An alternative approach to constructing solution processable multifunctional materials: Their structure, properties, and application in high-performance organic light-emitting diodes

Article abstract of DOI:10.1002/adfm.201000474

A new series of full hydrocarbons, namely 4,4'-(9,9'-(1,3-phenylene) bis(9H-fluorene-9,9-diyl))bis(N,N-diphenylaniline) (DTPAFB), N,N'-(4,4'- (9,9'-(1,3-phenylene)bis(9H-fluorene-9,9-diyl))bis(4,1-phenylene))bis(N^ phenylnaphthalen-1-amine) (DNPAFB), 1,3-bis(9-(4-(9H-carbazol-9-yl) phenyl)-9H-fluoren-9-yl)benzene, and 1,3-bis(9-(4-(3,6-di-tert-butyl-9H- carbazol-9-yl)phenyl)-9H-fluoren-9-yl)benzene, featuring a highly twisted tetra- hedral conformation, are designed and synthesized. Organic light-emitting diodes (OLEDs) comprising DNPAFB and DTPAFB as hole transporting layers and tris(quinolin-8-yloxy)aluminum as an emitter are made either by vacuum deposition or by solution processing, and show much higher maximum efficiencies than the commonly used N,N'-di(naphthalen-1-yl)-N,N'- diphenylbiphenyl-4,4'- diamine device (3.6 cd A^-1) of 7.0 cd A^-1 and 6.9 cd A^-1, respectively. In addition, the solution processed blue phosphorescent OLEDs employing the synthesized materials as hosts and iridium (III) bis[(4,6-di- fluorophenyl)-pyridinato-N, C²] picolinate (FIrpic) phosphor as an emitter present exciting results. For example, the DTPAFB device exhibits a brightness of 47 902 cd m^-2, a maximum luminescent efficiency of 24.3 cd A^-1, and a power efficiency of 13.0 lm W^-1. These results show that the devices are among the best solution processable blue phosphorescent OLEDs based on small molecules. Moreover, a new approach to constructing solution processable small molecules is proposed based on rigid and bulky fluorene and carbazole moieties combined in a highly twisted configuration, resulting in excellent solubility as well as chemical miscibility, without the need to introduce any solubilizing group such as an alkyl or alkoxy chain.

Full text of DOI:10.1002/adfm.201000474

www.afm-journal.de
www.MaterialsViews.com
An Alternative Approach to Constructing Solution
Processable Multifunctional Materials: Their Structure,
Properties, and Application in High-Performance Organic
Light-Emitting Diodes
By Shanghui Ye, Yunqi Liu,* Kun Lu, Weiping Wu, Chunyan Du, Ying Liu, Hongtao Liu,
Ti Wu, and Gui Yu
1. Introduction
A new series of full hydrocarbons, namely 4,4′-(9,9′-(1,3-phenylene)
Organic light-emitting diodes (OLEDs) have
bis(9H-fluorene-9,9-diyl))bis(N,N-diphenylaniline) (DTPAFB), N,N′-(4,4′-
been recognized as the next-generation
(9,9′-(1,3-phenylene)bis(9H-fluorene-9,9-diyl))bis(4,1-phenylene))bis(N-
phenylnaphthalen-1-amine) (DNPAFB), 1,3-bis(9-(4-(9H-carbazol-9-yl)
phenyl)-9H-fluoren-9-yl)benzene, and 1,3-bis(9-(4-(3,6-di-tert-butyl-9H-
carbazol-9-yl)phenyl)-9H-fluoren-9-yl)benzene, featuring a highly twisted tetra-
hedral conformation, are designed and synthesized. Organic light-emitting
diodes (OLEDs) comprising DNPAFB and DTPAFB as hole transporting
layers and tris(quinolin-8-yloxy)aluminum as an emitter are made either by
vacuum deposition or by solution processing, and show much higher max-
imum efficiencies than the commonly used N,N′-di(naphthalen-1-yl)-N,N′-
diphenylbiphenyl-4,4′-diamine device (3.6 cd A−1 ) of 7.0 cd A−1 and 6.9 cd A−1 ,
respectively. In addition, the solution processed blue phosphorescent OLEDs
employing the synthesized materials as hosts and iridium (III) bis[(4,6-di-
fluorophenyl)-pyridinato-N, C2] picolinate (FIrpic) phosphor as an emitter
present exciting results. For example, the DTPAFB device exhibits a bright-
ness of 47 902 cd m−2 , a maximum luminescent efficiency of 24.3 cd A−1 , and
a power efficiency of 13.0 lm W−1 . These results show that the devices are
among the best solution processable blue phosphorescent OLEDs based on
small molecules. Moreover, a new approach to constructing solution proc-
essable small molecules is proposed based on rigid and bulky fluorene and
carbazole moieties combined in a highly twisted configuration, resulting in
excellent solubility as well as chemical miscibility, without the need to intro-
duce any solubilizing group such as an alkyl or alkoxy chain.
high quality full color displays. Recently,
OLEDs have even extended their utiliza-
tion to solid state lighting with fluorescent
tube efficiency.[1] To date, the most efficient
OLEDs reported are based on phospho-
rescent emitters, due to the strong spin-
orbit coupling allowing the harvesting of
both singlet and triplet exciton emission
with the ability to reach 100% internal
quantum efficiency.[2–5] This has enabled
the fabrication of green emitting phos-
phorescent OLEDs with external quantum
efficiencies (EQEs) approaching 20%
and luminous efficacy on the order of
70−80 lm W−1 .[2,6] However, extending these
results to blue phosphorescent devices has
proved to be a challenge. The most efficient
blue phosphorescent emitter is still iridium
(III)
bis[(4,6-di-fluorophenyl)-pyridinato-
N, C2] picolinate (FIrpic). Very recently,
Kido’s group reported a FIrpic-based high
efficiency blue phosphorescent OLED in a
precise carrier- and exciton-confining mul-
tilayer structure with a record power effi-
ciency of 44 lm W−1 at the typical brightness
of 1000 cd m−2 .[7] Considering the potential
improvement by outcoupling, the external
power efficiency can be further improved to over 70 lm W−1 ,
close to the general efficiency of fluorescent bulbs. However,
the utilization of multiple layers in such small molecule devices
implies complex device fabrication methodologies, leading to
higher fabrication costs. For low cost general lighting applica-
tions or large area displays, a simple solution-based processing
approach is desirable, provided the efficiency of the devices is not
compromised. Generally, polymers allow easy access to solution
processing of large products with a low cost, but are generally of
lower purity than small molecules. Small molecules are advanta-
geous as they can be highly purified and vacuum deposited in a
multilayer architecture to achieve high efficiency.[8,9] To combine
both advantages in a single molecule is really a challenge.
∗
[ ] S. H. Ye, Prof. Y. Q. Liu, Dr. K. Lu, Dr. W. P. Wu, C. Y. Du, Y. Liu,
H. T. Liu, T. Wu, Prof. G. Yu
Beijing National Laboratory for Molecular Sciences
Key Laboratory of Organic Solids
Institute of Chemistry
Chinese Academy of Sciences
Beijing 100190 (P. R. China)
E-mail: liuyq@iccas.ac.cn
S. H. Ye, C. Y. Du, Y Liu, H. T. Liu, T. Wu
Graduate School
Chinese Academy of Science
Beijing 100039 (P. R. China)
DOI: 10.1002/adfm.201000474
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2010, 20, 3125–3135
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3125

www.afm-journal.de
www.MaterialsViews.com
The majority of OLEDs containing solution processed layers
have contained oligomers or conjugated polymers,[10–12] den-
drimers,[13,14] non-planar star-shaped molecules or conjugated
moleculesbearingsuchtypesofhyper-branchedsidegroups.[15–17]
Among them dendrimers seem to be a good candidate, whose
applications in green, orange, and red phosphorescent OLEDs
are well reported in the literature.[17–19] However, almost all of
them incorporate a solubilizing group such as an alkyl or alkoxy
chain to overcome their poor solubilities and film-forming
abilities. It is well known that alkyl or alkoxy groups are electri-
cally insulating, and introducing such groups into a molecule
will affect its conductivity. Even worse, by attaching alkyl or
alkoxy chains to a molecule, its glass-transistion temperature
(Tg) drops rapidly, which reduces morphological stability, espe-
cially when the molecule is small.[20,21] Thus it is very difficult
for small molecules to possess both morphological stability and
solution processability, especially for blue phosphorescent host
materials whose effective conjugation length must be very short
to encompass the phosphor emitter. As far as we know, there
are few reports on solution processed blue phosphorescent
OLEDs with small hosts in the literature.[22] To develop high
performance blue phosphorescent OLEDs capable of solution
manufacturing is thus very worth pursuing.
