University of Southern California Dissertations and Theses

Dopants for organic light-emitting devices

(USC Thesis Other)

Dopants for organic light-emitting devices

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content DOPANTS FOR ORGANIC LIGHT-EMITTING DEVICES

by
Evgueni Polikarpov

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

May 2008

Copyright 2008 Evgueni Polikarpov

ii
Acknowledgements
First of all, I would like to thank my remarkable advisor Professor Mark
Thompson. During my work at USC under his guidance my idea of what doing
science in the real world is like acquired much more detail. Equally important, I
can admit (with no concession) that here at USC, I spent the most fun, exciting,
and fulfilling time of my life. That is, for no small portion because I was fortunate
to work in a place run by a good person.
During my first two years of graduate school at USC I did take classes. In
my opinion, the quality of teaching and originality of the curricula ranged from
excellent to exceptional for most of the classes I attended, and if I choose an
educational profession, I certainly have many examples of what constitutes
excellent teaching, in terms of both methods and attitude. I would like to thank
USC Professors Stephen Bradforth, Thomas Flood, Robert Bau, Arieh Warshel,
and Anna Krylov for being great educators who put a lot of effort into designing
and teaching interesting courses. Also, my special thanks here go to Professor
Anna Krylov for her guidance and help during my first year at USC.
It is my pleasure to name my friends and colleagues who have been
students, postdocs, and staff in our research group and at USC Chemistry
Department, with whom I shared lab equipment and glassware, a camping tent in
the Sierra, TA duties, a monthly rent, and many other things, and with whom my

iii
graduate school experience was what it was: Vadim Adamovich, Bert Alleyn, Alex
Alexander, Carsten Borek, Jimmy Ly, Marco Curreli, Simona Garon, Biwu Ma,
Paulin Wajudi, Seogshin Kang, Laurent Griffe, Kristin Martinez, Dolores Perez,
Cody Schlenker, Libby Mayo, Azad Hassan, Peter Djurovich, Chao Wu, Rui
Zhang, Slava Diev, Wei Wei, Tissa Sajoto, Arnold Tamayo, Ken Hanson, Siyi
Wang, Jin Ho Oh, Askat Jailaubekov, Oleg Mironov, Dmitry Skvortsov, Ludmila
and Mikhail Slipchenko, Kirill Kuyanov, Igor Fedorov, Nikolai Markovsky, Boris
Karpichev, Sergey Malyk.
Jim Merritt, the USC scientific glassblowing shop director, was always a
great help not only in fixing glassware that I broke, but also in making custom-
designed setups that I used in some of my research projects described in this
Thesis. For serving on my screening, quals, and defense committee I thank
Professors Edward Goo Florian Mansfeld, Robert Bau, Stephen Bradforth, and
Roy Perianna. My special thanks go to Judy Hom for her generous help in so many
ways. Also, I thank Chemistry Department staff Michele Dea, Heather Connor,
and Jaime Avila for their help and advice.
Funding for my research came from various sources some of which I knew
(like the University of Southern California or the Universal Display Corporation),
others that I didn’t know, and I thank all the sponsors and funding agencies for
their financial support. From my personal perspective, the full credit for keeping

iv
research running and the labs well-stocked goes to my advisor Prof. Mark
Thompson anyway.
Finally, I would like to thank my parents Vladimir Stepanovich and
Tatyana Borisovna for their love and support.

v
List of Tables
Table 1 Fluorescence and phosphorescence peak maxima (measured at 77
K temperature), and singlet-triplet gaps in electronvolts for four
indolocarbazole derivatives .......................................................................... 107
Table 2 Quantum yields, fluorescent life times, and radiative rate constants
(k
R
= QE/ τ) for IC derivatives measured for degassed toluene
solutions at two different wavelengths using glass cells .............................. 114
Table 3 Quantum yield wavelength dependence for NPIC. Same sample in
toluene ........................................................................................................... 115
Table 4 Quantum yield wavelength dependence for MeIC. Same sample in
toluene ........................................................................................................... 116
Table 5 The effect of replacing glass quvettes with quartz ones. 340nm
excitation wavelength was used in all 4 measurements ................................ 116
Table 6 Excitation wavelength dependence of the quantum yield of the
BASF-083 fluorescent standard .................................................................... 117
Table 7 Wavelength dependence of quantum yield for AlQ
3
............................... 118
Table 8 Quinine-corrected quantum yields (QY
degas
), fluorescent life times
( τ), and radiative rate constants (k
R
= QY/ τ) for IC derivatives
measured for degassed toluene solutions at room temperature using
quartz cells .................................................................................................... 120
Table 9 Quantum yields for the films of DMIC and DTIC doped into CBP
at different doping concentrations measured under N
2
positive
pressure and in air (numbers in parentheses) ................................................ 124
Table 10 Comparison of quantum efficiency decrease with increasing the
doping concentration for DTIC and TBIC (fresh spin-cast from
CH
2
Cl
2
films held under positive pressure of nitrogen during QE
measurements) .............................................................................................. 131

vi
List of Figures
Figure 1 A schematic of a double heterostructure OLED consisting of a
hole transport layer (HTL), electron transport layer (ETL), emissive
layer (EML), and the electrodes ........................................................................ 2
Figure 2 Chemical structures, the Commission Internationale de
L’Eclairage (CIE) chromaticity coordinates of OLEDs and
phosphorescence spectra of iridium cyclometalated complexes. The
CIE coordinates for OLEDs with the ppy
2
Ir(acac), bt
2
Ir(acac),
pq
2
Ir(acac), and btp
2
Ir(acac) phosphorescent dopants (the circled
structures on the left) are marked with colored arrows. The CIE
coordinates of the phosphorescence spectra of the rest of
C^N
2
Ir(acac) complexes are also shown in square boxes on the CIE
diagram. The NTSC standard coordinates for the red, green, and
blue subpixels of a CRT are at the corners of the black
triangle(Lamansky, Djurovich et al. 2001) ....................................................... 8
Figure 3 Photoluminescence spectra of F
2
-ppyPt(acac) doped films
showing the spectral lineshape dependence on the doping level. The
spectra consist of aggregate and monomer emission components. At
doping concentration of 5.6%, the F
2
-ppyPt(acac) monomer-to
aggregate ratio in the film is balanced to produce white light.
Chemical structures of F
2
-ppyPt(acac) and its dimer are shown on
the right(Adamovich, Brooks et al. 2002) ...................................................... 14
Figure 4 Absorption spectrum of a dilute BDF solution in toluene ........................ 25
Figure 5 Cyclic voltammogram for decamrthylferrocene with ferrocene
as an internal reference (in DMF) ................................................................... 28
Figure 6 Cyclic voltammogram for BDF with decamrthylferrocene as an
internal reference (in DMF) ............................................................................ 29
Figure 7 Differential pulse voltammogram for ferrocene,
decamethylferrocene, and bis (diphenyl) ferrocene (BDF) in DMF ............... 29
Figure 8 Data for the devices with the structure
ITO/BDF(x)/NPD(200Å)/AlQ3(400Å)/LiF(10Å)/Al(1200Å), x =
50, 200, 500 Å. ................................................................................................ 33
Figure 9 Devices of the structure ITO/xHTL(300 Å)/NPD(100
Å)/AlQ3(400 Å)/LiF(10 Å)/Al(1200 Å), where xHTL = NPD, BDF,
or BDF:F4-TCNQ (4% wt) ............................................................................. 35

vii
Figure 10 Devices of the structure ITO/xHTL(300 Å)/NPD(100
Å)/AlQ3(400 Å)/LiF(10 Å)/Al(1200 Å), where xHTL = NPD, BDF,
or BDF:F4-TCNQ (25% wt) ........................................................................... 37
Figure 11 NPD/AlQ
3
devices with the AlQ
3
layer doped with DBC (15%
wt) at the cathode interface. The undoped reference device contained
a neat AlQ
3
layer of 50nm thickness. The structure of the DBC-
doped device is the inset on the left graph ...................................................... 45
Figure 12 NPD/AlQ
3
/BCP devices with the BCP layer doped with DBC
(15% and 25% wt) at the cathode interface. The undoped reference
device contained a neat AlQ
3
layer of 20nm thickness ................................... 46
Figure 13 SC5:DBC devices. The architecture of the device with SC5
EIL doped with 15% of DBC is shown on the left. The undoped
reference device contained a neat SC5 layer of the same thickness ............... 47
Figure 14 2.2.2-cryptand (left) and its complex with Li
+
(right), according
to B3LYP/6-31G* geometry optimization ...................................................... 51
Figure 15 SC5:2.2.2-cryptand devices. The architecture of the device
with SC5 EIL doped with 15% of the cryptand is shown on the left.
The undoped reference device contained a neat SC5 layer of the
same thickness ................................................................................................ 53
Figure 16 Devices with the structure
ITO/NPD(50nm)/AlQ
3
(30nm)/AlQ
3
:2.2.2-cryptand 15%
(20nm)/LiF(1nm)/Al(120nm). The reference undoped device
contained a neat AlQ
3
layer of 50nm thickness. The EL spectrum of
the doped device is shown as an inset on the left graph ................................. 55
Figure 17 Two isomers of diphenyl-1,4-distyrylbenzene ....................................... 60
Figure 18 Spirofluorene compound described by Chao et al(Chao, Lin et
al. 2005) as an OLED neat blue emitter .......................................................... 62
Figure 19 Derivatized fluorenes used as neat blue emitters for
OLEDs(Tao, Peng et al. 2005) ........................................................................ 63
Figure 20 Normalized electroluminescence spectra of the OLEDs
containing emissive layers of neat fluorenes(Tao, Peng et al. 2005) .............. 64
Figure 21 Room temperature solution emission, absorption and excitation
spectra of DPAVB (on the left) in DMF and 2-
Methyltetrahydrofuran (2MeTHF, measured in this work). ........................... 71
Figure 22 Emission and excitation spectra of DPAVB in 2MeTHF
measured at 77K. ............................................................................................ 72

viii
Figure 23 Device structures(Palilisa, Ma¨kinen et al. 2003) .................................. 77
Figure 24 Photoluminescence spectra of the films of NPD (closed
circles), PPSPP (closed squares), NPB:PPSPP (molar ratio 1:1)
(closed triangles), and EL spectra of bilayer and three-layer devices
based on PPSPP as the emissive/electron transport layer (open
circles) and PPSPP and PyPySPyPy as the emissive and electron
transport layers (open squares), respectively(Palilisa, Ma¨kinen et al.
2003). .............................................................................................................. 77
Figure 25 A schematic of a white OLED that consists of blue fluorescent
and red and green phosphorescent emitters doped into a common
host (shown in gray color). ............................................................................. 81
Figure 26 Chemical structures and valence orbital pictures for two
benzofluoranthene isomers BbF and BkF according to PM3
calculation. ...................................................................................................... 91
Figure 27 Room temperature excitation (empty circles), absorption (filled
circles), and emission (filled squares) spectra of BbF .................................... 92
Figure 28 Excitation (empty circles), absorption (filled circles), and
emission (filled squares) spectra of BbF recorded at 77 K ............................. 93
Figure 29 I-V characteristics (open squares), and luminance-voltage plot
(filled diamonds) of the OLED with BbF-doped emissive layer (see
text for the device structure) ........................................................................... 94
Figure 30 External quantum efficiency (open circles), and
electroluminescence spectra taken at different voltages (inset) for the
OLED with BbF as an emissive dopant .......................................................... 95
Figure 31 BbF devices with 100 Å layer of mCP between the hole-
transporting (NPD) and emissive (CBP:BbF) layers. ..................................... 96
Figure 32 Synthesis of N-substituted indolocarbazoles. ......................................... 98
Figure 33 The indolocarbazoles on the left were successfully prepared
using iodoarenes with sterically non-hindered ortho-positions
relative to iodine. Iodoarenes with fully-substituted carbons at 2,6-
positions showed no reactivity towards coupling with
indolocarbazole resulting in failed syntheses of the derivatives
shown on the right. ........................................................................................ 102
Figure 34 TBIC synthesis ..................................................................................... 103
Figure 35 Solution spectra of three indolocarbazole derivatives at room
temperature ................................................................................................... 104

