A new class of non-conjugated bipolar hybrid hosts for phosphorescent organic light-emitting diodes

Article abstract of DOI:10.1039/b909787b

Comprising hole- and electron-transporting moieties with flexible linkages, representative non-conjugated bipolar hybrids have been synthesized and characterized for a demonstration of their potential use as host materials for the fabrication of phosphore

Full text of DOI:10.1039/b909787b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
A new class of non-conjugated bipolar hybrid hosts for phosphorescent
organic light-emitting diodes†
a
Lichang Zeng, Thomas Y.-H. Lee, Paul B. Merkel and Shaw H. Chen
a
b
ac
*
Received 18th May 2009, Accepted 15th September 2009
First published as an Advance Article on the web 7th October 2009
DOI: 10.1039/b909787b
Comprising hole- and electron-transporting moieties with flexible linkages, representative non-
conjugated bipolar hybrids have been synthesized and characterized for a demonstration of their
potential use as host materials for the fabrication of phosphorescent organic light-emitting diodes. The
advantages of this material class include solution processing into amorphous films with elevated glass
transition temperatures, stability against phase separation and crystallization, and provision of
LUMO/HOMO levels and triplet energies contributed by the two independent moieties without
constraint by the electrochemical energy gap. While exciplex formation between the hole- and electron-
transporting moieties is inevitable, its adverse effects on spectral purity and device efficiency can be
avoided by trapping charges on triplet emitters, as demonstrated for Ir(mppy)3 in TRZ-3Cz(MP)2, and
TRZ-1Cz(MP)2. With these two bipolar hybrids and hole-transporting Cz(MP)2 as the host, the
maximum current efficiency of the bilayer PhOLED is achieved with TRZ-3Cz(MP)2, but the driving
voltage decreases monotonically with an increasing TRZ content.
transfer to the host because of the higher ET of the latter.
Introduction
Compared to exciton transfer from the host, charge trapping on
the emitter as the source of phosphorescence is advantageous in
terms of the higher internal quantum yield,6,9,10 less concentra-
tion quenching because of the lower doping level,9,11 and the
emission spectrum solely from the emitter,11,12 albeit at the higher
driving voltage.12
Since the invention of relatively efficient fluorescent organic
light-emitting diodes (OLEDs) in 1987,1 a new generation of flat-
panel displays has emerged with a potential for capturing
a substantial market share of consumable electronics, such as
television sets and computer monitors. While full-color OLED
displays require the emission of blue, green and red light, white
OLEDs are potentially useful for efficient and inexpensive solid-
state lighting and as backlights for liquid crystal displays.2–4
Compared to molecular materials that can be vacuum-deposited
into thin films, solution-processable materials, such as p-conju-
gated polymers and monodisperse oligomers, offer cost advan-
tage and ease of scale-up to large-area thin films. Fluorescence or
phosphorescence is responsible for light emission from organic
luminophores. Electrophosphorescence is superior to electro-
fluorescence in terms of readily accessible internal quantum
yield, 100 versus 25%. Despite the intensive efforts worldwide
over the past decade, device efficiency and lifetime have remained
critical issues. For the fabrication of an efficient phosphorescent
OLED, a triplet emitter is typically doped in a host material with
sufficiently high triplet energy, ET, to realize blue, green or red
emission.5–8 A higher ET of the host than the guest ensures
exciton transfer from the former to the latter where light emission
occurs. In cases where the triplet emitters serve as charge traps,
exciton formation is expected at the emitter without back-
Most of the existing triplet host materials are capable of
preferentially transporting holes or electrons.7,13,14 Charge
injection and transport layers are added between electrodes and
the emitting layer as needed to improve efficiency.5–8,13,15
Nevertheless, charge recombination tends to occur close to the
interface with the charge-transport layer for lack of bipolar
transport capability in general of the emitting layer.16,17 Under
a high current density pertaining to practical application,
confinement of excitons to the interfacial region could expedite
triplet-triplet annihilation, resulting in efficiency roll-off.18–21
Furthermore, a narrow recombination zone is detrimental to
operational stability because only a fraction of molecules
contribute to charge transport, exciton formation, and light
emission.22–24 To substantially improve device efficiency and
lifetime, it is imperative that excitons be evenly distributed
through the emitting layer and that the accumulation of charges
and excitons at interfaces be prevented. To this effect, it has been
demonstrated that mixed hosts can effectively decrease driving
voltage while improving device efficiency sustainable at high
current densities.11,12,25–30 A typical phosphorescent layer is
comprised of a host mixed with a charge-transport component at
25 to 50 wt%, to which 1 to 10 wt% of a triplet emitter is doped.
The desired bipolar transport capability entails a high concen-
tration of the charge-transport additive, at which doping level
phase separation is destined to take place over time unless
miscibility has been taken into account in the design of both
components, thus adversely affecting long-term operational
stability of OLEDs. Bipolar charge-transport host materials via
aDepartment of Chemical Engineering, University of Rochester, Rochester,
NY, 14627, USA. E-mail: shch@lle.rochester.edu
bDepartment of Chemistry, University of Rochester, Rochester, NY, 14627,
USA
cLaboratory for Laser Energetics, University of Rochester, 240 East River
Road, Rochester, NY, 14623, USA
† Electronic supplementary information (ESI) available: Synthesis
procedures, characterization data and POM images. See DOI:
10.1039/b909787b
8772 | J. Mater. Chem., 2009, 19, 8772–8781
This journal is ª The Royal Society of Chemistry 2009

View Article Online
chemical modification represent a viable approach to circum-
venting the potential phase-separation problem. A bipolar
compound can be constructed by chemically bonding an elec-
tron- and a hole-transport moiety with and without a finite extent
of p-conjugation between the two moieties, resulting in conju-
gated and non-conjugated bipolar compounds, respectively. For
all the conjugated bipolar host materials that have been devel-
oped for full-color emission,31–40 few non-conjugated bipolar
compounds have been reported,41,42 of which none carry a flex-
ible linkage consisting of s-bonds between the two moieties.
