.0446/0.1307
0.0522/0.1379
0.459/ À 0.261
0.0493/0.1215
0.0521/0.1251
0.366/ À 0.244
0.1209/0.3627
0.1479/0.3861
2.369/ À 0.616
1
2
À3
]
Chem. Eur. J. 2003, 9, 2745 ±2757
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2753

A. D. Schl¸ter, W. D. Neudorff et al.
FULL PAPER
CCDC-202685 (endo-syn-17), CCDC-202682 (19), CCDC-202681 (exo-22),
CCDC-202684 (endo-22), CCDC-202683 (23) contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cam-
bridge Crystallographic Data Centre, 12 Union Road, Cambridge
CB21EZ, UK; fax: (‡44)1223-336033; or deposit@ccdc.cam.uk).
Bis(bromomethyl)acenaphthene (9c): Compound 9c was prepared from
[
11c]
9b (20 g, 93 mmol) according to a published procedure
in 83% yield
(26.2 g). M.p. 157 ±159 8C (decomp) (lit:[
10c]
157 ±159 8C); H NMR: d ˆ
1
3
3.33 (s, 4H; H-1,2 ), 5.28 (s, 4H; CH
2
Br), 7.26 (d, J ˆ 7.23 Hz, 2H; H-3,8),
3
13
7.55 ppm (d, J ˆ 7.23 Hz, 2H; H-4,7); C NMR: d ˆ 30.16 (C-1,2), 36.81
(CH
2
Br), 119.96, 127.75, 129.44, 134.02, 140.92, 149.34 ppm (arom-C)
Computation: All calculations were performed with the HyperChem6.02
Pyracene (10): This was prepared by a modified Trost method:[11c]
A
[
37]
package. The AM1 method at the restricted Hartree ±Fock (RHF) level
and conjugate gradient minimization, Polak ±Ribiere approach, were
applied to obtain the optimized geometries. The convergence limit and
solution of 9c (28 g, 82 mmol) in dry diethyl ether (120 mL) was added to
an ice-cooled mixture of PhLi (50 mL, 90 mmol, 1.8m in cyclohexane/
diethyl ether) over a period of 10 min. The mixture was stirred for 4 h and
was then quenched with water (10 mL). The organic solvent was
evaporated, the residue dissolved in CH Cl (1 L), and the organic layer
À4
À1
gradient values (RMS) were maintained below 10 kcalmol
and
À1
À4
À1
0
0
.01 kcalä for macrocyclic structures and below 10 kcalmol and
2
2
À1
.1 kcalä for open-chain structures. Minimum structures for [5]cyclacene
was washed with water. Drying and evaporation of the solvent yielded
crude 10 which was further purified by digestion in warm diethyl ether.
After cooling to room temperature the precipitate was filtered and dried in
vacuum. This procedure was repeated once to yield 13.7 g (93%) of a
and [6]cyclacene could be obtained starting from preoptimization with the
implemented mm ‡ force-field followed by AM1 optimization applying a
À5
À1
À6
À1
convergence limit and an RMS of 10 kcalmol and 10 kcalä
,
respectively. These minimum structures were checked by frequency
analysis.
colorless solid (m.p. 205 ±209 8C), which was suitable for further work.
Sublimation of a sample raised the melting point to 212 ±216 8C (lit.:[11c]
1
General preparative methods: All readily available reagents were used
without further purification. Solvents were purified and dried by standard
procedures. All reactions were carried out under nitrogen. NMR spectra
214 ±217 8C); H NMR: d ˆ 3.42 (s, 8H; H-1,2,5,6), 7.19 ppm (s, 4H;
1
3
H-2,3,7,8); C NMR: d ˆ 31.60 (C-1,2,5,6), 120.33, 138.35, 140.81 ppm
‡
(arom-C); MS (80 eV, EI, 1008C): m/z (%): 180 (100) [M] .
were recorded in CDCl
3
(unless otherwise stated) either on Bruker WH
1
,2-Dihydropyracylene (12): Under nitrogen atmosphere, a solution of 10
2
70 MHz or AC 500 MHz spectrometers, with the solvent as the internal
(
13.7 g, 76 mmol) and TMEDA (30 mL) in dry cyclohexane (250 mL) was
standard. Mass spectra (MS) were obtained from Varian MAT771 or
MAT112S, respectively. The matrix used for fast atom bombardment
prepared. BuLi (110 mL of a 1.6m solution in hexane) was added to this
solution over a period of 10 min at room temperature. The color changed
quickly to dark red and then to dark green. After refluxing for 1 h, the
mixture was allowed to cool for 15 min and then added dropwise to an ice-
(
FAB) spectra was 3-nitrobenzyl alcohol (NOBA). Column chromatog-
raphy was carried out on silica gel60 (230 ±400 mesh, Merck). Thin-layer
chromatography (TLC) was performed on aluminum sheets coated with
Merck5554 silica gel60F, visualization by an ultraviolet (UV) lamp (l ˆ
cooled suspension of [Cu(ii)(acac)
2
] (50 g, 190 mmol) in dry cyclohexane
(
150 mL) over a period of 15 min. The mixture was stirred for another
2
54 nm and l ˆ 366 nm). Analytical size exclusion chromatography (SEC)
3
2 3
0 min and then poured onto a short column (Al O , basic, activity grade V,
was performed on Waters Assoc. 150-c Alc/GPC chromatograph on Waters
StyragelHR columns and with tetrahydrofuran (THF) as the mobile phase.
A Waters484 UV/Vis detector was used with a polystyrene standard.
Melting points were determined on a B¸chi500 and are uncorrected.
height of column 5 cm) and washed with hexane until the filtrate became
colorless. The solvent was evaporated in a vacuum, and the crude yellow-
brown product was purified by column chromatography (silica gel, hexane/
ethyl acetate 10:1, R
solid (m.p. 154 ±156 8C) (lit:
f
ˆ 0.57) to yield 10.01 g (74%) of a yellow crystalline
[
11c]
1
5
,6-Dibromoacenaphthene (9a): This was prepared by a modified Kasai
156 ±157 8C); H NMR: d ˆ 3.50 (s, 4H;
[
13]
3
method: a suspension of N-bromosuccinimide (NBS) (1 kg, 5.62 mol) in
DMF (2 L) was added in portions to an ice-cooled suspension of
acenaphthene (400 g, 2.59 mol) in DMF (500 mL) over a period of 5 h.
The temperature of mixture was not allowed exceed 158C. The mixture was
stirred for a further 12 h and then allowed to warm to room temperature.
The precipitate was filtered with suction, washed with ethanol (3 Â
H-1,2), 7.16 (s, 2H; H-5,6), 7.41 (d, J ˆ 6.9 Hz, 2H; H-3,8), 7.77 ppm (d,
3
13
J ˆ 6.9 Hz, 2H; H-4,7); C NMR: d ˆ 32.37 (C-1,2), 120.46 (C-3,8), 126.01
(C-3,4,7,8), 127.93 (C-8c), 128.22 (C-5,6), 134.90 (C-8b), 135.22 (C-4a,6a),
146.60 ppm (C-2a,8a).
