Synthesis and characterization of novel para- and meta-phenylenevinylene derivatives: Fine tuning of the electronic and optical properties of conjugated materials

Article abstract of DOI:10.1021/jp025748d

We report the synthesis of novel phenylenevinylene derivatives that allow for the introduction of meta versus para connections on the phenylene rings, as well as the incorporation of nitrogen atoms within the conjugated backbone and the attachment of elec

Full text of DOI:10.1021/jp025748d

6442
J. Phys. Chem. B 2002, 106, 6442-6450
Synthesis and Characterization of Novel para- and meta-Phenylenevinylene Derivatives:
Fine Tuning of the Electronic and Optical Properties of Conjugated Materials
L. Pascal,† J. J. Vanden Eynde,† Y. Van Haverbeke,*,† P. Dubois,*,‡ A. Michel,§ U. Rant,|
E. Zojer,*,| G. Leising,*, L. O. Van Dorn,# N. E. Gruhn,# J. Cornil,#,3 and J. L. Bre´das*,#,3
De´partement de Chimie Organique, UniVersite´ de Mons-Hainaut, Place du Parc 20, B-7000 Mons, Belgium,
SerVice des Mate´riaux Polyme`res et Composites, UniVersite´ de Mons-Hainaut, Place du Parc 20,
B-7000 Mons, Belgium, De´partement de Chimie Biologique, UniVersite´ de Mons-Hainaut, Place du Parc 20,
B-7000 Mons, Belgium, Institut fu¨r Festko¨rperphysik, Technische UniVersita¨t Graz, Petergasse 16,
A-8010 Graz, Austria, Science and Technology, AT&S AG, Fabriksgasse 13, A-8700 Leoben, Austria,
Department of Chemistry, The UniVersity of Arizona, 1306 East UniVersity BouleVard,
Tucson, Arizona 85721-0041, and SerVice de Chimie des Mate´riaux NouVeaux, UniVersite´ de Mons-Hainaut,
Place du Parc 20, B-7000 Mons, Belgium
ReceiVed: March 11, 2002; In Final Form: April 30, 2002
We report the synthesis of novel phenylenevinylene derivatives that allow for the introduction of meta versus
para connections on the phenylene rings, as well as the incorporation of nitrogen atoms within the conjugated
backbone and the attachment of electroactive substituents. We assess the impact of the various derivatization
schemes on the electronic and optical properties by means of gas-phase ultraviolet photoelectron spectroscopy
(UPS) and optical absorption and emission measurements; the evolution of the experimental data is further
supported by the results of quantum-chemical calculations. We demonstrate that the electronic and optical
properties of phenylenevinylene chains, and by extension those of many conjugated materials, can be tuned
over a large energy range by a tailored design of the molecular structures.
1. Introduction
within the organic medium to confine the charge carriers at
molecular interfaces and hence generate a preferential zone for
the radiative recombination of the electron-hole pairs.5 Other
strategies to balance the electron and hole injection and to
improve device performance consist of (i) attaching electron-
transport moieties on the conjugated polymer backbones6,7 or
at the end of flexible side chains to avoid changes in the color
of emitted light (as is typically observed upon direct attachment
of substituents on the backbone8), and (ii) blending polymers
with electron-transport and hole-transport units.9,10
Organic conjugated materials are now in the stage of
commercialization under the form of active elements in light-
emitting diodes (LED) used as pixels in low-resolution matrix
displays. A typical LED device is made by sandwiching organic
thin film(s) between two metal electrodes.1,2 The luminescence
signal is generated as a result of several processes:2 (i) injection
of electrons and holes in the organic layer from the cathode
and the anode, respectively, (ii) migration of the charges in
opposite directions under the influence of a static electric field,
(iii) recombination of the electrons and holes under the form
of intramolecular singlet and triplet excitons, and (iv) radiative
decay of the excitons to the ground state. The achievement of
high electroluminescence quantum yield requires a balanced
injection of the electrons and holes into the organic layer. This
is generally fulfilled by the design of multilayer devices in which
the emission layer is surrounded by a hole-transporting layer
or an electron-transporting layer or both;3,4 the main role of the
hole[electron]-transporting layer is to minimize the barrier
between the energy of the highest occupied level [the lowest
unoccupied level] of the material and the Fermi energy of the
metal at the electrode to ease the injection process. Moreover,
the use of several layers can help in creating energy barriers
Conjugated materials also emerge as attractive candidates for
the fabrication of photodiodes11-13 and, by extension, solar cells
and optical scanners.14 Such devices generally involve a binary
mixture of two chemically distinct conjugated systems, one with
a low ionization potential playing the role of the donor and the
other with a high electron affinity acting as the acceptor. The
conversion of the incident light into an electrical current requires
(i) light absorption by the donor or the acceptor or both, (ii)
migration of the excitons toward the interface between the two
materials (this step can be avoided in homogeneous blends),
(iii) exciton dissociation at the molecular interface into charge
carriers, for instance, by transferring the electron to the acceptor
while keeping the hole on the donor if the latter is initially
photoexcited (the efficiency of this photoinduced charge transfer
process critically depends on the relative energies of the frontier
levels of two materials in interaction),15 and (iv) charge transport
to the electrodes.
* To whom correspondence should be addressed.
† De´partement de Chimie Organique, Universite´ de Mons-Hainaut.
‡ Service des Mate´riaux Polyme`res et Composites, Universite´ de Mons-
Hainaut.
§ De´partement de Chimie Biologique, Universite´ de Mons-Hainaut.
These descriptions underscore that the design of efficient
LEDs and photodiodes requires a proper energy alignment of
the frontier electronic levels of the materials used in the active
layers and, hence, strategies to tailor the electronic structure of
conjugated materials. It is also of interest to develop routes to
| Technische Universita¨t Graz.
AT&S AG.
# The University of Arizona.
3 Service de Chimie des Mate´riaux Nouveaux, Universite´ de Mons-
Hainaut.
10.1021/jp025748d CCC: $22.00 © 2002 American Chemical Society
Published on Web 06/04/2002

Novel para- and meta-Phenylenevinylene Derivatives
J. Phys. Chem. B, Vol. 106, No. 25, 2002 6443
Figure 1. Sketch of the synthetic route and chemical structures of
compounds 1-8.
fine-tune the energies of the lowest singlet and triplet excited
states of such materials, which are primarily governed by their
one-electron structure. This is needed, for instance, to modulate
the color of the light emitted by electroluminescent diodes or
the efficiency of the energy-transfer processes exploited in
electrophosphorescent devices.16,17 Up to now, the search for
the best matching partners in a given application has been
achieved in many cases by empirical trials from a wide set of
available compounds. This task would be significantly facilitated
if the impact of various derivatization schemes could be
quantified and exploited to vary, in a controlled way and over
a large range, the energies of the frontier one-electron levels
and those of the excited states.
