Ladder-type pentaphenylenes and their polymers: Efficient blue-light emitters and electron-accepting materials via a commmon intermediate

Article abstract of DOI:10.1021/ja0398823

A new route to ladder-type pentaphenylenes has been developed in which both good hole-accepting p-type and electron-accepting n-type materials can be prepared from a common intermediate. This key intermediate is a pentaphenylene diester 5 obtained in high yield by Suzuki coupling of 2 equiv of fluorene boronates with 2,5-dibromoterephthalate. Addition of aryllithium followed by ring closure with boron trifluoride produced a blue-emitting ladder-type pentaphenylene. Bromination followed by reductive polymerization with nickel(0) gave new high molecular mass polymers, which show efficient blue emission with a very small Stokes shift. These polymers bridge the gap in emission between polyfluorenes and fully ladder-type polyphenylenes. An alternative ring closure of the dibromopentaphenylene diester 14 with acid made a diketone that is a good electron-accepting material, as it displays a reversible two-electron reduction. The reduction onset potential of -0.875 V against Ag/Ag⁺ corresponds to a lowest unoccupied molecular orbital (LUMO) energy level of 3.53 eV, comparable to the work function of magnesium, suggesting that this unit could be used to greatly increase the injection of electrons into polymers containing it in a light-emitting diode (LED) or solar cell. A red-emitting material was prepared by Suzuki coupling of the dibromopentaphenylene 10b with a perylene dye, thus offering the prospect of tuning the emission from pentaphenylene materials over the whole visible range by attachment of suitable dyes. Unoptimized single-layer organic LEDs that used 11b showed stable pure-blue emission with brightnesses of over 200 cd/m² at 7 V, with moderate efficiencies.

Full text of DOI:10.1021/ja0398823

Published on Web 05/12/2004
Ladder-Type Pentaphenylenes and Their Polymers: Efficient
Blue-Light Emitters and Electron-Accepting Materials via a
Common Intermediate
Josemon Jacob,† Stefan Sax,‡ Thomas Piok,‡ Emil J. W. List,‡
Andrew C. Grimsdale,† and Klaus Mu¨llen*,†
Contribution from the Max-Planck-Institute for Polymer Research, Ackermannweg 10,
D 55128 Mainz, Germany, Christian Doppler Laboratory “AdVanced Functional Materials”,
Institute of Solid State Physics, Graz UniVersity of Technology, Petersgasse 16,
A-8010 Graz, Austria, and Institute of Nanostructured Materials and Photonics,
Franz-Pichler-Strasse 30, A-8160 Weiz, Austria
Received December 1, 2003; E-mail: muellen@mpip-mainz.mpg.de
Abstract: A new route to ladder-type pentaphenylenes has been developed in which both good hole-
accepting p-type and electron-accepting n-type materials can be prepared from a common intermediate.
This key intermediate is a pentaphenylene diester 5 obtained in high yield by Suzuki coupling of 2 equiv
of fluorene boronates with 2,5-dibromoterephthalate. Addition of aryllithium followed by ring closure with
boron trifluoride produced a blue-emitting ladder-type pentaphenylene. Bromination followed by reductive
polymerization with nickel(0) gave new high molecular mass polymers, which show efficient blue emission
with a very small Stokes shift. These polymers bridge the gap in emission between polyfluorenes and fully
ladder-type polyphenylenes. An alternative ring closure of the dibromopentaphenylene diester 14 with acid
made a diketone that is a good electron-accepting material, as it displays a reversible two-electron reduction.
The reduction onset potential of -0.875 V against Ag/Ag+ corresponds to a lowest unoccupied molecular
orbital (LUMO) energy level of 3.53 eV, comparable to the work function of magnesium, suggesting that
this unit could be used to greatly increase the injection of electrons into polymers containing it in a light-
emitting diode (LED) or solar cell. A red-emitting material was prepared by Suzuki coupling of the
dibromopentaphenylene 10b with a perylene dye, thus offering the prospect of tuning the emission from
pentaphenylene materials over the whole visible range by attachment of suitable dyes. Unoptimized single-
layer organic LEDs that used 11b showed stable pure-blue emission with brightnesses of over 200 cd/m2
at 7 V, with moderate efficiencies.
Introduction
accepting components in photovoltaic devices, and to increase
the efficiency of LEDs by improving electron injection and
transport.
Phenylene-based polymers are one of the most important
Conjugated organic oligomers and polymers are the subject
of considerable current interest as potential active materials in
electronic applications such as organic light-emitting diodes
(OLEDs),1,2 plastic lasers,3 photovoltaic devices (solar cells),4
and field-effect transistors.5 Two of the most important goals
of research on organic materials for electronic applications are
the obtaining of stable blue emission,6 which is particularly
required for full-color displays, and the preparation of good
electron-accepting and -transporting (n-type) materials. Most
organic materials are p-type materials, which accept and
transport holes better than electrons, but n-type materials are
urgently needed for use in n-type transistors, as the electron-
classes of conjugated materials for electronic applications. In
particular they are of great interest as potential blue-emitting
materials in applications such as LEDs and polymer lasers. Three
classes of polyphenylenes that have been of particular interest
are polyfluorenes (PF, 1),7 polyindenofluorenes (PIF, 2),8,9 and
ladder-type poly-p-phenylenes (LPPP, 3)10 (Figure 1). These
materials all show efficient blue or blue-green emission in
solution with the emission color red-shifting as the chain rigidity
increases from 420 nm for 1, through 430 nm for 2, to 450 nm
for 3. However, the blue emission in the solid state is unstable,
due to the rapid appearance of long-wavelength emission
bands.7,9,10 These were initially assigned to aggregation of the
polymer chains,7,9,11 but more recently it has been shown that
† Max-Planck-Institute for Polymer Research.
‡ Graz University of Technology and Institute of Nanostructured Materials
and Photonics.
(1) Kraft, A.; Grimsdale, A. C.; Holmes, A. B. Angew. Chem., Int. Ed. 1998,
37, 402.
(2) Mitschke, U.; Ba¨uerle, P. J. Mater. Chem. 2002, 10, 1471.
(3) McGehee, M. D.; Heeger, A. J. AdV. Mater. 2002, 12, 1655.
(4) Brabec, C. J.; Sariciftci, N. S.; Hummelen, J. C. AdV. Funct. Mater. 2001,
11, 15.
(7) Neher, D. Macromol. Rapid Commun. 2001, 22, 1365.
(8) Setayesh, S.; Marsitzky, D.; Mu¨llen, K. Macromolecules 2000, 33, 2016.