Scheme 1. Synthetic route of the target compounds.
In this contribution, we present another way to construct
stable and solution processable small molecules, which is
based on a highly twisted molecular configuration comprising
a rigid and bulky phenyl-substituted fluorene dimer central
core and an aromatic triphenylamine or carbazole side group,
without employing problematic alkyl or alkoxy groups. The
newly synthesized molecules are 4,4′-(9,9′-(1,3-phenylene)
bis(9H-fluorene-9,9-diyl))bis(N,N-diphenylaniline) (DTPAFB),
N,N′-(4,4′-(9,9′-(1,3-phenylene)bis(9H-fluorene-9,9-diyl))bis(4,1-
present in the Experimental section. The design concept is based
on the following consideration: triphenylamine or carbazole are
high mobility hole-transporting materials possessing triplet
energy levels of about 3.04 and 3.01 eV, respectively,[23,24] and
have been widely used as hole transporting layers as well as host
materials for phosphorescent emitters. Fluorene has a rigid and
bulky skeleton with a large triplet energy of 2.9 eV,[25] and thus
connecting two fluorene units at the C−9 position by the phenyl
group to form a highly twisted tetrahedron greatly improves the
chemical and thermal stability as well as the solubility due to its
highly twisted configuration, sharply contrasting with the con-
ventional diarylfluorene derivatives that are very rigid but usu-
ally show low solubility and a high tendency to crystallize due
to strong π−π intermolecular interactions.[26] In our molecular
system, all π−π intermolecular interactions are avoided due to
its highly twisted configuration. To prove our molecular design
rational, we also synthesized an alkyl substituted molecule by
the conventional method, incorporating a tert-butyl group into
one of our molecules, and found that the alkyl group has no
obvious effect on the solubility but has some effect on the elec-
tronic properties, which are to be discussed in the following
section. Moreover, considering the highest occupied molecular
orbital (HOMO) energy levels of the fluorene central core, we
incorporate an aromatic amine or carbazole. It has been dem-
onstrated that the ionization potential of triphenylamine may be
varied between 4.9 and 5.9 eV by substitution,[27] thus allowing
us to further fine tune the properties of our target molecules by
suitable substitution for good charge injection.
phenylene))bis(N-phenylnaphthalen-1-amine)
(DNPAFB),
1,3-bis(9-(4-(9H-carbazol-9-yl)phenyl)-9H-fluoren-9-yl)benzene
(DCPFB), 1,3-bis(9-(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenyl)-
9H-fluoren-9-yl)benzene (DTCPFB). This series of molecules
exhibits excellent thermal and morphological stability, good
film-forming ability and solubility as well as high triplet energy
(2.82−2.33 eV), making them very promising candidates for
optoelectronics. With them, solution processed blue phospho-
rescent OLEDs show satisfying results.
2. Results and Discussion
2.1. Synthesis of the Compounds
The synthetic routes and chemical structures of DTPAFB,
DNPAFB, DCPFB, and DTCPFB are shown in Scheme 1.
From the commercially available starting material isoph-
thaloyl and bromobenzene through Friedel-Crafts benzoyla-
tion gives 1,3-bis(4-bromobenzoyl)benzene, which was treated
with biphenyl-2-yllithium followed by cyclization to produce
1,3-bis(9-(4-bromophenyl)-9H-fluoren-9-yl)benzene (DBPFB). A
palladium-catalyzed amination coupling reaction, in the pres-
ence of Pd(OAc)2 and P(t-Bu)3, between DBPFB and the corre-
sponding diphenylamine, 1-naphthylphenylamine or carbazole
yields the target products. Detailed synthetic procedures are
2.2. Single Crystal X-ray Characterization and Molecular
Packing Properties
For the unique advantages of small molecule OLEDs, more
understanding is needed of the correlation between molecular
©
3126 wileyonlinelibrary.com
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2010, 20, 3125–3135

www.afm-journal.de
www.MaterialsViews.com
Similar to DTPAFB crystals, the DCPFB cell also consists of
four molecules. However, the whole molecule adopts a more
twisted tetrahedral configuration than DTPAFB crystal, with
dihedral angles of 99.1° between fluorenyl rings and phenyl
rings, 113.7° between the central phenyl ring and the carbazole
ring, and 100.4° between the carbazole ring and the fluorenyl
ring. Two carbazole rings stretch out toward the same sides at
an angle of 102.7°. In this crystal system, two kinds of short
contacts exist: fluorenyl ring to fluorenyl ring, edge to edge
(C16···C17′, C17···C16′ in lengths of 3.396 Å and 3.396 Å,
respectively) and carbazole ring to fluorenyl edge to edge
(C1–H···C17, 2.774 Å) in three dimensions, resulting in an
orthorhombic crystalline structure.
The DNAPFB single crystal, however, is different from the
above-mentioned two crystals. It consists of two molecules in
one unit cell and ascribes to a triclinic crystal system. It adopts
a tortile tetrahedral architecture, with a dihedral angle of 103.2°
between the fluorenyl and central phenyl rings, 103.8° between
the peripheral phenyl and central phenyl rings, and 100.2°
between the fluorenyl and peripheral phenyl rings, while the
two 1-naphthylphenylamine moieties are almost perpendicular
to each other, with a dihedral angle of 94.5°. In this single
crystal, we noted an equienergy conformational isomer of the
naphthyl group, as shown in Figure 1c. Two short contacts
between triphenylamine (C16–H, C16) and the naphthyl ring
edge to edge, and naphthyl ring to naphthyl ring edge to edge
exist. The latter interaction is dominated by the bulky naphthyl
rings, leading to a complicated interlaced arrangement with no
π-stacking.
Figure 1. ORTEP drawings and related labeling scheme with 50% prob-
ability of thermal ellipsoids and molecules packing in one unit cell of
DTPAFB (a) and DCPFB (b) and DNPAFB (c); solvents and hydrogen
atoms have been omitted for clarity.
It must be noted that DTCPFB single crystal was not
obtained even though numerous efforts were made, which
may be attributed to weak intermolecular interactions due to
the introduction of the tert-butyl group. However, the DTCPFB
molecule should, in principle, also adopt a similar but looser
architecture in the solid state to the DCPFB crystal, considering
its similar building structure. Based on the above analysis, we
conclude that DTPAFB, DNPAFB and DCPFB all adopt a tet-
rahedral structure, in which all aromatic π−π intermolecular
interactions are greatly weakened and a large free volume is
generated, functionally similar to a star-shaped configuration,[29]
thus providing a chance for the solvent molecule to penetrate,
suggesting excellent solubilities.
structure and physical properties. So we have carefully fos-
tered single crystals of DTPAFB, DNPAFB, and DCPFB in
solution, suitable for X-ray diffraction (XRD) analysis. Each
unit cell of DTPAFB crystals, on average, contains four
molecules with no solvent molecule contained. In the crystal
there are two tetrahedral sp3-hybridized carbons of fluorenyl
with dihedral angles of 104.9° between the fluorenyl units
(C1–C13) and the central phenyl ring (C14–C16), 115.9°
between the central phenyl ring and the peripheral phenyl
ring (C30–C32), and 105.7° between the fluorenyl ring and
peripheral ring, while two fluorenyl rings (C1–C13 and
C17–C29) are twisted from each other by 106.5°, and there
are two triarylamine units extended toward the same side but
twisted with a torsion angle of 93.1° at a distance of 9.9 Å
(Figure 1). On the whole, DTPAFB is a highly twisted mol-
ecule with no symmetric plane or center. In the packing state,
two kinds of short contacts exist between fluorenyl (C2–H,
C3–H) and triphenylamine C (C–H···C) with a distance of
2.859 Å and fluorenyl C and triphenylamine C–H (C···H–C,
2.895 Å). These values are all within the van der Waals radius,
suggesting intermolecular interactions exist between the flu-
orenyl rings and the triphenylamine rings. Alternating inter-
actions between the fluorenyl rings and triphenylamine rings
result in a monoclinic crystal system. In this crystallography,
no cofacial π−π stacking is observed between the fluorenyl
rings and triphenylamine rings, which limits or eliminates
an important mechanism of exciton transfer that can quench
solid-state photoluminescence (PL) or shift its wavelength rel-
ative to isolated molecular behavior.[28]
2.3. Thermochemical Properties and Film-Forming Ability
For a better insight into the structure-property relationship,
thermal properties were measured. All of our compounds show
high thermal stability as determined by thermogravimetric
analysis (TGA), with the decomposition temperature (Td), cor-
responding to 5%-weight-loss, in the range of 452−476 °C
(Table 1). The Tg explored by differential scanning calorimetry
(DSC) reached 147 °C for DTPAFB, 162 °C for DNPAFB, and
223 °C for DTCPFB, while DCPFB showed no obvious thermal
transition during repeated thermal scanning in a temperature
range of 25−300 °C. In the first thermal scanning process,
DTPAFB underwent a glass transition at 147 °C and endo-
thermic recrystallization at 216 °C, followed by melting at 287 °C,
but in the subsequent heating scan, only a glass transition
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2010, 20, 3125–3135
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3127

www.afm-journal.de
www.MaterialsViews.com
T a b le 1 . Thermal, photophysical, and electrochemical properties of the compounds.