Figure 2. Gas-phase UPS spectra of compounds 5-8 (left). The spectra
are shown in the order of decreasing HOMO binding energy from top
to bottom; corresponding simulations of the UPS spectra at the INDO
level are shown in the right panel.
quantum-chemical approach as a predictive tool for designing
molecules prior to chemical synthesis.
2. Geometric and Electronic Properties
The chemical structures of the molecular compounds under
study are illustrated in Figure 1. The most stable conformers
are characterized by the following: (i) the coplanarity of the
vinylene linkages and the pyrimidine rings; it can be noted that
a 180° rotation of the pyrimidine units leads to a less-stable
structure due to the activation of steric effects between adjacent
hydrogen atoms; (ii) a torsion angle on the order of 15°-25°
between the plane of the vinylene units and that of the adjacent
benzene rings as a result of steric interactions between hydrogen
atoms; (iii) a full conjugation between the π-atomic orbitals in
the methoxy and aldehyde groups and those of the aromatic
rings to which they are attached, as already shown by previous
calculations for methoxy substituents.21
On the basis of these structural data, we investigate the way
in which the energy of the HOMO level is affected by the
various derivatization schemes. To do so, we report in Figure
2 the gas-phase UPS spectra of the three-ring oligomers together
with the corresponding intermediate neglect of differential
overlap (INDO) simulations; we also collect in Table 1 the
energy of the HOMO and LUMO levels calculated at the INDO
level for all of the compounds. The experimental results show
that the energy of the HOMO level spans a range of ∼0.8 eV
among the three-ring compounds.
We have recently demonstrated that (di)methyldiazines readily
react with aromatic aldehydes, in a hot aqueous solution of
sodium hydroxide 5 M and in the presence of a catalytic amount
(10%) of a quaternary ammonium salt, to afford the expected
condensation compounds.18 In this contribution, we apply this
efficient protocol to synthesize novel oligophenylenevinylene
derivatives, the chemical structures of which are shown in Figure
1. The focus on conjugated structures derived from oligo(para-
phenylenevinylene)s is motivated by the tremendous interest
devoted to the parent polymer (PPV) and its substituted
derivatives for applications in optoelectronic devices.3,10,12,15,19,20
The compounds synthesized here are obtained in short reaction
times (less than 1 h) and excellent yields (over 70%). In addition,
the electronic and optical properties can be modulated by (i)
varying the number of repeating units, (ii) controlling the para
versus meta connections on the aromatic rings, (iii) varying the
number of nitrogen atoms involved along the conjugated path,
or (iv) attaching electroactive substituents on the rings. These
compounds thus offer a unique opportunity to study on a
quantitative basis the impact of the various derivatization
schemes on the electronic and optical properties of the chains,
which we assess by means of gas-phase ultraviolet photoelectron
spectroscopy (UPS) and optical absorption and emission meas-
urements. We demonstrate that cooperation between various
effects can lead to significant fluctuations in the energies of
the one-electron levels and excited states and can be exploited
for a fine-tuning of these properties. The experimental data are
further supported by the results of semiempirical quantum-
chemical calculations; the very good agreement observed
between theory and experiment also validates the use of our
We first focus on the way in which the meta connections
impact the electronic structure of the oligomers. Comparison
between the experimental and theoretical UPS spectra allows
us to assign the peak around -7.9 eV in 8 to the optical
signature of the HOMO level and the doublet observed at higher
binding energy in 7 to the HOMO and HOMO-1 levels. The
influence of the meta versus para connection is illustrated by
the change in the HOMO level energy when going from 8 to 7;

6444 J. Phys. Chem. B, Vol. 106, No. 25, 2002
Pascal et al.
preliminary step toward organic magnets.23 Note also that the
theoretical estimates of the HOMO energy are underestimated
by ∼0.3 eV with respect to the experimental values. This
contrasts with a similar joint study performed on the 4,4′-bis-
(N-m-tolyl-N-phenylamino)biphenyl (TPD) molecule and sub-
stituted derivatives in which a remarkable quantitative agreement
was found between the INDO calculations and the gas-phase
UPS data.24 On the basis of previous calculations performed in
gas phase, we attribute the present discrepancy to the slight
changes in the total energy of PPV-like oligomers when
modulating the torsion angles between the vinylene units and
the aromatic rings up to ∼40°; this may lead to actual torsion
angles larger than those obtained at the AM1 level21 (and hence
to larger values of the HOMO energy).
TABLE 1: Experimental and Theoretical Energies of the
HOMO Level of Compounds 1-8 as Estimated from
Gas-Phase UPS Spectra and INDO Calculations,
Respectively, and the Energy of the LUMO Level Calculated
at the INDO Level
HOMO/INDOa (eV) HOMO/UPSa (eV) LUMO/INDO (eV)
1
2
3
4
5
6
7
8
-7.98
-0.54
-0.52
-1.07
-0.65
-0.63
-1.24
-0.68
-0.90
-7.74
-8.27
-7.77 (-8.02)
-7.58 (-7.82)
-8.18 (-8.46)
-7.90
-8.11 (-8.33)
-7.61 (-7.85)
-8.42 (-8.70)
-8.19
-7.67
-7.93
a Values in parentheses are the energy of the HOMO-1 level when
the UPS peak with the lowest binding energy corresponds to the
superposition of two electronic levels.
We now turn to the role played by the nitrogen atoms in
determining the energy of the frontier electronic levels in the
various compounds. The introduction of four nitrogen atoms to
yield compounds 7 and 8 induces at the INDO level a
stabilization on the order of 0.4-0.5 eV for both the HOMO
and LUMO levels with respect to the corresponding hydrocarbon
derivatives; this trend is in agreement with the results obtained
in a previous theoretical study based on INDO calculations.25
However, the stabilization of the HOMO and LUMO levels does
not exclusively correlate with the number of nitrogen atoms in
the molecule and is actually also very sensitive to their positions
along the molecular backbone. Although no general trends
prevail, the changes in the one-electron structure between two
compounds differing only by the position and number of
nitrogen atoms along the conjugated backbone can be rational-
ized by the results of our quantum-chemical calculations. This
is demonstrated in Figure 2; the evolution of the intense peak
observed around -8.0 eV in the UPS spectrum of 4 (which
corresponds to the superposition of the highest two occupied
levels) into a doublet in 7 is well-reproduced by the theoretical
simulations. The doublet in 7 is made of two individual peaks
separated by 0.45 eV in the UPS spectrum, while a value of
0.35 eV is obtained at the INDO level. The theoretical results
further indicate that the HOMO level of 4 has an equivalent
level stabilized by ∼0.5 eV in 7 in which it corresponds to the
H-1 level, see Figure 3. This evolution is rationalized by the
fact that the nitrogen atoms in 4 are located on sites possessing
a very small electronic density on the HOMO level, in contrast
to the situation found for the equivalent level in 7. For similar
reasons, the H-1 level of 4 is destabilized by 0.13 eV in 7 in
which it becomes the highest occupied level. Globally, the
calculations thus show that the HOMO level is stabilized by
0.13 eV when going from 4 to 7, while the UPS spectrum
provides an estimate on the order of 0.07 eV.
