A Diels-Alder crosslinkable host polymer for improved PLED performance: The impact on solution processed doped device and multilayer device performance

Article abstract of DOI:10.1039/c2jm14591j

We report on the synthesis of a polyfluorene derivative, PFO(X), with furan pendant groups capable of Diels-Alder crosslinking with a maleimide containing small molecule passive crosslinker (PC) and a maleimide containing red emitting donor-acceptor-donor dopant molecule, bE-BTD(X). It was initially intended that a blend of these three components would afford a system where the dopant concentration could be increased to the point where complete energy transfer from the host polymer to the emissive dopant would be achieved. Because such systems often suffer from quenching and shifts in emission maxima indicative of emitter aggregation, it was hypothesized that crosslinking the emissive dopant with the host polymer would lead to de-aggregation of the dopant emitter. In thin films of PFO(X) and bE-BTD(X), a 16 nm bathochromic shift is observed in the emission maximum when the dopant concentration is increased from 1% to 8%, suggesting that the dopant is aggregating. In similar films where PC is included and the film is heated to affect crosslinking, a comparable 16 nm shift in the emission maximum is observed indicating that aggregation is still occurring and not affected by the heating step. Similar decreases in luminance are observed independent of whether the heating step is included. Not unexpectedly, however, crosslinking does afford an insoluble network that allows for the subsequent solution deposition of additional layers. When an electron transport layer (ETL) is used in PFO(X)/PC devices, increases of 190% and 490% are observed in luminance and luminous efficiency, respectively, relative to devices without an ETL indicating that this Diels-Alder crosslinkable system is amenable to multilayer deposition by solution methods. When bE-BTD(X) is included as the dopant emitter, similar increases in luminance and luminous efficiency are observed with the ETL included compared to devices where this layer is omitted.

Full text of DOI:10.1039/c2jm14591j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 3004
www.rsc.org/materials
PAPER
A Diels–Alder crosslinkable host polymer for improved PLED performance:
the impact on solution processed doped device and multilayer device
performance†
*
Dinesh G. (Dan) Patel, Kenneth R. Graham and John R. Reynolds
Received 15th September 2011, Accepted 3rd December 2011
DOI: 10.1039/c2jm14591j
We report on the synthesis of a polyfluorene derivative, PFO(X), with furan pendant groups capable of
Diels–Alder crosslinking with a maleimide containing small molecule passive crosslinker (PC) and
a maleimide containing red emitting donor–acceptor–donor dopant molecule, bE-BTD(X). It was
initially intended that a blend of these three components would afford a system where the dopant
concentration could be increased to the point where complete energy transfer from the host polymer to
the emissive dopant would be achieved. Because such systems often suffer from quenching and shifts in
emission maxima indicative of emitter aggregation, it was hypothesized that crosslinking the emissive
dopant with the host polymer would lead to de-aggregation of the dopant emitter. In thin films of PFO
(X) and bE-BTD(X), a 16 nm bathochromic shift is observed in the emission maximum when the
dopant concentration is increased from 1% to 8%, suggesting that the dopant is aggregating. In similar
films where PC is included and the film is heated to affect crosslinking, a comparable 16 nm shift in the
emission maximum is observed indicating that aggregation is still occurring and not affected by the
heating step. Similar decreases in luminance are observed independent of whether the heating step is
included. Not unexpectedly, however, crosslinking does afford an insoluble network that allows for the
subsequent solution deposition of additional layers. When an electron transport layer (ETL) is used in
PFO(X)/PC devices, increases of 190% and 490% are observed in luminance and luminous efficiency,
respectively, relative to devices without an ETL indicating that this Diels–Alder crosslinkable system is
amenable to multilayer deposition by solution methods. When bE-BTD(X) is included as the dopant
emitter, similar increases in luminance and luminous efficiency are observed with the ETL included
compared to devices where this layer is omitted.
amenable to the construction of large area devices. When
Introduction
considering that polymer based light emitting diodes (PLEDs)
are capable of displaying high brightness with relatively high
luminous efficiency they emerge as promising alternatives to the
classic inorganic devices.2,3
Conjugated polymer based light emitting devices have been
under intense, sustained investigation since the first observation
of light emission from a phenylene-vinylene polymer in 1990.1
Since then, both all-polymer devices and polymer-dopant
devices, systems in which a small molecule emitter is doped into
a polymer host matrix, have moved forward as attractive alter-
natives to the classic inorganic- and metal-based devices with
polymer materials offering several advantages. For example,
polymeric materials offer ease of synthesis, high tunability of
electronic properties, and a level of processability that is
Although PLEDs are a very attractive alternative lighting
source due to their ease of processability, they generally have
efficiencies which lag behind those of vapour deposited organic
light emitting diodes (OLEDs). This is attributed to two main
factors. First, the use of small molecules in vapour deposited
devices allows for higher purity materials to be obtained through
purification techniques such as sublimation,4 and second, the
vapour deposition process allows the facile fabrication of
multilayer device architectures.5,6 These multilayer devices
consist of electron transporting/hole blocking and hole trans-
porting/electron blocking layers sandwiching the emissive layer,
resulting in devices with external quantum efficiencies of
ꢀ20%.2,5,7,8 The use of appropriate materials leads to enhanced
charge injection and confinement of charges to the emissive layer,
thus leading to lower operating powers and reduced efficiency
The George and Josephine Butler Polymer Laboratories, Department of
Chemistry, Center for Macromolecular Science and Engineering,
University of Florida, Gainesville, Florida, 32611, USA. E-mail:
reynolds@chem.ufl.edu
† Electronic supplementary information (ESI) available: AFM images of
PFO(X) and PC films, synthetic details for compound 4 and PFOX(X),
NMR characterization of all new compounds. See DOI:
10.1039/c2jm14591j
3004 | J. Mater. Chem., 2012, 22, 3004–3014
This journal is ª The Royal Society of Chemistry 2012

View Article Online
losses due to leakage current and non-radiative charge recom-
bination at the electrodes.
aggregation and the ability to create insoluble films allowing for
solution processed multilayer devices. Additionally, the synthetic
approach to crosslinking employed does not require the use of
any additional initiators, high temperatures, or exposure to
potentially harmful UV irradiation as some previous approaches
have required.13
The realization of multilayer solution processed PLEDs is
significantly more difficult when compared to vapour deposited
OLEDs, however, because sequential layer deposition in PLEDs
requires either the use of orthogonal solvents9,10 to process the
different layers or that layers be rendered insoluble after depo-
sition. Due to the similar solubility of electronically active
organic materials the use of orthogonal solvents has seen more
limited use, the one ubiquitous exception being PEDOT:PSS,
which is spin cast from water and is insoluble in common organic
solvents thus allowing the solution processing of the active layer.
To date, the latter approach has been achieved through either
a loss of solubilizing groups11 or through post deposition cross-
linking.12,13 The loss of solubilizing groups was famously used in
the first demonstration of electroluminescence from poly(para-
phenylene vinylene).1 The use of crosslinking is more versatile,
and consequently it is emerging as a promising means of fabri-
cating multilayer solution processed devices as detailed in
a review by Marder and coworkers.13 Crosslinking can occur
thermally or be photo- or chemically initiated and proceed with
or without the formation of by-products. Styryl groups, for
example, have found use as cross-linkable groups in polymers
used in the emissive layer14,15 or hole transporting16 layer. Simi-
larly, trifluorovinylether functionalized polymers have been used
where a 2 + 2 cycloaddition upon heating gives a hexa-
fluorocyclobutane group, resulting in a loss in polymer solu-
bility.17,18 Some crosslinkable functional groups such as
oxetanes19–22 and siloxanes23 are known to leave side products
including photoacid initiator residue (for oxetanes) and water
(for siloxanes) which have detrimental effects on device perfor-
mance. The formation of insoluble polymer networks from
soluble and crosslinkable precursors, regardless of crosslinking
method, usually affords devices that have enhanced efficiency
over single layer devices.13,24
Results and discussion
Design strategy
Using the NLO work of Jen for inspiration,30,32–34 we envisioned
a system composed of a conjugated host polymer with pendant
furan groups and a dopant NIR emitter molecule containing the
complementary dienophile maleimide; a bis-maleimide passive
(non-emitting) crosslinker would also be employed to ensure
complete crosslinking as the dopant molecule would only be
included at a few weight percent. To the best of our knowledge
this Diels–Alder crosslinking approach has not yet been applied
to PLEDs and has the advantage that, as a part of the cross-
linking process, there are no by-products or reactive intermedi-
ates that would lead to degradation of the polymer or emitter.
