Counting chromophores in conjugated polymers

Article abstract of DOI:10.1002/anie.200461784

The emitting species in conjugated polymers have been studied by single-molecule spectroscopy (see picture). The number of emissive centers (chromophores) varies as a function of molecular weight and chain length. A large ladder-type undecaphenylene was used as a molecular ruler to estimate the size of chromophores in the polymer and to enable a direct assessment of the polymeric conjugation length.

Full text of DOI:10.1002/anie.200461784

Communications
Conjugated Polymers
Counting Chromophores in Conjugated
Polymers**
Florian Schindler, Josemon Jacob,
Andrew C. Grimsdale, Ullrich Scherf, Klaus Mꢀllen,
John M. Lupton,* and Jochen Feldmann
The development of novel materials for organic optoelec-
tronics based on conjugated polymers (CPs) has been
paralleled by the quest to understand the nature of the
fundamental emitting species.[1] CPs can have molecular
weights of hundreds of thousands, comprising thousands of
repeat units, yet the delocalization of p electrons is thought to
be limited by structural and chemical defects on the chain and
leads to the formation of “isolated” chromophore units on the
CP. Although this idea is generally upheld, there is little direct
evidence for this model; the evidence reported until now is
derived mainly from site-selective fluorescence investiga-
tions.[2] The size of the emitter has also been widely debated in
the literature as the size defines the key difference between
polymeric and small-molecule organic semiconductors.[1–6]
A
simple way to assess the size of the chromophore is to study
model oligomeric compounds and establish the critical
oligomer length for which absorption and emission spectra
correspond to the polymeric analogue.[1,4] However, electron–
hole correlations are very strong in conjugated systems, thus
this approach only provides insight into the delocalization or
coherence length of the exciton but not the so-called diagonal
length; that is, the actual length of the conjugated segment
over which the electron–hole pair may move.[6] These two
physical quantities can be very different, such as in the case of
defect-free polydiacetylene.[7]
Single-molecule (SM) detection and spectroscopy (SMS)
are powerful tools to help to construct an understanding of
the basic electronic properties of molecules.[8] SMS has
[*] F. Schindler, Dr. J. M. Lupton, Prof. Dr. J. Feldmann
Photonics and Optoelectronics Group
Physics Department and CeNS
Ludwig-Maximilians-Universitꢀt, 80799 Munich (Germany)
Fax: (+49)89-2180-3441
E-mail: john.lupton@physik.uni-muenchen.de
Dr. J. Jacob, Dr. A. C. Grimsdale, Prof. Dr. K. Mꢁllen
Max Planck Institute for Polymer Research
Ackermannweg 10, 55128 Mainz (Germany)
Prof. Dr. U. Scherf
FB Chemie
Universitꢀt Wuppertal
Gauss-Str. 20, 42097 Wuppertal (Germany)
[**] The authors are grateful to W. Stadler, A. Helfrich, J. Schnee, and E.
Preis for technical assistance, the Deutsche Forschungsgemein-
schaft for financial support (grant nos. SFB 486 and SFB 625), as
well as the Schwerpunktprogramm organische Feldeffekttransisto-
ren, the Gottfried Wilhelm Leibniz award, and the Bundesministe-
rium fꢁr Bildung und Forschung (project 13N8615 OLAS) for
funding.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
1520
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200461784
Angew. Chem. Int. Ed. 2005, 44, 1520 –1525

Angewandte
Chemie
yielded insight into the nature of emission in a number of
well-defined multichromophoric assemblies[9–11] as well as
demonstrating that the emission from CPs can occur from
trap states which can be reversibly quenched.[12–17] We
recently showed that SMS can provide unique information
on the presence of chromophores in CPs and may be used to
investigate intramolecular interchromophoric couplings.[18–21]
Single chromophores exhibit universal photophysical proper-
ties independent of the details of the materials chemistry, thus
enabling the derivation of universal structure–property rela-
tionships.[21] Herein we address the fundamental issue of how
the number of chromophores varies with molecular size (i.e.
the number of repeat units) and present the first study by
high-resolution SMS on model oligomeric compounds. The
complexity of the SM emission spectra varies with chain
length and results from the contribution of individual spectral
features of single chromophores. Comparison of these data
with those of oligomeric compounds allows a direct differ-
entiation between the single-charge carrier and exciton
confinement lengths. Most importantly, we demonstrate that
a single CP molecule can support multichromophoric emis-
sion, which is not trap-state limited.