Figure 3. LCAO pattern of the highest occupied levels in compounds
4, 7, and 8 (assumed to be in a planar conformation). The size and
color of the circles over the atoms correspond to the amplitude and
sign of the LCAO coefficient associated with the π atomic orbitals.
The arrows denote the location of the nitrogen atoms in compounds 4
and 7.
the UPS data show a significant stabilization of the HOMO level
by 0.26 eV upon introduction of the meta connection, which is
fully reproduced by the INDO calculations. In contrast, the
calculations indicate that the LUMO level is destabilized by
∼0.2 eV. This leads to a total increase in the HOMO-LUMO
difference by some 0.46 eV when going from 8 to 7; note that
similar results are calculated between the corresponding hydro-
carbon derivatives. This evolution is rationalized by the linear
combination of atomic orbitals (LCAO) patterns of the HOMO
and LUMO levels in the two compounds showing a break of
conjugation in the central ring when the meta connection is
introduced (as illustrated for the HOMO level in Figure 3). The
reduction in the degree of delocalization by meta bridging has
also been demonstrated by previous AM1 calculations performed
on m-phenylenevinylene oligomers containing from two to seven
rings; these results pointed to the very slow evolution of the
HOMO and LUMO energy with chain size.22 The rupture of
conjugation induced by meta connections has been largely
exploited in the design of high-spin radical cations as a
Next, we address the influence of electroactive substituents
attached on the conjugated backbone by comparing the UPS
spectra of 4, 5, and 6. The INDO simulations indicate for the
three compounds that the lowest binding energy peak corre-
sponds to the overlapping signature of the HOMO and HOMO-1
levels. A peak fitting of the experimental line shapes can
however provide the energy of the individual levels; this
procedure yields energy separations between the highest two
occupied levels that are in full agreement with the INDO
calculations (see Table 1). The UPS data and the calculations
show that the attachment of formyl groups on the terminal
carbon atoms of 4 to give 6 leads to a significant stabilization
of the HOMO level by 0.3 and 0.4 eV, respectively. The
influence of substituents is governed by the interplay between
the inductive effects taking place in the σ skeleton and the
mesomer effects associated with the π electrons. The stabiliza-

Novel para- and meta-Phenylenevinylene Derivatives
J. Phys. Chem. B, Vol. 106, No. 25, 2002 6445
TABLE 2: Experimental and Theoretical Transition
Energies of the Lowest Optical Transition and Associated
Molar Extinction Coefficient (Emax, in L/mol cm) of
Compounds 1-8a
tion of the HOMO level upon formyl substitution results from
their σ-acceptor and π-acceptor characters, which both depopu-
late the electronic density over the molecular backbone and
hence stabilize the frontier electronic levels. The INDO calcula-
tions yield a stabilization of the LUMO level by 0.6 eV when
attaching the aldehyde groups; the asymmetry in the stabilization
of the HOMO and LUMO levels will have strong implications
on the optical properties of the oligomers, as discussed in the
next section. In contrast, both the experimental and theoretical
results indicate that the introduction of methoxy groups leads
to a destabilization of the HOMO level; this confirms that the
effects linked to the π-donor character of the methoxy groups
dominate over those associated with their σ-acceptor character.
Similarly, the LUMO level is slightly destabilized upon methoxy
substitution as a result of the counteracting influence of the
inductive and mesomer effects. The HOMO shift estimated at
the INDO level is underestimated with respect to the experi-
mental value (0.2 versus 0.5 eV); this discrepancy cannot be
attributed to the INDO parametrization, which has been shown
to provide results comparing remarkably well to gas-phase UPS
spectra in the case of the TPD molecule and methoxy-substituted
derivatives.24 We attribute this discrepancy once again to the
underestimation of the actual torsion angles in the gas-phase
conformations; the good quantitative agreement observed
between the experimental and theoretical HOMO energy for 5
suggests that the conformation of this molecule is less twisted
than that in the other compounds and is actually very close to
that calculated at the AM1 level. We stress that the electronic
properties could be further tuned by varying the nature, number,
and position of the substituents over the conjugated backbone.26
Very similar trends are calculated for the two-ring compounds
when going from the unsubstituted backbone 1 to the derivatives
substituted by a methoxy 2 and an aldehyde 3 group; the HOMO
energy in these compounds is, however, systematically stabilized
with respect to the values obtained for the corresponding three-
ring oligomers because of the shortening of the conjugated
backbone.
INDO/SCI
dichloromethane
ethanol
energy (eV) ꢀmax energy (eV) ꢀmax energy (eV) ꢀmax
1
2
3
4
5
6
7
8
4.09
3.96
3.96
3.76
3.67
3.66
4.02
3.62
28 500
31 000
36 000
34 000
38 500
46 500
52 500
52 000
3.96
3.67
3.81
3.59
3.45
3.51
3.94
3.44
28 500
28 500
40 000
36 500
36 000
50 000
53 000
51 000
3.92
3.63
3.80
3.59
3.36
3.49
3.89
3.42
28 500
28 500
36 000
34 500
36 500
50 000
52 500
53 000
a The calculated molar extinction coefficients have been scaled with
respect to the experimental value obtained for compound 1.
compounds with a high electron affinity are required in order
to (i) allow for the injection of electrons in organic layers from
environmentally stable metallic electrodes, characterized by high
work functions, (ii) promote exciton dissociation in molecular
blends used in photodiodes, or (iii) ensure efficient electron
transport not hampered by trapping associated with oxygen
contamination.27,28
3. Optical Properties
The lowest transition energy of oligo(paraphenylenevinylene)s
decreases nearly linearly as a function of the inverse number
of monomer units along the chains, as evidenced by experi-
mental29,30 and calculated31 absorption and photoluminescence
spectra. As a result of this chain-length evolution, short
oligomers emit in the blue while longer chains are characterized
by a yellow-green emission;31 the energy of the lowest optical
transition can thus be varied by up to 1 eV when modulating
the chain length, with the steepest evolution being observed
among the short chains. It is thus of interest to explore the way
in which the absorption and emission properties are affected
for short chains by the incorporation of para versus meta
connections as well as by the introduction of nitrogen atoms or
electroactive substituents or both along the backbone; to do so,
we analyze below the absorption and emission spectra of
compounds 1-8.