(9) Grimsdale, A. C.; Lecle`re, Ph.; Lazzaroni, R.; MacKenzie, J. D.; Murphy,
C.; Setayesh, S.; Silva, C.; Friend, R. H.; Mu¨llen, K. AdV. Funct. Mater.
2002, 12, 729.
(5) Dimitrakopoulos, C. D.; Malefant, P. R. L. AdV. Mater. 2002, 14, 99.
(6) Kim, D. Y.; Cho, H. N.; Kim, C. Y. Prog. Polym. Sci. 2000, 25, 1089.
(10) Scherf, U. J. Mater. Chem. 1999, 9, 1853.
9
10.1021/ja0398823 CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 6987-6995
6987

A R T I C L E S
Jacob et al.
Figure 1. Polyphenylene-based conjugated polymers.
the emission from PFs arises from emissive fluorenone defects
formed by oxidation at the methine bridge.12 There is also
evidence suggesting that similar ketone defects may be respon-
sible for the long-wavelength emission from PIF13 and LPPP.14
Substitution of the bridgehead hydrogens in LPPP (3a) with
methyl groups to give Me-LPPP (3b)10 leads to suppression of
the long-wavelength emission seen for 3a but unfortunately also
red-shifts the emission, so that 3b is a blue-green emitter with
an emission maximum at 470 nm. We have shown that
attachment of aryl substituents to the bridgehead positions of
PFs15,16 and PIFs17 suppresses the long-wavelength emission,
due to both a lower susceptibility of the materials to oxidation
and a reduction of exciton migration to any defect sites that
might be formed. However, the emission maxima for these
materials are around 425-435 nm, where the eye is not
particularly sensitive, and an emission maximum around 440-
450 nm is more desirable for maximum apparent brightness,
i.e., better conversion of charge carriers to light observable by
the human eye. A structure intermediate between PIF and
Me-LPPP would show an emission maximum in this region. A
suitable polymer is thus one in which the repeating unit is a
bridged pentaphenylene. Such units have been prepared by
Scherf and Grimme18 and by Xia and Advincula19 and reported
to show photoluminescence (PL) emission maxima close to 430
nm, suggesting an emission maximum for the corresponding
polymer of around 440 nm.
electrode materials. It has been previously shown that the hole
injection into PFs can be improved by incorporation of
triarylamine units within the chain20 as end-capping units21 or
as side chains.16 However, the problem of improving the
electron-accepting properties remains. We have previously
demonstrated that poly(9-fluorenone) (4) is a good electron-
accepting material,22 but this polymer is insoluble and has to
be prepared by acid treatment of a polyketal precursor, which
leads to problems in device fabrication as the material must be
dedoped lest residual acid damage the electrodes. A soluble
analogue of 4 would thus be a potential electron-transporting/
emissive material. We anticipated that a ladder-type pentaphenyl
with one or two ketone groups at the bridges would show good
electron-accepting properties and that the polymer should be
soluble.
We chose the pentaphenylene diester 5 (Figure 2) as the key
intermediate in our new synthesis of ladder-type pentaphen-
ylenes. We foresaw that this could be readily converted into a
blue-emitting p-type compound 6, intended as a monomer for
a blue-emitting polymer, and into a diketone 7, which we
expected to be a good electron-accepting (n-type) material. In
this way one could facilely prepare p- or n-type materials from
a common intermediate.
We now report the short, efficient synthesis of the desired
ladder-type pentaphenylenes, a blue-emitting polymer based on
6, and an electron-accepting copolymer containing the diketone
7.
A second problem with phenylene-based polymers is that of
charge injection due to the large differences between the energy
levels for the polymer orbitals and the work functions of typical
Results and Discussion
The synthesis of the blue-emitting monomer is shown in
Scheme 1. Suzuki coupling of the fluorene-2-boronate ester 8
with the dibromoterephthalate 9 generated the key intermediate,
the pentaphenylenediester 5, in 92% yield. Addition of an excess
of 4-alkylphenyllithium produced a diol that was facilely ring-
closed by use of BF3 etherate to generate the ladder-type
pentaphenylenes 6a (73%) and 6b (95%). Bromination of 6 by
CuBr2 on alumina23 gave the monomers 10a (56%) and 10b
(91%). The polymers 11 were synthesized by a nickel(0)-
mediated Yamamoto-type polymerization. Polymer 11a had only
limited solubility in tetrahydrofuran (THF), toluene, and dichlo-
(11) (a) Lemmer, U.; Heun, S.; Mahrt, R. F.; Scherf, U.; Hopmeier, M.; Siegner,
U.; Go¨bel, E. O.; Mu¨llen, K.; Ba¨ssler, H. Chem. Phys. Lett. 1995, 240,
373. (b) Pannozo, S.; Vial, J.-C.; Kervalla, Y.; Ste´phan, O. J. Appl. Phys.
2002, 92, 3495. (c) Herz, L. M.; Phillips, R. T. Phys. ReV. B 2000, 61,
13691. (d) Zeng, G.; Chua, W. L. S. J.; Huang, W. Macromolecules 2002,
35, 6907.
(12) (a) Scherf, U.; List, E. J. W. AdV. Mater. 2002, 14, 477. (b) Gong, X.;
Iyer, P. K.; Moses, D.; Bazan, G. C.; Heeger, A. J.; Xiao, S. S. AdV. Funct.
Mater. 2003, 13, 325. (c) Gaal, M.; List, E. J. W.; Scherf, U. Macromol-
ecules 2003, 36, 4236.
(13) Keivanidis, P. E.; Lupton, J. M.; Im, C.; Grimsdale, A. C.; Mu¨llen, K.
Unpublished results.
(14) Lupton, J. M. Chem. Phys. Lett. 2002, 365, 366.
(15) (a) Setayesh, S.; Grimsdale, A. C.; Weil, T.; Enkelmann, V.; Mu¨llen, K.;
Meghdadi, F.; List, E. J. W.; Leising, G. J. Am. Chem. Soc. 2001, 123,
946. (b) Pogantsch, A.; Wenzl, F. P.; List, E. J. W.; Grimsdale, A. C.;
Mu¨llen, K.; Leising, G. AdV. Mater. 2002, 14, 1061.
(16) Ego, C.; Grimsdale, A. C.; Uckert, F.; Yu, G.; Srdanov, G.; Mu¨llen, K.
AdV. Mater. 2002, 14, 809.
(20) Bernius, M.; Inbasekaran, M.; O’Brien, J.; Wu, W. AdV. Mater. 2000, 12,
1737.
(17) (a) Jacob, J.; Oldridge, L.; Zhang, J.; Gaal, M.; List, E. J. W.; Grimsdale,
A. C.; Mu¨llen, K. Curr. Appl. Phys. 2004, 4, 339. (b) Jacob, J.; Zhang, J.;
Grimsdale, A. C.; Mu¨llen, K.; Gaal, M.; List, E. J. W. Macromolecules
2003, 36, 8240.