1,2,4b,5,10,10a-Hexahydro-7,8-dibromo-5,10-epoxy-6,9-dihexylbenzo[k]cy-
clopenta[c,d]fluoranthene (14): A solution of 12 (3.56 g, 19.8 mmol) and
1
5
00 mL), and purified by stirring over night in refluxing ethanol (3 L).
[
16]
3
(16.9 g, 19.8 mmol) in toluene (300 mL) was prepared and refluxed for
Cooling to room temperature, filtration, washing with ethanol, and drying
in vacuo yielded 206 g (25%) of a beige crystalline solid (m.p. 169 ±172 8C)
that was suitable for further work. Recrystallization of a sample in
2
4 h. After cooling to room temperature, the solvent was evaporated in
vacuo. To remove a large part of the tetraphenylbenzene byproduct
(
(
ꢀ80%), the residue was suspended in a 2:1 mixture of hexane/toluene
[
12]
chloroform raised the melting point to 173 ±175 8C (lit.: 174 ±176 8C).
50 mL) and stirred for 30 min. The suspension was filtered with suction,
1
3
H NMR: d ˆ 3.28 (s, 4H; H-1,2), 7.06 (d, J ˆ 7.49 Hz, 2H; H-3,8), 7.76 ppm
and the residue washed with a 2:1 mixture of hexane/toluene (2 Â 25 mL).
The combined organic layers were concentrated to dryness in vacuo. The
isomers were purified and separated by column chromatography (silica gel,
3
13
(
1
d, J ˆ 7.49 Hz, 2H; H-4,7); C NMR (68 MHz, CDCl
3
): d ˆ 29.99 (C-1,2),
14.31, 120.87, 131.80, 135.77, 141.75, 147 ppm (arom-C)
5
,6-Bis(hydroxymethyl)acenaphthene (9b): A suspension of 9a (40 g,
hexane/toluene 2:1) to give 2.12 g of exo-14 (m.p. 126 ±127 8C, R
and 8.61 g of endo-14 (m.p. 92 ±94 8C, R
ˆ 0.14) in a combined yield of
7%.
f
ˆ 0.20)
1
29 mmol) in dry diethyl ether was prepared under nitrogen and cooled to
f
À308C. BuLi (1.6m, 170 mL, 272 mmol) in hexane was added within
9
1
0 min. After further 30 min of stirring, paraformaldehyde (8 g, dried over
exo-14: 1H NMR: d ˆ 0.96 (t, 6H, J ˆ 7.1 Hz ; CH
3
), 1.3 ±1.6 (m, 12H;
3
P
2
O
5
) was added at once. The mixture was stirred for 3 h at À308C and for
alkyl-CH
3
2
), 1.73 (m, 4H; b-CH
2
), 2.95 (m, 4H; a-CH
2
), 3.43 (s, 4H; H-1,2),
a further 16 h while warming to room temperature. The reaction was
quenched by adding HCl (40 mL, 25%). After 1 h of stirring, the colorless
precipitate was filtered with suction, suspended in diethyl ether (500 mL),
filtered again, and dried in vacuum at 508C. The dry material was powdered
and suspended in 1% HCl for 2 h, filtered, washed with ethanol (2 Â
3
.92 (s, 2H; H-4b,10a), 5.41 (s, 2H; H-5,10), 7.28 (d, J ˆ 6.9 Hz, 2H;
3
13
H-3,12), 7.37 ppm (d, J ˆ 6.9 Hz, 2H; H-4,11); C NMR (125 MHz): d ˆ
1
8
4.12, 22.61, 29.46, 31.56, 31.69, 35.09 (alkyl-C and C-1,2), 55.14 (C-4b,10a),
3.44 (C-5,10), 120.67, 126.16, 134.34, 138.13, 140.41, 142.49, 144.56 ppm
‡
(
C
arom-C); MS (80 eV, EI, 2408C): m/z (%): 622 (5.6) [M] , 440 (100) [M À
1
00 mL), and dried again in vacuum at 508C yielding 20.5 g (74%) of a
‡
‡
‡
14
H
10] , 365 (62.4) [M À C14
H
10 À Br] 178 (69.4) [C14
H10] ; elemental
colorless solid (m.p. 201 ±203 8C), which was suitable for further work. Two
recrystallizations from dioxane raised the melting point to 201 ±204 8C
analysis calcd (%) for C34
H 6.29.
H
38Br
2
O (622.47): C 65.60, H 6.15; found: C 65.89,
[
11c]
1
(
(
lit.:
202 ±205 8C). H NMR ([D
6
]DMSO): d ˆ 2.15 (s, 4H; H-1,2), 3.85
d, 3J ˆ 5.3 Hz, 4H; CH
OH), 4.25 (t, J ˆ 5.3 Hz, 2H; OH), 6.09 (d, J ˆ
3
3
endo-14: H NMR: d ˆ 0.91 (t, J ˆ 6.6 Hz, 6H; CH
1
3
2
3
), 1.2 ±1.6 (m, 16H;
), 3.27 (m, AA'BB', 4H; H-1,2),
3
3
7
.0 Hz, 2H; H-3,8 or H-4,7), 6.35 ppm (d, J ˆ 7.0 Hz, 2H; H-3,8 or H-4,7);
alkyl-CH
2
), 2.48 (t, J ˆ 7.7 Hz, 4H; a-CH
2
1
3
3
H NMR ([D
6
]DMSO/D
2
O): 1.96 (s, 4H; H-1,2), 3.67 (s, 4H; CH
2
OH), 5.95
4.65 (m, J ˆ 3.4 Hz, 2H; H-4b,10a), 5.73 (m, J ˆ 3.4 Hz, 2H; H-5,10), 7.00
3
3
3
3
(
d, J ˆ 7.0 Hz, 2H; H-3,8 or H-4,7), 6.34 ppm (d, J ˆ 7.0 Hz, 2H; H-3,8 or
(d, J ˆ 6.9 Hz, 2H; H-3,12), 7.14 ppm (d, J ˆ 6.9 Hz, 2H; H-4,11);
1
3
13
H-4,7); C NMR ([D
6
]DMSO): d ˆ 29.46 (C-1,2), 63.05 (CH
2
OH), 118.96,
C NMR (125 MHz): d ˆ 14.13, 22.63, 28.81, 29.53, 31.38, 31.62, 35.22
1
1
H
28.53, 129.70, 134.11, 140.24, 145.92 ppm (arom-C); MS (80 eV, EI,
(alkyl-C and C-1,2), 54.11 (C-4b,10a), 80.85 (C-5,10), 119.80, 120.76, 125.27,
134.41, 135.71, 138.09, 139.51, 14.72, 142.51 (arom-C); MS (80 eV, EI,
2108C): m/z (%): 622 (6.8) [M] , 444 (100) [M À C14
‡
‡
008C): m/z (%): 214 (30) [M] , 196 (100) [M À H
2
O] , 167 (69) [M À
‡
‡
‡
2
O À HCO] .