The optical absorption spectra of compounds 1-8 have been
recorded in dilute solutions (C ) 2 × 10-5 M) of ethanol and
dichloromethane at room temperature. These two solvents have
been chosen to investigate the influence of solvent polarity
(dielectric constants of 9.08 and 24.30 for dichloromethane and
ethanol, respectively) on the absorption spectra. We collect in
Table 2 the experimental transition energy and molar extinction
coefficient of the lowest absorption band in the eight com-
pounds. The spectra are found to be very similar in the two
solvents, except for a slight red shift of the transition energies
when going from dichloromethane to ethanol, thus with increas-
ing solvent polarity. We also report in Table 2 the transition
energies calculated at the INDO/SCI level for the oligomers,
assumed in first approximation to be in a planar conformation,
together with the relative intensities of the lowest absorption
band scaled with respect to the experimental value obtained for
compound 1. Despite a very good agreement between the
experimental and simulated spectra, we observe that the
calculated transition energies are slightly overestimated with
respect to the experimental values. Because calculations per-
formed on twisted backbones would increase the energy
difference between the theoretical and experimental values, we
attribute most of this shift to the neglect of the solvent effects
in our calculations; the latter are expected to stabilize the energy
All together, our joint theoretical and experimental charac-
terization demonstrates that the HOMO and LUMO energies
of oligo(phenylenevinylene) chains can be deeply modulated
by exploiting various derivatization schemes, which can act
cooperatively. We can formulate general guidelines as follows,
starting from a paraphenylenevinylene oligomer of a given size
as a reference compound: (i) the HOMO level is stabilized by
shortening the chain size, introducing nitrogen atoms, inserting
meta-linkages, and attaching substituents with a π-acceptor or
σ-acceptor character; it gets destabilized by increasing the chain
length and substituting the backbone with groups having a
π-donor or σ-donor character; (ii) the LUMO is stabilized by
the increase in chain length, the introduction of nitrogen atoms,
and the attachment of substituents with a π-acceptor or
σ-acceptor character; this level is destabilized by the reduction
of chain length, the incorporation of meta-connections, and the
attachment of groups with a π-donor or σ-donor character.
The very good agreement observed between the experimental
data and the theoretical calculations suggests that a molecular
design aimed at matching a specific electronic property can be
first accomplished at the theoretical level prior to chemical
synthesis. In particular, the results show that the electron affinity
of a three-ring meta-phenylenevinylene oligomer can be in-
creased by 1.0 eV following the introduction of nitrogen atoms
in the phenylene rings and the attachment of aldehyde groups
(with π-acceptor and σ-acceptor characters). This is of great
interest for application in (opto)electronic devices in which

6446 J. Phys. Chem. B, Vol. 106, No. 25, 2002
Pascal et al.
Figure 4. Optical absorption spectra of compounds 7 and 8 (top) and compounds 4, 5, and 6 (bottom) in dichloromethane.
of the lowest excited state because of the partial charge-transfer
taking place in all of the compounds from the phenylene to the
pyrimidine rings.
corresponding hydrocarbon backbone (i.e., the three-ring PPV
oligomer). This is driven by the fact that both transitions are
primarily described by a HOMO-LUMO excitation and that
the incorporation of nitrogen atoms leads to an almost symmetric
stabilization of the HOMO and LUMO levels (0.53 and 0.45
eV, respectively). Thus, in para-connected chains, we can
conclude that the nitrogen atoms strongly impact the one-
electron structure of the chains but only slightly affect the energy
of the lowest absorption and emission bands. The situation is
much more complex in derivatives in which the lowest optical
absorption band does not originate from a single excited state
mostly described by a HOMO-LUMO excitation, as is the case
for the oligomers incorporating meta connections. The calcula-
tions indicate that compound 7 and the corresponding hydro-
carbon chain are both characterized by a single absorption band
centered around 4.0 eV, resulting from the superposition of three
distinct excited states; in contrast, they predict for compound 4
that this band is split into two features around 3.7 and 4.5 eV
with an approximate 10/6 intensity ratio, which is fully
consistent with the experimental absorption spectra.
We display in Figure 4 the absorption spectra of compounds
7 and 8 obtained in dichloromethane. The lowest absorption
band has a similar intensity for the two compounds; however,
the transition energy of the oligomer with para connections on
the central ring is red shifted by 0.5 eV with respect to the
oligomer displaying meta connections, in full agreement with
the quantum-chemical calculations. The lowest absorption band
of 8 originates from the lowest excited state of the oligomer,
which is primarily described by an electronic excitation from
HOMO to LUMO; in contrast, the calculations show that the
lowest absorption band of 7 results from the superposition of
three distinct excited states. These results demonstrate that the
introduction of meta connections prevents an efficient delocal-
ization of the electronic wave function over the conjugated
backbone and, hence, a significant red shift of the lowest optical
transition with chain length; this is supported by experimental
data32,33 and previous calculations on oligo(metaphenylene-
vinylene)s showing a vanishingly small bathochromic shift of
the lowest optical transition in chains containing up to five
monomer units.22
We have also investigated the impact of the electroactive
substituents on the optical properties of the oligomers. Figure
4 also illustrates the way in which the absorption of compound
4 is modified when attaching methoxy or formyl groups on the
terminal carbon atoms. In both cases, a slight red shift of the
The energy and intensity of the lowest optical transition of
compound 8 are very similar to those computed for the

Novel para- and meta-Phenylenevinylene Derivatives
J. Phys. Chem. B, Vol. 106, No. 25, 2002 6447
Figure 5. Emission spectra of compounds 4-8 measured in dichloromethane. We report in parentheses the excitation energy used for each compound;
the spectra have been normalized with respect to the emission of compound 4.