prepared by a modification of existing routes to pentapheny-
lenes.[22] Starting with 2,5-dibromodimethylterephthalate 4,
two sequential Suzuki couplings generated a pentaphenylene
diester bearing a trimethylsilyl group at one end. This group
was converted through the bromide to a boronic ester group
to produce the end-functionalized pentamer intermediate 7 in
an overall yield of 14% from 3. The corresponding deboro-
nated compound (arising from addition of two equivalents of
3 to 4) was present as a major ( ꢀ 33mol%) contaminant in all
the earlier steps but could be removed by chromatography
only at this stage. The problem in removing this contami-
nation together with the low overall yield of 7 illustrate the
difficulties inherent in the synthesis of defined ladder-type
oligomers. Another Suzuki coupling of 7 with 4 gave the
undecaphenylene hexaester 8 (68%), which upon addition of
excess aryllithium followed by Lewis acid catalyzed Friedel–
Crafts alkylation was converted into the undecamer 2 (56%
yield from 8, 3% overall yield based on 3). The oligomer 2
was characterized by microanalysis, NMR, UV/Vis absorb-
ance, and photoluminescence (PL) spectroscopy, and mass
spectrometry. Full details of the synthesis and character-
ization of 2 are given in the Supporting Information.
As a suitable polymer we chose methyl-substituted
poly(para-phenylene) (MeLPPP, 1), and as a model oligomer
we chose a ladder-type undecaphenylene 2, which is one of
the longest poly(para-phenylene) oligomers reported thus far
and which was specifically synthesized to clarify the nature of
emission in CPs. The synthetic pathway to undecaphenylene 2
is shown in Scheme 1. The key intermediate in the prepara-
tion of 2 is a singly end-functionalized pentaphenylene
We studied the PL spectra of single molecules of the rigid
rod ladder-type CP 1 and the undecamer model compound 2
dispersed in a polystyrene matrix. Figure 1 shows spectra
recorded both in solution at room temperature and as single
molecules at 5 K. The spectra of the undecamer are blue-
shifted by 8 nm with respect to the CP, but are otherwise
rather similar with maxima at around 450 nm. The spectra of
the SM are substantially narrower (1–2-nm width) than those
Scheme 1. Synthesis of the ladder-type undecamer 2.
Angew. Chem. Int. Ed. 2005, 44, 1520 –1525
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1521

Communications
Figure 1. Single-molecule (SM) photoluminescence spectra at 5 K of
a) the conjugated polymer (CP) MeLPPP (1) and b) the model
oligomer 2. Dotted lines (g) show PL spectra of the ensembles in
solution at room temperature.
of the solution ensemble and reveal a distinct splitting of the
vibronic 0–1 transition around 480 nm. SMS shows a signifi-
cantly higher spectral resolution which allowed us to pinpoint
the nature of emission on the polymer chain. To do this, the
chain length of the CP was varied from approximately 62 to
around 165 phenylene units (Mr ꢀ 25 kDa to Mr ꢀ 67 kDa;
polydispersity ratio (PD) = 1.8–4.7), with the larger CPs
displaying a somewhat higher polydispersity ratio.
The different single CPs were compared directly with the
undecamer model compound. Figure 2a shows the 0–0
transition of a typical PL spectrum of a single oligomer. In
comparison, a characteristic PL spectrum of the short-chain
CP is presented in Figure 2b. One clear peak is identified in
both cases. The maximum of the short-chain CP is red-shifted
by a few nm and is somewhat narrower ( ꢀ 1 nm) relative to
that of the oligomer, but otherwise they are similar in shape.
The ensemble spectrum of the CP is found to be virtually
independent of molecular weight (not shown). The SM
spectrum of the long-chain CP in Figure 2c, however, is
more complex. We identify 5 distinct peaks, all of which have
the same shape and width as the single peak in panel b, and
for better comparison the PL peak of the short-chain CP is
superimposed on each of the 5 lines (dotted curves). We
conclude that the emitting species is similar in all cases and is
almost independent of chain length, which simply controls the
number of emitting units. The histograms in Figure 2d show a
statistical analysis of peak positions and line widths for a total
of 161 SMs that exhibit resolvable peaks in the PL spectra.