Compound
DTPAFB
DNPAFB
DCPFB
Eg (opt.) [eV]
3.65
ET [EV]
2.79
HOMO [c] [eV]
LUMO [d] [eV]
1.74
Tg/Tm/Td [a] [°C]
147/287/452
λ
max (em) [b] [nm]
369
435
359
350
5.39
5.33
5.64
5.58
3.24
2.33
2.09
162/>300/472
N.D./>300/476
223/>300/445
3.47
2.77
2.10
DTCPFB
3.54
2.82
2.11
[a] Obtained from DSC and TGA measurement. [b] Measured in dilute dichloromethane solution. [c] Deduced from cyclic voltammetry. [d] Estimated from the optical band
gap together with HOMO values.
example, 0.345 vs. 0.311 nm for the DTPAFB film before and
after annealing, and 0.322 vs. 0.295 nm for the DNPAFB film.
No aggregation, microcrystallinity or phase separation occurred
in any of the doped films, demonstrating that the dopant was
homogeneously dispersed and that our compounds are good
solid hosts for the phosphor emitter. Because of their excellent
thermal and morphological stabilities, we were able to prepare
homogeneous and amorphous thin films of these compounds
either through thermal vacuum deposition or by spin-coating.
without recrystallization or melting was observed at 147 °C,
indicating that a stable amorphous glass state was formed (for
details, see Figure S1). Compared with N,N′-di(naphthalen-1-yl)-
N,N′-diphenylbiphenyl-4,4′-diamine (NPB) (Tg = 98 °C), a com-
monly used material in OLEDs, our compounds show greatly
improved thermal resistance and a comparably high Tg to the
material TTPPPA (Tg = 202 °C).[30] A high Td makes it possible
to endure high temperatures accompanied by thermal vacuum
sublimation during device fabrication or further purification
by gradient sublimation. All the compounds, except DCPFB,
which showed no glass transition within the experimental tem-
perature range, displayed intrinsic amorphous properties, a
very desirable feature for photoelectronic applications.
2.4. Electrochemical Characterization
To investigate the electrochemical properties and energy levels,
cyclic voltammetry (CV) measurements were performed in a
conventional three electrode configuration in dichloromethane
solution with n-Bu4NPF6 as the supporting electrolyte. DTPAFB,
DNPAFB and DTCPFB showed reversible oxidation and reduc-
tion behavior during the repeated anodic voltage sweep in the
range of 0−1.6 V, but DCPFB displayed an irreversible oxidation
peak, as shown in Figure 3. The irreversible behavior of DCPFB
may be attributed to oxidation polymerization at the electro-
chemically active sites (C−3 and C−6 of the carbazole moiety),
as demonstrated in the literature.[31,32] Irreversible behavior
indicates unstable electrochemical doping, which is unfavorable
Furthermore, these compounds exhibit very good film-
forming ability and chemical miscibility with the phosphores-
cent dopant, FIrpic, as investigated by atomic force micros-
copy (AFM) imaging. As shown in Figure 2, the film spin-cast
from a chlorobenzene solution that contained 30% (by weight)
FIrpic showed a very smooth and homogeneous surface, with a
root-mean-square (RMS) roughness of 0.345 nm for DTPAFB,
0.322 nm for DNPAFB, 0.375 nm for DCPFB, and 0.376 nm
for DTCPFB film. More importantly, the doped films exhib-
ited high thermal stability as proved by annealing the films at
110 °C for 11 h under nitrogen and measuring the surfaces,
which showed negligible change based on the RMS values, for
Figure 2. AFM images of DTPAFB, DNPAFB, DCPFB, and DTCPFB surfaces, each doped with 30% FIrpic: top) before heating; bottom) after heating
at 110 °C for 11 h under nitrogen atmosphere.
©
3128 wileyonlinelibrary.com
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2010, 20, 3125–3135

www.afm-journal.de
www.MaterialsViews.com
2.5. Photophysical Properties
Figure 4 shows the ultraviolet (UV) absorption and pho-
toluminescence (PL) spectra of the compounds in dilute
dichloromethane solutions. DCPFB and DTCPFB displayed
characteristic absorptions of the hybrid units. For example,
DCPFB displayed three distinct absorption bands located at
around 286, 298, and 308 nm, consistent with its parent flu-
orene, 9,9-diphenylfluorene, along with two lowest energy
absorption bands at 329 and 340 nm, which resembles those
of carbazole derivatives.[36,37] DTCPFB also showed similar
absorption patterns with a 2−4 nm hypochromic shift relative
to that of DCPFB due to the steric effect of two tert-butyl groups
at the C−3 and C−6 positions of carbazole. However, DTPAFB
and DNPAFB revealed a featureless absorption peak, which
might be due to spectra overlap of triphenylamine moieties and
fluorene moieties, and only one broad band with a maximum
at 308 nm, but DNPAFB presented another weak band at the
longest wavelength around 350 nm, which can be attributed to
the n−π∗ transition of the 1-naphthylamine unit. The solubili-
ties of our compounds were further investigated by dissolving
our compounds in variant solvents and it was found that they
are soluble in most solvents such as THF, toluene, and chlo-
roform (for details, see Table S1). From the above analysis, we
concluded that the basic construction units’ properties are pre-
served while their solubilities are enhanced via our molecular
design.
Under UV excitation, these compounds, except DNAPFB,
which exhibited blue emission whose peak located at 435 nm,
showed characteristic emission with peaks at 350, 358, and
368 nm for DTCPFB, DCPFB, and DTPAFB, respectively. To
further explore the possible application of these compounds as
phosphorescent host materials, phosphorescent spectra at low
temperature of 77 K were also measured, and the key parame-
ters are listed in Table 1 (for details, see Figure S2). The highest
energy 0−0 phosphorescent emission was taken for calculating
the triplet energy level, giving a value of 2.82 eV for DTCPFB,
2.77 eV for DCPFB, and 2.79 eV for DTPAFB, higher than the
reported value of the commonly used triplet blue-emitter, FIrpic
(2.62 eV), while DNPAFB gives the lowest value of 2.33 eV, cor-
responding to green emission. A higher triplet energy level of a
host material is a provision for effective confinement of the tri-
plet exciton on the guest and for the consequential prevention
of back energy transfer from the dopant to the host materials.[38]
Figure 3. Cyclic voltammograms of DTAPFB, DNPAPFB, DCPFB, and
DTCPFB in dilute dichloromethane solution.
for device performance. By the introduction of two tert-butyl
groups at the C−3 and C−6 positions of the carbazole ring
this problem was overcome, as proved by the reversible redox
behavior of DTCPFB. The introduction of these groups at the
C−3 and C−6 positions not only shifts the oxidation potential to
a slightly lower level but also ensures stable redox behavior.