Fig. 1 shows the crosslinking chemistry using maleimide and
furan as well our envisioned system with a generic crosslinkable
polymer and small molecule crosslinking agent. As with the NLO
systems, we hoped that the thermally induced crosslinking step
would simultaneously break up dopant molecule aggregates
while also ‘‘freezing’’ the dopant molecules in this non-aggre-
gated state (not shown in Fig. 1).
For the initial experiments, as detailed in this manuscript, we
chose to use a polyfluorene (PF) derivative as our host material.
The PF class of polymers presents several advantages including
a wide bandgap, ease of synthesis, and the ability to include the
required crosslinkable pendant groups. The wide band gap is
desirable as it should facilitate good energy transfer to the
emissive dopant oligomer. We elected to synthesize our PF host
Our group is interested in using conjugated polymer hosts with
red to near-IR (NIR) emitting dopant molecules.25–28 Using
€
either Forster energy transfer or charge trapping with the
appropriate polymer host allows for emission from the dopant
molecule.29 In order to obtain complete quenching of host
emission a sufficient amount of dopant must be added.25,26 At the
required dopant levels, unfortunately, small shifts in the emission
band and a decrease in device performance (as measured by
luminance and luminous efficiency) are both observed. We
believe that this phenomenon is the result of aggregation of the
dopant molecules, which in some instance has been observed by
atomic force microscopy (AFM).25 Therefore, it is necessary to
inhibit the aggregation of dopant molecules to achieve higher
efficiency devices.
As an analogy to host polymer-dopant emitter systems,
nonlinear optical (NLO) systems typically suffer losses in
responsivity from aggregation of strongly dipolar chromophores.
This problem has been addressed by Jen and coworkers30–32 using
a combination of heating, poling, and crosslinking through
a Diels–Alder reaction to afford devices with enhanced and
prolonged nonlinear optical responses. With this in mind, we
chose to apply a similar strategy to PLED architectures and
devices. The crosslinking approach we have taken was intended
to serve two primary purposes, including the reduction of
Fig. 1 Depiction of the Diels–Alder reaction between furan and mal-
eimide and the crosslinking that would occur in a film composed of
a crosslinkable polymer and a separate crosslinking agent.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 3004–3014 | 3005

View Article Online
hours, end capping is affected by first adding a small amount of
the commercially available pinacol ester of phenylboronic acid
followed by the addition of bromobenzene. The required
monomer 2 was obtained by the reaction of 2,7-dibromofluorene
(3) with 1,6-dibromohexane in aqueous sodium hydroxide with
tetrabutylammonium bromide functioning as a phase transfer
catalyst to give 4, which is reacted with deprotonated furfuryl
alcohol giving 2 in 48% yield. PFO(X) was purified by Soxhlet
extraction with methanol, hexane, and finally chloroform to
obtain the high molecular weight fraction used in devices.
Residual palladium was removed with the palladium scavenger
1
diethylammonium diethyldithiocarbamate. The H NMR spec-
trum and elemental analysis of PFO(X) are both consistent with
a polyfluorene containing two furan groups on every other repeat
unit. Additionally, gel permeation chromatography (GPC) data
shows that PFO(X) has a monomodal molecular weight distri-
bution with an Mn of 38,047 g molꢁ1 and a polydispersity index
(PDI) of 1.87.
Chart 1 Molecular structures of the host polymer PFO(X), maleimide
containing passive crosslinker PC, the red emitting oligomeric dopant
bE-BTD(X), and electron transport material TAZ.
While PFO(X) initially showed excellent solubility, after
a matter of weeks the solubility decreased as evidenced by
particulates that were insoluble in chlorobenzene solution, even
after heating and prolonged stirring. It is hypothesized that some
acidic impurities, known to be present in chloroform,38
contaminated the polymer sample after removal of solvent from
the final Soxhlet extraction. As furan is acid sensitive, these acidic
impurities may have lead to slow crosslinking between the
pendant groups. Though the level of crosslinking is suspected to
be low, it was sufficiently high to decrease the polymer solubility.
To test this hypothesis, a second quantity of PFO(X) was
synthesized and purified as before, the exception being that the
chloroform used for the final Soxhlet extraction was distilled
from potassium carbonate immediately before use thereby
removing any acidic impurities. This second batch has, over
a period of several months, maintained its excellent solubility.
The passive crosslinker PC was synthesized by slight modifi-
cation of a previously reported procedure.39 Briefly, furan and
maleimide are reacted to give their Diels–Alder adduct (FM-a in
Scheme 2). This adduct protects maleimide from any base
polymer with furfuryloxy pendant groups to give PFO(X) (Chart
1) given the precedence for a Suzuki polycondensation in the
presence of furanyl pendant groups.35 Additionally, the Suzuki
methodology is one of the most often utilized methods to achieve
PFs.36,37 To ensure effective crosslinking, PFO(X) is designed as
an alternating copolymer with the 9,9-difurfuryloxyhexyl
pendant groups appearing on every other fluorene unit.
Because maleimide is easily reacted in Sn2 fashion with alkyl
halide containing compounds, we chose to use it as the corre-
sponding dieneophile in both the passive crosslinker and NIR
dopant emitter. The small molecule 1,6-dimaleimidohexane
would serve as the passive crosslinker (PC, Chart 1). An emissive
donor–acceptor-donor dopant oligomer, bis-(6-maleimidohexyl)
ethylenedioxythiophene-benzothiadiazole (bE-BTD(X), Chart
1), is terminated with maleimidohexane groups that allow this
emitter to crosslink as well. For comparison purposes in our
study, we also present results obtained from a commercially
available poly(9,9-dioctylfluorene) purchased from American
Dye Source, PFO, which is unable to undergo crosslinking
reactions.
Synthesis
As shown in Scheme 1, PFO(X) is obtained by the coupling of
2,7-bispinocalatoboron-9,9-dioctylfluorene (1) and 2,7-dibromo-
9,9-di(6-furfuryloxyhexyl)fluorene (2) under Suzuki coupling
conditions. After the polymerization is allowed to proceed for 48
Scheme 2 Synthesis of crosslinkable, maleimide terminated donor–
acceptor-donor oligomer bE-BTD(X).
Scheme 1 Synthesis of PFO(X) and furan containing monomer 2.
3006 | J. Mater. Chem., 2012, 22, 3004–3014
This journal is ª The Royal Society of Chemistry 2012

View Article Online
promoted decomposition given that FM-a is reacted with 1,6-
diboromohexane in the presence of potassium carbonate in the
subsequent step. The resulting product is then heated to affect
a retro Diels–Alder reaction yielding PC.
supported by the surface roughness and crystalline domains
observed by AFM; not surprisingly, this crystallinity and
roughness is lost on heating of the films as shown in Supporting
Information (Figure S1†). Two additional transitions are also
observed in this first heating cycle, including an exotherm start-
ing at ꢀ110 ꢂC which corresponds to the Diels–Alder cross-
linking. This same feature is not observed on the second heating
cycle, thereby indicating that crosslinking is complete within the
detection limits of the instruments. Observed after the PC melt
transition is a broad endotherm that is indicative of the Diels–
Alder de-crosslinking process (retro Diels–Alder reaction).