The distribution in peak energies is closely matched by the
ensemble spectra recorded at 5 K (solid lines). The spectra of
the single undecamer are on average twice as broad as the CP
spectra and are blue-shifted by 10 nm. There is a certain
overlap between the histograms of the CPs and the oligomers,
so that a few spectra of CPs are indistinguishable from those
of the oligomers. This suggests that the emitting species can
indeed be identical in the undecamer and the CP; that is, a
subunit on the chain which consists of 11 benzene rings makes
up the conjugated segments. It should be stressed that
although the line widths of the spectra of the SMs are
Figure 2. Single-molecule PL emission spectra at 5 K of the the model
oligomer and the conjugated polymer MeLPPP: a) the undecamer, b) a
short-chain CP (Mr ꢀ25 kDa), and c) a long-chain CP (Mr ꢀ67 kDa).
Each of the individual peaks in the emission from the long-chain CP is
identical in shape to the single peak in the emission from the short-
chain CP (this single peak is superimposed for each peak in part c;
g). The sketches indicate the difference between exciton size (black)
and conjugation length (gray). d) The distribution of peak energies (l)
and line widths (w) for the short-chain CP and the model oligomer,
with ensemble spectra overlaid (c; n=occurrence).
broader for the oligomer than those for the CP, the
distribution in peak positions is considerably narrower as a
direct consequence of the monodispersity of the oligomer.
1522
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 1520 –1525

Angewandte
Chemie
Note that in contrast the spectra of the ensemble at room
temperature are equally broad and thus do not provide a
measure of disorder.
Evidently, segments as short as 11 rings can constitute the
emitting species in CPs. The chain may be over 13 times as
long as this, so that a single chain should support multiple
chromophores emitting simultaneously. The number of units
should therefore depend on the chain length. Figure 3 shows
of the excitation (black). This effect has been studied in detail
by using quantum-chemical models,[6] but has so far been hard
to assess experimentally. Our data suggest that the absorbing
unit should be larger than the emitting unit. Above a length of
11–12 benzene rings, the Coulombic interactions as well as
phonon-mediated self-trapping are so strong that the exciton
size remains independent of the length of the conjugated
segment. The conjugation length may therefore substantially
exceed the exciton size. For a chain length of 165 rings, we
find 2.8 lines on average which provides an upper estimate of
the conjugation length of 59, approximately five times the size
of the exciton. Even if we account for the possibility of the
chain containing structural defects that lead to a branching of
the rigid rod structure and to different polarizations of the
individual conjugated segments,[18] which would not all be
excited by the linearly polarized laser light used, at most 50%
of the conjugated units on the chain would escape excitation.
The effective length of the optically active segment on the
chain is therefore at least twice the size of the exciton. As the
number of chromophores varies approximately linearly with
chain length, we propose that the exciton localization length
in this class of materials is generally at least twice the exciton
coherence size of roughly 12 phenylene units.
This conclusion is compatible with previous observations
in highly ordered and virtually defect-free polydiacetylenes.[7]
The huge optical non-linearities, which can be well-described
by a phase–space-filling model of one-dimensional excitons[7c]
as well as by ultrafast intramolecular migration of excitation
energy in single chains,[7b] are signatures of extended p con-
jugation with limited exciton size. Evidence for conjugated
units that are larger in absorption than in emission is also seen
in more-disordered materials for which the red shift in
fluorescence saturates considerably sooner with increasing
chain length than the red shift in absorption.[4g] This effect,
which is in part related to structural relaxation (although this
is very weak indeed in the ladder-type polymers[19]), has also
been investigated theoretically,[24,25] as a precise knowledge of
the size of the emitting and absorbing unit is imperative for
understanding and modeling intramolecular energy trans-
fer.[25]
The conclusion of extended conjugations naturally raises
the question of the origin of static disorder, which gives rise to
a distribution in chromophore energies and the associated
inhomogeneous broadening witnessed in Figure 2d. The
common notion is that topological defects lead to a confine-
ment of the excitation and thus to a modification in transition
energy. Our data suggest that the origin of disorder is more
subtle, although clear scissions of the p system do occur.