The HOMO energy levels of the four compounds were calcu-
lated using ferrocene as an internal standard (4.8 eV relative to
vacuum level), to be 5.33, 5.39, 5.64, and 5.58 eV for DNPAFB,
DTPAFB, DCPFB, and DTCPFB respectively, according to the
corresponding onset potentials of 0.93, 0.99, 1.22, and 1.18 V,
respectively, vs. Ag/AgCl reference electrode. The lowest
unoccupied molecular orbit (LUMO) energy levels were esti-
mated based on the HOMO and optical band gaps to be 2.09,
1.74, 2.10, and 2.11 eV for DNPAFB, DTPAFB, DCPFB, and
DTCPFB, respectively. The HOMO and LUMO energy levels
of DCPFB and DTCPFB are in agreement with those of PVK
or other carbazole derivatives (LUMO and HOMO, 2.0 − 2.2
and 5.6 − 5.8 eV),[33,34] and those of DNPAFB and DTPAFB are
very consistent with those of most reported values of triphe-
nylamine or phenylnaphthylamine (NPB and TPD, HOMOs
5.3 − 5.4 eV).[35] Similar HOMO and LUMO values of the com-
ponent moieties to those of the hybrid complex indicates that
the energy levels of the target compounds can be tuned through
a non-conjugated connecting method, and further improved
our molecular design rational.
Figure 4. UV absorption (a) and PL spectra (b) of DTPAFB, DNPAFB, DCPFB, and DTCPFB in dilute dichloromethane solution (2 × 10−5 M).
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2010, 20, 3125–3135
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3129

www.afm-journal.de
www.MaterialsViews.com
Based on the results of the measurements, we could expect that
our molecules should have potential applications in blue phos-
phorescent OLEDs, except DNPAFB, which would serve as a
green or red phosphorescent OLED host.
fabricated similar devices using the thermal vacuum deposition
method employing DNPAFB and DTPAFB as the HTLs and
found comparable performances with the solution processed
devices (for details, see Figure S3). This demonstrates that, with
our compounds, either method may be employed to fabricate
devices. In the NPB/Alq3 system, hole mobility is two orders of
magnitude higher than the electron mobility, which usually cre-
ates recombination areas near the emitting-layer/cathode inter-
face, thus decreasing the device efficiency.[39] Because the EL
spectra for all three devices are almost the same, the improve-
ment in device performance is unrelated to the spectrum of
emission (Figure S4). Thus the significant enhancement in the
brightness and efficiency of devices based on our molecules is
attributed to the better balanced charge recombination in the
emitting layer. As compared to NPB, the effective conjugation
length of our molecules is short, and the twisted configuration
in the phenyl-substituted fluorene dimer core structure could
increase the carrier hopping distance, consequently decreasing
the hole mobility, and thus resulting in a better balance between
the hole and electron fluxes. This proposal was tested by meas-
uring the hole mobility of the three compounds by fabricating
hole-only devices in the configuration of ITO/poly(3,4-ethylene
dioxythiophene):poly(styrene sulfonate) (PEDOT:PSS)/HTL/Au.
The results show that NPB possesses a higher hole mobility
than our compounds (Figure S5). This conclusion is also
consistent with the current density and voltage characteristic
curves in which the NPB device is shown to have a higher cur-
rent density at the same applied voltage than the DNPAFB
or DTPAFB devices. These results suggest that DNPAFB and
DTPAFB are suitable hole-transporting materials; especially
DTPAFB which is a very promising hole-transporting material
capable of replacing NPB.
2.6. Electroluminescent Properties
The uniqueness of our molecules lies in their highly twisted
structure, which endows good solubility and thus enables
solution processing. In order to confirm our hypotheses, we
performed the following experiments. Firstly, we considered
the suitablity of the energy levels of DTPAFB and DNPAFB
for hole-transporting layers: their HOMOs lie in the range
from −5.3 eV to −5.4 eV, very close to that of NPB (HOMO =
−5.4 eV), the extensively used hole-transporting material in
OLEDs. We fabricated devices employing DTPAFB or DNPAFB
as the hole transporting layer (HTL) in a simple configuration
of indium tin oxide (ITO)/HTL (40 nm)/tris(8-hydroxyquin-
oline)aluminum (Alq3) (50 nm)/cathode and explored their
hole injection and transporting abilities. For the purpose of
comparison, we also fabricated a control device using NPB as
the HTL in the same device architecture. Both DTPAFB and
DNPAFB were spin-coated from solution, while NPB and Alq3
were deposited by vacuum deposition. Green emission from
Alq3 with a peak at 524 nm was observed, suggesting that the
charge recombination is localized in the Alq3 emissive layer,
while DTPAFB and DNPAFB functioned purely as hole trans-
porters without causing exciplex formation at the interface with
Alq3. This may be attributed to the proper HOMO energy level
which is favorable for hole injection and transporting from
the ITO anode to the Alq3 layer. The current density–voltage–
luminance (J−V−L) curves are shown in Figure 5 and the key
parameters extracted are listed in Table 2. It is obvious that the
current density of DTPAFB and DNPAFB devices are compa-
rable to the NPB standard device, but the luminance varied
greatly with DNPAFB and DTPAFB showing significantly
higher luminance (46269, 37391 cd m−2 for the DNPAFB and
DTPAFB devices respectively, vs. 28229 cd m−2 for the NPB
device). More importantly, the efficiency, in general, was greatly
enhanced, with a maximum current efficiency of 7.0 and
6.9 cd A−1 for the DTPAFB and DNPAFB devices, respectively,
while the control device is only 3.6 cd A−1 . Moreover, we also
We were interested not only in pursuing a solution proc-
essable hole-transporting material, but also in developing
high performance phosphorescent host materials for OLED
applications, and thus extended our experiments to phospho-
rescent OLEDs. To simplify the fabrication procedure, double-
layer devices were fabricated in the configuration of ITO/
PEDOT:PSS/Host: (10%∼30%, by weight) FIrpic (40 nm)/ETL
(30 nm)/cathode, using 1,3,5-tris(N-phenylbenzimidizol-2-yl)
benzene (TPBI) as the ETL for its low HOMO/LUMO energy
levels, which is beneficial for blocking holes and facilitating
electron injection and transport. PEDOT:PSS is used to smooth
Figure 5. J−V−L characteristics (a) and luminance efficiency as a function of current density (b) curves for the DNPAFB, DTPAFB and NPB based
devices. Open symbol, luminance curves; solid symbol, current density curves.
©
3130 wileyonlinelibrary.com
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2010, 20, 3125–3135

www.afm-journal.de
www.MaterialsViews.com
T a b le 2 . EL performances of DTPAFB, DNPAFB, and NPB devices in the configuration of ITO/HTL/Alq/cathode.
max [cd m−2
37391.4
46269.0
28229.4
]
LEmax [cd A−1
]
PEmax [lm W−1
]
LE [cd A−1 ] [b]
PE [lm W−1 ] [b]
λmax (em) [nm] [c]
HTL
V
on [V] [a]
L
DTPAFB
DNPAFB
NPB
4.2
7.0
6.9
3.6
2.8
2.4
2.2
6.3
6.1
3.3
2.7
2.3
2.0
524
524
524
4.8
3.2
[a] Recorded at 1 cd m−2 . [b] Obtained at a brightness of 1000 cd m−2 . [c] EL spectra maxima recorded at 8 V.
the ITO surface. The synthesized compounds with high triplet
energy (DTPAFB, DCPFB, and DTCPFB) were subjected to fab-
rication of phosphorescent OLEDs. For a host-dopant system,
doping concentration greatly affects device performance, and
thus we investigated device performance using four different
doping concentrations.
curves are plotted as a function of luminance in Figure 7, and
the key parameters extracted are summarized in Table 3. The
highest luminance efficiency was achieved in the 30% doped
DTPAFB device with maximum luminance efficiency of
24.3 cd A−1 , a power efficiency of 13.0 lm W−1 at a current density
of 5.3 mA cm−2 , corresponding to a luminance of 1286 cd m−2 .
As far as we know, this performance is among the best for solu-
tion processed blue phosphorescent OLEDs based on small
molecular weight host materials and the FIrpic dopant.[22] The
comprehensive performance is comparable to the most efficient
blue phosphorescent OLEDs based on a carbazole dendritic
host (15.4 lm W, 27.6 cd A).[42] The 10% doped device displayed
the lowest maximum efficiency of 14.5 cd A−1 , 7.9 lm W−1 . In
general, the device performance is enhanced with increasing
emitter concentration. The best performance among the four
DTPAFB devices is ascribed to the 30% doped device.