TABLE 3: Experimental Transition Energies (in eV) of the
Emission Band of Compounds 4-8 in Dichloromethane and
Ethanol and Corresponding Shift Defined as the Energy
Difference between the Maximum of the Absorption and
Emission Bands and the Fluorescence Quantum Yield (ΦF)
Measured in Ethanol
lowest optical transition is observed upon substitution. Because
the lowest absorption feature of 4 observed around 3.6 eV is
mainly described by a HOMO-LUMO excitation, the red shift
is rationalized by the fact that the methoxy [formyl] groups lead
to an asymmetric destabilization [stabilization] of the frontier
electronic levels, which reduces the electronic band gap, as
discussed in the previous section. These trends are supported
by experimental and theoretical data reported on oligo(para-
phenylenevinylene)s substituted by cyano or methoxy groups
or both.26 A similar evolution is also observed for the two-ring
compounds when going from 1 to the substituted derivatives 2
and 3; their lowest optical transition is, however, blue-shifted
and reduced in intensity with respect to the three-ring oligomers
as a result of the chain shortening.
dichloromethane
ethanol
shift
energy (eV)
shift
energy (eV)
ΦF
4
5
6
7
8
3.12
2.82
3.05
3.20
3.11
0.47
0.63
0.46
0.74
0.33
3.01
2.64
2.79
3.08
2.97
0.58
0.72
0.72
0.81
0.45
0.034
0.076
0.010
0.012
0.034
spectra are weakly affected by the solvent (a small red shift is
observed when going from dichloromethane to ethanol). The
close match observed between the absorption and excitation
spectra for all of the compounds confirms that the photogener-
ated electron-hole pairs decay into the lowest excited state
before giving rise to light emission. Table 3 indicates that the
energy difference between the transition energy at the maximum
of the absorption and emission bands varies significantly among
the compounds. This reflects the amplitude of the geometry
relaxation taking place in the lowest excited state upon
photoexcitation and is expected to decrease with conjugation
length;30 this is consistent with the marked increase in this
energy difference when going from compounds 8 to 7 (i.e., when
breaking conjugation by the introduction of meta connections).
Because of these variations in the amplitude of the geometry
relaxation effects, the evolution of the transition energy of the
emission band among the various compounds does not fully
parallel that observed for the lowest absorption band; this points
to the fact that geometry relaxation effects have to be taken
into account at the theoretical level to provide accurate locations
of the energy of the lowest excited state in its equilibrium
geometry.
In summary, the theoretical and experimental data demon-
strate that meta connections prevent a progressive red shift of
the lowest optical transition in phenylenevinylene oligomers of
increasing length; this is of particular interest in the search for
new attractive candidates to be used as blue emitters in
optoelectronic devices. The energy of the lowest optical
transition can be further fine-tuned by attaching electroactive
substituents or incorporating nitrogen atoms along the conju-
gated backbone or both. The results also show that nitrogen
atoms can be used in some cases to strongly modify the one-
electron properties of conjugated chains (i.e., the ionization
potential and electron affinity) without affecting their absorption
and emission properties. We emphasize that the energy of the
lowest absorption band can also be modulated by the protonation
of the nitrogen atoms involved in the conjugated path,34 as
described in previous experimental studies;35 a detailed descrip-
tion of such effects is beyond the scope of the present paper
and will be discussed elsewhere.
We report in Figure 5 the emission spectra of the three-ring
compounds recorded in dichloromethane and collect in Table
3 the measured λmax values; we also report in Table 3 the λmax
values obtained in ethanol, as well as the corresponding
fluorescence quantum yields. Except for compound 8, the
emission spectra are structureless, thus pointing to a pronounced
conformational disorder of the emitting chains in solution. As
is the case for the absorption spectra, the λmax of the emission
The fluorescence quantum yields in compounds 4-8 are
rather low (systematically below 10%) when compared to those
in oligo(para-phenylenevinylene)s, typically characterized by
values between 50% and 90%;29 values above 50% have also
been reported for phenylenevinylene backbones incorporating

6448 J. Phys. Chem. B, Vol. 106, No. 25, 2002
Pascal et al.
both meta and para connections.37 In contrast, the efficiency of
light emission of the stilbene molecule in solution is generally
much lower as a result of a trans-cis photoisomerization
mechanism38 (note that no fluorescence signal is detected for
the two-ring oligomers, 1-3). Recent experimental studies have
also indicated that the fluorescence quantum yield is strongly
reduced for PPV oligomers displaying torsions along the
backbone upon substitution.38 The low quantum efficiency of
the three-ring compounds under study thus points to the
existence of a mechanism competing with the radiative decay
route in the lowest excited state. This quenching can be
associated with the following: (i) an intersystem crossing
process (ISC) from the singlet to the triplet manifold induced
by torsions along the conjugated backbone38,39 and leading to
the population of the lowest triplet excited state (and possibly
to a weak phosphorescence signal at longer time scales40); the
broadness of the emission bands is consistent with the presence
of torsions along the backbone, which are required for the ISC
process to occur in conjugated backbones without heavy atoms;
or (ii) a photoinduced twisted intramolecular charge transfer
(TICT) process leading to a complete separation between the
electron and hole as a result of a twist of the central part of the
chain; this mechanism has been identified in aminostilbazonium
ions.41
These torsions responsible for reducing the fluorescence
quantum yield are likely to be induced by a subtle interplay
between the presence of nitrogen atoms along the backbone and
solvation effects. This is supported by the change in fluorescence
quantum yield observed for the most emissive compound 5 when
going from ethanol to dicholoromethane (from 0.076 to 0.041,
respectively). It is worth stressing that the solid-state packing
will tend to planarize the chains in the absence of significant
steric effects and, hence, to strongly reduce the efficiency of
the ISC process; similarly, the possible formation of the highly
twisted excited states involved in the TICT process should be
strongly hindered in the solid state.
Department of Energy. The gas-phase UPS were measured at
the Center for Gas-Phase Electron Spectroscopy at The Uni-
versity of Arizona. J.C. is an FNRS research associate; L.P.
acknowledges a grant from the Fonds pour la Formation a` la
Recherche dans l’Industrie et dans l’Agriculture (FRIA).
Appendix: Experimental and Theoretical Section
Synthesis. 4-Methyl- (a) or 4,6-dimethylpyrimidine (b) reacts
with benzaldehyde (c) in a hot aqueous solution of sodium
hydroxide (5 M) and in the absence of any organic solvent to
yield compounds 1-8 (see Figure 1).18 The reactions do not
occur without a catalyst; the addition of a quaternary ammonium
salt (such as Aliquat 336 or tetrabutylammonium hydrogen
sulfate; 10 mol %) substantially favors the kinetics of formation
of the expected condensation compounds; the latter are obtained
in short reaction times (less than 1 h) and excellent yields (more
than 70%). Interestingly, the crude products can be isolated by
a mere filtration thus limiting the use of an organic solvent for
purification by recrystallization (except for compounds 3 and
6, see below). Solvents and reagents are commercially available
(Acros Organics, Aldrich Co) and were used without further
purification. Melting points (uncorrected) were determined on
a Electrothermal 9100 apparatus. NMR spectra (CDCl3) were
recorded on a JEOL JNP-PMX 60 spectrometer (60 MHz for
1
1H at 1.4 T) or a Bruker AMX spectrometer (300 MHz for H
and 75 MHz for 13C at 7.0 T). Chemical shifts are given in
ppm using TMS as internal reference. IR spectra were recorded
on a Perkin-Elmer FTIR 1760K spectrophotometer. The ele-
mental analyses were carried out at the Departement de Chimie
Pharmaceutique of the Universite´ de Lie`ge (Belgium). High-
resolution mass spectra have been recorded on a Micromass
Auto Spec 6F, resolving power ca. 10 000 (5% valley), electron
ionization, 70 eV energy, 200 µA trap current, PFK reference
compound. Compounds (yield) 1 (90%),42 2 (90%),43 3 (80%),44
4 (90%),45 and 5 (90%)46 have already been described in the
literature.