Instead, we propose that the dielectric environment of the
individual chromophores, that is, both the surrounding matrix
and the remaining chain, leads to an effective potential-
energy landscape, which is different for each chromophore
and which slightly modifies the transition energies between
different chromophores. This situation is conceptually similar
to the description of energetic disorder in single dye
molecules for which a comparably large degree of disorder
is observed despite the absence of a distribution in conjuga-
tion lengths.[8,9] Further evidence for this picture comes from
the fact that the line widths in the PL spectra of the single-
&
Figure 3. : Variation of the average number of peaks (chromophores)
in the PL spectra (nchrom) with chain length (lchain, number of benzene
rings). : The average intensity per peak for a total of 108 molecules
*
(I/nchrom).
the variation in the number of emission lines, that is,
chromophores, with chain length which were extracted from
a total of 200 SMs. Note that the data point at a chain length
of 138 rings was determined for a phenyl-substituted rather
than a methyl-substituted LPPP, which otherwise has identical
fluorescence properties.[23] A monotononic increase in the
number of units was observed which provides clear evidence
for a disruption of the p-electron system along the chain and
the formation of more-or-less isolated conjugated subunits.
The average PL intensity per chromophore line (circles) is
also indicated. This is virtually independent of chain length,
but increases almost tenfold upon going from the oligomer to
the CP. The PL intensity is both a measure of the absorption
and the emission efficiency. As the fluorescence yields of the
ensemble are comparable for the two materials, this differ-
ence in intensity can only be an effect of a difference in
absorption. The emitting unit on the CP must be comparable
to the undecamer in size, but the remainder of the conjugated
segment apparently leads to a substantial increase in oscil-
lator strength and absorption.
The effective conjugation length for absorption thus
increases with chain length but not the effective conjugation
length for emission. This effect may in part be explicable by
efficient light harvesting on the CP chain by ultrafast energy
transfer between adjacent chromophore units, which we were
recently able to identify by studying the polarization of
absorption and emission of single chains.[18] However, the
conjugation length (marked in gray in the sketch in
Figure 2)—the distance over which an excitation may move
coherently—may be substantially larger than the actual size
Angew. Chem. Int. Ed. 2005, 44, 1520 –1525
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1523

Communications
molecule oligomer are substantially greater than those of the
Time-resolved measurements pinpoint a substantial dif-
ference in the fluorescence from the oligomer and the CP.
Figure 4 inset shows three PL lifetime measurements
recorded for 2 and two CPs (with Mr ꢀ 25 kDa and Mr
ꢀ 67 kDa). The oligomer has a PL lifetime of 490 ps, whereas
this is reduced significantly to 360 ps in the short-chain CP.
Increasing the chain length by a factor of 2.5 leads to a 10%
decrease in the lifetime to 330 ps. Although the size of the p-
conjugated segment occupied by the exciton in the CP must
be of the order of the dimensions of the oligomer (i.e. 11 to 12
phenylene units) as the transition energies of 1 and 2 are
virtually equal, the remainder of the CP chain clearly affects
spontaneous emission. This is in part caused by highly
efficient intramolecular energy transfer[20] and the presence
of singlet scavenging defects, which have been shown to lead
to a chain-length-dependent decrease in the lifetime in
polyfluorenes.[26] However, the change in lifetime is much
smaller upon increasing the length of the CP than in going
from the oligomer to the CP. The conjugation length of the CP
must substantially exceed the undecamer size, which in turn is
a measure of the actual size of the excitation rather than of the
conjugated length. Whereas the excited state on the oligomer
has the nature of a molecular excitation, the emission from
the CP has excitonic character[1] as it arises from coherently
coupled conjugated segments extending further than the
effective spatial delocalization of the excited state. A conse-
quence of this is that vibrational coupling is weaker in the CP
than in the oligomer, which is clearly seen in the reduced
intensity of the 0–1 transition. Whereas molecular excitations
display very strong vibrational coupling, excitations in coher-
ently delocalized systems such as J-aggregates or highly
conjugated polydiacetylenes couple only weakly to vibra-
tions.[7] Furthermore, extension of the p-electron system also
leads to an increase in oscillator strength and consequently to a
reduction in fluorescence lifetime, in agreement with the
substantial acceleration in the emission rate observed in the
CP. These ensemble observations are fully consistent with our
results on counting the number of chromophores in single CP
chains and relating this to the dimensions of oligomers.
CP. As the undecamer is just of the length where the exciton is
still slightly confined by the limited p-electron delocalization,
a conclusion is that confinement of the exciton wavefunction
by complete disruption of the p system leads to line broad-
ening in similar fashion to J-aggregates, for example.[1] The
distribution in emission wavelengths in the CP is therefore
most likely not a direct consequence of the distribution in
conjugation lengths, which should lead to substantial line
broadening.