Conjugated and non-conjugated bipolar hybrid molecules
without a flexible linkage tend to be rigid and bulky, thus limiting
solubility and the ability to form morphologically stable glassy
films. This study aims at a new class of non-conjugated bipolar
hybrids comprising aliphatic linkages between the two charge-
transport moieties. Representative compounds have been
synthesized and characterized to assess their potential for use as
hosts for triplet emitters. Furthermore, the ability of bipolar
hybrid hosts to modulate charge injection into and transport
through the emitting layer has been unraveled through the
fabrication and characterization of PhOLEDs.
representative non-conjugated bipolar hybrid compounds with
propylene linkages, TRZ-1Cz(MP)2, TRZ-3Cz(MP)2, and
OXD-2Cz(MP)2 that were synthesized for an investigation of
their thermal, morphological, electrochemical, fluorescence, and
phosphorescence properties. These hybrid compounds consist of
a hole-transport Cz(MP)243–45 and an electron-transport TRZ46–49
or OXD.27,30,49,50
The results from thermogravimetric analysis compiled in
Fig. 1 reveal their thermal stability to 400 ꢀC at about 5 wt%
weight loss. Solid morphologies of the three independent
building blocks, three bipolar compounds, and their corre-
sponding mixtures, TRZ:1Cz(MP)2, TRZ:3Cz(MP)2, and
OXD:2Cz(MP)2, were characterized by differential scanning
calorimetry and hot-stage polarizing optical microscopy. The
DSC thermograms compiled in Fig. 2 indicate that the building
blocks and their mixtures are crystalline, semicrystalline or
amorphous with a glass transition temperature, Tg, below 65 ꢀC,
while all the hybrid compounds are amorphous with a Tg near or
ꢀ
above 100 C.
Results and discussion
In addition to precluding phase separation, the flexible linkages
connecting the two charge-transport moieties in the non-conju-
gated bipolar hybrid molecules serve to increase entropy because
of the more abundant conformations, which is conducive to
solubility in benign solvents to facilitate materials purification
and solution processing. Furthermore, the increased entropy
with flexible linkages presents a higher free energy barrier to
crystallization from a glassy state, thereby improving morpho-
logical stability against crystallization over relatively rigid
conjugated and non-conjugated bipolar hybrid molecules
without flexible linkages. Depicted in Chart 1 are three
Fig. 1 TGA thermograms of hybrid compounds recorded at a heating
rate of 10 ꢀC/min under nitrogen atmosphere. The decomposition
temperatures at a weight loss of 5% are 399, 403 and 407 ꢀC for TRZ-
1Cz(MP)2, TRZ-3Cz(MP)2 and OXD-2Cz(MP)2, respectively.
Chart 1 Representative non-conjugated bipolar compounds as well as independent electron- and hole-transport moieties with their thermal transition
temperatures determined by DSC heating scans shown in Fig. 2 below. Symbols: G, glassy; K, crystalline; I, isotropic.
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 8772–8781 | 8773

View Article Online
Fig. 3 Polarizing optical micrographs of films from spin-cast chloro-
benzene of OXD:2Cz(MP)2 mixture (a) before and (b) after thermal
annealing at 32 ꢀC for 3 days; (c) that of an OXD-2Cz(MP)2 film before
and after thermal annealing at 100 ꢀC for 3 days; and (d) that of an
OXD:PVK mixture at 30:70 mass ratio after thermal annealing at 100 ꢀC
for 3 days.
reduced temperature and annealing time as for OXD:2Cz(MP)2,
indicating that the hybrid compound is not vulnerable to phase
separation and that it is resistant to thermally activated crystal-
lization. Poly(N-vinylcarbazole), PVK, mixed with OXD at a
70:30 mass ratio was employed as the host for iridium(III) bis[2-
0
(4,6-difluorophenyl)-pyridinato-N, C2 ]picolinate (FIrpic).27 The
pristine, spin-cast film of this bipolar host material was found to
be amorphous with the same optical micrograph as shown in
Fig. 3a or c, but phase separation and/or crystallization emerged
ꢀ
in Fig. 3d upon thermal annealing at 100 C for 3 days. More-
over, the crystallites observed under polarizing optical micros-
copy were found to melt at 230 ꢀC, close to the melting point of
OXD.
Fig. 2 DSC heating and cooling scans at ꢂ20 ꢀC/min of samples
comprising (a) hole- and electron-transport moieties, (b) mixtures
thereof, and (c) non-conjugated bipolar compounds that have been pre-
heated to beyond their melting points followed by quenching to ꢁ30 ꢀC.
Symbols: G, glassy; K, crystalline; I, isotropic.