H
10] , 365 (52) [M À
2754
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2745 ±2757

Belt Region of Fullerenes
2745±2757
isomers were separated by column chromatography (silica gel, CH
2
Cl
ˆ 0.27)
2
) to
‡
‡
C
C
14
H
H
10 À Br] 178 (51) [C14
H10] ; elemental analysis calcd (%) for
give 424 mg of exo-syn-16 (R
in a total yield of 50%.
f
ˆ 0.15) and 457 mg of exo-anti-16 (R
f
34
2
38Br O (622.47): C 65.60, H 6.15; found: C 65.56, H 6.13.
4
b,5,10,10a-Tetrahydro-7,8-dibromo-5,10-epoxy-6,9-dihexylbenzo[k]cyclo-
penta[c,d]fluoranthene (15):
exo-anti-16: 1H NMR (500 MHz, CDCl
CH ), 1.45 (m, 14H; alkyl-CH ), 1.67 (m, 2H; b-CH
): d ˆ 0.95 (t, 6H, J ˆ 6.7 Hz;
3
3
3
2
2
), 2.71 (m, 2H; a-
endo-15: DDQ (3.2 g, 14 mmol) was added to a refluxing solution of endo-
4 (8.6 g, 13.8 mmol) in toluene (150 mL). After refluxing for 90 min, the
solution was allowed to cool for 15 min and was then concentrated to a
volume of 80 mL in vacuo. The product was isolated from this solution by
column chromatography (silica gel, toluene) as 7.3 g of a yellow solid (R
CH2), 2.81 (m, 2H; a-CH2), 3.95 (s, 2H; H-4b,12a), 5.38 (s, 2H; H-7,10),
1
3
5
6
.83 (s, 2H; H-5,12), 7.03 (s, 2H; H-8,9), 7.13 (s, 2H; H-1,2), 7.51 (d, J ˆ
3
13
.8 Hz, 2H; H-4,13), 7.78 ppm (d, J ˆ 6.8 Hz, 2H; H-3,14); C NMR
(
125 MHz, CDCl
3
): d ˆ 14.15, 22.60, 29.28, 30.24, 31.39, 31.71 (alkyl-C),
f
ˆ
5
1
6.12 (C-4b,12a), 81.04 (C-5,12 or C-7,10), 81.87 (C-7,10 or C-5,12), 120.62,
26.08, 126.95, 129.04, 136.46, 137.21, 142.14, 142.92, 144.42, 147.15 ppm
1
3
0
(
2
.54) in 85% yield; H NMR: d ˆ 0.96 (t, J ˆ 6.5 Hz, 6H; CH
3
), 1.2 ±1.6
3
4
m, 16H; alkyl-CH
2
), 2.54 (t, J ˆ 7.6 Hz, 4H; a-CH
2
), 4.71 (m, J ˆ 3.7 Hz,
‡
(
C
arom-C); MS (80 eV, EI, 2208C): m/z (%): 528 (1.65) [M] , 352 (100) [M À
4
H; H-4b,10a), 5,74 (m, J ˆ 3.7 Hz, 2H; H-5,10), 6.96 (s, 2H; H-1,2), 7.27
‡
‡
8
14
H
8
] , 176 (9.52) [C14
H
] ; HRMS calcd for C38
H
40
O
2
: 528.302831; found:
3
3
(
d, J ˆ 6.9 Hz, 2H; H-4,11), 7.51 ppm (d, J ˆ 6.9 Hz, 2H; H-3,12);
5
28.30475.
1
3
C NMR: d ˆ 14.10, 22.63, 28.85, 29.53, 31.64, 35.30 (alkyl-C), 54.96 (C-
b,10a), 80.32 (C-5,10), 120.83, 125.27, 125.68, 127.16, 129.00, 134.45, 136.55,
36.67, 141.23, 141.45 ppm (arom-C and C-1,2); MS (80 eV, EI, 1308C): m/z
1
3
exo-syn-16: H NMR (500 MHz, CDCl
3
): d ˆ 0.96 (t, 6H, J ˆ 6.7 Hz, CH
3
),
4
1
(
1
.42 (m, 12H; alkyl-CH ), 1.68 (m, 4H; b-CH
2
2
), 2.79 (m, 4H; a-CH
2
), 3.99
‡
‡
‡
(s, 2H; H-4b,12a), 5.40 (s, 2H; H-7,10), 5.82 (s, 2H; H-5,12), 7.12 (s, 2H;
%): 620 (5.6) [M] , 444 (71.6) [M À C14
H
8
] , 365 (100) [M À C14
H
8
À Br] ;
3
H-8,9), 7.15 (s, 2H; H-1,2), 7.54 (d, J ˆ 6.8 Hz, 2H; H-4,13), 7.80 ppm (d,
elemental analysis calcd (%) for C34
2
H36Br O (620.46): C 65.82, H 5.85;
3
13
2
2
H, J ˆ 6.8 Hz; H-3,14); C NMR (125 MHz, CDCl ): d ˆ 14.11, 22.58,
3
found: C 65.94, H 5.62.
9.27, 30.21, 31.15, 31.77 (alkyl-C), 56.34 (C-4b,12a), 81.12 (C-5,12 or
exo-15: Prepared from exo-14 (2.12 g, 3.4 mmol) by a procedure similar to
C-7,10), 81.95 (C-7,10 or C-5,12), 120.41, 125.82, 126.07, 126.95, 129.02,
36.42, 137.39, 142.17, 144.42, 143.05, 147.63 ppm; MS (80 eV): m/z (%): 528
that described above to give 1.52 g of a yellow solid in 72% yield (R
f
ˆ 0.72,
1
m.p. 152 ±154 8C); 1H NMR: d ˆ 0.97 (t, J ˆ 6.9 Hz, 6H; CH
3
), 1.38 (m,
3
‡
‡
‡
(
2.92) [M] , 352 (100) [M À C14
H ] , 176 (17.79) [C14H ] ; HRMS calcd for
8
8
1
2H; alkyl-CH
2
), 1.62 (m, 4H; b-CH
2
), 2.95 (m, 4H; a-CH
2
), 3.97 (s, 2H;
38 40 2
C H O
: 528.302831; found: 528.30646
3
H-4b,10a), 5.43 (s, 2H; H-5,10), 7.14 (s, 2H; H-1,2), 7.52 (d, 2H, J ˆ
3
13
4b,5,7,7a,8,11,11a,12,14,14a-Decahydro-8,11-carbonyl-5,14:7,12-diepoxy-
,13-dihexyl-8,9,10,11-tetraphenylacenaphtheno[1',8':2,3,4]cyclopenta[1,2-
6
(
4
1
(
.86 Hz, H-3,12), 7.79 ppm (d, 2H, J ˆ 6.86 Hz, H-4,11);
C NMR
6
125 MHz): d ˆ 14.13, 22.61, 29.47, 29.49, 31.57, 35.13 (alkyl-C), 55.92 (C-
b]naphthacene (17): As a representative procedure for the synthesis of the
four isomers of 17, the one for endo-anti-17 is given. A suspension of endo-
anti-16 (830 mg, 1.57 mmol) and tetracyclone (604 mg, 1.57 mmol) in
ethanol (100 mL) was refluxed for 2 h under nitrogen atmosphere. The
mixture was dried in vacuum and the residue purified by column
b,10a), 82.53 (C-5,10), 120.73, 126.15, 126.35, 126.95, 129.16, 134.48,
36.64, 137.10, 143.72, 144.12 ppm (arom-C); MS (80 eV, EI, 2308C): m/z
‡
‡
‡
%): 620 (1.9) [M] , 444 (38.2) [M À C14
H
8
] , 365 (33) [M À C14
H
8
À Br] ,
‡
1
6
76 (100) [C14
H
8
] ; elemental analysis calcd (%) for C34
H
36Br
2
O (620.46): C
5.82, H 5.85; found: C 65.66, H 5.80.