During the second heat heating cycle this same transition appears
but is quite broad. Pure PFO(X) shows none of the features
observed in the PFO(X)/PC blend which supports our assign-
ment of the observed transitions.
The synthesis of bE-BTD(X) is outlined in Scheme 2.
Commercially available 3,4-ethylenedioxythiophene (5) is treated
with n-butyllithium followed by quenching of the anion with
Br-C6-OTHP to give 6. Compound 6 is then deprotonated with
n-butyllithium and reacted with trimethyltin chloride to give
stanylated compound 7. A Stille coupling between 4,7-dibro-
mobenzo[c]-1,2,5-thiadiazole (8) and 7 gives tetrahydropyranyl
(THP) protected 9. THP protecting group removal is achieved in
methanol with catalytic hydrochloric acid to afford dialcohol 10,
which is tosylated to give 11. The tosyl group is displaced with
FM-a under basic conditions and the crude material heated in
toluene to drive the retro-Diels–Alder reaction giving bE-BTD
(X) in 47% yield after column chromatography. It should be
noted that attempts to isolate bE-BTD(X) without a heating step
lead to complicated product mixtures owing to the presence of
endo and exo isomers of the FM-a component, which have
different retention factors.
Spectral characterization
The spectral characteristics of PFO(X) were compared to
commercially available PFO. Both materials exhibit nearly
identical absorption and emission spectra, with lmax abs at 388
nm and lmax PL at 416 nm for both materials in chloroform as
shown in Fig. 3. This suggests that the furan pendant groups do
not significantly influence the electronic structure of the poly-
fluorene backbone through any sort of p–p interactions at the
concentrations examined. Thus, we believe that PFO(X) will
behave similarly to non-crosslinkable PFO in our device studies.
To study the effects of the furan pendant groups and the
presence of PC on the solid-state behaviour of PFO(X), we
examined the film photoluminescence spectrum of PFO(X) both
before and after crosslinking with PC and compared it to that of
commercially available PFO (Fig. 4). Films were spin cast onto
PEDOT:PSS coated glass substrates with PFO(X):PC
(1.00 : 0.29 weight ratio) solutions in chlorobenzene, at which
there is a 1 : 1 mole ratio of furan to maleimide. Thermal treat-
ment was carried out in an Ar atmosphere glovebox (typically
<0.1 ppm O2 and H2O) at 130 ꢂC for ten minutes to induce
crosslinking. The PFO solutions were spin cast from chloro-
benzene as well, but were not subjected to the thermal annealing
step. A 100 nm thick layer of aluminum was deposited on top of
the polymer layer to protect the films from atmospheric water
Thermal characterization
Thermogravimetric analysis (TGA) and differential scanning
calorimetery (DSC) were used to examine pure PFO(X), pure
bE-BTD(X), and a mixture of PFO(X) and PC. We observe by
TGA that a mixture of PFO(X) and PC (1 : 1 maleimide:furan) is
stable up to 190 ꢂC while bE-BTD(X) is stable up to 300 ꢂC,
thereby indicating that thermal decomposition would not be
a problem during device operation. Additionally, these experi-
ments show that crosslinking takes place in the solid state and the
data are provided in the Supporting Information.†
The DSC experiments show that pure PC has a melting tran-
ꢂ
sition at 141 C, while pure bE-BTD(X) shows a melting transi-
tion at 128 ꢂC. This PC melting transition is observed in the first
heating cycle when a mixture of PFO(X) and PC (1 : 1 furan/
ꢂ
maleimide) is heated from 20 to 180 C (Fig. 2). This suggests
that some crystalline domains of pure PC exist in the film and is
Fig. 2 TGA scan of the first and second heating cycles of a PFO(X):PC
mixture and the second heating cycle of PFO(X) without crosslinker
present.
Fig. 3 Normalized solution absorption and fluorescence spectra of PFO
(X) and commercially available PFO in chloroform.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 3004–3014 | 3007

View Article Online
defined, presumably due to increased ordering imparted by the
thermal treatment. It is also worth mentioning that the PFO film
is thicker than the PFO(X)/PC films, with an absorbance of 0.91
vs. 0.36 at 380 nm respectively. The increased thickness of the
PFO film may result in more self-absorption, thereby increasing
the relative intensity of the longer wavelength emission.47
Initial device studies
With the solid-state photoluminescence studies indicating that
PFO(X)/PC films behave similarly to commercially available
PFO, we turned our attention to electroluminescence studies on
films of PFO(X) both with and without PC, and before and after
heating and crosslinking, in a simple device architecture con-
sisting of ITO/PEDOT:PSS/polymer/LiF/Ca/Al. These electro-
luminescence spectra are presented in Fig. 5. Comparisons are
made with an analogous PFO device. When biased at 8 V, all
films showed strong electroluminescence. Commercially avail-
able PFO has emission maxima at lmax EL ¼ 435 nm and 466 nm
typical of PFs;48 the longer wavelength emission between 500 and
600 nm is indicative of aggregation, excimer formation, and/or
keto defects as observed in the solid state photoluminescence
spectrum in our experiments and as reported previously.40,42
PFO(X) without PC exhibits a blue shifted (11 nm, main peak)
electroluminescence spectrum relative to commercially available
PFO and decreased emission intensity at longer wavelengths. We
attribute this observation to the presence of the furan pendant
groups, which decrease intermolecular interactions between the
PF backbones while also possibly shortening the effective
conjugation length due to increased side chain bulkiness.41,43 The
electroluminescence spectrum of a PFO(X)/PC film shows
a further decrease in emission at longer wavelengths when
compared to pure PFO(X). This suggests that PC plays a role in
breaking up aggregation in PFO(X), thus decreasing excimer
formation. Alternatively, the electron deficient maleimide groups
in PC could help inhibit the formation of long wavelength
emitting keto defects as was observed when PFO is blended with
electron accepting oxadiazole containing oligomers and small
molecules.49 After heating the PFO(X)/PC film a slight blue shift
in emission occurs (6 nm, main peak) relative to the unheated
Fig. 4 Film photoluminescence of commercially available PFO and
PFO(X)/PC before and after heating and crosslinking.
and oxygen during photoluminescence measurements, which
were performed outside of the glovebox.
Spectral analysis of PFO(X)/PC film emission, presented in
Fig. 4, shows small differences when compared to the solution
emission spectrum of both PFO and PFO(X), but most impor-
tantly demonstrates that the heating and crosslinking processes
do not lead to a significant modification in the solid-state emis-
sion from the PFO(X)/PC blend. The solid-state photo-
luminescence spectra of PFO(X) and commercially available
PFO do show some differences, with PFO showing broader
emission with a more intense 0–1 transition (448 nm) relative to
PFO(X). This PFO emission spectrum is similar to that of
a dihexyl-substituted PF and other PF derivatives previously
reported.40–43 In a study by Bliznyuk and coworkers,40 the longer
wavelength band was attributed to emission from excimers
formed as a result of PF aggregation. This aggregation seems to
be minimized in PFO(X), which is likely attributed to the furan
pendant groups disrupting intermolecular packing and hence
limiting p–p interactions in the fluorene backbone. This
decreased aggregation results in an emission spectrum that is
more similar to that of the solution spectra and is consistent with
a report by Setayesh43 and co-workers, who examined a PF
substituted with dendron side chains. Crystal structure analysis
of a model compound and the emission spectrum suggest that
aggregation is prevented by the dendrons, which act as shields to
prevent p–p interactions between neighbouring PF chains.