General Procedure for the Preparation of Compounds 1, 2,
4, 5, and 7. A stoichiometric mixture of the (di)methylpyrimi-
dine and the aromatic aldehyde in an aqueous solution (25 mL
for 5 mmol of the heterocycle) of sodium hydroxide (5 M)
containing Aliquat 336 or tetrabutylammonium hydrogen sulfate
(10 mol % vs the heterocycle) was heated under reflux for 1 h.
After cooling, the crude product was filtered off and recrystal-
lized.
Preparation of Compound 3. A stoichiometric mixture of
4-methylpyrimidine and benzene-1,4-dicarboxaldehyde mono-
(diethyl acetal) in an aqueous solution (25 mL for 5 mmol of
the heterocycle) of sodium hydroxide (5 M) containing Aliquat
336 or tetrabutylammonium hydrogen sulfate (10 mol % versus
the heterocycle) was heated under reflux for 1 h. After cooling,
the reaction mixture was extracted with dichloromethane (75
mL). The solution was dried on magnesium sulfate, and the
solvent was removed by evaporation under reduced pressure.
The residue, 4-{2-[4-(diethoxymethyl)phenyl]ethenyl}pyrimidine,
was dissolved in acetone (50 mL), and a small amount (0.5
mL) of concentrated hydrochloric acid was added under
agitation. Compound 3 precipited in the acetone and has been
filtered off and recrystallized.
4. Conclusions and Perspectives
In this work, we have shown that the electronic and optical
properties of a class of conjugated compounds can be widely
modulated through molecular engineering. This has been
illustrated by comparing the properties of novel phenylene-
vinylene derivatives, in which chain legnth, incorporation of
nitrogen atoms, attachment of electroactive substituents, and
introduction of meta versus para connections can be controlled
through our synthetic approach.
We have assessed on a quantitative basis the impact of the
various derivatization schemes on the electronic and optical
properties of the chains by means of gas-phase UPS, optical
absorption, and fluorescence measurements. The evolution of
the properties among the derivatives has been further rationalized
by the results of quantum-chemical calculations, which can thus
provide useful guidelines for molecular engineering. We stress
that the present results are applicable to many other conjugated
oligomeric backbones.
Acknowledgment. The work in Mons is partly supported
by the Belgian Federal Governement (PAI 4/11), the European
Community for general support in the frame of “Objectif 1:
Materia Nova”, and the Belgian National Fund for Scientific
Research (FNRS/FRFC). The work at Arizona is partly sup-
ported by National Science Foundation (Grant CHEM-0078819),
the Office of Naval Research, the IBM Shared University
Research Program, the Petroleum Research Fund, and the
Preparation of Compound 6. The procedure used to obtain
compound 3 was applied to 4,6-dimethylpyrimidine.
Preparation of Compound 8. A stoichiometric mixture of
4-methylpyrimidine and 4-[2-(4-pyrimidyl)ethenyl]benzaldehyde
3 in an aqueous solution (25 mL for 5 mmol of the heterocycle)
of sodium hydroxide (5 M) containing Aliquat 336 or tetra-

Novel para- and meta-Phenylenevinylene Derivatives
J. Phys. Chem. B, Vol. 106, No. 25, 2002 6449
butylammonium hydrogen sulfate (10 mol % vs the heterocycle)
was heated under reflux for 1 h. After cooling, the crude product
was filtered off and recrystallized.
yield of 0.75 in ethanol,49 was taken as the reference compound.
We stress that the PPV-like oligomer and the reference
compound all emit in a wavelength range at which the spectral
sensitivity of the fluorimeter is almost constant.
4-[2-(4-Pyrimidinyl)ethenyl]-4′-[2-(6-pyrimidinyl)ethenyl]-
benzaldehyde (6). Yield: 85%. M.p. (CH3CN) 230 °C (decom-
Theoretical Characterization. The geometries of all of the
molecular compounds were optimized with the help of the semi-
empirical Hartree-Fock AM1 (Austin Model 1) Hamiltonian,
which has been parametrized to accurately reproduce ground-
state geometric structures.50 All of the internal degrees of
freedom are optimized to provide the structural data used as
input in the simulation of the gas-phase UPS spectra. On the
basis of the optimized geometries, we have calculated the one-
electron structure of the compounds by means of the spectro-
scopic version of the semiempirical intermediate neglect of
differential overlap (INDO) Hamiltonian developed by Zerner
and co-workers.51 The UPS spectra are simulated at the INDO
level within Koopmans’ approximation, as described in refs 52
and 53, in which the choice of our approach to describe the
nature of the highest occupied levels of organic conjugated
materials was validated. The optical absorption spectra are
computed by coupling the INDO Hamiltonian to a single
configuration interaction (SCI) scheme; all of the π levels of
the molecules are involved in the generation of the singly excited
configurations to ensure size consistency. Our previous studies
have demonstrated that this technique is successful in describing
the nature and the energies of the lowest excited states of organic
conjugated molecules.54,55
1
position). H NMR δ 10.0 (2H, s), 9.1 (1H, s), 8.0 (2H, d, 16
Hz), 7.9 (4H, d, 8 Hz), 7.7 (4H, d, 10 Hz), 7.3 (1H, s), 7.2 (2H,
d, 16 Hz) ppm. 13C NMR not available (solubility too low). IR
(KBr) 1696, 1602, 1576, 810, 515 cm-1. Anal. Calcd. for
C22H16N2O2: C, 77.63; H, 4.74; N, 8.23. Found: C, 76.88; H,
4.86; N, 8.21. HRMS Calcd. for C22H16N2O2: 340.121 178.
Found: 340.121 140.
4,4′-(1,3-Phenylenedi-2,1-ethenediyl)bis-pyrimidine (7).