(21) Miteva, T.; Meisel, A.; Knoll, W.; Nothofer, H.-G.; Scherf, U.; Mu¨ller, D.
C.; Meerholz, K.; Yasuda, A. AdV. Mater. 2001, 13, 565.
(22) (a) Uckert, F.; Setayesh, S.; Mu¨llen, K. Macromolecules 2001, 34, 4519.
(b) Uckert, F.; Tak, Y.-H.; Mu¨llen, K.; Ba¨ssler, H. AdV. Mater. 2000, 12,
905.
(18) Scherf, U.; Grimme, J. Macromol. Chem. Phys. 1996, 197, 2297.
(19) Xia, C.; Advincula, R. C. Macromolecules 2001, 34, 6922.
(23) Ranger, M.; Rondeau, D.; Leclerc, M. Macromolecules 1997, 30, 7686.
9
6988 J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004

New Route to Ladder-Type Pentaphenylenes
A R T I C L E S
Figure 2. Pentaphenylene monomers and the key intermediate in their synthesis.
Scheme 1. Synthesis of Poly(pentaphenylene)s 11
romethane, but 11b was readily soluble in all these solvents.
Gel-permeation chromatography (GPC) analysis with PPP (poly-
p-phenylene) standards showed number-averaged molecular
masses (Mn) of 4.38 × 104 and 1.36 × 105 for 11a and 11b,
respectively, corresponding to degrees of polymerization of
about 31 and 83, with polydispersities of 3.27 and 2.19.
The presence of bridgehead hydrogens due to incomplete
alkylation has been shown to promote the formation of ketone
defects in polyfluorenes,12a and so we tested the starting fluorene
derivative and the pentaphenylenes 6 for their presence. We
have found that upon addition of butyllithium to incompletely
alkylated fluorenes in dry THF the appearance of the distinct
red color of a fluorenyl anion can be clearly seen, even when
the bridgehead hydrogens are not detectable by NMR, thus
providing a simple but effective qualitative test for incomplete
alkylation of fluorenes. No such color change was seen during
the preparation of the boronate 8, which we take as good
evidence that there are no residual bridgehead hydrogens present
in 8. We also observed no color change upon testing solutions
of the pentaphenylenes 6 with butyllithium, from which we
conclude that they contain no detectable bridgehead hydrogens.
Furthermore, no evidence of ketones was seen in the Fourier
transform infrared (FTIR) spectra of the polymers 11.
absorption and emission spectra of 11b are given in Figure 3.
The polymers displayed blue photoluminescence (PL) in solution
and both polymers 11a and 11b had an absorption maximum
at 434 nm and an emission maximum at 445 nm. The optical
properties are very similar to those of LPPP, with a very small
Stokes shift between the absorption (434 nm) and emission (445
nm) maxima, indicative of the high degree of rigidity of the
As expected, the optical properties of the polymers 11 were
intermediate between those of PIF and LPPP. The solution
Figure 3. Absorption and emission spectra of polymer 11b in solution
(THF).
9
J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004 6989

A R T I C L E S
Jacob et al.
Scheme 2. Synthesis of Pentaphenylenedione 15
structure. A secondary emission peak is seen around 470 nm
with a slight tail into the yellow region of the spectrum. The
position of the emission maximum lies about midway between
those of PIF (434 nm) and LPPP (450 nm), so that with these
polymers the gap between unbridged and ladder-type poly-
phenylenes has been closed.
When the absorption and emission spectra of polymer 11b
in film (Figure 4) and solution (Figure 3) are compared, very
similar spectra with no bathochromic shift are observed. This,
together with the observation of a small Stokes shift, is not
unexpected due to the ladder-type structure of the polymer
backbone. The photoinduced absorption (PA) spectrum displays
remarkable similarities to those of polymers 1 and 3b,24 with
distinct features at 900, 700, and 620 nm. Given the similarity
in the chemical structures of polymers 3b and 11b, the feature
at 900 nm is assigned to a triplet-triplet absorption, whereas
the features at 700 and 620 nm are assigned to the absorption
of a polaron and its vibronic progression.
Moreover, when the relative intensities of the PA signals are
analyzed as compared to films of polymer 3b recorded under
comparable conditions (excitation density, film thickness, tem-
perature),25 it has to be noted that polymer 11b does show a
dramatically lower polaron as well as triplet exciton steady-
state density under similar conditions. This is attributed to the
fact that polymer 11b does possess a low energetic disorder in
the bulk material, which is also reflected in the steep onset of
PL in the films. Since a high steady-state polaron as well as
triplet exciton density represents a serious drawback in OLED
and solid-state laser applications due to nonradiative quenching
of singlet excitons at trapped polarons and triplet excitons,25,26
polymer 11b is thus a promising candidate for such applications.
As the bromination of the pentaphenylene 6b requires careful
monitoring to avoid overbromination, we investigated the
possibility of using a trimethylsilyl group as a synthon to permit
controlled introduction of the bromines (Scheme 2). 2,7-
Dibromo-9,9-dioctylfluorene (12) was converted via the mono-
silyl compound to the boronate 13. Again no evidence of
residual bridgehead hydrogens was seen during the addition of
butyllithium to 12.
Suzuki coupling of this with the terephthalate 9 gave a crude
disilylpentaphenylene in 86% yield. We found that it was not
possible to satisfactorily purify this compound, and so the crude
product was converted to the dibromide 14 with sodium acetate
and bromine (72%). Because the bromodesilylation occurs far
(24) List, E. J. W.; Partee, J.; Shinar, J.; Scherf, U.; Mu¨llen, K.; Graupner, W.;
Petritsch, K.; Zojer, E.; Leising, G. Phys. ReV. B 2000, 61, 10807.
(25) List, E. J. W.; Kim, C.-H.; Naik, A. K., Scherf, U.; Leising, G.; Graupne,
W.; Shinar, J. Phys. ReV. B 2000, 64, 155204.
(26) List, E. J. W.; Scherf, U.; Mu¨llen, K.; Graupner, W.; Kim, C.-H.; Shinar,
J. Phys. ReV. B 2002, 66, 235031.
Figure 4. Photoluminescence excitation (red), emission (blue), and
photoinduced absorption (green) spectra of polymer 11b film.
9
6990 J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004

New Route to Ladder-Type Pentaphenylenes
A R T I C L E S
Figure 5. Absorption and emission spectra of 7 in THF.
