Crosslinkable TAPC-based hole-transport materials for solution-processed organic light-emitting diodes with reduced efficiency roll-off

Article abstract of DOI:10.1002/adfm.201201197

1-Bis[4-[N,N-di(4-tolyl)amino]phenyl]-cyclohexane (TAPC) has been widely used in xerography and organic light-emitting diodes (OLEDs), but derivatives are little known. Here, a new series of solution-processable, crosslinkable hole conductors based on TAPC with varying highest occupied molecular orbital (HOMO) energies from -5.23 eV to -5.69 eV is implemented in blue phosphorescent OLEDs. Their superior perfomance compared with the well-known N4,N4,N4′, N4′-tetraphenylbiphenyl-4,4′-diamine (TPDs) analogues regarding hole-injection and mobility, electron and exciton blocking capabilities, efficiency, and efficiency roll-off is demonstrated. Overall, the TAPC-based devices feature higher luminous and power efficiency over a broader range of brightness levels and reduced efficiency roll off. A systematic broadening of the emission zone is observed as the hole-injection barrier between the anode and the hole-transporting layer increased. The influence of varying hole-injection and electron-blocking properties is studied using different crosslinkable hole conductors of the 4,4′-(cyclohexane-1,1-diyl)bis(N,N- dip-tolylaniline) (TAPC) family in blue phosphorescent organic light-emitting diodes (OLEDs) consisting of polyvinylcarbazole (PVK)/bis[(4,6-difluorophenyl) pyridinato-N,C²](picolinato)iridium(III) (FIrpic)/1,3-bis(5-(4-tert- butylphenyl)-1,3,4-oxadiazol-2-yl)benzene (OXD-7)-based solution-processed devices. Copyright

Full text of DOI:10.1002/adfm.201201197

www.afm-journal.de
www.MaterialsViews.com
Crosslinkable TAPC-Based Hole-Transport Materials for
Solution-Processed Organic Light-Emitting Diodes with
Reduced Efficiency Roll-Off
Georgios Liaptsis and Klaus Meerholz*
-fluorene[4,7,8] derivatives as well as con-
jugated[5] or non-conjugated[9] arylamine
containing polymers in organic electronics
aiming improved hole-transport as well as
hole-injection into the emissive organic
1-Bis[4-[N,N-di(4-tolyl)amino]phenyl]-cyclohexane (TAPC) has been widely
used in xerography and organic light-emitting diodes (OLEDs), but deriva-
tives are little known. Here, a new series of solution-processable, crosslink-
able hole conductors based on TAPC with varying highest occupied molecular
material of OLEDs using low cost fabrica-
orbital (HOMO) energies from –5.23 eV to –5.69 eV is implemented in blue
phosphorescent OLEDs. Their superior perfomance compared with the well-
known N4,N4,N4′,N4′-tetraphenylbiphenyl-4,4′-diamine (TPDs) analogues
regarding hole-injection and mobility, electron and exciton blocking capabili-
ties, efficiency, and efficiency roll-off is demonstrated. Overall, the TAPC-
based devices feature higher luminous and power efficiency over a broader
range of brightness levels and reduced efficiency roll off. A systematic broad-
ening of the emission zone is observed as the hole-injection barrier between
the anode and the hole-transporting layer increased.
tion techniques.
In this study, we introduce a new series
of crosslinkable hole-conducting materials
based on 1-bis[4-[N,N-di(4-tolyl)amino]
phenyl]-cyclohexane (XTAPC), which is
well known for its high hole mobility
and high triplet energy[10,11] and has been
widely used as hole injection layer for
OLEDs[12] or xerographic applications.[13]
Due to their smaller conjugated system,
TAPCs generally feature a larger bandgap
compared to TPD-type compounds (e.g.,
the commonly used aNPD). Thus, for sim-
1. Introduction
ilar HOMO (highest occupied molecular orbital) levels, higher
LUMO (lowest unoccupied molecular orbital) energies are
expected, promising improved electron-blocking capabilities.
Further, due to the enhanced triplet energy of TAPC compared
to TPD evaporated small-molecule OLEDs based on TAPC
using FIrpic as dopand showed better device performance than
the TPD analogues.[14] We will show that the HOMO of the
TAPC based compounds can be varied by introducing electroni-
cally withdrawing or donating groups to the aromatic core. We
investigate the use of the TAPCs in blue-emitting, phosphores-
cent OLEDs (PHOLEDs). Compared with the XTPDs, XTAPCs
generally show greatly improved performance. In addition,
a systematic shift of the emission color is observed when the
injection barrier between the poly(3,4-ethylenedioxythiophene
):poly(styrenesulfonate) (PEDOT:PSS) anode and the adjacent
hole conductor varies, indicative of a change of the width of the
emission zone (EMZ) inside the device. This is confirmed by
numerical simulations. Overall, the TAPC-based devices feature
higher luminous and power efficiency over a broader range of
brightness levels.