ix
Figure 36 Solution spectra of three indolocarbazole derivatives at 77 K ............. 105
Figure 37 Absorption, emission (RT and 77 K) and excitation spectra of
MeIC in solution ........................................................................................... 106
Figure 38 Current-voltage characteristics, external quantum efficiencies,
luminance, and electroluminescence spectra (recorded at 10 Volts)
for the devices with indolocarbazole-doped emissive layers (see text
for device structures). .................................................................................... 108
Figure 39 Change in fluorescent lifetime and quantum yield upon bubble-
aerating the degassed solution of DMIC in toluene; time between 1
st

and last recorded points is approx. 1 hour .................................................... 111
Figure 40 Change in fluorescent lifetime and quantum yield for DMIC
upon N
2
-degassing the toluene solutions. The graph on the left was
recorded for excitation wavelength of 340nm, the one on the right –
383nm ........................................................................................................... 112
Figure 41 MeIC absorption spectrum in toluene .................................................. 113
Figure 42 Differences in absorption and excitation for the same solution
of NPIC in toluene. The two spectra were normalized to the height
of the 340 nm peak. ....................................................................................... 119
Figure 43 Extinction coefficients of ICs normalized to the high energy
absorption peak of NPIC. .............................................................................. 122
Figure 44 Emission spectrum of NPIC in 2MeTHF recorded at 77 K. The
phosphorescence peak with the λ
max
= 488 nm is shown on the inset. ......... 123
Figure 45 Absorption, emission (RT and 77 K) and excitation spectra of
DMIC in solution .......................................................................................... 125
Figure 46 Emission decay plot for DMIC neat thin film recorded at room
temperature ................................................................................................... 126
Figure 47 Emission spectra of neat solids and thin films of DTIC, MeIC,
and NPIC. The red-shifted broad emission and the corresponding
excitation spectra for DTIC are shown on the bottom-right. ........................ 127
Figure 48 Normalized room temperature emission spectra of TBIC in the
neat film, doped CBP film, and neat crystalline solid. A life-time
plot for the neat crystalline sample is shown as an inset .............................. 130

x
Figure 49 Current-Voltage and Brightness-Voltage plots (left), and
external quantum efficiency and EL spectrum (right) for the device
with the structure ITO/NPD(400Å)/CBP:TBIC
5%(250Å)/BCP(150Å)/AlQ3(250Å)/LiF(10Å)/Al(1200Å) ......................... 132
Figure 50 Chemical structures of the ultra-high band gap hosts and
phosphorescent emitters discussed in this Chapter ....................................... 137
Figure 51 Photoluminescence spectra from thin films with BtpIr as an
emitter doped into 4PA and CBP hosts, and the solution
photoluminescence of BtpIr at room temperature ........................................ 139
Figure 52 The device architecture. UGH is either tetraphenyladamantane
(4pa) or (p-bis(triphenylsilyl)benzene) (UGH2). .......................................... 140
Figure 53 Current-voltage plots for the devices containing emissive
layers of BtpIr doped into 4PA. Device structures:
ITO/NPD(40nm)/EBL(x)/4PA:BtpIr 7%
(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm), EBL = mCP(10nm)
(red), Irppz(10nm) (blue), or no blocking layer (black). .............................. 141
Figure 54 Brightness-voltage data for the devices containing emissive
layers of BtpIr doped into 4PA. Device structures:
ITO/NPD(40nm)/EBL(x)/4PA:BtpIr 7%
(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm), EBL = mCP(10nm)
(red), Irppz(10nm) (blue), or no blocking layer (black). External
quantum efficiencies are shown as an inset. ................................................. 142
Figure 55 Electroluminescence spectra of BtpIr/4PA devices with no
electron-blocking layer. Device structure:
ITO/NPD(40nm)/4PA:BtpIr
7%(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm) ............................................. 143
Figure 56 Electroluminescence spectra of BtpIr/4PA devices with an
mCP electron-blocking layer. Device structure:
ITO/NPD(40nm)/mCP(10nm)/4PA:BtpIr
7%(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm) ............................................. 143
Figure 57 Electroluminescence spectra of BtpIr/4PA devices with an
Irppz electron-blocking layer. Device structure:
ITO/NPD(40nm)/mCP(10nm)/4PA:BtpIr
7%(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm) ............................................. 144

xi
Figure 58 Energy levels (in electronvolts) for the 4PA:BtpIr device. The
dashed box corresponds to the energy levels of mCP. The HOMO
values were obtained using UPS(D'Andrade, Datta et al. 2005); the
available LUMO values are taken from the work of Adamovich et
al(Adamovich, Brooks et al. 2002) ............................................................... 145
Figure 59 I-V, brightness-voltage, electroluminescence spectrum, and
quantum efficiency plots for the
ITO/NPD(40nm)/Irppz(10nm)/UGH2:BtpIr,
7%(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm) devices ................................ 146
Figure 60 Current-voltage and brightness-voltage plots (on the right), and
the E. Q. E. plot (on the right) for the devices with the structure
ITO/NPD(40nm)4PA:PQIr, 7%
(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm). .................................................. 147
Figure 61 Electroluminescence spectra of the OLEDs with emissive
layers consisting of BtpIr doped into 4PA, UGH2, and CBP (7%
doping by weight).
a
CBP:BtpIr EL data taken from Lamansky et
al(Lamansky, Djurovich et al. 2001). The devices structures
correspond to the one shown on Figure 3. .................................................... 148
Figure 62 Electroluminescence spectra of the devices with PQIr doped
into 4PA and CBP hosts (device structures ITO/NPD(40nm)/4PA:
PQIr 7% (25nm)/BCP(30nm)/LiF(1nm)/(120nm);
ITO/NPD(40nm)/CBP: PQIr 7%
(25nm)/BCP(15nm)/AlQ
3
(25nm)LiF(1nm)/(120nm) and in fluid
dichloromethane solution. ............................................................................. 150

xii
Abstract
Organic light-emitting devices (OLEDs) with two-component layers
consisting of a guest material (dopant) dispersed in a matrix (host) are examined in
this work. Both emissive organic materials as dopants and non-emissive
conductivity dopants are discussed. For the conductivity doping, lithium-
coordinating agents including cryptands and crown ether derivatives have been
used as dopants for electron-transport layers to improve the device operating
voltages and luminance. Devices with oxidatively doped hole-transport layers
consisting of novel hole-transporting ferrocene derivatives are also described. For
the emissive doping, the spectral line shape dependence on the varied host
materials for the emission of phosphorescent iridium complexes is examined. It
has been shown that the value of Huang-Rhys parameters for the phosphorescence
spectra decreases several times when non-polar ultra-high-band-gap hosts (UGHs)
are used, compared to conventional carbazole-based host materials. Finally, a new
class of efficient blue fluorescent dopants specifically designed as blue
components for the white OLEDs with separate channels of harvesting blue singlet
and red and green triplet excitons has been characterized.

xiii
Table of Contents
Acknowledgements ii
List of Tables v
List of Figures vi
Abstract xii
Chapter 1. Introduction 1
Electroluminescence of organic materials; principles of OLED operation 1
OLEDs with two-component layers; Doping 4
Phosphorescent OLEDs 6
Redox dopants to increase conductivity of charge transport layers 10
White OLEDs 11
OLEDs in RGB displays 17

Chapter 2. Ferrocene derivatives as materials for OLED hole-transporting layers 19
Introduction 19
Metallocenes: redox-tunable systems 19
Doped HTLs 21
Experimental 23
Synthesis of ferrocene derivatives 23
Materials characterization 24
OLEDs fabrication 25
Results and discussion 26
Materials synthesis 26
Electrochemistry 27
Ferrocenes for OLED hole-transporting materials 31
F4-TCNQ-doped BDF devices 34
Ferrocenes: potential HTLs for red and NIR emitters 38
Conclusions 39

Chapter 3. Lithium-chelating dopants for the electron-transport layers of OLEDs
with LiF/Al cathodes 40
Introduction 40
Experimental 42
Results and Discussion 42
Conclusion 56

xiv

Chapter 4. Blue fluorescent materials for WOLEDs with separate channels for
triplet and singlet exciton harvesting 58
Introduction 58
Overview of blue fluorescent emitters for OLEDs 59
Mixed fl/ph WOLEDs 80
Experimental 82
Results and discussion 87
Fluoranthenes 89
Indolocarbazoles 97
Blue OLEDs with IC dopants as emitters 107
Conclusions 132

Chapter 5. Spectral line shapes of iridium cylometalated complexes doped into
ultra-high band-gap hosts (UGHs) 134
Introduction 134
Experimental 136
Results and discussion 137
Conclusions 151

References 152

1
Chapter 1. Introduction
Electroluminescence of organic materials; principles of OLED
operation
Electroluminescence of organic molecules has been a well-known phenomenon
for more than 50 years (Fromhold Jr. 1963; (Pope, Kallmann et al. 1963). However, it
wasn’t until the late 1980’s that it became promising for practical use. Successful
application of organic luminescence in light-emitting devices required device structures
that overcame the problems associated with the high resistivity of organic materials, and
achieved a well-balanced charge injection from the electrodes into organics. These two
problems were solved by Tang and van Slyke (Tang, VanSlyke 1987) with the thin film
heterostructure concept for the organic LEDs (OLEDs). Figure 1 shows a schematic of a
double heterostructure OLED consisting of three organic layers sandwiched between the
electrodes.