Figure 6. CV of diketone 15.
more quickly than the uncatalyzed bromination of the penta-
phenylene, by stopping the reaction immediately after the
bromodesilylation was complete we avoided any overbromina-
tion of 14. The monomers 10 could then be obtained by addition
of aryllithium and ring closure, but in view of the longer
synthesis and lower overall yield, this route is inferior to that
previously described. Treatment of 14 with concentrated sulfuric
acid at 165 °C was found to induce a double intramolecular
Friedel-Crafts acylation to produce the desired diketone 7 in
69% yield. By contrast, treatment of the unbrominated diester
5 under the same condition gave incomplete ring closure. Figure
5 shows the optical properties of 7 in solution. The unusually
large gap between the absorption and emission maxima we
interpret as arising from the presence of two chromophore
unitssthe pentaphenylene backbone and the aryldiketone sys-
tem. The absorption spectrum closely matches that for 6,
indicating that the absorption is from the former, but the
expected emission around 400 nm is very weak, and instead
the major emission occurs at 619 nm from the diketone system,
with a corresponding weak absorption band at 550 nm, due to
internal energy transfer between the two systems. More detailed
studies including polarized absorption and emission measure-
ments are underway to confirm this hypothesis.
To investigate the redox properties of this system, we treated
7 with an excess of benzeneboronic acid under Suzuki coupling
conditions to obtain the model compound 15. Cyclic voltam-
metry was performed on 15 against Ag/AgCl with a ferrocene/
ferrocenium internal standard27 (Figure 6). The diketone showed
fully reversible p- and n-doping, with two reduction peaks at
-1.0 and -1.35 V, which we assign to the reduction of the
diketone to a quinonoidal dianion, and an oxidation peak at 1.46
V. The onsets for reduction and oxidation occurred at -0.875
and 1.31 V, from which we estimate the lowest unoccupied and
highest occupied molecular orbital (LUMO and HOMO) energy
levels to be 3.53 and 5.71 eV, respectively, given an energy
level of 4.4 eV for Ag/AgCl.7 This corresponds to an electro-
chemical band gap of 2.18 eV. The estimated LUMO energy
means that there should be small barriers to injection of electrons
into materials containing this unit from relatively air-stable
metals such as magnesium (work function 3.7 eV) or aluminum
(4.2 eV), thus obviating the need to use unstable low work
function metals such as calcium as cathodes. Polymerization
of 7 with nickel(0) gave an insoluble material, but copolymer-
ization of 7 (20 mol %) with 6 produced a soluble polymer
with good electron-accepting properties that is currently being
evaluated as an electron-accepting component in photovoltaic
devices. The results of those investigations will be published
separately.
The pentaphenylenes 10 are blue emitters with PL maxima
at 424 nm in solution. We have previously shown that efficient
tuning of EL emission from polyfluorenes across the visible
spectrum can be accomplished by introduction of perylene dyes
into the polymer chain.28 Since the emission maxima of 10 are
close to those of polyfluorenes (425 nm), we anticipated that
equally efficient energy transfer would occur if they were to be
substituted with such dyes. Accordingly, we treated 10b with a
slight excess of the perylene boronate 16 in a Suzuki-type
coupling (Scheme 3). The chromatographically purified product
was confirmed by NMR and matrix-assisted laser desorption
ionization time-of-flight spectroscopy (MALDI-TOF) to be the
diadduct 17. Figure 7 shows the optical properties of 17
measured in THF. The absorption from the pentaphenylene
backbone is characterized by the maximum at 397 nm, and the
absorption from the dye, by the maximum at 520 nm. As
evidenced from Figure 7, there is some energy transfer in
solution from the pentaphenylene core to the dye as the relative
intensity of the emission maxima for the dye at 620 nm is greater
than the relative intensity of the dye’s absorption maximum.
This is in contrast to the behavior of the fluorene-perylene
copolymers mentioned above, where no energy transfer was seen
in solution,28 but this is explicable by the much lower amounts
of dye (only 1-5 mol %) present in those materials. As in the
aforementioned polymers, the solid-state PL shows that nearly
complete energy transfer occurs so that the emission is almost
exclusively from the dye. In addition to being a potential red-
emitting material for LEDs or polymer lasers, this material is a
suitable substrate for single-molecule spectroscopy experiments
to probe the energy transfer processes between the chromo-
phores.29 Such studies are now underway and will be reported
separately. By attaching different dyes to 10b, it should be
(28) Ego, C.; Marsitzky, D.; Becker, S.; Zhang, J.; Grimsdale, A. C.; Mu¨llen,
K.; MacKenzie, J. D.; Silva, C.; Friend, R. H. J. Am. Chem. Soc. 2003,
125, 437.
(29) Ja¨ckel, F.; De Feyter, S.; Hofkens, J.; Ko¨hn, F.; De Schryver, F. C.; Ego,
C.; Grimsdale, A.; Mu¨llen, K. Chem. Phys. Lett. 2002, 362, 534.
(27) Janietz, S.; Bradley, D. D. C.; Grell, M.; Geibler, C.; Inbasekaran, M.;
Woo, E. P. Appl. Phys. Lett. 1998, 73, 2453.
9
J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004 6991

A R T I C L E S
Jacob et al.
Scheme 3. Synthesis of Dye-Substituted Pentaphenylene
possible to tune the emission color from blue to red or even
into the near-infrared, and this is currently under investigation.
Electroluminescence Measurements on Polymer 11b.
LEDs were prepared with configuration indium tin oxide (ITO)/
poly(3,4-ethylenedioxythiophene) doped with poly(styrene sul-
fonate) (PEDOT:PSS)/11b/Ca/Al. As seen in Figure 8, the best
devices showed blue emission (CIE30 coordinates 0.17, 0.09)
with a luminance of over 200 cd/m2 that was stable during
several minutes of operation at 6-7 V (j ) 78 mA) in a
glovebox environment. As compared to devices fabricated from
polymers 1 or 2 under identical conditions, no oxidative
degradation of the polymer, i.e., the formation of keto-type
defects and a corresponding emission band around 530 nm,
could be observed. Best results were obtained from devices with
rather thin emissive layers of ca. 50 nm. Thicker emissive layers
not only produced devices with lower efficiencies but also
caused rather strong self-absorption, resulting in less pure blue
emission. However, these unoptimized single-layer devices show
high leaking currents in reverse bias and only moderate
efficiencies (0.25-0.5 cd/A), indicating that electron-transport-
ing layers are required to obtain maximum efficiency. Investiga-
tions into device optimization are underway and will be reported
separately.