Organic light-emitting diodes (OLEDs) are of strong interest
for display and solid-state lighting applications owing to their
low power consumption, wide viewing angle and low produc-
tion costs.[1,2] They consist of one or more organic layer(s) sand-
wiched between a transparent indium-tin-oxide (ITO) anode
and a reflective metal cathode (back contact). For improved
device characteristics, like the luminous efficiency, efficiency
roll-off and high brightness, a multi-layer design is favorable
over single-layer OLEDs.[3] In vacuum-processed devices, this is
straight forward. In order to realize a multi-layer OLED proc-
essed from solution, it is necessary to form insoluble films
before depositing the subsequent layers. In the past, we have
introduced oxetane moieties to small-molecule precursors,
allowing crosslinking of the materials through photoinitiated
cationic ring-opening polymerization (CROP) without impeding
their electrooptic properties.[4] Subsequently, additional layers
can be deposited.[4,5]
The focus of our and others[6] work has been on hole con-
ductors based on tetraphenylene-benzidine, -phenylene,
G. Liaptsis, Prof. K. Meerholz
Physikalische Chemie
2. Results and Discussion
Universität zu Köln
Luxemburger Straße 116, 50939 Köln, Germany
E-mail: klaus.meerholz@uni-koeln.de
2.1. Synthesis, Electrochemical and Photophysical Properties
Crosslinkable hole-transport materials (XHTMs) based on the
TAPC motif were synthesized by subsequent Pd(0)-catalyzed
DOI: 10.1002/adfm.201201197
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201201197
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1

www.afm-journal.de
www.MaterialsViews.com
Scheme 1. General reaction scheme for the synthesis of X-TAPCs using subsequent Buchwald-Hartwig amination reactions.
Buchwald-Hartwig cross-coupling reactions[15,16] between
4,4′-(cyclohexane-1,1-diyl)dianiline and the corresponding bro-
moarene under basic conditions (Scheme 1) in acceptable to
good overall yields (Table 1). Firstly, different bromoarenes (var-
ying in the respective substituents w through z; Table 1) were
coupled to the corresponding dianiline. Secondly, hexylether-
and oxetane-group-containing bromoarenes were coupled with
the bis-diphenylamine to the final products. The side chain
increases the solubility, and the oxetane units can be reacted
via cationic ring-opening polymerization (CROP) to crosslink
the films after deposition for subsequent multi-layer fabrica-
tion. The chemical structures of the TAPCs are shown in Figure
S1 (Supporting Information). All compounds were fully char-
expression (1), the oxidation potentials can be converted into
the HOMO energies versus vacuum level (Table 1):
Ei= (5.1 0.05)eV + Eox(0.9 0.1)eV V−1
(1)
where –5.1 eV is the HOMO energy of ferrocene, and 0.9 is a
solvent-dependent correction factor.[17]
UV/Vis absorption spectra were recorded in dichlo-
romethane to obtain the transition energy from ground state
(S0) to the first excited state (S1). A multi-Gaussian fit was used
to model the absorption spectra, and the maximum of the most
bathochromic Gaussian curve was assigned to the transition
energy S0 ⇒ S1. All new HTMs show a main absorption band
around 4 eV. The X-TAPCs 7 and 9 featuring trifluoro-methyl
substituents show a second, smaller and bathochromically
shifted absorption around 3.5 eV. The energy level of the LUMO
was estimated by adding the transition energy to the HOMO
level (Table 1). Additionally, low-temperature photolumines-
cence (PL) spectra of films of most of the novel X-TAPCs were
recorded at 80 K to obtain their triplet energies. Therefore, the
PL intensity was integrated for 5 ms 300 μs after the excitation
pulse. For all measured compounds a triplet level of 2.87 eV
was observed from the onset of photoluminescence (Figure S3,
1
acterized by H-NMR, 13C-APT NMR, high resolution ESI-MS
spectroscopy, and elementary analysis. The new compounds
X-TAPC 4 is most closely related to the parent compound
TAPC itself (tetra-alkyl substitution; see Figure S1, Supporting
Information).
The electrochemical properties of the new compounds were
investigated by cyclic voltammetry in dichloromethane with
tetrabutylammonium hexafluorophosphate as supporting elec-
trolyte and ferrocene/ferrocenium as internal standard. All
compounds exhibit two reversible oxidation steps. Using the
T a b le 1 . Synthesis, yield and electronic properties. Overall yield of the new X-TAPCs depending on the different substituents (w, x, y, z) on their aro-
matic core (see Figure 1). First and second oxidation potential measured by cyclic voltammetry calibrated vs. the ferrocene/ferrocenium redox couple,
wavelength of maximum absorption, and calculated HOMO and LUMO energies.