With these optimized device parameters in mind, we fab-
ricated two other devices employing DCPFB and DTCPFB as
host and FIrpic phosphorescent emitter as dopant. The device
performances are also presented in Figure 6 with the key
parameters summarized in Table 3. Both devices exhibit high
performance comparable to the DTPAFB device; for instance,
the DCPFB device displays a maximum luminance efficiency
of 25.6 cd A−1 , 12.7 lm W−1 , while the DTCPB device shows
a maximum luminance efficiency of 19.5 cd A−1 , 10.7 lm W−1 .
The maximum brightness reached is 15 717 cd m−2 at 14.1 V
and 22 040 cd m−2 at 10.2 V, for DCPFB and DTCPFB respec-
tively. The turn-on voltage of the DTCPFB device (Von = 4.5 V) is
lower relative to the DCPFB device (Von = 5.1 V) which could be
attributed to the HOMO energy level of this device being better
matched with the ITO anode, as discussed above. Although the
DCPFB device shows a higher current efficiency than DTPAFB
device, its power efficiency is a little lower than the latter due
Figure 6 presents the J−V−L characteristics of DTPAFB
devices doped with 10%, 15%, 20%, and 30% of FIrpic phos-
phors. According to the J−V characteristic curves of the above
device systems, an increase in the dopant concentration leads
to a decrease in driving voltage, with the minimum driving
voltage achieved at 30% doping level. The opposite trend was
observed in the V−L curves, that is, the luminance increased
with increasing doping level, with the highest luminance
reached in the 30% doped device, with a turn-on voltage
(voltage at 1 cd m−2 ) decreasing from 4.8 V to 4.5 V, 4.1 V, and
3.7 V for the 10%, 15%, 20%, and 30% doped DTPAFB devices
respectively. This indicates that the charge might be injected
directly into the dopant molecules. When the doping concentra-
tion is beyond 10% an additional electron transporting channel
is formed by charges hopping between the dopant sites,[40,41] in
addition to the hole transporting channel on the host materials.
Moreover, FIrpic is an emitter, and increasing its concentration
simultaneously increased the triplet concentration to generate
more excitons, and thus the luminance increased. But further
increasing the doping concentration would induce triplet-
triplet annihilation and concentration quenching, and thus
the luminance would not be further increased. So the optimal
doping concentration is 30%. The maximum luminance of the
DTPAFB device is 47 902 cd m−2 at 12 V corresponding to the
30% doping level, while the 10% doped device gave a maximum
luminance of 19 822 cd m−2 at 13.5 V. The other two devices
exhibited intermediate luminance. The luminance efficiency
Figure 6. J−V−L characteristics for the DTPAFB, DCPFB, and DTCPFB devices.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2010, 20, 3125–3135
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3131

www.afm-journal.de
www.MaterialsViews.com
Figure 7. Luminance efficiency as a function of luminance characteristics (a) and EL spectra of DTPAFB, DCPFB, and DTCPFB devices (b) at 8 V.
to the higher operating voltage. In general, all the devices show
stable efficiency roll-off, for example, the maximum efficiency
decreased from 24.3 cd A−1 to 24.1 A−1 at 1000 cd m−2 for the
DTPAFB device. Another thing worth noting is that the max-
imum efficiency was achieved at high luminance (in the range
of 100 ∼ 1000 cd m−2 ), which is more practical for commercial
applications;[43] even at high brightness of 1000 cd m−2 , these
devices display gratifying levels of efficiency (22.6 cd A−1 and
16.8 cd A−1 for the DCPDB and DTCPFB devices respectively).
For comparison, we also prepared a control device employing
poly(9-vinylcarbazole) (PVK) as a host material and FIrpic as
the phosphorescent dopant in the same device architecture, and
found the maximum luminance efficiency to be 13.8 cd A−1 or
8.03 lm W−1 (for details, see Figure S6), a value comparable to
that reported by Wang, et al. [42] Compared with the PVK device,
our devices show pronounced improvement in luminance
efficiency and brightness, with decreased operating voltage.
As demonstrated in the literature, the deep HOMO energy
level of PVK (5.9 eV)[44] accompanied by a hole injection bar-
rier accounts for its poor efficiency. For our host materials the
hole injection problem is overcome. The HOMO energy level
range (5.3 ∼ 5.6 eV) is closer to the Fermi level of commonly
used anodes such as ITO and/or extensively used hole-injection
layers such as PEDOT:PSS (5.2 eV) (for details, see Figure S7).
This makes high current densities and consequently high lumi-
nance levels possible at moderate operation voltages in our
blue phosphorescent OLEDs (Von = 3.7 V, 100 cd m−2 at 4.7 V,
and 47 902 cd m−2 at 12 V for the DTPAFB device). Among the
three host materials, DTPAFB showed the best performance;
its HOMO energy level being well matched to the ITO anode
relative to those of DCPFB and DTCPFB might be respon-
sible for this difference. Although DTCPFB incorporates more
four alkyl groups than DCPFB, solubility does not seem to be
a determining factor, for both were soluble enough to form
high quality films for solution-cast device fabrication, yet the
electrochemical properties of DTCPFB were better tuned than
the latter, leading to lower operation voltages (5.6 V vs. 6.4 V at
100 cd m−2 , 6.6 V vs. 7.8 V at 1000 cd m−2 ). In all, all of our
synthesized compounds were capable of forming beautiful thin
films for solution processable high performance OLEDs.
As for the EL spectra of the phosphorescent OLEDs based
on the three host materials, all devices exhibited a similar max-
imum luminescence wavelength at around 470 nm, which orig-
inates from the triplet emission of the FIrpic emitter. Further-
more, no additional emission coming from the host materials
was observed, indicative of efficient energy transfer from the
host to the FIrpic dopant. However, the relative intensity of the
shoulder at 500 nm increased in the order of DCPFB, DTPAFB,
and DTCPFB host materials, which is attributed to an optical
effect due to the recombination zone shifting.[45] In the case of
FIrpic-based devices, Wu et al. have shown that as the distance
between the cathode and light emitting layer increases, the rel-
ative intensity of the shoulder at 500 nm in the EL spectrum
increases.[46] Thus, the variation in the EL spectrum can be
explained as follows: as the hole transporting ability increases
from DCPFB to DTPAFB to DTCPFB (inferred from the J–V
curves) the recombination zone would shift from the HTL/
EML side to the center of the EML due to different charge (hole)
carrier mobilities of the host materials. But on the whole, no
obvious difference exists in the EL spectra of the three devices.
T a b le 3 . EL characteristics of FIrpic doped blue phosphorescent devices.
max [cd m−2
47902.8
33680.2
23013.6
19822.8
15717.6
22040.4
]
LEmax [CD A−1
]
PEmax [lm W−1
]
LE [cd A−1 ] [c]
24.1
PE [lm W−1 ] [c]
Host
Flrpic [%]
Von [V] [a]
3.7
CIE [x, y] [b]
0.187, 0.375
0.187, 0.375
0.187, 0.375
0.187, 0.375
0.189, 0.365
0.193, 0.369
L
DTPAFB
DTPAFB
DTPAFB
DTPAFB
DCPFB
DTCPFB
30
20
15
10
30
30
24.3
13.0
11.2
8.6
12.3
9.6
7.1
5.0
9.5
8.3
4.1
20.4
19.5
4.5
17.2
16.7
4.8
14.5
7.9
12.4
5.1
25.6
12.7
10.7
22.6
4.5
19.5
16.8
[a] Recorded at 1 cd m−2 . [b] Measured at 8 V. [c] Values recorded at a luminance of 1000 cd m−2
.
©
3132 wileyonlinelibrary.com
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2010, 20, 3125–3135

www.afm-journal.de
www.MaterialsViews.com
scanning calorimeter (DSC7). The voltammetry characterization of the
redox process was performed with a computer-controlled CHI 600C
Electrochemical Workstation using a Pt disk working electrode of
0.01 cm2, a Pt wire counter electrode and a Ag/AgCl reference electrode
in a conventional three-electrode cell. XRD measurements were carried
out in the reflection system at RT using a Rigaku MM-007 XRD system
(Mo Kα radiation, λ = 0.71073 Å). Tapping-mode AFM was carried out
on spots randomly chosen.
We found the EL spectra to be independent of applied voltage
and current density within experimental error. As shown in
Figure 7 the Commission Internationale de L’Eclairage (CIE)
coordinates are (0.187, 0.375), (0.189, 0.365), and (0.193, 0.369)
for DTPAFB, DCPFB, and DTCPFB devices, respectively, cor-
responding to light blue emission.