Finally, to highlight the relevance of our conclusions to
bulk materials for devices, we demonstrate that the ensemble
photophysical properties of 1 and 2 are also consistent with
the picture of polymeric chromophores substantially exceed-
ing the size of oligomers. Figure 4a shows a direct comparison
Figure 4. a) Ensemble fluorescence spectra of the undecamer (c)
and polymer (g) in dilute solution in toluene. The spectrum of the
oligomer is shown as recorded as well as shifted by 8 nm to overlap
that of the polymer (a) for better comparison. Inset: The PL decay
*
of the oligomer (&) and two conjugated polymers (Mr ꢀ25 kDa ( );
!
Mr ꢀ67 kDa ( )). b) Fluorescence spectra of the oligomer (c) and
the polymer (g) in the solid state, again with the spectrum of the
oligomer shifted and superimposed (a).
In summary, we have correlated the number of lines
observed in the fluorescence spectra from single CP mole-
cules directly with the average chain length. Our data suggest
that the origin of disorder in CPs is not caused by a statistical
distribution of defects confining the exciton but rather the
polarizing influence of the local environment. Understanding
the nature of the chromophores, the origin of disorder, and in
particular the interactions between chromophores on a chain
is absolutely vital to optimizing both electroluminescent and
photovoltaic CP devices. Our results provide strong evidence
that chromophoric emission is intrinsic to the nature of
extended p-electron systems and not directly controlled by
defects, which would be detrimental to device operation.
of the PL spectra of 2 and the CP ( ꢀ 25 kDa) in solution at
room temperature. The spectra are very similar in shape. The
PL spectrum of the oligomer is also shown red-shifted by 8 nm
and superimposed (dashed line) on that of the CP (dotted
line) for comparison. The spectral width of the monodisperse
oligomer and the polydisperse CP are almost identical, in
agreement with our conclusion that the CP spectrum is not
primarily broadened owing to a distribution of different
conjugation lengths. The CP spectrum does exhibit a slight
broadening to shorter wavelengths relative to the oligomer
which may indeed be an indication of the presence of
conjugated segments with effective lengths equal to or shorter
than the oligomer. The similarity in the PL spectra of the CP
and the oligomer is even more striking in the case of spin-
coated bulk films, which are compared in Figure 4b. Besides
the reduced intensity of the vibronic sideband in the case of 1
and the blue shift of 10 nm, the spectra are identical to within
1 nm, thus suggesting that 2 may indeed be an excellent model
for 1 in the bulk phase.
Received: August 25, 2004
Revised: November 3, 2004
Published online: January 31, 2005
Keywords: chromophores · luminescence · polymers ·
.
single-molecule studies
1524
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 1520 –1525

Angewandte
Chemie
[1] M. Pope, C. E. Swenberg, Electronic Processes in Organic
Crystals and Polymers, 2nd ed., Oxford University Press,
Oxford, 1999.
[2] H. Bꢀssler, B. Schweitzer, Acc. Chem. Res. 1999, 32, 173.
[3] Y. V. Romanosvkii, H. Bꢀssler, U. Scherf, Chem. Phys. 2002, 276, 321.
[4] a) J. Grimme, M. Kreyenschmidt, F. Uckert, K. Mꢁllen, U.
Scherf, Adv. Mater. 1995, 7, 292; b) H. S. Woo, O. Lhost, S. C.
Graham, D. D. C. Bradley, R. H. Friend, C. Quattrocchi, J.-L.
Brꢂdas, R. Schenk, K. Mꢁllen, Synth. Met. 1993, 59, 13; c) H.
Meier, U. Stamach, H. Kolshorn, Acta Polym. 1997, 48, 379; d) T.
Kirschbaum, P. Bꢀuerle, Synth. Met. 2001, 119, 127; e) J. J.
Apperloo, L. Groenendaal, H. Verheyen, M. Jayakannan,
R. A. J. Janssen, A. Dkhissi, D. Beljonne, R. Lazzaroni, J.-L.
Brꢂdas, Chem. Eur. J. 2002, 8, 2384; f) F. Wudl, S. P. Bitler, J. Am.
Chem. Soc. 1986, 108, 4685; g) G. Klaerner, R. D. Miller,
Macromolecules 1998, 31, 2007.
