University of Southern California Dissertations and Theses

New materials and device structure for organic light-emitting diodes

(USC Thesis Other)

New materials and device structure for organic light-emitting diodes

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

NEW MATERIALS AND DEVICE STRUCUTURE FOR
ORGANIC LIGHT-EMITTING DIODES

by

Chao Wu

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(MATERIALS SCIENCE)

December 2010

Copyright 2010 Chao Wu

ii
Acknowledgements

From the bottom of my heart, I feel full of gratitude for my advisor Professor
Mark Thompson, who not only led me into this exciting research field, also helped me
navigate through my entire research work with his knowledge and wisdom. My
appreciation to your enormous patience, care and understanding is beyond any words.
I would also like to thank my parents and Rui. Your ubiquitous support and love
in every aspect of my life is heartfelt.
In addition, I want to say “thank you” to Professor Chongwu Zhou and Professor
Edward Goo, who are on my committee. Your precious time and support along the way
is sincerely appreciated.
In the end, I would like to extend my warmest thanks to all the colleagues and
friends I met and shared time with at USC.
I always believe that actions speak louder than words. What you mean to me will
easily be seen for the years to come.

iii
Table of Contents
Acknowledgements ............................................................................................................. ii
List of Tables ...................................................................................................................... v
List of Figures………………………………………………………………………….....vi
Abstract……..………………………………………………………………………........xii

Chapter 1 Introduction
1.1 OLED .........................................................................................................................1
1.2 Structure and operation mechanism ...........................................................................2
1.2.1 Device structure ...................................................................................................2
1.2.2 Charge injection ..................................................................................................3
1.2.3 Charge conduction ...............................................................................................5
1.2.4 Exciton formation ................................................................................................8
1.2.5 Recombination and energy transfer ...................................................................10
1.2.6 Efficiency ..........................................................................................................11
1.3 Evolvement of OLED technology ...........................................................................12
1.3.1 Electroluminescence ..........................................................................................12
1.3.2 First heterojuction OLED ..................................................................................13
1.3.3 Doping ...............................................................................................................13
1.3.4 Application of phosphorescent materials in OLED ..........................................15
1.3.5 Polymers in OLED ............................................................................................17
1.4 White OLED (WOLED) ..........................................................................................18
1.5 Chapter 1 References ...............................................................................................21
Chapter 2 Study of energy transfer and triplet exciton diffusion in hole
transporting host materials
2.1 Abstract ....................................................................................................................25
2.2 Introduction ..............................................................................................................25
2.3 Results and discussion .............................................................................................28
2.3.1 Electrochemistry .............................................................................................. 29
2.3.2 Photophysical properties .................................................................................. 32
2.3.3 Undoped devices .............................................................................................. 35
2.3.4 Doped devices .................................................................................................. 36
2.3.5 Theoretical analysis ......................................................................................... 45
2.4 Chapter Conclusions ................................................................................................49
2.5 Experimental section ................................................................................................50
2.6 Supporting information ............................................................................................54
2.7 Chapter 2 References ...............................................................................................61

iv
Chapter 3 Simplified structure for two-component and three-component
WOLED
3.1 Abstract ....................................................................................................................64
3.2 Introduction ..............................................................................................................65
3.3 Results and discussion .............................................................................................68
3.3.1 Photophysics of TPP and IrDBQ isomers ........................................................ 69
3.3.2 Quantum yield and lifetime ............................................................................. 73
3.3.3 WOLED devices .............................................................................................. 74
3.3.3.1 Triplet exciton diffusion study on TPP ..................................................... 74
3.3.3.2 Two-component and three-component WOLED ...................................... 77
3.4 Chapter Conclusions ................................................................................................81
3.5 Experimental section ................................................................................................82
3.6 Chapter 3 References ...............................................................................................85
Chapter 4 Study of ion-paired Iridium complexes (soft salts) and their application
in OLED
4.1 Abstract ....................................................................................................................87
4.2 Introduction ..............................................................................................................88
4.3 Results and discussion .............................................................................................89
4.3.1 Photophysics and quenching study .................................................................. 91
4.3.2 Quantum yield and lifetime ............................................................................. 95
4.3.3 HOMO and LUMO energy .............................................................................. 96
4.3.4 OLED Studies .................................................................................................. 98
4.4 Chapter Conclusions ..............................................................................................103
4.5 Experimental section ..............................................................................................104
4.6 Supporting informaiton ..........................................................................................111
4.7 Chapter 4 References .............................................................................................116
Chapter 5 Thermally cross-linkable electron transporting material for bottom-
emitting inverted OLEDs
5.1 Abstract ..................................................................................................................119
5.2 Introduction ............................................................................................................120
5.3 Results and discussion ...........................................................................................124
5.3.1 Polymerization ............................................................................................... 125
5.3.2 Bottom-emitting IOLED ................................................................................ 130
5.4 Chapter Conclusions ..............................................................................................137
5.5 Experimental section ..............................................................................................138
5.6 Chapter 5 References .............................................................................................141
Comprehensive Bibliography.......................................................................................145

v
List of Tables
Table 2.1 Redox, photoluminescence, energy levels and quantum yield
data for each material used in devices

31

Table 2.2 External quantum efficiency (EQE) and voltage data of TDDP,
TPD, NDDP and NNP devices at 10 mA/cm
2

37

Table 3.1 QY and lifetime data of TPP samples 74

Table 4.1 Redox potentials, energy levels, lifetimes and quantum yield
data of the soft salts and their component ions under nitrogen

91

Table 4.2 Performance of the OLEDs made using different soft salts 102

Table 4.S1 Life times (s) of the neat films which were prepared according to
the same procedure described in the paper.

115

Table 5.1 The thickness of the film before and after the treatments 128

vi
List of Figures
Figure 1.1 Examples of products using OLED technology 2

Figure 1.2 Classic heterjunction OLED structure 3

Figure 1.3 Hole and electron injection from the electrodes 4

Figure 1.4 Examples of common hole-transporting (top) and electron-
transporting (bottom) materials

7

Figure 1.5 Diagram of possible relaxation pathways for the excitons
formed

9

Figure 1.6 Förster and Dexter energy transfer mechanism 10

Figure 1.7 Charge movement and recombination mechanism in
host/dopant system

14
Figure 1.8 State-mixing in complexes containing heavy metals
(exemplified with Ir)

16
Figure 2.1 Device structure and chemical structures of the moluecules used
in this study. Horizontal lines and numbers on the lines in
device structure represent HOMO and LUMO energy levels
(eV).

30

Figure 2.2 PL spectra of TDDP and NNP in 2-MeTHF solution at 77K 33

Figure 2.3 PL spectra in 2-MeTHF solution and neat film for TDDP and
NNP

34

vii
Figure 2.4 PL spectra of TDDP and TDDP:BCP (1:1) film compared with
the EL spectra from undoped TDDP devices with and without
an mCP interlayer

35

Figure 2.5 EL spectra from F-type TDDP devices at J=10mA/cm
2
38

Figure 2.6 EL spectra from P-type TDDP and NNP devices at
J=10mA/cm
2

40
Figure 2.7 EL spectra from double-doped and control devices when
J=10mA/cm
2
. The structure of the undoped and double-doped
devices is given in the text. The ADN-doped and PQIr-doped
device has the structure of NNP (300 Ǻ)/5%ADN:NNP (100
Ǻ)/mCP (100 Ǻ)/BCP (400 Ǻ)/LiF (10 Ǻ)/Al (1200 Ǻ) and
NNP (200 Ǻ)/8%PQIr:NNP (100 Ǻ)/NNP (200 Ǻ)/mCP (100
Ǻ)/BCP (400 Ǻ)/LiF (10 Ǻ)/Al (1200 Ǻ), respectively.

42
Figure 2.8 EL spectra from P-III devices using different hosts at
J=10mA/cm
2
. The spectra are normalized to the blue
fluorescence peak maxima. Inset: Comparison of EQE among
P-III devices using different hosts.

43
Figure 2.9 EQE of P-III devices with host materials that do not contain
naphthyl groups

45
Figure 2.10 Spin density distribution of triplet states from DFT calculations.
Top: DDP (left) and TPD (right); bottom: NDDP (left) and
NNP (right)

47
Figure 2.S1 Solution and film emission spectra for TPD

54
Figure 2.S2 Solution and film emission spectra for TTP 55

viii
Figure 2.S3 Solution and film emission spectra for TTTP 55

Figure 2.S4 Solution and film emission spectra for NDDP 56

Figure 2.S5 EL spectra of P type TPD devices 56

Figure 2.S6 EL spectra of P type NDDP devices 57

Figure 2.S7 J-V characteristics of P type and undoped TDDP devices 57

Figure 2.S8 J-V characteristics of P type and undoped TPD devices 58

Figure 2.S9 J-V characteristics of P type and undoped NDDP devices 58

Figure 2.S10 J-V characteristics of P type and undoped NNP devices 59

Figure 2.S11 PL of 2% C6-doped TDDP film 59

Figure 2.S12 EQEs of F type TDDP devices

60
Figure 2.S13 EQEs of P type TDDP devices 60

Figure 3.1 Two-component and three-component WOLED using
simplified structure

67

Figure 3.2 Molecular structure of TPP and IrDBQ isomers 68

Figure 3.3 Photophysics of TPP 69

ix
Figure 3.4 Emission and absorption spectrum of fac- (top) and mer-
(bottom) IrDBQ

71

Figure 3.5 The emission spectra of pure fac-, mer- IrDBQ and three
mixtures

72

Figure 3.6 HPLC reports for the thermally-deposited films of the mixtures,
95:5 (left), 65:35 (middle) and 25:75 (right)

72

Figure 3.7 Cyclic voltammetry and differential pulse voltammetry (inset)
measurement of TPP (peak at 0.3V is the internal reference,
decamethylferrocene)

75

Figure 3.8 EL of the fluorescence-based OLED using TPP as the host 75

Figure 3.9 The comparison of EL and EQE between TPP and NNP device 76

Figure 3.10 The EL of two-component (left) and three-component (right)
WOLED

78

Figure 3.11 EQE and J-V curves of the tow-component and three-
component WOLED

79

Figure 4.1 Structure of the soft salts studied 90

Figure 4.2 Absorption of C2Cl and photoluminescence spectrum of C2Cl
and A3Na

92

Figure 4.3 Photoluminescence of C2A3 at different concentrations (all in
degassed acetonitrile, excitation wavelength = 350nm)

93

x
Figure 4.4 Stern-Volmer plot of the quenching study between C2 and A3
and the numerical fitting of K
q

94

Figure 4.5 Energy levels of the materials used in the OLEDs

96
Figure 4.6 Comparison of cyclic voltammetry between C2A3 and its
component ions

97

Figure 4.7 EL of the devices made of three soft salts 100

Figure 4.8 J-V characteristics (top) and EQE (bottom) of the devices using
different soft salts

101

Figure 4.S1 Cyclic voltammetry of A1 111

Figure 4.S2 Cyclic voltammetry of A2 111

Figure 4.S3 Cyclic voltammetry of C1A1 112

Figure 4.S4 Cyclic voltammetry of C1A2 112

Figure 4.S5 Photoluminescence of C1 113

Figure 4.S6 Photoluminescence of A1 113

Figure 4.S7 Photoluminescence of A3 114

Figure 4.S8 Photoluminescence of C1A1 at 10
-6
M 114

xi
Figure 4.S9 Photoluminescence of C1A3 at 10
-6
M 115

Figure 5.1 Structure of TPBI 124

Figure 5.2 DSC measurements starting with monomer 126

Figure 5.3 FT-IR spectra of the film before and after the thermal treatment 127

Figure 5.4 Absorption Spectra of the film before and after the treatment 129

Figure 5.5 Effect of different ITO cleaning methods on the J-V
characteristics and EQE

131

Figure 5.6 Comparison of cross-linked polymer with TPBI and Alq
3
as the
ETL

133

Figure 5.7 The EL spectrum of three phosphorescent IOLED 134

Figure 5.8 J-V characteristics and EQE of three phosphorescent IOLEDs
with different ETL

136

xii
Abstract

Organic light-emitting diode (OLED) refers to any light-emitting diode (LED)
using exclusively molecular and polymeric materials. Since the first demonstration in
late 80s, OLED has been the topic of intense interest to both scientific and industrial
community. With the progresses made on improving performance, the most advanced
OLEDs have demonstrated near unity quantum efficiencies and over tens of thousands of
hour-long lifetime. They have been developed to the point that that they are
commercially available in small, hand-held, full color displays and showed great promise
in lighting applications as well. All the aforementioned achievements were driven by a
series of new materials, device structures and fabrication techniques, which will continue
to be the force moving OLED technology towards wider success. This dissertation
follows the mainstream of the OLED research, trying to incorporate materials with
attractive properties into both conventional and newly-designed device structures, in the
mean time demonstrating the advantages in simplicity and performance.
Chapter 1 briefly introduces the basics of OLED and the breakthroughs in the
history of OLED research, including the first heterojuction device, doping concept and
the application of phosphorescence materials.
Chapter 2 focuses on understanding how the molecular structure affects a
property of profound significance to OLED, triplet exciton diffusion length. The study
uses a simple and effective device structure to probe the energy transfer and triplet

xiii
exciton diffusion in this paper. The deliberate mismatch of the HOMO between hosts
and dopants and the good hole mobility of host materials are key to the success. We are
able to correlate the device performance to the triplet exciton diffusion length and explain
the huge discrepancy, based on the data obtained from experiments and theoretical
calculations.
The versatility of the device structure discussed in Chapter 2 is proven in Chapter
3, with both two-component and three-component White OLEDs (WOLED) being
fabricated using this design. Thanks to the new broad-band yellow emitter, two-
component WOLED demonstrated excellent quantum efficiency and superior
electroluminescence stability. The study shows the promise of this simple structure in
yielding comparable performance to its complicated conventional counterpart.
Chapter 4 unveils a new type of materials called “soft slats”. In this dissertation,
“soft salt” are composed of an organometallic cation and an organometallic anion, both
Ir-based. The attractive features include solution-processing ready, amipolar and flexible
in property tuning. Energy transfer between the ions in solution is observed, and found to
take place at diffusion controlled rates. OLEDs prepared using a simple structure yielded
external quantum efficiencies close to theoretical maximum, if the front-orbital energies
between two ions are aligned properly.
In the end, a thermally cross-linkable electron-transporting polymer is studied
with the emphasis on the application in bottom-emitting inverted OLED in Chapter 5.
The polymer layer can be easily prepared by baking the spin-casted monomer film under

xiv
mild conditions. And it is shown that without the aid of interfacial electron injection
layer, the polymer lowers the electron injection barrier, resulting in the encouraging
device performance.

1
Chapter 1 Introduction

1.1 OLED
Organic electroluminescence (EL) is the emission of light generated electrically
from organic materials, which was first studied in the 1960s.
1
In 1987, combining
suitable materials and structure with modern thin film deposition techniques, a team at
Kodak introduced first two-layer heterojunction organic light-emitting diode (OLED).
2

Three years later, the Friend group at the Cambridge reported first conducting polymer-
based LED (PLED).
3
Since then, there has been growing interests and research activities
in this exciting field. Enormous progress has been made in the improvement of color,
efficiency and device lifetime. The increasing interest is overwhelmingly fueled by the
hope of applying OLED technology in flat panel displays and for lighting purposes. Till
today, various products have been commercialized or demonstrated, exemplified with
Google Nexus One cell phone and Philips ceiling OLED lamps in Figure 1.1.
The future holds tremendous opportunity for the low cost and high performance
offered by OLEDs. With the advantages of low power consumption and high brightness,
full color displays that use OELDs is challenging and may eventually replace traditional
technologies, such as liquid-crystal displays (LCDs) and cathode ray tube (CRT), in
applications from mobile phones to computer monitors. In addition, OLEDs can be
deposited on flexible plastic or metal foils, eliminating the fragile and heavy glass
substrates. Moreover, OLED displays emit light without the pronounced directionality

2
inherent in LCD viewing, all with efficiencies significantly higher than can be obtained
with incandescent light bulbs, the backlight source used in LCD and CRT. Another
important application for OLEDs is in lighting. The high efficiencies and excellent color
qualities for white OLEDs (WOLED) make them promising candidates for replacing
conventional incandescent light sources.

Figure 1.1 Examples of products using OLED technology

1.2 Structure and operation mechanism
1.2.1 Device structure
An OLED is comprised of multi-layer, amorphous, thin organic or organometallic
films that are sandwiched between two electrodes, one of which needs to be transparent
so that light can be perceived. The classic structure is shown in Figure 1.2, which is
consisted of a hole-transporting layer (HTL), electron-transporting layer (ETL) and
emissive layer (EML). The device thicknesses of 2000 Å or less is critical for lowering

3
drive voltages to the 5-10 V level, because the bulk properties of most organic molecules
are more similar to insulators than metallic conductors. This thickness also ensures that
devices are transparent so that light absorption by organic materials can be minimized.
OLEDs can be fabricated on different types of substrates ranging from rigid glass to
flexible polymers coated with a conductive layer, usually as the anode. Small organic
molecules are usually grown as smooth thin films by vapor deposition (sublimation) onto
the substrate under high vacuum conditions (< 10
-5
torr), because they are typically
amorphous glasses, without the need for epitaxial growth to maintain uniformity over
large area. Polymers, which are not volatile enough for vapor deposition, are typically
spin coated or printed onto the substrate.

Figure 1.2 Classic heterjunction OLED structure

1.2.2 Charge injection
Charge injection is the first step in the operation mechanism. Materials selection
for electrodes and the organic layers next to them, HTL and ETL, is crucial for effective

4
charge injection. The electrodes are typically selected to match the orbital energies of the
organic materials as closely as possible. The anode is usually a transparent
semiconducting layer consisting of metal oxide, such as indium tin oxide (ITO), whose
work function is close to the highest occupied molecular orbital (HOMO) energies of the
organic materials acting as HTL. Low work function metals (e.g. Ca, Al, Mg) are often
vapor deposited on top of the ETL as the cathode.

Figure 1.3 Hole and electron injection from the electrodes

In order to understand the operation of an OLED in a pictorial way, the front
orbitals, highest occupied molecular orbital (HOMO) and the lowest unoccupied
molecular orbital (LUMO), of the materials involved are usually represented by the base
and top of rectangles. Figure 1.3 illustrates what happens, when the electric field is
applied to an OLED. The vertical axis in the figure is energy and the width of each
rectangle stands for the thickness of the layer. Each Holes are injected by removing one

5
electron from the HOMO of the molecule in HTL, making it a positively charged radical
cation. On the other side, the material adjacent to the cathode (ETL) gets reduced. The
electrons are extracted from the metal cathode with a low work function and added to the
LUMO, making the molecules in ETL negatively charged radical anions. The HOMO
energy is related to the ionization potential and solution oxidation potential,
4
and the
LUMO energy is related to the electron affinity and solution reduction potential.
5

The energy difference between the electrode and the relevant MO is usually small
enough that both holes and electrons can be injected either thermally or under a small
driving voltage. However, there exist other channels for electron injection, tunneling
6

and midgap states.
7
It is widely accepted that reaction and damage occur at the interface
between cathode and ETL after metal deposition. The studies have shown that the metal
may penetrate 50Å or more into the organic film. The interaction between the metal and
organic molecules in ETL ranges from minor “solvation” effect to forming reduced
organic species and even bond cleavage.
8
These interactions lead to the formation of new
species with so-called midgap states, different energy levels that fall between those of the
cathode and the pristine organic materials. Injection of electrons into these midgap states
will be much more facile, comparing to pristine ETL materials.

1.2.3 Charge conduction
Charge conduction in organic device materials occurs through a hopping
mechanism. The carriers hop via the HOMOs and LUMOs of neighboring molecules

6
toward the opposite electrode. This process can be seen as an outer sphere electron-
transfer reaction between a charged and a neutral species and is a result of the applied
potential.
Different theories have been developed to study the charge conduction. The
Marcus theory, or polaron model was first introduced in 1990s.
9,10
The theory predicts
the mobility of charge carriers in organic solids based on the amount of reorganization
energy, an energy barrier, involved in the process of electron transfer. The higher the
barrier, the poorer the charge conductance will be. The scale of the barrier is determined
by the degree of structural distortion the carrier molecules undergo to reach the transition
state of the charge-transfer reaction. Based on this theory, an efficient charge conductor
will have a small structural difference between neutral and ionized species.
The disorder model sees charges migrate in an amorphous film consisting of
molecules with dipoles having random directions.
11
Every molecule in the film is
surrounded by different local field, depending on the intermolecular spacing and
orientation. As a result, some sites in the film have surrounding field which stabilizes the
carrier and some have the destabilizing one, which make those spots traps and antitrap,
accordingly. Beside the micro-potential environment, permanent dipoles generated by
combining all randomly positioned anisotropic molecules also affect the behavior of the
carriers.
It is noted that the charge mobility in organic materials is orders of magnitude
smaller than in crystalline inorganic semiconductors. For example, crystalline silicon has

7
the charge mobility around 1000 cm
2
V
-1
S
-1
,
12
while the best performance from organic
materials used in OLED does not exceed 10
-4
cm
2
V
-1
S
-1
.
13
This is mainly due to the
relatively small overlap of electronic wavefunctions of neighboring molecules, and other
irregularities of molecular packing within the layers. In addition, most organic materials
conduct one type of charge more efficiently than the other, which is the reason that
different materials are designed and chosen to function as HTL and ETL. Most common
hole-transporting materials are triarylamine derivatives and on the other hand, some
heterocyclic aromatic compounds, either pure organic or metal complexes, are known to
be good electron-transporting materials. Some examples are shown in Figure 1.4.

Figure 1.4 Examples of common hole-transporting (top) and electron-transporting
(bottom) materials

8
1.2.4 Exciton formation
The holes and electrons injected from the opposite electrodes will eventually get
close and meet somewhere inside the device. The process called exciton formation
occurs. One less obvious and underappreciated factor making this process possible is the
mismatch of the energy levels between EML and neighboring layers. As an example, the
deep HOMO of the ETL and the shallow LUMO of the HTL in Figure 1.3 provides the
energy barrier. At the interface, holes and electrons are “halted” from moving towards
the opposite electrode, increasing the chance for them to attract and recombine.
Assuming an electron (in LUMO of a reduced molecule) is approaching a hole (in
HOMO of an oxidized molecule) through hopping between neighboring molecules, when
they are still far apart, the electron move independently. However, once the electron gets
to a distance at which the Coulombic attraction of the positively-charged hole can be felt,
two charges commit to each other and stay bounded, barring other perturbations breaking
the correlation. Induced by this attraction, an exciton is formed. Based on the quantum
mechanics, each hole and electron has a spin, either up or down. The result is that the
exciton formed could be in any one of the four possible states, one singlet state and one
triplet state which contains three substates. From simple statistics, the ratio of singlets to
triplets can be derived to be 1:3.
14
The energy difference between singlet and triplet
states is undistinguishable until the electron reaches the nearest neighbor molecule of the
hole. The cause of it is the overlap of the hole and electron wavefunctions and the

9
exchange interaction. The electron will then jump to the same molecule where the hole is
residing on, essentially equal to bringing it to its excited state.
Unlike typical inorganic semiconductors, the organic materials possess relatively
low dielectric constant. The excitons formed in organic solids are Frenkel-type excitons
whose binding energies are much larger, comparing to that of inorganic semiconductors.
Consequently the radii are much smaller. Understandably, the singlet/ triplet ratio is also
different for inorganic semiconductors, where the exciton binding energy is small. Holes
and electrons with antiparallel spins recombine via inter-band radiation, while those with
parallel spins simply will not recombine.

Figure 1.5 Diagram of possible relaxation pathways for the excitons formed
15

10
1.2.5 Recombination and energy transfer
After an exciton is formed, it is usually deactivated through recombination or
energy transfer. Figure 1.5 illustrates the possible relaxation pathways. The organic
molecules often exhibit an efficient fluorescent emission, as the transition S
1
S
0

happens. However, the intersystem crossing (S
1
T
1
) rate and relaxation rate from
T
1
S
0
are very small. Consequently, the decay from triplet excited state occurs
normally non-radiatively through heat under ambient conditions. If excitons are formed
on certain compounds which possess strong spin-orbital coupling, transition metal
complexes for example, intersystem crossing becomes a competitive pathway for the
S
1
S
0
transition. In these cases, fluorescence is not observable, instead, the electron in
S
1
state will switch to T
1
state efficiently. Moreover, the radiation from T
1
S
0
becomes
sufficiently efficient so that phosphorescence can be detected at room temperature. As a
result, excitons from all four possible states can be utilized and converted into light,
which is a three-fold improvement in luminescence efficiency, in principle.

Figure 1.6 Förster and Dexter energy transfer mechanism

11
Besides recombination, excitons can transfer their energy to another molecule in
close proximity through two mechanisms, Förster and Dexter (Figure 1.6). Förster
mechanism involves a Coulombic interaction between two dipoles (participating
molecules), one in its excited state and one in ground state. While the dipole in excited
state relaxes, the Coulombic interaction induces the excitation of the other dipole through
space. The Förster energy transfer needs a good overlap integral between the two
transition state dipole moment to be efficient and usually has an effective radius of 30Å
or more. Extinction coefficient, distance and quantum yield are among other important
parameters which may affect the energy transfer through Förster mechanism.
16

Dexter mechanism is essentially a concerted electron exchange process between
two neighboring molecules. During this process, energy transfer is realized through the
formation of a transient exciplex between two molecules involved. The rate of Dexter
energy transfer is also strongly influenced by the integral overlap, along with the distance.
The rate of Dexter transfer drops exponentially, when the separation gets larger.
17

1.2.6 Efficiency
In general, the external efficiency, η
EL
, of an OLED can be summarized by the
following equation:
η
EL
=γη
PL
η
C
η
S
Eqn. 1.1

where γ can be estimated from the number of excitons formed per unit volume and unit
time, divided by the charge carriers passing through. It is obvious that the charge balance

12
is critical to this variable. η
PL
is the photoluminescence quantum efficiency of the emitter.
In most cases, only the relaxation of singlet excitons is spin-allowed, in which way 75%
of the energy will be lost from triplet excitons. However, phosphorescent emitters are
able to utilize otherwise irradiative triplets to give η
PL
a value close to 1. η
C
is the
fraction of photons coupled out of the device, which is related to device structure and
refractive indices of the materials used. Empirically, this factor is estimated to be around
20%. η
S
is the ratio of radiative transition from singlet states. In absence of competing
radiationless transitions, the value is generally set to 1.

1.3 Evolvement of OLED technology
1.3.1 Electroluminescence
The discovery of electroluminescence (EL) can be dated back to 1907, when
Captain Henry Joseph Round observed yellow light from a silicon carbide (SiC) detector
with current passing through.
18
In 1965, the first report of EL in organic materials was
made by Helfrich and Schneider,
19
who applied 50-1000 V to millimeter-sized anthracene
crystals and observed fluorescence. The promise of EL from organic materials in
practical applications was demonstrated by a team at Kodak with a thin-film (~ 100nm)
two-layer organic electroluminescent device two decades later.
2
This simple device
opened the door to an interesting technology based on organic EL, organic light-emitting
diodes (OLED). Thanks to the attention from both scientific and industrial community,

13
OLED has undergone tremendous development and has been successfully implemented
into applications such as displays and lighting.

1.3.2 First Heterojuction OLED
Before the successful application of organic luminescence in light-emitting
devices, two problems need to be addressed: the high resistivity of organic materials and
imbalanced charge injection from the electrodes. These two problems were solved by
Tang and van Slyke with the thin film heterostructure concept for the OLEDs. Their
device was consisted of two different layers, 4,4’-bis[N-(1-napthyl)-N-phenyl-amino]
biphenyl (α-NPB) and tris-(8-hydroxyquinoline) aluminum (AlQ
3
). This simply device
achieved an external quantum efficiency (photons/electron) of ~1% and exhibited green
emission from AlQ
3
. This is due to the fact that AlQ
3
has a smaller HOMO-LUMO
energy gap than NPD does, making energy transfer from NPD to AlQ
3
favorable.
Additionally, NPD transports holes more quickly than AlQ
3
transports electrons.
Therefore, the excitons are expected to generate within the AlQ
3
layer, near the
NPD/AlQ
3
interface.

1.3.3 Doping
Shortly after the report of first efficient heterojuction OLED, the host-dopant
concept was adopted in order to further improve the efficiency.
20
It is well known that
most organic materials suffer significant self-quenching in neat solid-state, while the

14
host-guest design solves the problem by dispersing small amount emissive dopant, which
has smaller HOMO-LUMO energy gap, into a wider gap, non-emissive matrix layer.
Within such system, energy will eventually be converted to light from dopant molecules,
which results in another benefit of doping, the color of the device can be tuned by
incorporating dopants of different energy.
21

Figure 1.7 Charge movement and recombination mechanism in host/dopant system

Two preliminary mechanisms of charge recombination and emission can happen
in the host-dopant system, after opposite charges are injected into HTL and ETL in a
similar fashion to an undoped device and later meet in EML: energy transfer from host to
dopant and charge trapping by low-energy dopant molecules (Figure 1.7). In the first
scenario, opposite charges form excitons at host molecules and rapidly dump the energy
to dopant molecules in proximity, where radiative decay takes place, yielding light. In

15
the second mechanism, without any involvement of the host, a pair of migrating opposite
charges is confined at a dopant molecule, before forming excitons and recombining.