Keto-defects have also been implicated in the longer wavelength
emission bands of PFs, and a report by Bradley and coworkers
showed the necessity of close contacts (fluorenone-fluorenone
dipole–dipole interaction) for this emission in a series of fluorene
oligomers containing fluorenone ‘‘defects’’.44 Work by Veinot45,46
has shown that polyaromatic ethers serve to reduce the dipole–
dipole interaction in PF films. By analogy we believe that the
alkyloxyfuran side groups in PFO(X) and maleimide in PC can
behave in much the same manner, with the PFO(X) backbone
interacting preferably though p-interactions with either furan or
maleimide and through dipole–dipole interactions with mal-
eimide thus minimizing the unwanted longer wavelength emis-
sion. The overall spectral effect is a more blue emission from
PFO(X)/PC films. Additionally, after thermal treatment the
peaks/shoulders due to vibrational structure43 become better
Fig. 5 Electroluminescence spectra of PFO (black), PFO(X) (red), and
PFO(X)/PC films with (green) and without (blue) a thermal crosslinking
step.
3008 | J. Mater. Chem., 2012, 22, 3004–3014
This journal is ª The Royal Society of Chemistry 2012

View Article Online
films, which is contrary to the expected red-shift in emission
commonly observed after thermal treatment.41 This blue shift
was not observed in the PL spectrum after thermal treatment and
its origin is currently unclear.
bE-BTD(X) doped devices
Our original hypothesis was that a thermal treatment and
crosslinking step would reduce or eliminate the aggregation of
dopant emitter molecules in PFO(X) devices. Towards that
goal, we examined the electroluminescence profiles of control
devices consisting of PFO(X) with bE-BTD(X), no PC, and no
thermal treatment to induce crosslinking. These control devices
were then compared with devices containing PFO(X), bE-BTD
(X), and PC in which the films were thermally treated to induce
crosslinking and thereby potentially reduce aggregation. The
amount of bE-BTD(X) was varied from 1 to 8 weight percent
and the electroluminescence spectra examined to determine
whether or not the heating and crosslinking step would serve to
break up aggregation of the dopant emitter. The device archi-
tecture consisted of ITO/PEDOT:PSS/active layer/LiF/Ca/Al,
where the active layer consists of bE-BTD(X):PFO(X):PC or
bE-BTD(X):PFO(X).
The electroluminescence spectra of the control devices are
shown in Fig. 6a. Observed is effective but incomplete energy
transfer from PFO(X) to bE-BTD(X) as shown by the strong
emission from bE-BTD(X) in the red region of the spectrum.
Upon increasing the amount of bE-BTD(X) from 1 to 8 weight
percent, a 16 nm bathochromic shift in the emission maximum of
bE-BTD(X) from 636 to 652 nm is observed. Lower concentra-
tions of bE-BTD(X) were not applied as these would have
resulted in increased PFO(X) emission. The bathochromic
spectral shift suggests the formation of aggregates at higher
dopant concentrations. With increasing concentrations,
a decrease in PFO(X) emission is also observed, with minimal
PFO(X) emission at 8 weight percent bE-BTD(X). The 1% bE-
BTD(X) device displays a maximum luminance of 900 cd mꢁ2
,
whereas the device with 8% bE-BTD(X) displays a maximum
luminance of only 220 cd mꢁ2 as evident in Fig. 6b. The luminous
efficiency (hL) data (Fig. 6c) displays a similar trend as the
luminance, with the maximum hL decreasing from 0.33 cd Aꢁ1
for the 1% device to 0.12 cd Aꢁ1 for the 8% bE-BTD(X) device.
Given the bathochromic shift observed in the EL spectra upon
increasing bE-BTD(X) concentration, it is likely that the
decreased device performance is due to aggregate quenching.26
The turn on voltages increase from 4.75 V for the 1% bE-BTD(X)
device to 6.0 V for the 8% bE-BTD(X) device. In addition, the
current density decreases as the concentration of bE-BTD(X)
increases as shown in SI Figure S15.† This decrease in current
likely results from bE-BTD(X) or bE-BTD(X) aggregates acting
as sites for charge trapping in the device.50
Fig. 6 Electroluminescence spectra (a), luminance-voltage plots (b), and
luminous efficiency vs. current density plots (c) for bE-BTD(X) doped
PFO(X) films without PC.
electroluminescence spectra, luminance, and luminous efficiency
data is presented in Fig. 7.
In an attempt to prevent the spectral modification and
decrease in device performance that occurs with increasing
dopant concentration, we examined the effects of crosslinking on
bE-BTD(X) doped PFO(X) devices. These devices, after the
active layer was spin cast, were heated to 130 ꢂC for ten minutes
in an argon atmosphere glovebox to induce crosslinking. As in
the control devices where PC was omitted, the concentration of
emitter was varied from 1 to 8 weight percent and the
As in the devices without PC, there is a decrease in PFO(X)
emission with increasing concentration of bE-BTD(X), but the
dopant emitter does not completely quench emission from PFO
(X) at the concentrations employed. There is also a bathochromic
shift in the emission maximum of bE-BTD(X), with lmax EL
629 nm at 1% bE-BTD(X) and lmax EL ¼ 645 nm at 8% bE-BTD
(X), which is a difference of 16 nm just as in the devices without
¼
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 3004–3014 | 3009

View Article Online
increasing bE-BTD(X) dopant concentration, with nearly iden-
tical luminance maxima as observed for the device with no PC as
shown in Fig. 6b. The hL data for the crosslinked devices shown
in Fig. 7c indicates the same trend of decreasing hL with
increasing bE-BTD(X) concentration; however, the hL values
remain lower at all concentrations when compared to the device
without PC (0.18 cd Aꢁ1 and 0.33 cd Aꢁ1 for the 1% bE-BTD(X)
devices with and without PC, respectively). The turn on voltages
increase from 5.3 V to 6.0 V as the concentration of bE-BTD(X)
is increased from 1 to 8%. As with the devices with no PC, the
current density decreases as the concentration of bE-BTD(X) is
increased (SI Figure S16†).
Multilayer devices with PFO(X)
The fabrication of multilayer, all-solution processed organic light
emitting diodes is an important topic in the field of organic
electronics.13 It is known that single layer devices usually offer
poorer performance relative to multilayer devices, where dedi-
cated hole transport/electron blocking and electron transport/
hole blocking materials can increase the probability that charge
recombination occurs in the desired layer.24 Based on our
observation that PFO(X) can crosslink efficiently to give an
insoluble network allowing for the deposition of subsequent
layers, we examined its performance as a blue emitter in
a multilayered device constructed using the commonly employed
electron transport material 3-(4-biphenylyl)-4-phenyl-5-(4-tert-
butylphenyl)-1,2,4-triazole (TAZ, Chart 1) in PFO at a ratio of
4 : 1 by weight.51,52 In this device the active layer, consisting of
PFO(X) and PC, was spin cast onto the PEDOT:PSS coated
substrate, thermally treated to induce cross-linking, and the
TAZ:PFO electron transport layer (ETL) spun cast on top.
The device was completed by thermal vapour deposition of the
LiF/Ca/Al cathode.
To examine whether the addition of an ETL could offer
enhanced performance relative to the single layer device, we first
assembled devices with PFO(X)/PC as the active layer both with
and without the ETL. The luminance and luminous efficiency
data for these devices are shown in Fig. 8. While the single layer
PFO(X)/PC device shows a lower turn-on voltage, 5.5 V
compared to 7.8 V with the ETL, the PFO(X)/PC device with
the ETL shows greater luminance at nearly all voltages
(Fig. 8a). The higher turn on voltage is expected given that the
multilayer device is thicker, thus resulting in a decreased electric
field in the device at a specific applied voltage. The maximum
luminance reached in the device with the ETL is 2280 cd mꢁ2 as
compared to only 780 cd mꢁ2 for the device with no ETL.