X-TAPC
n
w
x
y
z
Yield
[%]
66
EOx1
[V]
EOx2
[V]
EHOMO
[eV]
ELUMO
[eV]
λmax
[nm]
306
306
303
309
303
296
μSCLC
[cm2 V s−1
]
1.17 × 10−6
1.98 × 10−7
2.06 × 10−8
5.20 × 10−8
5.21 × 10−8
4.35 × 10−9
2.45 × 10−9
1
2
3
4
5
6
7
1
0
1
0
0
1
1
H
H
H
H
H
H
H
F
OMe
OMe
F
H
H
H
H
H
F
0.140
0.220
0.290
0.295
0.390
0.465
0.515
0.220
0.305
0.355
0.390
0.470
0.520
0.580
–5.23
–5.30
–5.36
–5.37
–5.45
–5.52
–5.56
–1.22
–1.31
–1.31
–1.39
–1.41
–1.47
–2.12
66
H
45
H
Me
F
44
H
53
H
F
64
CF3
H
CF3
H
32
294, 350
(sh)a)
1.53 × 10−8
1.23 × 10−9
8
9
0
0
H
F
F
F
70
62
0.590
0.650
0.665
0.745
300
–5.63
–5.69
–1.63
–2.19
CF3
H
CF3
H
296, 347
(sh)a)
XF6TPDb)
PVKb)
0.460
0.640
346
–5.51
–5.90
–5.89
–6.30
–2.17
–2.20
–3.10
–2.40
1.42 × 10−9
FIrpicb)
OXD-7b)
a) (sh) shoulder; b) HOMO and LUMO energies have been taken from ref. [14].
©
2
wileyonlinelibrary.com
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201201197

www.afm-journal.de
www.MaterialsViews.com
Energy / eV
Supporting Information), which is much higher than the triplet
energy of TPDs (2.34 eV).[14] Note that delayed fluorescence is
convoluted in the spectra of X-TAPC 4. The high triplet state
of the X- TAPC prevents energy back transfer from the triplet
level of FIrpic (2.7 eV) to the adjacent hole transport layer.
-1
CsF / Al
-2
-3
-4
LUMO OXD-7
2.2. Single-Carrier Devices
O
O
N
N
N
N
The single-carrier devices consisting of the corresponding
X-TAPC layer sandwiched between an ITO/PEDOT anode and
a non-injecting gold cathode were investigated. Assuming an
ohmic contact between the injecting electrode and the active
material and trap-free volume transport, the current-voltage
(J–V) characteristics can be described by the Mott-Gurney Law:
F
N
Ir
O
N
O
F
N
F
PEDOT:PSS
-5
n
N
F
9
8
V2
d2
-6
HOMO PVK
JSCLC
=
μhε0εr
(2)
Figure 1. Energy level diagram for compounds used in blue PHOLEDs.
HOMO and LUMO energies for Firpic are shown in black bars. PVK HOMO
energy is shown as black line and OXD-7 LUMO energy is shown as dotted
line. Inset: chemical structures of the compounds used in the EML.
Here, μh is the hole mobility, V the applied voltage, d the active
layer thickness, and ε0 and εr are the permittivity of free space
and the relative permittivity of the material (which is set to 3 in
the following calculations), respectively.
thickness of 80 nm.[17,19–20] Finally, the cathode (4 nm CsF and
100 nm Al) was deposited via thermal evaporation.[21,22]
Hole-mobilities were determined by fitting the SCLC regime
of the J–V curves using Equation 2. Formally, the assump-
tion of an ohmic contact is fulfilled for compounds 1 to 4 only
(injection barrier between PEDOT:PSS and HOMO of the cor-
responding X-TAPC < 0.3 eV), but due to the rather low mobili-
ties μSCLC featured by the other compounds (Table 1), the SCLC
conditions are maintained for all compounds despite of the
large injection barriers.[18] There seems to be a rough correla-
tion between μSCLC and the molecular polarity, resulting from
the alkoxy- (donor) and fluorine-, trifluoro-methyl (acceptor)
substitution on the TAPC aromatic core. Compared to XF6TPD,
the TPD-type hole conductor, which gave the best PHOLED
performance in previous experiments all X-TAPCs show an
improved or similar hole-mobility.