chromatography (silica gel, toluene) to give a pale yellow solid (R
f
ˆ
4
b,5,7,10,12,12a-Hexahydro-5,12,7,10-diepoxy-6,11-dihexylcyclopenta[c,d]-
naphtho[2,3-k]fluoranthene (16):
1
0
.38). Yield: 960 mg (67%); m.p. 187 ±189 8C (decomp.); H NMR
3
(
500 MHz, CDCl
3
): d ˆ 0.97 (t, 6H, J ˆ 6.8 Hz; CH
3
), 1.40 (m, 8H; alkyl-
CH ), 1.57 (m, 6H; b,g-CH ), 1.69 (m, 2H; b-CH ), 2.48 (m, 2H; a-CH ),
endo-syn-16 and endo-anti-16: A solution of endo-15 (5.44 g, 8.8 mmol) in
dry toluene (350 mL) and dry furan (30 mL) was prepared under nitrogen.
This solution was cooled to À308C and 6 mL of a 1.6m solution of BuLi
were added dropwise over a period of 1 h. The solution was stirred for an
additional 2 h at À308C. The reaction was quenched by adding water
2
2
2
2
4
2.61 (m, 2H; a-CH ), 2.82 (s, 2H; H-7a,11a), 4.75 (m,
H-4b,14a), 5.45 (s, 2H; H-7,12), 5.80 (m,
2
J ˆ 3.7 Hz, 2H;
4
J ˆ 3.7 Hz, 2H; H-5,14), 6.73 (m,
4H; phenyl-H), 6.89 (m, 4H; phenyl-H), 6.93 (s, 2H; H-1,2), 6.96 (m, 2H;
phenyl-H), 7,30 (m, 8H; phenyl-H), 7.35 (m, 4H; phenyl-H and H-4,15),
(
5 mL). After warming to room temperature, the volume of the mixture
7.51 ppm (d,
3
J ˆ 6.9 Hz, 2H, H-3 ±16); 13C NMR (125 MHz, CDCl ): d ˆ
3
was reduced to 30 mL in a vacuum. This solution was subjected to
chromatography (silica gel, toluene) to give the mixture of isomers as a
14.15, 22.76, 29.66, 30.25, 31.10, 31.46 (alkyl-C), 46.94 (C-7a,11a), 54.97 (C-
4b,14a), 64.06 (C-8,11), 79.51 (C-5,14 or C-7,12), 79.59 (C-7,12 or C-5,14),
120.47, 124.54, 125.18, 126.55, 127.16, 127.25, 127.37, 128.10, 128.90, 129.31,
129.76, 134.97, 135.64, 136.53, 136.78, 138.27, 140.79, 142.99, 143.90 (arom-C
1
yellow oil (R
f
ˆ 0.28). The syn/anti ratio was 1:1.5 by H NMR spectroscopy.
The isomers were separated by column chromatography (silica gel, hexane/
methylene chloride 2:1) to yield 930 mg of endo-anti-16 (R
yellow oil and 1.53 g endo-syn-16 (R
f
ˆ 0.23) as a
and C-1,2), 196.52 ppm (carbonyl-C); MS (FAB
[M‡H]
‡
): m/z (%): 913 (0.2)
f
ˆ 0.17) as a yellow solid (m.p. 91 ±
‡
.
9
38C) in a total yield of 53%.
endo-syn-17: Yellow crystals (R
f
(CH
2
Cl
2
) ˆ 0.43); yield: 1.42 g (54%)
1
3
endo-syn-16: H NMR (500 MHz): d ˆ 0.91 (t, J ˆ 6.9 Hz, 6H; CH
3
), 1.35
along with 664 mg (18%) of the colorless bisadduct. Single crystals of endo-
syn-17 for X-ray crystallographic analysis were obtained by recrystalliza-
(
m, 16H; alkyl-CH
2
), 2.00 (m, 2H; a-CH
2
), 2.33 (m, 2H; a-CH
2
), 4.63 (m,
4
4
J ˆ 4.2 Hz, 2H; H-4b,12a), 5.28 (s, 2H; H-7,10), 5.68 (m, J ˆ 4.6 Hz, 2H;
1
3
tion from ethanol. H NMR (500 MHz, CDCl
H; CH
CH ), 1.62 (m, 4H; b-CH
3
): d ˆ 0.88 (t, J ˆ 6.6 Hz,
3
H-5,12), 6.08 (s, 2H; H-8,9), 6.89 (s, 2H; H-1,2), 7.18 (d, J ˆ 6.9 Hz, 2H;
6
3 2
), 1.23 (s, 2H; H-7a,11a), 1.32 (m, 8H; alkyl-CH ), 1.40 (m, 4H; g-
3
13
H-4,13), 7.39 ppm (d, J ˆ 6.9 Hz, 2H; H-3,14); C NMR (125 MHz,
2
4
2
), 2.17 (m, 2H; a-CH
2
), 2.51 (m, 2H; a-CH
2
), 4.73
CDCl
4
1
3
): d ˆ 14.08, 22.56, 29.34, 30.72, 30.83, 31.68 (alkyl-C), 55.01 (C-
b,12a), 80.02 (C-5,12 or C-7,10), 80.71 (C-7,10 or C-5,12), 120.23, 124.42,
25.88, 128.44, 135.89, 136.01, 138.27, 141.29, 142.59, 146.55 ppm; MS
4
(
2
m, J ˆ 3.9 Hz, 2H; H-4b,14a), 5.28 (s, 2H; H-7,12), 5.80 (m, J ˆ 3.9 Hz,
H, H-5,14), 6.19 (s, 2H; H-1,2), 6.77 (m, 4H; phenyl-H), 6.91 (m, 4H), 6.97
(
m, 2H), 7.22 (m, 6H), 7.27 (m, 2H), 7.36 (m, 2H), 7.43 ppm (m, 4H;
‡
‡
(
80 eV, EI, 1908C): m/z (%): 528 (1.84) [M] , 352 (100) [M À C14
H
8
] , 176
13
phenyl-H and H-3,4,15,16); C NMR (125 MHz, CDCl
2
3
): d ˆ 14.10, 22.48,
‡
(
14.81) [C14
H
8
] ; HRMS calcd for C38
H
40
O
2
: 528.302831; found: 528.30395.