1.3.4 Application of phosphorescent materials in OLED
The holes and electrons in OLEDs are odd electron species with an equal
distribution of m
s
= ± ½. Thus, when holes and electrons recombine to form excitons, a
statistical mixture of singlet (anti-parallel electron spins) and triplet (parallel spins)
excitons are generated.
22,23
This leads to a population of excitons that is 25% singlet and
75% triplet, which has a profound effect on OLED efficiency. At the beginning,
fluorescent laser dyes that emit only from singlet states were used as the dopants. The
doping percentage is usually low (~1%), because energy can be transferred efficiently
from the host to fluorescent dyes via long range coulombic interaction (Förster
mechanism), as shown in Figure 1.6. However, the inherent drawback of using
fluorescent dyes is that they can only collect 25% of the excitons electrically generated.
24

Due to the fact that efficient Förster energy transfer requires spin conservation,
25
the
other 75% of the energy is lost to nonradiative decays.
In the late 1990’s a new family of emissive dopants was found to give significant
increases in OLED efficiency. The best examples are metal complexes containing heavy
metal atoms, such as Ir, Pt and Ru. Generally speaking, phosphorescence from organic
molecules is a spin forbidden process, which has lifetimes on the order of milliseconds to
seconds and usually can only be observed at low temperature. However, with the

16
presence of heavy atoms, the lifetime of phosphorescence can be greatly reduced by the
strong spin-orbit coupling.
25
Phosphorescent dopants readily convert all singlets to
triplets and emit efficiently from the triplet state on the order of microseconds.
26,27

Phosphors also have a large Stokes shift between absorption and emission, therefore they
are less susceptible to self-quenching. By harvesting all triplet and singlet excitons, a
carefully designed device has been demonstrated to emit with nearly 100% internal
efficiency.
28

Figure 1.8 State-mixing in complexes containing heavy metals (exemplified with Ir)

The emission energy for organometallic phosphors is closely related to the
structure of organic ligands. A great number of efficient phosphorescent emitters have
been developed, covering most of the visible spectrum.
29,30
The emission from these
complexes originate from their lowest energy triplet excited state, a mixture of the triplet
(cyclometalating) ligand-centered state (
3
LC) and the singlet and triplet metal-to-ligand

17
charge transfer (
1,3
MLCT) state, according to spectroscopic analysis (Figure 1.8). By
changing the ratio of the two states, the emission energy of the complex can vary
considerablly.
30
One good example is that, with different ancillary (“non-emissive”)
ligands, the amount of MLCT character in the excited state can be modified. Even
though the cyclometalating ligands are the same, a shift up to 50nm in the spectrum has
been reported.
31

Since the first introduction of Ir based phosphors in OLEDs,
32
over 200 different
Ir complexes have been successfully applied into OLEDs, most of which demonstrated
external quantum efficiencies (EQE) greater than 8%.
33
Several groups have reported the
devices with EQE > 20%, corresponding to a close to 100% internal efficiencies, using
Ir-based materials.
34-36

1.3.5 Polymers in OLED
The EL in conjugated polymers and polymer-based LED (PLED) was first
reported in 1990 using poly(para-phenylene vinylene) (PPV).
3
In a typical PLED, a thin
film of an emissive polymer is sandwiched between two electrodes. Since the efficiency
of the device depends heavily upon the balanced charge injection, polymers used in an
LED must be designed to have the HOMO and LUMO matching desired electrodes as
closely as possible. These problems may be partially overcome by adding charge-
transporting layers whose energy levels are intermediate between those of the electrode
and emissive material. However, the fabrication of such multilayer structures using

18
inexpensive and timesaving techniques, such as spin casting and ink jet printing, often is
nontrivial. Issues like cross-solubility and morphology control are among those most
encountered.
Several strategies have been developed to overcome the dissolution issues. The
simplest approach is to use materials soluble in orthogonal solvent systems for each
layer.
37
Other methods involving a liquid buffer layer
38
and self-assembled layer
39
have
also been introduced to prepare multilayer structures. Using cross-linkable materials
perhaps is the most elegant strategy among them all. Materials with cross-linking unit
can be either spin-cast or ink jet printed and then transformed into insoluble films in situ
by light or heat treatment. This provides an ideal solution to avoid tedious synthesis and
purification that are often encountered for polymers. Higher purity layers obtained from
this approach also result in better reproducibility of device results. Moreover, it provides
better morphological stability
40
and reduced crystallization
41
which are the main
problems frequently occurred in small-molecule- based devices.

1.4 White OLED (WOLED)
Up to date, OLED has been widely used in commercial products like displays and
illumination devices, where devices emitting white light are successfully implemented.
Benefits, such as low driving voltage, high efficiency, and potentials in growing on large
scale are very appealing. Since OLEDs generally have very high external quantum and
power efficiencies for monochromatic light emission, a combination of them in a single

19
device to achieve white light was a natural solution. However, several challenges require
extreme cautions during materials selection and device structure design. One inherent
problem is the voltage dependence of the emission spectrum caused by the drifting of the
recombination zone, a result of imbalanced charge injection and transportation.
42
Other
literatures have shown that the order of introducing each dopant,
43,44
dopant concentration
in the each layers
45
and the thickness of each layer
46
are all important factors to achieve
quality white emission and good efficiency. It is easy to see that improving the
performance of WOLED is very important to the success of OLED in the future.
CIE coordinates refer the x,y-coordinates of (0.33, 0.33) in the chromaticity
coordinate system to an ideal white light source. Another quantitative measure, the color
rendering index (CRI), describes the color shift which occurs when an object is
illuminated by the light source and a reference source of comparable color temperature.
The CRI values can range from 0 to 100, with 100 indicating no color shift.
The most common method generating white light is RGB color model, in which
red, green and blue (component) light are superimposed at certain intensity ratio to create
a white light to human eyes. Therefore, the most intuitive way to construct WOLED is to
use three emitters making up the different components and covering the visible spectrum
evenly. The simplest structure of a three-component WOLED dopes all three emitters
into single emissive layer.
47-49
The use of phosphorescent emitters in this design
generally leads to highly efficient devices,
50
despite the readily energy transfer among
dopants which requires careful adjustment on doping percentages to achieve well-

20
balanced white emission. In the hope of improving the performance of this simple
WOLED, attempts to place dopants in different emissive layers were also reported. Both
fluorescent
47,51,52
and phosphorescent
53
emitters were successfully used in this stacked
concept. However, the device structure becomes more complicated,
54
because of the
increase number of layers and the manipulation of the recombination zone.
White light can also be obtained by combining only two component colors. One
of the approaches takes advantage of excimer emission from planar Platinum
complexes,
55,56
which is originated from the bounded excited states when complex-
complex distance is small.
57-61
Dependent on the complex structure and doping
percentage,
55
excimer emission, generally red-shifted and broad, and the emission from
monomeric species can be simultaneously generated from single dopant to create the
white emission. Unfortunately, the nature of excimer emission decides inherently that it
is not an efficient way to convert excitons into light.
A relatively new approach starting to gain popularity is to use four emitters, blue,
green, yellow and red.
62
To obtain more balanced white emission, the yellow phosphor
was introduced to fill the gap between green and red region, which is very sensitive to
human eyes.

21
1.5 Chapter 1 References

(1) Pope, M.; Kallmann, H. P.; Magnante, P. J. J. Chem. Phys. 1963, 38, 2042.
(2) Tang, C. W.; VanSlyke, S. A. Appl. Phys. Lett. 1987, 51, 913.
(3) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay, K.;
Friend, R. H.; Burns, P. L.; Holmes, A. B. Nature 1990, 347, 539.
(4) D'Andrade, B. W.; Datta, S.; Forrest, S. R.; Djurovich, P.; Polikarpov, E.;
Thompson, M. E. Org. Electron. 2005, 6, 11.
(5) Djurovich, P. I.; Mayo, E. I.; Forrest, S. R.; Thompson, M. E. Org. Electron. 2009,
10, 515.
(6) Bushuk, S.; Douglas, W.; Kalvinkovskaya, Y.; Klapshina, L.; Rubinov, A.;
Bushuk, B.; Stupak, A. J. Lumin. 2003, 105, 81.
(7) Schubert, E. F. Light-Emitting Diodes; Cambridge University Press: Cambridge,
UK, 2003.
(8) Lees, A. J. Chem. Rev. 1987, 87, 711.
(9) Marcus, R. A. J. Chem. Phys. 1956, 24, 966.
(10) Marcus, R. A. Angew. Chem. Int. Ed. 1993, 32, 1111.
(11) Bassler, H. Phys. Status Solidi B 1993, 175, 15.
(12) Sze, S. M. Physics of Semiconductor Devices; 2nd ed.; John Wiley: New York,
NY 1981.
(13) Muccini, M. Nat. Mater. 2006, 5, 605.
(14) Baldo, M. A.; O'Brien, D. F.; Thompson, M. E.; Forrest, S. R. Phys. Rev. B 1999,
60, 14422.
(15) Yersin, H. In Transition Metal And Rare Earth Compounds III 2004, 241, 1.
(16) Förster, T. Göttingen: Vandenhoeck and Ruprecht 1951.
(17) Dexter, D. L. J. Chem. Phys. 1953, 21, 836.

22
(18) Round, H. J. Electr. World 1907, 19, 309.
(19) Helfrich, W.; Schneider, W. G. Phys. Rev. Lett. 1965, 14, 229.
(20) Tang, C. W.; VanSlyke, S. A.; Chen, C. H. J. Appl. Phys. 1989, 65, 3610.
(21) Shoustikov, A. A.; You, Y.; Thompson, M. E. IEEE J. Sel. Top. Quantum
Electron. 1998, 4, 3.
(22) O'Brien, D. F.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett.
1999, 74, 442.
(23) Segal, M.; Baldo, M. A.; Holmes, R. J.; Forrest, S. R.; Soos, Z. G. Phys. Rev. B
2003, 68, 075211.
(24) Brown, A. R.; Pichler, K.; Greenham, N. C.; Bradley, D. D. C.; Friend, R. H.;
Holmes, A. B. Chem. Phys. Lett. 1993, 210, 61.
(25) Turro, N. J. Modern Molecular Photochemistry; Benjamin/Cummings: Menlo
Park, CA, 1978.
(26) Justel, T.; Nikol, H.; Ronda, C. Angew. Chem. Int. Ed. 1998, 37, 3085.
(27) Aziz, H.; Popovic, Z. D.; Hu, N. X. Appl. Phys. Lett. 2002, 81, 370.
(28) Adachi, C.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. J. Appl. Phys. 2001, 90,
5048.
(29) Brooks, J.; Babayan, Y.; Lamansky, S.; Djurovich, P. I.; Tsyba, I.; Bau, R.;
Thompson, M. E. Inorg. Chem. 2002, 41, 3055.
(30) Li, J.; Djurovich, P. I.; Alleyne, B. D.; Yousufuddin, M.; Ho, N. N.; Thomas, J. C.;
Peters, J. C.; Bau, R.; Thompson, M. E. Inorg. Chem. 2005, 44, 1713.
(31) Holmes, R. J.; D’Andrade, B. W.; Forrest, S. R.; Ren, X.; Li, J.; Thompson, M. E.
Appl. Phys. Lett. 2003, 83, 3818.
(32) Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest, S. R.
Appl. Phys. Lett. 1999, 75, 4.
(33) Crabtree, R. H.; Mingos, D. M. Comprehensive organometallic chemistry III;
Elsevier: Oxford, UK, 2007; Vol. 12.

23
(34) Tsuboyama, A.; Iwawaki, H.; Furugori, M.; Mukaide, T.; Kamatani, J.; Igawa, S.;
Moriyama, T.; Miura, S.; Takiguchi, T.; Okada, S.; Hoshino, M.; Ueno, K. J. Am. Chem.
Soc. 2003, 125, 12971.
(35) Watanabe, S.; Agata, Y.; Tanaka, D.; Kido, J. J. Photopolym. Sci. Technol. 2005,
18, 83.
(36) Meerheim, R.; Walzer, K.; Pfeiffer, M.; Leo, K. Appl. Phys. Lett. 2006, 89,
061111.
(37) Ma, W. L.; Iyer, P. K.; Gong, X.; Liu, B.; Moses, D.; Bazan, G. C.; Heeger, A. J.
Adv. Mater. 2005, 17, 274.
(38) Tseng, R. J.; Chiechi, R. C.; Wudl, F.; Yang, Y. Appl. Phys. Lett. 2006, 88,
093512.
(39) Ho, P. K. H.; Granstrom, M.; Friend, R. H.; Greenham, N. C. Adv. Mater. 1998,
10, 769.
(40) Bellmann, E.; Shaheen, S. E.; Grubbs, R. H.; Marder, S. R.; Kippelen, B.;
Peyghambarian, N. Chem. Mater. 1999, 11, 399.
(41) Fenter, P.; Schreiber, F.; Bulovic, V.; Forrest, S. R. Chem. Phys. Lett. 1997, 277,
521.
(42) Sirringhaus, H. Nat. Mater. 2003, 2, 641.
(43) Zhu, X. L.; Sun, J. X.; Yu, X. M.; Wong, M.; Kwok, H. S. Jpn. J. Appl. Phys.,
Part 1 2007, 46, 4054.
(44) Baek, H. I.; Lee, C. H. J. Phys. D-Appl. Phys. 2008, 41, 124504.
(45) Ho, C. L.; Wong, W. Y.; Zhou, G. H.; Yao, B.; Xie, Z. Y.; Wang, L. X. Adv.
Funct. Mater. 2007, 17, 2925.
(46) D'Andrade, B. W.; Forrest, S. R. Adv. Mater. 2004, 16, 1585.
(47) Kido, J.; Kimura, M.; Nagai, K. Science 1995, 267, 1332.
(48) Tasch, S.; List, E. J. W.; Ekstrom, O.; Graupner, W.; Leising, G.; Schlichting, P.;
Rohr, U.; Geerts, Y.; Scherf, U.; Mullen, K. Appl. Phys. Lett. 1997, 71, 2883.
(49) Kawamura, Y.; Yanagida, S.; Forrest, S. R. J. Appl. Phys. 2002, 92, 87.
(50) D'Andrade, B. W.; Holmes, R. J.; Forrest, S. R. Adv. Mater. 2004, 16, 624.

24
(51) Jiang, X. Y.; Zhang, Z. L.; Zhao, W. M.; Zhu, W. Q.; Zhang, B. X.; Xu, S. H. J.
Phys. D-Appl. Phys. 2000, 33, 473.
(52) Huang, L.; Wang, K. Z.; Huang, C. H.; Gao, D. Q.; Jin, L. P. Synth. Met. 2002,
128, 241.
(53) D'Andrade, B. W.; Thompson, M. E.; Forrest, S. R. Adv. Mater. 2002, 14, 147.
(54) Sun, Y. R.; Giebink, N. C.; Kanno, H.; Ma, B. W.; Thompson, M. E.; Forrest, S.
R. Nature 2006, 440, 908.
(55) Adamovich, V.; Brooks, J.; Tamayo, A.; Alexander, A.; Djurovich, P. I.;
D’Andrade, B. W.; Adachi, C.; Forrest, S. R.; Thompson, M. E. New J. Chem. 2002, 26,
1171.
(56) D'Andrade, B. W.; Brooks, J.; Adamovich, V.; Thompson, M. E.; Forrest, S. R.
Adv. Mater. 2002, 14, 1032.
(57) Cheung, T. C.; Cheung, K. K.; Peng, S. M.; Che, C. M. J. Chem. Soc., Dalton
Trans. 1996, 8, 1645.
(58) Zheng, G. Y.; Rillema, D. P. Inorg. Chem. 1998, 37, 1392.
(59) Buchner, R.; Cunningham, C. T.; Field, J. S.; Haines, R. J.; McMillin, D. R.;
Summerton, G. C. J. Chem. Soc., Dalton Trans. 1999, 5, 711.
(60) Lai, S. W.; Lam, H. W.; Lu, W.; Cheung, K. K.; Che, C. M. Organometallics
2002, 21, 226.
(61) Lu, W.; Mi, B. X.; Chan, M. C. W.; Hui, Z.; Zhu, N. Y.; Lee, S. T.; Che, C. M.
Chem. Commun. 2002, 3, 206.
(62) Yu, X. M.; Zhou, G. J.; Lam, C. S.; Wong, W. Y.; Zhu, X. L.; Sun, J. X.; Wong,
M.; Kwok, H. S. J. Organomet. Chem. 2008, 693, 1518.

25
Chapter 2 Study of energy transfer and triplet exciton diffusion in hole
transporting host materials

2.1 Abstract
A device structure is used in which the hole transporting layer (HTL) of an OLED
is doped with either fluorescent or phosphorescent emitters, i.e. anode/HTL-host/hole
blocker/electron transporting layer/cathode. The HTL-hosts have higher HOMO energy
allowing holes to be transported without being trapped by dopant molecules, avoiding
direct recombination on the dopant. The unconventional mismatch of HOMO energies
between host and dopant allow for the study of energy transfer in these host/guest
systems and triplet exciton diffusion in the HTL-host layers of OLED devices, without
the complication of charge trapping at dopants. The host materials examined here are
tetraaryl-p-phenylenediamines. Data shows that Förster energy transfer between these
hosts and emissive dopant in devices is inefficient. Triplet exciton diffusion in these host
materials is closely related to molecular structure and the degree of intermolecular
interaction. Host materials that contain naphthyl groups demonstrate longer triplet
exciton diffusion lengths than those with phenyl substituents, consistent with DFT
calculations and photophysical measurements.

2.2 Introduction
Shortly after the first heterojunction organic light emitting diode (OLED) was
reported,
1
doped emitter systems were introduced into OLEDs and found to lead to more

26
efficient devices with tunable emission color.
2
To obtain emission from doped emitters,
the efficient harvesting of excitons by dopant molecules is crucial. Localization of
excitons on dopants can occur either through direct charge trapping or by energy transfer
from excitons formed in the host matrix. Both mechanisms are believed to be involved in
electroluminescence, although in many cases, charge trapping is found to be the principal
mechanism for dopant emission.
3-5
Consequently, host materials that are used in OLEDs
are often designed and optimized to promote charge trapping and subsequent carrier
recombination at the dopants. Although high efficiencies have been achieved, the
relative contribution these two mechanisms have on exciton formation, and the individual
roles they play in the performance of OLEDs, is hard to distinguish and often missing
from discussion.
Two types of excitons, singlet and triplet, can be created upon charge
recombination in a host material. The exciton migrates to the dopant by intermolecular
energy transfer processes. Singlet excitons can be transferred either through Förster
mechanism, a process involving dipole-dipole interaction mediated energy transfer, or by
hopping among neighboring molecules through electron-exchange (Dexter) mechanism.
Förster transfer involves a fairly long range interaction, which can lead to efficient energy
transfer over 8 nm,
6
however, efficient transfer typically takes place at less than 5 nm for
organic materials.
7
Since Dexter transfer requires electron exchange, it is only efficient
for nearest neighbor energy transfer. The short lifetimes of singlet excitons (nanoseconds)
lead to diffusion lengths of 10’s of nm. Due to the low oscillator strengths for electronic

27
transitions in organic triplet materials, Förster energy transfer is inefficient.
8
Thus, for
organic materials, triplet excitons are transferred principally through a Dexter process.
Despite the short range of Dexter transfer, the longer lifetime of triplets (often greater
than milliseconds) enables them to diffuse a significant distance. Triplet excitons have
been reported to migrate more than 100 nm in organic amorphous films,
9
however,
distances are typically closer to 100 Ǻ.
10
Efficient Dexter transfer relies on small
separations and direct orbital overlap between donor-acceptor pairs.
11
Studies have found
that triplet energy transfer follows Marcus theory for both electron and hole transfer due
to the similarity to weak coupling, so that transfer is favorable in systems with small
reorganization energies and large overlap integrals between the molecular orbitals of the
interacting species.
12,13

The importance of exciton diffusion to OLED device performance has been
realized since the early development of OLEDs.
2, 14
The most commonly used approach
to study exciton diffusion is to insert a doped layer into devices to serve as a sensing
layer at various distances away from the recombination zone in the OLED. Based on the
photophysical or electroluminescence data obtained, the exciton diffusion process can be
modeled and analyzed.
15-20
Exciton diffusion in OLED devices has only been
investigated using a limited number of materials, such as Alq
3
and CBP. In those studies,
charges were trapped or partially trapped in the doped region, since the dopant molecules
have shallower (less negative) HOMO and deeper (more negative) LUMO compared to

28
the host. Therefore, the location of the recombination zone is not well confined to the
interfacial region in these devices.
In this paper, we explore a new device structure that avoids charge trapping on
dopants to study energy transfer and triplet exciton diffusion in OLEDs. The selection of
the host is key to the success of this study and a series of symmetric tetraaryl-p-
phenylenediamine derivatives have been found to be good candidates for such host
materials. Tetraaryl-p-phenylenediamine are known to have tunable HOMO levels, high
thermal stability, good hole mobility and high singlet and triplet energy.
21, 22
These
attractive features make it possible for tetraaryl-p-phenylenediamine derivatives to
simplify the device structure by being used as both hole injection material and host.
Using new host materials along with the improved device architecture, singlet and triplet
excitons are independently harvested in devices, and the origin of dopant emission
(through charge trapping or energy transfer) is easily confirmed. The effect that the
molecular structure of hosts has on triplet exciton diffusion in the solid state is also
investigated.

2.3 Results and Discussion
The device architecture used for the studies reported here involves a simple
structure, which is designed to avoid charge trapping at dopant molecules and provide a
single recombination zone at the host-HTL/ETL interface (Figure 2.1). This structure
requires host materials to have a higher (less negative) HOMO energy than the dopants
(C6 and PQIr in this study). The high HOMO energy also enables the host to serve as an

29
effective hole injecting layer, simplifying the structure. The injected holes will travel
through host matrix, without being trapped by dopant molecules, and reach the host/ETL
interface, forming a single recombination zone. It is important for both of the singlet and
triplet energies of the host to be greater than those of the dopant so that the energy
transfer is favorable in the host/guest system for both fluorescent and phosphorescent
dopants. Tetraaryl-p-phenylenediamine derivatives fulfill all of the criteria for host
materials in the architecture described above. The molecular structure of the
phenylenediamine derivatives studied in this paper are given in Figure 2.1. The
unsubstituted compound, DDP, was found to crystallize rapidly after vacuum deposition
as a thin film. Two approaches were adopted to modify the molecular structure and give
materials that form amorphous films suitable for device studies. Adding alkyl groups to
the peripheral phenyl groups at (TTP, TTTP and TDDP) gives compounds that form
glassy thin films. Alternatively, the central phenylene was replaced with a bulkier 2,6-
naphthylene, NDDP. NNP, with naphthyl groups at periphery, was also synthesized as a
complementary structure. For comparative purposes, a widely used host material in
OLEDs, TPD, was also chosen due to the similarity of its molecular structure and
electronic properties to that of the other candidate host compounds.

2.3.1 Electrochemistry
Electrochemical data was used to estimate the HOMO energies for the materials.
Measurements were carried out using cyclic voltammetry (CV) and the

30
oxidation/reduction potentials are listed in Table 2.1. All host candidates undergo two
separate, fully reversible oxidation processes. The HOMO and LUMO energies were
calculated from the first oxidation and reduction potential respectively, using previously
published correlations.
23, 24
All of the host candidates are more easily oxidized than the
dopants chosen here, putting their HOMO levels above those of the dopants. The
designed mismatch of HOMO energy between the host and dopant should prevent hole
trapping by the dopant. The estimated HOMO energies are close to that of ITO (4.7 eV),
which should facilitate efficient hole injection into host layer.

O O N
S
N
Coumarin 6 (C6)

N
Ir
O
O
2
PQIr

Figure 2.1 Device structure and chemical structures of the molecules used in this study.
Horizontal lines and numbers on the lines in device structure represent HOMO and
LUMO energy levels (eV).

31
Figure 2.1, continued
mCP
N N

N N
BCP

N N
TPD

N N
DDP

N N
TTP

N N
NNP

N N
TDDP

N
N
NDDP

Table 2.1 Redox, photoluminescence, energy levels and quantum yield data for each
material used in devices

Molecule E
1/2
Ox

[V]
a
E
1/2
Re

[V]
a

Emission [nm]

HOMO/LUMO
[eV]
Quantum
Yield [%]
d
Fl
b
Ph
c
TPD 0.36 / 0.47 -2.93 394 / 402 505 -5.1 / -1.3 0.19
DDP 0.18 / 0.60 -2.65 390 / 398
e
452 -4.8 / -1.6 0.06
TTP 0.16 / 0.62 -2.65 393 / 404 454 -4.8 / -1.6 0.05

32
Table 2.1, continued
TTTP 0.13 / 0.56 -2.71 398 / 414 456 -4.8 / -1.6 0.10
TDDP 0.08 / 0.49 -2.70 399 / 397 458 -4.7 / -1.6 0.03
NDDP 0.24 / 0.55 -2.72 423 / 428 505 -4.9 / -1.5 0.05
NNP 0.18 / 0.58 -2.66 393 / 448 508 -4.8 / -1.6 0.05
C6 0.65 -1.97 487 / 502
f
-- -5.5 / -2.4 0.69
f

PQIr 0.45 -2.19 -- 598
g
-5.2 / -2.2 0.72
g

mCP 0.78 / 0.95 -2.75 338 / 350 408 -5.7 / -1.5

--
BCP 1.13 -2.67 359 / 390 473 -6.2 / -1.6

--
a
Measured using decamethylferrocene (-0.47V versus ferrocene) as internal reference
and then converted to ferrocene.
b
In the order shown are fluorescent peak maxima in
solution and for spin-coated film.
c
The phosphorescent peak with highest energy at 77K
d
Measured using spin-coated film.
e
The film crystalizes slowly.
f
Measured from a 2%
C6-doped polystyrene film under N
2
.
g
Measured from an 8% PQIr-doped polystyrene
film under N
2
.

2.3.2 Photophysical Properties
Photoluminescence data are also provided in Table 2.1. All of the hosts fluoresce
with λ
max
values between 390–425 nm in solution at room temperature. The alkyl groups
and peripheral naphthyl groups have only small impact on the energy of fluorescence
peak position while the central naphthyl group in NDDP leads to a red-shift of ~30 nm.
The E
0-0
values of the triplet state were estimated from the highest energy peak of their
phosphorescence spectrum obtained, in a dilute frozen solution of 2-MeTHF at 77K.
Figure 2.2 shows the emission spectra of TDDP and NNP at 77K. Phosphorescence is
dominant in the case of TDDP, while NNP spectrum mainly displays fluorescence. The

33
difference comes from the energy separation between the S
1
and T
1
states, 0.4eV for
TDDP and 0.7eV for NNP. For compounds with similar electronic configurations, which
is the case here, the energy gap between states controls the relative rate of intersystem
crossing.
8
The TTP, TTTP and TDDP derivatives, consisting of a p-phenylenediamine
core and only phenyl or alkyl-phenyl substitution at N, have triplet energies near 455 nm.
Extending the π conjugation, such as with biphenyl and naphthyl cored materials, is
expected to lower the triplet energy. Thus, TPD, NNP and NDDP have triplet energies
near 505 nm. The triplet energies of all the hosts are well above that of the
phosphorescence dopant used in this study, PQIr (E
0-0
= 575 nm or 2.16 eV).

350 400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity / a.u.
Wavelength / nm
TDDP
NNP

Figure 2.2 PL spectra of TDDP and NNP in 2-MeTHF solution at 77K

34
350 400 450 500 550
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity / a.u.
Wavelength / nm
TDDP Solution
TDDP Film

350 400 450 500 550
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity / a.u.
Wavelength / nm
NNP Solution
NNP Film

Figure 2.3 PL spectra in 2-MeTHF solution and neat film for TDDP and NNP

Emission spectra of TDDP and NNP in solution and neat thin films are compared
in Figure 2.3. TDDP shows no difference between solution and film, whereas emission
from NNP film displays a considerable red-shift (55 nm). Of all of the hosts studied here,
only NNP exhibits such marked red-shift from thin film (see Supporting Information).