Electroluminescence Optimized to the Sensitivity of the
Human Eye. Figure 9 shows a comparison of the typical
electroluminescence spectra obtained for polymers 1, 2, 3b, and
11b and a plot of the eye-sensitivity curve. As one can see, the
EL spectrum of 1 hardly overlaps at all with the region where
the eye is sensitive to light, meaning that for this polymer one
is generating light in OLEDs that the eye cannot see (such
polymers will thus show intrinsically low yields in terms of
Figure 7. Absorption and emission spectra of 17 in solution (THF).
Figure 8. Luminance (dashed line) and current density (solid line) from
ITO/PEDOT:PSS/11b/Ca/Al. (Inset) Electroluminescence spectra from the
same OLED.
Figure 9. Comparison of the typical electroluminescence spectra obtained
for polymers 1, 2, 3b, and 11b and a plot of the eye-sensitivity curve.
9
6992 J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004

New Route to Ladder-Type Pentaphenylenes
A R T I C L E S
thick films deposited by solution coating onto precleaned ITO as a
working electrode with an area of 0.2 cm2. After being coated, the
films were dried in a vacuum oven for 10 min. The measurements were
carried out in acetonitrile solution containing 0.1 M tetrabutylammo-
nium perchlorate as the supporting electrolyte, with Ag/AgCl as
reference electrode and a platinum wire as counterelectrode, respec-
tively, and an internal ferrocene/ferrocenium standard.
candelas per ampere). Compared to polymer 1, polymer 2 is
already better suited for the human eye. An analysis of the
sensitivity curve shows almost twice as high a sensitivity as
for the peak emission of polymer 1. The fact that both polymers
give rise to a very similar color impression to the human eye is
also well documented in the typical CIE color coordinates,30
where for polymer 1 one typically finds x ) 0.17 and y ) 0.14,
compared with x ) 0.18 and y ) 0.13 for polymer 2. On the
other hand, the EL spectrum of polymer 3b which has proven
to be rather stable against oxidative degradation, gives rise to a
blue-green emission that corresponds to CIE coordinates of x
) 0.20 and y ) 0.27, which is already too much in the green
spectral region to be utilized in a red-green-blue (RGB)
display.
The UV absorption measurements were performed on a Perkin-Elmer
Lambda 15 spectrophotometer and the PL measurements in solution
on a SPEX Fluorolog 2 type 212 steady-state fluorometer. Photolumi-
nescence and excitation spectra of the polymer films were recorded on
a Shimadzu RF5301 spectrofluorometer. For the photoinduced absorp-
tion measurements, the samples were mounted in a cryostat and cooled
to 80 K. A standard halogen lamp operated at 160 W was used as a
transmission source and an Ar+ laser operated in multilane UV mode
(3.53 eV, 3.41 eV) was used as excitation source.
As can be seen from Figure 9, polymer 11b, which has a
chemical structure intermediate between those of polymers 2
and 3b, has its maximum at a region where the human eye is 3
times as sensitive as for polymer 2 and 6 times as sensitive as
for polymer 1, while producing an emission with CIE coordi-
nates of x ) 0.18 and y ) 0.17 that are very close to those of
polymers 1 and 2. Thus polymer 11b not only has the advantage
of being chemically more stable against oxidative degradation
in the device but also of generating a sufficiently pure blue
emission spectrum optimized to the human eye.
The ITO/PEDOT-PSS/11b/Ca/Al OLEDs were fabricated by spin-
casting the filtered PEDOT-PSS suspension onto cleaned ITO substrates
and drying the PEDOT-PSS for 1 h in an argon atmosphere at 70 °C
and for 2 h in a vacuum at 120 °C, followed by spin-casting of the
toluene solutions (4 g/L) of polymer 11b. After the devices were dried
in a vacuum at 110 °C, the top electrodes were deposited by evaporation
at a base pressure of 2 × 10-6 mbar. Current-voltage characteristics
were recorded on a Keithley 236 SMU. The luminance was measured
by a Si photodiode in an integrating sphere, and the measured voltage
was scaled to the absolute luminance measured with a Minolta LS-
100 luminance meter at a single point in the upper part of the
represented characteristics. The EL spectra were recorded on an Oriel
Multispec CCD spectrometer and have been corrected accordingly.
Synthesis of the Diester 5. The boronate ester 8 (4.65 g, 9.0 mmol),
diester 9 (1.52 g, 4.39 mmol), and K2CO3 (1.21 g, 8.8 mmol) were
dissolved in THF (40 mL) and water (20 mL) in a 100 mL Schlenk
flask. The solution was purged with argon for 20 min, and then tetrakis-
(triphenylphosphine)palladium (152 mg, 0.03 equiv) was added and
the reaction was heated with stirring at 85 °C. The reaction was followed
by TLC and after 16 h was worked up. The cooled mixture was
extracted with diethyl ether, and the extract was washed with saturated
salt and then dried over MgSO4. The crude product so obtained was
purified by chromatography on silica with 0-5% ethyl acetate in hexane
as eluent. The product 5 was isolated as a thick viscous oil (3.94 g,
Conclusions
Ladder-type pentaphenylenes have been prepared by short
efficient syntheses via a pentaphenylene intermediate. Addition
of aryllithiums followed by ring closure produced monomers
for polymers that show bright blue emission with a very small
Stokes shift. Coupling of the monomer with a perylene dye
offers an efficient method to tune the emission color across the
visible spectrum. Treatment of the intermediate with acid gave
a diketone that shows good electron-accepting properties. In
this way both p- and n-type materials can be readily obtained
from a common intermediate. Single-layer OLEDs with 11b
showed stable pure-blue emission with brightnesses of over 200
cd/m2 at 7 V, with efficiencies that leave room for device
optimization. These initial results, however, clearly demonstrate
that polymer 11b has considerable potential as an emissive
material for stable pure-blue emitting devices due to its having
intrinsic advantages over other blue-emitting polymers in terms
of its emission matching to the sensitivity of the human eye,
being chemically more stable toward oxidative degradation and
having a low intrinsic structural defect concentration.
1
92%). H NMR: (CD2Cl2) δ 7.88 (s, 2H), 7.78 (m, 4H), 7.40-7.33
(m, 10H), 3.65 (s, 6H), 2.00 (t, 8H, J ) 8.5 Hz), 1.21-1.07 (m, 40H),
0.62-0.83 (m, 20H). 13C NMR: δ 169.26, 151.74, 151.62, 141.89,
141.56, 141.43, 139.54, 134.37, 132.44, 128.02, 127.82, 127.59, 123.84,
123.73, 120.55, 120.29, 55.96, 52.78, 41.16, 32.54, 30.81, 30.05, 29.99,
24.56, 23.35, 14.58. FDMS: m/z 971.3 Elemental analysis: calcd C
84.07, H 9.34; found C 84.12, H 9.32.