Concomitant with the increase in luminance is a greater than
five-fold increase in luminous efficiency, 1.41 compared to 0.24
cd Aꢁ1 upon addition of the TAZ:PFO ETL as shown in
Fig. 8b.
Fig. 7 Electroluminescence spectra (a), luminance-voltage plots (b), and
luminous efficiency vs. current density plots (c) for devices containing
PFO(X), bE-BTD(X), and PC in the active layer. Crosslinking was
affected by heating the devices at 130 ꢂC for ten minutes after spin
coating.
The effect of an ETL on the emission of the bE-BTD(X) doped
devices was probed by assembling devices with PFO(X)/PC
active layers containing 1 to 8 weight percent of bE-BTD(X). The
active layer films were thermally crosslinked as previously
described, followed by solution deposition of the TAZ:PFO ETL
and completed by thermal vapour deposition of a LiF/Ca/Al
cathode. The data obtained from these devices are shown in
Fig. 9.
PC and no crosslinking (Fig. 7a). The shift in the emission
maxima with increasing bE-BTD(X) concentration suggests that
crosslinking either does not break up aggregates or is not suffi-
cient to keep aggregates from forming as the device cools during
the crosslinking step. This supposition is supported by the
luminance data as presented in Fig. 7b. This data shows that even
after heating and crosslinking the luminance decreases with
3010 | J. Mater. Chem., 2012, 22, 3004–3014
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Fig. 8 Luminance (a) and luminous efficiency (b) data for PFO(X)
devices with and without a TAZ:PFO ETL.
The electroluminescence spectra (Fig. 9a) show incomplete
quenching of emission from the host polymer, PFO(X) or PFO,
as in the case of the devices with no ETL; however, when
compared to devices without an ETL, the devices with an ETL
show greater host emission. This is likely due to emission from
PFO in the ETL or a shift in the recombination zone to the active
layer:ETL interface where the relative concentration of bE-BTD
(X) may be lower. The luminance and hL data presented in
Fig. 9b and 9c clearly show that the devices with ETL layers have
increased luminance and hL with increases from 870 cd mꢁ2 and
0.18 cd Aꢁ1 to 2530 cd mꢁ2 and 1.8 cd Aꢁ1 for the devices without
and with an ETL respectively. This corresponds to a ꢀ3 fold
increase in the luminance and an order of magnitude increase in
hL upon addition of the ETL.
Fig. 9 Electroluminescence spectra (a), luminance-voltage plots (b), and
luminous efficiency vs. current density plots (c) for crosslinked bE-BTD
(X) doped PFO(X)/PC devices with a TAZ:PFO ETL.
TAZ in the ETL.51,52 The observed higher luminance and hL
figures, along with the decreased leakage and operating currents,
supports that the ETL helps facilitate charge recombination in
the active layer. Importantly, the deposition of the ETL without
disrupting the active layer was only made possible through the
insolubility of the active layer afforded by crosslinking, some-
thing that is not possible with standard PFO and other non-
crosslinkable PFs.
As was observed for the PFO(X)/PC devices upon addition of
an ETL, the turn on voltage increases from 5.3 V to 7.8 V for the
1% bE-BTD(X) devices upon addition of the ETL. In accor-
dance with the higher hL, the devices with ETLs display lower
leakage currents prior to turn on and decreased operating
currents relative to the devices with no ETL (SI Figure S16†).
This trend can be explained by the hole blocking nature of the
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 3004–3014 | 3011

View Article Online
Fabrication and characterization of PLEDS
Conclusions
The multilayer PLEDs were fabricated on 25 by 25 mm pre-
patterned indium tin oxide (ITO) coated glass substrates with
a sheet resistance of 20 U/,. The substrates were cleaned in
ultrasonic baths of sodium dodecyl sulfate in deionized water,
deionized water, acetone, and isopropyl alcohol consecutively for
15 min each and then exposed to oxygen plasma for 20 min.
Immediately following the substrates were spin coated with a 40
nm thick layer of PEDOT:PSS (Baytron P VP Al4083) and dried
To summarize, a PF host polymer with furan pendant groups has
been synthesized and interfaced with a maleimide terminated
passive crosslinker (PC) and donor–acceptor-donor red emitter
bE-BTD(X) dopant molecule. Crosslinking and decrosslinking
are observed by DSC experiments. We find that heating and
concomitant Diels–Alder crosslinking does not lead to reduced
aggregation of bE-BTD(X) as evidenced by EL spectra.
However, we report that crosslinking does allow for the forma-
tion of an insoluble network permitting the solution deposition
of a subsequent electron transport layer. While the resulting
multilayered devices show increased PFO emission, they also
show enhanced luminance and luminous efficiency. This
approach to the fabrication of multilayered structures has
implications for PLED assembly and could lead to all-solution-
processed multilayer devices with efficiencies equal to that of
vapour deposited OLEDs currently available.
ꢂ
at 130 C on a hotplate in an argon atmosphere glovebox (<0.1
ppm O2 and H2O). The active layer was deposited in the glove-
box through spin coating a 7 mg mLꢁ1 PFO or PFO(X) solution
in deoxygenated anhydrous chlorobenzene (Sigma Aldrich) or
a 9.5–10 mg mLꢁ1 (total weight) PFO(X) + PC + bE-BTD(X)
solution in deoxygenated anhydrous chlorobenzene. For all
devices individual solutions containing only one material were
prepared and stirred overnight, with the PFO and PFO(X)
solutions stirred at 50 ꢂC to facilitate dissolution and PC and bE-
BTD(X) solutions stirred at room temperature. The solutions
were then combined and stirred without heating for a maximum
of 2 h to avoid crosslinking in solution prior to spin coating. In
crosslinked devices the concentrations were adjusted such that
the total number of maleimides on bE-BTD(X) and PC combined
were equal to the number of furan groups. The devices which
were thermally treated were placed on a 130 ꢂC hotplate in the Ar
atmosphere glovebox for 10 min. In the multilayer devices the
substrates were allowed to cool following heating and then spin
cast with a TAZ:PFO (4 : 1 by weight) layer with a total
concentration of 12.5 mg mLꢁ1. All solutions were filtered with
0.45 mm PTFE filters using glass syringes prior to spin coating. A
1 nm layer of LiF, 10 nm layer of Ca, and a 100 nm layer of Al
were thermally evaporated at a pressure of 1 ꢃ 10ꢁ6 mbar
through shadow masks to create the cathodes. The pixel areas
were 7.07 mm2 with each substrate featuring 8 independently
addressable pixels. Voltage, current density, luminance, and
luminous efficiency characteristics of the PLEDs were measured
under ambient conditions using a custom built LabVIEW
program with a Keithley 2400 source meter and a UDT Instru-
ments optometer with a calibrated Si photodiode. In combina-
tion with the device spectra and cross-calibration with a Minolta
CS100 luminance meter, the current measured from the Si
photodiode was converted to luminance. EL spectra were taken
with an ISA SPEX Triax 180 spectrograph with the devices
driven at a constant current using a Keithley 2400 source meter.
Luminance and luminous efficiency values reported are the
average of 5–6 pixels.
Experimental section
Materials and methods
Reagents were obtained from Sigma-Aldrich and purified
according to previously published methods before use.38 Furfuryl
alcohol was distilled under vacuum from anhydrous potassium
carbonate immediately before use; the sodium hydride used
was a 60 wt. % dispersion in mineral oil. Compounds 1,53 3,54,55
4,56 6–8,25 and FM-a57 were synthesized according to published
procedures. PFO(X) was synthesized using usual conditions for
the Suzuki polycondensation reaction and the details are given
in the supporting information.† All NMR spectroscopic
experiments were performed on a Varian Gemini 300 Spec-
1
trometer operating at 300 MHz for H nuclei and 75 MHz for
13C nuclei. Chemical shifts are referenced to CDCl3 internal
standard (7.26 ppm for 1H and 77.0 ppm for 13C). Solution
UV-Vis and fluorescence measurements were performed in
chloroform on a PerkinElmer Lambda 25 UV-Vis spectrometer
and Jobin-Yvon Fluorolog-3 spectrofluorimeter. Elemental
analyses were carried out in house. Gel permeation chroma-
tography experiments were conducted on a Waters Associates
GPCV2000 liquid chromatography system with an internal
differential refractive index detector and two Waters Styragel
HR-5E columns (10 mm PD, 7.8 mm i.d., 300 mm length) using
THF as the mobile phase. Differential scanning calorimetry
(DSC) was performeꢂd using a Thermal Analysis (TA) Q1000 at
a heating rate of 10 C minꢁ1 under helium purge. Calibrations
were made using indium and freshly distilled n-octane as the
standards for peak temperature transitions and indium for the
enthalpy standard. All samples were prepared in hermetically
sealed pans and were run using an empty pan as a reference.
The samples were scanned for multiple cycles to remove
recrystallization differences between samples, and the results
were reported from the second scan cycle. Thermogravimetric
analysis (TGA) was performed on a TA Q5000 using the
dynamic high-resolution analysis. PFO was purchased from
American Dye Source with POSS end groups (ADS BE229)
and stored in argon atmosphere glovebox (typically <0.1 ppm
O2 and H2O) until use.
Synthetic details
Compound 9. In an oven-dried flask fitted with a water-cooled
reflux condenser was combined compound 7 (1.017 g, 2.08
mmol), compound 8 (0.175 g, 0.59 mmol), and THF. The reac-
tion was degassed by three freeze-pump-thaw cycles, and then Pd
(PPh3)2Cl2 (0.020 g, 0.03 mmol) was added followed by a single
freeze-pump-thaw cycle. After backfilling with nitrogen and
warming to room temperature the reaction was refluxed over-
night. Upon cooling to room temperature the reaction was
diluted with methylene chloride (100 mL) and washed with
3012 | J. Mater. Chem., 2012, 22, 3004–3014
This journal is ª The Royal Society of Chemistry 2012

View Article Online
saturated, aqueous NaHCO3 (2 ꢃ 30 mL), brine (30 mL), then
dried over MgSO4. After filtration and removal of solvent under
reduced pressure the remaining dark residue was chromato-
graphed on silica eluting with a methylene chloride/ethyl acetate
gradient (0 to 4% ethyl acetate). The major compound was iso-
lated and upon removal of solvent a dark, maroon solid was
to remove the furan protecting group. The solvent was removed
under high vacuum and the remaining solid chromatographed
on silica eluting with
a methylene chloride/ethyl acetate
gradient (0 to 8% ethyl acetate). The major eluted spot was
collected and afforded a dark red solid upon removal of solvent
(0.043 g, 40%). 1H NMR (300 MHz, CDCl3) d 8.32 (s, 2H), 6.66
(s, 4H), 4.36 (m, 4H), 4.29 (m, 4H), 3.51 (t, 4H, J ¼ 7.4 Hz),
2.72 (t, 4H, J ¼ 7.4 Hz); 13C NMR (75 MHz, CDCl3) d 170.8,
152.2, 140.1, 137.6, 133.9, 126.1, 123.2, 120.5, 109.7, 65.0, 64.3,
37.9, 30.2, 28.7, 26.5, 25.9. HRMS (ESI) calculated for
C38H38N4O8S3 [M]+ 774.1846, found m/z 774.1832. Anal.
Calcd.: C 58.90; H, 4.94; N, 7.23; found C, 58.85; H, 4.86; N,
6.92.
1
obtained (0.420 g, 90%). H NMR (300 MHz, CDCl3) d 8.33 (s,
2H), 4.58 (m, 2H), 4.38 (m, 4H), 4.29 (m, 4H), 3.87 (m, 2H), 3.75
(m, 1H), 3.72 (m, 1H), 3.50 (m, 2H), 3.41 (m, 1H), 3.37 (m, 1H),
2.74 (t, 4H, J ¼ 7.7 Hz), 1.82 ꢁ 1.43 (m, 28H); 13C NMR (75
MHz, CDCl3) d 152.2, 140.0, 137.5, 126.1, 123.1, 120.8, 109.6,
98.8, 67.5, 65.0, 64.2, 62.2, 30.8, 30.4, 29.6, 29.0, 26.0, 25.5, 19.7.
HRMS (ESI) calculated for C40H53N2O8S3 [M + H]+ 785.2959,
found m/z 785.2947.
1,6-bismaleimidohexane (PC).39 In a flask was combined FM-
a (1.000 g, 6.06 mmol) and K2CO3 (1.190 g, 8.64 mmol). The
solids were placed under vacuum for 1 h, and then the flask was
backfilled with nitrogen. DMF (15 mL) and 1,6-dibromohexane
(0.44 mL, 2.88 mmol) were added and the reaction heated to 40
ꢂC overnight. The now reddish reaction was cooled to room
temperature and poured into methylene chloride (200 mL) and
washed with water (3 ꢃ 100 mL). The organic phase was
collected and dried over MgSO4, filtered, and solvent removed
under reduced pressure to obtain a red coloured solid that was
refluxed in toluene overnight. Upon removal of toluene under
high vacuum the remaining solid was purified on a short column
of silica eluting with methylene chloride to give a white, crys-
Compound 10. To a flask containing 9 (0.375 g, 0.48 mmol) was
added methanol (4 mL) and dichloromethane (10 mL). With
stirring, concentrated hydrochloric acid (2 mL) was added
dropwise. The reaction was allowed to stir at room temperature
overnight under nitrogen, then poured into methylene chloride
(75 mL) and washed with saturated, aqueous sodium bicar-
bonate (2 ꢃ 50 mL) and brine (50 mL). The organic phase was
then dried over MgSO4, filtered, and solvent removed under
reduced pressue to afford a dark maroon solid that was spec-
1
troscopically pure (0.283 g, 96%). H NMR (300 MHz, CDCl3)
d 8.33 (s, 2H), 4.37 (m, 4H), 4.30 (m, 4H), 3.65 (m, 4H), 2.74
(t, 4H, J ¼ 7.5 Hz), 1.72 (m, 4H), 1.59 (m, 4H), 1.43 (m, 8H), 1.24
(m, 2H). HRMS (ESI) calculated for C30H37N2O6S3 [M + H]+
617.1808, found m/z 617.1814.
1
talline solid after removal of solvent (0.422 g, 53%). H NMR
(300 MHz, CDCl3) d 6.68 (s, 4H), 3.50 (t, 4H, J ¼ 7.2 Hz), 1.58
(m, 4H), 1.28 (m, 4H).