35
The most logical explanation for this observation is that the naphthyl groups lead to
strong π-π interactions in solid states, since similar bathochromic shifts are typically
considered to be a signature of strong intermolecular interaction.
25

350 400 450 500 550 600 650
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity / a.u.
Wavelength / nm
PL: TDDP
PL: TDDP + BCP
EL: no mCP
EL: with mCP

Figure 2.4 PL spectra of TDDP and TDDP:BCP (1:1) film compared with the EL spectra
from undoped TDDP devices with and without an mCP interlayer

2.3.3 Undoped Devices
Undoped devices with the structure of ITO/host (400 Ǻ)/BCP (400 Ǻ)/LiF (10
Ǻ)/Al (1200 Ǻ) were investigated as controls. To our surprise, all the
electroluminescence (EL) from devices showed red shifted emission relative to all of the
components in the device. TDDP is taken as an example and spectra are shown in Figure
2.4 where the broad, low-energy emission peaks around 500nm. While this signal did not
match any of the materials in the device, it did match the PL of the thin film composed of
a 50/50 mixture of TDDP and BCP, suggesting that the origin of the low energy EL is an

36
exciplex formed at the host/BCP interface. Similar exciplex formation between m-
MTDATA and BPhen has also been reported.
26
In order to suppress the exciplex
emission, a thin layer of mCP (100 Ǻ) was inserted between the host and BCP (ITO/host
(400 Ǻ)/mCP (100 Ǻ)/BCP (400 Ǻ)/LiF (10 Ǻ)/Al (1200 Ǻ)). These devices give only
host emission (Figure 2.4). While mCP is often employed in OLEDs as an electron
blocking layer, the large energy mismatch between the HOMO of mCP and the diammine
hosts (> 0.7 eV) should enable mCP to serve as an effective hole blocking layer. The
high triplet energy of mCP (408 nm, 3.04 eV) should also prevent excitons from
migrating into the BCP layer.

2.3.4 Doped Devices
In this study, OLEDs were fabricated using either 2 wt% doping of a green
fluorescent emitter (C6, emission λ
max
= 490 nm) or 8 wt% of a orange-red
phosphorescent emitter (PQIr, emission λ
max
= 600 nm) as an exciton sensing layer in the
hole transporting layer (HTL). Doped devices have a general structure of ITO/host (300-
X Ǻ)/doped layer (100 Ǻ)/host (X Ǻ)/mCP(100 Ǻ)/BCP (400 Ǻ)/LiF (10 Ǻ)/Al (1200 Ǻ).
Devices where X= 0, 50, 100 and 200Ǻ will be referred to as type I, II, III and IV
respectively in the discussion. For abbreviation purposes, F and P are used in front of
device to indicate what kind of dopant is used, namely fluorescent (C6) or
phosphorescent (PQIr). All the F and P-type devices were made using the host molecules
that are listed in Figure 2.1. The external efficiencies and voltages of the devices at

37
10mA cm
-2
for four host materials, TDDP, TPD, NDDP and NNP, are compared in Table
2.2.

Table 2.2 External quantum efficiency (EQE) and voltage data of TDDP, TPD, NDDP
and NNP devices at 10 mA/cm
2

EQE [%] / Voltage [V]
a

Undoped F-I F-II F-III P-I P-II P-III P-IV
TDDP 0.38 / 7.5 0.96 / 8.6 0.43 / 7.9 0.39 / 8.6 5.29 / 6.9 1.02 / 7.2 0.44 / 7.4 --
TPD 0.61 / 5.8 0.66 / 6.0 0.64 / 5.9 0.90 / 5.9 3.94 / 6.1 1.84 / 5.9 1.37 / 5.9 --
NDDP 0.80 / 4.9 0.68 / 6.2 0.76 / 5.9 0.77 / 6.1 2.34 / 6.2 2.85 / 5.8 2.86 / 6.0 --
NNP 0.59 / 5.0 0.67 / 6.2 0.53 / 6.2 0.46 / 6.1 2.49 / 6.1 3.23 / 6.0 5.48 6.0 4.20 / 5.8

a
The data are reproducible with an error bar less than 10%.

For the three arylenediamine based hosts listed in Table 2.2 (TDDP, NDDP and
NNP), the voltages at 10 mA cm
-2
are nearly constant for all of the doped devices,
regardless of the identity of the dopant or its location in the device, and higher than those
of their undoped counterparts. This increase in voltage may be due to either charge
trapping or scattering by the dopant.
27
The data suggests charge scattering to be the main
factor, because the voltage increase is independent of the different HOMO levels and
doping percentages of the two dopants (PQIr and C6).
28
The devices become more
resistive, as the holes can be deflected by dopant molecules whose HOMO levels are
deeper, on their way to the recombination zone. Data in Table 2.1 also suggests that
holes are not trapped by dopants. According to the table, the HOMO of TPD is the

38
closest to both dopants. This means that hole migration should be least affected in doped
TPD matrices, as evidenced by the near constant voltages for all of the TPD based
devices, including the undoped one. Previous work has also established that neither C6
nor Ir(ppy)
3
trap holes when TPD is used in such host/guest systems.
14, 28
Hole trapping
by either C6 or PQIr in these new hosts is even less likely to occur, since the hosts
studied in this work all have HOMO levels above that of TPD, and the HOMO energy of
PQIr is similar to that of Ir(ppy)
3.
25

350 400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity / a. u.
Wavelength / nm
F-I
F-II
F-III
Undoped

Figure 2.5 EL spectra from F-type TDDP devices at J=10mA/cm
2

The EL spectra from all of the undoped, mCP blocked devices only display
fluorescence from the host, whereas the spectra from doped devices vary significantly
depending on the dopant and position of the sensing layer. Figure 2.5 shows the EL
spectra from undoped and F-type devices using a TDDP host. Emission from C6 can be

39
observed in device F-I, whereas in the EL spectra of F-II and F-III device, only host
emission is seen. The same results have also been obtained for the other three host
materials. This difference in EL spectra is reflected by EQE numbers. At 10mA cm
-2
,
the F-I device has EQE of 1%, as opposed to 0.4% for the F-II, F-III and undoped devices.
The absence of C6 emission in the F-II and F-III devices lowers the efficiency to match
the undoped device. The EL spectra suggest that there is only one recombination zone
near the host/mCP interface in this architecture, and that the recombination zone appears
to be stationary since the EL spectra of all F-type devices are voltage-independent. A
C6-doped layer that is 50 Ǻ or more away from the mCP/host interface is ineffectively
sensitized through the Förster mechanism at this distance. The TDDP emission consists
of 57% of the EL spectrum of F-I device, indicating that even shorter range energy
transfer from the host to the dopant is rather inefficient. This is confirmed by the
photoluminescence observed from a spin-coated TDDP film doped with 2% C6
(Supporting Information), which matches the EL spectrum. There are a couple of
possible reasons for the poor luminescence efficiency. One is that the limited spectra
overlap between the emission of hosts and the absorption of C6 results in a relatively
short Förster energy transfer radius (about 25Ǻ). The other is C6 derivatives are known
to undergo a significant degree of self-quenching when doped at concentrations greater
than 1% into a host matrix.
29
The fact that the F-I device displays significant emission
from TDDP provides further evidence that charge is not trapped on the C6 dopant.

40
Previous work has shown exclusive emission from C6 when doped into a host with a
deep HOMO energy such as PVK.
30

400 500 600 700
0.00
2.50x10
12
5.00x10
12
Photon Counts
Wavelength / nm
P-I
P-II
P-III
350 400 450 500 550 600 650 700 750
0.0
2.0x10
12
4.0x10
12
6.0x10
12
8.0x10
12
P-I
P-II
P-III
P-IV
Photon Counts
Wavelength / nm

Figure 2.6 EL spectra from P-type TDDP and NNP devices at J=10mA/cm
2

We also examined PQIr as a dopant, with the intent to collect both singlet and
triplet excitons. The performance of P-type devices also depends on the position of the

41
sensing layer. The EL spectra of P-type TDDP devices are plotted in photon counts in
Figure 2.6. These plots allow us to easily see how the contribution to EQE from
fluorescence and phosphorescence changes for different types of devices. The P-I TDDP
device gives nearly exclusive emission from PQIr (95.4 % of the EL spectrum), while
little or no PQIr emission was observed in P-II and P-III devices. It is possible that some
emission in the P-I device occurs from direct recombination on the PQIr dopant. Several
factors that could favor this direct recombination include a charge build-up at host/mCP
interface, the high doping percentage of PQIr and the small difference in energy between
the HOMO of TDDP and PQIr. However, a small contribution of host emission in the
EL spectra suggests that recombination did take place on host molecules followed by the
energy transfer from host to dopant. The decreasing phosphorescent contribution in the
other TDDP devices eventually lowers the EQE of the P-III device to that of an undoped
device (Table 2.2). In contrast, all the P-type NNP devices show strong PQIr emission.
The contribution from PQIr to the EL spectra decreases as the dopant is placed further
from the recombination zone, but it still accounts for 67% of the observed emission in the
P-IV device (Figure 2.6).
To confirm that PQIr is excited by triplet energy transfer and not by charge
trapping followed by carrier recombination or singlet energy transfer, we decided to
employ a triplet filter and prepared a set of P-type devices using NNP as the host. 9,10-
Di(2-naphthyl)anthracene (ADN) is an ideal choice for the triplet filter, because it has a
high oxidation potential and a triplet energy that is well below those of NNP and PQIr.
31

42
The structure of the double-doped (PQIr and ADN) P-type device is NNP (200
Ǻ)/8%PQIr:NNP (50 Ǻ)/NNP (50 Ǻ)/5%ADN:NNP (100 Ǻ)/NNP (100 Ǻ)/mCP (100
Ǻ)/BCP (400 Ǻ)/LiF (10 Ǻ)/Al (1200 Ǻ). In this device, ADN will neither trap holes nor
collect singlet energy from recombination zone, as it is too far from NNP/mCP interface.
On the other hand, it will efficiently trap triplet excitons before they can reach PQIr.
350 400 450 500 550 600 650 700 750
0.0
5.0x10
11
1.0x10
12
1.5x10
12
2.0x10
12
2.5x10
12
3.0x10
12
Wavelength / nm
Photon Counts
Undoped NNP
Double-doped
ADN-doped
PQIr-doped

Figure 2.7 EL spectra from double-doped and control devices when J=10mA/cm
2
.

The EL spectrum of this double-doped device, along with other control OLEDs
are provided in Figure 2.7. The ADN-doped and PQIr-doped device has the structure of
NNP (300 Ǻ)/5%ADN:NNP (100 Ǻ)/mCP (100 Ǻ)/BCP (400 Ǻ)/LiF (10 Ǻ)/Al (1200 Ǻ)
and NNP (200 Ǻ)/8%PQIr:NNP (100 Ǻ)/NNP (200 Ǻ)/mCP (100 Ǻ)/BCP (400 Ǻ)/LiF
(10 Ǻ)/Al (1200 Ǻ), respectively.It is evident that the addition of the ADN doped layer
completely eliminates the PQIr emission from double-doped device, whereas the PQIr

43
emission is reasonably strong in control device where PQIr layer is at the same distance
away from NNP/mCP interface. This result is in agreement with previous data and
verifies that the PQIr emission from this device structure is due to triplet exciton energy
transfer, rather than charge trapping.
400 500 600 700
0
1
2
3
4
5
6
EQE
TDDP 0.44
TPD 1.4
NDDP 2.9
NNP 5.5
Nornalized Intensity / a.u.
Wavelength / nm

Figure 2.8 EL spectra from P-III devices using different hosts at J=10mA/cm
2
. The
spectra are normalized to the blue fluorescence peak maxima. Inset: Comparison of
EQE among P-III devices using different hosts.

The behavior of P-type TDDP devices shows that triplet excitons fail to migrate
further than 50 Ǻ in the TDDP matrix. EQEs of P-I devices follow the trend TDDP >
TPD > NDDP ≈ NNP, while the order for P-III devices is reversed, NNP > NDDP >
TPD > TDDP (Table 2.2). Since the impact of trapping and scattering by the dopant is
similar in the various hosts, this phenomenon can be reasonably explained by the
difference in transport properties of triplet excitons in these materials. The P-I device

44
using TDDP has the highest EQE among all the host materials, a result that could be due
to the short triplet exciton diffusion length which allows the PQIr dopant to efficiently
harvest triplets confined within the narrow recombination zone of the device. Using the
same argument, TDDP gives solely host emission when used in the P-III architecture,
since triplet excitons cannot diffuse through the spacer layer and be collected by PQIr
molecules located further away. In contrast, P-III devices using the other three hosts
yield much higher EQE, benefiting from the fact that more triplet excitons are able to
diffuse out of recombination zone. In other words, triplet excitons can diffuse a relatively
greater distance in these matrices. Figure 2.8 shows the contribution of PQIr for the four
hosts studied here, from which the diffusivity of triplet excitons in different host
materials can be qualitatively compared. The exciton diffusion length of TDP has been
reported to be 170 Ǻ using photocurrent spectroscopy.
32
However, based on our data, the
triplet exciton diffusion length in TPD is far from ideal and could be well below the
reported value. The contradiction may also come from the limitations of photocurrent
spectroscopy approach itself.
33

Other DDP derivatives with various alkyl substituents were also used as hosts in
P-III devices and found to have EQEs that decrease in the order of TTP > TTTP > TDDP
(Figure 2.9). The trend of EQEs can be correlated to the size and degree of alkyl
substitution in the molecule, because it has been shown from a previous study that triplet
states do not extend beyond alkyl groups in aromatic compounds.
34
Consequently, alkyl

45
groups act as a spacer to increase the separation and decrease the electronic coupling
between neighboring molecules.

0.1 1 10 100 1000
0.01
0.1
1
Quantum Efficiency / %
Current Density / mA/cm
2
TPD
TTP
TTTP
TDDP

Figure 2.9 EQE of P-III devices with host materials that do not contain naphthyl groups

2.3.5 Theoretical Analysis
To rationalize why the range of diffusion lengths is so different among the various
host materials, particularly those that contain naphthyl groups such as NNP, it is
worthwhile to examine the details of triplet exciton diffusion. Triplet energy transfer can
be modeled as a random diffusion process of triplet excitons. The diffusion length is
described in Equation 2.1.
τ D L = (2.1)
Here, D is the diffusion coefficient and τ is the lifetime. Since the lifetime of
phosphorescence among these host molecules are comparable, the diffusion length should

46
depend solely on D. The diffusion coefficient is proportional to the rate constant for
triplet energy transfer, k, according to Equation 2.2,
35

0
4 N DR k
eff
π = (2.2)
Here, R
eff
is the largest collision diameter for the two molecules undergoing triplet
energy transfer and N
0
is Avogadro’s number. In addition, the rate constant, k, should
follow the Golden Rule given in Equation 2.3 for a nonadiabatic process in a weakly
coupled system.
36

T k
T k
V k
B
B
πλ
λ π
4
) 4 / exp( 2 2 −
=
η
(2.3)
Here, λ is the reorganization energy and V is the electronic coupling term.
Therefore, based on the theoretical derivation, materials with a small λ and large V should
have large rate constants for energy transfer needed to support long diffusion lengths for
triplet excitons.
Estimates for triplet reorganization energies can be obtained from calculations
using density function theory (DFT). Following the approach described in the
experimental section, values of λ for triplet exciton transfer for DDP (an analogue of
TDD), TPD, NDDP and NNP were computed to be 0.62, 0.85, 0.61 and 0.75 eV,
respectively. The calculated reorganization energies for the compounds do not follow the
order of diffusion lengths estimated from the OLED data. For example, the large
reorganization energy of NNP does not correlate with the high conductivity of triplet
excitons demonstrated by the P-III devices made with NNP. The absence of any

47
correlation between λ and the triplet diffusion length directs attention to the coupling
term, V, in order to explain the trends among the host materials.

Figure 2.10 Spin density distribution of triplet states from DFT calculations. Top: DDP
(left) and TPD (right); bottom: NDDP (left) and NNP (right)

To our best knowledge, it has not been possible to accurately predict V in
disordered materials using theoretical models. Therefore, some qualitative approaches
need to be adopted to get an estimate of V. In our case, two key questions need to be
addressed before we begin the analysis: where are triplet states located in the molecule
and to what extent do those states overlap intermolecularly? To answer the first question
the spin density distribution, which reflects the location of triplet states in a molecule,

48
was determined using DFT calculations (Figure 2.10). It can be seen that the spin density
is mainly localized on the central aromatic ring(s) in DDP, TDP and NDDP, whereas the
spin resides on the outer naphthyl group in NNP. The PL spectra of these molecules
provide information to the second question. As shown in Table 2.1 and Figure 2.3, the
emission peak of TDDP film is identical to that in solution, whereas a small red-shift is
observed for TPD and NDDP (Supporting Information). This suggests that the small
aromatic π-system at the core of TDDP is unable to form effective intermolecular π-π
interactions, while the extended conjugated biphenyl and naphthyl systems of TPD and
NDDP lead to weak, but observable interactions. Additionally, the t-butyl groups in
TDDP provide extra steric hindrance to further minimize the effects of π-stacking. In
contrast, the fluorescent emission maximum of NNP undergoes a shift from 393 nm in
solution to 448 nm when measured in a thin film. The large red-shift from solution to
film is characteristic for π-π interactions between aromatic moieties caused mainly by
intermolecular overlap of π-orbitals in the conjugated system. Furthermore, this
noncovalent interaction is stronger when the amount of overlap is greater. It implies that
the flat, electron-rich naphthyl groups in NNP molecules are likely responsible for this
type of π-interaction during aggregation in the solid state. The naphthyl groups at
periphery should have little difficulty to encounter adjacent peers and undergo π-stacking
of naphthyl groups. As a result, the degree of π-π stacking in the film should show a
trend TDDP < TPD ≈ NDDP < NNP. Since the π overlap provides pathways, the value
of coupling term, V, among these four molecules should generally follow the same trend.

49
It is important to recognize that the moieties which tend to stack in these molecules are
exactly where triplet states are located. With plenty of accessible naphthyl groups in
NNP, this well coupled matrix offers the most effective path for diffusion of triplet
excitons.

2.4 Chapter Conclusions
A simple and effective device structure is reported to study energy transfer and
triplet exciton diffusion in this paper. This device structure has been shown to have a
single, stable recombination zone and is able to eliminate charge trapping by dopants, a
process that is often difficult to exclude in studies on triplet exciton diffusion. A
deliberate mismatch between HOMO energy of host/dopant and good hole mobility of
triarylamines are key to the success. Good overlap of π-system MOs is crucial to
extending the triplet exciton diffusion length, with naphthyl groups providing more π-
overlap in solid state than either phenyl or biphenyl groups. Despite having a larger
reorganization energy, NNP displays longest triplet exciton diffusion length as a host,
which leads to the best device performance. The location of the naphthyl groups also has
an effect on triplet exciton diffusion depending on specific molecular structure. NNP
demonstrates that exposing naphthyl groups to the periphery of the molecule extends the
diffusion length of triplet excitons. Bulky alky groups, such as t-butyl, on peripheral
aromatic rings inhibit the rate of exciton hopping by increasing separation between
adjacent molecules.

50
2.5 Experimental section
Synthesis The tetraaryl-p-phenylenediamine derivatives were synthesized by
Buchwald-Hartwig coupling using a catalyst combination of Pd(OAc)
2
and P(t-Bu)
3
.

37

Mass spectra were recorded on an HP 5973 mass spectrometer using electron ionization
70eV. Elemental analysis was carried out by Microanalysis Laboratory at University of
Illinois, Urbana-Champaign. NMR spectra were measured on a Bruker AC 250 MHz
spectrometer. The glass transition temperatures (T
g
) and melting points were measured
using a TA Instruments 910 differential scanning calorimeter (DSC). For T
g

measurements, the samples were initially heated up to 400
o
C at a rate of 10
o
C per minute,
then cooled using the quench cooling accessory provided with the instrument. After that,
the same samples were heated up at the rate of 5
o
C several times followed by natural
cooling after each heating cycle. Coumarin 6 (C6) and BCP were purchased from
Aldrich Chemical Co., PQIr
38
was obtained from Universal Display Corp. and mCP was
synthesized according to the published literature procedure.
39
All other starting materials
and solvents were purchased from commercial sources and used without further
purification.
Method A: A three-neck round-bottomed flask was charged with 1,4-
dibromobenzene or 2,6-dibromonaphthalene (1 equiv, 5mmol), NaO
t
Bu (2.5 equiv),
Pd(OAc)
2
(0.05 equiv) and P(t-Bu)
3
(0.15 equiv). Xylene was added such that the
concentration of 1,4-dibromobenzene or 2,6-dibromonaphthalene was 0.7 mol l
-1
. The
chosen diarylamines (2.5 equiv) were then added against a stream of nitrogen and the

51
mixture was refluxed overnight. After the reaction was stopped and cooled to room
termperature, the crude products were either precipitated with hexane or obtained by
removing solvent under reduced pressure and then chromatographed on a column of
silica gel (hexane/dichloromethane—5:1) to give the final product. Before
characterization and deposition, all products were gradient sublimed using a three-zone
furnace at the pressure of 10
-6
Torr. The temperature differences between neighboring
zones are 10
o
C and 60
o
C, e.g. 180/170/120°C for the purification of DDP. The yield was
calculated based on the product collected after sublimation and relative to limiting
reagent, bromides.
Method B: Same with Method A except the flask was charged with N,N′-
Diphenyl-p-phenylenediamine, which is limiting reagent in this case, base and catalysts
first and added with 1-Bromo-4-tert-butylbenzene later.
1,4-Bis(diphenylamino)benzene (DDP). DDP was synthesized using method A
from 1,4-dibromobenzene and diphenylamine, sublimed at 180
o
C with 68% yield. Anal.
calcd.: C 87.35; H 5.86; N 6.79; found: C 87.33; H 5.68; N 6.92.
1
H NMR (C
6
D
6
δ): 6.81
(t, 4H), 6.94 (s, 4H), 7.03 (t, 8H), 7.11 (d, 8H); MS m/z 412; T
g
: N/A; T
m
: 201
o
C.
1,4-Bis(phenyl-m-tolylamino)benzene (TTP). TTP was synthesized using method
A from 1,4-dibromobenzene and 3-methyl-N-phenylaniline, sublimed at 185
o
C with 56%
yield. Anal. calcd.: C 87.24; H 6.41; N 6.36; found: C 87.19; H 6.31; N 6.49.
1
H NMR
(C
6
D
6
δ): 1.97 (s, 6H), 6.69 (t, 2H), 6.81 (t, 2H), 6.98-7.07 (m, 14H), 7.17 (m, 4H); MS
m/z 440; T
g
: 36
o
C; T
m
: 174
o
C.

52
1,4-Bis(phenyl-4-tert-butylphenylamino)benzene (TTTP). TTTP was synthesized
using method B from 1,4-dibromobenzene and 4-tert-butyl-N-phenylaniline, sublimed at
215
o
C with 76% yield. Anal. calcd.: C 86.98; H 7.68; N 5.34; found: C 87.01; H 7.86; N
5.51.
1
H NMR (C
6
D
6
δ): 1.19 (s, 18H), 6.80 (t, 2H), 6.99 (s, 4H), 7.03-7.06 (d, 4H),
7.13-7.17 (m, 12H); MS m/z 524; T
g
: 72
o
C; T
m
: N/A.
1,4-Bis(di-4-tert-butylphenylamino)benzene (TDDP). TDDP was synthesized
using method A from 1,4-dibromobenzene and bis(4-tert-butylphenyl)amine, sublimed at
285
o
C with 81% yield. Anal. calcd.: C 86.74; H 8.86; N 4.40; found: C 86.88; H 8.93; N
4.21.
1
H NMR (C
6
D
6
δ): 1.22 (s, 36H), 7.06 (s, 4H), 7.18-7.19 (d, 16H); MS m/z 636; T
g
:
N/A; T
m
: 317
o
C.
1,4-Bis(2-naphthylphenylamino)benzene (NNP). NNP was synthesized using
method A from 1,4-dibromobenzene and 2-naphthylphenylamine, sublimed at 235
o
C
with 90% yield. Anal. calcd.: C 89.03; H 5.51; N 5.46; found: C 89.08; H 5.29; N 5.60.
1
H NMR (C
6
D
6
δ): 6.85 (t, 2H) 7.03 (s, 4H), 7.07 (t, 4H) 7.14-7.19 (m, 8H), 7.31-7.33 (d,
4H), 7.48-7.50 (d, 2H), 7.54-7.57 (t, 4H); MS m/z 512; T
g
: 76
o
C; T
m
: 200
o
C.
2,6-Bis(diphenylamino)naphthalene (NDDP). NDDP was synthesized using
method A from diphenylamine and 2,6-dibromonaphthalene, sublimed at 275
o
C with
89% yield. Anal. calcd.: C 88.28; H 5.67; N 6.06; found: C 88.42; H 5.60; N 6.22.
1
H
NMR (C
6
D
6
δ): 6.84 (t, 4H), 7.05 (t, 8H), 7.11-7.14 (m, 12H), 7.43 (d, 2H); MS m/z 462;
T
g
: N/A; T
m
: 277
o
C.

53
Characterization methods Oxidation and reduction potentials were measured by
cyclic voltammetry (CV) using a EG&G Instruments model 283 potentiostat. CV scans
were recorded using a Ag wire as a pseudo-reference electrode at a scan rate of 100mV/s,
in dry and degassed ethylene carbonate/dimethylcarbonate (1:1) mixture with 0.1M
tetrabutylammonium hexafluorophosphate as electrolyte. Decamethylferrocene (-0.47V
vs. ferrocene) were used as an internal reference. Both photoluminescence (PL) and
electroluminescence (EL) emission spectra were obtained by a PTI QuantaMaster model
C-60SE spectrofluorometer, equipped with a 928 PMT detector and corrected for detector
response. 2-Methyltetrahydrofuran was used as solvent for solution PL. Films used for
obtaining PL spectra and quantum yield were spin-coated on quartz substrates in air at
room temperature using CH
2
Cl
2
as solvent. Quantum yield was measured by a
Hamamatsu PL Quantum Yield Measurement System (C9920-01). Förster radii were
calculated using PhotocamCAD HD1.1.
Theoretical calculations were done using Titan version 1.0.7 to estimate the
reorganization energies during the process of triplet exciton transfer. The reorganization
energy of the acceptor was obtained by subtracting the T
1
energy in its optimized
geometry from the T
1
energy in the geometry of the optimized S
0
state. The
reorganization energy of the donor was computed in a similar way by subtracting the S
0

energy in the optimized geometry of T
1
state from the S
0
energy in optimized ground
state.
40
The numbers provided in this paper are the sum of the reorganization energies for
a pair of donor and acceptor.

54
OLED fabrication and testing The OLEDs were grown on pre-cleaned indium
tin oxide (ITO) coated glass substrates with sheet resistance of 20 Ω /sq. All compounds
were purified using temperature gradient vacuum sublimation prior to deposition.
Organic layers were deposited by thermal evaporation from resistively heated tantalum
boats at a rate of around 2Ǻ/s, after which a shadow mask was placed on the substrate
and the cathode consisting 10Ǻ of LiF and 1200Ǻ of Al was deposited. The devices were
tested in air within 1h after fabrication. Light coming out from front surface was
collected by a UV-818 Si photocathode leading to a Keithley 2400 SourceMeter/2000
multimeter coupled to a Newport 1835-C optical meter. Device current-voltage and
light-intensity characteristics were measured using the LabVIEW program by National
Instruments.