In Figure 2, the maximum luminous efficiency of the
PHOLEDs containing a single X-TAPC layer is plotted as a
function of their HOMO energy. The device efficiency increases
from 13.7 Cd A−1 for 1 (HOMO energy –5.23 eV) to >18 Cd
A−1 for 4–8 and decreases again to 15.7 Cd A−1 for 9 (–5.69 eV).
For comparison we also included data of our previous work on
X-TPDs. Obviously, compared to X-TPDs with similar HOMO
levels the device performance is significantly improved by up
to a factor of 2 for compounds 1 and 9, featuring the highest
or lowest HOMO level, respectively. We attribute this to the
enhanced electron/exciton blocking ability of the TAPCs at the
HTL/EML interface.
For both device series, X-TPDs and X-TAPCs, the maximum
luminous efficiency is obtained when the hole-injection bar-
rier between PEDOT:PSS (–5.1 eV) and PVK (–5.9 eV) is
divided into two almost identical steps (i. e. when HOMO is
at ca. –5.5 eV), reaching slightly higher efficiencies (up to
18.5 Cd A−1 ) compared with the reference device without HTL
(REF: 18.1 Cd A−1 ). However, while the maximum luminous
efficiency clearly peaks for the TPD-series, it is less distinct
for the X-TAPCs. Please note, that devices based on 3, 6 and 7
featuring a donor and acceptor substitution pattern on the aro-
matic units do not follow the trend of the other compounds.
This might be explained by an increased permanent dipole
moment of these molecules, resulting in a decreased hole-
mobility (compare Section 2.2).[23,24] Such compounds were not
investigated in our previous study on TPDs.
Table 2 summarizes the performance of the full device
series and compares the results with those obtained for devices
without a HTL (REF) or with the best TPD derivative (XF6TPD).
Compared to the REF device, all other devices feature reduced
onset voltages despite the fact that the total layer thickness is
increased by 10 nm (one HTL) or even 20 nm (double HTL).
The lowest onset voltage is achieved using X-TAPC 1. The
2.3. Single-HTL OLEDs
In our previous work, we introduced different hole-trans-
port layers based on crosslinkable TPDs (X-TPDs) between
PEDOT:PSS and the emitter layer in blue-phosphorescent
OLEDs (PHOLEDs), in which the emitting layer (EML) con-
sisted of PVK/OXD–7/FIrpic (68/27/5 wt%; see Figure 1 for
chemical structures).[17] We demonstrated that the maximum
luminous efficiency correlated with the HOMO energy of
the HTL, the best performance was achieved, when the hole-
injection barrier between PEDOT:PSS and the PVK matrix was
divided into two equal steps. However, the electron blocking
properties of the TPDs, which also varied within this series,
were not considered.
To examine the new crosslinkable, large-bandgap X-TAPCs
in terms of their device performance we fabricated similar blue
PHOLEDs. The device configuration was as follows: a 35 nm
thin layer PEDOT:PSS was spincast on ITO-coated glass fol-
lowed by deposition of a 10 nm thin layer of a X-TAPC. After
crosslinking of the HTL, the EML was spin-cast on top in a
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201201197
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3

www.afm-journal.de
www.MaterialsViews.com
device starts to emit (EL > 0.1 Cd/A) at an applied voltage of
2.85 V. Within the X-TAPC series the onset voltage increases
up to 4.50 V using X-TAPC 9, which is still superior to the REF
device (4.70 V). The maximum performance of devices con-
taining XTAPC 4 and even more so the double-HTL device 2/5
exceed the performance of REF, demonstrating a superior per-
formance within optimized HTL due to more facilitated hole-
injection and increased electron blocking ability.
2.4. Color Shift
Within the entire device series, the CIE color coordinates of the
emission shift from greenish blue for device using X-TAPC 1
as HTL (x = 0.194, y = 0.392)to a deeper blue for X-TAPC 9 (x =
0.172, y = 0.364; Table 2).
This might be attributed to a spatial variation of the emission
zone (EMZ) within the EML, which is largely affected by the
charge-carrier balance inside the EML. This balance is influ-
enced by both, charge-injection and charge-transport properties
of the layers.[25,26] By using the software Etfos 1.5 (Fluxim AG)
the EMZ can be qualitatively derived from the electrolumines-
cence (EL) spectra.[27,28] As input for the internal emission spec-
trum of the EML the normalized photoluminescence spectrum
of an 80 nm film of the PVK/OXD-7/FIrpic blend was used.
From the best fit of the simulation to the experimental EL
spectra (Figure 3a) at constant current densities of 1 mA cm−2 ,
the EMZs of the devices were obtained. The results for the device
containing X-TAPCs 1, 4, 9 and the REF device are shown in
Figure 3b. While the EMZ is narrow for compound 1, its width
increases approximately linearly (Figure 3c) when the injection
barrier between PEODT and X-TAPC increases. Simultaneously,
the height of the EMZ is reduced owing to a more extended
hole distribution within the EML at comparable brightness’s.
This affects the outcoupling of discrete wavelengths resulting
Figure 2. Maximum luminous efficiency of blue PHOLEDs with single-
layer XTAPCs 1–9 (filled and open squares). The maximum luminous
efficiency of the reference device (REF) without HTL is shown as dashed
line. Data obtained using X-TPDs are included for comparison (open up
triangles).[11] The bottom axis shows the HOMO energy of the respective
compound (Table 1); the energy levels of the injection layer PEDOT:PSS
and of PVK, the matrix material used in the EML are indicated by arrows.
The top axis shows the injection barrier between PEDOT:PSS and the
XHTL assuming energy level alignment. The solid lines are guide to the
eye. The vertical dashed line indicates the case, when the injection barrier
is subdivided into two identical steps.
T a b le 2 . Device data for single-, double-, and no-HTL devices Maximum luminous efficiency LEmax and the corresponding voltage U, luminous effi-
ciency at a brightness of 1000 Cd m−2 LE1000 and the corresponding voltage U, maximum power efficiency PEmax and the corresponding voltage U,
onset voltage Uon, and CIE-values (x,y) measured at a current density of 1 mA cm−2
.
LEmax [Cd A−1
]
LE1000 [Cd A−1
]
PE max [lm W−1
]
Uon [V] at 0.1 [Cd m−2
]
X-TAPC
U [V]
U [V]
U [V]
CIE
x
y
1
13.7
15.6
15.8
18.2
18.4
17.2
16.4
18.5
15.7
18.4
17.6
19.1
18.1
6.50
7.25
7.75
6.75
7.25
7.25
7.50
7.75
7.75
7.00
8.00
7.25
7.25
11.3
14.0
15.3
16.8
18.1
15.7
14.3
16.7
13.6
17.3
14.0
16.3
18.0
9.50
9.00
8.75
9.00
8.75
9.00
9.75
9.50
10.00
9.50
12.00
9.50
7.50
8.1
8.1
7.8
10.4
9.8
8.0
8.0
8.3
6.8
11.3
9.8
9.1
8.8
4.50
5.50
5.00
4.50
4.50
6.00
5.75
6.25
6.75
4.00
4.50
6.00
6.50
2.85
2.95
3.10
3.10
3.30
3.70
3.95
4.10
4.50
3.25
3.75
4.10
4.70
0.1939
0.1811
0.1712
0.1761
0.1702
0.1676
0.1712
0.1711
0.1731
0.1729
0.1723
0.1697
0.1652
0.3918
0.3769
0.3660
0.3681
0.3689
0.3590
0.3627
0.3610
0.3643
0.3612
0.3599
0.3578
0.3606
2
3
4
5
6
7
8
9
2/ 5
5/ 8
XF6TPD
REF
©
4
wileyonlinelibrary.com
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201201197

www.afm-journal.de
www.MaterialsViews.com
(c)
(a)
(b)
(d)
Figure 3. a) Normalized experimental EL spectra of the devices based on 1 (filled squares), 4 (open circles), 9 (filled up triangles), and REF device (open
down triangles) measured at 1 mA cm−2 . The inset diagram shows the corresponding CIE x and y values (Table 2). b) Relative distribution of the EMZ in
the OLEDs simulated from their corresponding EL-spectra (see subfigure a). c) Width (FWHM) of the emission zone (EMZ) for the above devices as a
function of the injection barrier between PEDOT:PSS and the XHTL. The solid black line is a guide to the eye. d) Operating voltage (left axis and closed
symbols) and brightness (right axis, open symbols) of devices operated at a current density of 1 mA/cm2 as a function of the injection barrier. The solid
lines are guide to the eye. The vertical dashed lines in (c,d) indicate in the case, when the injection barrier is divided into two identical steps.
in a color shift of the EL-spectra from greenish blue to deeper
blue due to the variation of the X-TAPCs. In comparison, the
REF device exhibits the broadest and least intense EMZ pro-
file. Note, that the emission intensity is roughly constant for all
devices when operated at 1 mA cm−2 (Figure 3d, open squares),
indicating constant efficiency. The drive voltage increases
slightly (by ca. 1 V) due to the variation of the hole-injection
barrier (Figure 3d, closed squares).
In addition, we studied the changes of the emission for dif-
ferent drive voltages. While there is a significant broadening
of the EMZ for devices based on 1 and 4, this effect is almost
vanished in 9 (Figure S2 (Supporting Information) and bars
in Figure 3a). This is explained due to an increasing displace-
ment of the holes towards the cathode when the electric field is
increased. By contrast, the effect is reversed in the REF device,
probably due to enhanced electron injection.
next to the EML for 2/5 interlayers, or next to PEDOT:PSS for
5/8 interlayers. These double-HTL devices show also high effi-
ciency (17.6 Cd A−1 at 8.00 V and 18.4 Cd A−1 at only 7.00 V
despite its increased device thickness of 20 nm) and the highest
power efficacy of 11.3 lm W−1 at 4.00 V for the 2/5 interlayers
(REF: 8.8lm W−1 at 6.50 V, best single-HTL: 10.4 lm W−1 at
4.50 V) (Figure 4; Table 2) due to facilitated hole injection.
Until now, we have only compared the optimum performance
(maximum values). Now, we will consider the full voltage range:
i) As already pointed out above, the maximum luminous effi-
ciency [cd/A] is very similar in all devices and is observed at
operating voltages of ca. 7–8 V. This indicates that the bal-
ance between holes and electrons is ideal under these condi-
tions). By contrast, distinct differences are observed for lower
and/or higher voltages, hinting that the hole/electron ratio
becomes non-ideal. Fortunately, the introduction of the HTL
significantly broadens the J–V curves compared to the REF
device, which features a rather narrow maximum.
2.5. Double-HTL OLEDs and Comparison with TPDs
ii) The power efficacy [lm/W] is similar for U > 8 V, whereas
larger differences are observed for U < 7 V. Again, introduc-
tion of the HTL improves the performance at all voltages
compared to the REF device.
For further optimization, devices with a double-HTL structure
were fabricated. In these cases, the maximum injection barrier
for holes was identical (0.35 eV), while its position varies: it is
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201201197
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
5

www.afm-journal.de
www.MaterialsViews.com
(a)
(c)
(b)
(d)
Figure 4. Luminous efficiency (a,c) and power efficacy (b,d) as a function of the operating voltage. a,b) Single-HTL devices based on 4 (EHOMO
=
–5.37 eV; filled squares), 5 (EHOMO = –5.45 eV; open circles), and XF6TPD (EHOMO = –5.50 eV; open up triangles). c,d) Double-HTL devices based on
2/5 (EHOMO = –5.30/EHOMO = –5.45 eV; filled squares) and 5/8 (EHOMO = –5.45 eV/EHOMO = –5.63 eV; open circles). In both cases, the reference device
REF without HTL is shown in open down triangles.
iii) Directcomparisonoftheperformanceofdevices5andXF6TPD,
which have similar HOMO energy (–5.51 eV and –5.45 eV, re-
spectively; see Table 1) shows strongly improved performance
for XTAPC 5 at low voltages (light levels below 100 Cd m−2 ) as
well as improved efficiency roll-off at high voltages (Figure 4).
This finding is clear evidence for the better electron-blocking
capabilities of XTAPC-5 compared with XF6TPD, the barrier
for electrons increases by ca. 0.7 eV (see Table 1).
We demonstrated the beneficial impact onto the performance
of prototypical blue PHOLEDs. While the maximum luminous
efficiency was essentially unchanged, the power efficiency and
efficiency roll-off were improved due to the enhanced electron-
blocking capabilities of the TAPCs as compared to TPDs. Using
optical simulations, it was demonstrated for the first time that
the EMZ broadens as the hole-injection barrier increases.
iv) The best overall performance regarding both, power efficacy
(>11 lm/W) and luminous efficiency (>18 Cd/A), as well as
for the efficiency roll-off is obtained with the double-HTL de-
vice 2/5, while the REF device is worst (Figure 4c,d; Table 2).
4. Experimental Section
Synthesis: All chemicals were used as received from commercial
suppliers. Buchwald-Hartwig amination reactions were carried out
under argon atmosphere in degassed and dry toluene. A mixture of
corresponding dianiline and appropriate equivalents of bromoarene
using 1 mole percent of Pd2(dba)3, 1.6 mole percent of P(tBu)3 and
1.5 equivalents of NaOtBu per reacting NH were heated for 2 to
24 h at temperatures between 40 and 80 °C. Completion of the reactions
was monitored by TLC and ESI-MS, if so, the reaction was diluted
with EtOAc, washed with water, re-extracted with EtOAc and dried
over MgSO4. After evaporation of the solvent, the desired product was
purified by mid-pressure liquid chromatography (MPLC) eluting with
toluene/ethylacetate mixtures at a constant flow of 20 mL/min.
Nuclear Magnetic Resonance Spectroscopy: The 1H-NMR or 13C-
NMR spectra were recorded on a Bruker DPX 300 spectrometer using
chloroform-d1 as the solvent.
3. Conclusion
In conclusion, we have introduced a novel class of crosslinkable
hole conductors based on the TAPC motif for multilayer OLED
fabrication from solution. These compounds generally feature a
larger bandgap and a higher triplet energy level than the com-
monly used TPD-type hole conductors. As a result, for com-
pounds with similar HOMO level the electron and/or excition
blocking capabilities of the TAPC derivatives are more favorable.
©
6
wileyonlinelibrary.com
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201201197

www.afm-journal.de
www.MaterialsViews.com
High Resolution ESI-MS: The high resolution ESI-MS spectra were
recorded on a Thermo Scientific LTQ Orbitrap XL.
Elementary Analysis: The elementary analysis was performed on a
Euro EA 3000 (CHNS), HEKAtech.
Dr. Dirk Hertel and Dr. Ronald Alle for fruitful discussions. Aditionally,
the authors thank Radouane Nagim for the HR-ESI-MS measurements
and Silke Kremer for the EA measurement. All persons mentioned above
are with the University of Cologne.
Electrochemical Measurements: Cyclic voltammetry was performed
with an EG&G Instruments Potentiostat/Galvanostat Model 283 in 10−3
molar solutions of appropriate HTM in dry dichlormethane with 0.1 M
tetrabutylammonium hexafluorophosphate as supporting electrolyte.
Platinum wires were used for working and counter electrodes whereas
a silver wire was used for reference electrode. All measurements were
carried out at room temperature without inert gas atmosphere and were
correlated to the ferrocenium/ferrocene redox couple as internal standard.
UV/VIS absorption spectroscopy: UV/VIS absorption spectra were
recorded with a CARY 50, VARIAN spectrometer from 10−5 molar
dichlormethane solution in a 10 mm diameter quartz cuvette.
Fluorescence Spectroscopy: Fluorescence spectra from a 80 nm thin
spincast film was recorded with a CARY ECLIPSE VARIAN spectrometer.
Low Temperature Photoluminescence Spectroscopy: The compounds
were excited with the third harmonic (355nm) of a Nd:YAG (Spitlight,
Innolas). The laser operated at a repetition rate of 10 Hz. The PL was
collected and focused onto the entrance slit of the monochromator and
detected by an intensified gateable ICCD camera (Roper Scientific).
Single-Carrier Devices: Single carrier devices were fabricated on ITO-
coated glass substrates which were thoroughly cleaned and treated
with ozone plasma. A layer of PEDOT:PSS (Baytron P, AI4083, Heraeus
Clevios) was spincast onto the substrates and backed at 120°C for 120 s.
The substrates were transferred to a nitrogen filled glove-box where the
crosslinkable HTMs were spincast from 20.0 mg mL−1 toluene solutions
containing 1.51 wt% 4-octyloxydiphenyliodonium hexafluoroantimonate
(OPPI) as photoinitiator. The films were irradiated with UV light
(310 nm) for 10 s and backed at 110 °C to promote crosslinking yielding
a 100 nm thick film. The spin coating and crosslinking procedure was
done twice to yield a layer thickness of 200 nm.
OLED Fabrication and Measurement: The OLEDs were fabricated on
ITO-coated glass substrates which were thoroughly cleaned and treated
with ozone plasma. A layer of PEDOT:PSS (Baytron P, AI4083, Heraeus
Clevios) was spincast onto the substrates and backed at 120 °C for 120 s.
The substrates were transferred to a nitrogen filled glove-box where the
crosslinkable HTMs were spincast from 3.0 mg mL−1 toluene solutions
containing 4 mol% 4-octyloxydiphenyliodonium hexafluoroantimonate
(OPPI) as photoinitiator. The films were irradiated with UV light (310 nm)
for 10 s and backed at 110 °C to promote crosslinking. Afterwards the
emitting blend consisting of PVK, OXD-7 (28 wt%) and FIrpic (5 wt%)
in chlorobenzene (15 mg mL−1 ) was spincast on top. Subsequently, the
cathode was deposited by thermal evaporation at 10−9 bar. The current-
voltage-luminescence characteristics were measured with an Keithley
2400 amperometer and a calibrated photodiode.
Simulation: For optical simulation the software Etfos 1.5 (Fluxim
AG) which uses the transfer-matrix formalism was used. As internal
spectra for simulations a normalized photoluminescence spectrum
of a 80 nm film on glass of the emitting blend consisting of PVK,
OXD-7 (28 wt%) and FIrpic (5 wt%) spincast from chlorobenzene
(15 mg mL−1 ) was used. The thickness of the different layers were set
as follows: 100 nm Alumina-cathode, 80 nm EML, 10 nm HTL, 35 nm
PEDOT:PSS, 130 nm ITO.
Received: May 1, 2012
Revised: July 10, 2012
Published online:
[1] 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. Bredas, M. Logdlund, W. R. Salaneck, Nature 1999, 397, 121.
[2] C. D. Muller, A. Falcou, N. Reckefuss, M. Rojahn, V. Wiederhirn,
P. Rudati, H. Frohne, O. Nuyken, H. Becker, K. Meerholz, Nature
2003, 421, 829.
[3] S. Reineke, F. Lindner, G. Schwartz, N. Seidler, K. Walzer, B. Lussem,
K. Leo, Nature 2009, 459, 234.
[4] M. S. Bayerl, T. Braig, O. Nuyken, D. C. Muller, M. Gross,
K. Meerholz, Macromol. Rapid Commun. 1999, 20, 224.
[5] S. Jungermann, N. Riegel, D. Muller, K. Meerholz, O. Nuyken,
Macromolecules 2006, 39, 8911.
[6] F. Huang, Y. J. Cheng, Y. Zhang, M. S. Liu, A. K. Y. Jen, J. Mater.
Chem. 2008, 18, 4495.
[7] X. H. Yang, D. C. Muller, D. Neher, K. Meerholz, Adv. Mater. 2006,
18, 948.
[8] D. C. Muller, T. Braig, H. G. Nothofer, M. Arnoldi, M. Gross,
U. Scherf, O. Nuyken, K. Meerholz, ChemPhysChem 2000, 1, 207.
[9] J. Schelter, G. F. Mielke, A. Kohnen, J. Wies, S. Kober, O. Nuyken,
K. Meerholz, Macromol. Rapid Commun. 2010, 31, 1560.
[10] P. M. Borsenberger, L. Pautmeier, R. Richert, H. Bassler, J. Chem.
Phys. 1991, 94, 8276.
[11] K. Goushi, R. Kwong, J. J. Brown, H. Sasabe, C. Adachi, J. Appl. Phys.
2004, 95, 7798.
[12] C. W. Tang, S. A. Vanslyke, Appl. Phys. Lett. 1987, 51, 913.
[13] P. M. Borsenberger, D. S. Weiss, Organic Photoreceptors for Xerog-
raphy, 59 Optical Engineering Series, Marcel Dekker Inc, 1998.
[14] J. Lee, N. Chopra, S. H. Eom, Y. Zheng, J. G. Xue, F. So, J. M. Shi,
Appl Phys Lett 2008, 93, 123306.
[15] J. F. Hartwig, Synlett 1997, 329.
[16] S. L. Buchwald, Top. Curr. Chem. 2001, 219, 131.
[17] P. Zacharias, M. C. Gather, M. Rojahn, O. Nuyken, K. Meerholz,
Angew. Chem. Int. Ed. 2007, 46, 4388.
[18] U. Wolf, S. Barth, H. Bassler, Appl. Phys. Lett. 1999, 75, 2035.
[19] M. K. Mathai, V. E. Choong, S. A. Choulis, B. Krummacher, F. So,
Appl. Phys. Lett. 2006, 88, 243512.
[20] X. H. Yang, F. Jaiser, S. Klinger, D. Neher, Appl. Phys. Lett. 2006, 88,
3.
[21] P. Piromreun, H. Oh, Y. L. Shen, G. G. Malliaras, J. C. Scott,
P. J. Brock, Appl. Phys. Lett. 2000, 77, 2403.
[22] M. Y. Chan, S. L. Lai, M. K. Fung, C. S. Lee, S. T. Lee, Chem. Phys.
Lett. 2003, 374, 215.
[23] M. Vanderauweraer, F. C. Deschryver, P. M. Borsenberger, H. Bassler,
Adv. Mater. 1994, 6, 199.
[24] K. L. Tong, S. K. So, H. F. Ng, L. M. Leung, M. Y. Yeung, C. F. Lo,
Synth. Met. 2004, 147, 199.
[25] G. G. Malliaras, J. C. Scott, J. Appl. Phys. 1998, 83, 5399.
[26] Z. X. Wu, L. D. Wang, G. T. Lei, Y. Qiu, J. Appl. Phys. 2005, 97, 103105.
[27] M. Flammich, M. C. Gather, N. Danz, D. Michaelis, K. Meerholz,
Appl. Phys. Lett. 2009, 95, 263306.
[28] S. L. M. van Mensfoort, M. Carvelli, M. Megens, D. Wehenkel,
M. Bartyzel, H. Greiner, R. A. J. Janssen, R. Coehoorn, Nat. Phot-
onics 2010, 4, 329.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
The authors thank Dr. Dirk Hertel for experimental help within low
temperature photoluminescence spectroscopy. Further, they thank
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201201197
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
7