8.97, 30.89, 31.53, 31.76 (alkyl-C), 46.00 (C-7a,11a), 55.12 (C-4b,14a), 63.75
endo-anti-16: 1H NMR (500 MHz, CDCl3): d ˆ 0.90(t, J ˆ 6.8 Hz, 6H;
3
(C-8,11), 79.62 (C-5,14 or C-7,12), 80.17 (C-7,12 or C-5,14), 120.47, 124.15,
124.37, 125.92, 126.50, 127.04, 127.36, 127.95, 129.18, 129.34, 129.70, 135.00,
135.44, 135.52, 136.34, 138.26, 140.70, 142.66, 144.75 (arom-C and C-1,2),
3 2 2 2
CH ), 1.28 (m, 14H; alkyl-CH ), 1.44 (m, 2H; b-CH ), 2.26 (m, 2H; a-CH ),
2
.40 (m, 2H; a-CH
2
), 4.67 (m, 4J ˆ 4.3 Hz, 2H; H-4b,12a), 5.33 (s, 2H;
4
‡
‡
H-7,10), 5.68 (m, J ˆ 4.3 Hz, 2H; H-5,12), 6.79 (s, 2H; H-8,9), 6.92 (s, 2H;
195.31 ppm (carbonyl-C); MS (FAB ): m/z (%): 913 (2) [M‡H] ;
3
3
H-1,2), 7.25 (d, J ˆ 6.8 Hz, 2H; H-4,13), 7.48 ppm (d, 2H, J ˆ 6.8 Hz;
H-3,14); 13C NMR (125 MHz, CDCl
): d ˆ 14.09, 22.56, 29.33, 30.27, 31.06,
1.69 (alkyl-C), 54.94 (C-4b,12a), 79.57 (C-5,12 or C-7,10), 80.49 (C-7,10 or
60 3
elemental analysis calcd (%) for C67H O (913.19): C 88.12, H 6.62;
found: C 87.82, H 6.68.
3
3
1,4,4a,6b,7,9,9a,10,13,13a,14,16,16a,18b-Tetradecahydro-1,4:10,13-dicar-
C-5,12), 120.37, 125.04, 125.77, 127.05, 128.05, 136.39, 136.71, 138.74, 142.21,
43.02, 146.06 ppm; MS (80 eV, EI, 2158C): m/z (%): 528 (4.67) [M] , 352
bonyl-7,16:9,14-diepoxy-8,15-dihexyl-1,2,3,4,10,11,12,13-octaphenyl-fluo-
‡
1
(
5
1
rantheno[3',4':2,3,4]cyclopenta[1,2-b]naphthacene (bisadduct): H NMR:
‡
‡
100) [M À C14
H
8
] , 176 (14) [C14
H
8
] ; HRMS calcd for
C
38
H
40
O
2
:
3
d ˆ 0.90 (t, J 6.4 Hz, 6H; CH
3
), 1.36 (m, 12H; CH
2
), 1.57 (s, 2H; H-9a,13a),
), 3.79 (m,
28.302831; found: 528.30420.
1
.64 (m, 4H; b-CH
2
), 2.12 (m, 2H; a-CH
2
), 2.48 (m, 2H; a-CH
2
exo-syn-16 and exo-anti-16: The procedure was similar to that used for
endo-syn-16 and endo-anti-16 starting from exo-15 (1.73 g, 3.3 mmol). The
2H; H-6a,16a), 4.03 (s, 2H; H-4a,18b), 5.40 (s, 2H; H-9,14), 5.73 (m, 2H;
H-7,16), 5,97 (d, J ˆ 7 Hz, 2H; H-5,18 or H-6,17), 6.61 (d, J ˆ 7 Hz, 2H;
3
3
Chem. Eur. J. 2003, 9, 2745 ±2757
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2755

A. D. Schl¸ter, W. D. Neudorff et al.
FULL PAPER
H-5,18 or H-6,17), 6.78 (m, 4H; phenyl-H), 7.10 (m, 26H; phenyl-H),
vacuo. The isomers were purified and separated by column chromatog-
raphy (silica, hexane/ethyl acetate 3:1) to give 5.87 g of endo-22 (m.p. 190 ±
1
3
7
3
(
.49 ppm (m, 10H; phenyl-H); C NMR: d ˆ 14.04, 22.44, 28.95, 30.81,
1.50, 31.64 (alkyl-C), 46.27 (C-9a,13a), 53.81 (C-4a,18b or C-6b,16a), 55.78
C-4a,18b or C-6b,16a), 63.48 (C-1,4 or C-10,13), 64.14 (C-1,4 or C-10,13),
1918C, R
f
ˆ 0.37 ) and 1.42 g of exo-22 (m.p. 166 ±167 8C, R ˆ 0.43) in an
f
overall yield of 87% as colorless solids. Single crystals for X-ray crystallo-
graphic analysis were grown from ethanol.
7
1
1
1
9.91 (C-7,16 or C-9,14), 80.40 (C-7,16 or C-9,14), 120.43, 122.61, 124.16,
26.26, 126.33, 126.79, 126.98, 127.10, 127.56, 128.25, 129.58, 129.76, 129.93,
34.66, 134.74, 134.83, 135.37, 136.02, 138.20, 138.51, 138.93, 141.39, 144.23,
1
3
endo-22: H NMR: d ˆ 4.64 (m, J ˆ 4.0 Hz, 2H; H-6b,12a), 5.71 (m, 2H,
4
J ˆ 4.0 Hz, 2H; H-7,12), 6.59 (AA'BB', 2H; H-8,11 or H-9,10), 6.68
‡
45.08, 197.06 (carbonyl-C), 201.94 ppm (carbonyl-C); MS (FAB) : m/z
3
(
AA'BB', 2H; H-8,11 or H-9,10), 7.19 (d, J ˆ 6.6 Hz, 4H; H-1,6), 7.29 (dd,
‡
(
%): 1298 (1.5) [M‡H] .
3
3
3
J ˆ 6.6 Hz, J ˆ 8.0 Hz, 2H; H-2,5), 7.37 ppm (d, J ˆ 8.0 Hz, 2H; H-3,4);
13
1
exo-anti-17: Pale yellow solid; yield: 376 mg (67%); H NMR (500 MHz,
C NMR: d ˆ 52.05 (C-6b,12a), 81.73 (C-7,12), 119.29, 120.35, 123.09,
3
CDCl
g-CH
3
): d ˆ 0.93 (t, 6H, J ˆ 7.1 Hz ; CH
3
), 1.44 (m, 8H; CH
2
), 1.63 (m, 4H;
),
125.52, 127.24, 131.21, 142.18, 142.27 ppm (arom-C); MS (80 eV, EI, 808C):
‡
‡
‡
), 1.83 (m, 4H; b-CH
2
), 2.91 (m, 2H; a-CH ), 2.96 (m, 2H; a-CH
2
2
m/z (%): 270 (76) [M] , 252 (26) [M À H
2
O] , 152 (5) [C12
H ] , 118 (100)
8
2
‡
3
2
.07 (s, 2H; H-7a,11a), 3.98 (s, 2H; H-4b,14a), 5.50 (s, 2H; H-7,12), 5.90 (s,
H; H-5,14), 6.91 (m, 4H; phenyl-H), 6.98 (m, 6H; phenyl-H), 7.15 (s, 2H;
[M À C12
H
8
] ; elemental analysis calcd (%) for C20
H14O (270.32): C 88.86,
H 5.22; found: C 88.80, H 5.11.
H-1,2), 7.33 (m, 2H; phenyl-H), 7.40 (m, 4H; phenyl-H), 7.48 (m, 4H;
exo-22: 1H NMR: d ˆ 3.93 (s, 2H; H-6b,12a), 5.48 (s, 2H; H-7,12), 7.29
3
3
phenyl-H), 7.56 (d, J ˆ 6.7 Hz, 2H; H-4,15), 7.80 ppm (d, J ˆ 6.7 Hz, 2H;
(
AA'BB', 2H; H-8,11 or H-9,10), 7.46 (AA'BB'; 2H, H-8,11 or H-9,10), 7.56
H-3,16); 13C NMR (125 MHz): d ˆ 14.03, 22.58, 29.64, 30.83, 31.56, 31.61
3
13
(
m, 4H; H-1,6 and H-2,5), 7.74 ppm (d, J ˆ 7.8 Hz, 2H; H-3,4); C NMR:
(
alkyl-C), 46.57 (C-7a,11a), 55.97 (C-4b,14a), 64.11 (C-8,11), 81.65 (C-5,14
d ˆ 53.28 (C-6b,12a), 84.30 (C-7,12), 119.49, 119.53, 123.54, 126.94, 127.96,
or C-7,12), 82.58 (C-7,12 or C-5,14), 120.56, 124.57, 125.99, 126.59, 126.86,
1
(
31.37, 143.95, 145.87 ppm (arom-C); MS (80 eV, EI, 1308C): m/z (%): 270
1
1
27.29, 127.39, 128.17, 128.98, 129.27, 129.78, 135.01, 135.38, 136.43, 137.12,
38.43, 143.73, 144.06, 144.80 (arom-C and C-1,2), 196.39 ppm (carbonyl-
‡
‡
‡
‡
8
18) [M] , 252 (3) [M À H
2
O] , 152 (12) [C12
H
8
] , 118 (100) [M À C12
H ] ;
elemental analysis calcd (%) for C20
C 88.83, H 5.29.
H14O (270.32): C 88.86, H 5.22; found:
‡
‡
C); MS (FAB ): m/z (%): 913 (1) [M‡H] .
exo-syn-17: Pale yellow solid; yield: 501 mg (75%); H NMR (500 MHz,
CDCl
1
rel-(1R,4S,4aS,9S,9aS,17bR,18R,22bR)-1,4:9,18-Diepoxy-8,19,23,24-tetra-
hexyl-1,4,4a,9,9a,17b,18,22b-octahydro-2,14:3,13-dimethenodiinde-
3
3
): d ˆ 0.99 (t, 6H, J ˆ 7.2 Hz ; CH
3
), 1.48 (m, 8H; alkyl-CH
), 2.91 (m, 2H; a-CH ), 3.01 (m, 2H; a-
), 3.20 (s, 2H; H-7a,11a), 4.06 (s, 2H; H-4b,14a), 5.46 (s, 2H; H-7,12),
2
), 1.66
(
m, 4H; g-CH
2
), 1.83 (m, 4H; b-CH
2
2
no[1,2,3-c,d:1',2',3'-c',d']benzo[2,3-j:5,6-j']difluoranthene (23): A solution
of 19 (60 mg, 0.06 mmol) in toluene (5 mL) was prepared under nitrogen
atmosphere and heated to reflux. p-TosH monohydrate (45 mg) was added
and the mixture was refluxed for 16 h. After cooling to room temperature,
the solution was washed three times with water and the organic layer was
dried in vacuo. The residue was purified by column chromatography (silica
gel, CH Cl )to give 27 mg of a pale yellow green solid (m.p. >3108C) in
CH
2
5
.89 (s, 2H; H-5,14), 6.92 (m, 4H; phenyl-H), 7.01 (m, 6H; phenyl-H), 7.14
(
s, 2H; H-1,2), 7.33 (m, 2H; phenyl-H), 7.40 (m, 4H; phenyl-H), 7.51 (m,
3
3
4
2
3
5
1
1
H; phenyl-H), 7.54 (d, J ˆ 6.8 Hz, 2H; H-4,15), 7.80 ppm (d, J ˆ 6.8 Hz,
H; H-3,16); 13C NMR (125 MHz): d ˆ 14.14, 22.68, 28.88, 31.21, 31.70,
1.77 (alkyl-C), 47.05 (C-7a,11a), 56.35 (C-4b,14a), 64.30 (C-8,11), 80.21 (C-
,14 or C-7,12), 81.98 (C-7,12 or C-5,14), 120.59, 124.62, 126.14, 126.73,
27.05, 127.38, 127.53, 128.24, 129.11, 129.46, 129.91, 135.14, 135.72, 136.61,
37.26, 138.54, 144.01, 144.40, 144.99 (arom-C and C-1,2), 196.79 ppm
2
2
47% yield. Single crystals for X-ray crystallographic analysis were grown
by the diffusion of hexane into a solution of 23 in chloroform; H NMR:
1
3
d ˆ 0.87 (t,
J ˆ 6.3 Hz, 12H; CH ), 1.1 ±1.5 (m, 24H; alkyl-CH ), 1.58 (m,
3
2
‡
‡
4
(
carbonyl-C); MS (FAB ): m/z (%): 913 (3) [M‡H] ; elemental analysis
8H; b-CH ), 2.49 (m, 4H; a-CH ), 2.74 (m, 4H; a-CH ), 4.77 (m, J ˆ
2
2
2
calcd (%) for C67 (913.19): C 88.12, H 6.62; found: C 88.36, H 6.33.
H
60
O
3
3.9 Hz, 4H; H-4a,9a,17b,22b), 5.92 (m,
4
J ˆ 3.9 Hz, 4H; H-1,4,9,18), 7.26 (d,
3
rel-(1R,4S,4aS,6bS,7R,9S,9aS,11bS,12R,15S,15aR,17bR,18R,20-
S,20aR,22bR)-1,4:7,20:9,18:12,15-Tetraepoxy-8,19,23,24-tetrahexyl-
J ˆ 6.9 Hz, 4H; H-5,10,17,21 or H-6,11,16,22), 7.47 (s, 4H; H-7,12,15,20),
3 13
7.49 ppm (d, J ˆ 6.9 Hz, 4H, H-5,10,17,21 or H-6,11,16,22); C NMR
(125 MHz): d ˆ 14.11, 22.61, 29.58, 30.93, 31.52, 31.76 (alkyl-C), 55.91 (C-
4a,9a,17a,22b), 81.66 (C-1,4,9,18), 116.25, 119.39, 121.43, 128.87, 131.00,
133.49, 134.47, 136.28, 137.24, 137.44, 142.31 ppm (arom-C); MS (80 eV, EI,
1
,4,4a,6b,7,9,9a,11b,12,15,15a,17b,18,20,20a,22b-hexadecahydro-2,14:3,13-
dimethenodiindeno[1,2,3-c,d:1',2',3'-c',d']benzo[2,3-j:5,6-j']difluoranthene
19): Typical procedure: A solution of exo-syn-19 (230 mg, 0.25 mmol) in
(
‡
2‡
toluene (10 mL) was prepared under nitrogen and refluxed for 2 d. After
cooling to room temperature and removal of solvent in vacuo, the residue
2608C): m/z (%): 968 (38) [M] , 484 (100) [M À C36
H36O or M] , 413 (19)
‡
[M À C36
H
36O À C
5
H
11] .
was purified by column chromatography (CH
2
Cl
2
) to afford a colorless
solid (m.p. 286 ±287 8C, decomp) in 43% yield (R
were grown by diffusion of ethanol into a chloroform solution. H NMR:
f
ˆ 0.25). Single crystals
1
3
Acknowledgement
d ˆ 0.88 (t, J ˆ 6.4 Hz, 12H; CH
3
), 1.31 (m, 24H; alkyl-CH
2
), 1.50 (s, 4H;
), 2.34 (m, 4H;
H-6b,11b,15a,20a), 1.56 (m, 8H; b-CH
2
), 2.00 (m, 4H; a-CH
2
4
We thank the Fonds der Chemischen Industrie for financial support of this
work.
a-CH
2
), 4.71 (m, J ˆ 3.7 Hz, 4H; H-4a,9a,17b,22b), 5.13 (s, 4H;
4
3
H-7,12,15,20), 5.73 (m, J ˆ 3.7 Hz, 4H; H-1,4,9,18), 7.03 (d, J ˆ 6.9 Hz,
3
4
H; H-5,10,17,21 or H-6,11,16,22), 7.19 ppm (d, J ˆ 6.9 Hz, 4H;
H-5,10,17,21 or H-6,11,16,22); 1H NMR ([D
]benzene): d ˆ 0.89 (t, J ˆ
3
6
6
8
4
.6 Hz, 12H; CH
3
), 1.22 (m, 16H; alkyl-CH
2
), 1.35 (m, 8H; g-CH
2
), 1.50 (m,
), 2.28 (m,
[
[
1] a) A. D. Schl¸ter, Adv. Mater. 1991, 3, 282 ±291; b) ™Synthesis of
Polymers∫: A. D. Schl¸ter, Material Science and Technology Series,
Wiley-VCH Weinheim, 1999, pp. 459 ±484
H; a-CH
H; a-CH
2
), 1.65 (s, 4H; H-6b,11b,15a,20a), 2.03 (m, 4H; a-CH
2
4
2
), 4.54 (m, J ˆ 3.7 Hz, 4H; H-4a,9a,17b,22b), 5.34 (s, 4H;
4
3
H-7,12,15,20), 5.68 (m, J ˆ 3.7 Hz, 4H; H-1,4,9,18), 6.77 (d, J ˆ 6.9 Hz,
2] a) A. Godt, V. Enkelmann, A. D. Schl¸ter, Angew. Chem. 1989, 101,
3
4
H; H-5,10,17,21 or H-6,11,16,22), 6.87 (d, J ˆ 6.9 Hz, 4H; H-5,10,17,21 or
1
704 ±1706; Angew. Chem. Int. Ed. Engl. 1989, 28, 1680 ±1682; b) O.
1
3
H-6,11,16,22); C NMR (125 MHz): d ˆ 14.10, 22.54, 29.19, 31.15, 31.70,
1.75 (alkyl-C), 54.69 (C-4a,9a,17a,22b), 57.23 (C-6b,11b,15a,17a), 80.60 (C-
,4,9,18), 81.56 (C-7,12,15,20), 120.21, 121.16, 125.29, 138.62, 139.39, 139.60,
Kintzel, A. D. Schl¸ter, Acta Polymer. 1997, 48, 212 ±214; c) O.
Kintzel, P. Luger, M. Weber, A. D. Schl¸ter, Eur. J. Org. Chem. 1998,
3
1
9
9 ±105.
1
40.67, 141.80, 144.05 ppm (arom-C); MS (80 eV, EI, 1408C): m/z (%): 1004
[
3] a) P. R. Ashton, G. R. Brown, N. S. Isaacs, D. Giuffrida, F. H. Kohnke,
J. P. Mathias, A. M. Z. Slawin, D. R. Smith, J. F. Stoddart, D. J.
Williams, J. Am. Chem. Soc. 1992, 114, 6330 ±6353; b) J. Benkhoff,
R. Boese, F.-G. Kl‰rner, A. E. Wigger, Tetrahedron Lett. 1994, 35, 73 ±
76; c) R. M. Cory, C. L. McPhail, Tetrahedron Lett. 1996, 37, 1987 ±
1990; d) R. M. Cory, C. L. McPhail, Tetrahedron Lett. 1996, 37, 1983 ±
1986; e) F. Vˆgle, Top. Curr. Chem. 1983, 115, 153 ±157.
‡
‡
‡
(
1.8) [M] , 828 (0.9) [M À C14
H
8
] , 502 (18) [M À C36
38
H O ] , 326 (100)
2
‡
[
M À C36
H
38
O
2
À C14H
8
] ; elemental analysis calcd (%) for
C
72
H
76
O
4
(
1005.33): C 86.02, H 7.62; found: C 85.78, H 7.58.
6
a,7,12,12a-Tetrahydro-7,12-epoxybenzo[k]fluoranthene (22): A solution
of acenaphthylene (4.77 g, 31 mmol) and of the isobenzofuran precursor
2
[
20]
1
(16.4 g, 31 mmol) in toluene (250 mL) was refluxed for 16 h. After
cooling to room temperature, the solvent was evaporated in vacuo. The
residue was digested in hot ethanol (250 mL) for 30 min and filtered while
hot; this was repeated once. This procedure removed most of the
tetraphenylbenzene byproduct. The combined filtrates were dried in
[4] J.-I. Aihara, J. Chem. Soc. Perkin Trans. 2 1994, 971 ±974.