Compound 11. A round bottom flask was charged with 10
(0.110 g, 0.18 mmol) and anhydrous pyridine (2 mL). The
ꢂ
2,7-dibromo-9,9-bis(6-furfuryloxyhexyl)-9H-fluorene (2). In
a 200 mL Schlenk flask was combined DMF (30 mL) and fur-
fuꢂryl alcohol (1.99 mL, 23.1 mmol). The solution was cooled to
0 C and NaH (60 weight % suspension in mineral oil, 0.923 g,
23.1 mmol) was added in small portions over approximately 30
min. The deprotonation reaction was allowed to stir for an
additional 30 min, then compound 4 (2.500 g, 3.85 mmol) was
added in one portion. The reaction was stirred for 30 min at 0 ꢂC,
solution was cooled to 0 C and toluenesulfonyl chloride (0.136
g, 0.71 mmol) was added in a single portion. The reaction was
stirred at 0 ꢂC for 2.5 h, then stored at ꢁ18 ꢂC overnight at
which point TLC showed no starting material. The dark red
solution was diluted with ether (150 mL) and washed with
aqueous HCl (1M, 4 ꢃ 50 mL) and brine (50 mL). The ether
phase was then dried over MgSO4, filtered, and solvent removed
under reduced pressure to give a dark red solid that was further
purified on a silica column eluting with methylene chloride.
Upon removal of solvent a dark red solid was obtained (0.134 g,
ꢂ
then heated to 50 C overnight. Upon cooling to room temper-
ature the reaction was poured into water (100 mL) and extracted
with methylene chloride (4 ꢃ 50 mL). The combined organic
phases were washed once with brine (50 mL) and dried over
MgSO4, filtered, and solvent removed under reduced pressure to
give a grey-brown oil. The crude product was adsorbed onto
silica and chromatographed on silica eluting with a hexanes/ethyl
acetate gradient (0 to 15% ethyl acetate). The second eluted
compound was collected as a clear, colourless oil upon removal
1
81%). H NMR (300 MHz, CDCl3) d 8.34 (s, 2H), 7.78 (d, 4H,
J ¼ 8.1 Hz), 7.33 (d, 4H, J ¼ 8.1 Hz), 4.37 (m, 4H), 4.29 (m,
4H), 4.02 (t, 4H, J ¼ 6.5 Hz), 2.69 (t, 4H, J ¼ 7.5 Hz), 2.43 (s,
6H), 1.65 (m, 8H), 1.35 (m, 8H); 13C NMR (75 MHz, CDCl3)
d 152.2, 144.6, 137.6, 133.0, 129.7, 127.7, 126.1, 123.1, 120.3,
109.6, 70.5, 65.0, 64.2, 30.0, 28.7, 28.3, 25.7, 25.0, 21.6. HRMS
(ESI) calculated for C44H49N2O10S5 [M + H]+ 925.1985, found
m/z 925.1944.
1
of solvent (1.714 g, 65%). H NMR (300 MHz, CDCl3) d 7.51
(dd, 2H, J ¼ 0.5 Hz, 8.0 Hz), 7.44 (dd, 2H, J ¼ 1.7 Hz, 8.0 Hz),
7.42 (dd, 2H, J ¼ 0.5 Hz, 1.7 Hz), 7.37 (dd, 2H, J ¼ 0.8 Hz, 1.8
Hz), 6.32 (dd, 2H, J ¼ 1.8 Hz, 3.2 Hz), 6.27 (dd, 2H, J ¼ 0.5 Hz,
3.2 Hz), 4.37 (s, 4H), 3.34 (t, 4 H, J ¼ 6.6 Hz), 1.89 (m, 4H), 1.40
(m, 4H), 1.08 (m, 8H), 0.58 (m, 4H); 13C NMR (75 MHz, CDCl3)
d 152.2, 151,9, 142.4, 138.8, 130.0, 125.9, 121.3, 121.0, 110.0,
108.8, 70.0, 64.5, 55.5, 40.0, 29.5, 29.3, 25.6, 23.5. HRMS (ESI)
calculated for C35H41Br2O4 [M + H]+ 685.1349, found m/z
685.1361. Anal. Calcd.: C 61.41; H, 5.89; N, 0.00; found C, 61.50;
H, 5.70; N, 0.00.
bE-BTD(X). A flask was charged with 11 (0.130 g, 0.14
mmol), FM-a (0.093 g), and K2CO3 (0.078 g) and cooled in an
ice bath. Anhydrous DMF (2 mL) was added and the reaction
stirred while slowly warming to room temperature overnight,
then poured into methylene chloride (100 mL) and washed with
water (3 ꢃ 50 mL). The organic phase was dried over MgSO4,
filtered, and solvent removed under reduced pressure to give
a red solid that was dissolved in toluene and refluxed overnight
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 3004–3014 | 3013

View Article Online
26 J. R. Sommer, R. T. Farley, K. R. Graham, Y. Yang, J. R. Reynolds,
J. Xue and K. S. Schanze, ACS Appl. Mater. Interfaces, 2009, 1, 274–
278.
Acknowledgements
The authors gratefully acknowledge the financial support of the
U.S. Army (project No. W31P4Q-08-1-0003). The views and
conclusions contained in this document are those of the authors
and should not be interpreted as representing the official policies,
either expressed or implied, of the U.S. Army or the U.S.
Government. D.G.P. would like to thank Paula Delgado and
Bora Inci for help with the thermal analysis experiments.
27 Y. Yang, R. T. Farley, T. T. Steckler, S.-H. Eom, J. R. Reynolds,
K. S. Schanze and J. Xue, Appl. Phys. Lett., 2008, 93, 163305.
28 Y. Yang, R. T. Farley, T. T. Steckler, S.-H. Eom, J. R. Reynolds,
K. S. Schanze and J. Xue, J. Appl. Phys., 2009, 106, 044509.
29 V. Cleave, G. Yahioglu, P. Le Barny, D. H. Hwang, A. B. Holmes,
R. H. Friend and N. Tessler, Adv. Mater., 2001, 13, 44–47.
30 T.-D. Kim, J. Luo, J.-W. Ka, S. Hau, Y. Q. Tian, Z. W. Shi,
N. M. Tucker, S. H. Jang, J.-W. Kang and A. K.-Y. Jen, Adv.
Mater., 2006, 18, 3038–3042.
31 T.-D. Kim, J. Luo, Y. Tian, J.-W. Ka, N. M. Tucker, M. Haller,
J.-W. Kang and A. K.-Y. Jen, Macromolecules, 2006, 39, 1676–1680.
32 J. Luo, M. Haller, H. Li, T.-D. Kim and A. K.-Y. Jen, Adv. Mater.,
2003, 15, 1635–1638.
33 M. Haller, J. Luo, H. X. Li, T.-D. Kim, Y. Liao, B. H. Robinson,
L. R. Dalton and A. K.-Y. Jen, Macromolecules, 2004, 37, 688–690.
34 Z. Shi, J. Luo, S. Huang, Y.-J. Cheng, T.-D. Kim, B. M. Polishak,
X.-H. Zhou, Y. Tian, S.-H. Jang, D. B. Knorr, R. M. Overney,
T. R. Younkin and A. K.-Y. Jen, Macromolecules, 2009, 42, 2438–
2445.
References
1 J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Marks,
K. Mackay, R. H. Friend, P. L. Burn and A. B. Holmes, Nature,
1990, 347, 539–541.
€
2 X. Yang, D. C. Muller, D. Neher and K. Meerholz, Adv. Mater.,
2006, 18, 948–954.
3 K. T. Kamtekar, A. P. Monkman and M. R. Bryce, Adv. Mater.,
2010, 22, 572–582.
35 F. He, Y. Tang, M. Yu, S. Wang, Y. Li and D. Zhu, Adv. Funct.
Mater., 2007, 17, 996–1002.
4 H. G. Jeon, M. Inoue, N. Hirarnatsu, M. Ichikawa and Y. Taniguchi,
Org. Electron., 2008, 9, 903–905.
5 C. Adachi, M. A. Baldo, M. E. Thompson and S. R. Forrest, J. Appl.
Phys., 2001, 90, 5048–5051.
˜
36 Y. Sangvikar, K. Fischer, M. Schmidt, A. D. SchluAater and
ꢀ
J. Sakamoto, Org. Lett., 2009, 11, 4112–4115.
37 A. Yokoyama, H. Suzuki, Y. Kubota, K. Ohuchi, H. Higashimura
and T. Yokozawa, J. Am. Chem. Soc., 2007, 129, 7236–7237.
38 W. L. F. Armarego and C. L. L. Chai, Purification of laboratory
chemicals, Elsevier/Butterworth-Heinemann, Boston, 2009.
39 J. Zhu, A. J. Kell and M. S. Workentin, Org. Lett., 2006, 8, 4993–
4996.
6 Y. Zheng, S.-H. Eom, N. Chopra, J. W. Lee, F. So and J. Xue, Appl.
Phys. Lett., 2008, 92, 223301.
7 S. H. Eom, Y. Zheng, E. Wrzesniewski, J. Lee, N. Chopra, F. So and
J. Xue, Org. Electron., 2009, 10, 686–691.
8 S. H. Eom, Y. Zheng, E. Wrzesniewski, J. Lee, N. Chopra, F. So and
J. G. Xue, Appl. Phys. Lett., 2009, 94(15).
€
40 V. N. Bliznyuk, S. A. Carter, J. C. Scott, G. Klarner, R. D. Miller and
€
9 S. T. Kim, D.-H. Hwang, X. C. Li, J. Gruner, R. H. Friend,
D. C. Miller, Macromolecules, 1999, 32, 361–369.
41 M. Grell, D. D. C. Bradley, G. Ungar, J. Hill and K. S. Whitehead,
Macromolecules, 1999, 32, 5810–5817.
A. B. Holmes and H. K. Shim, Adv. Mater., 1996, 8, 979–982.
10 W. Ma, P. K. Iyer, X. Gong, B. Liu, D. Moses, G. C. Bazan and
A. J. Heeger, Adv. Mater., 2005, 17, 274–277.
42 E. J. W. List, R. Guentner, P. S. de Freitas and U. Scherf, Adv.
Mater., 2002, 14, 374–378.
11 A. R. Brown and D. D. C. Bradley, Appl. Phys. Lett., 1992, 61,
2793.
€
43 S. Setayesh, A. C. Grimsdale, T. Weil, V. Enkelmann, K. Mullen,
12 H. Yan, P. Lee, N. R. Armstrong, A. Graham, G. A. Evmenenko,
P. Dutta and T. J. Marks, J. Am. Chem. Soc., 2005, 127, 3172–
3183.
F. Meghdadi, E. J. W. List and G. Leising, J. Am. Chem. Soc.,
2001, 123, 946–953.
44 K. L. Chan, M. Sims, S. I. Pascu, M. Ariu, A. B. Holmes and
D. D. C. Bradley, Adv. Funct. Mater., 2009, 19, 2147–2154.
45 S. L. McFarlane, D. G. Piercey, L. S. Coumont, R. T. Tucker,
M. D. Fleischauer, M. J. Brett and J. G. C. Veinot,
Macromolecules, 2009, 42, 591–598.
13 C. A. Zuniga, S. Barlow and S. R. Marder, Chem. Mater., 2011, 23,
658–681.
14 L. D. Bozano, K. R. Carter, V. Y. Lee, R. D. Miller, R. DiPietro and
J. C. Scott, J. Appl. Phys., 2003, 94, 3061–3068.
15 G. Klarner, J. I. Lee, V. Y. Lee, E. Chan, J. P. Chen, A. Nelson,
D. Markiewicz, R. Siemens, J. C. Scott and R. D. Miller, Chem.
Mater., 1999, 11, 1800–1805.
46 M. R. Sirtonski, S. L. McFarlane and J. G. C. Veinot, J. Mater.
Chem., 2010, 20, 8147–8152.
47 C. M. Martin, S. Guha, M. Chandrasekhar, H. R. Chandrasekhar,
R. Guentner, P. Scanduicci de Freitas and U. Scherf, Phys. Rev. B:
Condens. Matter, 2003, 68, 115203.
48 L. Romaner, A. Pogantsch, P. S. de Freitas, U. Scherf, M. Gaal,
E. Zojer and E. J. W. List, Adv. Funct. Mater., 2003, 13, 597–601.
49 J. H. Ahn, C. Wang, I. F. Perepichka, M. R. Bryce and M. C. Petty, J.
Mater. Chem., 2007, 17, 2996–3001.
50 H.-M. Liu, J. He, P.-F. Wang, H.-Z. Xie, X.-H. Zhang, C.-S. Lee,
B.-Q. Sun and Y.-J. Xia, Appl. Phys. Lett., 2005, 87, 221103.
51 J. Kido, M. Kimura and K. Nagai, Science, 1995, 267, 1332–1334.
52 M.-H. Tsai, H.-W. Lin, H.-C. Su, T.-H. Ke, C.-c. Wu, F.-C. Fang,
Y.-L. Liao, K.-T. Wong and C.-I. Wu, Adv. Mater., 2006, 18, 1216–
1220.
53 M. Ranger, D. Rondeau and M. Leclerc, Macromolecules, 1997, 30,
7686–7691.
54 J. Ding, M. Day, G. Robertson and J. Roovers, Macromolecules,
2002, 35, 3474–3483.
55 M. Ghaemy and M. Barghamadi, J. Appl. Polym. Sci., 2009, 114,
3464–3471.
56 X.-H. Zhou, J.-C. Yan and J. Pei, Macromolecules, 2004, 37, 7078–
7080.
16 G. K. Paul, J. Mwaura, A. A. Argun, P. Taranekar and
J. R. Reynolds, Macromolecules, 2006, 39, 7789–7792.
17 M. S. Liu, Y. H. Niu, J. W. Ka, H. L. Yip, F. Huang, J. D. Luo,
T. D. Kim and A. K. Y. Jen, Macromolecules, 2008, 41, 9570–9580.
18 Y. H. Niu, M. S. Liu, J. W. Ka and A. K. Y. Jen, Appl. Phys. Lett.,
2006, 88(9).
19 A. Charas, H. Alves, J. M. G. Martinho, L. Alcacer, O. Fenwick,
F. Cacialli and J. Morgado, Synth. Met., 2008, 158, 643–653.
20 M. J. Cho, M. H. Kim, K. N. Kim, J. I. Jin, Y. M. Kim, Y. W. Park,
B. K. Ju and D. H. Choi, Synth. Met., 2008, 158, 785–789.
21 K. Lu, Y. L. Guo, Y. Q. Liu, C. A. Di, T. Li, Z. M. Wei, G. Yu,
C. Y. Du and S. H. Ye, Macromolecules, 2009, 42, 3222–3226.
22 J. Schelter, G. F. Mielke, A. Kohnen, J. Wies, S. Kober, O. Nuyken
and K. Meerholz, Macromol. Rapid Commun., 2010, 31, 1560–1567.
23 H. Yan, B. J. Scott, Q. L. Huang and T. J. Marks, Adv. Mater., 2004,
16, 1948–1953.
24 A. P. Kulkarni, C. J. Tonzola, A. Babel and S. A. Jenekhe, Chem.
Mater., 2004, 16, 4556–4573.
25 S. Ellinger, K. R. Graham, P. Shi, R. T. Farley, T. T. Steckler,
R. N. Brookins, P. Taranekar, J. Mei, L. A. Padilha, T. R. Ensley,
H. Hu, S. Webster, D. J. Hagan, E. W. Van Stryland,
K. S. Schanze and J. R. Reynolds, Chem. Mater., 2011, 23, 3805–
3817.
57 K. Pilgram, M. Zupan and R. Skiles, J. Heterocycl. Chem., 1970, 7,
629–633.
3014 | J. Mater. Chem., 2012, 22, 3004–3014
This journal is ª The Royal Society of Chemistry 2012