Device Fabrication and Measurement: DTPAFB, DNPAFB, DCPFB, and
DTCPFB were synthesized and purified by recrystallization from toluene
solution. 1,3,5-tris(1-phenyl-1H-benzo[d]imidazol-2-yl)benzene (TPBI)
was purchased from Aldrich Chemical Company and was purified by
thermal sublimation before use. FIrpic was synthesized by our group
and was purified by sublimation before use. In a general procedure,
indium-tin oxide (ITO)-coated glass substrates were etched, patterned,
and washed with detergent, deionized water, acetone, and ethanol in
turn. PEDOT:PSS (Baytron P Al 4083) was spin coated onto the ITO
substrates under ambient atmosphere yielding a layer of 30 nm, which
were baked at 120 °C for 30 min to remove residual water. On top of the
PEDOT:PSS layer an emitting layer was spin coated from chlorobenzene
solution and was annealed at 80 °C under vacuum for 30 min to remove
any residual solvent. The thickness of the EML was about 40 nm.
Subsequently, a 30-nm-thick electron transporting and hole blocking
layer was thermally evaporated on top of the EML at a base pressure of
8 × 10−5 Pa. In all devices, a thin layer of Ca about 8 nm in thickness was
deposited as an electron injection layer; then a 100-nm-thick Ag capping
layer was deposited through a 2-mm array in a shadow mask. The active
area of the device was 6 mm2. A quartz crystal oscillator placed near the
substrate was used to determine the thickness of each layer, which was
calibrated ex situ using an Ambios Technology XP-2 surface profilometer.
UV–vis absorption and fluorescence spectra were collected with a
Hitachi U−3010 and Hitachi F−4500 spectrophotometer, respectively. EL
spectra and chromaticity coordinates were measured with a SpectraScan
PR650 photometer. J–V–L measurements were recorded simultaneously
using a Keithley 4200 semiconductor parameter analyzer and a Newport
multifunction 2835−C optical meter, with luminance measured in the
forward direction. All device characterizations were carried out under
ambient laboratory air at room temperature.
3. Conclusions
A new series of solution processable small molecules based on
a highly twisted phenyl-linked fluorene dimer skeleton func-
tionalized by grafting a peripheral group of triphenylamine or
carbazole have been designed and synthesized. This series of
compounds feature an aromatically rigid and bulky π-plane in
a tetrahedral configuration without any alkyl or alkoxy chain,
yet exhibit excellent solubility, thermal and morphological sta-
bility as well as electrochemical stability, enabling both solution
processing and thermal vacuum deposition methods to fabricate
OLEDs. OLEDs fabricated by both methods in a simple device
structure of HTL/Alq employing DNPAFB and DTPAFB as a
HTL show higher efficiency relative to an NPB control device.
Compared with NPB, our materials show great improvements
in Tg and efficiency, and thus they could be ideal replacements
for the widely used NPB in optoelectronic devices. Moreover,
we fabricated blue phosphorescent devices utilizing a solution
processing method incorporating our molecules as host mate-
rials and FIrpic as a dopant in a bilayer structure, with satis-
fying results. The best blue phosphorescent OLED is attributed
to the DTPAFB device doped with 30% FIrpic phosphor, with
maximum power efficiency of 13.0 lm W−1 , luminance effi-
ciency of 24.3 cd A−1 , and a maximum brightness of 47 902 cd
m−2 at 12 V. The DCPFB device also shows high efficiency of
12.7 lm W−1 , 25.6 cd A−1 , but DTCPFB displays a humble effi-
ciency compared with the former two devices. All the devices
present relatively low turn-on voltage in the range of 3.7 ∼ 5.1 V,
making low voltage operation possible. This represents a new
step forward toward solution processable OLEDs based on
small molecules. Furthermore, we discussed the molecular
structure, photophysical, and electrochemical properties of our
materials as well as their multifunctional applications such as
acting as hole transporting layers and as host materials for blue
phosphorescent OLEDs. Most importantly, we provided a new
method to constructing solution processable small molecules
based on rigid and bulkily aromatic moieties combined in a
highly twisted configuration, without incorporating any solubi-
lizing group.
Preparation of 1,3-Bis(4-bromobenzoyl)benzene: (Modified Procedure)[47]
20.7 g (0.155 mol) anhydrous powdered aluminum chloride was added
to a stirred solution of 15 g (0.074 mol) isophthaloyl dichloride and
100 mL bromobenzene. After an exothermic reaction, the solution was
stirred at room temperature for 9 h and then heated at 90 °C for 2 h.
After cooling, the solution was poured into cold methanol to precipitate
a white solid which was isolated by filtration. The crude product was
recrystallized from toluene to yield white crystals. Yield: 31.4g, 95.3%.
1H NMR (400 MHz, CDCl3, δ): 8.13 (s, 1H), 8.01−7.99 (d, J = 8 Hz, 2H),
7.70−7.64 (m, 9H). 13C NMR (400 MHz, CDCl3, δ): 195.9, 194.8, 138.1,
137.7, 137.5, 137.0, 135.7, 133.9, 133.7, 133.5, 133.1, 132.0, 131.7, 131.2,
131.0, 130.0, 128.9, 128.8, 128.7, 128.3. MS (EI): m/z (100%) calcd. for
C20H12Br2O2, 444.1; found, 446 (100).
1,3-Bis(9-(4-bromophenyl)-9H-fluoren-9-yl)benzene
(DBPFB):
2-bromobiphenyl (3.495 g, 15 mmol) and THF (50 mL) were mixed and
cooled to −78 °C and then n-BuLi (2.5 M in hexane, 6 mL, 15 mmol)
was added dropwise. The whole solution was stirred at this temperature
for 45 min followed by drop-wise addition of a solution of 1,3-Bis(4-
bromobenzoyl)benzene (4.441 g, 10 mmol) in THF (100 mL). The
resulting mixture was gradually warmed to ambient temperature and
kept stirring for 12 h, after which 50 mL saturated aqueous NaHCO3
was added to quench the reaction. The mixture was extracted with
CH2Cl2 (3 × 60 mL) and the combined organic layers were dried over
MgSO4, filtered and evaporated under reduced pressure. The crude
residue was dissolved in acetic acid (100 mL) and a catalytic amount of
aqueous HCl (12 N) was added, and then the whole mixture was warmed
to reflux for 10 h. After cooling to room temperature, the mixture was
condensed under reduced pressure. The crude residue was purified
by recrystallization from toluene to afford 6.874 g pure DBPFB, white
solid (yield: 96%). 1H NMR (400 MHz, CDCl3, δ): 7.75−7.73 (d, J =
4. Experimental Section
General Information: 1H NMR and 13C NMR measurements
were carried out on a Bruker DMX 400 NMR spectrometer. MS
spectrometry (MALDI-TOF-MS) was carried out on a Bruker BIFLEX
III mass spectrometer. Elemental analyses were carried out on a Carlo-
Erba 1160 elemental analyzer. TGA was carried out using a Perkin-
Elmer thermogravimeter (Model TGA7) under a dry nitrogen gas
flow at a heating rate of 10 °C min−1. Tg and melting phase-transition
temperature (Tm) were determined by differential scanning calorimetry
(DSC) at a heating rate of 10 °C min−1 using a Perkin-Elmer differential
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2010, 20, 3125–3135
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3133

www.afm-journal.de
www.MaterialsViews.com
3,6-Di-tert-butyl-9H-carbazole: 3,6-di-tert-butyl-9H-carbazole
8.0 Hz, 4H), 7.39 (s, 1H), 7.37−7.34 (t, J = 7.2 Hz, 4H), 7.31−7.29 (d, J =
8.0 Hz, 4H), 7.24−7.22 (d, J = 7.2 Hz, 3H), 7.20−7.18 (d, J = 8.0 Hz, 4H),
6.96−6.94 (d, J = 8.0 Hz, 5H), 6.94 (s, 1H), 6.84−6.82 (m, 2H). 13C NMR
(400 MHz, CDCl3, δ): 150.7, 145.5, 145.2, 140.2, 131.4, 130.4, 130.1,
128.1, 127.9, 126.1, 125.9, 120.9, 120.5, 65.1. MS (EI): m/z (100%) calcd.
for C44H28Br2, 716.5; found, 716 (100).
was prepared according to previously reported procedure.[48]
Supporting Information
4,4′-(9,9′-(1,3-Phenylene)bis(9H-fluorene-9,9-diyl))bis(N,N-
Supporting Information is available from the Wiley Online Library or
from the author.
diphenylaniline) (DTPAFB):
A mix of DBPFB (1 mmol, 0.716 g),
diphenylamine (0.347 g, 2.05 mmol), Pd(OAc)2 (0.014 g, 4 mmol) and
NaOtBu (0.169 g, 2.05 mmol) in 4 mL of p-xylene and 0.7 mL PtBu3
was stirred at 120 °C for 8 h under nitrogen atmosphere. After cooling
to room temperature 20 mL water and 60 mL dichloromethane were
added, and the organic phase was separated and concentrated under
reduced pressure to give a gray solid, which was subjected to column
chromatography (dichloromethane and petroleum ether as eluent,
1:8, v/v, Rf = 0.3) to yield the target product of DTPAFB. 1H NMR
(400 MHz, CDCl3, δ): 7.74 (d, J = 7.2 Hz, 4H), 7.51 (s, 1H), 7.33 (t,
J = 7.2 Hz, 4H), 7.28 (d, J = 4.4 Hz, 4H), 7.23−7.17 (m, 12 H), 7.05 (d, J =
5.2 Hz, 8H), 7.00−6.96 (m, 8H), 6.90 (t, J = 4.4 Hz, 5H), 6.8−6.79 (m, 2H).
13C NMR (400 MHz, CDCl3, δ): 151.31, 147.86, 146.34, 146.20,
140.25, 140.10, 130.80, 129.36, 129.15, 127.74, 127.53, 126.30, 125.46,
124.35, 123.37, 122.86, 120.24, 65.08. MS (MALDI-TOF): m/z calcd.
for C68H48N2, 892.4; found, 892.7 (m+). Elemental anal. calcd (%) for
C68H48N2: C, 91.45; H, 5.42; N, 3.14; Found: C, 91.43; H, 5.44; N, 3.12.
N,N′-(4,4′-(9,9′-(1,3-Phenylene)bis(9H-fluorene-9,9-diyl))bis(4,1-
phenylene))bis(N-phenylnaphthalen-1-amine) (DNPAFB): DNPAFB was
prepared similarly to DTPAFB using N-phenylnaphthalen-1-amine to
replace diphenylamine. 1H NMR (400 MHz, CDCl3, δ): 7.88 (d, J =
8.4 Hz, 2H), 7.84 (d, J = 8.4 Hz, 2H), 7.75 (d, J = 8.0 Hz, 2H), 7.69 (d, J =
7.6 Hz, 4H), 7.50 (s, 1H), 7.44 (t, J = 7.6 Hz, 2H), 7.37 (t, J = 7.6 Hz,
2H), 7.31−7.21 (m, 8H), 7.18 (t, J = 8.0 Hz, 7H), 7.14 (s, 1H), 7.04 (t, J =
7.6 Hz, 4H), 6.99 (d, J = 7.6 Hz, 4H), 6.91−6.86 (m, 10H), 6.82 (d, J =
7.6 Hz, 1H), 6.73 (d, J = 7.6 Hz, 2H). 13C NMR (400 MHz, CDCl3, δ):
151.24, 148.48, 146.91, 146.22, 143.61, 140.18, 139.29, 135.42, 131.39,
131.06, 129.19, 129.07, 128.51, 127.63, 127.42, 127.33, 126.53, 126.27,
125.26, 124.34, 121.65, 121.44, 120.15, 64.96. MS (MALDI-TOF): m/z
calcd. for C76H52N2, 992.4; found, 992.6 (M+). Elemental anal. calcd (%)
for C76H52N2: C, 91.90; H, 5.28; N, 2.82; Found: C, 91.89; H, 5.30; N,
2.81.
Acknowledgements
We acknowledge financial support from the National Natural Science
Foundation of China (20825208, 60736004, 20721061), the National
Major State Basic Research Development Program (2009CB623603),
and the Chinese Academy of Sciences.
Received: March 12, 2010
Revised: May 15, 2010
Published online: July 30, 2010
[1] S. Reineke, F. Lindner, G. Schwartz, N. Seidler, K. Walzer, B. Lüssem,
K. Leo, Nature 2009, 459, 234.
[2] C. Adachi, M. A. Baldo, M. E. Thompson, S. R. Forrest, J. Appl. Phys.
2001, 90, 5048.
[3] S. O. Jeon, K. S. Yook, C. W. Joo, J. Y. Lee, K. Y. Ko, J. Y. Park,
Y. G. Baek, Appl. Phys. Lett. 2008, 93, 063306.
[4] Y. Kawamura, K. Goushi, J. Brooks, J. J. Brown, H. Sasabe, C. Adachi,
Appl. Phys. Lett. 2005, 86, 071104.
[5] Y. R. Sun, N. C. Giebink, H. Kanno, B. W. Ma, M. E. Thompson,
S. R. Forrest, Nature 2006, 440, 908.
[6] M. Ikai, S. Tokito, Y. Sakamoto, T. Suzuki, Y. Taga, Appl. Phys. Lett.
2001, 79, 156.
[7] S. J. Su, E. Gonmori, H. Sasabe, J. Kido, Adv. Mater. 2008, 20, 4189.
[8] D. A. Pardo, G. E. Jabbour, N. Peyghambarian, Adv. Mater. 2000, 12,
1249.
[9] B. J. de Gans, P. C. Duineveld, U. S. Schubert, Adv. Mater. 2004, 16,
203.
1,3-Bis(9-(4-(9H-carbazol-9-yl)phenyl)-9H-fluoren-9-yl)benzene
(DCPFB): A mixture of DBPFB (1 mmol, 0.716 g), 9H-carbazole (0.343 g,
2.05 mmol), Pd(OAc)2 (0.014 g,
4 mmol) and K2CO3 (0.345 g,
2.50 mmol) in 4 mL of p-xylene and 0.7 mL PtBu3 was stirred at 120 °C for
8 h under nitrogen atmosphere. After cooling to room temperature 20 mL
water and 60 dichloromethane were added, and the organic phase was
separated and concentrated under reduced pressure to give a gray solid,
which was subjected to column chromatography (dichloromethane and
petroleum ether as eluents, 1:8, v/v, Rf = 0.3) to yield the target product
[10] R. H. Friend, R. W. Gymer, A. B. Holmes, J. H. Burroughes,
R. N. Marks, C. Taliani, D. D. C. Bradley, D. A. Dos Santos,
J. L. Brédas, M. Lögdlund, W. R. Salaneck, Nature 1999, 397, 121.
[11] C. Ulbricht, B. Beyer, C. Friebe, A. Winter, U. S. Schubert, Adv. Mater.
2009, 21, 4418.
[12] X. H. Yang, D. C. Müller, D. Neher, K. Meerholz, Adv. Mater. 2006,
18, 948.
1
of DCPFB. H NMR (400 MHz, CDCl3, δ): 8.13 (d, J = 7.6 Hz, 4H), 7.81
(d, J = 7.6 Hz, 4H), 7.52 (s, 1H), 7.41−7.36 (m, 24H), 7.34−7.28 (m, 8H),
7.24 (d, J = 1.7 Hz, 1H), 7.09−7.02 (m, 3H). 13C NMR (400 MHz, CDCl3,
δ): 150.93, 145.86, 145.20, 140.89, 140.32, 139.58, 136.24, 129.72, 127.92,
126.71, 126.24, 125.94, 123.46, 120.43, 120.01, 119.55, 110.67, 109.91,
65.37. MS (MALDI–TOF): m/z (100%) calcd. for C68H44N2, 889.09;
found, 888.8 (M+). Elemental anal. calcd (%) for C68H44N2: C, 91.86; H,
4.99; N, 3.15; Found: C, 91.86, H, 4.50, N, 3.14.
1,3-Bis(9-(4-(3,6-di-tert-butyl-9H-carbazol-9-yl)phenyl)-9H-fluoren-9-yl)
benzene (DTCPFB): DTCPFB was prepared as per DCPFB using 3,6-di-tert-
butyl-9H-carbazole to replace 9H-carbazole. 1H NMR (400 MHz,
CDCl3, δ): 8.11 (d, J = 1.4 Hz, 4H), 7.80 (d, J = 7.3 Hz, 4H), 7.55 (s,
1H), 7.40 (s, 4H), 7.39−7.38 (m, 9H), 7.36−7.35 (m, 5H), 7.32 (t,
J = 7.2 Hz, 7H), 7.29 (s, 2H), 7.02 (m, 1H), 6.98−6.96 (m, 2H), 1.43
(s, 36H). 13C NMR (400 MHz, CDCl3, δ): 150.83, 145.81, 144.38, 142.73,
140.15, 139.07, 136.62, 129.46, 127.75, 127.67, 126.11, 125.74, 123.42,
123.24, 120.26, 116.13, 109.18, 65.20, 34.63, 31.94. MS (MALDI-TOF):
m/z calcd. for C84H76N2, 1112.6; found, 1112.7 (M+). Elemental anal.
calcd (%) for C84H76N2: C, 90.60; H, 6.88; N, 2.52 Found: C, 90.58; H,
6.89; N, 2.51.
[13] S. C. Lo, R. E. Harding, C. P. Shipley, S. G. Stevenson, P. L. Burn,
I. D. W. Samuel, J. Am. Chem. Soc. 2009, 131, 16681.
[14] S. J. Lee, J. S. Park, M. Song, I. A. Shin, Y. I. Kim, J. W. Lee,
J. W. Kang, Y. S. Gal, S. Kang, J. Y. Lee, S. H. Jung, H. S. Kim,
M. Y. Chae, S. H. Jin, Adv. Funct. Mater. 2009, 19, 2205.
[15] Z. Y. Ge, T. Hayakawa, S. Ando, M. Ueda, T. Akiike, H. Miyamoto,
T. Kajita, M. A. Kakimoto, Adv. Funct. Mater. 2008, 18, 584.
[16] C. Huang, C. G. Zhen, S. P. Su, K. P. Loh, Z. K. Chen, Org. Lett.
2005, 7, 391.
[17] A. W. Freeman, S. C. Koene, P. R. L. Malenfant, M. E. Thompson,
J. M. J. Fréchet, J. Am. Chem. Soc. 2000, 122, 12385.
[18] J. Q. Ding, J. Gao, Y. X. Cheng, Z. Y. Xie, L. X. Wang, D. G. Ma,
X. B. Jing, F. S. Wang, Adv. Funct. Mater. 2006, 16, 575.
[19] A. Adronov, J. M. J. Fréchet, Chem. Commun. 2000, 1701.
[20] D. Katsis, Y. H. Geng, J. J. Ou, S. W. Culligan, A. Trajkovska,
S. H. Chen, L. J. Rothberg, Chem. Mater. 2002, 14, 1332.
[21] A. L. Kanibolotsky, R. Berridge, P. J. Skabara, I. F. Perepichka,
D. D. C. Bradley, M. Koeberg, J. Am. Chem. Soc. 2004, 126, 13695.
©
3134 wileyonlinelibrary.com
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2010, 20, 3125–3135

www.afm-journal.de
www.MaterialsViews.com
[22] L. D. Hou, L. Duan, J. Qiao, W. Li, D. Q. Zhang, Y. Qiu, Appl. Phys.
Lett. 2008, 92, 263301.
[23] S. Tokito, T. Iijima, Y. Suzuri, H. Kita, T. Tsuzuki, F. Sato, Appl. Phys.
Lett. 2003, 83, 569.
[24] P. I. Shih, C. H. Chien, T. I. Wu, C. F. Shu, Adv. Funct. Mater. 2007,
17, 3514.
[25] P. I. Shih, C. H. Chien, C. Y. Chuang, C. F. Shu, C. H. Yang, J. H. Chen,
Y. Chi, J. Mater. Chem. 2007, 17, 1692.
[26] F. Liu, C. Tang, Q. Q. Chen, S. Z. Li, H. B. Wu, L. H. Xie, B. Peng,
W. Wei, Y. Cao, W. Huang, Org. Electron. 2009, 10, 256.
[27] E. T. Seo, R. F. Nelson, J. M. Fritsch, L. S. Marcoux, D. W. Leedy,
R. N. Adams, J. Am. Chem. Soc. 1966, 88, 3498.
[28] Z. X. Wang, H. X. Shao, J. C. Ye, L. Zhang, P. Lu, Adv. Funct. Mater.
2007, 17, 253.
[36] P. I. Shih, C. L. Chiang, A. K. Dixit, C. K. Chen, M. C. Yuan,
R. Y. Lee, C. T. Chen, E. W. G. Diau, C. F. Shu, Org. Lett. 2006, 8,
2799.
[37] K. T. Wong, Y. M. Chen, Y. T. Lin, H. C. Su, C. C. Wu, Org. Lett. 2005,
7, 5361.
[38] R. J. Holmes, S. R. Forrest, Y. J. Tung, R. C. Kwong, J. J. Brown,
S. Garon, M. E. Thompson, Appl. Phys. Lett. 2003, 82, 2422.
[39] S. Barth, P. Müller, H. Riel, P. F. Seidler, W. Riess, H. Vestweber,
H. Bässler, J. Appl. Phys. 2001, 89, 3711.
[40] R. J. Holmes, B. W. D’Andrade, S. R. Forrest, X. Ren, J. Li,
M. E. Thompson, Appl. Phys. Lett. 2003, 83, 3818.
[41] Y. L. Tung, L. S. Chen, Y. Chi, P. T. Chou, Y. M. Cheng, E. Y. Li,
G. H. Lee, C. F. Shu, T. I. Wu, A. J. Carty, Adv. Funct. Mater. 2006,
16, 1615.
[29] Z. Ge, T. Hayakawa, S. Ando, M. Ueda, T. Akiike, H. Miyamoto,
T. Kajita, M. A. Kakimoto, Chem. Mater. 2008, 20, 2532.
[30] Q. X. Tong, S. L. Lai, M. Y. Chan, K. H. Lai, J. X. Tang, H. L. Kwong,
C. S. Lee, S. T. Lee, Chem. Mater. 2007, 19, 5851.
[31] M. H. Tsai, H. W. Lin, H. C. Su, T. H. Ke, C. C. Wu, F. C. Fang,
Y. L. Liao, K. T. Wong, C. I. Wu, Adv. Mater. 2006, 18, 1216.
[32] K. Brunner, A. van Dijken, H. Börner, J. J. A. M. Bastiaansen,
N. M. M. Kiggen, B. M. W. Langeveld, J. Am. Chem. Soc. 2004, 126,
6035.
[33] X. H. Yang, D. Neher, D. Hertel, T. K. Däubler, Adv. Mater. 2004, 16,
161.
[34] G. F. He, S. C. Chang, F. C. Chen, Y. F. Li, Y. Yang, Appl. Phys. Lett.
2002, 81, 1509.
[42] J. Q. Ding, B. H. Zhang, J. H. Lü, Z. Y. Xie, L. X. Wang, X. B. Jing,
F. S. Wang, Adv. Mater. 2009, 21, 4983.
[43] P. I. Shih, C. Y. Chuang, C. H. Chien, E. W. G. Diau, C. F. Shu, Adv.
Funct. Mater. 2007, 17, 3141.
[44] A. van Dijken, J. J. A. M. Bastiaansen, N. M. M. Kiggen,
B. M. W. Langeveld, C. Rothe, A. Monkman, I. Bach, P. Stössel,
K. Brunner, J. Am. Chem. Soc. 2004, 126, 7718.
[45] J. Lee, J. I. Lee, J. Y. Lee, H. Y. Chu, Org. Electron. 2009, 10, 1529.
[46] Z. X. Wu, L. D. Wang, G. T. Lei, Y. Qiu, J. Appl. Phys. 2005, 97,
103105.
[47] V. Percec, R. S. Clough, P. L. Rinaldi, V. E. Litman, Macromolecules
1994, 27, 1535.
[48] N. D. McClenaghan, R. Passalacqua, F. Loiseau, S. Campagna,
B. Verheyde, A. Hameurlaine, W. Dehaen, J. Am. Chem. Soc. 2003,
125, 5356.
[35] Q. L. Huang, J. F. Li, T. J. Marks, G. A. Evmenenko, P. Dutta, J. Appl.
Phys. 2007, 101, 093101.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2010, 20, 3125–3135
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3135