Synthesis of the Pentamers 6. (a) A solution of 4-octylbromoben-
zene (3.05 mL, 3.4 g, 12.6 mmol) in dry THF (40 mL) in a 250 mL
Schlenk flask, was cooled to -78 °C in an acetone/dry ice bath.
n-Butyllithium in hexane (8.0 mL, 1.6 M, 12.8 mmol) was then added
and the mixture was stirred for 20 min. Then a solution of the diester
5 (2.05 g, 2.1 mmol) in dry THF (40 mL) was added dropwise with
stirring and the solution was slowly allowed to warm to room
temperature. The mixture was stirred overnight and then quenched with
brine. The mixture was extracted with diethyl ether, and the extract
was washed with salt and dried over MgSO4. The crude product was
chromatographed on silica with 0-5% ethyl acetate in hexane as eluent
to give the diol as a thick viscous oil. (3.39 g, 96%). 1H NMR: δ 7.61
(m, 2H), 7.45 (d, 2H, J ) 7.6 Hz), 7.29 (m, 6H), 7.12 (m, 16H), 6.76
(m, 4H), 6.69 (s, 2H), 3.11 (s, 2H), 2.61 (t, 8H, J ) 7.6 Hz), 1.77-
0.45 (m, 128H). 13C NMR: δ 151.63, 151.23, 146.41, 144.19, 142.74,
141.71, 141.21, 141.05, 139.29, 136.29, 129.29, 128.76, 128.58, 127.93,
127.48, 124.32, 123.63, 120.35, 120.22, 83.91, 55.75, 40.55, 36.30,
32.72, 32.60, 32.45, 30.73, 30.30, 30.12, 30.09, 30.07, 30.04, 24.56,
23.46, 23.39, 14.65, 14.61. FDMS: m/z 1668.3.
Experimental Section
All solvents were purified and freshly distilled prior to use according
to literature procedures. 9,9-Dioctylfluorene-2-boronate ester 8, dimethyl
2,5-dibromoterephthalate 9, and 2,7-dibromo-9,9-dioctylfluorene were
synthesized according to literature procedures. 1H and 13C NMR spectra
were recorded on a Bruker DRX 250 (250 and 62.5 MHz respectively).
UV-Vis data were recorded on a Perkin-Elmer Lambda 9. Gel-
permeation chromatography (GPC) analysis was performed with PL
gel columns (103 and 104 Å pore widths) connected to a UV-vis
detector against poly-p-phenylene31 and polystyrene standards with
narrow weight distributions. Cyclic voltammetry was performed on an
EG&G Princeton Applied Research potentiostat, model 270, on 2 µm
(30) CIE Standard Color Matching Functions 1931 2° observer.
(31) Vanhee, S.; Rulkens, R.; Lehmann, U.; Rosenauer, C.; Schulze, M.; Ko¨hler,
W.; Wegner, G. Macromolecules 1996, 29, 5136.
9
J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004 6993

A R T I C L E S
Jacob et al.
With R′ ) t-Bu, the corresponding diol is isolated as a colorless
solid in 54% yield. H NMR: δ 7.64 (m, 2H), 7.47 (d, 2H, J ) 7.9
stirred for 4 h. The precipitated yellow solid was redissolved in THF
(40 mL) and added dropwise to methanol (200 mL). The resulting solid
was filtered off and subjected to Soxhlet extraction for 2 days in acetone.
The residue was then redissolved in THF and precipitated again from
methanol, filtered, washed with methanol, and dried. Isolated yield of
11b ) 508 mg (96%). Elemental analysis: calcd C 89.86, H 10.14;
found C 89.27, H 10.15. GPC analysis: Mn ) 1.36 × 105, Mw ) 2.98
1
Hz), 7.37-7.13 (m, 20H), 6.79 (d, 4H, J ) 7.9 Hz), 6.74 (s, 2H), 3.15
(s, 2H), 1.73-0.84 (m, 96H), 0.45 (m, 8H). 13C NMR: δ 151.58,
151.12, 150.90, 146.30, 144.07, 141.81, 141.23, 141.06, 139.05, 136.56,
129.49, 128.64, 127.92, 127.48, 125.48, 124.00, 123.64, 120.45, 120.35,
83.87, 55.72, 40.42, 35.17, 32.59, 31.95, 30.66, 30.01, 24.47, 23.44,
23.39, 14.63. FDMS: m/z ) 1442.4.
× 105, and D ) 2.19 (against PPP standard); Mn ) 2.64 × 105, Mw
7.37 × 105, and D ) 2.79 (against PS standard).
)
(b) Ring closure: The diol (3.30 g, 1.98 mmol) was dissolved in
dichloromethane (30 mL, used as purchased), and BF3 etherate (0.20
mL) was added with stirring at room temperature. The colorless solution
turned deep brown immediately upon addition and became light yellow
within minutes. After 10 min, methanol (50 mL) was added to the
solution, whereupon the pentamer 6 started to precipitate as a colorless
solid. The mixture was stirred for 12 h and then the solid was collected
by filtration, washed with methanol, and dried. The product was
redissolved in dichloromethane and precipitated again by addition of
For R′ ) t-Bu, starting with 300 mg of the monomer 10a, the
polymer 11a was isolated in 96% yield. Elemental analysis: calcd C
90.54, H 9.46; found C 89.95, H 9.41. GPC analysis: Mn ) 4.38 ×
104, Mw ) 1.43 × 105, and D ) 3.27 (against PPP standard); Mn )
7.05 × 104, Mw ) 3.28 × 105, and D ) 4.65 (against PS standard).
Synthesis of 7-Trimethylsilyl-9,9-di-n-octylfluorene-2-boronate
(13). n-Butyllithium in hexane (6.0 mL, 1.6 M, 9.6 mmol) was added
to a solution of 2,7-dibromo-9,9-di-n-octylfluorene (12) (5.00 g, 9.13
mmol) in dry THF (50 mL) in a Schlenk flask at -78 °C, and the
mixture was stirred for 20 min. Then chlorotrimethylsilane (1.5 mL,
1.29 g, 1.3 equiv) was added and the reaction was allowed to slowly
warm to room temperature. The mixture was stirred overnight and the
reaction was quenched with saturated brine. The product was extracted
into diethyl ether, and the extract was washed with brine and dried
over MgSO4. The crude product was chromatographed on silica with
hexane as eluent to give 2-bromo-7-trimethylsilyl-9,9-di-n-octylfluorene
as a colorless oil (3.50 g, 71%).
1
methanol. Isolated yield of 6b ) 3.06 g (95%). H NMR: δ 7.83 (s,
2H), 7.63 (m, 6H), 7.33-7.11 (m, 22H), 2.58 (t, 8H, J ) 7.6 Hz), 2.02
(t, 3H, J ) 7.9 Hz), 1.63-0.62 (m, 120H). 13C NMR: δ 152.80, 152.05,
151.88, 151.55, 144.35, 142.29, 141.83, 140.91, 140.87, 140.25, 129.11,
128.97, 127.63, 127.37, 123.65, 120.34, 118.16, 117.82, 115.35, 65.15,
55.61, 41.39, 36.28, 32.65, 32.55, 32.30, 30.82, 30.30, 30.22, 30.03,
30.01, 29.98, 24.67, 23.43, 23.34, 14.64, 14.60. FDMS: m/z 1633.2.
Elemental analysis: calcd C 89.75, H 10.25; found C 89.77, H 10.19.
For R′ ) t-Bu, yield of 6a ) 73% as a colorless solid. 1H NMR: δ
7.84 (s, 2H), 7.69 (d, 4H, J ) 8.4 Hz), 7.61 (m, 2H), 7.34-7.26 (m,
22H), 2.01 (t, 8H, J ) 8.2 Hz), 1.29-0.61 (m, 96H). FDMS: m/z
1407.3.
n-Butyllithium in hexane (4.4 mL, 1.6 M, mmol) was added to a
solution of 2-bromo-7-trimethylsilyl-9,9-di-n-octylfluorene (3.47 g, 6.41
mmol) in dry THF (75 mL) in a Schlenk flask at -78 °C, and the
mixture was stirred for 20 min. 2-Isopropoxy-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane (1.44 mL, 1.31 g, 7.04 mmol) was added and the reaction
was stirred for 24 h after warming to room temperature. The reaction
was then quenched with brine, and the product was extracted into diethyl
ether. The extract was washed with brine and dried over MgSO4. The
crude product was chromatographed on silica with 0-3% ethyl acetate
Synthesis of Monomers 10. The pentaphenylene 6b (0.80 g, 0.49
mmol) was added to carbon tetrachloride (30 mL) in a 100 mL round-
bottomed flask, followed by CuBr2 on alumina (2.0 g).23 The reaction
was heated under reflux with stirring and monitored by FDMS, which
showed nearly quantitative formation of the dibromide after 14 h. The
reaction mixture was concentrated, and the residue was loaded on silica
and chromatographed on silica with hexane as eluent. The product 10b
1
in hexane as eluent to give 13 as a colorless solid (2.67 g, 71%). H
1
was isolated as a light yellow solid (0.80 g, 91%). H NMR: δ 7.82
NMR (CD2Cl2): δ 7.73 (m, 4H), 7.51 (m, 2H), 1.99 (t, 4H, J ) 8.2
Hz), 1.36-0.54 (m, 42H), 0.31 (s, 9H). 13C NMR: δ 151.60, 151.34,
149.76, 145.29, 142.69, 141.07, 134.71, 133.01, 130.08, 129.03, 120.50,
120.25, 84.93, 56.22, 41.26, 33.01, 32.84, 31.11, 30.37, 25.98, 24.95,
23.85, 15.08, 0.01. FDMS: m/z 588.90.
(s, 2H), 7.66 (d, 4H, J ) 9.1 Hz), 7.42 (m, 6H), 7.24 (d, 8H, J ) 8.2
Hz), 7.12 (d, 8H, J ) 8.2 Hz), 2.58 (t, 8H, J ) 7.3 Hz), 2.00 (t, 8H,
J ) 7.6 Hz), 1.59 (m, 8H), 1.31-0.78 (m, 112H). 13C NMR: δ 154.19,
152.90, 152.30, 151.23, 144.17, 142.38, 140.97, 140.85, 140.76, 140.69,
130.52, 129.14, 128.95, 126.92, 121.79, 121.51, 118.29, 117.97, 115.44,
65.17, 55.99, 41.28, 36.26, 32.65, 32.53, 32.29, 30.73, 30.69, 30.29,
30.22, 30.00, 29.97, 24.62, 23.43, 23.35, 14.64, 14.60. FDMS: m/z
1780.50. Elemental analysis: calcd C 81.84, H 9.23; found C 81.99,
H 9.24.
For R′ ) t-Bu, isolated yield of 10a ) 56% as a light yellow solid.
1H NMR: δ 7.85 (s, 2H), 7.68 (d, 4H, J ) 11 Hz), 7.51-7.25 (m,
22H), 2.01 (t, 8H, J ) 6.6 Hz), 1.30-1.07 (m, 76H), 0.82-0.64 (m,
20H). 13C NMR: δ 154.21, 152.91, 152.30, 151.26, 150.36, 143.88,
140.94, 140.84, 140.77, 140.69, 130.53, 128.72, 126.92, 126.13, 121.83,
121.56, 118.35, 118.01, 115.54, 64.99, 56.02, 41.28, 35.08, 32.52, 31.86,
30.71, 29.95, 24.61, 24.55, 23.34, 14.59. FDMS: m/z 1565.90.
Elemental analysis: calcd C 81.30, H 8.50; found C 81.13, H 8.55.
Synthesis of Poly(pentaphenylene)s 11. Bis(cyclooctadiene)nickel
(215.0 mg, 2.4 equiv), cyclooctadiene (96 µL, 2.4 equiv), and 2,2′-
bipyridine (122.3 mg, 2.4 equiv) were dissolved in dry toluene (4 mL)
and dry N,N-dimethylformamide (4 mL) in a Schlenk flask within a
glovebox. The mixture was heated at 60 °C with stirring under argon
for 20 min to generate the catalyst, and then a solution of the dibromide
6b (584 mg, 0.33 mmol) in dry toluene (8 mL) was added. The reaction
was heated at 75 °C for 2 days, whereupon the solution became so
viscous that it ceased to stir. Then a mixture of toluene (4 mL) and
bromobenzene (0.10 mL) was added and the mixture was heated at 75
°C for an additional 12 h. The mixture was then poured into a mixture
of methanol and concentrated hydrochloric acid (1:1, 300 mL) and
Synthesis of Dibromopentaphenylene 14. The boronate ester 13
(2.45 g, 4.16 mmol), the diester 9 (697 mg, 1.98 mmol), and K2CO3
(821 mg) were dissolved in a mixture of THF (20 mL) and H2O (10
mL) in a Schlenk flask, and the mixture was purged with argon for 20
min. To this was added tetrakis(triphenylphosphine)palladium (114 mg,
0.05 equiv), and the mixture was heated at 85 °C for 20 h. The cooled
mixture was extracted with diethyl ether, and the extract was washed
with brine and dried over MgSO4. The residue was chromatographed
on silica with 0-3% ethyl acetate in hexane to give the crude
disilylpentaphenylene (1.89 g, 86%).
In a Schlenk flask, the diester (1.89 g, 1.69 mmol) was added to
dry THF (34 mL) along with anhydrous sodium acetate (278 mg, 2
equiv) and cooled to 0 °C. Bromine (0.37 mL, 1.15 g, 4.2 equiv) was
added and the mixture was stirred for 20 min. The reaction was
quenched by addition of triethylamine (1.89 mL, 8 equiv), followed
by an excess of aqueous Na2SO3. The product was extracted into diethyl
ether, and the extract was washed with sodium sulfite solution and
dried over MgSO4. The residue was recrystallized from hot hexane to
1
give the dibromide 14 as shiny colorless crystals (1.37 g, 72%). H
NMR (CD2Cl2): δ 7.87 (s, 2H), 7.77 (d, 2H, J ) 7.9 Hz), 7.64 (d, 2H,
J ) 7.9 Hz), 7.53-7.31 (m, 8H), 3.64 (s, 6H), 1.99 (t, 8H, J ) 8.8
Hz), 1.27-0.62 (m, 60H). 13C NMR: δ 169.06, 154.06, 151.27, 141.89,
140.58, 140.49, 140.05, 134.32, 132.51, 130.76, 128.06, 127.04, 123.87,
122.01, 121.96, 120.46, 56.36, 52.82, 41.08, 32.54, 30.74, 30.03, 29.99,
9
6994 J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004

New Route to Ladder-Type Pentaphenylenes
A R T I C L E S
24.52, 23.36, 14.59. FDMS: m/z 1129.30. Elemental analysis: calcd
C 72.33, H 7.85; found C 72.75, H 7.91.
Synthesis of Dye End-Capped Pentaphenylene 17. The dibromide
10b (143 mg, 0.08 mmol), peryleneboronate ester 16 (100 mg, 0.16
mmol), K2CO3 (22 mg, 0.16 mmol), tetrahydrofuran (6 mL), and water
(3 mL) were mixed together in a Schlenk flask which was purged with
argon for 15 min. Tetrakis(triphenylphosphine)palladium (2.8 mg, 0.002
mmol) was added and the mixture was heated at 80 °C for 16 h. The
product was extracted into dichloromethane, and the extract was washed
with brine and dried over MgSO4. The crude product was chromato-
graphed on silica with 0-12% ethyl acetate in hexane as eluent to
Synthesis of Dibromopentaphenylenedione 7. The dibromide 14
(1.37 g) was added to 40 mL of 80% H2SO4 (made from 4 mL of H2O
and 36 mL of concentrated H2SO4) and the mixture was heated with
stirring at 165 °C for 3 h, during which time the white solid turned
brown. The acid was decanted off and the product was extracted into
dichloromethane, and the extract was washed with sodium hydrogen
carbonate solution and dried over MgSO4. The crude product was
purified by chromatography on silica with 0-3% ethyl acetate in hexane
as eluent to give the diketone 7 as a gray solid (0.88 g, 69%). 1H NMR
(CD2Cl2): δ 7.97 (s, 2H), 7.86 (s, 2H), 7.65-7.49 (m, 8H), 2.03 (t,
8H, J ) 8.5 Hz), 1.25-0.62 (m, 60H). 13C NMR: δ 192.91, 159.79,
153.77, 146.79, 144.19, 142.53, 140.81, 139.69, 134.70, 131.11, 127.10,
122.66, 122.21, 116.50, 116.38, 116.27, 56.96, 40.88, 32.53, 30.66,
29.97, 29.95, 24.60, 23.37, 14.60. Elemental analysis: calcd C 74.42,
H 7.57; found C 74.63, H 7.60.
1
give the diadduct 17 as a red solid (36 mg, 17%). H NMR: 8.67-
8.52 (m, 12H), 8.10 (d, 2H, J ) 8.5 Hz), 7.89-7.50 (m, 18H), 7.49-
7.15 (m, 20H), 2.78 (septet, 4H, J ) 6.6 Hz), 2.61 (t, 8H, J ) 7.9 Hz),
2.11 (t, 8H, J ) 7.3 Hz), 1.63 (m, 8H), 1.40-1.14 (m, 104H), 0.90-
0.78 (m, 32H). 13C NMR: 164.86, 152.98, 152.37, 152.27, 152.02,
146.93, 144.91, 144.30, 142.40, 141.87, 141.34, 140.95, 140.70, 139.10,
138.62, 138.48, 133.54, 132.67, 132.62, 132.45, 131.41, 130.09, 130.04,
129.55, 129.23, 129.18, 129.11, 129.02, 127.73, 127.70, 125.57, 124.89,
124.77, 124.55, 121.69, 121.53, 121.18, 120.90, 120.55, 118.34, 118.26,
118.18, 115.55, 115.49, 65.24, 55.88, 53.54, 41.36, 36.30, 32.66, 32.57,
32.31, 30.83, 30.30, 30.23, 30.08, 30.00, 29.86, 24.87, 24.49, 23.44,
23.37, 14.64, 14.62. MALDI-TOF: m/z 2594.0.
Synthesis of 15. The diketone 7 (127.0 mg, 0.12 mmol), phenyl-
boronic acid (36.5 mg, 0.30 mmol), K2CO3 (33 mg, 0.24 mmol),
tetrahydrofuran (6.0 mL), and water (3.0 mL) were mixed in a Schlenk
flask which was purged with argon for 15 min. Tetrakis(triphenylphos-
phine)palladium (5.5 mg, 5.0 µmmol) was added and the mixture was
heated with stirring for 14 h at 85 °C. The product was extracted into
ether, and the extract was washed with saturated brine and dried over
MgSO4. The residue was chromatographed on silica with 0-4% ethyl
acetate in hexane as eluent to recover 15 as a gray solid (91.0 mg,
Acknowledgment. This work was supported by the Bundes-
ministerium fu¨r Bildung und Forschung (Projects 13N8165
OLAS and 13N8215 OLED) and by DuPont Displays. J.J.
acknowledges the Alexander von Humboldt Stiftung for the
grant of a research fellowship. The CDL-AFM is a key member
of the long-term AT&S research activities.
1
72%). H NMR: δ 8.02 (s, 2H), 7.86 (s, 2H), 7.83 (d, 2H, J ) 7.9
Hz), 7.71-7.62 (m, 10H), 7.49 (t, 4H, J ) 6.9 Hz), 7.37 (t, 2H, J )
7.2 Hz), 2.11 (t, 8H, J ) 8.4 Hz), 1.18-1.07 (m, 40H), 0.80-0.71 (m,
20H). FDMS: m/z 1059.0.
JA0398823
9
J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004 6995