[5] G. H. Gelinck, J. J. Piet, B. R. Wegewijs, K. Mꢁllen, J. Wildeman,
G. Hadziioannou, J. M. Warman, Phys. Rev. B 2000, 62, 1489.
[6] S. Mukamel, S. Tretiak, T. Wagersreiter, V. Chernyak, Science
1997, 277, 781.
[7] a) R. Lꢂcuiller, J. Berrꢂhar, J. D. Ganiꢃre, C. Lapersonne-Meyer,
P. Lavallard, M. Schott, Phys. Rev. B 2002, 66, 125205; b) T.
Guillet, J. Berrꢂhar, R. Grousson, J. Kovensky, C. Lapersonne-
Meyer, M. Schott, V. Voliotis, Phys. Rev. Lett. 2001, 87, 087401;
c) B. I. Greene, J. Orenstein, R. R. Millard, L. R. Williams, Phys.
Rev. Lett. 1987, 58, 2750.
[8] W. E. Moerner, M. Orrit, Science 1999, 283, 1670.
[9] T. Christ, F. Kulzer, T. Weil, K. Mꢁllen, T. Baschꢂ, Chem. Phys.
Lett. 2003, 372, 878.
[10] P. Tinnefeld, K. D. Weston, T. Vosch, M. Cotlet, T. Weil, J.
Hofkens, K. Mꢁllen, F. C. De Schryver, M. Sauer, J. Am. Chem.
Soc. 2002, 124, 14310.
[11] T. Gensch, J. Hofkens, A. Heirmann, K. Tsuda, W. Verheijen, T.
Vosch, T. Christ, T. Baschꢂ, K. Mꢁllen, F. C. De Schryver, Angew.
Chem. 1999, 111, 3970; Angew. Chem. Int. Ed. 1999, 38, 3752.
[12] D. A. Vanden Bout, W. T. Yip, D. H. Hu, D. K. Fu, T. M. Swager,
P. F. Barbara, Science 1997, 277, 1074.
[13] T. Huser, M. Yan, L. J. Rothberg, Proc. Natl. Acad. Sci. USA
2000, 97, 11187.
[14] J. Yu, D. Hu, P. F. Barbara, Science 2000, 289, 1327.
[15] C. W. Hollars, S. M. Lane, T. Huser, Chem. Phys. Lett. 2003, 370,
393.
[16] J. D. White, J. H. Hsu, S. C. Yang, W. S. Fann, G. Y. Pern, S. A.
Chen, J. Chem. Phys. 2001, 114, 3848.
[17] W.-T. Yip, D. Hu, J. Yu, D. A. Vanden Bout, P. F. Barbara, J.
Phys. Chem. A 1998, 102, 7564.
[18] J. G. Mꢁller, J. M. Lupton, J. Feldmann, U. Lemmer, U. Scherf,
Appl. Phys. Lett. 2004, 84, 1183.
[19] J. G. Mꢁller, M. Anni, U. Scherf, J. M. Lupton, J. Feldmann,
Phys. Rev. B 2004, 70, 035205.
[20] J. G. Mꢁller, U. Lemmer, G. Raschke, M. Anni, U. Scherf, J. M.
Lupton, J. Feldmann, Phys. Rev. Lett. 2003, 91, 267403.
[21] F. Schindler, J. M. Lupton, J. Feldmann, U. Scherf, Proc. Natl.
Acad. Sci. USA 2004, 101, 14695.
[22] a) C. Xia, R. C. Advincula, Macromolecules 2001, 34, 6922; b) J.
Jacob, S. Sax, T. Piok, E. J. W. List, A. C. Grimsdale, K. Mꢁllen,
J. Am. Chem. Soc. 2004, 126, 6987.
[23] J. M. Lupton, A. Pogantsch, T. Piok, E. J. W. List, S. Patil, U.
Scherf, Phys. Rev. Lett. 2002, 89, 167401.
[24] S. Tretiak, A. Saxena, R. L. Martin, A. R. Bishop, Phys. Rev.
Lett. 2002, 89, 097402.
[25] D. Beljonne, G. Pourtois, C. Silva, E. Hennebicq, L. M. Herz,
R. H. Friend, G. D. Scholes, S. Setayesh, K. Mꢁllen, J. L. Brꢂdas,
Proc. Natl. Acad. Sci. USA 2002, 99, 10982.
[26] J. M. Lupton, M. R. Craig, E. W. Meijer, Appl. Phys. Lett. 2002,
80, 4489.
Angew. Chem. Int. Ed. 2005, 44, 1520 –1525
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1525