2.6 Supporting information
350 400 450 500 550
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity / a.u.
Wavelength / nm
TPD Solution
TPD Film

Figure 2.S1 Solution and film emission spectra for TPD

55
350 400 450 500 550
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity / a.u.
Wavelength / nm
TTP Solution
TTP Film

Figure 2.S2 Solution and film emission spectra for TTP

350 400 450 500 550
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity / a.u.
Wavelength / nm
TTTP Solution
TTTP Film

Figure 2.S3 Solution and film emission spectra for TTTP

56
350 400 450 500 550
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity / a.u.
Wavelength / nm
NDDP Solution
NDDP Film

Figure 2.S4 Solution and film emission spectra for NDDP

350 400 450 500 550 600 650 700 750
0.0
5.0x10
11
1.0x10
12
1.5x10
12
2.0x10
12
2.5x10
12
Photon Counts
Wavelength / nm
P-I
P-II
P-III

Figure 2.S5 EL spectra of P type TPD devices

57
350 400 450 500 550 600 650 700 750
0.0
5.0x10
11
1.0x10
12
1.5x10
12
2.0x10
12
2.5x10
12
3.0x10
12
Photon Counts
Wavelength / nm
P-I
P-II
P-III

Figure 2.S6 EL spectra of P type NDDP devices

0 2 4 6 8 10 12
0
50
100
150
200
250
300
Voltage / V
Current Density / mA/cm
2
P-I
P-II
P-III
Undoped

Figure 2.S7 J-V characteristics of P type and undoped TDDP devices

58
0 2 4 6 8 10
0
50
100
150
200
Voltage / V
Current Density / mA/cm
2
P-I
P-II
P-III
Undoped

Figure 2.S8 J-V characteristics of P type and undoped TPD devices

0 2 4 6 8 10 12
0
100
200
300
400
500
Current Density / mA/cm
2
Voltage / V
P-I
P-II
P-III
Undoped

Figure 2.S9 J-V characteristics of P type and undoped NDDP devices

59
0 2 4 6 8 10
0
50
100
150
200
Voltage / V
Current Density / mA/cm
2
P-I
P-II
P-III
Undoped

Figure 2.S10 J-V characteristics of P type and undoped NNP devices

350 400 450 500 550 600 650 700 750
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity / a.u.
Wavelength / nm

Figure 2.S11 PL of 2% C6-doped TDDP film

60
0.1 1 10 100 1000
0.1
1
F-I
F-II
F-III
Undoped
Quantum Efficiency / %
Current Density / mA/cm
2

Figure 2.S12 EQEs of F type TDDP devices

0.1 1 10 100 1000
0.1
1
10
P-I
P-II
P-III
Undoped
Quantum Efficiency / %
Current Density / mA/cm
2

Figure 2.S13 EQEs of P type TDDP devices

61
2.7 Chapter 2 References

(1) Tang, C. W.; VanSlyke, S. A. Appl. Phys. Lett. 1987, 51, 913.
(2) Tang, C. W.; VanSlyke, S. A.; Chen, C. H. J. Appl. Phys. 1989, 65, 3610.
(3) Utsugi, K.; Takano, S. J. Electrochem. Soc. 1992, 139, 3610.
(4) Kido, J.; Shionoya, H.; Nagai, K. Appl. Phys. Lett. 1995, 67, 2281.
(5) Holmes, R. J.; D’Andrade, B. W.; Forrest, S. R.; Ren, X.; Li, J.; Thompson, M. E.
Appl. Phys. Lett. 2003, 83, 3818.
(6) Khan, A. L. T.; Sreearunothai, P.; Herz, L. M.; Banach, M. J.; Kohler, A. Phys.
Rev. B 2004, 69.
(7) Lacowicz, J. R. Principles of Fluorescence Spectroscopy, Springer, New York,
NY 2006.
(8) Turro, N. J. Modern Molecular Photochemistry, University Science Books,
Sausalito, CA 1991.
(9) Baldo, M. A.; O'Brien, D. F.; Thompson, M. E.; Forrest, S. R. Phys. Rev. B 1999,
60, 14422.
(10) Itaya, A.; Okamoto, K.; Kusabayashi, S. Bull. Chem. Soc. Jpn. 1976, 49, 2037.
(11) Dexter, D. L. J. Chem. Phys. 1953, 21, 836.
(12) Closs, G. L.; Johnson, M. D.; Miller, J. R.; Piotrowiak, P. J. Am. Chem. Soc. 1989,
111, 3751.
(13) Marcus, R. A. Angew. Chem. Int. Ed. 1993, 32, 1111.
(14) Baldo, M. A.; Forrest, S. R. Phys. Rev. B 2000, 62, 10958.
(15) Beierlein, T. A.; Ruhstaller, B.; Gundlach, D. J.; Riel, H.; Karg, S.; Rost, C.;
Riess, W. Synth. Met. 2003, 138, 213.
(16) Zhou, Y. C.; Ma, L. L.; Zhou, J.; Ding, X. M.; Hou, X. Y. Phys. Rev. B 2007, 75,
132202.
(17) D'Andrade, B. W.; Thompson, M. E.; Forrest, S. R. Adv. Mater. 2002, 14, 147.

62
(18) Sun, Y. R.; Giebink, N. C.; Kanno, H.; Ma, B. W.; Thompson, M. E.; Forrest, S.
R. Nature 2006, 440, 908.
(19) Giebink, N. C.; Sun, Y.; Forrest, S. R. Org. Electron. 2006, 7, 375.
(20) Schwartz, G.; Pfeiffer, M.; Reineke, S.; Walzer, K.; Leo, K. Adv. Mater. 2007, 19,
3672.
(21) Koene, B. E.; Loy, D. E.; Thompson, M. E. Chem. Mater. 1998, 10, 2235.
(22) Sakanoue, K.; Motoda, M.; Sugimoto, M.; Sakaki, S. J. Phys. Chem. A 1999, 103,
5551.
(23) D'Andrade, B. W.; Datta, S.; Forrest, S. R.; Djurovich, P. I.; Polikarpov, E.;
Thompson, M. E. Org. Electron. 2005, 6, 11.
(24) Djurovich, P. I.; Mayo, E. I.; Forrest, S. R.; Thompson, M. E. Org. Electron. 2009,
10, 515.
(25) Pope, C. E. S. M. Electronic Processes in Organic Crystals and Polymers, Oxford
University Press: New York, NY 1999.
(26) Wang, D.; Li, W.; Chu, B.; Su, Z.; Bi, D.; Zhang, D.; Zhu, J.; Yan, F.; Chen, Y.;
Tsuboi, T. Appl. Phys. Lett. 2008, 92, 3.
(27) Tsung, K. K.; So, S. K. Appl. Phys. Lett. 2008, 92, 103315.
(28) Uchida, M.; Adachi, C.; Koyama, T.; Taniguchi, Y. J. Appl. Phys. 1999, 86, 1680.
(29) Pschenitzka, F.; Sturm, J. C. Appl. Phys. Lett. 2001, 79, 4354.
(30) Jiang, X. Z.; Register, R. A.; Killeen, K. A.; Thompson, M. E.; Pschenitzka, F.;
Hebner, T. R.; Sturm, J. C. J. Appl. Phys. 2002, 91, 6717.
(31) Shi, J.; Tang, C. W. Appl. Phys. Lett. 2002, 80, 3201.
(32) Yang, C. L.; Tang, Z. K.; Ge, W. K.; Wang, J. N.; Zhang, Z. L.; Jian, X. Y. Appl.
Phys. Lett. 2003, 83, 1737.
(33) Rim, S. B.; Peumans, P. J. Appl. Phys. 2008, 103, 124515.
(34) Tse, S. C.; So, S. K.; Yeung, M. Y.; Lo, C. F.; Wen, S. W.; Chen, C. H. Chem.
Phys. Lett. 2006, 422, 354.

63
(35) Ohno, T.; Nozaki, K.; Nakamura, M.; Motojima, Y.; Tsushima, M.; Ikeda, N.
Inorg. Chem. 2007, 46, 8859.
(36) Dirac, P. A. M. Proc. R. Soc. Lond. Series A 1927, 114, 243.
(37) Yamamoto, T.; Nishiyama, M.; Koie, Y. Tetrahedron Lett. 1998, 39, 2367.
(38) Lamansky, S.; Djurovich, P. I.; Murphy, D.; Abdel-Razzaq, F.; Kwong, R.; Tsyba,
I.; Botz, M.; Mui, B.; Thompson, M. E. Inorg. Chem. 2001, 40, 1704.
(39) Adamovich, V.; Brooks, J.; Tamayo, A.; Alexander, A.; Djurovich, P. I.;
D’Andrade, B. W.; Adachi, C.; Forrest, S. R.; Thompson, M. E. New J. Chem. 2002, 26,
1171.
(40) Place, I.; Farran, A.; Deshayes, K.; Piotrowiak, P. J. Am. Chem. Soc. 1998, 120,
12626.

64
Chapter 3 Simplified structure for two-component and three-component
WOLED

3.1 Abstract
A simple and versatile WOLED device structure is introduced in this chapter.
The structure takes advantage of triplet exciton diffusion to achieve high efficiency and
balanced white emission. The singlet excitons emit directly from TPP molecules in
recombination zone, while the triplet excitons diffuse towards the phosphorescent
emitter(s) doped in the center of the host layer. The host molecule, TPP, and the
broadband emitter, IrDBQ, were developed to possess desirable properties for this
structure. The attractive properties of TPP include emitting blue, high triplet energy and
good triplet exciton conductivity, and the mixture of IrDBQ isomers exhibits wide yellow
emission which complements the TPP emission well for a two-component white light.
The triplet energy of TPP is high enough to be used with not only yellow emitters, but
also green ones, which allows both three-component and two-component device design to
be implemented into this WOLED structure. The two-component WOLED demonstrated
good efficiency with maximum of close to 10% and stable emission with a CIE
coordinate of (0.40, 0.41), although the breadth of the spectrum is not optimal. On the
other side, the three-component WOLED exhibited a more complete spectrum in visible
region with similar high efficiency, but the emission becomes voltage-dependent,

65
possibly due to the energy transfer between dopants and the drift of the recombination
zone.

3.2 Introduction
OLED has evolved into a commercialization-ready technology, thanks to the
breakthroughs in the past couple of decades. Tang et al. demonstrated first
heterojunction OLED,
1
and then the doping concept which opened the door to
tremendous amount of interest of practical implication.
2
Heavy-metal complexes
containing Pt and Ir were later found to be very efficient phosphorescent emitters, which
utilize 100% excitons and give far better device performance.
3-6
OLED has several
advantages over existing technology with its light weight, compatibility with flexible
substrates and superior power efficiency. Up to date, OLED has been widely used in
commercial products like displays and illumination devices, where devices emitting
white light are successfully implemented. It is easy to see that improving the
performance of WOLED is very important to the success of OLED in the future.
The most common method generating white light is RGB color model, in which
red, green and blue (component) light are superimposed at certain intensity ratio to create
a white light to human eyes. Therefore, the most intuitive way to construct WOLED is to
use three emitters making up the different components and covering the visible spectrum
evenly. The simplest structure of a three-component WOLED dopes all three emitters
into single emissive layer.
7-9
The use of phosphorescent emitters in this design generally

66
leads to highly efficient devices,
10
despite the readily energy transfer among dopants
which requires careful adjustment on doping percentages to achieve well-balanced white
emission. In the hope of improving the performance of this simple WOLED, attempts to
place dopants in different emissive layers were also reported. Both fluorescent
11-13
and
phosphorescent
14
emitters were successfully used in this stacked concept. However, the
device structure becomes more complicated,
15
because of the increase number of layers
and the manipulation of the recombination zone.
White light can also be obtained by combining only two component colors. One
of the approaches takes advantage of excimer emission from planar Platinum
complexes,
16,17
which is originated from the bounded excited states when complex-
complex distance is small.
18-22
Dependent on the complex structure and doping
percentage,
17
excimer emission, generally red-shifted and broad, and the emission from
monomeric species can be simultaneously generated from single dopant to create the
white emission. Unfortunately, the nature of excimer emission decides inherently that it
is not an efficient way to convert excitons into light.
With all the effort in developing WOLED, most structures contain more than five
different layers and up to ten different materials. The lack of simplicity in device
structure drives the production cost high and prevents WOLED from being competitive
and successful in the market. In this chapter, a simple, yet versatile WOLED structure is
introduced. The framework of the design can be applied to both three-component and
two-component WOLED. Borrowed the idea from excimer emission, an efficient Ir-

67
based broadband emitter was used in this simplified structure to complement high-energy
blue light and improve the efficiency.

Figure 3.1 Two-component and three-component WOLED using simplified structure

68
3.3 Results and Discussion
The structure discussed in Chapter 2 provides an interesting approach to fabricate
efficient WOLEDs with a simple architecture (Figure 3.1). The single recombination
zone is at the host/mCP interface, where singlet excitons emit through host molecules
instantaneously. Triplet excitons formed will diffuse into the host layer and captured by
the low-energy phosphors.
The host material has to meet three important criteria for this device structure to
work: emitting blue, a triplet energy higher than phosphorescent emitter(s) and
conducting triplet excitons well. Unfortunately, the phenylenediamine derivatives
studied in the Chapter 2 all fall short in these regards, because of the trade-off between
the high triplet energy and good triplet exciton conductivity. It is also challenging to
reduce the S
1
/T
1
gap so that the fluorescence of the host is blue while the triplet energy
can still be high enough for the phosphorescent emitters.

N N
TPP

Figure 3.2 Molecular structure of TPP and IrDBQ isomers

69
Triphenylene is selected as a substituent, because of its flat and conjugated π
system and high triplet energy.
23
The structure of the triphenyl derivative, TPP, is shown
in Figure 3.2. For two-component WOLED, a broad-band yellow emitter is pivotal to
make up a large part of the spectrum (500nm-700nm) so that a balanced white light can
be achieved. As already being studied, isomers of IrDBQ (Figure 3.2) demonstrate a
broad emission with peaks around 550nm to 600nm, well complementing the blue
emission from the host molecule (TPP).
24

300 400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
Emission: Solution at 77K
Emission: Powder at 77K
Excitation: Solution at 77K

Figure 3.3 Photophysics of TPP

3.3.1 Photophysics of TPP and IrDBQ isomers

The photophysical properties of TPP are shown in Figure 3.3. In dilute solution at
low temperature, TPP exhibits strong phosphorescence with well-resolved structure

70
peaking at 478nm (2.6eV) and 514nm. This suggests that the high triplet energy of
triphenylene is able to be retained after the coupling, matching the goal of the design.
The dominance of phosphorescence is mainly due to the small separation between S
1
and
T
1
state, which corresponds to only 0.3eV. One important feature of TPP is, unlike many
other derivatives in this family, its phosphorescence can still be clearly distinguished in
solid state at 77K, providing a rare opportunity to learn how triplet states are affected by
the change of the state. Based on the Figure 3.3, most characteristics of the emission
spectrum from solution remain intact. However, the whole spectrum is red-shifted, with
phosphorescence suffering more energy loss (0.12eV) from the aggregation effect,
compared to fluorescence (0.06eV). As a result, the S
1
/T
1
gap is enlarged and the
presence of fluorescence becomes larger. In terms of devices, TPP is an interesting
candidate for the host, thanks to its blue fluorescence and high triplet energy around
500nm.
The synthesis and characterization of fac- and mer- IrDBQ have been reported.
24

The photophysical properties of two isomers are compared in Figure 3.4. Although the
absorption spectra are almost identical for two molecules, they have very distinct
emission, in terms of the peak position. Fac-IrDBQ emission peaks at 540nm at RT,
while the maximum for mer-IrDBQ is at 600nm. Both emissions rise around 500nm, but
mer-IrDBQ has a broader emission tailing into deep red region, with a half-intensity
width (HIW) of 127nm, versus 97nm for the fac-IrDBQ. However, neither of the two
isomers is ideal as the broad-band yellow emitter for the two-component WOLED, since

71
the requirement on both peak position and width needs to be met. The dilemma can be
solved by using the mixtures of the isomers, thanks to their photo and thermal stability.
The isomerization was only found to happen when T>400
o
C under vacuum, allowing two
isomers to be evaporated at any pre-determined ratio without changing the contents.
300 400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
Soln. RT
Soln. 77K
Absorption RT

300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
Soln. RT
Soln. 77K
Absorption RT
Wavelength (nm)
Normalized Intensity (a.u.)

Figure 3.4 Emission and absorption spectrum of fac- (top) and mer- (bottom) IrDBQ

72
500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
fac - IrDBQ
mer - IrDBQ
95:5 fac:mer
65:35 fac:mer
25:75 fac:mer

Figure 3.5 The emission spectra of pure fac-, mer- IrDBQ and three mixtures

Figure 3.6 HPLC reports for the thermally-deposited films of the mixtures, 95:5 (left),
65:35 (middle) and 25:75 (right)

Figure 3.5 shows the progression of the emission spectra from different mixtures
of the isomers, 95:5, 65:35 and 25:75 (all fac- to mer- ratio). As the percentage of fac-
IrDBQ decreases, more mer-IrDBQ characteristics can be observed in the spectra of three
mixtures. The peak red-shifts from 540nm to 556nm, 570nm and 576nm and the HIW

73
increases to 106nm, 106nm and 110nm, respectively. The HPLC results were collected
from the films of the mixtures, which were thermally deposited under vacuum (Figure
3.6). No peak in the reports suggests the impurities were generated during the deposition
process and based on the integrated area under the peaks, the ratios remain the same after
the deposition.

3.3.2 Quantum yield and lifetime
Quantum yield (QY) and lifetime were measured for TPP samples and compiled
in Table 3.1. Comparing to other phenylenediamine derivatives studied in Chapter 2,
TPP has a higher QY in both solution and solid state. One interesting observation is that
the QY of the neat TPP film is the same with when it dispersed in PMMA. This indicates
the lack of stacking among TPP molecules in solid state, unexpected with the big, flat and
conjugated triphenylenyl groups at periphery. Irppy and IrDBQ doped TPP films were
also studied in order to see if they will be quenched by the host molecules. According to
the data, the QY of Irppy is significantly lower after doped into TPP,
25
while that of
IrDBQ remains close to the number measured without quenching reagent.
24
This is not
surprising considering the triplet energy of Irppy is almost identical with the one of TPP,
around 2.48eV. Without too much energy barrier to overcome, some of the excited Irppy
molecules could easily be quenched by the surrounding TPP. On the other hand, the
slightly lower energy of fac-IrDBQ seems to be enough to prevent the energy loss to the

74
matrix, which is very important for this host/guest combination to be successful in the
two-component WOLED device.

Table 3.1 QY and lifetime data of TPP samples

Solution
a
Films Doped Films
b

10% in PMMA Neat 10% Irppy in TPP 10% IrDBQ in TPP
QY 0.22 0.11 0.11 0.29 0.43
Life time (s) 5.6×10
-9
2.0×10
-9
3.6×10
-9
3.5×10
-7
1.6×10
-6

a
diluted in 2-MeTHF and degassed
b
spin-casted from toluene.

3.3.3 WOLED devices
3.3.3.1 Triplet exciton diffusion study on TPP
As having been discussed in Chapter 2, it is important to know TPP is solely
responsible for the hole conduction and deprives dopants of any role in exciton formation.
The redox potentials shown in Figure 3.7 suggest that TPP is much easier to be oxidized
(0.49 V versus decamethylferrocene or 0.02 V versus ferrocene), and possesses a
shallower HOMO than the dopants (C6 and PQIr). The peaks in cyclic voltammetry
measurement are weak due to the poor solubility of TPP in DMF, therefore differential
pulse voltammetry, another electrochemical experiment with greater sensitivity, was also
conducted.

75

Figure 3.7 Cyclic voltammetry and differential pulse voltammetry (inset) measurement
of TPP (peak at 0.3V is the internal reference, decamethylferrocene)

400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
C6-doped

Figure 3.8 EL of the fluorescence-based OLED using TPP as the host

Both fluorescence (C6-doped) and phosphorescence-based (PQIr-doped) control
devices (TPP 200Å/doped TPP 100Å/TPP 100Å/mCP 100Å/BCP 400Å) were then made
-2 0
-5.0x10
-5
0.0
5.0x10
-5
Current (A)
Potential (V)
-2 0 2
0.0
I Delta (A)
Potential (V)

76
to ensure that singlet and triplet excitons are harvested separately in space. According to
the EL spectra, TPP kept singlet energy from transferring to the doped layer (Figure 3.8),
while conducted triplet excitons well to exhibit strong PQIr emission.

350 400 450 500 550 600 650 700 750 800
0.0
0.5
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
TPP
NNP

0.01 0.1 1 10 100
1
10
Quantum Efficiency (%)
Current Density (mA/cm
2
)
NNP
TPP

Figure 3.9 The comparison of EL and EQE between TPP and NNP device

77
The performance of TPP in the phosphorescence-based device is evaluated by
comparing with NNP, which is the best among the group of molecules studied (Figure
3.9). The EL and EQE seem to lead to different conclusions to the comparison.
Apparently, TPP device has a higher EQE (maximum at 8%) than NNP device, with less
contribution from PQIr. This fact contradicts with the consistency between stronger PQIr
emission and higher EQE seen in other devices in Chapter 2. The uncharacteristic
correlation is worth a more careful look at what the EQE contributed by PQIr is. Based
on the integration of the EL spectrum, the ratio between phosphorescence and
fluorescence can be easily obtained (3.14 and 8.48 for TPP and NNP, respectively) and
further used to break down the total EQE. For example, the EQE of TPP and NNP
device is 6.48% and 5.48% at 1mA/cm
2
. Simple calculation yields 4.90% and 4.91% as
the phosphorescence part of the EQE for TPP and NNP, indicating that almost the same
amount of triplet excitons were sensitized by the PQIr in two devices. In other words, the
triplet excitons diffuse comparably well in both matrices, and the relatively strong
fluorescence in TPP device could be attributed to the higher QY of TPP.

3.3.3.2 Two-component and three-component WOLED
As it has been shown that TPP has good triplet exciton transport capability and
proper singlet and triplet energy to be used with phosphorescent emitters, two-component
(TPP (200 Ǻ)/8% IrDBQ:TPP (100 Ǻ)/TPP (100 Ǻ)/mCP (100 Ǻ)/BCP (400 Ǻ)) and
three-component (TPP (250 Ǻ)/2% PQIr:TPP (20 Ǻ)/5% Irppy:TPP (80 Ǻ)/TPP (100

78
Ǻ)/mCP (100 Ǻ)/BCP (400 Ǻ)) WOLED devices were made with the structure illustrated
in the Figure 3.1. A mixture of 65% fac- and 35% mer-IrDBQ was used to better
complement TPP emission and produce good white light in two-component WOLEDs.
350 400 450 500 550 600 650 700 750 800
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
6V
7V
8V
9V
10V

400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
6V
7V
8V
9V
10V

Figure 3.10 The EL of two-component (left) and three-component (right) WOLED

The EL spectra of two devices at various voltages are provided in Figure 3.10.
Two-component device demonstrates the excellent stable emission over a wide range of

79
the voltages. The CIE coordinate of the emission is (0.40, 0.41), excellent for lighting
purposes, while the CRI value is only 61. The reason for the unsatisfactory CRI value is
mainly due to the absence or weak emission at certain wavelengths such as 450 nm to
520 nm and 650 nm to 800 nm. Because of the inherent limitation of the two-component
WOLED, a more efficient blue fluorescent emitter (host) and broader yellow
phosphorescent emitter will be critical for the improvement in that category. On the
other hand, three-component WOLED exhibits a much more balanced but, as a trade-off,
very voltage-dependent EL spectrum. When the voltage goes up, namely more charges
are injected into the device, green emission is significantly suppressed, which is usually
attributed to the energy transfer between Irppy and PQIr.

0.01 0.1 1 10 100
1
10
Quantum Efficiency (%)
Current Density (mA/cm
2
)
Two-component
Three-component

Figure 3.11 EQE and J-V curves of the tow-component and three-component WOLED

80
Figure 3.11, continued
0 5 10
0
100
200
300
Current Density (mA/cm
2
)
Voltage (V)
Two-componet
Three-component

Efficiency wise, both devices score the maximum close to 10% at low current
density (Figure 3.11). However, the two-component device demonstrated a much smaller
roll-off within the meaningful range of current density, compared to the three-component
one. The EQE drops moderately from 8.1% to 6.0%, when current density increases
1000 times from 0.1 mA/cm
2
and 100 mA/cm
2
. The contrast is even greater between the
J-V characteristics of two devices (Figure 3.11). Although both devices have almost the
same thickness, the conductivity of the two-component device is as much as 15 times
higher than that of three-component one. The charge injection does not appear to be
related to the huge disparity, because two devices have the same turn-on voltage of 3.3V.
Consequently, the charge transport in the three-component device was somehow hindered.
The difficulty of moving charges will give rise to the shift of the recombination zone,

81
which rationalizes the aforementioned unstable emission spectrum. The possible
explanations to the poor conductivity include the extra interface between Irppy- and
PQIr-doped layer and the contamination brought in during the deposition process. To
summarize, there still exist challenges to improve the quality and stability of the white
light emission and overall EQE. Nevertheless, this simplified structure is versatile in
terms of materials selection and scheme design, at the same time, capable of delivering
the quality performance which is comparable to other WOLEDs.

3.4 Chapter Conclusions
Two-component and three-component WOLEDs with good EQE were
demonstrated using the simplified structure proposed in Chapter 2. The host material,
TPP, was designed to have a blue fluorescence and high triplet energy to accommodate
the phosphorescent emitters. The phosphorescence of TPP was found red-shifted from
solution to solid state at 77K. The aggregation seems to stabilize the triplet excited state
more, because the degree of the red-shift for phosphorescence is greater than that for
fluorescence. The mixture of fac- and mer-IrDBQ was selected to be the broad-band
yellow emitter for the two-component WOLED, thanks to their great stability under light
and heat. By combining two isomers, the emission peak and the width can be tuned to
create better white light. The two-component WOLED showed great voltage-
independent emission with a CIE coordinate of (0.40, 0.41), however, the CRI value is
only 61, which is mainly due to the inherent limitation of the two-component design.

82
This problem is alleviated in three-component devices, where two phosphorescent
emitters, Irppy and PQIr, were used to balance the emission. The drawback of this
approach is the energy transfer between two dopants causing the variation of the EL over
voltages and the inferior conductivity.

3.5 Experimental section
Synthesis BCP were purchased from Aldrich Chemical Co. Irppy and PQIr were
obtained from Universal Display Corp. Broad-band yellow emitter, IrDBQ, was
synthesized by Dr. Bossi, following the reported route.
24
mCP and 2-Bromotriphenylene,
the precursor for TPP, were synthesized according to the published procedures.
26
All
other starting materials and solvents were purchased from commercial sources and used
without further purification.
1,4-Bis(2-triphenylphenylamino)benzene (TPP). A three-neck round-bottomed
flask was charged with N,N′-Diphenyl-p-phenylenediamine (1 equiv, 5mmol), NaO
t
Bu
(2.5 equiv), Pd(OAc)
2
(0.05 equiv) and P(t-Bu)
3
(0.15 equiv), which then were dissolved
in xylene. The 2-Bromotriphenylene (2.5 equiv) were added against a stream of nitrogen
and the mixture was refluxed overnight. After the reaction was stopped and cooled to
room temperature, the crude products were either precipitated with hexane or obtained by
removing solvent under reduced pressure and then purified on a column of silica gel
(hexane/dichloromethane—10:1) to afford yellow powder as the final product. Before
characterization and deposition, all products were gradient sublimed using a three-zone

83
furnace at the pressure of 10
-6
Torr. TPP was sublimed at 340
o
C with 66% yield. Anal.
calcd.: C 90.98; H 5.09; N 3.93; found: C 90.93; H 4.68; N 4.09. MS m/z 712.
Characterization Methods Oxidation and reduction potentials were measured by
cyclic voltammetry and differential pulse voltammetry using a EG&G Instruments model
283 potentiostat. Scans were recorded using a Ag wire as a pseudo-reference electrode at
a scan rate of 100mV/s, in dry DMF with 0.1M tetrabutylammonium
hexafluorophosphate as electrolyte. Decamethylferrocene (-0.47V vs. ferrocene) were
used as an internal reference. Mass spectra were recorded on an HP 5973 mass
spectrometer using electron ionization 70eV. Elemental analysis was carried out by
Microanalysis Laboratory at University of Illinois, Urbana-Champaign. Both
photoluminescence (PL) and electroluminescence (EL) emission spectra were obtained
by a PTI QuantaMaster model C-60SE spectrofluorometer, equipped with a 928 PMT
detector and corrected for detector response. 2-Methyltetrahydrofuran was used as
solvent for solution PL. Films used for obtaining PL spectra and quantum yield were
spin-casted on quartz substrates in air at room temperature. Quantum yield was measured
by a Hamamatsu PL Quantum Yield Measurement System (C9920-01).
OLED fabrication and testing The OLEDs were grown on pre-cleaned indium
tin oxide (ITO) coated glass substrates with sheet resistance of 20 Ω /sq. All compounds
were purified using temperature gradient vacuum sublimation prior to deposition.
Organic layers were deposited by thermal evaporation from resistively heated tantalum
boats at a rate of around 2Ǻ/s, after which a shadow mask was placed on the substrate

84
and the cathode consisting 10Ǻ of LiF and 1200Ǻ of Al was deposited. The devices were
tested in air within 1h after fabrication. Light coming out from front surface was
collected by a UV-818 Si photocathode leading to a Keithley 2400 SourceMeter/2000
multimeter coupled to a Newport 1835-C optical meter. Device current-voltage and light-
intensity characteristics were measured using the LabVIEW program by National
Instruments.

85
3.6 Chapter 3 References

(1) Tang, C. W.; VanSlyke, S. A. Appl. Phys. Lett. 1987, 51, 913.
(2) Tang, C. W.; VanSlyke, S. A.; Chen, C. H. J. Appl. Phys. 1989, 65, 3610.
(3) Kwong, R. C.; Sibley, S.; Dubovoy, T.; Baldo, M.; Forrest, S. R.; Thompson, M.
E. Chem. Mater. 1999, 11, 3709.
(4) Adachi, C.; Baldo, M. A.; Forrest, S. R.; Thompson, M. E. Appl. Phys. Lett. 2000,
77, 904.
(5) Ikai, M.; Tokito, S.; Sakamoto, Y.; Suzuki, T.; Taga, Y. Appl. Phys. Lett. 2001, 79,
156.
(6) Baldo, M. A.; O'Brien, D. F.; You, Y.; Shoustikov, A.; Sibley, S.; Thompson, M.
E.; Forrest, S. R. Nature 1998, 395, 151.
(7) Kawamura, Y.; Yanagida, S.; Forrest, S. R. J. Appl. Phys. 2002, 92, 87.
(8) Tasch, S.; List, E. J. W.; Ekstrom, O.; Graupner, W.; Leising, G.; Schlichting, P.;
Rohr, U.; Geerts, Y.; Scherf, U.; Mullen, K. Appl. Phys. Lett. 1997, 71, 2883.
(9) Kido, J.; Shionoya, H.; Nagai, K. Appl. Phys. Lett. 1995, 67, 2281.
(10) D'Andrade, B. W.; Holmes, R. J.; Forrest, S. R. Adv. Mater. 2004, 16, 624.
(11) Jiang, X. Y.; Zhang, Z. L.; Zhao, W. M.; Zhu, W. Q.; Zhang, B. X.; Xu, S. H. J.
Phys. D-Appl. Phys. 2000, 33, 473.
(12) Huang, L.; Wang, K. Z.; Huang, C. H.; Gao, D. Q.; Jin, L. P. Synth. Met. 2002,
128, 241.
(13) Kido, J.; Kimura, M.; Nagai, K. Science 1995, 267, 1332.
(14) D'Andrade, B. W.; Thompson, M. E.; Forrest, S. R. Adv. Mater. 2002, 14, 147.
(15) Sun, Y. R.; Giebink, N. C.; Kanno, H.; Ma, B. W.; Thompson, M. E.; Forrest, S.
R. Nature 2006, 440, 908.
(16) D'Andrade, B. W.; Brooks, J.; Adamovich, V.; Thompson, M. E.; Forrest, S. R.
Adv. Mater. 2002, 14, 1032.