2
anode on substrate
+
- cathode
ETL
HTL
EML

Figure 1 A schematic of a double heterostructure OLED consisting of a hole
transport layer (HTL), electron transport layer (ETL), emissive layer (EML), and the
electrodes

The organic layers adjacent to the cathode and anode are the electron transport
layer (ETL), and the hole transport layer (HTL), respectively. When the electric field is
applied to the device, the molecules of the organic material adjacent to the anode are
oxidized thus forming holes (i.e., molecules with a missing electron); conversely, the
material adjacent to the cathode gets reduced; the reduced molecules of the corresponding
ETL material are referred to as electrons.
Holes and electrons injected from the opposite electrodes migrate to the emissive
layer (EML) where they recombine to form excitons. The organic solids comprising the
OLED are composed of weakly interacting molecules. Unlike typical inorganic
semiconductors, the OLED materials possess relatively low dielectric constants, and the
excitons formed in organic solids are Frenkel-type excitons binding energies of which are

3
much larger than that of inorganic semiconductors, whereas the radii are much smaller. In
fact, the excitons in organic films exist in a form of excited states of the molecules.
Excitons migrate through the organic layers by energy transfer and eventually relax to the
ground state through either emission of a photon or non-radiative processes.
For the highly resistive organic films, the film thicknesses of 500 Å or less is
essential for lowering drive voltages to the 5-10 V level. Since electrons and holes in the
organics have different mobilities (the mobility of holes higher than that of electrons,
opposite to inorganic semiconductors), in a single-layer device the current is mainly
conducted by only one type of charge carriers, which results in efficiency losses: the
majority of carriers generated at the electrode-organic interface do not recombine to form
excitons. Thus, separate hole and electron conducting layers provide efficient charge
injection and carrier recombination.
Charge conduction in organic device materials occurs through a hopping
mechanism. “Charge hopping” is an outer sphere electron-transfer reaction between
charged and a neutral species that proceeds through an energy barrier referred to as
reorganization energy. The theory of this process was first developed by R. Marcus
(Marcus 1956; (Marcus 1993). The Marcus theory predicts the mobility of charge carriers
in organic solids based on the amount of reorganization energy involved in the process of
electron transfer: the higher the barrier the poorer the charge conductance. The height of
the barrier is determined by the degree of structural distortion the carrier molecules

4
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. Triarylamines satisfy these requirements, and they were first to be used
in the OLEDs as HTL materials (Tang, VanSlyke 1987). Among the classes of
compounds that also meet this structural requirement but have never been examined as
HTL materials are ferrocenes. Chapter 2 of this Thesis will discuss hole-transport
properties of ferrocenes.
OLEDs with twocomponent layers; Doping
Shortly after the introduction of thin film heterostructure based OLEDs it was
demonstrated that two-component emissive layers consisting of emitter molecules doped
into a host matrix increased the device electroluminescent efficiency. This is attributed to
improving charge recombination and exciton confinement in the emissive layer of the
OLED, as well as eliminating self quenching of the emitting dopants (Tang, VanSlyke et
al. 1989).
Typically, a doped emissive layer contains 5 molar % or lower of the emitter
uniformly dispersed throughout the host matrix. The opposite charges are injected into
HTL and ETL layers from the two electrodes in a similar fashion to an undoped device.
When holes and electrons have migrated to the interfaces with the doped emissive layer,
the electron transfer reaction occurs between the molecules of these charge transport
layers and those of the host material. So for a device with a doped emissive layer, charge

5
recombination and exciton formation occurs on the host. The process is summarized in
the following scheme:

HTL
+•
+ Host° → HTL° + Host
+•
, (1)
ETL
-•
+ Host° → ETL° + Host
-•
, (2)
Host
+•
+ Host
-•
→ Host* + Host°. (3)

Here, HTL
+•
and ETL
-•
are a hole on an HTL and an electron on an ETL
materials, respectively; Host* is an exciton formed on a host molecule, and “°” denotes
ground states of the species involved.
Initially formed on a host molecule, the exciton then gets transferred onto a guest
emitter molecule, which then radiatively relaxes to the ground state to produce a photon
of light:

Host* + Guest° → Host° + Guest*, (4)
Guest* → Guest + h ν (5)

6
Two distinct mechanisms of energy transfer (4) have been outlined in literature
(Dexter 1953; (Gilbert, Bagot 1991). When an excited host molecule (the donor) relaxes
to its ground state, the released energy can be transferred to the acceptor molecule
through dipole-dipole interactions. This is called Förster mechanism. It is characterized
by long distances through which such energy transfer can proceed: up to 100Å in some
favorable cases. The Dexter mechanism, on the other hand, occurs between molecules
separated by short distances of the dimension of one molecule or shorter, and constitutes
direct electron transfer between the donor and acceptor. Whereas the Förster mechanism
has an advantage of being efficient through large distances in the molecular solid, the
Dexter mechanism is the only way the energy can be transferred between molecules
which differ in spin multiplicities.
As an alternative to energy transfer, an exciton formation on the guest emitter
molecule can result from direct charge trapping by the emitter. This mechanism is, in
general, more efficient than energy transfer since the latter isn’t lossless.
Phosphorescent OLEDs
The hole and electron in OLEDs are odd electron species with an equal
distribution of m
s
= ± ½. Thus, when the hole and electron recombine to form the exciton
(such as in Equation 3), a statistical mixture of singlet and triplet excitons are generated
(Baldo, O'Brien et al. 1999; (Segal, Baldo et al. 2003). This leads to a population of
excitons that is 25% singlet and 75% triplet and has a profound effect on OLED

7
efficiency. Most of the emitting dopants developed for OLEDs prior to the late 1990’s
emit from fluorescent states, which only utilize the singlet fraction of formed excitons
(Shoustikov, You et al. 1998). This limits the internal quantum efficiency of
fluorescence based devices to 25%, corresponding to an external efficiency (front face) of
only about 5%. In the late 1990’s a new family of emissive dopants was introduced that
gave marked increases in OLED efficiency. The key to this enhanced efficiency was the
recognition that the triplet exciton fraction is more important than the singlet. Efficient
harvesting of triplet excitons requires a phosphorescent dopant, which will trap both
singlet and triplet excitons. An added requirement for the phosphorescent dopant is that
it should have a radiative lifetime comparable to the RC time constant of the OLED,
which is typically in the microsecond time scale. The best way to achieve both high
phosphorescence efficiency and a radiative lifetime on the order of microseconds is to
incorporate a heavy metal atom into the dopant, whose spin orbit coupling will efficiently
promote intersystem crossing between singlet and triplet states. The most commonly
used metal for this purpose is Ir, however, efficient phosphorescent dopants have been
also prepared with other heavy metals as well, including Pt, Ru, Re, Au and Os.
Figure 2 shows the structures of a number of Ir based organometallic dopants and
the CIE color coordinates (CIE = Commission Internationale de L’Eclairage) for many of
them. OLEDs have been prepared with the four circled dopants and their CIE
coordinates labeled with colored arrows.

8

Figure 2 Chemical structures, the Commission Internationale de L’Eclairage
(CIE) chromaticity coordinates of OLEDs and phosphorescence spectra of iridium
cyclometalated complexes. The CIE coordinates for OLEDs with the ppy
2
Ir(acac),
bt
2
Ir(acac), pq
2
Ir(acac), and btp
2
Ir(acac) phosphorescent dopants (the circled structures
on the left) are marked with colored arrows. The CIE coordinates of the phosphorescence
spectra of the rest of C^N
2
Ir(acac) complexes are also shown in square boxes on the CIE
diagram. The NTSC standard coordinates for the red, green, and blue subpixels of a CRT
are at the corners of the black triangle (Lamansky, Djurovich et al. 2001)

Since the first introduction of Ir based phosphors in OLEDs (Baldo, Lamansky et
al. 1999), close to 200 different Ir complexes have been incorporated into OLEDs, most
giving external efficiencies of 8% or greater (Crabtree, Mingos et al. 2007). Several

9
groups have reported the use of Ir based materials in optimized devices can give external
efficiencies > 20%, corresponding to internal efficiencies of close to 100% (Tsuboyama,
Iwawaki et al. 2003; (Watanabe, Agata et al. 2005; (Meerheim, Walzer et al. 2006).
The emission energy for organometallic phosphors are closely related to the
structure of organic ligands, making it possible to design a series of efficient
phosphorescent emitters that covers most of the visible spectrum (Brooks, Babayan et al.
2002; (Li, Djurovich et al. 2005). The metal center of the complex is also a variable for
tuning the emission properties of the complex, which can be used to fine tune the
emission energy. The emission from a transition metal complex originates from its
lowest energy triplet excited state. Spectroscopic analysis shows that this state is
predominantly localized on the cyclometalating ligands, mixed with singlet metal-to-
ligand charge transfer (
1
MLCT) character. Modification of ancillary (“non-emissive”)
ligands affects the energy of the metal orbitals and thus the amount of
1
MLCT character
in the excited state. Varying the ratio of ligand centered to
1
MLCT character directly
affects the energy of the mixed excited state (Li, Djurovich et al. 2005). Thus, with
modification of the ancillary ligand in (F
2
ppy)
2
Ir(L^X) complexes (L^X = ancillary
ligand) it is possible to shift the emission energy of the complex from 458 to 512 nm.
One of the deep blue complexes of this series [(F
2
ppy)
2
Ir(pz
2
Bpz
2
)] has been used to
fabricate OLEDs which give external efficiencies > 11% (Holmes, D'Andrade et al.
2003).

10
Redox dopants to increase conductivity of charge transport layers
So far, the concept of doping was described here in a context of enhancing
luminescence efficiency of the device that elimination of concentration quenching brings
about when an emitter is diluted within the host matrix. As pointed out above, resistivity
of organic films is one of the problems that need to be addresses to improve the OLEDs
operational characteristics. High resistivity means higher operating voltages, hence faster
degradation and lower power efficiency of the device. Doping charge transport layers
with non-emissive redox dopants can affect the number of charge carriers available to
conduct electric current through the device. With that in mind, materials with sufficiently
high oxidizing ability have been developed that can be used as dopants for hole transport
layers. Employing such dopants allowed for up to an order-of-magnitude decrease in
turn-on voltage for devices with certain HTL materials compared to the undoped devices
(Blochwitz, Pfeiffer et al. 1998).
A similar concept can be envisioned to improve the conductivity of electron
transport layers by doping them with reducing agents. It appears however that organic n-
dopants are more difficult to design; instead, the most common materials for ETL n-
doping have been alkali and alkali-earth metals (Johansson, Osada et al. 1999;
(Piromreun, Oh et al. 2000; (Parthasarathy, Shen et al. 2001; (Zhou, Pfeiffer et al. 2002).
Their use is not free from drawbacks such as high diffusion rates of M
+
-ions within the

11
film resulting in the progressing in time emission quenching, so the search for suitable
organic/organometallic dopant for ETLs will continue.
A new class of ETL dopants that targets the already mentioned problem of
metallic dopant diffusion is the alkali-metal chelating complexes such as crown ethers
and cryptands. In Chapter 3 of this Thesis the use of these dopants in conjunction with
alkali-metal doping will be discussed in a context of enhancing electron injection from
the cathode into OLED ETLs.
White OLEDs
An important potential application for LEDs is in illumination. The requirements
for devices that serve as illumination sources are somewhat different than the
monochromatic OLEDs described above. OLEDs targeted for RGB displays have to give
electroluminescent spectra with a relatively narrow lineshape centered on the peak
wavelength. On the other hand, an illumination source is meant to approximate the
blackbody solar spectrum and needs to have a broad lineshape with roughly equal
intensity across the entire visible spectrum. Therefore, in order to attain complete
coverage across the visible spectrum, an OLED used for illumination purposes typically
employs multiple emitters are that are either co-deposited into a single emissive layer or
distributed into different layers or regions of the device. A number of the different device
architectures have been reported to achieve efficient white electroluminescence.

12
Most white organic LEDs (WOLEDs) utilize luminescence from several different
colored emitters such that the combined output covers the visible spectrum uniformly.
While WOLEDs with less than three distinct emitters have been reported, the most
common approach in WOLEDs is to use three, i.e. blue, green and red. One of the
simplest device architectures involves mixing blue, green and red dopants into a single
emissive layer, such that the sum of the three emission spectra covers the visible
spectrum (Kido, Shionoya et al. 1995; (Tasch, List et al. 1997; (Kawamura, Yanagida et
al. 2002). The use of phosphorescent emitters in a triple doped emissive layer can then
lead to highly efficient devices. However, using three dopants in a single layer is
problematic because energy readily transfers from the higher energy blue dopant to the
green dopant and from the green dopant to the red dopant. Therefore, careful adjustment
of the concentration of each dopant is required to achieve a well-balanced emission color,
with doping levels in the ratio blue > green >> red. In order to get well-balanced white
emission, the doping level of the red dopant typically needs to be well below 1%.
One solution to the inter-dopant energy transfer problem is to segregate the dyes
into different layers. Efficient WOLEDs have been prepared using this stacked concept
with either fluorescent (Kido, Shionoya et al. 1995; (Jiang, Zhang et al. 2000; (Ko, Tao
2001) or phosphorescent emitters. More simplified structures have also been described
that use dual component fluorescent blue and orange emitters doped into separate layers
(Jiang, Zhang et al. 2000; (Yang, Jin et al. 2000). While stacking the emitters in separate

13
layers eliminates these energy transfer problems, the device architecture can become
significantly more complicated due to difficulties in achieving balanced carrier
recombination and exciton localization within each of the emitting layers.
The use of planar platinum based dopants makes it possible to prepare a
broadband emitting (white) OLED with only a single dopant, contrary to approaches
described above, which utilized two or three different emitters. Figure 3 shows how the
white color is achieved by combining the emission from the monomer (blue) and the
aggregate (yellow-to-red) of the same organometallic platinum complex, giving an
emission spectrum that covers the full range of visible wavelengths.

14

Figure 3 Photoluminescence spectra of F
2
-ppyPt(acac) doped films showing the
spectral lineshape dependence on the doping level. The spectra consist of aggregate and
monomer emission components. At doping concentration of 5.6%, the F
2
-ppyPt(acac)
monomer-to aggregate ratio in the film is balanced to produce white light. Chemical
structures of F
2
-ppyPt(acac) and its dimer are shown on the right (Adamovich, Brooks et
al. 2002)

The ratio of the monomer-to aggregate emission is controlled by both the doping
concentration and the steric bulk of the dopant (Adamovich, Brooks et al. 2002).
Increasing the steric bulk on the dopant impedes the aggregate formation, whereas
increasing the dopant concentration favors it. Minimizing the number of dopants

15
significantly reduces the complexity of the device. It has recently been shown that
devices based on the monomer-aggregate approach to broadband emission can be used to
achieve external efficiencies of 15-20% (Cocchi, Kalinowski et al. 2007; (Williams,
Haavisto et al. 2007).
A significant advance has recently been reported for phosphorescent based
WOLEDs. A WOLED in which blue, green and red phosphors are used to generate a
broad spectrum white OLED have been prepared (Nakayama, Hiyama et al. 2007). The
device gave an efficiency of 64 lm/W at a brightness of 1000 cd/m
2
. This efficiency
exceeds compact fluorescent sources and is close that of fluorescent tube sources (ca. 75-
90 lm/W). Moreover this device gave a device lifetime of greater than 10,000 hours at
this brightness. These values are more than a factor of two high that the previous records
for OLEDs and clearly show that OLEDs have a bright future in lighting.
White light is composed of roughly 25% blue, with the balance covering the
energies between green and red. It also happens that the excitons, reformed on
recombination of the hole and electron in the OLED, are formed in a ratio of 25%
singlets to 75% triplets. The similarity in blue fraction of white light and the singlet
fraction suggests an alternative approach to achieving high efficiency white emission:
couple the singlet excitons to a blue fluorescent dopant and the triplets to phosphors
covering the green and red portions of the spectrum. Such an implementation of
combined fluorescent and phosphorescent emission has proven to have a number of

16
advantages. Introduction of the stable fluorescent blue is expected to alleviate a well-
known problem of limited operational life times of blue components of WOLEDs. The
shape of the quantum efficiency-current density plots of the phosphorescent three-
component WOLED, typically shows a steep roll-off of the efficiency curve at higher
current densities soon after the efficiency achieves its maximum (Sun, Giebnic et al.
2006). Triplet-triplet annihilation responsible for the unwanted efficiency decrease at
high currents is decreased in this device, since the triplets are at lower concentration in
the middle of the emissive layer than near the ETL or HTL interface where they are
formed.
Since these “hybrid” fluorescent/phosphorescent devices contain a blue
fluorescent component, they face the problem associated with any blue emitters for
OLED: finding an emitter with high efficiency, color purity, and operation stability is for
the blue is more difficult than for other color components. Along with these general
requirements, a blue component for fl/ph WOLEDs will need to satisfy other conditions
related to specific molecular energy level alignments to ensure transfer of singlet excitons
formed on the host onto a fluorescent blue emitter, and migration of the triplets to the
green and red phosphors. Why it is important for such an emitter to have its excited
triplet and singlet energy levels to have small separation, also the systematic way of
approaching the search of such materials, and our findings on indolocarbazoles, the
materials with small ΔE(S-T), will be looked at in Chapter 4 of this work.

17
OLEDs in RGB displays
Flat panel RGB displays are one of the primary applications of OLEDs. In the
field of flat displays, there is no competition to OLEDs from silicon since the latter is a
very poor emitter. The III-V inorganic semiconductors can make very bright and robust
LEDs, however, besides poorer color tunability, the cost of available technology
precludes them from being used for large-area displays: manufacturing cost per one pixel
is two orders of magnitude higher for III-V semiconductors that for LCD-based displays.
Thus the major competitor to OLED technology in the field of full color displays are
LCDs. Advantages of power efficiency, front-emission allowing for direct viewing
(opposed to backlit LCDs), and prospect of compatibility with flexible plastic substrates
will allow OLEDs compete with LCDs in this area.
The requirements for the spectral line shapes OLEDs need to satisfy in RGB
displays and discussed above white light sources for illumination are opposite. For an
illumination source even when the white light is to be composed from RGB components
each emitter needs to have as broad lineshape as possible to ensure good overlap so that
the resulting WOLED emission matches the solar spectrum. Contrary, the emitting
components for a full-color display are required to have narrow lineshapes for maximum
color purity. The dependence of an emission maximum on the host matrix it is doped into
is a known phenomenon, and was used in color tuning of electroluminescence (Bulovic,
Deshpande et al. 1999). The role that a host material plays in changing the lineshape of

18
the dopant emission, and properties of the host which such lineshapes are most sensitive
to, will be discussed in Chapter 5 of this Thesis with specific examples on OLEDs with
red phosphorescent emitters doped into several hosts.

19
Chapter 2. Ferrocene derivatives as materials for OLED
holetransporting layers
Introduction
Metallocenes: redox-tunable systems
Since the discovery of ferrocene (Kealy, Pauson 1951; (Wilkinson, Rosenblum et
al. 1952) a number of its properties promising for various applications in organic
electronics have been found. Among such properties are high thermal and chemical
stability that’s important for device operational lifetimes; relative ease of derivatization
and synthesis of complex ferrocene-containing structures; reversible and robust redox
couple (only weakly perturbed by solvent and other environmental changes); redox
tunability through the substitution in the cyclopentadiene rings; small reorganization
energy upon electron transfer due to small structural differences between the neutral and
oxidizes states.
One of the important characteristics of materials for organic electronic devices is
the ability to conduct charge. A direct relationship was established between the rates of
electron transfer and charge conduction in organic films (Nitzan 2001). Due to fast
electron transfer, one would expect efficient charge conductance in solid ferrocene.
Indeed, the electrical conductivity of ferrocene films was found to be much higher than

20
that of many molecular organic solids (Bradley, Hammes 1963; (Mallik, Bhattacharjee
1989; (Bhattacharjee, Mallik 1990; (Togni 1995; (Nath Bera, Mallik 2003). Adding
ferrocene groups turns otherwise insulating polymers into semiconductors; the latter
effect observed for polyvinylferrocene was proven to be enhanced by doping with ClO
4-
-
anions (Pittman, Surynarayanan et al. 1976; (Shirota, Kakuta et al. 1984; (Hunter, Tyler
et al. 1987; (Mariani, Abruna 1987). Moreover, theoretical studies of transport properties
suggest that there is a significant metallic character of the conduction through ferrocene
(Uehara, Belosludov et al. 2006; (Uehara, Igarashi et al. 2006). Due to their high
conductance and efficient electron transfer, ferrocene-containing systems were a subject
of investigation as components of conductive polymers and electron-transfer mediators in
polymeric bioorganic materials (Horwitz, Suhu et al. 1992; (Tyutnev, Saenko et al. 2001;
(Zhang, Wan et al. 2004; (Getty, Engtrakul et al. 2005; (Winther-Jensen, Chen et al.
2005; (Xiao, Brune et al. 2006; (Brisset, Navarro et al. 2007; (Carolan, Forster et al.
2007; (Farre, Spinelli et al. 2007; (Radhakrishnan, Paul 2007).
It had been understood that the unsubstituted Cp
2
M species have a limited use for
thin film devices because of the difficulties associated with device manufacturing: the
sublimation temperature of ferrocene under reduced pressure is ca. 50°C. This high
volatility makes it impossible to thermally deposit thin films under vacuum. In the
attempt to decrease the volatility, bulky substituents were introduced into the Cp-rings of
ferrocene. The redox potential of one of the synthesized ferrocene derivatives 1,1’-

21
bis(biphenyl)ferrocene (BDF) was found to be the same as that of ferrocene. BDF had
much lower volatility than ferrocene optimal not only for thermal deposition of thin films
in vacuum but also suitable for ultraviolet photoelectron spectroscopy (UPS)
measurements.
To the best of our knowledge ferrocene or its derivatives haven’t been
investigated as hole-transport materials for organic devices. 1,1’-bis(biphenyl)ferrocene
first discussed in our published work (D'Andrade, Datta et al. 2005) possesses low
enough volatility to be used in vacuum thermal deposition of thin films. Here in this
chapter, its properties as a hole-transporting material will be examined.
Doped HTLs
Resistivity of organic materials is high compared to that of inorganic
semiconductors. The amount of charge that passes through the material depends on
several factors, and among those is charge carrier concentration. Holes are quasi-particles
that constitute vacant electrons. The number of such vacancies and, hence the hole
concentration can be increased by adding acceptor molecules to the HTL. 2,3,5,6-
tetrafluoro-7,7,8,8-tetracyano-p-quinodimethane (F4-TCNQ) the chemical structure of
which is shown below is an oxidative agent that is known to reduce the turn-on voltage of
OLEDs when it is used as a dopant for HTLs (Blochwitz, Pfeiffer et al. 1998; (Zhou,
Pfeiffer et al. 2002).

22

F
F
F
F
N N
NN

When doped into an HTL composed of vanadium phtalocyanine, F4-TCNQ gives
rise to order of magnitude increased conductance, and the OLED turn-on voltage drop
from 50 V to ~5 V (Blochwitz, Pfeiffer et al. 1998).
The reduction potential of F4-TCNQ was reported as +0.53 V vs SCE in
acetonitrile solution (Wheland, Gillson 1976). This is 0.13 V higher than ferrocene
meaning that F4-TCNQ has a potential to oxidize ferrocene (and BDF which has the
same oxidation potential). With that in mind, we also looked at p-doping of the ferrocene
hole-transporting layers with F
4
-TCNQ to evaluate the effect of oxidative doping on the
charge conductance in the device and hence its performance.

23
Experimental
Ferrocene and decamethylferrocene were obtained from Aldrich and purified by
sublimation before use. Cp
2
Fe sublimation temperature under reduced pressure was
found to be +50°C. The sublimation temperature for decamethylferrocene (Me
5
C
5
)
2
Fe
was 100-105°C.
Synthesis of ferrocene derivatives
1,1’-bis(phenyl)ferrocene was synthesized from ferrocene-1,1’-diboronic acid and
bromobenzene on PdCl
2
(dppf) catalyst in DME according to a published procedure
(Knapp, Rehahn 1993).
Fe
B
OH
OH
B
OH
OH
I 2 +
PdCl
2
dppf NaOH
DME, reflux
Fe

1,1’-bis(biphenyl)ferrocene (BDF) was synthesized using the adjusted method
described by Knapp and Rehahn (Knapp, Rehahn 1993). A sample procedure: A mixture
of ferrocene-1,1’-diboronic acid (10g), 1,4-bromobiphenyl (17g), DME (275 ml), 3M
aqueous NaOH (50 ml) and PdCl
2
(dppf) (0.5g) was refluxed with vigorous stirring for 7
days. Chloroform (1000 ml) and water (200 ml) were added and the layers separated. The
organic layer was washed with water (3x500ml), dried over MgSO
4
, and the solvent was
removed under reduced pressure. The residue was purified by column chromatography

24
(silica gel, eluent: chloroform). The product was resublimed two times under high
vacuum (T
subl
~220-225
o
C). Mass after two sublimations: 1.43g.
During the course of this study one of the starting materials ferrocene-1,1’-
diboronic acid was discontinued and no longer commercially available. To continue
experiments with BDF, 1,1'-bis(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-Ferrocene
(1,1'-Ferrocenediboronic acid bis(pinacol) ester) was obtained commercially and used as
a replacement for the ferroceneboronic acid in the method analogous to the one described
above.
Fe
B
O
O
B
O
O
I 2 +
PdCl
2
dppf NaOH
DME, reflux
Fe

Materials characterization
The oxidation potentials of ferrocene derivatives were measured versus
ferrocene/ferrocenium internal reference in dilute DMF solutions using differential pulse
voltammetry at voltage scan rates of 100mV/s. The reversibility of the oxidation was
confirmed by cyclic voltammetry.
The absence of significant absorption in the ~600 nm region allowed for the
measurement of the thickness of a BDF film by ellipsometry (the BDF absorption
spectrum is shown on Figure 4).

25
400 450 500 550 600 650
-0.02
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
Absorbance (a. u.)
Wavelength, nm
200 300 400 500 600 700 800
0.0
0.5
1.0
1.5
2.0
Absorbance (a. u.)
Wavelength, nm
BDF in toluene

Figure 4 Absorption spectrum of a dilute BDF solution in toluene

The absolute thickness of a vapor-deposited BDF film determined by ellipsometry
was used in estimation of the BDF film density. The density of the film was calculated as
a ratio of the mass of the film measured by a calibrated quartz crystal monitor to the
measured absolute thickness.
OLEDs fabrication
Patterned ITO substrates were obtained from Thin Film Devices Inc.
(www.tfdinc.com) Photoresist was spin-coated onto the substrates to absorb and remove
foreign particles which are the source of pinholes and short circuits in the device. The

26
photoresist-coated substrates were baked at 90°C for 10 minutes, washed with acetone,
and sonicated in acetone for 10 min prior to UV-O
3
plasma cleaning. UV-O
3
plasma was
applied for 10 min.
F4-TCNQ (95%) was purchased from TCI America and used as received. BDF
and F4-TCNQ were deposited from double-well boats due to high volatility.
Organic light emitting devices were grown on a glass substrate pre-coated with a
∼100-nm thick layer of indium-tin-oxide (ITO) having a sheet resistance of ∼40 Ω/ .
Substrates were cleaned with solvents and then cleaned by exposure to UV-ozone
ambient for 10 minutes. After cleaning, the substrates were immediately loaded into a
thermal evaporation system operating at a base pressure of ∼1x10
-6
Torr. Device
cathodes consisting of a 10-Å-thick layer of Li or LiF followed by a 1000-Å-thick layer
of aluminum were deposited trough a shadow mask.
Results and discussion
Materials synthesis
When the ferroceneboronic acid pinacol ester was used as a starting material
instead of the boronic acid as shown on Figure AAA, the yields of BDF went down
significantly (total 1-3% yield after purification by sublimation). The main product in the
reaction of the pinacol ester with 4-halobihpenyls (bromo- and iodo-precursors behaved
similarly) was the mono-substituted ferrocene where only one of the two pinacol groups

27
was replaced by biphenyl. This mono-substituted product was isolated, purified, and
characterized by mass spectrometry and NMR. Interestingly, when attempts were made
to use this product in the reaction with iodobiphenyl, no traces of the target compound
was detected after several days of refluxing the reaction mixture:

Fe
B
O
O
I
+
PdCl
2
dppf NaOH
DME, reflux
no reaction

The fact that BDF forms in low yields from the pinacol ester, and the
monosubstituted species does not show any reactivity under the corresponding reaction
conditions may indicate trimolecular nature of the reaction of 4-iodobiphenyl with
ferrocenediboronic acid pinacol ester.
Electrochemistry
Studies of solution electrochemistry revealed that the new ferrocene derivative
BDF has the oxidation potential identical to that of unsubstituted ferrocene. This is
demonstrated on Figures 5-7:

28
-500 0 500 1000
-30
-20
-10
0
10
20
30
40
Current, μA
Voltage, mV
ferrocene + decamethylferrocene

Figure 5 Cyclic voltammogram for decamrthylferrocene with ferrocene as an
internal reference (in DMF)

29
-500 0 500 1000
-10
-5
0
5
10
15
Current, μA
Voltage, mV
decamethylferrocene + BDF

Figure 6 Cyclic voltammogram for BDF with decamrthylferrocene as an internal
reference (in DMF)
-500 0 500 1000
-6
-4
-2
0
2
4
6
decamethylferrocene +
ferrocene + BDF
Current, μA
Voltage, mV

Figure 7 Differential pulse voltammogram for ferrocene, decamethylferrocene,
and bis (diphenyl) ferrocene (BDF) in DMF

30
Referencing both the unsubstituted ferrocene and BDF against
decamethylferrocene as a common standard showed that derivatizing ferrocene with
byphenyl at each cyclopentadiene ring does not alter its oxidation potential. This
indicates that BDF and ferrocene should have similar HOMO energies. Due to high
volatility of the material, the HOMO energy of ferrocene had not been measured directly
by UPS. The corresponding HOMO value was estimated as a sum of the oxidation
potential of ferrocene and the energy of the normal hydrogen electrode (NHE) versus
vacuum (approx. -4.60 eV (Trasatti 1986)). Due to the presence of bulky groups in BDF,
its volatility appears to be low enough to allow the measurement of its HOMO energy
directly by UPS which gave the HOMO energy value for BDF of -4.76 eV (D'Andrade,
Datta et al. 2005).
Ferrocene is an important standard in organic/organometallic electrochemistry
due to its highly reversible Fc
+
/Fc redox couple, the potential of which is fairly
independent of the solvent. Furthermore, often the HOMO energy levels of OLED
materials are estimated using oxidation potentials obtained from solution
electrochemistry data. The direct method of probing HOMO energies is ultraviolet
photoelectron spectroscopy (UPS). However, cyclic voltammetry is much simpler, less
costly, and more widely available technique than the direct probing of the HOMO energy
by UPS.

31
Knowledge of the charge carrier orbital energy levels is very important for OLED
design and optimizing the device efficiency, since these energy levels determine the
charge injection barriers into the organics, and current rectification in a device (Lee,
Wang et al. 1999). HOMO and LUMO energies are used to describe the boundary
orbitals in isolated molecules. However, since organic molecular solids consist of weakly
interacting molecules, the band structure of such a solid can be described in terms of
HOMO and LUMO values. The offsets of the energy levels between the layers of
organics comprising the OLED act as potential energy barriers to the flow of charge –
and should be minimized for optimal device performance. Thus it was very desirable to
have a quantitatively accurate correlation between the solution electrochemistry data and
the UPS that would allow reliable prediction of HOMO energy levels based on solution
electrochemistry data. Obtaining a directly measured HOMO energy value for the
ferrocene established a common reference point for the electrochemically and directly
(by UPS) estimated HOMO levels of a number of organic materials (D'Andrade, Datta et
al. 2005).
Ferrocenes for OLED hole-transporting materials
As mentioned before in this Chapter, the sublimation temperature of bis(1,1’-
Phenylferrocene) (BPF) is 125°C under high vacuum. Attempts were made to use this
material for vacuum thermo-deposited devices. Even though using a high-resistance
double-well evaporation boat helped achieve reasonable evaporation rates, the attempts

32
were overall unsuccessful because the deposited BPF film, once deposited, showed fast
evaporation from the substrate into vacuum. According to the readings of a quartz
thickness monitor, the thickness of the film reduces from 400 Å to 205 Å in several
minutes. The resulting devices that contained an AlQ3 emissive layer, BPF hole-
transporting layer, and an NPD interlayer added to facilitate hole injection from BPF into
AlQ3, didn’t show any light emission.
BDF is significantly less volatile than BPF; however, double-well boats for
thermal evaporation of BDF are still preferable to achieve better control of the film
deposition rate. The data for the devices with BDF HTLs of varied thickness are given on
Figure 8

33
1 10 100
0.01
0.1
1
E.Q.E., %
Current Density, mA/cm
2
02468 10 12
1E-4
0.01
1
100
10000
Brightness, Cd/m
2
Voltage, V
0.1 1 10
1E-5
1E-3
0.1
10
1000
Current Density, mA/cm
2
Voltage, V
NPD200/AlQ400
BDF50/NPD200/AlQ400
BDF200/NPD200/AlQ400
BDF500/NPD200/AlQ400
450 500 550 600 650 700
0.0
0.2
0.4
0.6
0.8
1.0
Intensity, normalized
Wavelength, nm
BDF500/NPD200/AlQ400
9 V
8 V
7 V
6 V
10 V
11 V

Figure 8 Data for the devices with the structure
ITO/BDF(x)/NPD(200Å)/AlQ3(400Å)/LiF(10Å)/Al(1200Å), x = 50, 200, 500 Å.

The difference between the HOMOs of BDF (-4.76 eV) and HOMO of AlQ3 (-
5.65 eV) (D'Andrade, Datta et al. 2005) is -4.76 – (-5.65) = 0.89 eV. This value
corresponds to the energy barrier that holes need to overcome when injected from the
ferrocene HTL into the emissive layer. The barrier of almost 1 eV is quite high, and could
result in decreased of efficiency and high turn-on voltage of the device. In order to

34
improve hole injection into AlQ3, a thin interlayer of NPD, a material with an
intermediate HOMO energy of 5.30 eV was introduced between HTL and ETL/emissive
layers. From Figure 3 it can be seen that the BDF devices show turn-on voltage,
efficiency, and resistivity comparable to regular NPD/AlQ
3
devices for the combined
thicknesses of the hole transport layers of up to 700 Å. The slight decrease in luminance
for the device with the thickest BDF layer can be attributed to the increased resistivity,
which reflected in the I-V plot (at the total of 700 Å of the HTL, the conductance drop in
the device due to its thickness becomes noticeable, which could be expected).
F4-TCNQ-doped BDF devices
The device data that illustrates comparison between the devices with HTLs of
NPD, BDF, and BDF doped with F
4
-TCNQ (4% doping level) is shown on Figure 9.

35
024 68 10 12
0.01
1
100
10000
Brightness, Cd/m
2
Voltage, V
1 10 100
0.01
0.1
1
E. Q. E., %
Current density, mA/cm
2
024 68 10 12
0
2000
4000
6000
8000
10000
12000
Brightness, Cd/m
2
Voltage, V
02 46 8 10 12
0
200
400
600
800
1000
Current density, mA/cm
2
Voltage, V
BDF:F4TCNQ, 4% (300)/NPD(100)/AlQ
3
(400)
BDF(300)/NPD(100)/AlQ
3
(400)
NPD(400)/AlQ
3
(400)

Figure 9 Devices of the structure ITO/xHTL(300 Å)/NPD(100 Å)/AlQ3(400
Å)/LiF(10 Å)/Al(1200 Å), where xHTL = NPD, BDF, or BDF:F4-TCNQ (4% wt)

Unlike the set of devices discussed in connection to the BDF layer thickness
dependence of the conductance and brightness (Figure 8), Figure 9 shows comparison of
the devices of the same total thickness. The set of devices on Figure 9 was prepared to
compare the conductance, brightness and efficiency of the devices with NPD, neat BDF,
and F4-TCNQ-doped BDF HTLs of the same thickness. From the I-V characteristics it

36
can be seen that doping by F4-TCNQ increases the amount of current that at the same
voltage passes through the device with ferrocene-HTL compared to the undoped device.
In its turn, the undoped BDF-based device shows higher conductance than the NPD/AlQ3
device used as a control. Luminance of the devices changes accordingly with the change
in conductance: the doped BDF device was observed to be brighter than the undoped
BDF device, the latter exceeding in brightness of the standard NPD/AlQ3 device. These
observations also confirm the conclusions made for the BDF devices with the varied
ferrocene layer thicknesses: the charge transport properties of ferrocene are comparable
or exceed those for a common HTL like NPD.
The effect of F4-TCNQ doping increases with increased doping levels. Devices
with a F4-TCNQ-doped BDF hole transport layer that has 25% F4-TCNQ by weight are
shown on Figure 10. The structures of the devices are identical of those on Figure 9
except for a higher dopant concentration.

37
02 468 10 12
10
-4
10
-3
10
-2
10
-1
10
0
10
1
10
2
10
3
10
4
Brightness, Cd/m
2
Voltage, V
0.01 0.1 1 10
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
10
1
10
2
10
3
10
4
Current Density, mA/cm
2
Voltage, V
BDF:F4TCNQ(25%)300/NPD100/Alq400
BDF300/NPD100/Alq400
NPD400/Alq400
-1 0 1 23 45 67 8 9 101112
0.0
2.0x10
2
4.0x10
2
6.0x10
2
8.0x10
2
1.0x10
3
Current Density, mA/cm
2
Voltage, V
1 10 100 1000
1
E. Q. E. %
Current Density, mA/cm
2

Figure 10 Devices of the structure ITO/xHTL(300 Å)/NPD(100 Å)/AlQ3(400
Å)/LiF(10 Å)/Al(1200 Å), where xHTL = NPD, BDF, or BDF:F4-TCNQ (25% wt)

The turn-on voltages of all three types of devices are similar, and cannot be
decreased significantly since their values (2.5-2.6 Volts) are already very close to the
photon energy emitted by AlQ
3
(2.4 eV). The conductance, luminance, and quantum
efficiency, on the other hand, have a room for improvement, and are significantly higher
for F4-TCNQ-doped devices than those of standard NPD/AlQ3 or undoped BDF devices.
In the case of 4% doping levels, the difference between the standard and undoped BDF

38
device was of the same order as between the doped and undoped BDF devices. High
doping level made the doped device characteristics stand far apart from the other two
structures.
Ferrocenes: potential HTLs for red and NIR emitters
Ferrocene is a compound frequently employed as a quencher in kinetic studies of
photochemical reactions (Bhaumik, El-Sayed 1965; (Fry, Liu et al. 1996). Its ability to
quench triplets, however, has a lower limit on the energy scale: the sensitizers with triplet
energies significantly lower than 14000 cm
-1
are not quenched by ferrocene efficiently
(Kikuchi, Kikuchi et al. 1974). For the device applications, ferrocene being a strong
quencher will require a thin interface layer of some other hole-conducting material (such
as NPD used in our studies and described above) to screen the emitter from quenching by
ferrocene. However, if the wavelength of the OLED emissive material lies low enough,
the screening layer is not needed, and ferrocene derivatives can be exploited as HTLs to
their full advantage. Experimental measurement of the triplet energy of ferrocene’s was
described as somewhat challenging. The upper limit for the ferrocene triplet was reported
by M. Kikuchi et al (Kikuchi, Kikuchi et al. 1974) as 15000cm
-1
(~670 nm). In the
followed work of W. Herkstroeter (Herkstroeter 1975), estimation of the triplet energy
was made through flash kinetic spectrophotometry of quenching rates of an array of
sensitizers quenched by ferrocene, and the reported value for the ferrocene triplet level is
13300cm
-1
(750 nm). Thus, ferrocene derivatives such as BDF can be used as HTL

39
materials for OLEDs that emit in NIR region such as platinum
tetraphenyltetrabenzoporphyrin described by Borek et al (Borek, Hanson et al. 2007).
Conclusions
A low-volatility ferrocene derivative 1,1’-bis(biphenyl)ferrocene (BDF) has been
synthesized. Its electrochemistry was shown to be quantitatively similar to unsubstituted
ferrocene. The increased sublimation temperature of BDF compared to ferrocene allowed
for the direct measurement of its HOMO energy by UPS. This was the first direct
measurement of the ferrocene HOMO. Knowledge of the BDF HOMO in conjunction
with its electrochemistry allowed the use of BDF as a common reference for estimating
HOMO energy levels of an array of materials important in the OLED field. Also, BDF
has proven to be a hole conducting material. The OLEDs with BDF HTL layers showed
performance as good, or better than that of NPD-based standard devices. The
conductance, brightness, and quantum efficiency of BDF devices were further enhanced
by oxidative doping with F4-TCNQ. Ferrocenes can be excellent hole-transporting
materials for devices that will contain BDF interfaces with materials with shallower
HOMOs than AlQ
3
, and also, in particular, NIR-emitting OLEDs.

40
Chapter 3. Lithiumchelating dopants for the electron
transport layers of OLEDs with LiF/Al cathodes
Introduction
The efficiency of OLEDs built with small molecular weight materials is limited
by the voltage required to pass current though the highly resistive organic layers
comprising the devices. Doping the charge-transport layers of the device with electron
donors (in case of electron-transport layers, ETLs) or electron acceptors (for hole-
transport layers, HTLs) has been shown to increase conductivity of the device and, thus,
reduction of the driving voltage (Blochwitz, Pfeiffer et al. 1998; (Kido, Matsumoto
1998; (Liu, Pinto et al. 2001; (Parthasarathy, Shen et al. 2001; (Zhou, Pfeiffer et al. 2001;
(Zhou, Pfeiffer et al. 2002).
Significant improvement in device driving voltages and conductivity has been
achieved using redox doped charge-transport layers, particularly HTLs (Blochwitz,
Pfeiffer et al. 1998; (Zhou, Pfeiffer et al. 2001). For the ETLs, the most widely used
dopants at this moment are alkali metals, especially lithium (Blochwitz, Pfeiffer et al.
1998; (Jabbour G. E.; Kippelen 1998; (Johansson, Osada et al. 1999; (Piromreun, Oh et
al. 2000; (Parthasarathy, Shen et al. 2001; (Lee, Park et al. 2003). However, using
lithium metal introduces some problems. First, the Li dopant diffuses over time into the
ETL and can eventually diffuse into the emissive layer, leading to marked emission

41
quenching (it has been demonstrated that lithium has the propensity to migrate from the
interface into the organics as deep as 80nm (Parthasarathy, Shen et al. 2001)) . Second,
the number of carriers generated in the doping process is a small fraction of the amount
of Li that is present in the film. On the order of 1% of the Li leads to free carriers in the
film (Kido, Matsumoto 1998; (Parthasarathy, Shen et al. 2001).
Thin layers of LiF deposited at the metal/organic interface are the source of
lithium ions in the interfacial part of the ETL, and are known to improve electron
injection into organics. The ability of lithium cations to populate the interfacial part of the
organic ETL through the diffusion can be important in improving charge injection from
the cathode into the ETL. However, further penetration of lithium to the device emissive
zone is highly undesirable. To confine lithium ions within the thin part of the ETL
adjacent to the cathode, we propose to dope the interfacial part of the ETL with a strong
lithium-coordinating species. Crown ethers and cryptands are well-known to bind metal
ions including lithium cation (Myers 1980; (Michaux, Reisse 1982; (Steed 2001).
Besides enhancing the charge injection, immobilizing lithium cation by a chelating agent
can facilitate charge separation thus increasing the number of free charge carriers in the
ETL. This will ensure increase in conductivity of the ETL and, consequently, decrease of
the device operation voltage.

42
Experimental
Dibenzo-18-crown-6 and 2.2.2-cryptand were purchased from Aldrich, and
purified by sublimation before use.
2.2.2-cryptand complex with KI was prepared by dissolving KI (99.9%) and the
pre-sublimed 2.2.2-cryptand in absolute ethanol. The resulting solution was sonicated for
1 hour and left overnight at room temperature prior to solvent removal. Thus obtained
white solid was dried in the oven for 1 hour at 120°C.
Fluorene trimer F3 was synthesized according to the published procedure
(Sudhakar, Djurovich et al. 2003).
Organic light emitting devices were grown on ITO/glass substrates as described in
Chapter 2.
Results and Discussion
The chemical structures for electron-transport layer materials used in this work,
and the structure of the ETL dopant dibenzo-18-crown-6 (DBC) are shown below:

43
3
N N
O
O
O
O O
O
N
O
Al
3
BCP
SC5
F3
DBC
AlQ
3

The electron injection barrier which exists due to the energy offset between the
ETL LUMO and cathode workfunction affects the device driving voltage and brightness.
This barrier can be lowered if a positive species like a metal cation is present within the
part of the ETL adjacent to the cathode. Li
+
present at the organic-cathode interface can
shift the ETL LUMO level down due to electrostatic interaction of the positive ion with
the electron entering the ETL LUMO from the cathode, which constitutes electron

44
injection. Chelating can decrease the energy of Li
+
species residing within the ETL to
facilitate doping of the interfacial part of the organics with lithium ions.
Dibenzo-18-crown-6 (DBC) was our first choice as a chelating dopant for OLED
ETLs. 18-crown-6 is known to have good affinity towards Li
+
coordination (Michaux,
Reisse 1982). However, simple aliphatic crown ethers aren’t the most suitable for thermal
deposition because their volatility is too high. DBC is electrochemically inert material
having higher oxidation and lower reduction potentials than any of the carrier transporters
themselves, and its sublimation temperature of ~180°C is high enough for use in vacuum
thermal deposition systems.
OLEDs containing an interfacial part of the AlQ3 layer doped with DBC were
fabricated, and their metrics were compared with a standard NPD/ALQ
3
undoped device.
As seen from Figure 11, doping of AlQ
3
with DBC, however, does not change to the
device characteristics significantly.

45
02 46 8 10
0
2000
4000
C. D., mA/cm
2
Voltage, V
024 68 10
0
2000
4000
6000
8000
10000
Bridhtness, Cd/m
2
Voltage, V
partially doped AlQ3 layer
undoped
10 100 1000
0.01
0.1
1
Q. E., %
C. D., mA/cm
2
AlQ3:DBC 20nm
NPD 50nm
LiF 1nm
Al 120nm
ITO on glass
AlQ
3
30nm
N
O
Al
3
NPD 50nm
LiF 1nm
Al 120nm
ITO on glass
AlQ
3
50nm

Figure 11 NPD/AlQ
3
devices with the AlQ
3
layer doped with DBC (15% wt) at
the cathode interface. The undoped reference device contained a neat AlQ
3
layer of 50nm
thickness. The structure of the DBC-doped device is the inset on the left graph

The same holds true if we replace AlQ
3
with another common ETL/EIL material,
BCP (Figure 12). DBC-doping brought about no appreciable effect on the device
performance.

46
10 100 1000
1E-3
0.01
0.1
1
Q. E., %
Current density, mA/cm
2
undoped
doped 15%
doped 25%
0 2468 10
0
200
400
Current density, mA/cm
2
Voltage, V
02 468 10
0
2000
4000
6000
8000
10000
Brightness, Cd/m
2
Voltage, V
BCP:20nm
NPD 50nm
LiF 1nm
Al 120nm
ITO on glass
AlQ
3
30nm
BCP:DBC(x%) 20nm
NPD 50nm
LiF 1nm
Al 120nm
ITO on glass
AlQ
3
30nm

Figure 12 NPD/AlQ
3
/BCP devices with the BCP layer doped with DBC (15% and
25% wt) at the cathode interface. The undoped reference device contained a neat AlQ
3

layer of 20nm thickness

Both BCP and AlQ
3
contain heteroatoms, which makes these molecules capable
of coordinating lithium cations. In order to observe the lithium chelating effect by the
crown more clearly, for the ETL, we switched to less polar pure-hydrocarbon materials
containing no heteroatoms to provide coordination sites for Li
+
. 1,3,5-tris-phenyl-2-(4-
biphneyl)benzene (SC5) happens to be one of such materials. It has proven very difficult
to inject electrons from OLED cathode materials into the SC5 ETL. The result are high

47
voltage devices, that give quite low efficiencies. For these devices with SC5 ETLs, the
benefits of using DBC as a chelating dopant, are very clear: a 20 nm-thick SC5 layer
placed between AlQ
3
and the LiF/Al cathode shuts the standard NPD/AlQ
3
device off
almost completely. Once the SC5 layer is doped with DBC at the interface with the
cathode, the brightness and turn-on voltage of the device increase dramatically compared
to the device with no doping (Figure 13).
0 5 10 15
-500
0
500
1000
1500
2000
2500
3000
3500
Brightness, Cd/m
2
Voltage, V
0 5 10 15
0
50
100
150
200
Current density, mA/cm
2
Voltage, V
110 100
1E-3
0.01
0.1
1
Q. E., %
C.D., mA/cm
-1
SC5:DBC 20nm
NPD 50nm
LiF 1nm
Al 120nm
ITO on glass
AlQ3 30nm
SC5 20nm
NPD 50nm
LiF 1nm
Al 120nm
ITO on glass
AlQ3 30nm

Figure 13 SC5:DBC devices. The architecture of the device with SC5 EIL doped
with 15% of DBC is shown on the left. The undoped reference device contained a neat
SC5 layer of the same thickness

48

It has been shown that the complex formed of 18-crown-6 and lithium cation in
solution or gas phase is significantly non-planar (Anderson, Paulsen et al. 2003). To
prevent the doped DBC in the film from planarization and stacking, and encourage the
complex formation with Li
+
, we tried annealing the devices at 90°C after fabrication.
Perhaps for the fact that for BCP or AlQ
3
, and the 18-crown-6 the affinity towards Li
+

coordination are comparable, we didn’t observe appreciable doping effects when these
heteroatom-containing ETLs were used, even after the annealing of the DBC-doped
devices.
Lithium-coordinating dopants can be envisioned to have more significant effect if
we find a chelating agent that binds the lithium cation stronger then DBC does, and at the
same time is not prone to stacking and planarization in the solid thin film.
Attempts were made to apply Pd-catalyzed cross-coupling reactions to obtain the
derivatized crown ethers that are less susceptible to planarization. An exemplary reaction
scheme of the synthesis of a 15-crown-5 derivative is shown below:

49
O
O
O
O
O
O
O
O
O
O
Br
+
B(OH)
2
Pd(Ac)
2
, PPh
3
DME

Upon heating under high vacuum, this compound melts at 125°C, and remains
liquid up to 265°C, after which it decomposes without sublimation. Extra care was taken
to purify the material prior to sublimation. However, this didn’t help prevent the thermal
decomposition. Such problems of thermal instability were encountered for all
corresponding aromatic derivatives of 15-crown-5 and 18-crown-6:

50
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
Br
+
B(OH)
2
B(OH)
2
2
Pd(Ac)
2
, PPh
3
DME
decomposes at T>270
o
C

Cryptands are another class of molecules known to form strongly-bound adducts
with alkali metal ions (Dye, Ellaboudy 1984; (Wagner, Dye 1993). Our choice as a
ligand for Li
+
has been the 2.2.2-cryptand which has an adequate cavity to accommodate
lithium cations. The bonding scheme and DFT/B3LYP-optimized molecular structures of
the 2.2.2-cryptand and its complex with Li-cation are shown below on Figure 14:

51
OO
N
N
OO
O
O
2.2.2-cryptand

Figure 14 2.2.2-cryptand (left) and its complex with Li
+
(right), according to
B3LYP/6-31G* geometry optimization

The 2.2.2-cryptand evaporation temperature is 160-170°C (1.5x10
-7
torr). Even
though this compound was proven usable in our OLED fabrication system, its volatility is
quite high to make it require special attention during thermal evaporation. In order to
optimize deposition conditions, we prepared the complex of 2.2.2-cryptand with
potassium iodide. Unlike many other alkali metal halide complexes with 2.2.2-cryptand,
the KI complex does not coordinate solvent molecules when synthesized in solution
phase. The absence of coordinated solvent is important to avoid the doped layer

52
contamination when the MHal*cryptand complex is used thermal generation of the free
cryptand. Under high vacuum, the KI complex starts to release the free cryptand at
195°C.
Doping of SC5 ETL with the cryptand produced an effect similar to that of the
crown described previously: the cryptand-doped device showed over 1 V gain in turn-on
voltage, and an order of magnitude higher brightness and conductance compared to the
undoped NPD/ALQ3/SC5 device (Figure 15)

53
24 6 8 10 12
0
20
40
60
80
100
NPD/Alq/SC5/(SC5:cryptand15%)
NPD/Alq/SC5
C. D., mA/cm
2
Voltage, V
246 8 10 12
0
100
200
300
400
500
600
Brightness, Cd/m
2
Voltage, V
10 100
1E-3
0.01
0.1
1
Q. E., %
C. D., mA/m
2
SC5:crypt15% 10nm
NPD 50nm
LiF 1nm
Al 120nm
ITO on glass
AlQ
3
30nm
SC5 10nm

Figure 15 SC5:2.2.2-cryptand devices. The architecture of the device with SC5
EIL doped with 15% of the cryptand is shown on the left. The undoped reference device
contained a neat SC5 layer of the same thickness

Doping SC5 interfacial part with 2.2.2-cryptand results in significant
improvement in the turn-on voltage, and an order-of-magnitude increase in the brightness
of the device. Lowering the LUMO position of the host implies a better electron injection
from the cathode. Also, the complexation of Li
+
facilitates the charge separation giving
rise to larger charge carrier concentrations in the ETL, and improved conductivity. Either

54
the number of excitons formed in the emissive layer or their radiative rates doesn’t
change when the ETL is doped with lithium-chelating molecules such as DBC or
cryptand. Consecutively, ETL doping does not bring about significant changes in the
device quantum efficiency compared to the standard undoped devices.
Unlike what had been observed for DBC-doping of AlQ
3
and BCP, the NPD/AlQ
3

device on Figure 16 with a cryptand-doped ETL shows no dilution effect for AlQ
3
.
Contrary to that, both the brightness and the current passed through the device at a given
voltage increase with doping, whereas for the doping with DBC the effect is opposite.
The device data for 2.2.2-cryptand-doped AlQ
3
EIL is shown on Figure 3.6. Except for
the dopant, the device structures are analogous to those described earlier for DBC.

55
02 4 6 8 10 12
-100
0
100
200
300
400
500
600
700
800
900
1000
Current Density, mA/cm
2
Voltage, V
standard (undoped)
AlQ doped with cryptand
02468 10 12
0
2000
4000
6000
8000
10000
Brightness, Cd/m
2
Voltage, V
10 100 1000
0.01
0.1
1
Q. E., %
Current Density, mA/cm
2
Al 1100
LiF 10
AlQ3:crypt(15%) 150
AlQ3 250
NPD 400
ITO
Al 1100
LiF 10
AlQ3 400
NPD 400
ITO

Figure 16 Devices with the structure ITO/NPD(50nm)/AlQ
3
(30nm)/AlQ
3
:2.2.2-
cryptand 15% (20nm)/LiF(1nm)/Al(120nm). The reference undoped device contained a
neat AlQ
3
layer of 50nm thickness. The EL spectrum of the doped device is shown as an
inset on the left graph

Lithium cations are present within the interfacial part of the organics at the
ETL/cathode interface due to Li
+
penetration from a layer of LiF placed between the
aluminum and the ETL. It is well-known that a thin LiF layer is crucial for the devices
with aluminum cathodes. LiF is a wide band gap insulator, and how exactly it enhances
electron injection into an ETL is still a subject of discussion. Existing hypotheses on the

56
role of LiF in enhancing charge injection from Al-cathode into a commonly used ETL
material AlQ
3
include: lowering the Al workfunction (Shaheen, Jabbour et al. 1998); Li-
metal doping of AlQ
3
resulting in AlQ
3
-anion formation (Mason, Tang et al. 2001);
Fluorine doping of AlQ
3
(Grozea, Turak et al. 2002); Presence of interfacial dipoles
(Baldo, Forrest 2001); Band bending at the interface (Hung, Tang et al. 1997). The
problem with the metallic lithium doping hypothesis is that Li-metal formation through a
chemical reaction of LiF with Al is thermodynamically unfavorable (Mason, Tang et al.
2001). Furthermore, fluorine doping of AlQ
3
can be ruled out by the fact that Li metal as
dopant works as well instead of LiF; also, fluorine anion would raise the LUMO, and the
injection barrier would therefore rise. Instead, if we assume that the ETL at the interface
is doped with Li
+
when lithium ions diffuse from the LiF/organic interface deeper into the
organic layer, then the electrostatic interaction of Li-ions with the ETL material will
lower its LUMO. This means decrease in the energy difference between the cathode
workfunction and the electron-accepting orbital of the ETL. This Li
+
-diffusion model
thus helps explain the phenomenon of the electron injection enhancement by the LiF thin
layer.
Conclusion
It has been shown that stabilization of Li cation by coordination within the
interfacial part of ETL improves electron injection into ETL for the OLEDs with LiF/Al
cathodes. Chelating dopants significantly improve driving voltages (over 1 Volt) and

57
brightness of the devices (over a two order of magnitude increase) for some ETL
materials, especially those which do not have the metal-coordinating properties. Stronger
chelating dopants will likely to have a greater impact on the device driving voltage and
brightness. The effect of Li-chelation on the device operational life times should be
further investigated.

58
Chapter 4. Blue fluorescent materials for WOLEDs with
separate channels for triplet and singlet exciton harvesting
Introduction
White organic light-emitting devices (WOLEDs) have a potential to provide a
power-efficient low-cost technology for general-purpose illumination. WOLEDs reported
to date show significantly better power efficiencies than incandescent lamps (25-35
Lm/W vs ~15 Lm/W for incandescent sources (Sun, Giebnic et al. 2006)). One of the
unresolved problems that WOLED technology faces is that a robust and efficient blue-
emitting component of an RGB white device is hard to find. The project described in this
Chapter has a goal to explore a class of blue fluorescent materials indolocarbazoles that
possess electronic structure properties relevant to the white-emitting RGB devices with
blue fluorescent and red and green phosphorescent components.
Preceded by a short review of reported blue fluorescent materials for OLEDs, the
photophysical properties of indolocarbazoles and OLED devices with ICs as blue
emitters will be discussed in this Chapter.

59
Overview of blue fluorescent emitters for OLEDs
We will start the discussion of organic blue fluorescent materials for OLEDs with
a review of organic blue fluorophors. The review will not be exhaustive but rather
focused on examples relevant to electroluminescent devices.
A large number of organic compounds have electronic transitions that correspond
to emission in the blue region. However not all of them actually show detectable blue
fluorescence. This is due to existence of various deactivation pathways and/or poor
overlap of the orbitals involved in the electronic transition. The deactivation mechanisms
include intersystem crossing to triplet states, and internal conversion through
intramolecular motions. The degree of intersystem crossing is determined by the energy
gap between S
1
and T
1
states, the orbital symmetries, and presence of heavy atoms.
Vibrational deactivation is facilitated when the twisting or bending molecular motions are
available to couple with the electronic transition. Hence, an efficient blue fluorophor will,
in general, will be a rigid structure that contains no heavy atoms, and has slow
intersystem crossing rates.
Hydrocarbons
A number of aromatic hydrocarbons that emit in the visible satisfy the mentioned
above emission wavelength and efficiency requirements: their planar rigid structures
prevent vibronic deactivation; the rate of intersystem crossing is extremely low (as an
example, the phosphorescent life times for benzene and naphthalene are of the order of

60
seconds (Sixl, Schwoerer 1970)), therefore the fluorescent quantum efficiencies can
approach unity. Examples are perylene and anthracene derivatives (Nijegorodov,
Downey 1994; (Nijegorodov, Mabbs et al. 2001; (Shi, Tang 2002). A review on the
recent developments in the field of blue fluorescent OLEDs is largely concerned with the
blue emitters belonging to this class (Wen, Lee et al. 2005).
Synthesis, crystal structures, DSC data, photophysical properties, electrochemical
characterization, theoretical modeling, and OLED data on cis- and trans- isomers of 2,5-
Diphenyl-1,4-distyrylbenzene (DPDSB) were reported by Z. Xie et al (Xie, Yang et al.
2005) (structures shown on Figure 17)

Cis-DPDSB
Trans-DPDSB

Figure 17 Two isomers of diphenyl-1,4-distyrylbenzene

61
The fluorescence emission maxima for these two isomers are 435 and 444 nm in
crystalline solid and 400 nm for the trans-isomer in THF solution. The cis-isomer shows
no fluorescence in solution. It was found that presence of the torsion conformation brings
about a drop in quantum efficiency from 0.95 for trans-isomer to effectively zero for cis-
DPDSB. The OLEDs with the structures NPD/(cis/trans-DPDSB)/BCP/AlQ
3
/LiF-Al
showed blue electroluminescence with peak maxima at 440 and 464 nm and CIE
coordinates (0.1839, 0.1757) and (0.1797, 0.2189) for cis- and trans- isomers,
respectively. The devices showed moderate brightness; no E.Q.E. data for the devices
were reported.
Fluorene and spirofluorene derivatives can show high solution quantum
efficiencies (Hung, Liao et al. 2005; (Tang, Liu et al. 2006; (Liu, Lu et al. 2007). The
emission of fluorenes usually occurs not in the blue but in near-UV region. Near-UV
OLEDs containing neat emitter layers have reported by T.-C. Chao et al (Chao, Lin et al.
2005). The solution and thin film quantum yields of the discussed fluorenes range within
0.6-0.7. The device with the structure ITO/PEDT:PSS/TCTA/B2/TPBI/LiF/Al with a
fluorene core-containing B2 emitter (Figure 18) showed 3.6% external quantum
efficiency at 100Cd/m
2
. Electroluminescence emission peak λ
max
(B2) was 392 nm.

62
B2
Figure 18 Spirofluorene compound described by Chao et al (Chao, Lin et al.
2005) as an OLED neat blue emitter

The emission of pyrene-substituted fluorenes (Figure 19) and spirofluorenes (Tao,
Peng et al. 2005) is significantly red-shifted compared to that of the near-UV fluorene-
based emitters mentioned in (Chao, Lin et al. 2005). The fluorene derivatives shown on
the scheme are reported to have high solution quantum efficiencies of 0.68-0.78:

63
DPF
SDPF DPhDPF

Figure 19 Derivatized fluorenes used as neat blue emitters for OLEDs (Tao, Peng
et al. 2005)

These pyrene-derivatized fluorenes were used in blue OLEDs as neat emitters.
The device structures that correspond to the electroluminescence spectra on Figure 20 are
ITO/CuPc/NPD/fluorene derivative/AlQ
3
/Mg:Ag. Quantum efficiencies for the devices
were not reported.

64

Figure 20 Normalized electroluminescence spectra of the OLEDs containing
emissive layers of neat fluorenes (Tao, Peng et al. 2005)

Spyro-antracene was reported by D. Gebeyehu et al. as a neat blue emitter for
fluorescent OLEDs with the CIE coordinates of (0.14, 0.14) (Gebeyehu, Walzer et al.
2005).
Spyroanthracene

65

The optimized device structures employing doped hole-transport and multiple
charge-blocking layers allowed for the device turn-on voltages as low as 2.9 V, and the
external quantum efficiency of 3.7% at 100Cd/m
2
.
A recent communication (Wei, Chen 2007) described a stilbene-fluorene hybrid
with high efficiency sky-blue fluorescence. Cis-4,4’-bis(diphenylamino)slilbene/fluorene
hybrid structure of which is shown on the scheme below improves the emission
efficiency compared to the trans-configuration stilbene-fluorene hybrid described in
earlier reports (Rumi, Ehrlich et al. 2000) by confining the cis-isomer in a cyclic
framework.
N
N
Ph
Ph
Ph
Ph

This cyclic stilbene-fluorene hybrid is characterized by 0.9 quantum efficiency in
CH
2
Cl
2
solution, fluorescence λ
max
at 461 nm with the FWHM of 57 nm, and the Tg of
123°C. It is claimed that the OLEDs with this blue emitter have the efficiency two times
higher than for the corresponding open form systems. The device consisted of a neat
layer of stilbene-fluorene hybrid serving as both hole-transport and emissive layer; a

66
TPBI electron-transport layer, and a LiF/aluminum cathode. When this device structure is
modified with insertion of 20 nm of PEDOT:PSS layer between ITO and the
HTL/emissive layer, the reported external quantum efficiency increases from 1.94% to
7.87%. The details on how E.Q.E. is defined, or protocol for its measurement are not
given.
Indenofluorenes present an interesting class of blue emitters as they are non-
alternant structures (Michl, Thulstrup 1976) similar to indolocarbazoles (the latter are the
subject of discussion in this Chapter), and have been shown to fluoresce with quantum
yields reaching 0.7 (Nijegorodov, Downey 1994; (Merlet, Birau et al. 2002; (Hadizad,
Zhang et al. 2005).

2,6-substituted indenofluorenes possess irreversible oxidation and reduction. The
OLEDs with the structure ITO/indenofluorene(500)/PBD(500)/LiF(15)/Al were
fabricated with the indenofluorene derivative shown below as a fluorescent emitter
(Hadizad, Zhang et al. 2005).

67
S
S

This thiohpene-derivatized indenofluorene exhibits blue photoluminescence in
fluid solution. However, the broad electroluminescence spectrum of the device had a
peak maximum at 530-600 nm that also had a broad shoulder in the 450-500 nm region.
The color of the emission was described as yellowish-white, and the reason for large line
broadening was attributed to polycrystalline nature of the film. The device showed
brightness of 1400 Cd/m2 at 10V, and maximum luminance efficiency of 1 Cd/A at 9V.
Quantum efficiency was not reported.
R. Chiechi et al started investigating fluoranthene and its derivatives with the
intent to find a robust blue emitter that is easier to synthesize than fluorene or
spirofluorene (Chiechi, Tseng et al. 2006). They reported a two-step synthesis of 7,8,10-
triphenylfluoranthene (TPF) that exhibits strong luminescence and is bulky enough to
make it usable for vacuum thermal deposition.

68
TPF

Solution quantum yield of TPF was found to be solvent dependent: it was
reported as 0.38 in CH
2
Cl
2
and 0.52 in cyclohexane. It was also pointed out in that TFP
does not reach the maximum possible quantum yields for fluoranthenes: another
derivative, benzo[k]fluoranthene has a quantum yield of unity in solution.
Benzo[k]fluoranthene

In solid state, however, the TPF quantum yield measured using an integrating
sphere reaches 0.86 for the films of TPF doped into PMMA. The quantum yield of the

69
neat TPF was reported as 0.51. The high quantum yield of TPF in the PMMA matrix
underscores the role of rigid environment in suppressing TPF emission quenching via
non-radiative deactivation pathways. OLEDs of the structure
ITO/NPD/CBP/TFP/BCP/Ca-Al with TFP as a neat emitter had turn-on voltage 5 V, CIE
coordinates (0.117; 0.24), electroluminescence spectrum with λ
max
= 456 nm, and
maximum E.Q.E of 1.8%. Like fluorenes, fluoranthene derivatives belong to non-
alternant hydrocarbons that potentially have a singlet-triplet energy gap. This property
makes them also an interesting class of blue emitters for mixed
fluorescent/phosphorescent WOLEDs. Hence, photophysical properties of fluoranthenes
and the OLEDs having fluoranthenes as blue emitters will be discussed later in this
Chapter in Experimental and Discussion parts.
N- and O- heterocycles
Unlike aliphatic amines, the internal conversion in triarylamines is hindered thus
eliminating an important deactivation pathway. In addition to efficient fluorescence, the
presence of lone pairs on nitrogen atoms lowers the oxidation potential thus facilitating
charge transport through generally resistive organics.
Liao at al. (Liao, Lee et al. 2005) claimed an 8.7% quantum efficiency for their
blue doped fluorescent OLED that contained an emissive layer of DSA-Ph (structure
shown below) doped into 2-methyl-9,10-di(2-naphthyl)anthracene (MADN) host.

70
N
N

Given that the Q. E. value exceeds 5% (the theoretical limit for the external
quantum efficiency of fluorescent devices (Baldo, O'Brien et al. 1999)) it can be
concluded that the efficiency was measured using an integrated sphere that implies
collecting the light coming from all directions, not just the front (viewing) direction. The
front-collected external quantum efficiency η
ext
pertinent to display applications
constitutes 1/5 of the efficiency measured using integration spheres (Forrest, Bradley et
al. 2003). In the case of the blue device described in Ref 14 the η
ext
value will be
significantly lower than 2%.
In a framework of the project of white fluorescent/phosphorescent OLEDs, we
examined DPAVB, a compound different from DSA-Ph by four methyl groups:

71
N
N

250 300 350 400 450 500 550 600 650 700 750
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
Intensity, norm.
Wavelength, nm
emission, 2MeTHF
excitation, 2MeTHF
absorption, 2MeTHF
excitation, DMF
emission, DMF
absorption, DMF
all at room temperature

Figure 21 Room temperature solution emission, absorption and excitation spectra
of DPAVB (on the left) in DMF and 2-Methyltetrahydrofuran (2MeTHF, measured in
this work).

72

300 350 400 450 500 550 600 650
0.0
0.2
0.4
0.6
0.8
1.0
Intensity, norm.
Wavelength, nm
DPAVB emisson
DPAVB excitation
77K, in 2MeTHF

Figure 22 Emission and excitation spectra of DPAVB in 2MeTHF measured at
77K.

It will be explained later in this Chapter that one of the properties of a blue
fluorescent dopant needed to efficiently channel the singlets to a blue emitter and the
triplets to green and red phosphors is a small S-T energy gap of the fluorophore molecule.
The solution in 2-MeTHF did not show any triplet emission either at room temperature or
at 77K. