Synthesis of guanidinium-derived receptor libraries and screening for selective peptide receptors in water

Article abstract of DOI:10.1002/1521-3765(20020315)8:6<1424::AID-CHEM1424>3.0.CO;2-Z

A library of "tweezer" receptors, incorporating a guanidinium "head group" and two peptide derived side arms has been prepared on the solidphase using an orthogonally protected guanidinium scaffold 12. The library was screened with various tripeptide derivatives in an aqueous solvent system. A tweezer receptor 25 for the side chain protected tripeptide 19 was identified from the screening experiments. Receptor 25 was resynthesised and solution binding studies were carried out, which revealed that 25 binds to tripeptide 19 with K_a = 8.2 × 10⁴ ± 2.5 × 10⁴ (15% DMSO/H₂O, pH 8.75) and with appreciable selectivity over the tripeptide enantiomer 22 and the side chain deprotected tripeptide 20.

Full text of DOI:10.1002/1521-3765(20020315)8:6<1424::AID-CHEM1424>3.0.CO;2-Z

FULL PAPER
Synthesis of a Giant 222 Carbon Graphite Sheet
Christopher D. Simpson, J. Diedrich Brand, Alexander J. Berresheim, Laurence Przybilla,
Hans Joachim R‰der, and Klaus M¸llen*[a]
Abstract: In this paper we present the
synthesis and characterization of the so
far largest polycyclic aromatic hydro-
carbon (PAH), containing 222 carbon
atoms or 37 separate benzene units. First
a suitable three-dimensional oligophe-
nylene precursor molecule is built up by
a sequence of Diels Alder and cyclo-
trimerization reactions and then planar-
ized in the final step by oxidative cyclo-
dehydrogenation to the corresponding
hexagonal PAH. Structural proof is
based on isotopically resolved MAL-
DI-TOF mass spectra and electronic
characteristics are studied by UV/Vis
spectroscopy.
Keywords: carbon ¥ cyclodehydro-
genation
¥
graphite
¥
mass
spectrometry ¥ polycycles
sponding hexagonal, 222 carbon containing PAH (2). This
PAH, which up to now has only been considered in theoretical
articles,[4] is presently the biggest fully condensed aromatic
system actually having been prepared. Its diameter is
approximately 3 nm (Scheme 1).
Introduction
Extremely large polycyclic aromatic hydrocarbons (PAHs)
are for long being used by theoreticians as model compounds
for graphite.[1 Not so long ago the synthesis of such huge
5]
PAHs seemed out of reach due to decreasing solubility with
increasing size of the molecules. Hexa-peri-hexabenzocoro-
nene, described for the first time by Halleux, was for a long
time with its 42 carbon atoms the largest fully characterized
PAH.[6, 7] Applying the synthetic concept recently introduced
Results and Discussion
by us, it has been possible to produce a variety of extremely
The two synthetic approaches to the precursor molecule 1
(pathway I and II, see Scheme 2) are based on the selectivity
of the Diels Alder reaction regarding acetylenes carrying
substituents of varying steric demand. While mono-substitut-
ed phenylacetylenes react at temperatures around 1608C with
reasonable rates, diphenylacetylenes require a temperature of
or above 2008C to obtain an adequate reaction rate.
Triisopropylsilyl groups serve as protective groups for acety-
lenes, preventing the Diels Alder reaction from occurring.
This circumstance has been used by us recently in the
synthesis of polyphenylene dendrimers.[14]
13]
large PAHs with up to 132 carbon atoms.[8
This was
achieved employing a synthetic concept based on two key
steps. In the first step a fully characterizable, soluble
oligophenylene precursor was synthesized, which was then
in the second step planarized by oxidative cyclodehydroge-
nation to the corresponding PAH. For the synthesis of the
oligophenylene precursors, the cobaltoctacarbonyl-catalyzed
cyclotrimerisation of symmetrical diphenylacetylene deriva-
tives as well as the Diels Alder reaction of a phenyl- or
diphenylacetylene containing molecule and a tetraphenyl-
cyclopentadienone have proven to be useful. While the
cyclotrimerisation leads exclusively to hexagonal structures,
a range of different PAHs of varying structure and size are
accessible through the Diels Alder route. In this article we
describe two different synthetic routes to an oligophenylene
precursor (1) containing 37 separate benzene units as well as
its subsequent oxidative cyclodehydrogenation to the corre-
The first pathway (I) of obtaining 1 is the combination of
the Diels Alder reaction with the cobaltoctacarbonyl-cata-
17]
lyzed cyclotrimerisation.[15
According to Scheme 2 the
diphenylacetylene derivative 3 can be obtained by coupling
4,4'-dibromophenylacetylene (4) with trimethylsilylacety-
20]
lene[18
followed by the removal of the silyl protection
groups using potassium fluoride. Compound 3 reacts in a
Diels Alder reaction at 1908C in diphenyl ether with two
equivalents of tetraphenylcyclopentadienone (6) exclusively
at the outer acetylene units to give 7 in 66% yield. Despite the
large substituents on the triple bond, the cobaltoctacarbonyl-
catalyzed cyclotrimerisation of 7 can be carried out with a
yield of 71% obtaining the precursor oligophenylene 1.
[a] Prof. Dr. K. M¸llen, Dipl.-Chem. C. D. Simpson, Dr. J. D. Brand,
Dr. A. J. Berresheim, Dr. L. Przybilla, Dr. H. J. R‰der
Max-Planck-Institute for Polymer Research
Ackermannweg 10, 55128 Mainz (Germany)
Fax : (‡49) 6131-379100
E-mail: muellen@mpip-mainz.mpg.de
1424
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1424 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 6

1424 1429
The second synthetic path-
way (II) to 1 was opened by
the use of the sterically very
demanding
triisopropylsilyl
groups as substituents on the
outer triple bonds of 5b. Thu s,
when treating 5b with the cy-
clopentadienone 8, only the
central, sterically less hindered
triple bond of 5b underwent a
Diels Alder reaction to afford
9 in 72% yield. The reaction
was carried out in diphenyl
ether at 2008C, however, in this
case it took 11 days for the
starting materials to react com-
pletely. This extended reaction
time is explained by the in-
creased steric demand of the
reaction compared with path-
way (I) where the outer triple
bonds of 3 undergo the Diels
Alder reaction. On treating 9
-108 H
2
1
Scheme 1. Oxidative cyclodehydrogenation of the precursor molecule 1 to the PAH 2.
Br
Br
4
pathway II
pathway I
a)
e)
with
fluoride, the hexaethynyl-sub-
stituted hexaphenylbenzene
tetrabutylammonium
TMS
TMS
TIPS
TIPS
O
5a
5b
TIPS
(10) was obtained with a yield
of 89% which was then reacted
further with tetraphenylcyclo-
pentadienone (6) to obtain the
targeted precursor molecule 1
in 90% yield. Even though
being a very large oligopheny-
lene, 1 is very well soluble in
solvents such as dichlorome-
thane, chloroform, tetrahydro-
furan. It was possible to com-
pletely characterize this mole-
cule including assigning the
proton and carbon signals in
the corresponding NMR spec-
tra.
TIPS
f)
b)
8
TIPS
TIPS
TIPS
TIPS
3
O
TIPS
TIPS
c)
9
6
TIPS
TIPS
The high demands towards
the efficiency and selectivity of
the now following cyclodehy-
drogenation reaction can be
understood by looking at the
computer-generated model of 1
in Figure 1. It shows the three-
dimensional structure of the
dendritic molecule which has
to be completely planarized by
g)
7
10
À
aryl aryl bond formation to
d)
h)
obtain the flat PAH 2. The
depicted conformation, resem-
bling a six-bladed propeller, can
be assumed to be one of the
thermodynamically most likely
arrangements of the molecule
1
Scheme 2. The two different synthetic routes to the precursor molecule 1: a) trimethylsilylacetylene,
[Pd(PPh3)2Cl2], CuI, PPh3, NEt3/toluene, 808C, 93%; b) KF, DMF, RT, 83%; c) tetraphenylcyclopentadienone,
Ph2O, 1908C, 11 h, 66%; d) [Co2CO8], dioxane, 1008C, 5 d, 71%; e) triisopropylsilylacetylene, [Pd(PPh3)2Cl2],
CuI, PPh3, NEt3/toluene, 808C, 74%; f) Ph2O, 2008C, 11 d, 72%; g) Bu4NF, THF, RT, 89%; h) tetraphenylcy-
clopentadienone, Ph2O, 1908C, 11 h, 89%.
Chem. Eur. J. 2002, 8, No. 6
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1425 $ 17.50+.50/0
1425

K. M¸llen et al.
FULL PAPER
developed for MALDI-TOF mass spectrometry. This new
method, as well as a detailed characterization of 2 was
recently published in a separate paper.[23] The key feature of
this method is a modified way of sample preparation for
MALDI-TOF mass spectrometry. Until quite recently only
soluble samples could be characterized by MALDI-TOF mass
spectrometry, since the traditional method of sample prepa-
ration required a dissolving step for homogenization of
sample and matrix prior to analysis.[24]
Within the new method of sample preparation the use of
solvents could be avoided completely, which opened the
application of MALDI-TOF MS also for insoluble samples in
general. Only with the novel method of sample preparation
did it become possible to obtain well resolved mass spectra of
PAH 2 without fragmentation and cluster signals which is a
severe necessity for a reliable characterization of reaction pro-
ducts by mass spectrometry, especially when no other methods
can provide supplemental information to prove the results.
Figure 2a shows the MALDI-TOF mass spectrum of PAH
2. The most intense signal at 2708 gmolÀ1 represents exactly
the mass of the target molecule, whereas a series of low
intensity signals in the higher molecular region up to
2900 gmolÀ1 are due to trace amounts of side products.
These additional signals are due to intermediates during
synthesis which result from an incomplete cyclodehydroge-
nation and a partial chlorination. However, it should be
emphasized, that the signal intensities of the side products are
highly overestimated in relation to the signal of the desired
structure 2. A quantification approach, which is described
thoroughly in ref. [22], reveals that the side products have a
much higher desorption and ionization efficiency than PAH 2
which leads to an overestimation of the impurities by a factor
of 285 in the MALDI-TOF mass spectrum. Thus the purity of
2 can be calculated from the corrected peak areas in Figure 2a
to be 99%.
Figure 2b shows an expanded region of the molecular ion of
PAH 2 with isotopic resolution. The black bars below the
measured signals indicate the theoretical isotopic distribution
of PAH 2 which is calculated for the elemental composition of
the perfect structure (C222H42).
The close agreement of theoretical and measured signal
intensities is a strong proof for the desired structure, since it
proves the complete removal of 108 hydrogen atoms during
the formation of 54 new carbon bonds.
Although a mass spectrum is certainly not a definitive
structure proof in the customary sense, the existence of the
desired structure of PAH 2 can be assumed with high certainty
in this case because there is no reasonable other structure
which can be formed by the given reaction conditions to yield
an elemental composition of C222H42. Any defect structure
which may result from rearrangement reactions or incomplete
cyclodehydrogenation give rise to a higher number than 42
hydrogens in the molecule, leading to additional or over-
lapping isotopic distributions. An incomplete cyclodehydro-
Figure 1. Computer-simulated model of the three-dimensional structure of
1 (using the molecular-mechanics force field of Wavefunction Spartan Pro
1.0.5).
in solution. Single-crystal X-ray studies, which give an insight
into the molecular structure in the solid-state, confirm the
calculated shape in general, however, of the dendritic arms
only five are oriented almost perpendicular to the central ring,
while the sixth one is almost parallel to it. Details of these
X-ray studies will be published separately.[21]
The oxidative cyclodehydrogenation method used to con-
vert the oligophenylene 1 to the PAH 2 has been described
previously.[8 11] Due to the extremely low solubility of 2 in all
solvents, the successful cyclodehydrogenation could not be
proven by conventional analysis methods. For instance
solution NMR spectroscopy was completely out of the
question, but also solid-state 13C NMR spectroscopy resulted
only in broad unidentifiable signals, giving no further
information about the detailed structure of the analyte. Also
elementary analysis, a method usually very well suitable for
solid materials, could not be used since incomplete combus-
tion under soot formation of this high carbon content type of
molecule leads to inaccurate results. Even leaving the
combustion problem aside, the accuracy of the method is
not high enough to distinguish between a molecular compo-
sition of C222H44 (1) and for instance a molecule with a C222H46
composition or a mixture of the two.
The problems mentioned above led to the burden of
structural proof resting mainly on mass spectrometry. In the
case of smaller PAHs, laser desorption time of flight (LD-
TOF) mass spectrometry was shown to be a useful method.[22]
However, in the case of 2 only unstructured extremely broad
signals were detectable with LD-TOF, rendering the clear
proof of 2 impossible. This is due to the high molecular weight
of 2 (2708 gmolÀ1) which cannot be detected by laser
desorption mass spectrometry without fragmentation and
clustering phenomena.
À
genation reaction for example, where only one C C bond
remains unclosed would result in an elementary composition
of C222H44 instead of C222H42 for PAH 2, thus resulting in an
isotopic pattern similar to that in Figure 2b but shifted by two
mass units to higher molecular weight. Thus, the signal
A reliable and detailed characterization of 2 was not
possible until a new method of sample preparation was
1426
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1426 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 6

Giant Carbon Sheet
1424 1429
The conditions applied for cyclodehydrogenation are listed
in Table 1. Best results were obtained using a combination of
copper(ii) triflate/aluminium(iii) chloride in carbon disulfide
as cyclodehydrogenation reagents. Experiment 1, with just
one hour of reaction time, resulted in a broad distribution of
products with various degrees of dehydrogenation. Increasing
the reaction time to one day and the temperature to 308C led
Table 1. Reaction conditions employed in the conversion from 1 to 2.
Experiment no.
Reaction conditions
1
2
3
4
5
6
AlCl3/Cu(SO3CF3)2/CS2/RT/1 h
AlCl3/Cu(SO3CF3)2/CS2/308C/24 h
AlCl3/Cu(SO3CF3)2/CS2/308C/10 d
AlCl3/Cu(SO3CF3)2/CS2/RT/21 d
AlCl3/CuCl2/CS2/RT/10 d
FeCl3/nitromethane/CH2Cl2/RT/2 d
to an almost complete formation of 2 (Experiment 2), which
turned out to be the best conditions.
Other experiments using copper(ii) chloride/aluminium(iii)
chloride as oxidizing agents only lead to the increased
formation of chlorinated and partially dehydrogenated side
products next to our desired compound. Iron(iii) chloride on
the other hand only yielded partially cyclized products.
Special attention was also directed to a efficient removal of
metal ions during the workup procedure. The precipitate was
washed repeatedly and thoroughly with ammonium hydrox-
ide to remove copper ions and with hydrochloric acid to
remove aluminium ions. Analysis by transmission electron
microscopy (TEM) using energy dispersive X-ray spectro-
scopy (EDX) and electron energy loss spectroscopy (EELS)
methods showed no significant amount of inorganic residues.[25]
Previous work in our group has shown, that UV/Vis spectra
of insoluble molecules, which are not accessible to solution
spectroscopy, can be recorded as thin films smeared onto a
quartz substrate.[26] The UV/Vis spectrum of 2, depicted in
Figure 3, was measured by the same method. It shows
Figure 2. a) MALDI-TOF mass spectra obtained by applying the dry
sample preparation with tetracyanoquinodimethane (TCNQ) as matrix to
a) sample containing a significant amount of target molecule 2 (cyclo-
dehydrogenation reaction with copper(ii) triflate/aluminum(iii) chloride);
b) sample containing only side products (cyclodehydrogenation reaction
with iron(iii) chloride); c) mixture of the two preceding samples in a mass
ratio 1:1; d) isotopically resolved MALDI-TOF spectrum of 2. The bars
underneath the signal represent the calculated spectrum for C222H42
(enlargement of spectrum 2).
intensities within the isotopic distribution of the perfect
structure in Figure 2b would increase starting at the isotopic
peak at 2708.5 gmolÀ1 towards higher molecular weight and
therefore destroying the agreement of measured and simu-
lated signal intensities of the isotopic pattern of the pure
molecular ion of structure 2.
The fact that the progress and the efficiency of the
cyclodehydrogenation reaction can now be monitored by
MALDI-TOF mass spectrometry allowed us to optimize the
experimental conditions for the cyclodehydrogenation reaction.
Figure 3. UV/Vis spectrum of 2, measured of a thin film on a quartz
substrate.
Chem. Eur. J. 2002, 8, No. 6
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1427 $ 17.50+.50/0
1427

K. M¸llen et al.
FULL PAPER
48H); 13C NMR (75 MHz, CDCl3): d ˆ 157.7, 142.2, 142.0, 141.30, 140.27,
absorption over the complete visible range of the electronic
spectrum, with a maximum of the band at approximately
765 nm. The broad and unstructured pattern of the band can
be explained by the large number of electronic transitions
taking place in this giant PAH, which leads to the continuum-
like appearance in the visible range.
In summary, this successful synthesis of the so far largest
polycyclic aromatic hydrocarbon represents an important step
towards improved graphite model structures. It shows on one
side the efficiency but on the other side also the limits of the
synthetic concept employed, of planarizing large soluble
oligophenylene precursors to polycyclic aromatic hydrocar-
bons. The removal of 108 hydrogens from one molecule
represented a challenge in its own right but being able to
analyze and characterize the resulting product precisely by
MALDI-TOF mass spectrometry was also a prerequisite,
showing how strongly synthetic chemistry depends on the
advances of analytical methods.
140.1, 131.9, 131.8, 131.6, 131.2, 130.3, 130.1, 128.0, 127.5, 127.3, 127.0, 126.7,
‡
126.2, 125.8, 123.6, 119.3, 90.1; FD-MS: m/z (%): 938.1 (100) [M] ; elemental
analysis calcd (%) for C74H50: C 94.63, H 5.37; found: C 92.01, H 5.22.
1,2,3,4,5,6-Hexakis(4',5',6'-triphenyl-1,1':2',1''terphenyl)benzene (1) by cy-
clotrimerization: Di(4',5',6'-triphenyl-1,1':2',1''terphenyl)acetylene (7;
500 mg, 0.533 mmol) was dissolved in dioxane (100 mL) and degassed.
Then dicobaltoctacarbonyl (118 mg, 0.35 mmol) was added under argon
and the reaction mixture was refluxed for 15 h. After evaporating the
solvent, the crude product was purified by column chromatography on
silica gel with petroleum ether/ethylacetate (8:2) to afford 1 (355 mg, 71%)
1
as colorless solid. M.p. >3008C; H NMR (500 MHz, CDCl3): d ˆ 7.44 (s,
6H), 7.14 7.11 (m, 30H), 6.90 6.60 (m, 90H), 6.62 (d, 3J ˆ 8.14 Hz, 12H),
6.36 (d, 3J ˆ 8.14, 12H); 13C NMR (125 MHz, CDCl3): d ˆ 141.78, 141.67,
140.63, 140.54, 140.45, 140.08, 140.01, 139.92, 139.12, 139.00, 138.44, 138.13,
131.65, 131.56, 131.53, 131.17, 129.96, 128.33, 127.51, 126.85, 126.74, 126.53,
‡
126.18, 125.51, 125.26; FD-MS: m/z (%): 2815.4 (100) [M] ; elemental
analysis calcd (%) for C222H150: C 94.63, H 5.37; found: C 94.05, H 5.83.
4,4'-Di-(triisopropylsilylethinyl)tolane (5b): 4,4'-Dibromotolane (4; 8 g,
23.8 mmol), copper(i) iodide(0.91 g, 4.8 mol), triphenylphosphine (1.25 g,
4.77 mmol) and bis(triphenylphosphine)palladium(ii) dichloride (1.7 g,
2.3 mmol) were dissolved under inert atmosphere in triethylamine
(140 mL) and toluene (70 mL). Triisopropylsilylacetylene (10.42 g,
57.13 mmol) were added and the reaction mixture was stirred for 5 h at
808C. After cooling, the solution was poured into CH2Cl2 (300 mL) and 6m
HCl (200 mL). The organic phase was separated, washed with saturated
ammonium chloride solution and water, and dried over magnesium sulfate.
After evaporating the solvent, the crude product was purified by column
chromatography on silica gel with petroleum ether to afford 5b (9.53 g,
74%) as colorless solid. M.p. 1088C; 1H NMR (500 MHz, CD2Cl2): d ˆ 7.51
(m, 8H), 1.20 (m, 42H); 13C NMR (75 MHz, CD2Cl2): d ˆ 135.1, 134.5,
Experimental Section
General methods: 1H NMR and 13C NMR were recorded in CDCl3 and
CD2Cl2 on a Bruker DPX 250, Bruker 300 AMX and Bruker 500 DRX with
use of the solvent proton or carbon signal as internal standard. FD mass
spectra were obtained on a VG Instruments ZAB 2-SE-FPD. MALDI-
TOF mass spectra were measured on a Bruker Reflex II-TOF spectrometer
‡
126.7, 125.9, 109.6, 96.3, 94.1, 21.8, 14.4; FD-MS: m/z (%): 538.2 (100) [M] ;
using
a 337 nm nitrogen laser and 7,7,8,8-tetracyanoquinodimethane
elemental analysis calcd (%) for C36H50Si2: C 80.23, H 9.35; found: C 80.01,
H 9.50.
(TCNQ) as matrix. UV/Vis spectra were recorded on a Perkin Elmer
Lambda 9 machine. Elemental analysis was carried out on a Foss Heraeus
Vario EL. Chemicals were obtained from Fluka, Aldrich and Strem and
used as received.
1,2,3,4,5,6-Hexakis[4-(triisopropylsilylethinyl)phenyl]benzene (9): 4,4'-
Di(triisopropylsilylethinyl)tolane (5b; 565 mg, 1.05 mmol) and 2,3,4,5-
tetrakis-[4-(triisopropylsilylethinyl)phenyl]-cyclopentadienone (8; 1.26 g,
1.15 mmol) were dissolved in diphenyl ether (7 mL) and heated under an
argon atmosphere for 11 d at 2008C. After cooling CH2Cl2 (10 mL) was
added, the solution was slowly poured into methanol (100 mL) and the
precipitate was filtered. The crude product was again dissolved in CH2Cl2
(10 mL) and precipitated in methanol (250 mL). After drying in vacuo, 9
(1.23 g, 72%) was afforded as pale yellow solid. M.p. >3008C; 1H NMR
(200 MHz, CDCl3): d ˆ 7.05 (d, J ˆ 8.0 Hz, 12H), 6.73 (d, J ˆ 8.0 Hz, 12H),
1.10 (m, 126H); 13C NMR (50 MHz, CDCl3): d ˆ 140.5, 140.4, 131.4, 121.2,
Synthesis
4,4'-Di(trimethylsilylethinyl)tolane (5a): 4,4'-Dibromotolane (4; 10 g,
29 mmol), copper(i) iodide (1.13 g, 5.9 mol), triphenylphosphine (0.77 g,
2.93 mmol) and bis(triphenylphosphine)palladium(ii) dichloride (0.7 g)
were dissolved under inert atmosphere in piperidine (200 mL). Trimethyl-
silylacetylene (8.76 g, 89.3 mmol) was added and the reaction mixture was
stirred for 5 h at 808C. After cooling, the solution was poured into CH2Cl2
(300 mL) and 6m HCl (200 mL). The organic phase was separated, washed
with saturated ammonium chloride solution and water, and dried over
magnesium sulfate. After evaporating the solvent, the crude product was
purified by column chromatography on silica gel with petroleum ether/
ethylacetate (9:1) to afford 5a (5.9 g, 93%) as colorless crystals. M.p.
‡
107.8, 90.8, 19.1, 11.8; FD-MS: m/z (%): 1617.3 (100) [M] ; elemental
analysis calcd (%) for C108H150Si6: C 80.23, H 9.35; found: C 80.01, H 9.42.
1,2,3,4,5,6-Hexakis[4-ethinylphenyl]benzene (10): 1,2,3,4,5,6-Hexakis[4-
(triisopropylsilylethinyl)phenyl] benzene (9; 400 mg, 0.25 mmol) was
dissolved in THF (30 mL). To this solution tetrabutylammonium fluoride
trihydrate (700 mg, 2.23 mmol) in THF (10 mL) was added. After 2 h,
CH2Cl2 (500 mL) was added, the organic phase washed with water (2 Â
200 mL) and dried over MgSo4. After evaporating the solvent, the crude
product was purified by column chromatography on silica gel with
petroleum ether/CH2Cl2 (1:1) to afford 10 (150 mg, 89%) as colorless
1
1338C; H NMR (200 MHz, CDCl3): d ˆ 7.44 (s, 4H), 7.43 (s, 4H), 0.25 (s,
18H); 13C NMR (125 MHz, CDCl3): d ˆ 132.0, 131.5, 123.2, 123.1, 104.7,
‡
96.5, 91.0, 0.0; FD-MS: m/z (%): 369.9 (100) [M] .
4,4'-Di(ethinyl)tolane (3): 4,4'-Di(trimethylsilylethinyl)tolane (5a; 2 g,
5.4 mmol) was dissolved in dimethylformamide (400 mL). After adding
potassium fluoride (0.78 g), dissolved in water (1 mL), the reaction mixture
was stirred for 1 h at room temperature. Then CH2Cl2 (400 mL) was added,
the organic phase washed with 6m HCl and saturated ammonium chloride
solution and dried over MgSO4. After evaporating the solvent, the crude
product was purified by column chromatography on silica gel with CH2Cl2
to afford 3 (1.0 g, 83%) as colorless crystals. M.p. 1888C; 1H NMR
(250 MHz, CDCl3): d ˆ 7.47 (m, 8H), 3.17 (s, 2H); 13C NMR (75 MHz,
CDCl3): d ˆ 132.6, 132.0, 123.9, 122.7, 91.3, 83.7, 79.5; FD-MS: m/z (%):
1
solid. M.p. >3008C; H NMR (200 MHz, CD2Cl2): d ˆ 7.04 (d, J ˆ 8.0 Hz,
12H), 6.73 (d, J ˆ 8.0 Hz, 12H), 3.00 (s, 6H); 13C NMR (50 MHz, CD2Cl2);
d ˆ 141.3, 140.6, 131.9, 131.5, 120.3, 84.0, 77.9; FD-MS: m/z (%): 678.3 (100)
‡
[M] ; elemental analysis calcd (%) for C54H30: C 95.53, H 4.45; found: C
95.08, H 4.29.
1,2,3,4,5,6-Hexakis(4',5',6'-triphenyl-1,1':2',1''terphenyl)benzene (1) by
Diels Alder: 1,2,3,4,5,6-Hexakis[4-ethinylphenyl]benzene (10; 1.13 g,
1.67 mmol) and tetraphenylcyclopentadienone (6; 5.77 g, 15.00 mmol)
were dissolved in diphenyl ether (40 mL) and heated under an argon
atmosphere for 2 h at 1908C. After cooling CH2Cl2 (30 mL) was added, the
solution was slowly poured into methanol (500 mL) and the precipitate was
filtered. After soxhlet extraction over night with pentane, 1 (4.2 g, 90%)
was afforded as white solid (for spectroscopical data see above).
‡
226.1 (100) [M] ; elemental analysis calcd (%) for C18H10: C 95.55, H 4.45;
found: C 93.43, H 3.75.
Di-(4',5',6'-triphenyl-1,1':2',1''terphenyl)acetylene (7): 4,4'-Di(ethinyl)to-
lane (3; 520 mg, 2.48 mmol) and tetraphenylcyclopentadienone (6; 1.9 g,
4.95 mmol) were dissolved in diphenyl ether (10 mL) and heated under an
argon atmosphere for 11 h at 1708C. After cooling heptane (100 mL) was
added, the precipitate filtered and repeatedly washed with heptane. After
drying in vacuo, 7 (1.54 g, 66%) was afforded as pale yellow solid. M.p.
>3008C; 1H NMR (300 MHz, CDCl3): d ˆ 7.40 (s, 2H), 7.22 6.60 (m,
C222-PAH (2): Copper(ii) triflate (416 mg, 1.16 mmol) was dried com-
pletely under vacuum and heating. After cooling aluminium(iii) chloride
1428
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1428 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 6

Giant Carbon Sheet
1424 1429
(153 mg, 1.15 mmol) and dry CS2 (50 mL) were added under an argon
atmosphere. The mixture was suspended by intense stirring for 15 min,
warmed to 308C and then 1,2,3,4,5,6-hexakis(4',5,'6'-triphenyl-1,1':2',1''-
terphenyl)-benzene (1; 10 mg, 3.55 Â 10À3 mmol) dissolved in CS2 (5 mL)
was injected through a septum. After being stirred for 24 h, the reaction
was quenched by adding methanol (50 mL). The residue was collected by
filtration, washed successively with ammonium hydroxide solution, HCl,
water, ethanol, CS2, and CH2Cl2, and subsequently dried in vacuo to afford
2 (6 mg, 62%) as black powder. M.p. >3008C; MS (MALDI-TOF): m/z:
calcd for C222H42: 2706.3; found: 2706.5 (see main text for detailed
discussion).
[10] V. S. Iyer, K. Yoshimura, V. Enkelmann, R. Epsch, J. P. Rabe, K.
M¸llen, Angew. Chem. 1998, 110, 2843; Angew. Chem. Int. Ed. 1998,
37, 2696.
[11] F. Morgenroth, K. M¸llen, Tetrahedron 1997, 53, 15349.
[12] M. M¸ller, H. Mauermann-D¸ll, M. Wagner, V. Enkelmann, K.
M¸llen, Angew. Chem. 1995, 107, 1648; Angew. Chem. Int. Ed. 1995,
34, 1583.
[13] M. M¸ller, V. S. Iyer, C. K¸bel, V. Enkelmann, K. M¸llen, Angew.
Chem. 1997, 109, 1680; Angew. Chem. Int. Ed. 1997, 36, 1607.
[14] U. M. Wiesler, A. J. Berresheim, F. Morgenroth, G. Lieser, K. M¸llen,
Macromolecules 2001, 34, 187.
[15] R. Diereks, K. P. C. Vollhardt, J. Am. Chem. Soc. 1986, 108, 3150.
[16] K. P. C. Vollhardt, Acc. Chem. Res. 1977, 10, 1.
[17] K. P. C. Vollhardt, Angew. Chem. 1984, 106, 525; Angew. Chem. Int.
Ed. Engl. 1984, 23, 539.
[18] H. A. Dieck, F. R. J. Heck, Organomet. Chem. 1975, 93, 259.
[19] L. Cassar, J. Organomet. Chem. 1975, 93, 253.
[20] S. Takahashi, Y. Koroyama, K. Sonogashira, N. Hagihara, Synthesis
1980, 627.
Acknowledgement
The authors would like to thank Lileta Gherghel for measuring the UV/Vis
spectra and for the fruitful discussions. Financial support of the European
Commission (TMR-Programs SISITOMAS and DISCEL) is gratefully
acknowledged.
[21] R. Bauer, U. Wiesler, A. Berresheim, V. Enkelmann, K. M¸llen,
unpublished results.
[22] K. Yoshimura, L. Przybilla, S. Ito, J. D. Brand, M. Wehmeier, H. J.
R‰der, K. M¸llen, Macromol. Chem. Phys. 2001, 202, 215.
[23] L. Przybilla, J. D. Brand, K. Yoshimura, J. R‰der, K. M¸llen, Anal.
Chem. 2000, 72, 4591.
[24] H. J. R‰der, W. Schrepp, Acta Polymerica 1998, 49, 272.
[25] C. K¸bel, Ph. D. thesis, Johannes-Gutenberg-University, Mainz,
Germany, 1998; C. K¸bel, K. Eckhardt, V. Enkelmann, G. Wegner,
K. M¸llen, J. Mater. Chem. 2000, 10, 879.
[1] E. Clar, Aromatische Kohlenwasserstoffe–Polyzyklische Systeme,
Springer, Berlin, 1952.
[2] E. Clar, The Aromatic Sextet, Wiley, London, 1972.
[3] M. Zander, Polycyclische Aromaten, Teubner, Stuttgart, 1995.
[4] S. E. Stein, R. L. Brown, J. Am. Chem. Soc. 1987, 109, 3721.
[5] S. E. Stein, R. L. Brown, Pathways to Graphite: Properties of Very
Large Polybenzoid Hydrocarbons, Vol. 2, VCH, Weinheim, 1987.
[6] A. Halleux, R. H. Martin, G. S. D. King, Helv. Chim. Acta 1958, 41,
1177.
[7] A. J. Berresheim, M. M¸ller, K. M¸llen, Chem. Rev. 1999, 99, 1747.
[8] P. Herwig, M. M¸ller, K. M¸llen in NATO-ASI-Series C, Mathemat-
ical and Physical Science, Vol. 499 (Ed.: J. Michl), Kluwer, Dorn-
drecht, 1997, p. 167.
[26] F. Dˆtz, J. D. Brand, S. Ito, L. Gherghel, K. M¸llen, J. Am. Chem. Soc.
2000, 122, 7698.
Received: June 22, 2001
[9] V. S. Iyer, M. Wehmeier, J. D. Brand, M. A. Keegstra, K. M¸llen,
Angew. Chem. 1997, 109, 1675; Angew. Chem. Int. Ed. 1997, 36, 1604.
Revised: November 14, 2001 [F3361]
Chem. Eur. J. 2002, 8, No. 6
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0806-1429 $ 17.50+.50/0
1429