The electrochemical properties of the non-conjugated bipolar
compounds and their building blocks are characterized by cyclic
voltammetry. The oxidation and reduction scans are shown in
Fig. 4, illustrating that the oxidation and reduction scans of the
hybrids are represented by the composites of those of their
constituent hole- and electron-transport moieties. The key data
summarized in Table 1 indicates that the HOMO levels of TRZ-
1Cz(MP)2, TRZ-3Cz(MP)2 and OXD-2Cz(MP)2 are placed
at ꢁ5.2 eV, a value identical to that of the hole-transporting
Cz(MP)2. The LUMO levels of ꢁ2.6 eV for TRZ-1Cz(MP)2 and
TRZ-3Cz(MP)2, and ꢁ2.5 eV for OXD-2Cz(MP)2, are also
equal within error to those of the electron-transporting TRZ and
OXD, respectively. With a HOMO level at ꢁ5.2 eV, which is
close to the work function of PEDOT:PSS at 5.1 eV, and
a LUMO level at ꢁ2.5 to ꢁ2.6 eV, which is relatively close to the
work function of LiF/Al at 2.9 eV, these non-conjugated bipolar
compounds are expected to facilitate hole and electron injection
in phosphorescent OLED devices. With an interruption of p-
conjugation between the two moieties by a propylene linkage, the
HOMO and LUMO levels of non-conjugated bipolar hybrid
compounds are essentially imported from those of the hole- and
electron-transport moieties as independent chemical entities. The
The amorphous OXD:2Cz(MP)2 mixture observed under
differential scanning calorimetry and polarizing optical micros-
copy was further evaluated for phase separation via thermal
annealing at 32 ꢀC, viz. a reduced temperature, T/Tg ¼ 0.91, for
three days. While the pristine, spin-cast film of OXD:2Cz(MP)2
was amorphous under polarizing optical microscopy (Fig. 3a),
phase separation and/or crystallization occurred upon thermal
annealing at 32 ꢀC for 3 days (Fig. 3b). The thermally annealed
film was further characterized by polarizing optical microscopy
to yield melting points at 150 and 185 ꢀC, which fall between the
melting points of Cz(MP)2 (144 ꢀC) and OXD (240 ꢀC) also
determined by polarizing optical microscopy for spin-cast films
left at room temperature for 2 to 3 days. These results suggest
a complex phase behavior of the OXD:2Cz(MP)2 mixture. As
shown in Fig. 3c, the amorphous character persisted in the OXD-
ꢀ
2Cz(MP)2 film upon thermal annealing at 100 C, i.e. the same
8774 | J. Mater. Chem., 2009, 19, 8772–8781
This journal is ª The Royal Society of Chemistry 2009

View Article Online
LUMO and HOMO levels in conjugated bipolar hybrid
compounds, however, are generally affected by the finite extent
of p-conjugation between the two moieties.
In addition to phase separation detected by microscopy (see
Fig. 3), fluorescence spectroscopy was employed to uncover
molecular aggregation in non-conjugated bipolar compounds
and their equivalent mixtures. To facilitate data interpretation,
TRZ-1Cz(MP)2 was used along with TRZ:1Cz(MP)2 based on
the facts that TRZ was found to be essentially nonemissive.51 and
that OXD showed fluorescence in the 400 to 450 nm region
largely overlapping with that of Cz(MP)2. Approximately 45
nm-thick amorphous films of TRZ-1Cz(MP)2, TRZ:1Cz(MP)2,
and Cz(MP)2 prepared by spin-coating from chlorobenzene were
photoexcited at 360 nm, and their fluorescence spectra normal-
ized with film thickness are presented in Fig. 5. The pristine TRZ-
1Cz(MP)2 and TRZ:1Cz(MP)2 films exhibited similar
fluorescence in the same spectral range from 450 to 600 nm,
representing a red-shift from that of Cz(MP)2 by about 90 nm.
These broad and red-shifted fluorescence peaks at 2.4 eV origi-
nated from exciplex formation between TRZ and Cz(MP)2
moieties in the hybrid compound and their equimolar mixture, as
expected of the offset between the HOMO level of Cz(MP)2 and
the LUMO level of TRZ in addition to the Coulomb attraction
energy.52 The lower fluorescence intensity from the pristine film
of TRZ-1Cz(MP)2 than that of TRZ:1Cz(MP)2 could have
arisen from the less extent of inter- and/or intra-molecular exci-
plex formation in the former caused by steric hindrance in the
presence of a propylene spacer. Thermal annealing of the TRZ-
1Cz(MP)2 film at 20 ꢀC above its Tg for ½ h did not result in any
change in the amorphous character and the fluorescence spec-
trum. Upon thermal annealing under the same condition, poly-
crystalline domains emerged from the pristine TRZ:1Cz(MP)2
film with a melting point at 178 ꢀC, whꢀich is distinct from those
of TRZ and Cz(MP)2 at 242 and 144 C, respectively as noted
above. The polarizing optical micrographs of thermally annealed
TRZ-1Cz(MP)2 and TRZ:1Cz(MP)2 films are displayed in
Fig. S1 (ESI†). As shown in Fig. 5, phase separation accompa-
nied by crystallization led to much reduced, blue-shifted exciplex
emission as well as weak emission between 400 and 450 nm
attributable to Cz(MP)2. Thus, the hybrid compound is prefer-
able over its equivalent mixture from the standpoint of
morphological stability, but the issue of exciplex formation in
both material systems must be addressed. Whereas exciplex
emission involving bipolar mixed hosts is frequently observed in
photoluminescence, it is absent in electroluminescence where
triplet emitters serve as charge traps with improved device effi-
ciencies at lower driving voltages compared to the constituent
unipolar hosts.11,12,29 To ensure effective charge trapping on the
emitter, its LUMO level must be lower than that of the electron-
transport component, while its HOMO level must be higher than
that of the hole-transport component, an idea to be incorporated
in the design of non-conjugated bipolar hybrid compounds.
The absorption spectra of Cz(MP)2, OXD, and OXD-
2Cz(MP)2 at 10ꢁ7 to 10ꢁ6 M in chloroform are presented in
Fig. 6a, serving to identify selective photoexcitation wavelengths
for the determination of their lowest triplet energies, ET, to
within ꢂ0.02 eV through phosphorescence measurements.
Phosphorescence spectra were obtained for Cz(MP)2, OXD, and
OXD-2Cz(MP)2 at 10ꢁ4 M in ethyl acetate at ꢁ196 ꢀC (or 77 K).
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 8772–8781 | 8775

View Article Online
Fig. 4 Cyclic voltammetric scans of compounds in acetonitrile/toluene (1:1 by volume) at 10ꢁ3 M with 0.1 M tetrabutylammonium tetrafluoroborate as
the supporting electrolyte.