86
(17) Adamovich, V.; Brooks, J.; Tamayo, A.; Alexander, A.; Djurovich, P. I.;
D’Andrade, B. W.; Adachi, C. Forrest, S. R.; Thompson, M. E. New J. Chem. 2002, 26,
1171.
(18) Lu, W.; Mi, B. X.; Chan, M. C. W.; Hui, Z.; Zhu, N. Y.; Lee, S. T.; Che, C. M.
Chem. Commun. 2002, 3, 206.
(19) Lai, S. W.; Lam, H. W.; Lu, W.; Cheung, K. K.; Che, C. M. Organometallics
2002, 21, 226.
(20) Buchner, R.; Cunningham, C. T.; Field, J. S.; Haines, R. J.; McMillin, D. R.;
Summerton, G. C. J. Chem. Soc., Dalton Trans. 1999, 5, 711.
(21) Zheng, G. Y.; Rillema, D. P. Inorg. Chem. 1998, 37, 1392.
(22) Cheung, T. C.; Cheung, K. K.; Peng, S. M.; Che, C. M. J. Chem. Soc., Dalton
Trans. 1996, 8, 1645.
(23) Herkstroeter, W. G.; Hammond, G. S.; Lamola, A. A. J. Am. Chem. Soc. 1964, 86,
4537.
(24) Bossi, A.; Wu, C.; Djurovich, P. I.; Thompson, M. E. private communication,
manuscript submitted 2010.
(25) Lamansky, S.; Djurovich, P. I.; Murphy, D.; Abdel-Razzaq, F.; Lee, H.-F.;
Adachi, C.; Burrows, P. E.; Forrest, S. R.; Thompson, M. E. J. Am. Chem. Soc. 2001, 123,
4304.
(26) Barker, C. C.; Emmerson, R. G.; Periam, J. D. J. Am. Chem. Soc. 1955, 4, 4482.

87
Chapter 4 Study of ion-paired Iridium complexes (soft salts) and their
application in OLED

4.1 Abstract
Three Ir-based materials were synthesized through metathesis reaction between
halide and alkali metal salts of two cationic and three anionic Ir complexes, respectively.
The resulting “soft salt” complexes are composed of an organometallic cation and an
organometallic anion. The electrochemical and photophysical characterization of these
compounds is reported. The redox potentials of the soft salts are shown to be determined
by the lowest energy potentials of the two ions. Energy transfer between the ions in
solution is observed, and found to take place at diffusion controlled rates. Organic LEDs
were prepared with each of the three soft salts, using the simple structure of
anode/PVK/soft salt/BCP/cathode. The soft salts yielded maximal external quantum
efficiencies (EQE) ranging from 0.2% to 4.7%. The study suggests that the internal
energy alignment between two ions in the soft salts is responsible for the widely disparate
results. To achieve a high EQE it is critical to have the HOMO and LUMO values of one
of the ions fall between those of the other ion, i.e. one ion has both the lowest oxidation
potential and the least negative reduction potential.

88
4.2 Introduction
Since the groundbreaking report of heterojunction organic light-emitting diodes
(OLED) by Tang et al.,
1
an extensive amount of research interest has been drawn to this
field. A marked advance was made in OLED efficiencies when phosphorescent dopants
were introduced.
2-8
Among the most successful phosphorescent emitters are
cyclometalated Ir complexes, due to a number of important features, including strong
spin-orbit coupling,
9
which leads to phosphorescent lifetimes in the 1-10  sec range,
5

high phosphorescent efficiencies at room temperature
10
and good color tunability of the
emission energy, spanning the visible spectrum.
11,12
Both neutral and ionic Ir-based
complexes have these photophysical properties.
13,14
Neutral Ir-based complexes have
been used in OLED structures, consisting of distinct carrier transporting/blocking and
emitting layers.
15
Cationic Ir complexes have been used in light-emitting electrochemical
cells (LEC), which typically consist of a single active layer, responsible for both carrier
transport and light emission,
16-19
while the studies of anionic Ir complexes have been
mainly focused on their photophysics.
20-22

“Soft salt” is a term introduced to describe ionic materials that are composed of
only organometallic components, lacking halide, alkali metal or other ions commonly
present as counterions for these materials.
23
These salts are considered “soft”, because
their component ions are of significantly larger radii than simple ions so that the lattice
energies are expected to be lower and the ions are bonded mainly through van der Waals
force. Soft salts can be readily obtained in crystalline form,
24,25
and various cluster ions

89
and soft salts with different metals, such as Fe,
26
Cr,
27
Mo,
28
Os
29
and Ru
29
, have been
studied. De Cola, et al., have recently reported the crystal structure of an Ir-based soft
salt, closely related to one of the compounds discussed herein.
30

Mononuclear Ir-based soft salts have not been examined and are interesting
alternatives in OLEDs to the neutral materials that have been explored extensively in this
application. These soft salts can be easily synthesized through metathesis reactions and
show dual emission in solution, ambipolar charge conduction and good flexibility in
setting the HOMO and LUMO energy levels, due to the two independent functional
components in oppositely-charged ions. Herein, we present the synthesis,
characterization and OLED studies for three soft salts. We show that the alignment of the
energy levels within the soft salt can be adjusted by the ligand choice for each of the ions
and this energy alignment is a critical parameter controlling the performance of the
OLED.

4.3 Results and Discussion
The ionic Ir complexes were synthesized by refluxing bis-cyclometalated Ir(III)
dichloro-bridged dimer in the presence of an excess of an auxiliary ligand, (see equation
below). The net charge of the resulting Ir complex depends on the auxiliary ligand
selected. For example, when cyanide ions are present, two anions coordinate with Ir to
generate stable anionic Ir complexes. On the other hand, when neutral ligands, such as
diimines and isocyanides, are used, cationic complexes are obtained. By mixing two
oppositely charged Ir complexes in water, the soft salts were obtained through a simple

90
metathesis reaction, in moderate yield. The three soft salts studied in this paper are
composed of two different cations and three different anions. These five ions are
abbreviated based on their charge, namely C for cations and A for anions, followed by a
number to differentiate different ions. The structures of the ions and soft salts and their
corresponding acronyms are illustrated in Figure 4.1.

½ [(C^N)
2
Ir( -Cl)
2
Ir(C^N)
2
] + excess X
-
or L Cl
-
and (C^N)
2
IrX
2
-
or (C^N)
2
IrL
2
+

Figure 4.1 Structure of the soft salts studied

91
Table 4.1 Redox potentials, energy levels, lifetimes and quantum yield data of the soft
salts and their component ions under nitrogen

Redox (V) Energy (eV) Lifetime (μs)
c
QY (%) λ
max
(nm)
c

E
1/2
ox
/ E
1/2
re

HOMO / LUMO

Soln. Film
d

C1
a
1.21
b
/ -2.41
b
-6.3 / -1.9 36.7 38 3.7 458
C2
a
0.83 / -1.86 -5.8 / -2.6 0.43 21 16 586
A1
a
0.51 / -2.62 -5.3 / -1.7 4.0 70 4.8
e
472
A2
a
0.52 / -2.25 -5.3 / -2.1 3.4 78 3.2
e
572
A3
a
0.93 / -2.63 -5.9 / -1.7 4.1 70 7.8
e
448
C1A1 0.53 / -2.37 -5.3 / -2.0 12 / 3.8 87 7.0 470
C1A2 0.55 / -2.28 -5.4 / -2.1 8.6 / 3.1 74 13 456, 572
C2A3 0.82 / -1.87 -5.7 / -2.6 0.43 / 1.3 24 18 448, 586
a)
The counterion for C1 is OTf
-
, C2 is Cl
-
and Na
+
for all three anions.
b)
Ref. 21
c)

Measured in acetonitrile at the concentration below 10
-3
M
d)
Spin-coated from
acetonitrile solution at 3000 rpm for 40 sec followed by baking under vacuum at 90
o
C for
2 h. Measured under nitrogen
e)
The same as (d) except dissolved in acetonitrile / DMF
mixture solution (20:1).

4.3.1 Photophysics and quenching study
The emission data of the ions and soft salts is summarized in Table 4.1. The
photoluminescent (PL) spectra of the ion C2 and A3 in degassed acetonitrile solution are
presented in Figure 4.2 as an example. The emission spectrum of A3, which peaks at
around 450nm, is blue-shifted by the electron-withdrawing fluorines in the chelating
ligand with well-resolved vibronic structure. The structured emission is the result of a

92
triplet ligand-centered (
3
LC) transition on the cyclometalating ligands.
21
The rest of the
ions exhibit a broad and featureless spectrum, similar to that of C2 in Figure 4.2. Single-
ligand complexes C1, A1 and A2, are all expected to emit from metal-to-ligand charge-
transfer (MLCT) states, while the difference in C^N ligand is responsible for the shift in
peak position.
31,32
The mixed-ligand complex C2 showed a significantly red-shifted
emission compared to C1 and A1, because Irbipyridine (bpy) CT transition, rather than
Irtolylpyridine (tpy), is the lowest-energy one in this case.
33

200 300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
PL of C2
PL of A3
Absorption of C2
Absorption of A3
Normalized Intensity (a. u.)
Wavelength (nm)

Figure 4.2 Absorption of C2Cl and photoluminescence spectrum of C2Cl and A3Na

The PL spectra of the soft salts show interesting concentration dependence, a
result of one ion quenching the emission of the other. C2A3 serves as a good example
with the distinct spectrum from two ions, and three PL spectra measured at different
concentrations are compared in Figure 4.3. Based on the spectra, the ratio of two peaks

93
varies greatly depending on the solution concentration, indicating that the degree of
energy transfer between the ions is different. At the relatively low concentration of 10
-5
M and 350nm excitation, the emission is mainly from the anion, which is because the
quantum yield of the anion is much larger than that of the cation. This high-energy blue
emission diminishes, when the solution is concentrated, with cations serving as a
quencher of A3 emission, giving exclusive C2 emission at concentrations of 10
-3
M and
above.

400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0

Normalized Intensity (a. u.)
Wavelength (nm)
10
-3
M
10
-4
M
10
-5
M

Figure 4.3 Photoluminescence of C2A3 at different concentrations (all in degassed
acetonitrile, excitation wavelength = 350nm)

To study the energy transfer between two ions in C2A3, Stern-Volmer quenching
analysis was carried out. The lifetime of A3 in degassed acetonitrile solution with
various amounts of the quencher, C2, was recorded. The concentration of the A3 was
kept at 0.67μM in all samples and that of the quencher varied from 0 to 120μM. Based

94
on a bimolecular quenching model, the reciprocal of the lifetime of A3 is linearly
correlated to the concentration of the quencher ([Q]=[C2]), as observed for the C2 and
A3 in Figure 4.4. The quenching rate constant (K
q
) can be extracted by dividing the slope
of the fitted straight line by τ
0
(the lifetime with no quencher present). The calculation
yields a K
q
value of 1.71×10
10
M
-1
s
-1
, close to the diffusion limit in acetonitrile (2×10
10
M
-
1
s
-1
).
34
Quenching in of A3 emission occurs by either energy or electron transfer to C2.
35

A3 has a phosphorescence energy than C2, so energy transfer is a clear pathway for
quenching of A3 emission. While the A3 emission efficiency and lifetime are decreased
in C2A3 relative to A3 alone, the values for C3 are the same as those of C3 alone (see
below), suggesting that the energy transfer/quenching process is very efficient. Förster
energy transfer is unlikely to be an efficient pathway for energy transfer, because of the
poor overlap between the absorption of C2 and the emission of A3 (Figure 4.2). Electron
exchange (Dexter) energy transfer is the likely mechanism of energy transfer in this case.
0 2 4 6 8 10 12 14
2
4
6
8
10
1 / τ τ τ τ ( ( ( (x 10
5
s
-1
) ) ) )
[Q] (x 10
-5
M)
k
obs
= slope / τ τ τ τ
0
= 1.71 x 10
10
k
diff(ACN)
= 2 x 10
10

Figure 4.4 Stern-Volmer plot of the quenching study between C2 and A3 and the
numerical fitting of K
q

95
4.3.2 Quantum yield and lifetime
The quantum yields (QY) of the soft salts and their component ions were
measured in solution and as neat films (Table 4.1). In solution, the QYs increased
significantly after degassing, consistent with the efficient oxygen quenching of
phosphorescence from the complexes. The QYs of degassed solutions of C1A2 and
C2A3 match the values of the ion with lower energy in the pair, consistent with the
efficient energy transfer in soft salts. In films, the QYs are generally lower compared to
the values observed in degassed solutions, as expected due to self quenching. It is
interesting to see that the QY values are also sensitive to the size of the ligand and the
counter ion. For example, soft salts possess higher QYs than their component ions in
neat film, because the ions of the soft salts are markedly larger than halide or alkali metal
ions, leading to larger intermolecular spacing in the soft salts. Similarly, a bulkier ligand
can improve the QY by spacing the ions farther apart in the ionic solids, which explains
why C2Cl has a higher QY than C1Cl in film samples even though their QYs are quite
similar in solution.
The lifetime data of the individual ions measured in degassed acetonitrile solution
(Table 4.1) confirm the phosphorescent nature of the photoluminescence. The lifetime
measurement for the soft salts all yielded characteristic double-exponential curves,
representing different component ions. Because of the quenching through energy transfer,
the lifetimes of the component ions with higher triplet energies were all markedly
shortened, while the ones with lower energy were not affected.

96
4.3.3 HOMO and LUMO energy
The electrochemical properties of the soft salts and individual ions were examined
by cyclic voltammetry (CV). The measurements were performed in acetonitrile with
ferrocene as the internal reference. HOMO/LUMO levels were obtained from redox
potentials using previously published correlations, Table 4.1.
36,37
The energy levels are
also plotted in Figure 4.5 for comparison purposes. The HOMO/LUMO values of the
ions are affected by both the net electrical charge and the chelating ligand. For example,
C1A1 contains two ionic Ir complexes that have the same C^N chelating ligand, leading
to very similar emission and absorption energies, while their HOMO and LUMO energy
levels differ significantly. An anionic Ir complex (A1) is easier to be oxidized and harder
to be reduced than an analogous cationic complex, which translates into a higher HOMO
and LUMO than the cation (C1).

Figure 4.5 Energy levels of the materials used in the OLEDs

97
2.5 2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0 -2.5
C2
A3
Current
E (V vs. Ag/AgCl)
C2A3

Figure 4.6 Comparison of cyclic voltammetry between C2A3 and its component ions

The HOMO and the LUMO energy of the soft salts are determined by the lowest
energy potential of the pair of Ir complexes. The CV of C2A3 and its component ions
are given as an example in Figure 4.6. The first oxidation and reduction of C2A3
matches with the oxidation of A3 and the reduction of the C2, respectively. This
provides a simple solution to tuning the energy levels of the soft salts by changing the
chelating ligands of the individual ions. The idea can be seen by comparing C1A1 and
C1A2. C1A2 is obtained by replacing the tpy group of A1 with phenylquinoline (PQ)
groups, namely A2. In C1A2, the extended π-conjugation of the PQ ligand lowers the
reduction potential more than the positive charge raises it. As a result, C1A2 has both of
its lowest energy potentials on the anion, in contrast to C1A1 having the lowest energy
oxidation and reduction from HOMO of the anion and LUMO of the cation, respectively.

98
Using a similar strategy, the energy levels of C2A3 were tailored such that oxidation and
reduction both took place on the cation. The ability to adjust the energy levels
independently allows us to probe the role of carrier trapping on the cationic and anionic
component of the soft salt on OLED properties.

4.3.4 OLED studies
Soft salts were tested in both single layer LEC and hetero-structured OLED
devices. Typical LECs show delayed luminescence, which is a result of gradual charge
accumulation at electrode/organic interface, facilitating charge injection. When soft salts
were used in an LEC structure (anode/soft salt/cathode), characteristic LEC behavior was
not observed. There was no increase in current over a period of 30 minutes and the
devices failed to turn on under voltages ranging from 2.5V to 7V. This can be explained
by the absence of the small ions found in LECs which migrate in the applied electric field,
facilitating the charge injection in functional LECs. Charge injection in these soft-salt-
based LECs is difficult and imbalanced without the Ohmic contact created by migrating
mobile ions to establish a recombination zone in the middle of the soft salt layer. While
this simple device structure does not give efficient charge injection into the soft salt
materials, carrier transport in the soft salt layer is expected to be the same for both LEC
and OLED, namely relying on redox reactions of charged iridium complexes. Being a
mixture of cation and anion themselves, soft salts are ambipolar and expected to transport
both holes and electrons.

99
To lower the hole injection barriers in soft-salt-based devices, a PVK layer was
spin-coated on the ITO. This PVK layer also prevents agglomeration or crystallization of
the soft salt layer. The PVK/soft salt layers were capped with a thin film of BCP to
improve energy alignment between the cathode and the soft salts. The energy diagrams
for these three soft-salt-based OLEDs are shown in Figure 4.5. At the interface of PVK
and soft salt, the holes are injected by oxidizing the anions and hop between anionic
complexes inside the soft salt layer. The charge localization in soft salt layers is quite
different from that of the neutral materials typically used in OLEDs. In contrast to the
typical OLEDs, where the hole is present as a positive polaron, the hole in the soft salt
film forms a neutral species (oxidation of the anion) and the positive charge is
delocalized over the lattice, as it is represented by an excess of cations over what is
needed for the number of anions. The same is true for the electron, which corresponds to
an excess of anions over what is needed to compensate the number of cations in the
lattice, after the electron is injected. Conduction of charges through the soft salt is
expected to be higher than an analogous neutral lattice, due to the higher dielectric of the
highly-charged soft salt lattice. When a hole and an electron are localized on adjacent
molecules, the excess positive and negative lattice charge associated with each carrier are
neutralized and the two carriers are expected to be bound, just as positive and negative
polarons can be electrostatically attracted and bound in a neutral lattice.

100
400 450 500 550 600 650 700 750
0.0
0.2
0.4
0.6
0.8
1.0
C1A1
C1A2
C2A3
Normalized Intensity (a.u.)
Wavelength (nm)

Figure 4.7 EL of the devices made of three soft salts

The electroluminescence (EL) spectra of the devices are shown in Figure 4.7 and
the EQE and J-V characteristics of the devices are compared in Figure 4.8. The data
suggest that the relative position of the energy levels between two ions in the soft salt is
crucial to device performance. The component ions of C1A1 consist of the same
cyclometallating ligand, but different ancillary ligands, the latter imparting the charge to
the complex. The two ions have similar HOMO/LUMO gaps, but the molecular charges
markedly shift the HOMO and LUMO energies. Because the energy barriers between
two HOMOs and two LUMOs are both relatively high, it is difficult for carriers to be
exchanged between the reduced cation (the electron) and the oxidized anion (the hole),
which hinders exciton formation. As a result, there is no obvious preference for
recombination to take place on one ionic complex over the other. Although C1A1 has

101
the lowest PL QY in neat film among three soft salts, the 0.2 % EQE of the C1A1-based
device is far lower than would be expected based on the PL efficiency alone. This
suggests that poor matching of the HOMO/LUMO is the principal reason why C1A1
demonstrates the poorest performance among all three soft salts (Figure 4.8 and Table
4.2).
10 15 20 25
0
50
100
150
200
250
300
Current Density (mA/cm
2
)
Voltage (V)
C1A1
C1A2
C2A3

0.1 1 10 100 1000
1E-3
0.01
0.1
1
10
C1A1
C1A2
C2A3
Quantum Efficiency (%)
Current Density (mA/cm
2
)

Figure 4.8 J-V characteristics and EQE of the devices using different soft salts

102
In contrast to C1A1, C1A2 and C2A3 have a close energy match between one of
the frontier orbitals, close LUMO match for C1A3 and close HOMO match for C2A3.
Moreover, in these two soft salts, the energy levels of one ion are bracketed by those of
the other, suggesting that one of the ions may carry both the hole and electron. Thus,
there is a preference energetically to generate excitons on the ion with the smaller energy
gap. The anion of C1A2 and the cation of C2A3 serve as a trap for both holes and
electrons, which improves the efficiency of exciton formation. The advantage of this
energetic configuration is reflected in the markedly higher EQE values for C1A2 and
C2A3, i.e. 2.6 and 4.7%, respectively (Figure 4.8). Compared to the QY data of the films
in Table 1, the maximum EQE value of the devices made using C1A2 and C2A3 (Table
4.2) is comparable to the theoretical maximum (the internal efficiency is expected to be
3-5 times higher than EQE).
38

Table 4.2 Performance of the OLEDs made using different soft salts

λ
max, EL
(nm)
V
turn-on

(V)

L
max

(Cd / m
2
)

η
ext, max

(%)
C1A1 510 4 160 (24V) 0.2
C1A2 590 3 1971 (24V) 2.6
C2A3 586 2.5 7428 (21V) 4.7

Despite the good EQE, the device using C1A2 shows lower conductivity than the
one using C2A3 (Figure 4.8). A similar trend can be seen in the brightness and turn-on

103
voltage data (Table 4.2). The difference likely stems from the energy barrier for electron
injection. The LUMO of the cation of C1A2 (-1.94 eV) is 0.6 eV higher than that of the
cation in C2A3 (-2.54 eV), creating a larger barrier for electron injection from BCP. In
contrast, the effect on hole injection from the energy difference between HOMO of PVK
and the HOMO levels of two anions is comparable, based on the energy diagram (Figure
4.5).
Among the three soft salts, C2A3 offers the best HOMO/LUMO match-up with
hole and electron injection layer, and the energy levels between its component ions also
facilitate the charge recombination. The device data for C2A3-based OLEDs suggests
that soft salts have the potential to prepare OLEDs with properties comparable to those
made with neutral Ir phosphors.

4.4 Chapter Conclusions
Three Ir-based soft salts have been synthesized and characterized. Because each
soft salt is formed by a pair of ionic Ir complexes, through independently altering the
ligands, it is easy to tune the relative energetic properties, which have significant impact
on the photophysical, electrochemical and device performance. C2A3 shows the dual
emission in solution, with the ratio of two components in the spectrum dependent on the
soft salt concentration. Quenching studies show that the energy transfer process from the
blue to red emissive ion is at diffusion controlled limit. The internal match-up between
the energy levels of two ions is key to achieving good OLED performance. C2A3 gives

104
the best overall device performance with the maximal EQE of 4.7%, comparable to the
theoretical maximum based on the film PL efficiency. This result suggests that the soft-
salt-based devices have the potential to reach efficiencies comparable to neutral Ir-based
OLEDs. The key to achieving such high efficiencies is the preparation of soft salts with
thin film PL efficiencies in the 0.5-1.0 range, which are readily achieved for Ir-phosphor-
doped thin films.

4.5 Experimental section
General All the starting materials and solvents were purchased from commercial
sources and used without further purification. The
1
H and
13
C NMR spectra were
collected on a 400 MHz spectrometer at room temperature. Mass spectra were recorded
on an Applied Biosystems Voyager-DE STR mass spectrometer. IR data was obtained
from a Perkin Elmer Spectrum 2000 FT-IR spectrometer. Elemental analyses for the
three soft salts were carried out in an Heraeus Vario EL III elemental analyzer at NSC
Regional Advanced Instrument Center, National Taiwan University. Samples dried
under vacuum at room temperature yielded elemental analyses with higher-than-expected
percentage for carbon, consistent with solvent being trapped in the porous crystals.
30

Heating the samples under vacuum overnight at 100
o
C gave better CHN analyses (vide
infra).
General procedures for synthesis All experiments involving IrCl
3
·H
2
O or any
other Ir(III) species were carried out in an inert atmosphere. Cyclometalated Ir(III)

105
dichlorobridged dimmers were synthesized according to published procedures.
9,39,40
The
reaction conditions and purification procedures reported in this paper were not optimized.
[Ir(tpy)
2
(CN-t-Bu)
2
]Cl, C1: The characterization was previously published.
21

[Ir(tpy)
2
(
t
bpy)]Cl, C2: A mixture of iridium 2-(p-tolyl)pyridine dichloro-bridged
dimer (150 mg, 0.13 mmol) and 4,4’-di-tert-butyl-2,2'-bipyridine (79 mg, 0.29 mmol)
was dissolved in methanol (10 mL) and refluxed for 15 h. The solution was concentrated
and washed with hexane to afford pure product (185 mg, 84%) as a yellow solid.
1
H
NMR (DMSO-d
6
, 400 MHz) δ 8.86 (s, 2H), 8.20 (d, J = 8.0, 2H), 7.90 (td, J = 7.5, 1.4,
2H), 7.80 (d, J = 8.0, 2H), 7.76 (d, J = 5.9, 2H), 7.72 (dd, J = 5.9, 1.9, 2H), 7.55 (d, J =
5.8, 2H), 7.13 (td, J = 6.6, 1.4, 2H), 6.83 (dd, J = 7.9, 1.1 2H), 5.98 (s, 2H), 2.06 (s, 6H),
1.38 (s, 18H);
13
C NMR, (DMSO-d
6
, 100 MHz) δ 166.88, 163.43, 155.03, 151.36, 149.41,
148.59, 141.15, 139.52, 138.52, 131.61, 125.43, 124.94, 123.32, 123.10, 122.17, 119.60,
35.64, 29.95, 21.43; MS (MALDI) m/z 797 (M-Cl).
Na[Ir(tpy)
2
(CN)
2
] (A1): Iridium 2-(p-tolyl)pyridine dichloro-bridged dimer (500
mg, 0.44 mmol) was combined with sodium cyanide (261 mg, 5.32 mmol) in methanol
(50 mL) and refluxed with stirring for 15 h. The crude product was purified through
column chromatography on silica gel (DMF) to yield pure product (470 mg, 88%) as a
light yellow solid.
1
H NMR (DMSO-d
6
, 400 MHz) δ 9.48 (d, J = 6.0, 2H), 8.02 (d, J =
8.0, 2H), 7.86 (td, J = 8.0, 1.2, 2H), 7.55 (d, J = 8.0, 2H), 7.26 (td, J = 6.0, 1.2, 2H), 6.55
(dd, J = 8.0, 1.2, 2H), 5.89 (s, 2H), 1.92 (s, 6H);
13
C NMR, (DMSO-d
6
, 100 MHz) δ
167.84, 163.98, 153.07, 141.83, 137.00, 135.90, 131.55, 131.07, 123.52, 121.91, 120.95,

106
118.51, 21.41; IR: 2102, 2088 cm
-1
(terminal C≡N stretch); MS (MALDI) m/z 581 (M-
Na).
Na[Ir(pq)
2
(CN)
2
] (A2): Iridium phenylquinoline dichloro-bridged dimer (500 mg,
0.39 mmol) was combined with sodium cyanide (231 mg, 4.71 mmol) in methanol (50
mL) and refluxed with stirring for 15 h. The crude product was purified through column
chromatography on silica gel (DMF) to yield pure product (521 mg, 94%) as a light
yellow solid.
1
H NMR (DMSO-d
6
, 400 MHz) δ 10.13 (d, J = 8.0, 2H), 8.48 (d, J = 8.0,
2H), 8.26 (d, J = 8.0, 2H), 8.04 (dd, J = 8.0, 1.6, 2H), 7.80 (d, J = 8.0, 2H), 7.75 (td, J =
8.0, 1.6, 2H), 7.68 (td, J = 8.0, 1.6, 2H), 6.76 (td, J = 8.0, 1.2, 2H), 6.54 (td, J = 8.0, 1.2,
2H), 5.95 (dd, J = 8.0, 1.2, 2H);
13
C NMR, (DMSO-d
6
, 100 MHz) δ 171.46, 166.75,
148.75, 146.91, 138.25, 132.81, 131.52, 130.19, 129.80, 128.80, 127.85, 127.32, 126.17,
125.82, 120.00, 117.64; IR: 2107, 2087 cm
-1
(terminal C≡N stretch); MS (MALDI) m/z
653 (M-Na).
Na[Ir(dfppy)
2
(CN)
2
] (A3): Iridium 2-(2,4-Difluorophenyl)pyridine dichloro-
bridged dimer (300 mg, 0.25 mmol) was combined with sodium cyanide (145 mg, 2.96
mmol) in methanol (30 mL) and refluxed with stirring for 15 h. The crude product was
purified through column chromatography on silica gel (DMF) to yield pure product (270
mg, 84%) as a light yellow solid.
1
H NMR (DMSO-d
6
, 400 MHz) δ 9.55 (d, J = 6.0, 2H),
8.22 (d, J = 8.0, 2H), 8.03 (td, J = 8.0, 1.2, 2H), 7.44 (td, J = 6.0, 1.2, 2H), 6.63 (ddd, J =
12.8, 9.6, 2.4, 2H), 5.54 (dd, J = 8.0, 2.4, 2H);
13
C NMR, (DMSO-d
6
, 100 MHz) δ 170.2,
163.42, 154.97, 145.83, 139.76, 137.44, 132.17, 129.64, 124.72, 123.36, 121.62, 120.11,

107
59.49, 24.77, 20.70, 13.93; IR: 2114, 2106 cm
-1
(terminal C≡N stretch); MS (MALDI)
m/z 625 (M-Na).
C1A1: [Ir(tpy)
2
(CN)
2
]OTf (50 mg, 0.07 mmol) and Na[Ir(tpy)
2
(CN)
2
] (50 mg,
0.08 mmol) were added to water (10 mL). The reaction mixture was stirred for 1 h at RT
and then extracted with CH
2
Cl
2
. The combined organic extracts were dried over MgSO
4

and concentrated by rotary evaporation. The resulting solid was washed with ethyl ether
to afford soft salt 1 (49 mg, 65%) as a yellow solid.
1
H NMR (DMSO-d
6
, 400 MHz) δ
9.48 (d, J = 5.1, 2H), 8.98 (d, J = 5.8, 2H), 8.29 (d, J = 7.9, 2H), 8.17 (td, J = 7.8, 1.4,
2H), 8.00 (d, J = 7.8, 2H), 7.85 (td, J = 7.8, 1.6, 2H), 7.78 (d, J = 8.0, 2H), 7.55 (d, J =
8.0, 4H), 7.51 (td, J = 6.7, 1.5, 2H), 7.25 (td, J = 6.6, 1.4, 2H), 6.80 (dd, J = 7.9, 1.1, 2H),
6.55 (dd, J = 8.4, 1.8, 2H), 5.89 (s, 2H), 5.83 (s, 2H), 1.98 (s, 6H), 1.92 (s, 6H), 1.31 (s,
18H);
13
C NMR, (DMSO-d
6
, 100 MHz) δ 167.85, 166.60, 164.05, 153.30, 153.13, 153.10,
141.85, 141.28, 139.50, 139.23, 136.99, 135.88, 131.57, 131.01, 130.35, 124.82, 124.52,
124.13, 123.51, 121.91, 120.94, 120.43, 118.50, 94.48, 58.52, 29.51, 21.42; IR: 2187,
2161, 2101, 2093 cm
-1
(terminal C≡N stretch); MS (MALDI) m/z 695 (M
+
), 581 (M
-
).
Anal Calcd. for C
60
H
58
Ir
2
N
8
·2H
2
O: C, 54.94; H, 4.76; N, 8.54. Found: C, 54.62, H, 4.75,
N, 8.32.
C1A2: [Ir(tpy)
2
(CN)
2
]OTf (80 mg, 0.09 mmol) and Na[Ir(pq)
2
(CN)
2
] (80 mg,
0.11 mmol) were added to water (15 mL). The reaction mixture was stirred for 1 h at RT
and then extracted with CH
2
Cl
2
. The combined organic extracts were dried over MgSO
4

and concentrated by rotary evaporation. The resulting solid was washed with ethyl ether

108
to afford soft salt 2 (85 mg, 67%) as a yellow solid.
1
H NMR (DMSO-d
6
, 400 MHz) δ
10.13 (d, J = 8.8, 2H), 8.98 (d, J = 6.1, 2H), 8.46 (d, J = 8.6, 2H), 8.28 (d, J = 7.9, 2H),
8.24 (d, J = 9.0, 2H), 8.17 (td, J = 8.1, 1.5, 2H), 8.01 (dd, J = 7.9, 1.6, 2H), 7.78 (d, J =
8.0, 4H), 7.72 (td, J = 7.8, 1.7, 2H), 7.66 (td, J = 7.4, 1.1, 2H), 7.51 (td, J = 6.6, 1.5, 2H),
6.80 (dd, J = 7.9, 1.1, 2H), 6.73 (td, J = 7.2, 1.3, 2H), 6.51 (t, J = 7.3, 2H), 5.93 (d, J =
7.6, 2H), 5.84 (s, 2H), 1.98 (s, 6H), 1.30 (s, 18H);
13
C NMR, (DMSO-d
6
, 100 MHz) δ
171.42, 166.79, 166.57, 153.20, 153.02, 148.74, 146.86, 141.21, 139.44, 139.11, 138.15,
132.81, 131.50, 130.31, 130.08, 129.69, 128.71, 127.75, 127.27, 126.08, 125.72, 124.72,
124.38, 124.05, 120.35, 119.92, 117.54, 58.43, 29.47, 21.33; IR: 2183, 2159, 2105, 2088
cm
-1
(terminal C≡N stretch); MS (MALDI) m/z 695 (M
+
), 653 (M
-
). Anal Calcd. for
C
66
H
58
Ir
2
N
8
·2H
2
O: C, 57.29; H, 4.52; N, 8.10. Found: C, 57.29, H, 4.63, N, 8.04.
C2A3: [Ir(tpy)
2
(
t
bpy)]Cl (83 mg, 0.10 mmol) and Na[Ir(dfppy)
2
(CN)
2
] (80 mg,
0.12 mmol) were added to water (15 mL). The reaction mixture was stirred for 1 h at RT
and then extracted with CH
2
Cl
2
. The combined organic extracts were dried over MgSO
4

and concentrated by rotary evaporation. The resulting solid was washed with ethyl ether
to afford soft salt 3 (90 mg, 64%) as a yellow solid.
1
H NMR (DMSO-d
6
, 400 MHz) δ
9.54 (dd, J = 4.8, 1.0, 2H), 8.85 (s, 2H), 8.19 (d, J = 7.5, 4H), 8.03 (td, J = 8.3, 1.3, 2H),
7.90 (td, J = 8.2, 1.5, 2H), 7.80 (d, J = 8.0, 2H), 7.76 (d, J = 5.9, 2H), 7.71 (dd, J = 5.9,
1.9, 2H), 7.55 (d, J = 5.8, 2H), 7.44 (td, J = 6.6, 1.4, 2H), 7.12 (td, J = 6.7, 1.4, 2H), 6.84
(dd, J = 8.3, 1.3, 2H), 6.60 (ddd, J = 13.0, 9.4, 2.5, 2H), 5.98 (s, 2H), 5.52 (dd, J = 8.3,
2.5, 2H), 2.06 (s, 6H), 1.38 (s, 18H);
13
C NMR, (DMSO-d
6
, 100 MHz) δ 166.88, 163.43,

109
155.03, 151.36, 149.41, 148.59, 141.15, 139.52, 138.52, 131.61, 125.43, 124.94, 123.32,
123.10, 122.17, 119.60, 35.64, 29.95, 21.43; IR: 2113, 2106 cm
-1
(terminal C≡N stretch);
MS (MALDI) m/z 797 (M
+
), 625 (M
-
). Anal Calcd. for C
66
H
56
F
4
Ir
2
N
8
·H
2
O: C, 55.06; H,
4.06; N, 7.78. Found: C, 55.06, H, 4.57, N, 7.65.
Characterization methods Oxidation and reduction potentials were measured
by cyclic voltammetry (CV). CV scans were recorded at a scan rate of 100mV/s in dry
and degassed acetonitrile with 0.1M tetrabutylammonium hexafluorophosphate as
electrolyte. Ferrocene/ferrocenium (Cp
2
Fe/Cp
2
Fe
+
) redox couple was used as an internal
reference. A Pt wire and a glassy carbon rod were used as the counter and the working
electrode, respectively. An Ag wire was also used as a pseudoreference electrode.
The quenching study applied to determine the bimolecular quenching rate
constants is calculated according to Equation 4.1.
τ
0
/ τ = 1 + K
q
τ
0
[Q] 4.1
where τ and τ
0
are the excited state lifetime with and without the quencher, K
q
is the
experimental quenching rate constant and [Q] is the molar concentration of the quencher.
All the sample solutions for the quenching study had the same concentration for the
emitter, A3, at 0.67μM. The concentration of the quencher, C2, ranged from 0 to 120μM.
Lifetime measurements were performed on an IBH lifetime system after all the solution
samples were bubble-degassed with nitrogen for 5 minutes and excited at 380nm.
Both photoluminescence (PL) and electroluminescence (EL) emission spectra
were obtained by a PTI QuantaMaster model C-60SE spectrofluorometer, equipped with

110
a 928 PMT detector and corrected for detector response. Acetonitrile was used as solvent
for solution PL. Films used for obtaining PL spectra and quantum yield were spin-coated
on quartz substrates in air at room temperature from acetonitrile solutions. Quantum
yield was measured by a Hamamatsu PL Quantum Yield Measurement System (C9920-
01).
Device (OLED) fabrication and testing The OLEDs were grown on pre-
cleaned indium tin oxide (ITO) coated glass substrates with sheet resistance of 20 Ω /sq.
20mg PVK was dissolved in 1ml dichlorobenzene and filtered before being spin-coated
onto ITO at a rate of 3000rpm for 40s, followed by baking at 90
o
C for 1h under vacuum.
The soft salts were then spin-coated from the acetonitrile solution (30mg/ml) before the
baking under the same conditions. Thereafter, the substrates were transferred to the
vacuum chamber where the BCP layer was deposited by thermal evaporation from a
resistively heated tantalum boat at a rate of around 2Ǻ/s. A shadow mask was placed on
the substrates and the cathode consisting of 10Ǻ of LiF and 1200Ǻ of Al was
subsequently deposited. The devices were tested in air within 1h after fabrication. Light
coming out from the front surface was collected by a UV-818 Si photocathode leading to
a Keithley 2400 SourceMeter/2000 multimeter coupled to a Newport 1835-C optical
meter. Device current-voltage and light-intensity characteristics were measured using the
LabVIEW program by National Instruments.

111
4.6 Supporting Information
-2 0
-2.0x10
-4
-1.0x10
-4
0.0
1.0x10
-4
Current (A)
Potential (V)

Figure 4.S1 Cyclic voltammetry of A1
-2 0
-1.0x10
-4
0.0
1.0x10
-4
Current (A)
Potential (V)

Figure 4.S2 Cyclovoltammetry of A2

112
-2 0
-1.5x10
-4
-1.0x10
-4
-5.0x10
-5
0.0
5.0x10
-5
Current (A)
Potential (V)

Figure 4.S3 Cyclic voltammetry of C1A1
-2 0
-1.0x10
-4
-5.0x10
-5
0.0
5.0x10
-5
Current (A)
Potential (V)

Figure 4.S4 Cyclic voltammetry of C1A2

113
400 450 500 550 600 650
0.0
0.2
0.4
0.6
0.8
1.0
Intensity (a. u.)
Wavelength (nm)

Figure 4.S5 Photoluminescence of C1

400 450 500 550 600 650
0.0
0.2
0.4
0.6
0.8
1.0
Intensity (a. u.)
Wavelength (nm)

Figure 4.S6 Photoluminescence of A1

114
450 500 550 600 650 700
0.0
0.2
0.4
0.6
0.8
1.0
Intensity (a. u.)
Wavelength (nm)

Figure 4.S7 Photoluminescence of A3

400 450 500 550 600 650 700
0.0
0.2
0.4
0.6
0.8
1.0
Intensity (a. u.)
Wavelength (nm)

Figure 4.S8 Photoluminescence of C1A1 at 10
-6
M

115
450 500 550 600 650 700 750
0.0
0.2
0.4
0.6
0.8
1.0
Intensity (a. u.)
Wavelength (nm)

Figure 4.S9 Photoluminescence of C1A3 at 10
-6
M

Table 4.S1 Life times (s) of the neat films which were prepared according to the same
procedure described in the paper.

Before Baking After Baking
C1 3.33×10
-7
1.83×10
-6
1.80×10
-7
2.45×10
-6

C2 1.15×10
-7
4.26×10
-7
1.08×10
-7
4.03×10
-7

A1 1.85×10
-7
2.00×10
-6
4.09×10
-8
4.21×10
-7

A2 1.28×10
-7
7.61×10
-7
6.51×10
-8
2.14×10
-7

A3 7.12×10
-8
1.71×10
-7
6.52×10
-8
1.87×10
-7

C1A1 2.45×10
-8
2.21×10
-7
1.51×10
-7
7.54×10
-7

C1A2 2.70×10
-7
7.24×10
-7
1.89×10
-7
6.13×10
-7

C2A3 2.61×10
-7
5.86×10
-7
2.02×10
-7
5.66×10
-7

116
4.7 Chapter 4 References

(1) Tang, C. W.; VanSlyke, S. A. Appl. Phys. Lett. 1987, 51, 913.
(2) Kwong, R. C.; Sibley, S.; Dubovoy, T.; Baldo, M.; Forrest, S. R.; Thompson, M.
E. Chem. Mater. 1999, 11, 3709.
(3) Adachi, C.; Baldo, M. A.; Forrest, S. R.; Thompson, M. E. Appl. Phys. Lett. 2000,
77, 904.
(4) Baldo, M. A.; O'Brien, D. F.; You, Y.; Shoustikov, A.; Sibley, S.; Thompson, M.
E.; Forrest, S. R. Nature 1998, 395, 151.
(5) Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest, S. R.
Appl. Phys. Lett. 1999, 75, 4.
(6) Adachi, C.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. J. Appl. Phys. 2001, 90,
5048.
(7) Ikai, M.; Tokito, S.; Sakamoto, Y.; Suzuki, T.; Taga, Y. Appl. Phys. Lett. 2001, 79,
156.
(8) Markham, J. P. J.; Lo, S. C.; Magennis, S. W.; Burn, P. L.; Samuel, I. D. W. Appl.
Phys. Lett. 2002, 80, 2645.
(9) Sprouse, S.; King, K. A.; Spellane, P. J.; Watts, R. J. J. Am. Chem. Soc. 1984, 106,
6647.
(10) Sajoto, T.; Djurovich, P. I.; Tamayo, A. B.; Oxgaard, J.; Goddard, W. A.;
Thompson, M. E. J. Am. Chem. Soc. 2009, 131, 9813.
(11) Sajoto, T.; Djurovich, P. I.; Tamayo, A. B.; Yousufuddin, M.; Bau, R.; Thompson,
M. E.; Holmes, R. J.; Forrest, S. R. Inorg. Chem. 2005, 44, 7992.
(12) Lamansky, S.; Djurovich, P. I.; Murphy, D.; Abdel-Razzaq, F.; Lee, H.-F.;
Adachi, C.; Burrows, P. E.; Forrest, S. R.; Thompson, M. E. J. Am. Chem. Soc. 2001, 123,
4304.
(13) Lowry, M. S.; Hudson, W. R.; Pascal, R. A.; Bernhard, S. J. Am. Chem. Soc. 2004,
126, 14129.

117
(14) Lo, K. K. W.; Chan, J. S. W.; Lui, L. H.; Chung, C. K. Organometallics 2004, 23,
3108.
(15) Tang, C. W.; VanSlyke, S. A.; Chen, C. H. J. Appl. Phys. 1989, 65, 3610.
(16) Slinker, J. D.; Koh, C. Y.; Malliaras, G. G.; Lowry, M. S.; Bernhard, S. Appl.
Phys. Lett. 2005, 86, 173506.
(17) Tamayo, A. B.; Garon, S.; Sajoto, T.; Djurovich, P. I.; Tsyba, I. M.; Bau, R.;
Thompson, M. E. Inorg. Chem. 2005, 44, 8723.
(18) Graber, S.; Doyle, K.; Neuburger, M.; Housecroft, C. E.; Constable, E. C.; Costa,
R. D.; Orti, E.; Repetto, D.; Bolink, H. J. J. Am. Chem. Soc. 2008, 130, 14944.
(19) Su, H. C.; Chen, H. F.; Wu, C. C.; Wong, K. T. Chem.--Asian J. 2008, 3, 1922.
(20) Nazeeruddin, M. K.; Humphry-Baker, R.; Berner, D.; Rivier, S.; Zuppiroli, L.;
Graetzel, M. J. Am. Chem. Soc. 2003, 125, 8790.
(21) Li, J.; Djurovich, P. I.; Alleyne, B. D.; Yousufuddin, M.; Ho, N. N.; Thomas, J. C.;
Peters, J. C.; Bau, R.; Thompson, M. E. Inorg. Chem. 2005, 44, 1713.
(22) Di Censo, D.; Fantacci, S.; De Angelis, F.; Klein, C.; Evans, N.;
Kalyanasundaram, K.; Bolink, H. J.; Gratzel, M.; Nazeeruddin, M. K. Inorg. Chem. 2008,
47, 980.
(23) Green, M. L. H.; Hamnett, A.; Qin, J.; Baird, P.; Bandy, J. A.; Prout, K.;
Marseglia, E.; Obertelli, S. D. J. Chem. Soc.-Chem. Commun. 1987, 24, 1811.
(24) Basolo, F. Coord. Chem. Rev. 1968, 3, 213.
(25) Braga, D.; Grepioni, F. Organometallics 1992, 11, 711.
(26) Toan, T.; Teo, B. K.; Ferguson, J. A.; Meyer, T. J.; Dahl, L. F. J. Am. Chem. Soc.
1977, 99, 408.
(27) Pasynskii, A. A.; Eremenko, I. L.; Rakitin, Y. V.; Novotortsev, V. M.; Ellert, O.
G.; Kalinnikov, V. T.; Shklover, V. E.; Struchkov, Y. T.; Lindeman, S. V.; Kurbanov, T.
K.; Gasanov, G. S. J. Organomet. Chem. 1983, 248, 309.
(28) Bandy, J. A.; Davies, C. E.; Green, M. L. H.; Green, J. C.; Prout, K.; Rodgers, D.
P. S. J. Chem. Soc.-Chem. Commun. 1983, 23, 1395.

118
(29) McQueen, A. E. D.; Blake, A. J.; Stephenson, T. A.; Schroder, M.; Yellowlees, L.
J. J. Chem. Soc.-Chem. Commun. 1988, 23, 1533.
(30) Mauro, M; Schuermann, K. C.; Pretot, R.; Hafner, A.; Mercandelli, P.; Sironi, A.;
De Cola, L. Angew. Chem. Int. Ed. 2010, 49, 1222.
(31) Colombo, M. G.; Brunold, T. C.; Riedener, T.; Gudel, H. U.; Fortsch, M.; Burgi,
H. B. Inorg. Chem. 1994, 33, 545.
(32) Lamansky, S.; Djurovich, P. I.; Murphy, D.; Abdel-Razzaq, F.; Kwong, R.; Tsyba,
I.; Botz, M.; Mui, B.; R.;, B.; Thompson, M. E. Inorg. Chem. 2001, 40, 1704.
(33) Colombo, M. G.; Hauser, A.; Gudel, H. U. In Electronic and Vibronic Spectra of
Transition Metal Complexes I. Topics in Current Chemistry Series, 2
nd
ed.; Yersin, H.,
Ed.; Springer: Berlin, 1994; Vol. 171, pp. 143.
(34) Turro, N. J. Modern Molecular Photochemistry; University Science Books:
Sausalito, CA, 1991; pp. 314.
(35) Ma, B.; Djurovich, P. I.; Thompson, M. E. Coord. Chem. Rev. 2005, 249, 1501.
(36) D'Andrade, B. W.; Datta, S.; Forrest, S. R.; Djurovich, P.; Polikarpov, E.;
Thompson, M. E. Org. Electron. 2005, 6, 11.
(37) Djurovich, P. I.; Mayo, E. I.; Forrest, S. R.; Thompson, M. E. Org. Electron. 2009,
10, 515.
(38) Hung, L. S.; Chen, C. H. Mater. Sci. Eng., R 2002, 39, 143.
(39) Lohse, O.; Thevenin, P.; Waldvogel, E. Synlett 1999, 1, 45.
(40) Nonoyama, M. Bull. Chem. Soc. Jpn. 1974, 47, 767.

119
Chapter 5 Thermally cross-linkable electron transporting material for bottom-
emitting inverted OLEDs

5.1 Abstract
A thermally cross-linkable polymer is studied in this chapter as the electron-
transporting layer in bottom-emitting inverted OLEDs. The polymer layer can be
obtained by baking the spin-cast monomer layer at only 100
o
C under N
2
. The thermally-
initiated polymerization process is confirmed using various methods. Different cleaning
procedures for ITO were screened and solvent cleaning followed by UV-ozone was found
to provide the easiest electron injection at ITO/polymer interface. The polymer ETL was
then applied to both fluorescent and phosphorescent inverted OLEDs with simple
structures and compared with other common electron-transport materials. In general, the
EQE of the inverted OLEDs are lower than the numbers from regular OLEDs made with
the same structure, which can be attributed to the charge imbalance resulted from the
relatively poor electron injection. For the structure of ITO/ETL/Alq
3
/NPD/MoO
3
/Al, the
polymer ETL demonstrated better conductivity and EQE than the analoge, TPBI, while
the device using Alq
3
failed to turn on due to the lack of energy barrier to block the holes.
On the other hand, it is indicated that BCP has better hole-blocking capability than the
polymer and some recombination might happen inside the polymer layer without BCP
layer inserted.

120
5.2 Introduction
OLED has drawn tremendous amount of interest since the first heterojunction
device was reported.
1
The future holds tremendous opportunity for OLED, due to the
advantages in light weight, high contrast ratio, wide viewing angle and low power
consumption. Active-matrix OLED (AMOLED), a derivative of OLED technology, has
emerged as one of the main competitors for the market. AMOLED consists of pixels
deposited onto thin film transistors (TFTs) array to form a matrix of pixels, where every
pixel is controlled independently by the current flowing through. Since the majority of
the TFTs are n-type fabricated on amorphous silicon substrate, the conventional structure
of bottom-emitting OLED based on transparent ITO (indium tin oxide) /glass substrate
cannot be transplanted easily. Therefore, the most intuitive way circumventing this
problem would be to build the OLED with an inverted structure, where the bottom
cathode can be directly connected to a TFT.
Depending on from which side of the device the light is given out, there are two
types of inverted OLEDs (IOLED), namely top-emitting IOLED (from anode) and
bottom-emitting IOLED (from cathode). The former is composed of a stack of organic
layers deposited in exactly the reversed order of that in a conventional OLED structure
starting from the substrate. A reflective cathode on the substrate and a transparent anode
on the top are also needed. Top-emitting IOLED structure is favorable for its
compatibility with integrated circuits and high aperture ratio.
2
Many groups have opted
to use ITO as the top anode,
3-6
although the deposition of ITO is usually expansive and

121
destructive to organic materials.
7
Other candidates were also studied, such as Au
8
and
some metal oxides
9,10
. Unfortunately, they are often only semitransparent and induce the
microcavity effect, resulting in the shift of the EL spectra at different viewing angles.
Another challenge for top-emitting IOLEDs is to find a suitable bottom cathode, because
one of the most reliable cathodes, LiF/Al, will no longer work effectively in this
structure.
11
Intensive research has been done to alleviate this problem. Highly reactive
low-work-function metals, such as Mg, Ca and Li, were tried to deposited directly as the
cathode layer,
3,12,13
but the method varies greatly on the processing and not practical for
the industrial fabrication. Doping ETLs with Li or Cs has also been tried in hope to
realize the Ohmic injection by band bending at interfaces.
14,15
However, such alkali-
doping undermines the device stability and should be avoided, because of the diffusion
problem caused by metal dopants.
3,16
Another method, ultrathin interfacial electron
injection layer (EIL), seems to be the most effective and reliable alternative so far.
MgO,
17
MgAg-Alq
3
,
12
Alq
3
-LiF-Al trilayer,
10
aminoalkyl-substituted polfluorene
copolymers,
8
pentacene,
18
and PbO
19
are among many EIL candidates investigated. Two
models have been proposed to explain the enhanced electron injection by EILs, tunneling
probability enhanced model
4
and chemical reaction model
20
. The former claims that the
interfacial dipoles generated by interfacial EILs drop the voltage across the layer and
result in realigning the Fermi level of the cathode and the LUMO of the ETL, which
essentially reduces the electron injection barrier. The second model applies to EILs

122
containing alkali atoms whose high reactivity and low work functions help to lower the
energy barrier at the interface.
The bottom-emitting IOLED has the same sequence of organic layer with the top-
emitting version, only differing in that the light is directed out from the cathode by the
reflective anode. Efficient electron injection is also an important issue here, because
there are only limited number of materials were found suitable for the cathode. ITO is so
far the most commonly used cathode for the bottom-emitting IOLEDs, owing to its great
transparency and conductivity. Unfortunately, the high work function of ITO creates a
considerable injection barrier for electrons, requiring additional treatment in order to
work. For example, metals with low work function, such as Mg, were deposited as an
ultrathin interfacial layer between ITO and EILs.
13
Cs-containing compounds were found
liberating Cs while depositing and have been used to form the Cs interfacial layer in
situ.
14,21
Instead of neat ETL layer, Li and Cs doped ETLs were also introduced onto
ITO.
13,15,22
The drawbacks of such methods include the decreased transparency of the
cathode, difficulties in handling and the short operational time. Metal oxides are another
interesting group of materials with good charge injection property and high
transparency.
23,24

The success of applying polymer in IOLED is scarce in general. To my best
knowledge, there has been no record on using polymer as the ETL in the bottom-emitting
IOLED. The major challenge might come from the processing, because most of EIL
materials are soluble in common organic solvents and the dissolution problem usually

123
leads to poor electron injection. Several strategies have been developed to overcome the
dissolution issues. The simplest approach is to use materials soluble in orthogonal
solvent systems for each layer.
25
Other methods involving a liquid buffer layer
26
and
self-assembled layer
27
have also been introduced to prepare multilayer structures. Using
cross-linkable materials perhaps is the most elegant strategy among them all. Materials
with cross-linking unit can be either spin-cast or ink jet printed and then transformed into
insoluble films in situ by light or heat treatment. This provides an ideal solution to avoid
tedious synthesis and purification that are often encountered for polymers. Higher purity
layers obtained from this approach also result in better reproducibility of device results.
Moreover, it provides better morphological stability
28
and reduced crystallization
29
which
are the main problems frequently occurred in small-molecule- based devices.
In this chapter, an effort is made to introduce thermally cross-linkable polymer to
the bottom-emitting IOLED as the ETL, with no ultrathin interfacial EIL or doped layer
needed. This study is of practical significance trying to bridge the gap between
potentially simple and inexpensive fabrication technique with industrial design concept.
The monomer, namely cross-linking unit, is easy to synthesize and polymerize,
demonstrating great promise to compete with traditional small molecule ETL which
requires high vacuum and temperature to deposit.

124
5.3 Results and discussion
1,3,5-tris(1-phenylbenzo[d]imidazole-2-yl)benzene (TPBI, Figure 5.1), a trimer of
N-arylbenzimidazoles, is a wide band gap material demonstrating excellent electron
transporting and hole blocking capability in OLEDs.
30-32
Besides the deep HOMO
(around -6.5eV) and LUMO (-2.8eV), it also shows high T
g
around 120
o
C, thanks to the
“star-shaped” geometry.
33
The modification of TPBI has been demonstrated with success
by some research reporting the polymer
34-36
and dendritic
37-39
analogues to TPBI with
better physical properties and device performance.

N
N
N
N
N
N

Figure 5.1 Structure of TPBI

Styrene is known to undergo rapid polymerization to form polystyrene by heating
without any initiator.
40-42
From the practical point of view, the styryl group is a
promising thermal curable group, because it can be polymerized at lower temperature and
is easy to synthesize. Combining the two facts, the functionalization of N-

125
arylbenzimidazoles with styryl groups should provide a simple solution to find a good
thermally cross-linkable material for ETL.
COOH
Br Br NH
2
N
H
N N
Br Br
+
B
O O +
N N
1
2

Scheme 5.1 Synthetic routes for the monomer

The proposed monomer (2) and the synthetic route was illustrated in Scheme 5.1.
The precursor (1) was prepared through the double dehydration between 3,5-
dibromorbenzoic acid and N-phenyle-o-phenylenediamine in modest yield, with the
presence of the triphenylphosphine oxide and trifluoromethansulfonic anhydride as the
extraction reagents.
43
Phosphoryl compounds and triflic anhydrides are known to form
diphosphonium salts,
44
which can activate carboxyl groups. The reagent used in this
study was formed as white precipitates at -78
o
C in C
2
H
4
Cl
2
and then directly used in situ,
considering its hygroscopic nature. 2 was then synthesized by reacting 1 with
vinylboronic acid pinacol ester under standard Suzuki coupling conditions.

5.3.1 Polymerization
Differential Scanning Calorimetry (DSC) was used to study the thermal properties
of the monomer and resulted cross-linked polymer (Figure 5.2). The sample was heated
through three cycles at a ramping rate of 10
o
C/minute. The first scan shows a broad

126
exothermic peak (T
max
around 80
o
C), which is usually accredited to the heating-induced
cross-linking of styryl groups. After the initial scan, the broad peak did not appear and
no clear glass transition temperature (T
g
) can be observed up to 300
o
C. It is worth
pointing out that the cross-linking below 100
o
C is much lower than both reported
temperatures to polymerize styryl groups in similar compounds
42,45
and the temperature
needed for creating solvent-resistant network (170
o
C). However, the decomposition
temperature (T
d
) of 450
o
C, according to the third scan, suggests that the resulting cross-
linked network is fairly robust.
0 100 200 300 400 500
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
10
o
C/min 1st run
10
o
C/min 2nd run
10
o
C/min 3rd run
Heat Flow (mW)
Temperature (
o
C)

Figure 5.2 DSC measurements starting with monomer

2 is very soluble in common organic solvents, such as dichloromethane, THF and
toluene and can be easily spin-coated from toluene solution onto various substrates.
Other measurements that compare film properties before and after the treatments were

127
also conducted to confirm the cross-linking temperature. The films studied in this
chapter were prepared by spin-casting the 20 mg/ml toluene solution of 2 at 3000 rpm for
40 seconds under ambient conditions, followed by serial treatments including baking at
120
o
C for 1 hour under N
2
and immersion in boiling toluene for 20 minutes before the
measurements. Mother solution was filtered through a syringe filter with 0.45 μm PTFE
membrane before spin-casting. The obtained films remained smooth and transparent
throughout the process.
1000
96
98
100
Before Baking
After Baking @100
o
C
After Baking @250
o
C
% T
Wavenumber (cm
-1
)

Figure 5.3 FT-IR spectra of the film before and after the thermal treatment

128
The FT-IR spectra of the sample were examined (Figure 5.3) to further verify the
polymerization process. The film of 2 was prepared by spin-casting on a KBr salt plate
as the substrate. As reported in literature, pendant vinyl groups in the compounds with
unsaturated rings generally have absorption peaks between 950-1000 cm
-1
, exemplified
by 1-vinyl-2-pyrrolidone (985 cm
-1
) and styrene (990 cm
-1
).
46
It is reassuring to see that
the peaks around 1000 cm
-1
were significantly reduced in intensity by the baking at
100
o
C for 30 minutes and completely eliminated after the further treatment at 250
o
C for
30 minutes.

Table 5.1 The thickness of the film before and after the treatments

Before Baking 1-hour Baking 2-hour Baking After immersion
Thickness (nm) 62.6 56.0 55.4 51.3

Solvent resistance was first investigated by recording the thickness change of the
film using ellipsometer and the data is shown in Table 5.1. 10% thickness loss was
observed, after the first one-hour baking, which can be attributed to the solvent driven off
during the treatment and the smoothening of the film surface. The thickness of the layer
maintained unchanged when second-stage thermal treatment was finished, suggesting that
only a short baking period is needed to attain the stable form of the film. To gauge the
formed cross-linked network, solvent rinsing was applied and impressively,
approximately 92% of the material on the substrate survived the processing. In this case,

129
the thermally-initiated radical polymerization occurs under easy conditions and results in
quality film with good solvent resistance, giving significant advantages to 2 in reducing
the fabrication time and cost, considering that 200
o
C is routinely needed for other
thermally cross-linked polymers to form.
47-50

200 400 600
0.0
0.5
1.0
1.5
Before Baking
After Baking
After Immersion
Intensity (a. u.)
Wavelength (nm)

Figure 5.4 Absorption Spectra of the film before and after the treatment

The same film was monitored via UV/vis spectroscopy and the spectra are
compared in Figure 5.4. The absorption of the film cuts off around 350 nm and is
transparent over the entire visible region. The little interference with the visible light
qualifies this material as a suitable bottom layer for the OLED without affecting the light
output of the devices. The spectrum retained its line shape throughout the treatment,
except for the part between 200 and 300 nm. This might be attributed to the degree of the
polymerization and the amount of the solvent left in the sample, because toluene has a

130
strong absorption band around 250 nm.
51
In addition, the intensity of the absorption
underwent noticeable drop after baking and solvent rinsing, for which the loss of the
material could be held responsible.

5.3.2 Bottom-emitting IOLEDs
With the polymerization conditions figured out, the cross-linked polymer formed
from 2 was applied to a simple bottom-emitting IOLED structure as the ETL,
ITO/polymer (200Å)/Alq
3
(200 Å)/NPD (400 Å)/MoO
3
(80 Å)/Al (1200 Å). The ITO
substrates were cleaned using different procedures, because it is well known that the
work function of ITO varies significantly over the treatments.
52
2 was spin cast at 3000
RPM for 20 seconds on the ITO surface immediately after the cleaning is finished,
followed by the baking under N
2
at 120
o
C for 1 hour, which transforms the film into a
cross-linked network. Alq
3
and NPD were then deposited as the EML and HTL,
respectively. Al was chosen to be the anode for its high reflectivity. A MoO
3
layer was
inserted between Al and NPD to mitigate the high injection barrier for holes, a result
attributed to the interfacial dipoles generated from the electron transfer at interface.
53
By
design, holes and electrons will be confined and recombine in the Alq
3
layer because of
the deep HOMO of the polymer and shallow LUMO of NPD.
Two major cleaning protocols, solvent and photoresist cleaning, were applied,
followed by an optional 10-minute UV-ozone. The solvent cleaning involves the
immersion of the ITO substrates in boiling tetrachloroethane, acetone and ethanol for 5

131
minutes each. Cleaning using photoresist requires a series of treatment including
covering of the ITO surface with photoresist by spin casting, baking the coated substrates
at 80
o
C in air and rinsing off the photoresist layer using acetone.
5 10 15 20
0
100
200
300
400
500
600
Current Density (mA/cm
2
)
Voltage (V)
Solvent
Solvent + UV Ozone
Photoresist
Photoresist + UV Ozone
0.1 1 10 100 1000
1E-3
0.01
0.1
1
Solvent
Solvent + UV Ozone
Photoresist
Photoresist + UV Ozone
Quantum Efficiency (%)
Current Density (mA/cm
2
)

Figure 5.5 Effect of different ITO cleaning methods on the J-V characteristics and EQE

132
Device performance of the four devices using different cleaning methods is
compared in Figure 5.5. Since the only difference among the devices is the ITO surface,
the J-V characteristics perfectly reflect the disparity of the electron injection barrier.
Based on Figure 5.5, solvent cleaned ITO surfaces demonstrated superior conductivity in
general, comparing to the substrates treated with photoresist. It consequently implies the
work function of ITO is relatively low after this cleaning method, which leads to a better
alignment with the LUMO of Alq
3
. Interestingly, UV-ozone treatment appears to have
opposite influence on the work function of ITO in these two cases. It has been found that
UV-ozone effectively removes the carbon contamination on ITO surface and increases
the work function,
54
which, for IOLED, undesirably enlarges the energy barrier at the
interface. While this is true for photoresist-cleaned ITO, the additional UV-ozone
exposure after the solvent cleaning unexpectedly improved the electron injection and
conductivity.
Devices with solvent-cleaned ITO substrate also excel in EQE with the maximum
of 0.5%, regardless if UV-ozone treatment was conducted or not. In contrast, a two-fold
improvement was obtained with the UV-ozone exposure after photoresist cleaning. It is
obvious after comparing in both categories, that solvent cleaning combined with UV-
ozone treatment offers the best contact with cross-linked polymer in terms of injection
barrier and device performance.
To gauge the effectiveness of cross-linked polymer as the ETL, control devices
were made with 200Å of the monomer analogue, TPBI and the tradition electron-

133
transporting materials, Alq
3
, respectively (Figure 5.6). Alq
3
device behaved as a typical
hole- or electron-only device, failing to turn on, but with high conductivity, which is the.
Since holes are known to be injected into Alq
3
layer from NPD and electron injection is
much harder than hole’s in IOLED, it is not surprising that holes reached the cathode
before electrons were able to be injected in this case, especially without a hole-blocking
layer in the device.
5 10 15
0
100
200
300
400
500
600
Polymer
TPBI
Alq
3
Current Density (mA/cm
2
)
Voltage (V)

0.1 1 10 100 1000
1E-6
1E-5
1E-4
1E-3
0.01
0.1
1
Polymer
TPBI
Alq
3
Quantum Efficiency (%)
Current Density (mA/cm
2
)

Figure 5.6 Comparison of cross-linked polymer with TPBI and Alq
3
as the ETL

134
Unlike Alq
3
, TPBI and the analogous polymer have a deeper HOMO, which
provides a big energy barrier for holes to overcome. Before the bias is big enough for
that to happen, electrons are able to be injected and meet the trapped holes inside Alq
3

layer. As a result, both devices showed green electroluminescence from Alq
3
. On the
other hand, the similarity in structure did not translate into the same device performance.
Cross-linked polymer indisputably exhibited superior electron injection and transporting
capability, relative to TPBI device. In addition, the higher EQE implies that the polymer
is capable of blocking holes better, which promotes the chance of recombination,
therefore improves the efficiency.
400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)

Figure 5.7 The EL spectrum of three phosphorescent IOLED

To further demonstrate the potential of the cross-linking polymer, phosphorescent
IOLEDs were fabricated with the structure of ITO/ETL/8% Irppy in CBP (200 Å)/NPD

135
(400 Å)/MoO
3
(80 Å)/Al (1200 Å), where ETL are 200 Å polymer, 150 Å polymer +
100Å BCP and 200 Å BCP, respectively. Irppy is an efficient phosphorescent emitter,
while CBP is an ambipolar conductive host with big HOMO/LUMO gap. The widely-
acknowledged dopant/host combination as EML yields the EQE above 8% in regular
OLED with emission peak around 510.
55
The electroluminescence spectrum of the three
devices is solely consisted of the emission from Irppy, as shown in Figure 5.7. It is worth
mentioning that NPD emission is usually present in a regular NPD/Irppy:CBP/BCP
device, because of the electron leakage into NPD layer. In this study, the abscence of
NPD emission confirmed that the hole injection is far more efficient than electron
injection, keeping the recombination zone away from NPD/CBP interface.
The performance of the three devices, including EQE and J-V characteristics, is
compared in Figure 5.8. Based on the previous rationalization, the data suggests that
BCP is a better hole blocker than the cross-linking polymer, which is not surprising,
considering BCP has been widely applied as a hole-blocking layer in regular OLED.
Although various values have been assigned to the HOMO of BCP, it is well accepted
that the actual energy falls in the range between 6.5 – 6.7V,
56
slightly deeper than that of
TPBI. With additional 100 Å of BCP between polymer and CBP, the device suffered
significant conductivity drop comparing to polymer-only device. Since both
electrode/organic interfaces, namely charge injection barrier, are the same in these two
devices, the reasonable explanation is that the cross-linking polymer surrendered to the
increasing bias and allowed holes to inject before BCP would do. This is also in

136
agreement with the fact that the J-V characteristic of the polymer-BCP device resembles
the BCP device more as opposed to the polymer device.
5 10 15 20
0
100
200
300
400
500
Polymer
Polymer + BCP
BCP
Current Density (mA/cm
2
)
Voltage (V)

0.1 1 10 100 1000
1E-3
0.01
0.1
1
Quantum Efficiency (%)
Current Density (mA/cm
2
)
Polymer
Polymer + BCP
BCP

Figure 5.8 J-V characteristics and EQE of three phosphorescent IOLEDs with different
ETL

137
The EQE for the two BCP-containing devices benefits from the better hole-
blocking capability with the maximum reaching 3%, as opposed to the disappointing 1%
from the polymer device. The inferior EQE for the device without BCP implies that there
is a significant amount of the excitons were formed not only in the polymer layer, but
also too far from the doped CBP layer to be utilized. On the other hand, when holes were
more effectively confined within CBP layer by BCP, more excitons can be generated in
the proximity of Irppy molecules, which harvest and convert them into light. It is worth
pointing out that, in general, the EQE of the IOLEDs is lower than the regular device
with the same structure. The underachievement is mainly due to the poor charge balance
inside the IOLEDs, leaving the challenge of improving electron injection yet to be
addressed. Nevertheless, the study successfully demonstrated the application of the
thermally cross-linkable ETL in the bottom-emitting IOLED with simple structure and
easy processing conditions.

5.4 Chapter Conclusions
A monomer containing N-arylbenzimidazoles group has been synthesized and
shown undergoing polymerization at only 100
o
C. The thermally-initiated polymerization
process was confirmed by various methods, such as DSC, FT-IR, UV-Viz and
ellipsometry. The obtained polymer was applied to both fluorescent and phosphorescent
bottom-emitting IOLEDs as ETL with success. The study shows the device performance
is directly related to the cleaning protocols used on ITO and the hole-blocking capability

138
of the ETL. The poor electron injection forces the devices to rely on slowing down hole
movement to efficiently form exciton inside EML. Devices using BCP performed the
best among the electron-transporting materials investigated, while the cross-linked
polymer appears to have low electron injection barrier at the interface with ITO.

5.5 Experimental section
Synthesis All the starting materials and solvents were purchased from
commercial sources and used without further purification.
2-(3,5-dibromophenyl)-1-phenyl-1H-benzo[d]imidazole (1) To a solution of 5.56
g (20 mmol) of triphenylphosphine oxide in 30 ml of ethylene dichloride at 0
o
C was
added drop wise a solution of 1.57 ml (10 mmol) of triflic anhydride in 30 ml of ethylene
dichloride. After 15 minutes, white precipitate appeared and the mixture then was kept in
room temperature. To the extraction reagent, a solution containing 4 mmol of N
1
-
phenylbenzene-1,2-diamine and 5 mmol 3,5-dibromobenzoic acid was added in 10 ml
ethylene dichloride. The reaction mixture was then stirred overnight under ambient
conditions. After the reaction was stopped, the solution was washed with 5% aqueous
sodium bicarbonate solution, water and brine. The organic phase was then dried over
MgSO
4
and evaporated under reduced pressure. The obtained residue was passed
through a column of silica using hexane and ethyl acetate (from 50:1 to 10:1 gradually)
as the eluent. The fraction of 1 was concentrated under vacuum and further purified
using recrystallization with dichloromethane and hexane to afford white crystals with

139
44% yield.
1
H NMR (DMSO-d
6
δ): 7.90 (s, 1H), 7.82 (d, 1H), 7.62-7.64 (m, 5H), 7.50
(dd, 2H), 7.34 (quintet, 2H), 7.21 (dd, 1H); MS m/z 428.
2-(3,5-divinylphenyl)-1-phenyl-1H-benzo[d]imidazole (2) 1 equiv of 1, 2.2 equiv
of vinylboronic acid pinacol ester, 5 equiv of K
2
CO
3
was added against nitrogen flow
into a three-neck round bottom flask which contained the mixture of dioxane and water
with 3:1 ratio. The mixture was bubbled using nitrogen for 30 minutes, before 0.2 equiv
of Pd(PPh
3
)
4
was added. The mixture was left overnight refluxing, after additional 15
minutes of bubbling. The organic fraction of the reaction mixture was separated and
further washed with water and brine, before collected and dried over MgSO
4
. The
solvent was removed under reduced pressure and the remaining residue was separated by
chromatography using hexane and ethyl acetate as the eluent (from 100:1 to 10:1
gradually). The final product was obtained as colorless oil with 53% of yield. Extreme
caution is needed during the purification and handling process, because 2 is prone to
decomposition in solvents like dichloromethane and chloroform. Slow polymerization
was noticed, when not dissolved in solution and stored at low temperature.
1
H NMR
(DMSO-d
6
δ): 7.82 (d, 1H), 7.56-7.66 (m, 4H), 7.52 (s, 2H), 7.49 (d, 2H), 7.34 (t, 1H),
7.29 (t, 1H), 7.20 (d, 1H), 6.68 (d, 1H), 6.65 (d, 1H), 5.65 (d, 2H), 5.26 (d, 2H); MS m/z
322.
Characterization methods NMR spectra were determined with a Varian 400
NMR spectrometer. Mass spectra were recorded on an HP 5973 mass spectrometer using
electron ionization 70eV. DSC measurements were carried out on a Q10 differential

140
scanning calorimeter from TA Instruments. FT-IR and absorption spectra were collected
from a Bruker Vertex 80 FT-IR spectrometer and an Agilent 8453 UV-visible
spectrophotometer, respectively.
Polymerization procedure 2 was spin-coated at 3000 rpm for 30 seconds from
the toluene solution (10 mg/ml) filtered by syringe filters with 0.45 μm PTFE membrane,
followed by the thermal treatment under nitrogen at 120
o
C for 1h.
OLED fabrication and testing The bottom cathodes were first patterned using
shadow mask one on pre-cleaned indium tin oxide (ITO) coated glass substrates with
sheet resistance of 20 Ω /sq. Organic layers were then deposited through either thermal
evaporation from resistively heated tantalum boats at a rate of around 2Ǻ/s (for small
molecules) or spin-casting (for 2) followed by the thermal treatment to initiate the
polymerization. The top anode consisting MoO
3
and Al was deposited, after the shadow
mask two was placed on the substrate to form the defined device area. The devices were
tested in air within 1h after fabrication. Light coming out from front surface was
collected by a UV-818 Si photocathode leading to a Keithley 2400 SourceMeter/2000
multimeter coupled to a Newport 1835-C optical meter. The J-V and light-intensity
characteristics were measured by the LabVIEW program from the National Instruments.

141
5.6 Chapter 5 References

(1) Tang, C. W.; VanSlyke, S. A. Appl. Phys. Lett. 1987, 51, 913.
(2) Ji, W. Y.; Zhang, L. T.; Gao, R. X.; Zhang, L. M.; Xie, W. F.; Zhang, H. Z.; Li, B.
Opt. Express 2008, 16, 15489.
(3) Dobbertin, T.; Kroeger, M.; Heithecker, D.; Schneider, D.; Metzdorf, D.; Neuner,
H.; Becker, E.; Johannes, H. H.; Kowalsky, W. Appl. Phys. Lett. 2003, 82, 284.
(4) Zhao, J. M.; Zhan, Y. Q.; Zhang, S. T.; Wang, X. J.; Zhou, Y. C.; Wu, Y.; Wang,
Z. J.; Ding, X. M.; Hou, X. Y. Appl. phys. lett. 2004, 84, 5377.
(5) Kho, S.; Sohn, S.; Jung, D. G. Jpn. J. Appl. Phys., Part 2 2003, 42, L552.
(6) Kho, S.; Sohn, S. Y.; Jung, D.; Chae, H.; Boo, J.; Kim, B. J. Korean Phys. Soc.
2005, 46, 1224.
(7) Liao, L. S.; Hung, L. S.; Chan, W. C.; Ding, X. M.; Sham, T. K.; Bello, I.; Lee, C.
S.; Lee, S. T. Appl. phys. lett. 1999, 75, 1619.
(8) Hou, L. T.; Huang, F.; Zeng, W. J.; Peng, J. B.; Cao, Y. Appl. phys. lett. 2005, 87,
153509.
(9) Park, S. W.; Choi, J. M.; Kim, E.; Im, S. Appl. Surf. Sci. 2005, 244, 439.
(10) Chen, C. W.; Lin, C. L.; Wu, C. C. Appl. phys. lett. 2004, 85, 2469.
(11) Hung, L. S.; Tang, C. W.; Mason, M. G. Appl. Phys. Lett. 1997, 70, 152.
(12) Bulovic, V.; Tian, P.; Burrows, P. E.; Gokhale, M. R.; Forrest, S. R.; Thompson,
M. E. Appl. phys. lett. 1997, 70, 2954.
(13) Chu, T. Y.; Chen, S. Y.; Chen, J. F.; Chen, C. H. Jpn. J. Appl. Phys., Part 1 2006,
45, 4948.
(14) Xiong, T.; Wang, F. X.; Qiao, X. F.; Ma, D. G. Appl. phys. lett. 2008, 92, 263305.

142
(15) Zhou, X.; Pfeiffer, M.; Huang, J. S.; Blochwitz-Nimoth, J.; Qin, D. S.; Werner,
A.; Drechsel, J.; Maennig, B.; Leo, K. Appl. Phys. Lett. 2002, 81, 922.
(16) Wu, C. C.; Chen, C. W.; Lin, C. L.; Yang, C. J. J. Disp. Technol. 2005, 1, 248.
(17) Choi, H. W.; Kim, S. Y.; Kim, W. K.; Hong, K.; Lee, J. L. J. Appl. Phys. 2006,
100, 064106.
(18) Yun, C.; Cho, H.; Kang, H.; Lee, Y. M.; Park, Y.; Yoo, S. Appl. phys. Lett. 2009,
95, 053301.
(19) Wang, Q.; Wang, F. X.; Qiao, X. F.; Ma, D. G. Semicond. Sci. Technol. 2009, 24,
105027.
(20) Hung, L. S.; Zhang, R. Q.; He, P.; Mason, M. G. J. Phys. D 2002, 35, 103.
(21) Chen, S. Y.; Chu, T. Y.; Chen, J. F.; Su, C. Y.; Chen, C. H. Appl. Phys. Lett. 2006,
89, 053518.
(22) Chu, T. Y.; Chen, J. F.; Chen, S. Y.; Chen, C. J.; Chen, C. H. Appl. phys. Lett.
2006, 89, 053503.
(23) Bolink, H. J.; Coronado, E.; Repetto, D.; Sessolo, M.; Barea, E. M.; Bisquert, J.;
Garcia-Belmonte, G.; Prochazka, J.; Kavan, L. Adv. Funct. Mater. 2008, 18, 145.
(24) Lee, H.; Park, I.; Kwak, J.; Yoon, D. Y.; Lee, C. Appl. Phys. Lett. 2010, 96,
153306.
(25) Ma, W. L.; Iyer, P. K.; Gong, X.; Liu, B.; Moses, D.; Bazan, G. C.; Heeger, A. J.
Adv. Mater. 2005, 17, 274.
(26) Tseng, R. J.; Chiechi, R. C.; Wudl, F.; Yang, Y. Appl. Phys. Lett. 2006, 88,
093512.
(27) Ho, P. K. H.; Granstrom, M.; Friend, R. H.; Greenham, N. C. Adv. Mater. 1998,
10, 769.
(28) Bellmann, E.; Shaheen, S. E.; Grubbs, R. H.; Marder, S. R.; Kippelen, B.;
Peyghambarian, N. Chem. Mater. 1999, 11, 399.

143
(29) Fenter, P.; Schreiber, F.; Bulovic, V.; Forrest, S. R. Chem. Phys. Lett. 1997, 277,
521.
(30) Shi, S.; Tang, C. W.; Chen, C. H. US Patent No. 5,645,948, 1997.
(31) Gao, Z. Q.; Lee, C. S.; Bello, I.; Lee, S. T.; Wu, S. K.; Yan, Z. L.; Zhang, X. H.
Synth. Met. 1999, 105, 141.
(32) Tao, Y. T.; Balasubramaniam, E.; Danel, A.; Jarosz, B.; Tomasik, P. Appl. Phys.
Lett. 2000, 77, 1575.
(33) Debeaux, M.; Thesen, M. W.; Schneidenbach, D.; Hopf, H.; Janietz, S.; Kruger,
H.; Wedel, A.; Kowalsky, W.; Johannes, H. H. Adv. Funct. Mater. 2010, 20, 399.
(34) Pommerehne, J.; Selz, A.; Book, K.; Koch, F.; Zimmermann, U.; Unterlechner,
C.; Wendorff, J. H.; Heitz, W.; Bassler, H. Macromolecules 1997, 30, 8270.
(35) Grimsdale, A. C.; Chan, K. L.; Martin, R. E.; Jokisz, P. G.; Holmes, A. B. Chem.
Rev. 2009, 109, 897.
(36) Suzuki, M.; Tokito, S.; Sato, F.; Igarashi, T.; Kondo, K.; Koyama, T.; Yamaguchi,
T. Appl. phys. lett. 2005, 86, 103507.
(37) Ge, Z.; Hayakawa, T.; Ando, S.; Ueda, M.; Akiike, T.; Miyarnoto, H.; Kajita, T.;
Kakimoto, M. A. Chem. Mater. 2008, 20, 2532.
(38) Chen, C. H.; Huang, W. S.; Lai, M. Y.; Tsao, W. C.; Lin, J. T.; Wu, Y. H.; Ke, T.
H.; Chen, L. Y.; Wu, C. C. Adv. Funct. Mater. 2009, 19, 2661.
(39) Ge, Z.; Hayakawa, T.; Ando, S.; Ueda, M.; Akiike, T.; Miyamoto, H.; Kajita, T.;
Kakimoto, M. A. Adv. Funct. Mater. 2008, 18, 584.
(40) Khuong, K. S.; Jones, W. H.; Pryor, W. A.; Houk, K. N. J. Am. Chem. Soc. 2005,
127, 1265.
(41) Chong, Y. K.; Rizzardo, E.; Solomon, D. H. J. Am. Chem. Soc. 1983, 105, 7761.
(42) Mayo, F. R. J. Am. Chem. Soc. 1968, 90, 1289.
(43) Hendrickson, J. B.; Hussoin, S. J. Org. Chem. 1989, 54, 1144.

144
(44) Aaberg, A.; Gramstad, T.; Husebye, S. Tetrahedron Lett. 1979, 20, 2263.
(45) Cheng, Y. J.; Liu, M. S.; Zhang, Y.; Niu, Y. H.; Huang, F.; Ka, J. W.; Yip, H. L.;
Tian, Y. Q.; Jen, A. K. Y. Chem. Mater. 2008, 20, 413.
(46) Rege, R. V.; Webster, C. C.; Ostrow, J. D.; Carr, S. H.; Ohkubo, H. Biochem. J.
1984, 224, 871.
(47) Lim, B.; Hwang, J. T.; Kim, J. Y.; Ghim, J.; Vak, D.; Noh, Y. Y.; Lee, S. H.; Lee,
K.; Heeger, A. J.; Kim, D. Y. Org. Lett. 2006, 8, 4703.
(48) Liu, S.; Jiang, X. Z.; Ma, H.; Liu, M. S.; Jen, A. K. Y. Macromolecules 2000, 33,
3514.
(49) Niu, Y. H.; Liu, M. S.; Ka, J. W.; Jen, A. K. Y. Appl. phys. lett. 2006, 88, 093505.
(50) Ma, B.; Lauterwasser, F.; Deng, L.; Zonte, C. S.; Kim, B. J.; Frechet, J. M. J.
Chem. Mater. 2007, 19, 4827.
(51) Du, H.; Fuh, R. C. A.; Li, J. Z.; Corkan, L. A.; Lindsey, J. S. Photochem.
Photobiol. 1998, 68, 141.
(52) Destruel, P.; Bock, H.; Seguy, I.; Jolinat, P.; Oukachmih, M.; Bedel-Pereira, E.
Polym. Int. 2006, 55, 601.
(53) Wang, F. X.; Qiao, X. F.; Xiong, T.; Ma, D. G. Org. Electron. 2008, 9, 985.
(54) Sugiyama, K.; Ishii, H.; Ouchi, Y.; Seki, K. J. Appl. Phys. 2000, 87, 295.
(55) Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest, S. R.
Appl. Phys. Lett. 1999, 75, 4.
(56) Hill, I. G.; Kahn, A. J. Appl. Phys. 1999, 86, 4515.

145
Comprehensive Bibliography

Aaberg, A.; Gramstad, T.; Husebye, S. Tetrahedron Lett. 1979, 20, 2263.
Adachi, C.; Baldo, M. A.; Forrest, S. R.; Thompson, M. E. Appl. Phys. Lett. 2000, 77,
904.
Adachi, C.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. J. Appl. Phys. 2001, 90, 5048.
Adamovich, V.; Brooks, J.; Tamayo, A.; Alexander, A.; Djurovich, P. I.; D’Andrade, B.
W.; Adachi, C.; Forrest, S. R.; Thompson, M. E. New J. Chem. 2002, 26, 1171.
Aziz, H.; Popovic, Z. D.; Hu, N. X. Appl. Phys. Lett. 2002, 81, 370.
Baek, H. I.; Lee, C. H. J. Phys. D-Appl. Phys. 2008, 41, 124504.
Baldo, M. A.; Forrest, S. R. Phys. Rev. B 2000, 62, 10958.
Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest, S. R. Appl. Phys.
Lett. 1999, 75, 4.
Baldo, M. A.; O'Brien, D. F.; Thompson, M. E.; Forrest, S. R. Phys. Rev. B 1999, 60,
14422.
Baldo, M. A.; O'Brien, D. F.; You, Y.; Shoustikov, A.; Sibley, S.; Thompson, M. E.;
Forrest, S. R. Nature 1998, 395, 151.
Bandy, J. A.; Davies, C. E.; Green, M. L. H.; Green, J. C.; Prout, K.; Rodgers, D. P. S. J.
Chem. Soc.-Chem. Commun. 1983, 23, 1395.
Barker, C. C.; Emmerson, R. G.; Periam, J. D. J. Am. Chem. Soc. 1955, 4, 4482.
Basolo, F. Coord. Chem. Rev. 1968, 3, 213.
Bassler, H. Phys. Status Solidi B 1993, 175, 15.
Beierlein, T. A.; Ruhstaller, B.; Gundlach, D. J.; Riel, H.; Karg, S.; Rost, C.; Riess, W.
Synth. Met. 2003, 138, 213.
Bellmann, E.; Shaheen, S. E.; Grubbs, R. H.; Marder, S. R.; Kippelen, B.; Peyghambarian,
N. Chem. Mater. 1999, 11, 399.

146
Bolink, H. J.; Coronado, E.; Repetto, D.; Sessolo, M.; Barea, E. M.; Bisquert, J.; Garcia-
Belmonte, G.; Prochazka, J.; Kavan, L. Adv. Funct. Mater. 2008, 18, 145.
Bossi, A.; Wu, C.; Djurovich, P. I.; Thompson, M. E. private communication, manuscript
submitted 2010.
Braga, D.; Grepioni, F. Organometallics 1992, 11, 711.
Brooks, J.; Babayan, Y.; Lamansky, S.; Djurovich, P. I.; Tsyba, I.; Bau, R.; Thompson,
M. E. Inorg. Chem. 2002, 41, 3055.
Brown, A. R.; Pichler, K.; Greenham, N. C.; Bradley, D. D. C.; Friend, R. H.; Holmes, A.
B. Chem. Phys. Lett. 1993, 210, 61.
Buchner, R.; Cunningham, C. T.; Field, J. S.; Haines, R. J.; McMillin, D. R.; Summerton,
G. C. J. Chem. Soc., Dalton Trans. 1999, 5, 711.
Bulovic, V.; Tian, P.; Burrows, P. E.; Gokhale, M. R.; Forrest, S. R.; Thompson, M. E.
Appl. phys. lett. 1997, 70, 2954.
Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay, K.; Friend, R.
H.; Burns, P. L.; Holmes, A. B. Nature 1990, 347, 539.
Bushuk, S.; Douglas, W.; Kalvinkovskaya, Y.; Klapshina, L.; Rubinov, A.; Bushuk, B.;
Stupak, A. J. Lumin. 2003, 105, 81.
Chen, C. H.; Huang, W. S.; Lai, M. Y.; Tsao, W. C.; Lin, J. T.; Wu, Y. H.; Ke, T. H.;
Chen, L. Y.; Wu, C. C. Adv. Funct. Mater. 2009, 19, 2661.
Chen, C. W.; Lin, C. L.; Wu, C. C. Appl. phys. lett. 2004, 85, 2469.
Chen, S. Y.; Chu, T. Y.; Chen, J. F.; Su, C. Y.; Chen, C. H. Appl. Phys. Lett. 2006, 89,
053518.
Cheng, Y. J.; Liu, M. S.; Zhang, Y.; Niu, Y. H.; Huang, F.; Ka, J. W.; Yip, H. L.; Tian, Y.
Q.; Jen, A. K. Y. Chem. Mater. 2008, 20, 413.
Cheung, T. C.; Cheung, K. K.; Peng, S. M.; Che, C. M. J. Chem. Soc., Dalton Trans.
1996, 8, 1645.
Choi, H. W.; Kim, S. Y.; Kim, W. K.; Hong, K.; Lee, J. L. J. Appl. Phys. 2006, 100,
064106.
Chong, Y. K.; Rizzardo, E.; Solomon, D. H. J. Am. Chem. Soc. 1983, 105, 7761.

147
Chu, T. Y.; Chen, J. F.; Chen, S. Y.; Chen, C. J.; Chen, C. H. Appl. phys. Lett. 2006, 89,
053503.
Chu, T. Y.; Chen, S. Y.; Chen, J. F.; Chen, C. H. Jpn. J. Appl. Phys., Part 1 2006, 45,
4948.
Closs, G. L.; Johnson, M. D.; Miller, J. R.; Piotrowiak, P. J. Am. Chem. Soc. 1989, 111,
3751.
Colombo, M. G.; Brunold, T. C.; Riedener, T.; Gudel, H. U.; Fortsch, M.; Burgi, H. B.
Inorg. Chem. 1994, 33, 545.
Colombo, M. G.; Hauser, A.; Gudel, H. U. In Electronic and Vibronic Spectra of
Transition Metal Complexes I. Topics in Current Chemistry Series, 2
nd
ed.; Yersin, H.,
Ed.; Springer: Berlin, 1994; Vol. 171, pp. 143.
Crabtree, R. H.; Mingos, D. M. Comprehensive organometallic chemistry III; Elsevier:
Oxford, UK, 2007; Vol. 12.
D'Andrade, B. W.; Brooks, J.; Adamovich, V.; Thompson, M. E.; Forrest, S. R. Adv.
Mater. 2002, 14, 1032.
D'Andrade, B. W.; Datta, S.; Forrest, S. R.; Djurovich, P.; Polikarpov, E.; Thompson, M.
E. Org. Electron. 2005, 6, 11.
D'Andrade, B. W.; Holmes, R. J.; Forrest, S. R. Adv. Mater. 2004, 16, 624.
D'Andrade, B. W.; Forrest, S. R. Adv. Mater. 2004, 16, 1585.
D'Andrade, B. W.; Thompson, M. E.; Forrest, S. R. Adv. Mater. 2002, 14, 147.
Debeaux, M.; Thesen, M. W.; Schneidenbach, D.; Hopf, H.; Janietz, S.; Kruger, H.;
Wedel, A.; Kowalsky, W.; Johannes, H. H. Adv. Funct. Mater. 2010, 20, 399.
Destruel, P.; Bock, H.; Seguy, I.; Jolinat, P.; Oukachmih, M.; Bedel-Pereira, E. Polym.
Int. 2006, 55, 601.
Dexter, D. L. J. Chem. Phys. 1953, 21, 836.
Di Censo, D.; Fantacci, S.; De Angelis, F.; Klein, C.; Evans, N.; Kalyanasundaram, K.;
Bolink, H. J.; Gratzel, M.; Nazeeruddin, M. K. Inorg. Chem. 2008, 47, 980.
Dirac, P. A. M. Proc. R. Soc. Lond. Series A 1927, 114, 243.

148
Djurovich, P. I.; Mayo, E. I.; Forrest, S. R.; Thompson, M. E. Org. Electron. 2009, 10,
515.
Dobbertin, T.; Kroeger, M.; Heithecker, D.; Schneider, D.; Metzdorf, D.; Neuner, H.;
Becker, E.; Johannes, H. H.; Kowalsky, W. Appl. Phys. Lett. 2003, 82, 284.
Du, H.; Fuh, R. C. A.; Li, J. Z.; Corkan, L. A.; Lindsey, J. S. Photochem. Photobiol. 1998,
68, 141.
Fenter, P.; Schreiber, F.; Bulovic, V.; Forrest, S. R. Chem. Phys. Lett. 1997, 277, 521.
Förster, T. Göttingen: Vandenhoeck and Ruprecht 1951.
Gao, Z. Q.; Lee, C. S.; Bello, I.; Lee, S. T.; Wu, S. K.; Yan, Z. L.; Zhang, X. H. Synth.
Met. 1999, 105, 141.
Ge, Z.; Hayakawa, T.; Ando, S.; Ueda, M.; Akiike, T.; Miyamoto, H.; Kajita, T.;
Kakimoto, M. A. Adv. Funct. Mater. 2008, 18, 584.
Ge, Z; Hayakawa, T.; Ando, S.; Ueda, M.; Akiike, T.; Miyarnoto, H.; Kajita, T.;
Kakimoto, M. A. Chem. Mater. 2008, 20, 2532.
Giebink, N. C.; Sun, Y.; Forrest, S. R. Org. Electron. 2006, 7, 375.
Graber, S.; Doyle, K.; Neuburger, M.; Housecroft, C. E.; Constable, E. C.; Costa, R. D.;
Orti, E.; Repetto, D.; Bolink, H. J. J. Am. Chem. Soc. 2008, 130, 14944.
Green, M. L. H.; Hamnett, A.; Qin, J.; Baird, P.; Bandy, J. A.; Prout, K.; Marseglia, E.;
Obertelli, S. D. J. Chem. Soc.-Chem. Commun. 1987, 24, 1811.
Grimsdale, A. C.; Chan, K. L.; Martin, R. E.; Jokisz, P. G.; Holmes, A. B. Chem. Rev.
2009, 109, 897.
Helfrich, W.; Schneider, W. G. Phys. Rev. Lett. 1965, 14, 229.
Hendrickson, J. B.; Hussoin, S. J. Org. Chem. 1989, 54, 1144.
Herkstroeter, W. G.; Hammond, G. S.; Lamola, A. A. J. Am. Chem. Soc. 1964, 86, 4537.
Hill, I. G.; Kahn, A. J. Appl. Phys. 1999, 86, 4515.
Ho, C. L.; Wong, W. Y.; Zhou, G. H.; Yao, B.; Xie, Z. Y.; Wang, L. X. Adv. Funct.
Mater. 2007, 17, 2925.
Ho, P. K. H.; Granstrom, M.; Friend, R. H.; Greenham, N. C. Adv. Mater. 1998, 10, 769.

149
Holmes, R. J.; D’Andrade, B. W.; Forrest, S. R.; Ren, X.; Li, J.; Thompson, M. E. Appl.
Phys. Lett. 2003, 83, 3818.
Hou, L. T.; Huang, F.; Zeng, W. J.; Peng, J. B.; Cao, Y. Appl. phys. lett. 2005, 87,
153509.
Huang, L.; Wang, K. Z.; Huang, C. H.; Gao, D. Q.; Jin, L. P. Synth. Met. 2002, 128, 241.
Hung, L. S.; Chen, C. H. Mater. Sci. Eng., R 2002, 39, 143.
Hung, L. S.; Tang, C. W.; Mason, M. G. Appl. Phys. Lett. 1997, 70, 152.
Hung, L. S.; Zhang, R. Q.; He, P.; Mason, M. G. J. Phys. D 2002, 35, 103.
Ikai, M.; Tokito, S.; Sakamoto, Y.; Suzuki, T.; Taga, Y. Appl. Phys. Lett. 2001, 79, 156.
Itaya, A.; Okamoto, K.; Kusabayashi, S. Bull. Chem. Soc. Jpn. 1976, 49, 2037.
Ji, W. Y.; Zhang, L. T.; Gao, R. X.; Zhang, L. M.; Xie, W. F.; Zhang, H. Z.; Li, B. Opt.
Express 2008, 16, 15489.
Jiang, X. Y.; Zhang, Z. L.; Zhao, W. M.; Zhu, W. Q.; Zhang, B. X.; Xu, S. H. J. Phys. D-
Appl. Phys. 2000, 33, 473.
Jiang, X. Z.; Register, R. A.; Killeen, K. A.; Thompson, M. E.; Pschenitzka, F.; Hebner,
T. R.; Sturm, J. C. J. Appl. Phys. 2002, 91, 6717.
Justel, T.; Nikol, H.; Ronda, C. Angew. Chem. Int. Ed. 1998, 37, 3085.
Kawamura, Y.; Yanagida, S.; Forrest, S. R. J. Appl. Phys. 2002, 92, 87.
Khan, A. L. T.; Sreearunothai, P.; Herz, L. M.; Banach, M. J.; Kohler, A. Phys. Rev. B
2004, 69.
Kho, S.; Sohn, S.; Jung, D. G. Jpn. J. Appl. Phys., Part 2 2003, 42, L552.
Kho, S.; Sohn, S. Y.; Jung, D.; Chae, H.; Boo, J.; Kim, B. J. Korean Phys. Soc. 2005, 46,
1224.
Khuong, K. S.; Jones, W. H.; Pryor, W. A.; Houk, K. N. J. Am. Chem. Soc. 2005, 127,
1265.
Kido, J.; Kimura, M.; Nagai, K. Science 1995, 267, 1332.
Kido, J.; Shionoya, H.; Nagai, K. Appl. Phys. Lett. 1995, 67, 2281.

150
Koene, B. E.; Loy, D. E.; Thompson, M. E. Chem. Mater. 1998, 10, 2235.
Kwong, R. C.; Sibley, S.; Dubovoy, T.; Baldo, M.; Forrest, S. R.; Thompson, M. E.
Chem. Mater. 1999, 11, 3709.
Lacowicz, J. R. Principles of Fluorescence Spectroscopy, Springer: New York, NY 2006.
Lai, S. W.; Lam, H. W.; Lu, W.; Cheung, K. K.; Che, C. M. Organometallics 2002, 21,
226.
Lamansky, S.; Djurovich, P. I.; Murphy, D.; Abdel-Razzaq, F.; Kwong, R.; Tsyba, I.;
Botz, M.; Mui, B.; Thompson, M. E. Inorg. Chem. 2001, 40, 1704.
Lamansky, S.; Djurovich, P. I.; Murphy, D.; Abdel-Razzaq, F.; Lee, H.-F.; Adachi, C.;
Burrows, P. E.; Forrest, S. R.; Thompson, M. E. J. Am. Chem. Soc. 2001, 123, 4304.
Lee, H.; Park, I.; Kwak, J.; Yoon, D. Y.; Lee, C. Appl. Phys. Lett. 2010, 96, 153306.
Lees, A. J. Chem. Rev. 1987, 87, 711.
Li, J.; Djurovich, P. I.; Alleyne, B. D.; Yousufuddin, M.; Ho, N. N.; Thomas, J. C.; Peters,
J. C.; Bau, R.; Thompson, M. E. Inorg. Chem. 2005, 44, 1713.
Liao, L. S.; Hung, L. S.; Chan, W. C.; Ding, X. M.; Sham, T. K.; Bello, I.; Lee, C. S.; Lee,
S. T. Appl. phys. lett. 1999, 75, 1619.
Lim, B.; Hwang, J. T.; Kim, J. Y.; Ghim, J.; Vak, D.; Noh, Y. Y.; Lee, S. H.; Lee, K.;
Heeger, A. J.; Kim, D. Y. Org. Lett. 2006, 8, 4703.
Liu, S.; Jiang, X. Z.; Ma, H.; Liu, M. S.; Jen, A. K. Y. Macromolecules 2000, 33, 3514.
Lo, K. K. W.; Chan, J. S. W.; Lui, L. H.; Chung, C. K. Organometallics 2004, 23, 3108.
Lohse, O.; Thevenin, P.; Waldvogel, E. Synlett 1999, 1, 45.
Lowry, M. S.; Hudson, W. R.; Pascal, R. A.; Bernhard, S. J. Am. Chem. Soc. 2004, 126,
14129.
Lu, W.; Mi, B. X.; Chan, M. C. W.; Hui, Z.; Zhu, N. Y.; Lee, S. T.; Che, C. M. Chem.
Commun. 2002, 3, 206.
Ma, B.; Djurovich, P. I.; Thompson, M. E. Coord. Chem. Rev. 2005, 249, 1501.

151
Ma, B.; Lauterwasser, F.; Deng, L.; Zonte, C. S.; Kim, B. J.; Frechet, J. M. J. Chem.
Mater. 2007, 19, 4827.
Ma, W. L.; Iyer, P. K.; Gong, X.; Liu, B.; Moses, D.; Bazan, G. C.; Heeger, A. J. Adv.
Mater. 2005, 17, 274.
Marcus, R. A. Angew. Chem. Int. Ed. 1993, 32, 1111.
Marcus, R. A. J. Chem. Phys. 1956, 24, 966.
Markham, J. P. J.; Lo, S. C.; Magennis, S. W.; Burn, P. L.; Samuel, I. D. W. Appl. Phys.
Lett. 2002, 80, 2645.
Mauro, M; Schuermann, K. C.; Pretot, R.; Hafner, A.; Mercandelli, P.; Sironi, A.; De
Cola, L. Angew. Chem. Int. Ed. 2010, 49, 1222.
Mayo, F. R. J. Am. Chem. Soc. 1968, 90, 1289.
McQueen, A. E. D.; Blake, A. J.; Stephenson, T. A.; Schroder, M.; Yellowlees, L. J. J.
Chem. Soc.-Chem. Commun. 1988, 23, 1533.
Meerheim, R.; Walzer, K.; Pfeiffer, M.; Leo, K. Appl. Phys. Lett. 2006, 89, 061111.
Muccini, M. Nat. Mater. 2006, 5, 605.
Nazeeruddin, M. K.; Humphry-Baker, R.; Berner, D.; Rivier, S.; Zuppiroli, L.; Graetzel,
M. J. Am. Chem. Soc. 2003, 125, 8790.
Niu, Y. H.; Liu, M. S.; Ka, J. W.; Jen, A. K. Y. Appl. phys. lett. 2006, 88, 093505.
Nonoyama, M. Bull. Chem. Soc. Jpn. 1974, 47, 767.
O'Brien, D. F.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett. 1999, 74,
442.
Ohno, T.; Nozaki, K.; Nakamura, M.; Motojima, Y.; Tsushima, M.; Ikeda, N. Inorg.
Chem. 2007, 46, 8859.
Park, S. W.; Choi, J. M.; Kim, E.; Im, S. Appl. Surf. Sci. 2005, 244, 439.
Pasynskii, A. A.; Eremenko, I. L.; Rakitin, Y. V.; Novotortsev, V. M.; Ellert, O. G.;
Kalinnikov, V. T.; Shklover, V. E.; Struchkov, Y. T.; Lindeman, S. V.; Kurbanov, T. K.;
Gasanov, G. S. J. Organomet. Chem. 1983, 248, 309.
Place, I.; Farran, A.; Deshayes, K.; Piotrowiak, P. J. Am. Chem. Soc. 1998, 120, 12626.

152
Pommerehne, J.; Selz, A.; Book, K.; Koch, F.; Zimmermann, U.; Unterlechner, C.;
Wendorff, J. H.; Heitz, W.; Bassler, H. Macromolecules 1997, 30, 8270.
Pope, C. E. S. M. Electronic Processes in Organic Crystals and Polymers, Oxford
University Press: New York, 1999.
Pope, M.; Kallmann, H. P.; Magnante, P. J. J. Chem. Phys. 1963, 38, 2042.
Pschenitzka, F.; Sturm, J. C. Appl. Phys. Lett. 2001, 79, 4354.
Rege, R. V.; Webster, C. C.; Ostrow, J. D.; Carr, S. H.; Ohkubo, H. Biochem. J. 1984,
224, 871.
Rim, S. B.; Peumans, P. J. Appl. Phys. 2008, 103, 124515.
Round, H. J. Electr. World 1907, 19, 309.
Sajoto, T.; Djurovich, P. I.; Tamayo, A. B.; Oxgaard, J.; Goddard, W. A.; Thompson, M.
E. J. Am. Chem. Soc. 2009, 131, 9813.
Sajoto, T.; Djurovich, P. I.; Tamayo, A. B.; Yousufuddin, M.; Bau, R.; Thompson, M. E.;
Holmes, R. J.; Forrest, S. R. Inorg. Chem. 2005, 44, 7992.
Sakanoue, K.; Motoda, M.; Sugimoto, M.; Sakaki, S. J. Phys. Chem. A 1999, 103, 5551.
Schubert, E. F. Light-Emitting Diodes; Cambridge University Press: Cambridge, UK,
2003.
Schwartz, G.; Pfeiffer, M.; Reineke, S.; Walzer, K.; Leo, K. Adv. Mater. 2007, 19, 3672.
Segal, M.; Baldo, M. A.; Holmes, R. J.; Forrest, S. R.; Soos, Z. G. Phys. Rev. B 2003, 68,
075211.
Shi, J.; Tang, C. W. Appl. Phys. Lett. 2002, 80, 3201.
Shi, S.; Tang, C. W.; Chen, C. H. US Patent No. 5,645,948, 1997.
Shoustikov, A. A.; You, Y.; Thompson, M. E. IEEE J. Sel. Top. Quantum Electron. 1998,
4, 3.
Sirringhaus, H. Nat. Mater. 2003, 2, 641.
Slinker, J. D.; Koh, C. Y.; Malliaras, G. G.; Lowry, M. S.; Bernhard, S. Appl. Phys. Lett.
2005, 86, 173506.

153
Sprouse, S.; King, K. A.; Spellane, P. J.; Watts, R. J. J. Am. Chem. Soc. 1984, 106, 6647.
Su, H. C.; Chen, H. F.; Wu, C. C.; Wong, K. T. Chem.--Asian J. 2008, 3, 1922.
Sugiyama, K.; Ishii, H.; Ouchi, Y.; Seki, K. J. Appl. Phys. 2000, 87, 295.
Sun, Y. R.; Giebink, N. C.; Kanno, H.; Ma, B. W.; Thompson, M. E.; Forrest, S. R.
Nature 2006, 440, 908.
Suzuki, M.; Tokito, S.; Sato, F.; Igarashi, T.; Kondo, K.; Koyama, T.; Yamaguchi, T.
Appl. phys. lett. 2005, 86, 103507.
Sze, S. M. Physics of Semiconductor Devices; 2nd ed.; John Wiley: New York, 1981.
Tamayo, A. B.; Garon, S.; Sajoto, T.; Djurovich, P. I.; Tsyba, I. M.; Bau, R.; Thompson,
M. E. Inorg. Chem. 2005, 44, 8723.
Tang, C. W.; VanSlyke, S. A. Appl. Phys. Lett. 1987, 51, 913.
Tang, C. W.; VanSlyke, S. A.; Chen, C. H. J. Appl. Phys. 1989, 65, 3610.
Tao, Y. T.; Balasubramaniam, E.; Danel, A.; Jarosz, B.; Tomasik, P. Appl. Phys. Lett.
2000, 77, 1575.
Tasch, S.; List, E. J. W.; Ekstrom, O.; Graupner, W.; Leising, G.; Schlichting, P.; Rohr,
U.; Geerts, Y.; Scherf, U.; Mullen, K. Appl. Phys. Lett. 1997, 71, 2883.
Toan, T.; Teo, B. K.; Ferguson, J. A.; Meyer, T. J.; Dahl, L. F. J. Am. Chem. Soc. 1977,
99, 408.
Tse, S. C.; So, S. K.; Yeung, M. Y.; Lo, C. F.; Wen, S. W.; Chen, C. H. Chem. Phys. Lett.
2006, 422, 354.
Tseng, R. J.; Chiechi, R. C.; Wudl, F.; Yang, Y. Appl. Phys. Lett. 2006, 88, 093512.
Tsuboyama, A.; Iwawaki, H.; Furugori, M.; Mukaide, T.; Kamatani, J.; Igawa, S.;
Moriyama, T.; Miura, S.; Takiguchi, T.; Okada, S.; Hoshino, M.; Ueno, K. J. Am. Chem.
Soc. 2003, 125, 12971.
Tsung, K. K.; So, S. K. Appl. Phys. Lett. 2008, 92, 103315.
Turro, N. J. Modern Molecular Photochemistry; Benjamin/Cummings: Menlo Park, CA,
1978.
Uchida, M.; Adachi, C.; Koyama, T.; Taniguchi, Y. J. Appl. Phys. 1999, 86, 1680.

154
Utsugi, K.; Takano, S. J. Electrochem. Soc. 1992, 139, 3610.
Wang, D.; Li, W.; Chu, B.; Su, Z.; Bi, D.; Zhang, D.; Zhu, J.; Yan, F.; Chen, Y.; Tsuboi,
T. Appl. Phys. Lett. 2008, 92, 3.
Wang, F. X.; Qiao, X. F.; Xiong, T.; Ma, D. G. Org. Electron. 2008, 9, 985.
Wang, Q.; Wang, F. X.; Qiao, X. F.; Ma, D. G. Semicond. Sci. Technol. 2009, 24, 105027.
Watanabe, S.; Agata, Y.; Tanaka, D.; Kido, J. J. Photopolym. Sci. Technol. 2005, 18, 83.
Wu, C. C.; Chen, C. W.; Lin, C. L.; Yang, C. J. J. Disp. Technol. 2005, 1, 248.
Xiong, T.; Wang, F. X.; Qiao, X. F.; Ma, D. G. Appl. phys. lett. 2008, 92, 263305.
Yang, C. L.; Tang, Z. K.; Ge, W. K.; Wang, J. N.; Zhang, Z. L.; Jian, X. Y. Appl. Phys.
Lett. 2003, 83, 1737.
Yamamoto, T.; Nishiyama, M.; Koie, Y. Tetrahedron Lett. 1998, 39, 2367.
Yersin, H. In Transition Metal And Rare Earth Compounds III 2004, 241, 1.
Yu, X. M.; Zhou, G. J.; Lam, C. S.; Wong, W. Y.; Zhu, X. L.; Sun, J. X.; Wong, M.;
Kwok, H. S. J. Organomet. Chem. 2008, 693, 1518.
Yun, C.; Cho, H.; Kang, H.; Lee, Y. M.; Park, Y.; Yoo, S. Appl. phys. Lett. 2009, 95,
053301.
Zhao, J. M.; Zhan, Y. Q.; Zhang, S. T.; Wang, X. J.; Zhou, Y. C.; Wu, Y.; Wang, Z. J.;
Ding, X. M.; Hou, X. Y. Appl. phys. lett. 2004, 84, 5377.
Zheng, G. Y.; Rillema, D. P. Inorg. Chem. 1998, 37, 1392.
Zhou, X.; Pfeiffer, M.; Huang, J. S.; Blochwitz-Nimoth, J.; Qin, D. S.; Werner, A.;
Drechsel, J.; Maennig, B.; Leo, K. Appl. Phys. Lett. 2002, 81, 922.
Zhou, Y. C.; Ma, L. L.; Zhou, J.; Ding, X. M.; Hou, X. Y. Phys. Rev. B 2007, 75, 132202.
Zhu, X. L.; Sun, J. X.; Yu, X. M.; Wong, M.; Kwok, H. S. Jpn. J. Appl. Phys., Part 1
2007, 46, 4054.

Abstract (if available)

Abstract Organic light-emitting diode (OLED) refers to any light-emitting diode (LED) using exclusively molecular and polymeric materials. Since the first demonstration in late 80s, OLED has been the topic of intense interest to both scientific and industrial community. With the progresses made on improving performance, the most advanced OLEDs have demonstrated near unity quantum efficiencies and over tens of thousands of hour-long lifetime. They have been developed to the point that that they are commercially available in small, hand-held, full color displays and showed great promise in lighting applications as well. All the aforementioned achievements were driven by a series of new materials, device structures and fabrication techniques, which will continue to be the force moving OLED technology towards wider success. This dissertation follows the mainstream of the OLED research, trying to incorporate materials with attractive properties into both conventional and newly-designed device structures, in the mean time demonstrating the advantages in simplicity and performance.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

New materials for organic light emitting diodes

PDF

High energy hosts and blue emitters for phosphorescent organic light emitting diodes

PDF

Studies of molecular orientation using iridium phosphors and integration of corannulene into organic light emitting diodes (OLEDs)

PDF

Ir(III) and Pt(II) phosphorescent emitters in organic light emitting diodes: from materials development to light out-coupling

PDF

Development of N-type chromophores for organic photovoltaics, and thermally activated delayed fluorescence NHC complexes for organic light-emitting diodes

PDF

Integration of cost effective luminescent materials into organic light emitting diodes (OLEDs) and molecular approaches to improve the device efficiency

PDF

Novel pyrimidine-based hole blocking materials for long-lived and highly efficient organic light emitting diodes

PDF

Dopants for organic light-emitting devices

PDF

Molecular approaches to solve the blue problem in organic light emitting diode display and lighting applications

PDF

One-dimensional nanostructures for chemical sensing, transparent electronics, and energy conversion and storage devices

PDF

Zero-dimensional and one-dimensional nanostructured materials for application in photovoltaic cells

PDF

Nanomaterials for macroelectronics and energy storage device

PDF

Nanomaterials for energy storage devices and electronic/optoelectronic devices

PDF

Molecular design strategies for blue organic light emitting diodes

PDF

Qualitative quantum chemical description of the structure-color relationships in phosphorescent organometallic complexes

PDF

The performance of light emitting diode (LED) light curing units and dental radiometers

PDF

New materials and fabrication techniques and their utilization in organic photovoltaic devices

PDF

Printed electronics based on carbon nanotubes and two-dimensional transition metal dichalcogenides

PDF

Organic photovoltaics: from materials development to device application

PDF

Chemical vapor deposition of graphene and two-dimensional materials: synthesis, characterization, and applications

Asset Metadata

Creator Wu, Chao (author)

Core Title New materials and device structure for organic light-emitting diodes

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Materials Science

Publication Date 04/01/2011

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag cross-linkable electron-conducting material,inverted OLED,Ir-based soft salt,OAI-PMH Harvest,OLED,triplet exciton diffusion,WOLED

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Thompson, Mark E. (committee chair), Goo, Edward K. (committee member), Zhou, Chongwu (committee member)

Creator Email chaowu@usc.edu,chster@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3489

Unique identifier UC163089

Identifier etd-Wu-4112 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-411639 (legacy record id),usctheses-m3489 (legacy record id)

Legacy Identifier etd-Wu-4112.pdf

Dmrecord 411639

Document Type Dissertation

Rights Wu, Chao

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu