Pyrene as chromophore and electrophore: Encapsulation in a rigid polyphenylene shell

Article abstract of DOI:10.1002/chem.200500999

Starting from the fourfold ethynyl-substituted chromophore 1,3,6,8-tetraethynylpyrene as core, a series of polyphenylene dendrimers was prepared in high yield by combining divergent and convergent growth methods. The fluorescence quantum yields (Q_f>0.92) of the encapsulated pyrene chromophore were independent of the size of the polyphenylene shell. Fluorescence quenching studies and temperature-dependent fluorescence spectroscopy were performed to investigate the site isolation of the core. They indicate that a second-generation dendrimer layer is needed to efficiently shield the encapsulated pyrene and prevent aggregate formation. Alkalimetal reduction of the encapsulated pyrene core was carried out to afford the corresponding pyrene radical anions, for which hampered electron transfer to the core was observed with increasing dendrimer generation, which is further proof of the site isolation due to the polyphenylene shell. To improve film formation and solubility of the material, solubilizing alkyl chains were introduced on the periphery of the spherical particles. Furthermore, highly transparent films obtained by a simple drop-casting method showed blue emission mainly from the unaggregated species. The materials presented herein combine high quantum efficiency, good solubility, and improved film-forming properties, which make them possible candidates for several applications in electronic devices.

Full text of DOI:10.1002/chem.200500999

FULL PAPER
DOI: 10.1002/chem.200500999
Pyrene as Chromophore and Electrophore: Encapsulation in a Rigid
Polyphenylene Shell
Stefan Bernhardt, Marcel Kastler, Volker Enkelmann, Martin Baumgarten, and
Klaus Müllen*[a]
Abstract: Starting from the fourfold
ethynyl-substituted chromophore
1,3,6,8-tetraethynylpyrene as core,
dendrimer layer is needed to efficiently
shield the encapsulated pyrene and
prevent aggregate formation. Alkali-
metal reduction of the encapsulated
pyrene core was carried out to afford
the corresponding pyrene radical
anions, for which hampered electron
transfer to the core was observed with
increasing dendrimer generation, which
is further proof of the site isolation due
to the polyphenylene shell. To improve
film formation and solubility of the ma-
terial, solubilizing alkyl chains were in-
troduced on the periphery of the spher-
ical particles. Furthermore, highly
transparent films obtained by a simple
drop-casting method showed blue emis-
sion mainly from the unaggregated spe-
cies. The materials presented herein
combine high quantum efficiency, good
solubility, and improved film-forming
properties, which make them possible
candidates for several applications in
electronic devices.
a
series of polyphenylene dendrimers
was prepared in high yield by combin-
ing divergent and convergent growth
methods. The fluorescence quantum
yields (Qf >0.92) of the encapsulated
pyrene chromophore were independent
of the size of the polyphenylene shell.
Fluorescence quenching studies and
temperature-dependent
fluorescence
Keywords:
dendrimers
·
spectroscopy were performed to inves-
tigate the site isolation of the core.
They indicate that a second-generation
EPR spectroscopy · fluorescence ·
pyrenes · substituent effects
Introduction
and thin films containing pyrene derivatives turned out to
be promising candidates for several applications such as or-
ganic light-emitting diodes (OLEDs).[5]
The fluorescence properties of pyrene are well known and
characterized by long excited-state lifetimes[1] and distinct
solvatochromic shifts.[2] Furthermore, pyrene exhibits char-
acteristic excimer formation in concentrated solutions and
in the solid state, due to self-association of the polyaromatic
hydrocarbon moieties. However, this leads to a dramatic de-
crease in fluorescence and also to less defined, broadened
fluorescence spectra. Therefore, excimer formation of
pyrene can be used to study aggregation phenomena.[1b,3] In
addition, the sensitive solvatochromic shift of pyrene has
been used to exploit the inner structure and polarity of den-
drimers by introducing pyrene in the exterior, interior, or
throughout the dendritic structure.[4] Moreover, monolayers
Polyphenylene dendrimers have attracted great attention
due to their highly stiff and shape-persistent dendritic back-
bones[6] on the nanometer scale. Furthermore, this class of
dendrimers can be synthesized strictly monodisperse in high
yields, and this allows accurate control over the size of the
resulting spherical macromolecules. Recently, we have
shown that by attaching polyphenylene dendrons of differ-
ent generations, an encapsulated perylene chromophore can
be shielded from the surrounding medium.[7] However, a sig-
nificant decrease in fluorescence quantum yield was ob-
served with increasing dendrimer generation. Moore and
others combined luminescent chromophores with phenylace-
tylene dendrons and found energy transfer from peripheral-
ly attached donors to the emitting core.[8] Likewise, excita-
tion of polyphenylene dendrons leads to emission of encap-
sulated perylene, giving rise to an extra excitation band in
the UV region.[7a,9]
[a] S. Bernhardt, M. Kastler, Dr. V. Enkelmann, Dr. M. Baumgarten,
Prof. Dr. K. Müllen
Max-Planck-Institut für Polymerforschung
Ackermannweg 10, 55128 Mainz (Germany)
Fax: (+49)6131-379-351
Insight into the relationship between dendrimer structure
and core encapsulation can be derived from investigation of
encapsulated redox-active core moieties. For example, a de-
E-mail: muellen@mpip-mainz.mpg.de
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2006, 12, 6117 – 6128
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6117

creased reversibility of electrochemical charge transfer has
often been correlated with increasing dendrimer generation.
Porphyrin is perhaps the most investigated example of a
redox-active core to which different types of dendrimer
arms have been attached.[10] In this regard, Gorman et al.[11]
found that a rigid dendritic architecture hampered the elec-
tron-transfer rate to a iron sulfur core much more than a
flexible architecture.
Since the shielding of the encapsulated pyrene core was
expected to increase with increasing dendrimer generation,
the unsubstituted second- to fourth-generation dendrimers
(11, 20, 21) were synthesized by repetitive cycloadditions of
the branching unit 3,4-bis(4-triisopropylsilyethynylphenyl)-
2,5-diphenylcyclopentadienone[15] (8) and quantitative desi-
lylation of the protecting groups with tetrabutylammonium
fluoride (TBAF; Scheme 3).
Herein the encapsulation of a pyrene moiety in polyphe-
nylene dendrimers of different sizes is described. This is ex-
pected to separate the chromophores from each other due
to the stiff dendritic arms and therefore suppress aggrega-
tion in solution and in the solid state. Since amorphous
glassy films are a prerequisite for several electronic devices,
a large number of peripherally attached, branched alkyl
chains was introduced to improve the film-forming ability of
the dendronized pyrenes. The influence of the polypheny-
lene dendrons on the optical properties of the pyrene core
was investigated by UV/Vis absorption and fluorescence
spectroscopy. Additional insight into the relationship be-
tween dendrimer structure and core encapsulation was de-
rived from alkali-metal reduction of the pyrene dendrimers
with pyrene as electrophore.
Third-generation dendrimer 20 was synthesized by a com-
bination of convergent and divergent synthetic steps: firstly,
synthesis of second-generation dendron 19[16] and, secondly,
reaction of 19 with ethynyl-substituted first-generation den-
drimer 10. Due to the higher steric demand of dendron 19,
the yield of the Diels–Alder cycloaddition step was reduced.
To improve the film-formation ability and thus reduce the
crystallinity of the materials, alkyl chains were introduced
into second-generation derivative 11b by using the corre-
sponding alkyl-substituted tetraphenylcyclopentadienone 14.
These alkyl chains lowered significantly the isotropization
temperature of liquid-crystalline polyaromatic hydrocarbons,
and much higher solubility was observed.[17] Cyclopentadie-
none 14 was obtained by oxidation of substituted diphenyla-
cetylene 12 to corresponding benzil derivative 13 and subse-
quent Knoevenagel condensation with 1,3-diphenylpropan-
2-one.
Results
All higher generation dendrimers have good solubility in
common organic solvents, (e.g., 15 even in hexane) which
allows purification by column chromatography and full char-
acterization by standard spectroscopic techniques. The
1H NMR spectra showed well-separated and clearly assigna-
ble signals for the aromatic pyrene protons, as well as for
the ethynyl or triisopropylsilyl (TiPS) protons. For the
higher generation dendrimers, not all the aromatic signals
could be distinguished due to
Synthesis and characterization: To ensure a dense and there-
fore highly shielding polyphenylene shell around the pyrene
core, a large number of starting points for dendrimer growth
is desired. Thus, fourfold bromination of pyrene (1) in the
1,3,6,8-positions was carried out according to the litera-
ture[12] to obtain 2 (Scheme 1).
strong signal overlap. In the
case of the ethynyl- and TiPS-
substituted dendrimers, the in-
tensity ratios between aromatic
and aliphatic signals corre-
sponded to the expected values.
As an additional proof of struc-
ture, the single proton on the
pentaphenyl repeating units
(see Scheme 2) was used, since
Scheme 1. Synthesis of 1,3,6,8-tetraethynylpyrene (4). i) Bromine, nitrobenzene, 1608C, 90%; ii) 6 equiv trime-
thysilylethyne, [PdCl2
95%.
A
generation-dependent chemical
shifts were observed. The mon-
Hagihara–Sonogashira[13] cross-coupling of 2 with trime-
thysilylethyne afforded 3 in 83% yield, which was deprotect-
ed with K2CO3 in methanol to give activated pyrene core 4
in high yield. The preparation of structurally defined poly-
phenylene dendrimers with a pyrene core was realized by
repetitive coupling and activation steps based on Diels–
Alder cycloadditions[14] (Scheme 2). Despite of the low solu-
bility of 4, first-generation dendrimer 6 was obtained by
Diels–Alder cycloaddition with tetraphenylcyclopentadie-
none (5) in quantitative yield.
odispersity of the described dendrimers could easily be veri-
fied by MALDI-TOFmass spectrometry. For all dendrimers,
the calculated and experimentally determined m/z ratios
were in good agreement, even for fourth-generation den-
drimer 21 with a molecular weight of 23031 gmolꢀ1.
The growth of single crystals of suitable size and perfec-
tion is often difficult, mainly due to the conformational flex-
ibility of most dendritic backbones.[18] To gain further insight
into the spatial arrangement of the shielding polyphenylene
dendrons, first-generation dendrimer 23 bearing an addition-
al phenyl group on each dendritic branch was synthesized
6118
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6117 – 6128

Pyrenes
FULL PAPER
Scheme 2. Synthesis of first- and second-generation dendrimers 6, 11a, and 11b. Polyphenylene dendrimer 7 constructed from tetrahedral tetraphenylme-
thane as core.[15] i) 6 equiv 5, o-xylene, 1608C, 95%; ii) 6 equiv 8, o-xylene, 1608C, 71%; iii) TBAF, THF, 93%; iv)11a: 16 equiv 5, o-xylene, 1608C, 73%;
11b: 14 equiv 14, Ph2O, 2208C, 63%. v) 1 equiv I2, DMSO, 1708C, 40%; vi) 1 equiv 1,3-diphenylpropan-2-one, tert-butanol, 808C, 49%.
(Scheme 4). Due to the higher molecular symmetry of 23, a
more pronounced tendency towards crystallization was ex-
pected. Hagihara–Sonogashira cross-coupling of 2 with phe-
nylacetylene gave 1,3,6,8-tetrakis(phenylethynyl)pyrene
(22). Subsequent Diels–Alder reaction with tetraphenylcy-
clopentadienone (5) yielded first-generation dendrimer 23.
Crystals of 23 suitable for structure determination were
obtained from CH2Cl2 solution by slow evaporation at room
temperature.[19] A projection of the crystal structure is
shown in Figure 1.
The crystal contains eight well-ordered CH2Cl2 molecules
per dendrimer unit. The four central phenyl rings of the
pentaphenylbenzene substituents are oriented approximate-
ly perpendicular to the pyrene core. Thus, the dendrimer as-
sumes the shape of a cross.
The absorption in the visible region is due to the p–p*
transition of the pyrene core and showed a red-shift of up to
55 nm compared with unsubstituted pyrene (337 nm), in
accord with other phenyl-substituted pyrenes, for example,
1,3,6,8-tetraphenylpyrene (380 nm).[21] Additionally, the fine
structure of the pyrene absorption spectrum was lost due to
substitution with phenyl rings. The absorption band in the
UV region can be predominantly attributed to the polyphe-
nylene dendrons,[7a,9] as indicated by the linear increase in
the extinction coefficients e(l) with increasing number of at-
tached phenylene moieties. Going to higher generations has
no influence on the absorption maximum and extinction co-
efficients of the pyrene core.[22]
The emission spectra of 6, 11a, 20 and 21 (Figure 2), ob-
tained on excitation of the pyrene core at 390 nm displayed
a broad emission band at 425 nm, red-shifted compared to
the emission of parent pyrene (395 nm). No change in emis-
sion maximum or fluorescence intensity of the pyrene core
with changing dendrimer generation was observed. Excita-
tion of the polyphenylene dendrons at 310 nm resulted in
strong emission of the pyrene core at 425 nm, which indi-
cates efficient energy transfer from the polyphenylene den-
UV/Vis and EPR spectroscopic studies: In the absorption
spectra of the first- to fourth-generation dendrimers 6, 11a,
20, and 21 two distinct bands, one in the visible region (ca.
395 nm) and the other in the UV region (280–350 nm), were
observed (Figure 2).[20]
Chem. Eur. J. 2006, 12, 6117 – 6128
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6119

K. Müllen et al.
Scheme 3. Synthesis of third- and fourth-generation dendrimers 20 and 21. i) 12 equiv 8, o-xylene, 1508C, 76%; ii) TBAF, THF, 91%; iii) 32 equiv 8, o-
xylene, 1608C, 85%; iv) TBAF, THF, 89%; v) 65 equiv5, Ph2O, 1908C, 85%; vi) 16 equiv 19, o-xylene, 1708C, 55%.
drons to the pyrene core (for
further details, see Supporting
Information). The fluorescence
quantum yields Qf of 6, 11a, 20,
and 21 in CHCl3 were deter-
mined by using 9,10-diphenyl-
anthracene as reference chro-
mophore (Table 1).[23]
Pyrene exhibits sensitive sol-
vatochromic behavior in which
the relative intensity of emis-
sion bands is dependent on sol-
vent polarity.[2a,24] Webber and
co-workers used this effect to
investigate the diffusion of
Scheme 4. Synthetic route to first-generation dendrimer 23. i) 6 equiv phenylacetylene, [PdCl2
CuI, toluene/triethylamine, 808C, 92%; ii) 6 equiv 5, Ph2O, 2308C, 90%.
A
6120
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6117 – 6128

Pyrenes
FULL PAPER
When solutions of 10 in CHCl3 with different concentrations
were excited at 390 nm, a strong decrease in the overall flu-
orescence intensity was observed with increasing concentra-
tion (Figure 3a).
Figure 1. X-ray structure of 23 recrystallized from CH2Cl2.
Figure 3. a) Emission spectra of 10 with increasing concentration in
CHCl3. Excitation: 390 nm. b) Temperature-dependent fluorescence spec-
tra of dendronized pyrenes 10 and 11a (c=10ꢀ2 m) in tetrachloroethane,
normalized to their emission maximum. Excitation: 390 nm.
Figure 2. Absorption and emission spectra of dendronized pyrenes 6,
11a, 20, and 21 in CHCl3. Excitation: 390 nm. The emission spectra were
normalized to the same optical density at the excitation wavelength.
Table 1. Fluorescence quantum yields of dendronized pyrenes 6, 11a, 20,
and 21 in CHCl3.
6
11a
20
21
More detailed inspection of the spectra (inset in Fig-
ure 3a) showed a decrease in the emission band at 425 nm
and simultaneous increase in emission intensity at 470 nm.
This most probably indicates aggregation of the pyrene
cores in first-generation dendrimer.
Qf[a]
0.96ꢁ0.03
0.97ꢁ0.02
0.97ꢁ0.03
0.92ꢁ0.02
[a] Excitation: 350 nm; errors were estimated by using Gaussꢁs law of
propagation of error.
pyrene from block-copolymer micelles.[25] Hecht et al. deter-
mined the influence of chain length and solvent polarity on
the encapsulation of a poly(e-caprolactone)-dendronized
pyrene.[4b] They found a solvation-induced encapsulation
effect, mainly due to the structural extent and collapse of
the dendrimer backbone. This should also be tested for poly-
phenylene-dendronized pyrenes. Accordingly, TiPS-substi-
tuted dendrimers 9 and 17 as well as 1,3,6,8-tetraphenylpyr-
ene (TPP) as reference were dissolved in cyclohexane and
methanol. The TiPS groups were necessary to ensure solu-
bility of the dendrimers in both solvents. Emission spectra
showed a bathochromic shift of the pyrene emission with in-
creasing solvent polarity of Dl=9 nm (9), 6 nm (17), and
9 nm (TPP).
Since the temperature dependence of the monomer–exci-
mer equilibrium can provide information on the steric or
electronic environment of the pyrene moiety,[3,27] the tem-
perature-dependent fluorescence spectra of polyphenylene-
dendronized pyrenes 10 and 11a were recorded (excitation
at 390 nm, Figure 3b).[28] At room temperature (300 K), a
10ꢀ2 m solution of 10 in tetrachloroethane displayed a broad
emission spectrum with a maximum at 470 nm, which indi-
cates aggregation. A shoulder at 440 nm may be due to
emission from the unaggregated species. At 400 K a small
increase in the intensity of this band was noted, while strong
fluorescence at 470 nm was still detectable. In contrast,
second-generation dendrimer 11a with the same concentra-
tion at 300 K showed an emission spectrum (lmax =436 nm)
almost similar to that of the unaggregated species in dilute
solution (compare Figure 2). When the temperature was in-
In concentrated solutions (c>10ꢀ4 m), pyrene exhibits
strong characteristic excimer fluorescence at 475 nm.[26]
Chem. Eur. J. 2006, 12, 6117 – 6128
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6121

K. Müllen et al.
creased to 400 K no change in spectral shape was observed,
and this rules out underlying aggregate fluorescence at
lower temperature.
The fluorescence intensity of a fluorophore depends on
the natural lifetime of its first excited singlet state and on
the rate at which nonradiative processes deactivate it.
Quenching processes depend significantly on the location
and accessibility of the fluorophore within a macromolecular
structure. Therefore, fluorescent probes such as dansyl chlor-
ide are used especially in biochemistry to study the various
binding sites in large macromolecules.[29] Furthermore, for
pyrenyl moieties on the focal point of poly(amido) dendrim-
A
ers, decreasing Stern–Volmer quenching constants KSV were
found with increasing dendrimer generation.[4a] This was at-
tributed to a greater steric congestion with increasing den-
dron size making the chromophore less accessible. The
Stern–Volmer expression: [Eq. (1)]
F0=F ¼ 1þKSV½QŠ ¼ 1þhkqihti0½QŠ
ð1Þ
allows the quenching process to be investigated by measur-
ing the fluorescence intensity in the absence and presence of
the quencher, F0 and F, respectively. In Equation (1), KSV is
the Stern–Volmer quenching constant, hkqi<the average bi-
molecular quenching constant, hti0 the mean fluorophore
lifetime in the absence of the quencher, and [Q] the analyti-
cal concentration of the quencher. Comparing different fluo-
rophores with each other by means of their hkqi< values re-
quires determination of the mean fluorescence lifetime hti0.
We used nitromethane and N,N-dimethylaniline, both
well-known quenchers of pyrene fluorescence.[1b] The fluo-
rescence quenching of polyphenylene-dendronized pyrenes
and TPP was well described by the Stern–Volmer expression
(nitromethane: r2 >0.996; N,N-dimethylaniline: r2 >0.97)
(Figure 4a,b). By applying Equation (1), the Stern–Volmer
quenching constants KSV were determined from the slope of
the derived curves.
Figure 4. Stern–Volmer quenching plots for a) nitromethane, b) N,N-di-
methylaniline, and c) tetrachloromethane in CHCl3. Excitation: 360 nm.
Inset in b): Enlarged plot for N,N-dimethylaniline. F0 and F are the fluo-
rescence intensity in the absence and the presence of the quencher, re-
For nitromethane, the Stern–Volmer quenching constant
KSV decreased clearly on going from TPP to first-generation
~
&
dendrimer
6
and second-generation dendrimer 11a
*
*
spectively: : 6, &: 11a, : 20, : 21, : TPP.
(Table 2). In particular, no further decrease in KSV was ob-
served for third- and fourth-generation dendrimers 20 and
21, respectively.
N,N-Dimethylaniline as quencher had only a minor influ-
ence on the fluorescence intensity of the dendronized pyr-
enes. Therefore, experimental errors significantly contribute
to the obtained curves, as can be seen by the value of
r2 >0.97. The KSV values, derived
(Figure 4c). Similarly to nitromethane, KSV decreased from
TPP to first-generation dendrimer 6, but a further significant
decrease was only observed between third- and fourth-gen-
eration dendrimers 20 and 21, respectively (Table 2).
from a linear fit, were about 15
times smaller than for nitrome-
Table 2. Stern–Volmer quenching constants KSV for TPP, 6, 11a, 20, and 21 in CHCl3, respectively, obtained
thane and decreased significant-
ly only between the first- and
the second-generation dendrim-
with different quenching agents. Excitation: 360 nm.
[a]
KSV [mꢀ1
]
TPP
6
11a
20
21
ers
6 and 11a, respectively.
CH3NO2
17.27ꢁ1.18
6.98ꢁ0.17
8.41ꢁ0.42
5.72ꢁ0.28
0.33ꢁ0.04
6.20ꢁ0.60
3.43ꢁ0.19
0.19ꢁ0.03
6.35ꢁ0.71
3.77ꢁ0.22
0.16ꢁ0.04
5.30ꢁ0.35
3.57ꢁ0.12
0.19ꢁ0.04
0.62ꢁ0.18
When tetrachloromethane was
used as fluorescence quencher,
C6H5N
CCl4
A
diverse results were obtained [a] Errors were estimated by using Gaussꢁs law of propagation of error.
6122
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6117 – 6128

Pyrenes
FULL PAPER
The encapsulation effect of the dendronized pyrenes was
for charging. The EPR and UV/Vis spectra of the reduction
of 7 were similar to those observed after prolonged reduc-
tion of 11a, namely, a sharp signal (inset in Figure 5b) and
absorption maxima at 450 and 750 nm. On reduction of
fourth-generation dendrimer 21 a strong sharp signal imme-
diately appeared in the EPR spectrum (DHpp ꢂ0.1 mT, Fig-
ure 5b), as also found for 7.[33] However, after equilibrating
with neutral compound (comproportionation) a broad signal
with four shoulders (aH ꢂ0.2 mT) was found, similar to the
EPR spectrum of 11a, which strongly supports the assign-
ment of this signal to the radical monoanion of dendronized
pyrenes 11a and 21. The pyrene radical monoanion could
not be observed in the absorption spectra of 21 recorded
during reduction, mainly due to unresolved spectra. On fur-
ther reduction of 21 the sharp line in the EPR spectrum re-
appeared and revealed close similarity to the control sample
7. In all three samples the sharp line broadened on pro-
longed contact with the potassium mirror, but no defined
number of transferred electrons could be ascribed (DHpp
ꢂ0.6 mT).
also tested by means of the electron-transfer behavior of in-
creasingly shielded pyrene cores. Reduction of dendronized
pyrenes 11a and 21 was carried out on a potassium mirror
under high vacuum in absolute THFby using a technique
previously described.[30] Formation of the charged species
was followed by EPR and UV/Vis spectroscopy (for UV/Vis
spectra, see Supporting Information). On contact with the
potassium mirror the color of the solution of second-genera-
tion dendrimer 11a turned from light yellow to green, and
two absorption bands at 456 and 640 nm appeared in the
UV/Vis spectrum. This can be ascribed to formation of the
radical anion of the substituted pyrene.[31] The disappear-
ance of the absorption band of the neutral starting material
at 394 nm on prolonged contact suggested quantitative con-
version. The EPR spectroscopic monitoring showed first a
broad signal with four shoulders, which should be attributed
to the four protons at positions 4, 5, 9, and 10 of the pyrene
core (aH ꢂ0.2 mT, Figure 5a), whose splittings are very close
To investigate the solid-state photophysics of dendronized
pyrenes, the absorption and fluorescence spectra of alkyl-
chain-decorated second-generation dendrimer 11b were re-
corded. Films of good optical quality were obtained by
simple drop casting and spin coating from toluene solution
onto quartz substrates. The transparency of the films of den-
drimer 11b was much higher than for those made from
parent pyrene, mainly due to crystallization of the non-
dendronized species. Absorption and emission spectra of the
films showed a typical broadening relative to solution spec-
tra (Figure 6).
Figure 6. Absorption and emission spectra of 11b in a film prepared from
toluene solution. Excitation: 390 nm.
Figure 5. EPR spectra (RT, K, THF) a) 11a in the order of further reduc-
tion 1–3; b) 21 with initial spectrum 1, partial comproportionation 2, and
complete comproportionation 3. Inset in b): EPR spectrum for the reduc-
tion of 7.
Thin-film dendrimer 11b displayed an absorption maxi-
mum of at 393 nm, almost unshifted compared with solution
spectra. The absorption band in the UV region at 260 nm
can be assigned to the polyphenylene dendrons.[7a] Com-
pared with solution (337 nm), the absorption maximum of
unsubstituted pyrene in the film (337 nm) was not shifted at
all. The emission maximum of the film prepared from un-
substituted pyrene (446 nm) was red-shifted by 51 nm com-
pared to that in solution (395 nm), that is, emission in the
to those of the corresponding protons of the known pyrene
radical anion (aH =0.208 mT).[32]
On further reduction, the intensity of this signal was di-
minished, but instead of reaching a diamagnetic dianionic
state, a new sharp signal appeared in the center of the spec-
trum and increased in intensity (DHpp ꢂ0.1 mT). In a control
experiment, second-generation dendrimer 7 was considered
Chem. Eur. J. 2006, 12, 6117 – 6128
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6123

K. Müllen et al.
solid state originated mainly from aggregated species (exci-
tation: 390 nm). On the contrary, emission of 11b occurred
at 449 nm, that is, a bathochromic shift of only 20 nm com-
pared with solution spectra.
larger than those of N,N-dimethylaniline. For both quench-
ers almost equal kq values were reported.[35] It can thus be
suggested, that the different quenching behavior originates
from a size-exclusion effect provided by the polyphenylene
shell, which hampers diffusion of the larger quencher N,N-
dimethylaniline. Tetrachloromethane as fluorescence
quencher showed a further strong decrease in KSV between
dendrimers 20 and 21, probably due to an enhanced density
of phenyl rings in the outer shell of fourth-generation den-
drimer 21. This suggestion is somewhat confirmed by the
fact that steric congestion prevents defect-free growth of a
fifth-generation polyphenylene dendrimer when branching
unit 8 is used.[36]
Discussion
Monodisperse polyphenylene dendrimers with an encapsu-
lated pyrene moiety have been synthesized in high yield
starting from tetraethynyl-substituted pyrene core 4. The
crystal structure of the first-generation dendrimer 23
showed that the first-generation polyphenylene shell already
produces pronounced steric shielding of the pyrene core.
The shape persistence of the polyphenylene dendrons[6] sug-
gests that shielding would be even more pronounced for
higher generations. This expectation is indeed born out by
the fact that, compared to TPP as reference, third-genera-
tion dendrimer 17 showed a less solvatochromically shifted
emission maximum. One can attribute this to the creation of
a stable hydrophobic environment around the pyrene core,
due to the large number of surrounding phenyl rings. In con-
trast to dendrimers constructed from flexible building units,
the shape and conformation of the stiff polyphenylene den-
drons do not change with changing polarity of the surround-
ing medium.
No decrease in fluorescence quantum yield was observed
with increasing dendrimer generation. This is in contrast to
results from polyphenylene-dendronized perylenes, for
which the fluorescence quantum yield decreased from 0.83
for the first-generation dendrimer down to 0.73 for the
third-generation dendrimer.[7b] New nonradiative deactiva-
tion pathways were suggested in this case, which do not
seem to apply to a pyrene core.
Additional proof for efficient shielding of the core was
obtained by fluorescence quenching experiments with nitro-
methane and N,N-dimethylaniline. Note that KSV values can
only be compared with each other when the fluorescence
lifetimes of the investigated fluorophores are equal
[Eq. (1)]. Since detailed photophysical investigations were
not within the scope of this paper, fluorescence lifetimes of
the herein-described dendrimers will be reported in a future
paper. However, for polyphenylene dendrimers with pery-
lene diimide as core, fluorescence lifetimes hti0 were con-
stant for different sizes of the dendritic shell.[34] It can there-
fore be assumed that the dendrimers with pyrene as core
will also have constant fluorescence lifetimes irrespective of
generation. The KSV values significantly decreased on going
from TPP to first-generation dendrimer 6 and to second-
generation dendrimer 11a. Remarkably, KSV did not further
decrease for the third- and fourth-generation dendrimers.
This quenching behavior can not be ascribed only to smaller
diffusion constants of the larger dendrimers, since in that
case KSV would constantly decrease with increasing genera-
tion. Therefore, one can suggest that the second-generation
dendrons of 11a already produced the maximum achievable
steric shielding. The KSV values for nitromethane were much
In concentrated solutions (c>10ꢀ4 m) first-generation den-
drimer 10 displayed aggregation of the pyrene core, which
suggests that a first-generation polyphenylene shell is too
small to inhibit aggregation (Figure 3a). Surprisingly, tem-
perature-dependent fluorescence spectroscopy showed that,
even at high temperatures, a 10ꢀ2 m solution of the first-gen-
eration dendrimer still displayed mainly aggregation. A
higher temperature is normally expected to hamper self-as-
sociation due to increased rotational motion of the dendrons
within the molecules. However, the small increase in intensi-
ty of the absorption band at 440 nm indicated that at lower
concentration (c<10ꢀ2 m) 10 exists exclusively in the unasso-
ciated form at high temperature. Under the same conditions
second-generation dendrimer 11a displayed no aggregation
of the core, and this can again be attributed to the larger di-
ameter of the dendritic polyphenylene shell in the case of
11a (10: d = 2.5 nm, 11a: d = 3.9 nm).[37] This supports the
results derived from the quenching experiments. Moreover,
the small influence of temperature on the monomer–aggre-
gate equilibrium of dendronized pyrenes suggested a stiff,
temperature-independent arrangement of the dendritic
shell. This conclusion is confirmed by investigations on poly-
phenylene dendrimers by small-angle neutron scattering and
solid-state NMR spectroscopy.[6c,d] In the latter case the
slowed motion of the terminal phenyl groups was proven
and, furthermore, it was demonstrated that the dendrons
can not reorient even at high temperatures.
Alkali-metal reduction of second- and fourth-generation
dendrimers 11a and 21 produced the radical anion of the
pyrene core. However, significant differences were observed
during the reduction of the two dendrimers. With second-
generation dendrimer 11a the radical anion of pyrene was
formed immediately, as indicated by UV/Vis and EPR spec-
troscopy. On the contrary, in the case of fourth-generation
dendrimer 21 an intense sharp signal in the EPR spectrum
appeared at the beginning. The same sharp signal was ob-
served for reduction of polyphenylene dendrimer 7 without
a pyrene core under the same conditions. This leads to the
conclusion that charging of the polyphenylene shell of 7 and
21 immediately took place on contact with the potassium
mirror. The sharp EPR signal indicated a highly mobile
electron spin, which might be due to hopping between dif-
ferent oligomeric phenylene sites. Thus, no specific charged
bi- or triphenyl units could be identified.[38] After compro-
6124
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6117 – 6128

Pyrenes
FULL PAPER
portionation with neutral compound, the pyrene radical
anion of 21 was also detected. This suggests that a more
highly charged state was generated in the beginning, since
only in this case does comproportionation result in the radi-
cal anion. The observed differences in reduction between
11a and 21 can be attributed to the larger diameter of
fourth-generation dendrimer 21 (21: d=6.8 nm, 11a: d=
3.9 nm).[37] This hampered diffusion of electrons to the
pyrene core to such an extent that in the beginning only the
charged polyphenylene shell was detected by EPR spectro-
scopy. On the contrary, for 11a diffusion of electrons to the
core is easier due to the smaller distance between core and
outer dendritic shell; thus, the pyrene radical anion could
immediately be observed. A diffusion-controlled reduction
process can therefore be suggested, strongly influenced by
the density and diameter of the surrounding dendrimer
shell.
Experimental Section
General procedures: All starting materials, e.g., pyrene (1) and tetraphe-
nylcyclopentadienone (5), were obtained from commercial suppliers (Al-
drich, Fluka, Fischer, Strem, Acros) and were used without purification.
Solvents were used in HPLC grade as purchased. All atmosphere-sensi-
1
tive reactions were performed under argon using Schlenck techniques. H
and 13C NMR spectra were recorded on a Bruker AMX250, AC300,
AMX500, or 700Ultrashield NMR spectrometer. Specific protons in the
dendritic structures (positions shown in Scheme 2) are assigned as HG1
HG2, etc. for the corresponding dendrimer generation to which they
belong. FD mass spectra were recorded with a VG-Instruments ZAB 2-
,
SE-FDP. MALDI-TOF mass spectra were measured with
a Bruker
Reflex II and dithranol as matrix (molar ratio dithranol/sample 250:1).
UV/Vis absorption spectra were recorded on a Perkin-Elmer Lambda 9
spectrophotometer, and fluorescence spectra on a SPEX Fluorolog II
(212) spectrometer. For the temperature-dependent fluorescence meas-
urements an Oxford cryostat (GF1204) with sapphire windows was used.
EPR spectra were recorded on a CW X-band ESP 300 equipped with an
NMR gauss meter (Bruker ER 035), a frequency counter (Bruker ER
041 XK), and
a variable-temperature continuous-flow N2 cryostat
Since good film-forming ability is a major prerequisite for
compounds suitable for application in OLEDs, the optical
properties of films prepared from alkyl-decorated second-
generation dendrimer 11b were investigated. Films of good
quality and high transparency were obtained by simple drop
casting and spin coating. Small bathochromic shifts of the
emission maximum compared to solution spectra showed
that the emission in the film mainly occurred from unaggre-
gated species. Hence, the second-generation dendrons al-
ready suppress aggregation of pyrene, not only in solution
but also in film.
(Bruker B-VT 2000). Elemental analysis was carried out by the Micro-
analytical Laboratory of the University of Mainz, Mainz, Germany. Be-
cause of the high carbon content in some molecules, incomplete combus-
tion (sooting) may have resulted in lower values than expected for the
carbon content. The crystal structure determinations were carried out in
glass capillaries on a Nonius KCCD diffractometer with graphite-mono-
chromated MoKa radiation at 120 K. The structures were solved by direct
methods (SHELXS97). Refinement was done with anisotropic tempera-
tures factors for C, and the hydrogen atoms were refined with fixed iso-
tropic temperature factors in the riding mode.
General procedure for the Diels–Alder cycloaddition of ethynyl com-
pounds and tetraphenylcyclopentadiene derivatives: A mixture of the
ethynyl compound and tetraphenylcyclopentadienone derivative was re-
fluxed in o-xylene or diphenyl ether under argon atmosphere. In the case
of the unsubstituted dendrimers, the cooled reaction mixture was poured
into pentane to remove the excess of tetraphenylcyclopentadienone. The
precipitated product was filtered and the filter washed with pentane until
the filtrate became colorless. All crude products were purified by column
chromatography (silica gel, petroleum ether/CH2Cl2). Reactions were
conveniently monitored by MALDI-TOFmass spectrometry to ensure
their completeness.
Conclusion
A series of monodisperse polyphenylene dendrimers with an
encapsulated pyrene core has been prepared in high yield
by
a straightforward divergent growth method. Stern–
Volmer quenching experiments and temperature-dependent
fluorescence spectroscopy indicated that a second-genera-
tion dendrimer shell is sufficient for efficiently shielding the
pyrene core and thereby preventing aggregate formation.
Alkali-metal reduction of pyrene dendrimers yielded the
corresponding pyrene radical anions. Thereby, hampered
electron transfer to the core was observed with increasing
dendrimer generation, which further verifies the site-isola-
tion concept. From the synthetic point of view, it is interest-
ing that the improvements in the optical properties, such as
high quantum efficiency in solution (Qf >0.92), were already
achieved by applying a second-generation dendrimer shell,
and therefore gram-scale synthesis of these new materials is
possible.
General procedure for the desilylation of triisopropylsilylethynyl deriva-
tives: The triisopropylsilylethynyl derivative was dissolved in dry THF
and 1 equiv of TBAF(dissolved in THF) per triisopropylsilylethynyl
group was added under argon atmosphere. The end of the reaction (ca.
5–15 min) was determined by TLC. The reaction was quenched with
water, and the reaction mixture extracted with water and CH2Cl2. The or-
ganic phase was separated and dried over MgSO4. After having removed
the solvent under reduced pressure, the crude product was purified by
column chromatography (silica gel, petroleum ether/CH2Cl2).
1,3,6,8-Tetrabromopyrene (2): Pyrene (0.5 g, 2.47 mmol) was dissolved in
nitrobenzene (15 mL). Under vigorous stirring, bromine (1.62 g,
10.13 mmol) was added. After complete addition, the temperature was
increased to 1608C and maintained for 3 h. The cooled reaction suspen-
sion was poured into acetone and the precipitate filtered off. Further
drying of the precipitate in high vacuum gave the crude product, which
was used without further purification (1.15 g, 2.22 mmol, 90%). 1H NMR
(500 MHz, [D5]nitrobenzene, 433 K, TMS): d=8.44 (s, 2H; 2,7-pyrene
H), 8.42 ppm (s, 4H; 4,5,9,10-pyrene H); MS (FD, 8 kV): m/z (%): 517.98
(100) [M]+.
The herein-presented macromolecules are exciting new
materials that combine excellent optical features and an im-
proved film-forming ability. Thus, application in OLEDs can
be suggested and further exploration into this area is under-
way.
1,3,6,8-Tetrakis(trimethylsilanylethynyl)pyrene (3): Compound 2 (0.5 g,
0.97 mmol) was suspended in triethylamine (20 mL) and toluene (3 mL),
and bis(triphenylphosphine)palladium(ii) dichloride (136 mg, 0.19 mmol),
copper(i) iodide (73 mg, 0.39 mmol), and triphenylphosphine (101 mg,
0.39 mmol) were added. The flask was evacuated and flushed with argon.
While stirring, the reaction mixture was heated to 608C and trimethylsily-
lethyne (0.57 g, 5.8 mmol) was injected through a septum. After 15 min
Chem. Eur. J. 2006, 12, 6117 – 6128
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6125

K. Müllen et al.
of stirring at this temperature the reaction was heated to 808C and stir-
red overnight under argon atmosphere. The cooled reaction mixture was
diluted with CH2Cl2 and extracted with water. The organic phase was
dried over MgSO4, and the solvent was removed under reduced pressure.
The crude product was purified by column chromatography (silica gel,
petroleum ether) to afford 3 as an orange solid (470 mg, 0.8 mmol, 83%).
1H NMR (250 MHz, CD2Cl2, 300 K, TMS): d=8.59 (m, 4H; 4,5,9,10-
pyrene H), 8.27 (s, 2H; 2,7-pyrene H), 0.39 ppm (s, 36H; CH3); 13C NMR
(75 MHz, CD2Cl2, 300 K, TMS): d=134.7, 132.3, 127.3, 123.8, 119.1 (all
petroleum ether/CH2Cl2) to afford 11 as a yellow solid (360 mg, 76 mmol,
73%. 1H NMR (250 MHz, CD2Cl2, 300 K, TMS): d=7.78–7.61 (m, 6H;
pyrene H), 7.43–7.35 (m, 12H; HG1+G2), 7.16–6.53 ppm (m, 232H;
aromat. H); 13C NMR (75 MHz, [D8]THF, 300 K, TMS): d=142.8, 142.7,
142.0, 141.7, 141.4, 141.3, 141.1, 141.0, 140.1, 140.1, 140.0, 139.7, 139.4,
139.2, 138.9, 132.4, 132.0, 130.6, 129.5, 129.2, 129.1, 128.3, 127.6, 127.3,
126.9, 126.3, 126.0 ppm (all aromat. C); MALDI-TOF-MS: m/z (%):
4768 (100) [M]+; elemental analysis calcd (%) for C376H250: C 94.72, H
5.87; found C 94.00, H 5.87.
ꢀ ꢃ
ꢃ ꢀ
aromat. C), 102.8 (Ar C C), 102.1 ( C Si), 0.1 ppm (CH3); MS (FD,
8 kV): m/z (%): 587.24 (100) [M]+; elemental analysis calcd (%) for
C36H42Si4: C 73.65, H 7.21; found: C 73.14, H 7.55.
Compound 11b: Compounds 10 (18 mg, 9.4 mmol) and 14 (140 mg,
132 mmol) were dissolved in Ph2O (4 mL) and heated for 16 h at 2208C.
Purification was performed by column chromatography (silica gel, petro-
leum ether/CH2Cl2) to afford 11b as a yellow oil (60 mg, 5.9 mmol, 63%).
1H NMR (500 MHz, CD2Cl2, 300 K, TMS): d=7.80–6.42 (m, 234H;
aromat. H), 2.31 (m, 16H; CH), 1.27–1.19 (m, 672H; CH2), 0.88–
0.87 ppm (m, 96H; CH3); 13C NMR (125 MHz, CD2Cl2, 300 K, TMS): d=
168.2, 168.0, 156.9, 156.3, 142.8, 142.7, 142.5, 141.9, 141.7, 141.4, 141.4,
141.2, 141.0, 140.9, 140.9, 140.8, 140.2, 140.1, 139.9, 139.8, 139.6, 139.4,
139.3, 138.9, 138.7, 138.7, 138.5, 138.2, 135.6, 134.3, 133.9, 133.6, 133.3,
133.2, 133.0, 132.8, 132.3, 131.9, 131.8, 131.7, 131.5, 131.4, 131.4, 131.1,
130.7, 130.6, 130.4, 130.4, 130.3, 130.3, 130.2, 130.0, 129.9, 129.8, 129.5,
129.4, 129.3, 129.3, 129.2, 129.0, 128.9, 128.7, 128.6, 128.5, 128.4, 128.2,
128.0, 128.0, 127.9, 127.7, 127.6, 127.4, 126.6, 126.0, 125.6, 120.5 (all
aromat. C), 41.21, 40.77, 40.63, 40.22, 40.19, 39.71, 39.64, 39.51, 39.44 (all
aCH2/CH), 33.71, 33.62, 34.17, 34.00, 32.52, 30.65, 30.29, 30.25, 29.92,
27.10, 27.13, 23.23 (all CH2), 14.33 ppm (CH3); MALDI-TOF-MS: m/z
(%): 10161 (100) [M]+.
1,3,6,8-Tetraethynylpyrene (4): Compound 3 (1 g, 1.7 mmol) was suspend-
ed in methanol (100 mL). K2CO3 (1.9 g, 13.6 mmol) was added and the
reaction mixture stirred overnight and then poured into water (500 mL)
and filtered. The filter was washed with water until the filtrate was neu-
tral to afford 4 as a slightly brown solid (480 mg, 1.61 mmol, 95%).
1H NMR (300 MHz, [D8]THF, 300 K, TMS): d=8.70 (s, 4H; 4,5,8,10-
pyrene H), 8.36 (s, 2H; 2,7-pyrene H), 4.31 ppm (s, 4H; CH); 13C NMR
(75 MHz, [D8]THF, 300 K, TMS): d=135.6, 133.1, 127.7, 124.4, 119.2 (all
ꢃ
ꢀ ꢃ
ꢃ
aromat. C), 86.2 (Ar C C), 81.8 ppm ( CH); MS (FD, 8 kV): m/z (%):
298.08 (100) [M]+.
Compound 6: Compounds 4 (0.2 g, 0.67 mmol) and 5 (1.55 g, 4.03 mmol)
were heated in o-xylene (20 mL) for 12 h at 1608C. The reaction mixture
was poured into methanol and the precipitate filtered off. Compound 6
was obtained by repeated precipitation from ethanol (1.08 g, 0.63 mmol,
95%) until the red color of 5 had disappeared. 1H NMR (700 MHz,
C2D2Cl4, 373 K, TMS): d=7.86–7.40 (m, 6H; pyrene H), 7.24–6.52 ppm
(m, 84H; aromat. H); 13C NMR (75 MHz, CD2Cl2, 300 K, TMS): d=
142.1, 141.7, 141.1, 141.0, 140.7, 140.5, 140.3, 139.9, 131.9, 131.8, 131.3,
131.0, 130.3, 128.3, 128.2, 128.1, 127.4, 127.5, 126.9, 126.6, 126.4, 126.2,
125.6, 125.3, 125.2, 125.1 ppm (all aromat. C); MS (FD, 8 kV): m/z (%):
1725 (100) [M]+.
1,2-Bis[4-(2-decyltetradecyl)phenyl]ethane-1,2-dione (13): Compound 12
(2.2 g, 2.58 mmol) and iodine (330 mg, 1.3 mmol) were dissolved in
DMSO (30 mL). The mixture was stirred for 24 h at 1708C and the prod-
uct extracted with CH2Cl2 and water. The organic phase was removed in
vacuo and the residue was purified by column chromatography (silica
gel, hexane/ethyl acetate) to give 13 as a colorless oil (0.9 g, 1.0 mmol,
40%). 1H NMR (300 MHz, CD2Cl2, 300 K, TMS): d=7.86 (d, 3J
8.3 Hz, 4H; 2,2’,6,6’-aromat. H), 7.31 (d, 3J
(H,H)=8.3 Hz, 4H; 3,3’,5,5’-
aromat. H), 2.62 (d, 3J
(H,H)=7.0 Hz, 4H; aCH2), 1.66 (m, 2H; CH),
1.50–1.10 (m, 80H; CH2), 0.89 ppm (t, 3J
(H,H)=6.5 Hz, 12H; CH3);
A
Compound 9: Compounds
4 (250 mg, 0.84 mmol) and 8 (3.0 g,
AHCTREUNG
4.03 mmol) were heated in o-xylene (30 mL) for 12 h at 1608C. Purifica-
tion was performed by column chromatography (silica gel, petroleum
ether/CH2Cl2) to afford 9 as a yellow solid (1.88 g, 0.59 mmol, 71%).
1H NMR (250 MHz, CD2Cl2, 300 K, TMS): d=8.02–7.57 (m, 6H; pyrene
H), 7.42 (m, 4H; HG1), 7.19–6.61 (m, 72H; aromat. H), 1.10 ppm (s,
215H; CH/CH3); 13C NMR (75 MHz, [D8]THF, 300 K, TMS): d=142.3,
142.2, 141.9, 141.9, 141.8, 141.7, 141.7, 141.6, 141.6, 141.6, 141.6, 141.5,
141.5, 141.4, 141.3, 141.3, 141.2, 141.1, 140.9, 140.9, 140.7, 140.6, 140.5,
140.4, 140.4, 139.9, 139.6, 139.6, 139.5, 139.4, 139.4, 139.3, 137.4, 137.3,
137.2, 137.1, 136.8, 136.5, 136.0, 132.4, 132.4, 131.6, 131.3, 130.7, 130.6,
130.7, 129.7, 129.0, 128.9, 128.8, 128.7, 128.4, 128.0, 127.5, 127.2, 126.5,
AHCTREUNG
AHCTREUNG
13C NMR (75 MHz, CD2Cl2, 300 K, TMS): d=195.0 (C=O), 150.8 (4,4’-
aromat. C), 131.2 (2,2’,6,6’-aromat. C), 130.2 (3,3’,5,5’-aromat. C), 130.1
(1,1’-aromat. C), 41.2 (aCH2), 40.0 (CH), 33.5, 32.4, 30.3, 30.1, 29.8, 26.9,
23.1 (all CH2), 14.3 ppm (CH3); MS (FD, 8 kV): m/z (%): 441.8 (100)
[M]2+, 883.6 (45) [M]+.
3,4-Bis[4-(2-dodecyltetradecyl)phenyl]-2,5-diphenyl-cyclopenta-2,4-dien-
one (14): Compound 13 (1.37 g, 1.55 mmol) and 1,3-diphenylpropan-2-
one (326 mg, 1.55 mmol) were dissolved in degassed tBuOH (3 mL) at
808C. 1 mL of a 0.8m solution of tetrabutylammonium hydroxide in
methanol was diluted with tBuOH (3 mL) and added to the reaction mix-
ture. After 20 min the reaction was stopped by adding water (10 mL). Ex-
traction with CH2Cl2 and concentration of the organic layer gave a resi-
due which was further purified by column chromatography to obtain 14
ꢀ ꢃ
126.0, 125.2, 121.7, 121.4 (all aromat. C), 108.5 (Ar C ), 90.2, 90.0 (both
C Si), 19.0 (CH3), 12.2 ppm (CH); MALDI-TOF-MS: m/z (%): 3166
ꢃ ꢀ
(100) [M]+.
Compound 10: Compound 9 (1.43 g, 0.45 mmol) was dissolved in dry
THF(30 mL) and treated with TBAF(1.14 g, 3.6 mmol). Purification was
performed by column chromatography (silica gel, petroleum ether/
CH2Cl2) to afford 10 as a yellow solid (0.8 g, 0.42 mmol, 93%). 1H NMR
(250 MHz, CD2Cl2, 300 K, TMS): d=7.89–7.73 (m, 6H; pyrene H), 7.53–
as a purple oil (802 mg, 0.76 mmol, 49%). 1H NMR (300 MHz, CD2Cl2,
3
300 K, TMS): d=7.23 (s, 10H; aromat. H), 6.95 (d, J
A
aromat. H), 6.8 (d, 3J
aCH2), 1.40–1.10 (m, 80H; CH2), 0.88 ppm (t, 3J
ACHTREUNG
ꢃ
AHCTREUNG
7.43 (m, 4H; HG1), 7.17–6.60 (m, 80H; aromat. H), 3.02 ppm (m, 8H;
CH3); 13C NMR (75 MHz, CD2Cl2, 300 K, TMS): d=200.8 (C=O), 155.4,
143.2, 131.7, 130.8, 130.5, 129.6, 129.0, 128.3, 127.6, 125.3 (all aromat. C),
40.9 (aCH2), 40.0 (CH), 33.6, 32.3, 30.4, 30.1, 30.1, 29.8, 26.9, 23.1 (all
CH2), 14.3 ppm (CH3); MS (FD, 8 kV): m/z (%): 1058.2 (100) [M]+.
CH); 13C NMR (75 MHz, [D8]THF, 300 K, TMS): d=142.3, 142.2, 142.1,
141.9, 141.8, 141.7, 141.6, 141.6, 141.5, 141.5, 141.3, 141.3, 141.2, 141.2,
141.1, 140.9, 140.8, 140.7, 140.6, 140.6, 140.5, 140.4, 140.3, 140.2, 139.9,
139.6, 139.5, 139.5, 139.4, 139.4, 139.3, 137.3, 137.3, 137.3, 137.1, 136.9,
136.8, 136.5, 136.4, 136.1, 134.8, 134.6, 134.4, 134.2, 133.9, 132.7, 132.5,
132.3, 131.9, 131.6, 131.3, 130.7, 130.6, 130.1, 129.6, 128.9, 128.9, 128.7,
128.5, 128.1, 127.5, 127.3, 126.8, 126.5, 126.5, 126.3, 126.1, 126.0, 125.9,
Compound 15: Compounds 10 (300 mg, 0.16 mmol) and
8 (1.4 g,
1.88 mmol) were dissolved in o-xylene (10 mL) and heated for 20 h at
1508C. Purification was performed by column chromatography (silica gel,
ꢀ ꢃ
125.5, 125.4, 125.2, 125.1, 125.0, 121.0, 120.6 (all aromat. C), 84.2 (Ar C ),
petroleum ether/CH2Cl2) to afford 15 as
a yellow solid (0.91 g,
0.12 mmol, 76%). 1H NMR (250 MHz, CD2Cl2, 300 K, TMS): d=7.78–
7.61 (m, 6H; pyrene H), 7.42–7.31 (m, 12H; HG1+G2), 7.18–6.40 (m,
216H; aromat. H), 1.09 ppm (s, 336H; CH/CH3); 13C NMR (75 MHz,
[D8]THF, 300 K, TMS): d=142.6, 142.4, 141.9, 141.9, 141.8, 141.7, 141.4,
141.2, 141.0, 140.6, 140.5, 140.4, 140.3, 140.1, 140.0, 139.6, 139.5, 139.2,
139.2, 138.9, 137.4, 136.9, 136.6, 136.2, 132.3, 131.5, 131.2, 130.8, 130.6,
ꢃ
78.7, 78.6 (both CH), 67.4 ppm (Cq); MALDI-TOF-MS: m/z (%): 1916
(100) [M]+; elemental analysis calcd (%) for C152H90: C 95.27, H 4.73;
found C 94.61, H 4.91.
Compound 11a: Compounds 10 (200 mg, 0.1 mmol) and
1.67 mmol) were dissolved in o-xylene (10 mL) and heated for 16 h at
1608C. Purification was performed by column chromatography (silica gel,
5 (640 mg,
6126
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6117 – 6128

Pyrenes
FULL PAPER
130.1, 129.5, 129.2, 128.8, 128.5, 128.3, 127.8, 127.2, 127.0, 126.7, 126.7,
126.9, 126.3, 126.0 ppm (all aromat. C); MALDI-TOF-MS: m/z (%):
23019 (100) [M]+; elemental analysis calcd (%) for C1816H1210: C 94.70, H
5.30; found C 94.58, H 5.37.
ꢀ ꢃ
126.5, 123.8, 121.6, 121.3 (all aromat. C), 108.5, 108.5 (both Ar C ), 90.2,
ꢃ ꢀ
90.0 (both C Si), 67.4 (Cq), 19.0 (CH3), 12.2 ppm (CH); MALDI-TOF-
MS: m/z (%): 7657 (100) [M]+; elemental analysis calcd (%) for
1,3,6,8-Tetrakis(phenylethynyl)pyrene (22): Compound
2
(314 mg,
C
552H570Si16: C 86.62, H 7.51; found C 86.94, H 7.61.
61 mmol) was suspended in triethylamine (15 mL) and toluene (5 mL),
and bis(triphenylphosphine)palladium(ii) dichloride (130 mg, 0.18 mmol),
copper(i) iodide (45 mg, 0.24 mmol), and triphenylphosphine (64 mg,
0.24 mmol) were added. The flask was evacuated and flushed with argon.
Under stirring, the reaction mixture was heated to 608C and phenylacety-
lene (0.56 mL, 4.8 mmol) was injected through a septum. After 15 min of
stirring the reaction was heated to 808C and stirred overnight under
argon atmosphere. After cooling, the reaction mixture was filtered and
the orange precipitate 22 thoroughly washed with CH2Cl2 and acetone
(340 mg, 56.4 mmol, 92%). 1H NMR (250 MHz, CD2Cl2, 300 K, TMS):
d=7.78–7.69 (m, 6H; pyrene H), 7.56–7.02 ppm (m, 20H; aromat. H);
13C NMR (125 MHz, C2D2Cl4, 373 K, TMS): d=134.1 (1,3,6,8-pyrene C),
132.2, 132.1, 128.9, 128.7 (all aromat. C), 127.2 (2,7-pyrene C), 124.5,
Compound 16: Compound 15 (0.8 g, 0.1 mmol) was dissolved in dry THF
(15 mL) and treated with TBAF(518 mg, 1.67 mmol). Purification was
performed by column chromatography (silica gel, petroleum ether/
CH2Cl2) to afford 16 as a yellow solid (0.49 g, 95 mmol, 91%). 1H NMR
(250 MHz, CD2Cl2, 300 K, TMS): d=7.78–7.61 (m, 6H; pyrene H), 7.46–
7.32 (m, 12H; HG1+G2), 7.18–6.54 (m, 216H; aromat. H), 3.02 ppm (m,
16H; CH); 13C NMR (75 MHz, [D8]THF, 300 K, TMS): d=141.6, 141.3,
ꢃ
141.1, 139.5, 139.4, 139.0, 138.8, 138.5, 131.8, 131.5, 131.2, 130.8, 130.2,
129.0, 128.6, 128.1, 127.9, 127.4, 126.9, 126.2, 119.7, 119.4 (all aromat. C),
ꢀ ꢃ
ꢃ
83.8 (Ar C ), 77.3 ppm ( CH); MALDI-TOF-MS; m/z (%): 5147 (100)
[M]+; elemental analysis calcd (%) for C408H250: C 95.11, H 4.89; found C
94.38, H 5.76.
ꢀ ꢃ
123.6 (both aromat. C), 119.5 (4,5,9,10-pyrene C), 96.8 (Ar C ),
Compound 17: Compounds 16 (200 mg, 39 mmol) and
8 (0.93 g,
+
ꢀ ꢃ
88.1 ppm (pyrene C ); MS (FD, 8 kV): m/z (%): 603.1 (100) [M] ,
1.25 mmol) were dissolved in o-xylene (10 mL) and heated for 20 h at
1608C. Purification was performed by column chromatography (silica gel,
petroleum ether/CH2Cl2) to afford 17 as a yellow solid (0.55 g, 33 mmol,
85%). 1H NMR (250 MHz, CD2Cl2, 300 K, TMS): d=7.78–7.59 (m, 6H;
pyrene H), 7.40–7.27 (m, 28H; HG1–G3), 7.17–6.38 (m, 504H; aromat. H),
1.08 ppm (s, 672H; CH/CH3); 13C NMR (75 MHz, [D8]THF, 300 K,
TMS): d=142.8, 142.7, 142.7, 142.4, 141.9, 141.9, 141.8, 141.7, 141.4,
141.0, 140.9, 140.6, 140.1, 140.0, 139.9, 139.8, 139.5, 139.3, 139.2, 139.2,
139.1, 139.0, 138.8, 137.4, 136.9, 132.3, 132.0, 131.5, 131.2, 130.7, 130.6,
130.1, 129.4, 129.1, 128.5, 128.3, 127.8, 127.2, 126.9, 126.7, 126.5, 121.6,
301.7 (20) [M]2+
.
Compound 23: Compounds 22 (79 mg, 0.13 mmol) and
5 (300 mg,
0.78 mmol) were dissolved in Ph2O (5 mL) and heated for 24 h at 2308C.
Purification was performed by column chromatography (silica gel, petro-
leum ether/CH2Cl2) to afford 23 as a yellow solid (170 mg, 7.3 mmol,
90%). 1H NMR (500 MHz, C2D2Cl4, 373 K, TMS): d=7.31 (d, 3J
A
3
6.0 Hz, 2H; 2,7-pyrene H), 7.23 (d, J
A
H), 6.81–6.75 (m, 80H; aromat. H), 6.52 ppm (m, 20H; aromat. H);
13C NMR (500 MHz, C2D2Cl4, 373 K, TMS): d=141.6, 141.3, 140.8, 140.7,
138.6, 134.8, 134.0, 131.9, 131.8, 131.4, 126.6, 126.5, 126.3, 125.1,
125.0 ppm (all aromat. C); MS (FD, 8 kV): m/z (%): 2031.4 (100) [M]+.
ꢀ ꢃ
ꢃ
121.3 (all aromat. C), 108.5 (Ar C ), 90.2, 90.0 (both CH), 19.0 (CH3),
12.2 ppm (CH); MALDI-TOF-MS: m/z (%): 16640 (100) [M]+; elemen-
tal analysis calcd (%) for C1208H1210Si32: C 87.26, H 7.33; found C 86.22, H
7.79.
Compound 18: Compound 17 (0.4 g, 24.1 mmol) was dissolved in dry
THF(15 mL) and treated with TBAF(243 mg, 0.77 mmol). Purification
was performed by column chromatography (silica gel, petroleum ether/
Acknowledgements
Financial support by the Deutsche Forschungsgemeinschaft (SFB 625) is
gratefully acknowledged. We thank C. Beer for synthetic support.
CH2Cl2) to afford 18 as
a yellow solid (250 mg, 21.5 mmol, 89%).
1H NMR (250 MHz, CD2Cl2, 300 K, TMS): d=7.77–7.59 (m, 6H; pyrene
H), 7.42 (m, 8H; HG2), 7.38 (m, 16H; HG3), 7.31–7.29 (m, 4H; HG1),
ꢃ
7.13–6.44 (m, 504H; aromat. H), 3.04–2.00 ppm (m, 32H; CH);
[1] a) I. Carmichael, G. L. Hug, Handbook of Photochemistry, 2nd ed.,
Marcel Dekker, New York, 1993; b) J. B. Birks, Photophysics of Ar-
omatic Molecules, Wiley Interscience, London, 1970.
[2] a) E. M. S. Castanheira, J. M. G. Martinho, Chem. Phys. Lett. 1991,
185, 319–323; b) D. C. Dong, M. A. Winnik, Can. J. Chem. 1984, 62,
2560–2565.
13C NMR (75 MHz, [D8]THF, 300 K, TMS): d=142.8, 142.7, 142.3, 141.9,
141.8, 141.8, 141.7, 141.4, 141.0, 140.9, 140.5, 140.0, 139.9, 139.4, 139.3,
139.2, 139.2, 139.0, 138.8, 132.4, 132.0, 131.5, 131.2, 130.7, 130.6, 129.4,
129.1, 128.5, 128.3, 127.8, 127.2, 127.0, 126.9, 126.7, 126.5, 120.9, 120.6 (all
ꢀ ꢃ
ꢃ
aromat. C), 84.2 (Ar C ), 78.6 ppm ( CH); MALDI-TOF-MS: m/z (%):
11618 (100) [M]+.
[3] A. Thomas, S. Polarz, M. Antonietti, J. Phys. Chem. B 2003, 107,
5081–5087.
Compound 20: Compounds 10 (100 mg, 52 mmol) and 19 (950 mg,
0.83 mmol) were dissolved in o-xylene (10 mL) and heated for 2 d at
1708C. The solvent was removed under reduced pressure and the crude
product was purified by column chromatography (silica gel, petroleum
ether/CH2Cl2) to afford 20 as a yellow solid (310 mg, 29 mmol, 55%).
1H NMR (250 MHz, CD2Cl2, 300 K, TMS): d=7.76–7.60 (m, 6H; pyrene
H), 7.41 (m, 8H; HG2), 7.37 (m, 16H; HG3), 7.32–7.29 (m, 4H; HG1),
7.13–6.44 ppm (m, 536H; aromat. H); 13C NMR (75 MHz, [D8]THF,
300 K, TMS): d=142.8, 142.7, 142.0, 141.7, 141.4, 141.4, 141.3, 141.1,
141.0, 140.2, 140.1, 140.0, 140.0, 139.7, 139.5, 139.4, 139.4, 139.2, 138.9,
132.4, 132.0, 130.6, 129.6, 129.5, 129.2, 128.9, 128.3, 127.6, 127.3, 126.9,
126.5, 126.3, 126.3, 126.0 ppm (all aromat. C); MALDI-TOF-MS: m/z
(%): 10837 (100) [M]+; elemental analysis calcd (%) for C856H570: C
94.71, H 5.29; found C 94.50, H 5.24.
[4] a) C. M. Cardona, T. Wilkes, W. Ong, A. E. Kaifer, T. D. McCarley,
S. Pandey, G. A. Baker, M. N. Kane, S. N. Baker, F. V. Bright, J.
Phys. Chem. B 2002, 106, 8649–8656; b) S. Hecht, N. Vladimirov,
J. M. J. FrØchet, J. Am. Chem. Soc. 2001, 123, 18–25; c) L. A. Baker,
R. M. Crooks, Macromolecules 2000, 33, 9034–9039; d) R. Wilken,
J. Adams, Macromol. Rapid Commun. 1997, 18, 659–665.
[5] For OLEDs in general see, for example: a) A. P. Kulkarni, C. J. Ton-
zola, A. Babel, S. A. Jenekhe, Chem. Mater. 2004, 16, 4556–4573;
b) L. S. Hung, C. H. Chen, Mater. Sci. Eng. R 2002, 39, 143–222;
c) M. T. Bernius, M. Inbasekaran, J. OꢁBrien, W. S. Wu, Adv. Mater.
2000, 12, 1737–1750; d) Y. Cao, I. D. Parker, G. Yu, C. Zhang, A. J.
Heeger, Nature 1999, 397, 414–417. For pyrene containing devices,
see: e) W. L. Jia, T. McCormick, Q. D. Liu, H. Fukutani, M. Motala,
R. Y. Wang, Y. Tao, S. N. Wang, J. Mater. Chem. 2004, 14, 3344–
Compound 21: Compounds 18 (100 mg, 8.6 mmol) and
5 (420 mg,
0.56 mmol) were dissolved in Ph2O (8 mL) and heated for 5 d at 1908C.
Purification was performed by column chromatography (silica gel, petro-
leum ether/CH2Cl2) to afford 21 as a yellow solid (170 mg, 7.3 mmol,
85%). 1H NMR (250 MHz, CD2Cl2, 300 K, TMS): d=7.79–7.59 (m, 6H;
pyrene H), 7.41–7.30 (m, 58H; HG1–G4), 7.14–6.42 ppm (m, 1146H;
aromat. H); 13C NMR (75 MHz, [D8]THF, 300 K, TMS): d=142.8, 142.7,
142.0, 141.7, 141.4, 141.3, 141.1, 141.0, 140.1, 140.1, 140.0, 139.7, 139.4,
139.2, 138.9, 132.4, 132.0, 130.6, 129.5, 129.2, 129.1, 128.3, 127.6, 127.3,
´
´
3350; f) V. Cimrovµ, D. Vyprachticky, Appl. Phys. Lett. 2003, 82,
642–644; g) Y. Aso, T. Okai, Y. Kawaguchi, T. Otsubo, Chem. Lett.
2001, 420–421.
[6] a) M. Wind, U. M. Wiesler, K. Saalwächter, K. Müllen, H. W. Spiess,
Adv. Mater. 2001, 13, 752–756; b) H. Zhang, P. C. M. Grim, P. Fou-
bert, T. Vosch, P. Vanoppen, U. M. Wiesler, A. J. Berresheim, K.
Müllen, F. C. De Schryver, Langmuir 2000, 16, 9009–9014; c) S.
Rosenfeldt, N. Dingenouts, D. Potschke, M. Ballauff, A. J. Berre-
Chem. Eur. J. 2006, 12, 6117 – 6128
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6127

K. Müllen et al.
sheim, K. Müllen, P. Lindner, Angew. Chem. 2003, 115, 111–114;
Angew. Chem. Int. Ed. 2004, 43, 109–112; d) M. Wind, K. Saal-
wächter, U. M. Wiesler, K. Müllen, H. W. Spiess, Macromolecules
2002, 35, 10071–10086.
[22] Extinction coefficients are in line with values reported by others:
a) F. M. Winnik, M. A. Winnik, S. Tazuke, C. K. Ober, Macromole-
cules 1987, 20, 38–44; b) K. Hara, W. R. Ware, Chem. Phys. 1980,
51, 61–68.
[7] a) D. J. Liu, S. De Feyter, M. Cotlet, A. Stefan, U. M. Wiesler, A.
Herrmann, D. Grebel-Koehler, J. Q. Qu, K. Müllen, F. C. De Schryv-
er, Macromolecules 2003, 36, 5918–5925; b) A. Herrmann, T. Weil,
V. Sinigersky, U. M. Wiesler, T. Vosch, J. Hofkens, F. C. De Schryver,
K. Müllen, Chem. Eur. J. 2001, 7, 4844–4853.
[8] a) J. N. G. Pillow, M. Halim, J. M. Lupton, P. L. Burn, I. D. W.
Samuel, Macromolecules 1999, 32, 5985–5993; b) P. W. Wang, Y. J.
Liu, C. Devadoss, P. Bharathi, J. S. Moore, Adv. Mater. 1996, 8, 237.
[9] J. Q. Qu, D. J. Liu, S. De Feyter, J. Y. Zhang, F. C. De Schryver, K.
Müllen, J. Org. Chem. 2003, 68, 9802–9808.
[10] a) E. M. Harth, S. Hecht, B. Helms, E. E. Malmstrçm, J. M. J. FrØ-
chet, C. J. Hawker, J. Am. Chem. Soc. 2002, 124, 3926–3938; b) V.
Rozhkov, D. Wilson, S. Vinogradov, Macromolecules 2002, 35, 1991–
1993.
[11] C. B. Gorman, J. C. Smith, M. W. Hager, B. L. Parkhurst, H. Sierzpu-
towska-Gracz, C. A. Haney, J. Am. Chem. Soc. 1999, 121, 9958–
9966.
[12] a) K. Ogino, S. Iwashima, H. Inokuchi, Y. Harada, Bull. Chem. Soc.
Jpn. 1965, 38, 473–477; b) Vollmann, Liebigs Ann. Chem. 1937,
531, 1.
[13] a) Y. F. Wang, W. Deng, L. Liu, Q. X. Guo, Chin. J. Org. Chem.
2005, 25, 8–24; b) K. Sonogashira, S. Takahashi, J. Synth. Org.
Chem. Jpn. 1993, 51, 1053–1063; c) H. A. Dieck, F. R. Heck, J. Or-
ganomet. Chem. 1975, 93, 259–263.
[23] D. F. Eaton, Pure Appl. Chem. 1988, 60, 1107–1114. Qf: 0.90, sol-
vent: cyclohexane, excitation wavelength: 350 nm.
[24] D. C. Dong, M. A. Winnik, Can. J. Chem. 1984, 62, 2560–2565.
[25] a) M. Stepanek, K. Krijtova, K. Prochazka, Y. Teng, S. E. Webber, P.
Munk, Acta Polym. 1998, 49, 96–102; b) M. Stepanek, K. Krijtova,
Z. Limpouchova, K. Prochazka, Y. Teng, P. Munk, S. E. Webber,
Acta Polym. 1998, 49, 103–107.
[26] T. Fçrster, Angew. Chem. 1969, 81, 364–374; Angew. Chem. Int. Ed.
Engl. 1969, 8, 333–343.
[27] T. Keeling-Tucker, J. D. Brennan, Chem. Mater. 2001, 13, 3331–
3350.
[28] Compound 10 was chosen instead of 6 due to its better solubility.
The peripheral ethynyl groups were not expected to significantly en-
large the dendrimer dimensions. The absorption maxima of 10 and
11a did not shift with increasing temperature.
[29] a) J. Zhang, R. E. Campbell, A. Y. Ting, R. Y. Tsien, Nat. Rev. Mol.
Cell Biol. 2002, 3, 906–918; b) D. L. Taylor, Y. L. Wang, Nature
1980, 284, 405–410.
[30] M. Baumgarten, K. Müllen, Top. Curr. Chem. 1994, 169, 1–103.
[31] K. H. J. Buschow, J. Dieleman, G. J. Hoijtink, J. Chem. Phys. 1965,
42, 1993–1998.
[32] I. C. Lewis, L. S. Singer, J. Chem. Phys. 1965, 43, 2712–2727.
[33] For ease of comparison the EPR spectra of 7, 11a, and 21 are col-
lected in the Supporting Information.
[14] a) U. M. Wiesler, T. Weil, K. Müllen, Top. Curr. Chem. 2001, 212, 1–
40; b) F. Morgenroth, E. Reuther, K. Müllen, Angew. Chem. 1997,
109, 647–649; Angew. Chem. Int. Ed. Engl. 1997, 36, 631–634.
[15] F. Morgenroth, K. Müllen, Tetrahedron 1997, 53, 15349–15366.
[16] U. M. Wiesler, K. Müllen, Chem. Commun. 1999, 2293–2294.
[17] W. Pisula, M. Kastler, D. Wasserfallen, T. Pakula, K. Müllen, J. Am.
Chem. Soc. 2004, 126, 8074–8075.
[34] R. Pilot, E. Fron, F. C. De Schryver, Katholieke Universiteit Leuven
(Belgium), personal communication.
[35] Pyrene on the focal point of a first-generation poly(amido) dendri-
N
mer.[4a] kq nitromethane 5.66ꢁ0.14ꢂ109 mꢀ1 sꢀ1, kq N,N-dimethylani-
line 6.24ꢁ0.30ꢂ109 mꢀ1 sꢀ1
.
[36] U. M. Wiesler, Dissertation, Johannes Gutenberg Universität Mainz
(Germany), 2001.
[18] R. E. Bauer, V. Enkelmann, U. M. Wiesler, A. J. Berresheim, K.
Müllen, Chem. Eur. J. 2002, 8, 3858–3864.
[19] CCDC-285925 contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from the Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[20] The absorption maxima lmaxabs and emission maxima lmaxem and cor-
responding molar extinction coefficients e for unsubstituted pyrene
and the dendronized pyrenes are given in the Supporting Informa-
tion.
[37] Distances were derived from three-dimensional structures calculated
by the MMFF method. T. A. Halgren, J. Comput. Chem. 1996, 17,
490–519.
[38] a) T. Shida, S. Iwata, J. Am. Chem. Soc. 1973, 95, 3473–3483;
b) D. F. Lindow, C. N. Cortez, R. G. Harvey, J. Am. Chem. Soc.
1972, 94, 5406; c) T. Shida, S. Iwata, J. Chem. Phys. 1972, 56, 2858;
d) T. Shida, S. Iwata, J. Phys. Chem. 1971, 75, 2591; e) P. Balk, S. De-
bruijn, G. J. Hoijtink, Recl. Trav. Chim. Pays-Bas 1957, 76, 907–918.
Received: August 16, 2005
Revised: March 6, 2006
[21] I. B. Berlman, J. Phys. Chem. 1970, 74, 3085–3093.
6128
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 6117 – 6128