[5] a) H. S. Choi, K. S. Kim, Angew. Chem. 1999, 111, 2400 ±2402; Angew.
Chem. Int. Ed. 1999, 38, 2256 ±2258; b) K. N. Houk, P. S. Lee, M.
Nendel, J. Org. Chem. 2001, 66, 5517 ±5521
2756
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 2745 ±2757

Belt Region of Fullerenes
2745±2757
[
6] a) Y. Kuwatani, paper presented at the 10th International Symposium
on Novel Aromatics, San Diego, 2001; b) G. J. Bodwell, D. O. Miller,
R. J. Vermeij, Org. Lett. 2001, 3, 2093 ±2096.
[22] a) G. H. Olah, S. C. Narang, B. G. B. Gupta, R. Mahalta, Angew.
Chem. 1979, 91, 648 ±649; Angew. Chem. Int. Ed. Engl. 1979, 18, 612 ±
614; b) G. H. Olah, J. Org. Chem. 1979, 44, 1247 ±1251.
[23] M. J. Dewar, E. G. Zoebisch, E. F. Healy, J. J. P. Stewart, J. Am. Chem.
Soc. 1985, 107, 3902 ±3909.
[
7] a) A. D. Schl¸ter, M. Lˆffler, V. Enkelmann, Nature 1994, 368, 831 ±
8
34; b) M. Lˆffler, A. D. Schl¸ter, K. Ge˚ler, W. Saenger, J.-M.
Toussaint, J.-L. Bredas, Angew. Chem. 1994, 106, 2281 ±2284; Angew.
Chem. Int. Ed. Engl. 1994, 33, 2209 ±2212; c) B. Schlicke, H. Schirmer,
A. D. Schl¸ter, Adv. Mater. 1995, 7, 544 ±546.
[24] R. Notario, J.-L. Abboud, J. Phys. Chem. A 1998, 102, 5290 ±5297 and
references therein.
[25] a) L. T¸rker, THEOCHEM 2000, 531, 333 ±337; b) L. T¸rker, J. Mol.
Struct. 1997, 407, 217 ±220.
[
[
8] M. M. Boorum, Y. V. Vasiliev, T. Drewello, L. T. Scott, Science 2001,
2
94, 828 ±834, and references therein.
9] a) T. Sasaki, K. Kanematsu, K. Iizuka, J. Org. Chem. 1976, 41, 1105 ±
112; b) S. Mondal, T. K. Bandyopadhyay, A. J. Bhattacharya, Ind. J.
Chem. 1983, 22, 225 ±229.
10] a) K. Harano, M. Yasuda, K. Kanematsu, J. Org. Chem. 1982, 47,
[26] M. Kertesz, A. Ashertehrani, Macromolecules, 1996, 29, 940 ±945.
[27] W. D. Neudorff, Dissertation, 2002, Freie Universit‰t Berlin (Germa-
ny).
1
[28] PAs for [n]cyclacenes with an even-numbered n rise to exceptionally
[
[
[
high values (see the Supporting Information). This may be a
3
1
736 ±3743; b) W. Ried, R. Neidhardt, Liebigs Ann. Chem. 1970, 739,
55 ±158; c) V. V. Plemenkov, L. A. Yanykina, J. Org. Chem. USSR
consequence of the unreliability of the AM1 method for this class of
compounds on account of their energetically similar valence-isomeric
structures (see ref. [5]).
(Engl. Transl.) 1970, 6, 2049 ±2052.
11] a) A. G. Anderson, R. G. Anderson, J. Org. Chem. 1958, 23, 517 ±520;
b) H.-J. Bestmann, R. Armsen, H. Wagner, Chem. Ber. 1969, 102,
[29] K. B. Wiberg, Angew. Chem. 1986, 98, 312 ±322; Angew. Chem. Int.
Ed. Engl. 1986, 25, 312 ±322.
[30] ™JANAF Thermochemical Tables, 3rd ed.∫: M. W. Chase, Jr., C. A.
Davies, J. R. Downey, D. J. Frurip, R. A. McDonald, A. N. Syverud, J.
Phys. Chem. Ref. Data 1985, 14, (Suppl. 1).
2
259 ±2269; c) B. M. Trost, G. M. Bright, C. Frihart, D. Britelli, J. Am.
Chem. Soc. 1971, 93, 737 ±745.
12] B. M. Trost, D. Buhner, G. M. Bright, Tetrahedron Lett. 1973, 29,
2
787 ±2790.
[31] SAINTPLUS Software Reference Manual, Version 5.054, Bruker-
AXS, Madison, WI, 1997 ± 1998.
[32] R. H. Blessing, Acta Crystallogr. Sect A 1995, 51, 33 ±38; SADABS,
Bruker AXS 1998.
[
[
13] N. Tanaka, T. Kasai, Bull. Chem. Soc. Jpn. 1981, 54, 3020 ±3023
14] R. G. Harvey, L. Nazareno, H. Cho, J. Am. Chem. Soc. 1973, 95, 2376 ±
2
378
[
[
[
15] M. R. Churchill, J. Wormald, Chem. Commun. 1968, 1597 ±1598.
16] H. Meier, B. Rose, Liebigs Ann./Recl. 1997, 4, 663 ±670.
17] H. Hart, C.-Y. Lai, G. C. Nwokogu, S. Shamouilian, Tetrahedron 1987,
[33] G. M. Sheldrick, SHELX97 Programs for Crystal Structure Analysis
(Release 97 ±2), University of Gˆttingen, Gˆttingen (Germany),
1998.
4
3, 5203 ±5224.
[34] PLATON/PLUTON: a) A. L. Spek, Acta Crystallogr. Sect A 1990, 46,
C34; b) A. L. Spek, PLATON, A Multipurpose Crystallographic Tool,
Utrecht University, Utrecht (The Netherlands), 1998.
[35] ORTEP3 for Windows: L. J. Farrugia, J. Appl. Crystallogr. 1997, 30,
565.
[
[
[
18] exo-anti-17 could not be obtained in pure form.
19] T. Vogel, Dissertation, 1990, University of Mainz (Germany).
20] R. Packe-Wirth, V. Enkelmann, J. Mol. Struct. 1998, 448, 1 ±9, and
references therein.
[
21] H. Schirmer, A. D. Schl¸ter, V. Enkelmann, Chem. Ber. 1993, 126,
[36] WINGX: L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837 ±838.
[37] Hypercube, Inc. 1115 NW 4th Street, Gainesville, Florida (USA).
2
543 ±2546.
Received: February 6, 2003 [F4824]
Chem. Eur. J. 2003, 9, 2745 ±2757
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2757