3Cz(MP)2 to arrive at ET ¼ 3.03 and 2.75 eV, respectively, an
observation also consistent with highly efficient triplet energy
transfer from TRZ to the lower energy Cz(MP)2 moiety.
Furthermore, the ET value of a non-conjugated bipolar hybrid
compound is determined by the lower value of the two inde-
pendent moieties. The ET value of a conjugated bipolar
compound, however, is consistently less than those of the two
independent moieties because of the finite p-conjugation
between them.29,31,54
Let us proceed to assess yet another merit of non-conjugated
bipolar compounds compared to their conjugated counterparts.
In a conjugated bipolar compound, the ET value is consistently
less than EG, the LUMO–HOMO energy gap,31–40 by the sum of
55
singlet–triplet splitting, DST
,
and exciton binding energy, EB.56
Fig. 5 Fluorescence spectra with excitation at 360 nm of approximately
45 nm-thick, spin-cast films of Cz(MP)2, TRZ-1Cz(MP)2, and
TRZ:1Cz(MP)2; thermal annealing was performed at 20 ꢀC above Tg
under argon for ½ h.
As illustrated in Fig. 6b, Cz(MP)2 shows a 0,0-phosphorescence
band corresponding to an ET at 2.77 eV upon excitation at
360 nm. The OXD moiety shows only very weak phosphores-
cence due largely to inefficient intersystem crossing. Thus, butyl
iodide was added at 10 wt% to the OXD and OXD-2Cz(MP)2
solutions to enhance phosphorescence, particularly the 0,0-
vibronic transition,53 and to help confirm the identity of the
emitting state. With photoexcitation at 332 nm, the relatively
sharp highest-energy 0,0-vibronic band shown in Fig. 6c estab-
lishes an ET of 2.71 eV for OXD. When the hybrid compound
OXD-2Cz(MP)2 is excited at 360 nm (Fig. 6d), its phosphores-
cence spectrum is identical to that of OXD (Fig. 6c) even though
OXD is not excited at 360 nm. These results indicate highly
efficient triplet energy transfer from Cz(MP)2 (ET ¼ 2.77 eV) to
OXD (ET ¼ 2.71 eV) in OXD-2Cz(MP)2 and further suggest that
the two moieties retain their energy levels as independent entities
and that the effective ET of the hybrid equals that of the lower-ET
moiety due to efficient intramolecular triplet energy transfer.
Phosphorescence spectra were also collected for TRZ and TRZ-
Fig. 6 (a) UV-vis absorption spectra in molecular extinction coeffi-
cients, 3, of OXD-2Cz(MP)2, Cz(MP)2 and OXD. Phosphorescence
spectra of (b) Cz(MP)2, (c) OXD, and (d) OXD-2Cz(MP)2 at 77 ꢀK in
ethyl acetate at 10ꢁ4 M, for which the ET values were determined by the 0–
0 transitions as indicated by arrows.
8776 | J. Mater. Chem., 2009, 19, 8772–8781
This journal is ª The Royal Society of Chemistry 2009

View Article Online
Fig. 7 (a) Molecular structure and energy diagram of a conjugated bipolar compound CzOXD, and (b) molecular structure and energy diagram of
a non-conjugated bipolar compound TRZ-1Cz(MP)2 accompanied by those of Cz(MP)2 and TRZ.
For example, CzOXD has its HOMO/LUMO levels at ꢁ5.6/
ꢁ2.4 eV based on the reported half-wave potentials with an ET at
2.5 eV,38 which is 0.7 eV less than EG (see Fig. 7a).
Prepared by spin casting, the emitting layer consisted of a host
doped with green-emitting tris(2-(p-methylphenyl)pyridine)iri-
dium, Ir(mppy)3 at a 10:1 mass ratio. While MoO3 and CsF were
applied as the hole- and electron-injection layers, respectively,
TPBI was applied as the electron-transporting and hole-blocking
layer. A typical electroluminescence spectrum is shown as the inset
in Fig. 8a, consistent with the triplet emission of Ir(mppy)3,12 sug-
gesting effective confinement of triplet excitons on the emitter. This
is expected of the higher ETs of all three hosts than that of
Ir(mppy)3, 2.8 eV for Cz(MP)2 over 2.4 eV for Ir(mppy)3.57
Furthermore, Ir(mppy)3 has a HOMO level at ꢁ5.0 eV, 0.2 eV
higher than those of the hosts, thus serving as a hole trap to
preclude exciplex formation between the TRZ and Cz(MP)2
moieties.
On the other hand, the ET value of a non-conjugated bipolar
compound, such as TRZ-1Cz(MP)2, is not limited by its EG. The
oxidation potentials of TRZ and OXD and the reduction
potential of Cz(MP)2 are beyond our CV measurement range.
Their optical bandgaps were estimated at the onset of their
absorption spectra in dilute solutions to yield HOMO levels at
ꢁ6.7 and ꢁ6.2 eV for TRZ and OXD, respectively, and LUMO
level at ꢁ1.8 eV for Cz(MP)2. As shown in Fig. 7b, the ETs of
independent electron- and hole-transport moieties are less than
their respective EGs as in a typical conjugated system. The
Cz(MP)2 moiety is characterized by an ET and HOMO/LUMO
levels at 2.8 eV, and ꢁ5.2/ꢁ1.8 eV, respectively, while the TRZ
moiety carries an ET and HOMO/LUMO levels at 3.0 eV, and
ꢁ6.7/ꢁ2.6 eV, respectively (see Fig. 7b). The hybrid TRZ-
1Cz(MP)2 has an ET at 2.8 eV, as verified by low temperature
phosphorescence spectroscopy, and HOMO/LUMO levels at
ꢁ5.2/ꢁ2.6 eV by cyclic voltammetry. In contrast to CzOXD,
TRZ-1Cz(MP)2 has an ET greater than its EG and its HOMO/
LUMO levels corresponding to the hole- and electron-transport
moieties without modification in the absence of inter-moiety
conjugation. The caveat here is that the ETs of both the electron-
and hole-transport moieties must be greater than the hybrid’s EG
to arrive at a non-conjugated bipolar compound with an ET
greater than its EG, as is also applicable to TRZ-3Cz(MP)2 and
OXD-2Cz(MP)2.
It is also shown in Fig. 8a that driving voltage decreases with
an increasing TRZ content in the hybrid hosts, suggesting
improved electron injection from the adjacent TPBI into the
emitting layer. Current efficiency and luminance as functions of
Fig. 8 (a) Current density as a function of driving voltage for phos-
phorescent OLEDs with emitting layers comprising Cz(MP)2, TRZ-
3Cz(MP)2, and TRZ-1Cz(MP)2 doped with Ir(mppy)3 at a 10:1 mass
ratio. Inset: electroluminescence (EL) spectrum with TRZ-3Cz(MP)2 as
the host. (b) Luminance and current efficiency as functions of current
density for the same phosphorescent OLEDs as described in (a).
Bipolar hybrid TRZ-3Cz(MP)2 and TRZ-1Cz(MP)2 and hole-
transporting Cz(MP)2 were employed as hosts for the fabrication
of bilayer phosphorescent OLEDs in the device architecture, ITO/
MoO3(10 nm)/emitting layer(40–50 nm)/1,3,5-tris(N-phenyl-
benzimidazol-2-yl)benzene (TPBI)(30 nm)/CsF(1 nm)/Al(100 nm).
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 8772–8781 | 8777

View Article Online
current density are shown in Fig. 8b. Compared to Cz(MP)2, the
higher efficiency with TRZ-3Cz(MP)2 as the host is attributable
to the more balanced electron and hole fluxes through the
emitting layer, which leads to the more efficient electron-hole
recombination. Furthermore, the presence of electron-trans-
porting TRZ in TRZ-3Cz(MP)2 generates a broader charge
recombination zone to alleviate efficiency roll-off at high current
densities. As shown in Fig. 8b, at the current density of 0.5 mA/
cm2, the device with Cz(MP)2 as the host has a luminance of
105 cd/m2, corresponding to current efficiency of 21 cd/A and
external quantum efficiency of 5.9%, which diminish to 5.4 cd/A
and 1.5% at 100 mA/cm2, respectively, representing a 74% loss in
efficiency. In contrast, at the current density of 0.5 mA/cm2, the
device with TRZ-3Cz(MP)2 as the host has a luminance of
160 cd/m2, corresponding to current efficiency of 32 cd/A and
external quantum efficiency of 9.2%, which roll off to 16.8 cd/A
and 4.9% at 100 mA/cm2, respectively, representing a 47% loss in
efficiency. The efficiencies achieved with TRZ-3Cz(MP)2 as the
host are among the best of solution-processed phosphorescent
OLEDs using bipolar hosts.32,33,38,41 Implemented in more
sophisticated device architectures, optimum non-conjugated
bipolar hybrids can be expected to yield much higher efficiencies
than the bilayer device architecture as presently reported. With
a further increase in the TRZ content, however, TRZ-1Cz(MP)2
resulted in current efficiencies comparable to those with
Cz(MP)2 as the host, presumably because of exciton quenching
by MoO3 as the recombination zone is shifted toward the anode
at an increased electron transport.58 The results obtained to date
have demonstrated the potential of non-conjugated bipolar
hybrid hosts with flexible linkages for substantially improving
PhOLED device performance through optimization of change
injection into and transport through the emitting layer.
optical microscopy (DMLM, Leica, FP90 central processor and
FP82 hotstage, Mettler Toledo). Absorption spectra of dilute
solutions in chloroform at a concentration of 10ꢁ7 to 10ꢁ6 M were
acquired on a UV-vis-NIR spectrophotometer (Lambdaꢁ900,
Perkin-Elmer).
Electrochemical characterization
Cyclic voltammetry was conducted on an EC-Epsilon potentio-
stat (Bioanalytical Systems Inc.) at a concentration of 10ꢁ3 M in
acetonitrile/toluene (1:1 by volume) containing 0.1 M tetrae-
thylammonium tetrafluoroborate as the supporting electrolyte.
A silver/silver chloride (Ag/AgCl) wire, a platinum wire, and
a glassy carbon disk (3 mm diameter) were used as the reference,
counter, and working electrodes, respectively, to complete
a standard 3-electrode cell. The supporting electrolyte was
purified as described previously,59 and the solvents acetonitrile
and toluene were distilled over calcium hydride and sodium/
benzophenone, respectively. The dilute sample solutions in ace-
tonitrile:toluene (1:1 by volume) exhibit reversible reduction and
oxidation waves against the Ag/AgCl reference electrode. The
reduction and oxidation potentials were adjusted to ferrocene
serving as an internal standard with an oxidation potential of
0.51 ꢂ 0.02 V over Ag/AgCl. The resultant reduction and
oxidation potentials, E1/2(red) and E1/2(oxd), relative to (Fc/Fc+)
were used to calculate the LUMO and HOMO levels as ꢁ4.8eV ꢁ
qE1/2(red) and ꢁ4.8eV ꢁ qE1/2(oxd), respectively, where q is the
electron charge.60,61
Triplet energy measurement
Phosphorescence spectra were gathered using a Fluorolog-3
spectrofluorimeter (Jobin Yvon, Horiba) and were corrected for
the efficiency of the monochromator and the spectral response of
the photomultiplier tube. Samples (10ꢁ4 M) were dissolved in
ethyl acetate in NMR tubes and inserted into a small liquid
nitrogen Dewar to measure the phosphorescence spectra at 77 K.
As has been customary,54,62 the maximum of highest-energy 0–
0 vibronic band in the phosphorescence spectrum was assigned as
the energy of the lowest triplet state. Phosphorescence
measurements were also carried out with 10% butyl iodide added
to the ethyl acetate to enhance the 0–0 vibronic transition53 and
to help differentiate between phosphorescence and fluorescence.
Experimental section
Material synthesis and characterization
1H NMR spectra were acquired in CDCl3 with an Avanceꢁ400
spectrometer (400 MHz) at 298 K using trimethylsilane (TMS) as
an internal standard. Elemental analysis was carried out by
Quantitative Technologies, Inc. Molecular weights were
measured with a TofSpec2E MALD/I TOF mass spectrometer
(Micromass, Inc., Manchester, U.K.) with 2-[(2E)-3-(4-tert-
butylphenyl)-2-methylpropenylidene]malononitrile (DCTB) as
the matrix. The target compounds were synthesized and purified
according to Scheme 1 following the procedures described in the
ESI.†
Thin film preparation and characterization
Films of hybrid compounds and mixtures were prepared by spin
coating from 2 wt% chlorobenzene solutions at 2500 rpm on
microscope glass slides followed by drying under vacuum over-
night. The thicknesses of the resultant films were determined by
optical interferometry (Zygo New Views 5000). Thermal
annealing was performed under argon at elevated temperatures.
The film morphology, including melting points where applicable,
was characterized by hot-stage polarizing optical microscopy.
Photoluminescence was characterized using a spectrofluorimeter
(Quanta Master C-60SE, Photon Technology International) with
a liquid light guide directing excitation at 360 nm onto the sample
film at normal incidence.
Morphology, thermal stability, and phase transition
temperatures
Thermogravimetric analysis was performed in a TGA/DSC
system (SDT Q600, TA Instruments) at a ramping rate of 10 ꢀC/
min under a nitrogen flow of 50 ml/min. Thermal transition
temperatures were determined by differential scanning calorim-
etry (Perkin-Elmer DSC-7) with a continuous N2 purge at 20 ml/
min. Samples were preheated to above Tm and then cooled down
ꢀ
ꢀ
to ꢁ30 C at ꢁ100 C/min before the reported second heating
and cooling scans were recorded at 20 ꢀC/min. The nature of
phase transition was characterized by hot-stage polarizing
8778 | J. Mater. Chem., 2009, 19, 8772–8781
This journal is ª The Royal Society of Chemistry 2009

View Article Online
Scheme 1 Synthesis scheme of non-conjugated bipolar hybrids, TRZ-1Cz(MP)2, TRZ-3Cz(MP)2, OXD-2Cz(MP)2, and hole-transporting Cz(MP)2.
Phosphorescent OLED device fabrication and characterization
a source-measure unit (Keithley 2400) and a spectroradiometer
(PhotoResearch PR650). Only the front-view performance data
were collected.
Glass substrates coated with patterned ITO were thoroughly
cleaned and treated with oxygen plasma prior to deposition of
a 10 nm thick MoO3 layer by thermal evaporation at 0.1 nm/s.
The emitting layers comprising host:Ir(mppy)3 at a mass ratio of
10:1 were then prepared by spin-coating from 2.0 wt% toluene
solutions at 4000 rpm for 2 min in a nitrogen-filled glove box with
oxygen level less than 1 ppm. The resultant film thickness was 40–
50 nm determined by optical interferometry (Zygo NewView
5000). Layers of TPBI (30 nm) and CsF (1 nm) were then
consecutively deposited at rates of 0.1 nm/s and 0.02 nm/s,
respectively. The devices were completed by thermal evaporation
of Al (100 nm) at 1 nm/s through a shadow mask to define an
active area of 0.1 cm2. All evaporation processes were carried out
at a base pressure less than 4 ꢃ 10ꢁ6 Torr. All devices were
encapsulated with cover glass and glue for characterization with
Summary
Potentially useful as the host materials for the fabrication of
efficient and stable, single-layer phosphorescent OLEDs for
information display and solid-state lighting, a new class of non-
conjugated bipolar compounds have been synthesized and
characterized for their thermal, morphological, electrochemical,
fluorescence, and phosphorescence properties. Comprising hole-
and electron-transport moieties chemically bonded by an
aliphatic spacer, the potential of these materials has been
demonstrated for solution processing and for the formation of
thin films with elevated glass transition temperatures with
superior stability against phase separation and thermally
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 8772–8781 | 8779

View Article Online
12 X. H. Yang, F. Jaiser and D. Neher, in Highly Efficient OLED with
Phosphorescent Materials, ed. H. Yersin, Wiley-VCH, Weinheim,
2008, pp. 221–258.
13 X. F. Ren, J. Li, R. J. Holmes, P. I. Djurovich, S. R. Forrest and
M. E. Thompson, Chem. Mater., 2004, 16, 4743–4747.
14 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.
15 K. Walzer, B. Maenning, M. Pfeiffer and K. Leo, Chem. Rev., 2007,
107, 1233–1271.
16 C. Adachi, M. Baldo, M. E. Thompson and S. R. Forrest, J. Appl.
Phys., 2001, 90, 5048–5051.
17 S. Reineke, K. Walzer and K. Leo, Phys. Rev. B: Condens. Matter
Mater. Phys., 2007, 75, 125328.
18 M. A. Baldo, C. Adachi and S. R. Forrest, Phys. Rev. B: Condens.
Matter Mater. Phys., 2000, 62, 10967–10977.
19 N. C. Giebink and S. R. Forrest, Phys. Rev. B: Condens. Matter
Mater. Phys., 2008, 77, 235215.
20 S. Reineke, G. Schwartz, K. Walzer and K. Leo, Appl. Phys. Lett.,
2007, 91, 123508.
activated crystallization. Because of the absence of p-conjuga-
tion between the two charge-carrier moieties, the LUMO/
HOMO levels and the triplet energies of the two moieties as
independent entities are retained in the resultant non-conjugated
bipolar compounds. Furthermore, the flexibility in molecular
design is enhanced by the fact that the triplet energy of a non-
conjugated bipolar compound is not constrained by its electro-
chemical energy gap. Exciplex formation is inevitable in neat
films between the hole- and electron-transport moieties, but its
adverse effects on spectral purity and device efficiency can be
prevented by having triplet emitters serve as charge traps. All
these material traits are conducive to the optimization of prop-
erties for intended device applications. The current efficiencies of
PhOLEDs consisting of Ir(mppy)3 doped in Cz(MP)2, TRZ-
3Cz(MP)2, and TRZ-1Cz(MP)2 at an increasing TRZ content
reach the maximum at 32 cd/A with TRZ-3Cz(MP)2, which is
among the best of solution processed devices using bipolar hosts.
The driving voltage, however, decreases monotonically with an
increasing TRZ content, suggesting improved electron injection
from the adjacent TPBI layer into the emitting layer.
21 W. Staroske, M. Pfeiffer, K. Leo and M. Hoffmann, Phys. Rev. Lett.,
2007, 98, 197402.
22 B. D. Chin and C. Lee, Adv. Mater., 2007, 19, 2061–2066.
23 R. Meerheim, S. Scholz, S. Olthof, G. Schwartz, S. Reineke,
K. Walzer and K. Leo, J. Appl. Phys., 2008, 104, 014510.
24 N. C. Giebink, B. W. D’Andrade, M. S. Weaver, P. B. Machenzie,
J. J. Brown, M. E. Thompson and S. R. Forrest, J. Appl. Phys.,
2008, 103, 044509.
25 K. M. Vaeth and C. W. Tang, J. Appl. Phys., 2002, 92, 3447–3453.
26 X. Yang, D. Neher, D. Hertel and T. K. Daubler, Adv. Mater., 2004,
16, 161–166.
27 F. So, B. Krummacher, M. K. Mathai, D. Poplavsky, S. A. Choulis
and V.-E. Choong, J. Appl. Phys., 2007, 102, 091101.
28 S. H. Kim, J. Jang, K. S. Yook and J. Y. Lee, Appl. Phys. Lett., 2008,
92, 023513.
29 M. E. Kondakova, T. D. Pawlik, R. H. Young, D. J. Giesen,
D. Y. Kondakov, C. T. Brown, J. C. Deaton, J. Lenhard and
K. P. Klubek, J. Appl. Phys., 2008, 104, 094501.
30 H. B. Wu, J. H. Zou, F. Liu, L. Wang, A. Mikhailovsky, G. C. Bazan,
W. Yang and Y. Cao, Adv. Mater., 2008, 20, 696–702.
31 Z. Y. Ge, T. Hayakawa, S. Ando, M. Ueda, T. Akiike, H. Miyamoto,
T. Kajita and M. Kakimoto, Org. Lett., 2008, 10, 421–424.
32 Z. Y. Ge, T. Hayakawa, S. Ando, M. Ueda, T. Akiike, H. Miyamoto,
T. Kajita and M.-a. Kakimoto, Adv. Funct. Mater., 2008, 18, 584–590.
33 Z. Y. Ge, T. Hayakawa, S. Ando, M. Ueda, T. Akiike, H. Miyamoto,
T. Kajita and M.-a. Kakimoto, Chem. Mater., 2008, 20, 2532–2537.
34 X. Cai, A. B. Padmaperuma, L. S. Sapochak, P. A. Vecchi and
P. E. Burrows, Appl. Phys. Lett., 2008, 92, 083308.
35 S. J. Su, H. Sasabe, T. Takeda and J. Kido, Chem. Mater., 2008, 20,
1691–1693.
Acknowledgements
The authors thank Kevin Klubek and Andrew J. Hoteling of
Eastman Kodak Company for MALD/I-TOF analysis,
Professor Ching W. Tang of the Chemical Engineering Depart-
ment at the University of Rochester for his suggestion of bilayer
phosphorescent OLED device structures and access to the
PhOLED device fabrication and characterization facilities, and
Professor Hong Yang for access to the TGA analysis. They are
grateful for the financial support provided by the New York
State Energy Research and Development Authority. Additional
funding was provided by the Department of Energy Office of
Inertial Confinement Fusion under Cooperative Agreement No.
DE-FC52-08NA28302 with LLE. The support of DOE does not
constitute an endorsement by DOE of the views expressed in this
article.
36 M.-Y. Lai, C.-H. Chen, W.-S. Huang, J.-T. Lin, T.-H. Ke,
L.-Y. Chen, M.-H. Tsai and C.-c. Wu, Angew. Chem., Int. Ed.,
2008, 47, 581–585.
References
37 Y. Tao, Q. Wang, C. Yang, Q. Wang, Z. Zhang, T. Zou, J. Qin and
D. Ma, Angew. Chem., Int. Ed., 2008, 47, 8104–8107.
38 Y. T. Tao, Q. Wang, C. L. Yang, K. Zhang, Q. Wang, T. T. Zou,
J. G. Qin and D. G. Ma, J. Mater. Chem., 2008, 18, 4091–4096.
39 Z. Q. Gao, M. M. Luo, X. H. Sun, H. L. Tam, M. S. Wong, B. X. Mi,
P. F. Xia, K. W. Cheah and C. H. Chen, Adv. Mater., 2009, 21, 688–
692.
40 Y. T. Tao, Q. Wang, Y. Shang, C. L. Yang, J. Ao, J. G. Qin, D. G. Ma
and Z. G. Shuai, Chem. Commun., 2009, 77–79.
41 F.-M. Hsu, C.-H. Chien, P.-I. Shih and C.-F. Shu, Chem. Mater.,
2009, 21, 1017–1022.
42 F.-M. Hsu, C.-H. Chien, S.-F. Shu, C.-H. Lai, C.-C. Hsien,
K.-W. Wang and P.-T. Chou, Adv. Funct. Mater., 2009, 19, 2834.
43 K. Brunner, A. V. Dijken, H. Brner, J. J. A. M. Bastiaansen,
N. M. M. Kiggen and B. M. W. Langeveld, J. Am. Chem. Soc.,
2004, 126, 6035–6042.
ꢀ
44 I. Avilov, P. Marsal, J.-L. Bredas and D. Beljonne, Adv. Mater., 2004,
16, 1624–1629.
45 S.-O. Jeon, K. S. Yook, C. W. Joo, J. Y. Lee, K.-Y. Ko, J.-Y. Park
and Y. G. Baek, Appl. Phys. Lett., 2008, 93, 063306.
46 T. Ishi-I, K. Yaguma, T. Thiemann, M. Yashima, K. Ueno and
S. Mataka, Chem. Lett., 2004, 33, 1244–1245.
1 C. W. Tang and S. A. VanSlyke, Appl. Phys. Lett., 1987, 51, 913–915.
2 B. D’Andrade and S. R. Forrest, Adv. Mater., 2004, 16, 1585–1595.
3 A. Misra, P. Kumar and S. Kamalasana, Semicond. Sci. Technol.,
2006, 21, R35–R47.
4 G. Schwartz, S. Reineke, T. C. Rossenow, K. Walzer and K. Leo,
Adv. Funct. Mater., 2009, 19, 1319–1333.
5 M. A. Baldo, D. F. O’Brien, Y. You, A. Shoustikov, S. Sibley,
M. E. Thompson and S. R. Forrest, Nature, 1998, 395, 151–154.
6 S. Tokito, T. Iijima, Y. Suzuri, H. Kita, T. Tsuzuki and F. Sato, Appl.
Phys. Lett., 2003, 83, 569–571.
7 R. J. Holmes, S. R. Forrest, Y.-J. Tung, R. C. Kwong, J. J. Brown,
S. Garon and M. E. Thompson, Appl. Phys. Lett., 2003, 82, 2422–
2424.
8 Y. R. Sun, N. C. Giebink, H. Kanno, B. Ma, M. E. Thompson and
S. R. Forrest, Nature, 2006, 440, 908–912.
9 P. A. Lane, L. C. Palilis, D. F. O’Brien, C. Giebeler, A. J. Cadby,
D. G. Lidzey, A. J. Campbell, W. Blau and D. D. C. Bradley, Phys.
Rev. B: Condens. Matter Mater. Phys., 2001, 63, 235206.
10 R. J. Holmes, B. D’Andrade, S. R. Forrest, X. Ren, F. Li and
M. E. Thompson, Appl. Phys. Lett., 2003, 83, 3818–3815.
11 X. Gong, J. C. Ostrowski, D. Moses, G. C. Bazan and A. J. Heeger,
Adv. Funct. Mater., 2003, 13, 439–444.
8780 | J. Mater. Chem., 2009, 19, 8772–8781
This journal is ª The Royal Society of Chemistry 2009

View Article Online
ꢀ
56 J. L. Bredas, J. Cornil and A. J. Heeger, Adv. Mater., 1996, 8, 447–
452.
57 H. H. Liao, H. F. Meng, S. F. Horng, W. S. Lee, J. M. Yang,
C. C. Liu, J. T. Shy, F. C. Chen and C. S. Hsu, Phys. Rev. B:
Condens. Matter Mater. Phys., 2006, 74, 245211.
47 H. Inomata, K. Goushi, T. Masuko, T. Konno, T. Imai, H. Sasabe,
J. J. Brown and C. Adachi, Chem. Mater., 2004, 16, 1285–1291.
48 R. A. Klenkler, H. Aziz, A. Tran, Z. D. Popovic and G. Xu, Org.
Electron., 2008, 9, 285–290.
49 Y. Shirota and H. Kagenyma, Chem. Rev., 2007, 107, 953–1010.
50 T. Yasuda, Y. Yamaguchi, D. C. Zou and T. Tsutsui, Jpn. J. Appl.
Phys., 2002, 41, 5626.
58 D. Kabra, M. H. Song, B. Wenger, R. H. Friend and H. J. Snaith,
Adv. Mater., 2008, 20, 3447.
51 A. C.-A. Chen, J. U. Wallace, S. K.-H. Wei, L. C. Zeng, S. H. Chen
and T. N. Blanton, Chem. Mater., 2006, 18, 204–213.
52 A. Weller, in The Exciplex, ed., M. Gordon and W. R. Ware,
Academic Press, New York, 1975, pp. 23–38.
59 L. C. Zeng, F. Yan, S. K. H. Wei, S. W. Culligan and S. H. Chen, Adv.
Funct. Mater., 2009, 19, 1978–1986.
60 R. Fink, Y. Heischkel, M. Thelakkat, H. W. Schmidt, C. Jonda and
M. Huppauff, Chem. Mater., 1998, 10, 3620–3625.
€
53 G. G. Giachino and D. R. Kearns, J. Chem. Phys., 1970, 53, 3886–3891.
54 W. G. Herkstroeter, A. A. Lamola and G. S. Hammond, J. Am.
Chem. Soc., 1964, 86, 4537–4540.
61 J. Pommerehne, H. Vestweber, W. Guss, R. F. Mahrt, H. Bassler,
M. Porsch and J. Daub, Adv. Mater., 1995, 7, 551–554.
62 D. P. Specht, P. A. Martic and S. Farid, Tetrahedron, 1982, 38, 1203–
1206.
€
55 A. Kohler and D. Beljonne, Adv. Funct. Mater., 2004, 14, 11–18.
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 8772–8781 | 8781