Yield: 70%. M.p. (CH3CN) 182-4 °C. 1H NMR δ 9.2 (2H, s),
8.6 (2H, d, 6 Hz), 7.9 (2H, d, 16 Hz), 7.8 (1H, s), 7.6-7.1 (5H,
m), 7.0 (2H, d, 16 Hz). 13C NMR δ 161.9, 158.9, 157.5, 136.7,
136.2, 129.4, 128.3, 126.8, 126.2, 118.8 ppm. IR (KBr) 1636,
1572, 1461, 1386, 988 cm-1. Anal. Calcd. for C18H14N4: C,
75.50; H, 4.93; N, 19.57. Found: C, 75.40; H, 4.97; N, 19.58.
HRMS Calcd. for C18H14N4: 286.121 847. Found: 286.122 187.
4,4′-(1,4-Phenylenedi-2,1-ethenediyl)bis-pyrimidine (8).
Yield: 90%. M.p. (DMSO) 234-6 °C. 1H NMR δ 9.2 (2H, s),
8.7 (2H, d, 6 Hz), 7.9 (2H, d, 16 Hz), 7.6 (4H, s), 7.3 (2H, d,
6 Hz), 7.1 (2H, d, 16 Hz) ppm. 13C NMR δ 161.9, 158.9, 157.5,
136.7, 136.6, 128.3, 128.2, 126.3, 118.9 ppm. IR (KBr) 1634,
1568, 1385, 847, 564 cm-1. Anal. Calcd. for C18H14N4: C,
75.50; H, 4.93; N, 19.57. Found: C, 75.55; H, 4.87; N, 19.64.
HRMS Calcd. for C18H14N4: 286.121 847. Found: 286.121 283.
References and Notes
Experimental Characterization. The HeI gas-phase photo-
electron spectra of five different oligophenylenevinylenes were
recorded using an instrument that features a 36-cm radius, 8-cm
gap hemispherical analyzer (McPherson) and custom-designed
excitation source, sample cells, and detection and control
electronics that have been described in more detail previously.47
The oligophenylenevinylenes sublimed cleanly with no evidence
of decomposition products in the gas phase or as a solid residue.
The samples sublimed between 120 and 220 °C (10-4 Torr); 4
sublimed at 120-140 °C, 5 at 170-190 °C, 6 at 200-220 °C,
7 at 150-160 °C, and 8 at 160-180 °C. These temperatures
were monitored using a “K”-type thermocouple passed through
a vacuum feedthrough and attached directly to the sample cell.
The argon 2P3/2 ionization at 15.759 eV was used as an internal
calibration lock of the absolute ionization energy. The difference
(1) Sheats, J. R.; Antoniadis, H.; Hueschen, M.; Leonard, W.; Miller,
J.; Moon, R.; Roitman, D.; Stocking, A. Science 1996, 273, 884.
(2) Friend, R. H.; Gymer, R. W.; Holmes, A. B.; Burroughes, J. H.;
Marks, R. N.; Taliani, C.; Bradley, D. D. C.; dos Santos, D. A.; Bre´das, J.
L.; Lo¨gdlund, M.; Salaneck, W. R. Nature 1999, 397, 121.
(3) Tang, C. W.; van Slyke, S. A. Appl. Phys. Lett. 1987, 51, 913.
(4) Pardo, D. A.; Jabbour, G. E.; Peyghambarian, N. AdV. Mater. 2000,
12, 1249.
(5) Greenham, N. C.; Moratti, S. C.; Bradley, D. D. C.; Friend, R. H.;
Holmes, A. B. Nature 1993, 365, 268.
(6) Chung, S. J.; Kwon, K. Y.; Lee, S. W.; Jin, J. I.; Lee, C. H.; Lee,
C. E.; Park, Y. AdV. Mater. 1998, 10, 1112.
(7) Zheng, M.; Ding, L. M.; Gurel, E. E.; Lahti, P. M., Karasz, F. E.
Macromolecules 2001, 34, 4124.
(8) Lee, Y.; Chen, X.; Chen, S. A.; Wei, P. K.; Fann, W. S. J. Am.
Chem. Soc. 2001, 123, 2296.
(9) Cao, Y.; Parker, I. D.; Yu, G.; Zhang, C.; Heeger, A. J. Nature
1999, 397, 414.
(10) Ohmori, Y.; Hironaka, Y.; Yoshida, R.; Fujii, A.; Tada, N.; Yoshino,
2
2
between the argon P3/2 ionization and the methyl iodide E1/2
ionization at 9.538 eV was used to calibrate the ionization energy
scale. During data collection, the instrument resolution (meas-
ured using fwhm of the argon 2P3/2 peak) was 0.020-0.035 eV.
All data are intensity-corrected with an experimentally deter-
mined instrument analyzer sensitivity function. The HeI spec-
trum was also corrected for HeIâ resonance line emission from
the source, which is about 3% the intensity of the HeIR line
emission and at 1.869 eV higher photon energy. The first two
ionization peaks were fit using two asymmetric Gaussian
peaks.48 Ionization peak potentials are reproducible to (0.02
eV (∼3σ level).
K. Polym. AdV. Technol. 1997, 8, 403.
(11) Sariciftci, N. S.; Smilowitz, L.; Heeger, A. J.; Wudl, F. Science
1992, 258, 1474.
(12) Halls, J. J. M.; Walsh, C. A.; Greenham, N. C.; Marseglia, E. A.;
Friend, R. H.; Moratti, S. C.; Holmes, A. B. Nature 1995, 376, 498.
(13) Smolenyak, P.; Peterson, R.; Nebesny, K.; To¨rker, M.; O’Brien,
D. F.; Armstrong, N. R. J. Am. Chem. Soc. 1999, 121, 8628.
(14) Yu, G.; Wang, J.; McElvain, J.; Heeger, A. J. AdV. Mater. 1998,
17, 1431.
(15) Halls, J. J. M.; Cornil, J.; dos Santos, D. A.; Silbey, R.; Wang, D.
H.; Holmes, A. B.; Bre´das, J. L.; Friend, R. H. Phys. ReV. B 1999, 60,
5721.
(16) Baldo, M. A.; O’Brien, D. F.; You, Y.; Shoustikov, A.; Silbey, S.;
Thompson, M. E.; Forrest, S. R. Nature 1998, 395, 151.
(17) Cleave, V.; Yahioglu, G.; Le Barny, P.; Friend, R. H.; Tessler, N.
AdV. Mater. 1999, 11, 285.
(18) Vanden Eynde, J. J.; Pascal, L.; Dubois P.; Van Haverbeke Y. Synth.
Commun. 2001, 31, 3167.
The optical absorption spectra were recorded with a Varian
Cary 50 spectrophotometer in ethanol or in dichloromethane
(C ) 2 × 10-5 M). Excitation and emission spectra were
measured on a Shimadzu RF5301PC fluorimeter in dichloro-
methane (C ) 5 µg/mL). Fluorescence quantum yields, Φf, were
determined on a Perkin-Elmer MPF-2A fluorimeter from the
integrated fluorescence spectra of dilute solutions (absorbance
at the excitation wavelength, 350 nm, being lower than 0.1) in
ethanol; 7-amino-4-methylcoumarin, with a reported quantum
(19) Kraft, A.; Grimsdale, A. C.; Holmes, A. B. Angew. Chem., Int.
Ed. 1998, 37, 403.
(20) Primary Photoexcitations in Conjugated Polymers: Molecular
Exciton Versus Semiconductor Band Model; Sariciftci, N. S., Ed.; World
Scientific: Singapore, 1997.
(21) Lhost, O.; Bre´das, J. L. J. Chem. Phys. 1992, 96, 5279.
(22) Karabunarliev, S.; Baumgarten, M.; Tyutyulkov, N.; Mu¨llen, K. J.
Phys. Chem. 1994, 98, 11892.

6450 J. Phys. Chem. B, Vol. 106, No. 25, 2002
Pascal et al.
(23) Wienk, M. M.; Janssen, R. A. J. J. Am. Chem. Soc. 1997, 119,
4492.
(24) Cornil, J.; Gruhn, N. E.; dos Santos, D. A.; Malagoli, M.; Lee, P.
A.; Barlow, S.; Thayumanavan, S.; Marder, S. R.; Armstrong, N. R.; Bre´das,
J. L. J. Phys. Chem. A 2001, 105, 5206.
(25) Cornil, J.; Beljonne, D.; dos Santos, D. A.; Bre´das, J. L. Synth.
Met. 1996, 76, 101.
(26) Cornil, J.; dos Santos, D. A.; Beljonne, D.; Bre´das, J. L. J. Phys.
Chem. 1995, 99, 5604.
(38) Gierschner, J.; Lu¨er, L.; Oelkrug, D.; Musluoglu, E.; Behnisch, B.;
Hanack, M. AdV. Mater. 2000, 12, 757.
(39) Beljonne, D.; Shuai, Z.; Pourtois, G.; Bre´das, J. L. J. Phys. Chem.
A 2001, 105, 3899.
(40) Hertel, D.; Setayesh, S.; Nothofer, H. G.; Scherf, U.; Mu¨llen, K.;
Ba¨ssler, H. AdV. Mater. 2001, 13, 65.
(41) Ephardt, H.; Fromherz, P. J. Phys. Chem. 1989, 93, 7717.
(42) Gabriel, S.; Colman, J. Chem. Ber. 1903, 36, 3379.
(43) Haroutounian, S.; Katzenellenbogen, J. Tetrahedron 1995, 51, 1585.
(44) Hasegawa, M.; Endo, Y.; Aoyama, M.; Saigo, K. Bull. Chem. Soc.
Jpn. 1989, 62, 1556.
(45) Strul, M.; Zugravescu, I. ReV. Roum. Chim. 1971, 16, 1877.
(46) Reynolds, G. A.; Drexhage, K. H. Opt. Commun. 1975, 13, 222.
(47) Westcott, B. L.; Gruhn, N. E.; Enemark, J. H. J. Am. Chem. Soc.
1998, 120, 3382 and references therein.
(48) Lichtenberger, D. L.; Copenhaver, A. S. J. Electron Spectrosc. Relat.
Phenom. 1990, 50, 335.
(27) Horowitz, G. AdV. Mater. 1998, 10, 365.
(28) Katz, H. E.; Lovinger, A. J.; Johnson, J.; Kloc, C.; Siegrist, T.; Li,
W.; Lin, Y. Y.; Dodabalapur, A. Nature 2000, 404, 478.
(29) Peeters, E.; Ramos, A. M.; Meskers, S. C. J.; Janssen, R. A. J. J.
Chem. Phys. 2000, 112, 9445.
(30) Cornil, J.; Beljonne, D.; Heller, C. M.; Campbell, I. H.; Laurich,
B. K.; Smith, D. L.; Bradley, D. D. C.; Mu¨llen, K.; Bre´das, J. L. Chem.
Phys. Lett. 1997, 278, 139.
(31) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.;
Friend, R. H.; Burn, P. L.; Holmes, A. B. Nature 1990, 347, 359.
(32) Lane, P. A.; Cadby, A. J.; Mellor, H.; Martin, S. J.; Lidzey, D. G.;
Bradley, D. D. C.; Lipson, S. M.; O’Brien, D. F.; Blau, W. Phys. ReV. B
2000, 62, 15718.
(33) Liao, L.; Pang, Y.; Ding, L.; Karasz, F. E. Macromolecules 2001,
34, 7300.
(34) Pascal, L.; Vanden Eynde, J. J.; Van Haverbeke, Y.; Dubois, P.;
Cornil, J.; Bre´das, J. L. Manuscript in preparation.
(35) Egelhaaf, H. J.; Gierschner, J.; Oelkrug, D. Synth. Met. 1996, 83,
221.
(49) Reynolds, G. A.; Drexhage, K. H. Opt. Commun. 1975, 13, 222.
(50) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P. J.
Am. Chem. Soc. 1985, 107, 3902.
(51) Zerner, M. C.; Loew, G. H.; Kichner, R. F.; Mueller-Westerhoff,
U. T. J. Am. Chem. Soc. 1980, 102, 589.
(52) Cornil, J.; Vanderdonckt, S.; Lazzaroni, R.; dos Santos, D. A.; Thys,
G.; Geise, H. J.; Yu, L. M.; Szablewski, M.; Bloor, D.; Lo¨gdlund, M.;
Salaneck, W. R.; Gruhn, N. E.; Lichtenberger, D. L.; Lee, P. A.; Armstrong,
N. R.; Bre´das, J. L. Chem. Mater. 1999, 11, 2436.
(53) Hill, I. G.; Kahn, A.; Cornil, J.; dos Santos, D. A.; Bre´das, J. L.
Chem. Phys. Lett. 2000, 317, 444.
(36) Wang, H.; Helgeson, R.; Ma, B.; Wudl. F. J. Org. Chem. 2000,
65, 5862.
(37) Mu¨llner, R.; Winkler, B.; Stelzer, F.; Tasch, S.; Hochfilzer, C.;
Leising, G. Synth. Met. 1999, 105, 129.
(54) Cornil, J.; dos Santos, D. A.; Crispin, X. A.; Silbey, R.; Bre´das, J.
L. J. Am. Chem. Soc. 1998, 120, 1289.
(55) Bre´das, J. L.; Cornil, J.; Beljonne, D.; dos Santos, D. A.; Shuai, Z.
Acc. Chem. Res. 1999, 32, 267.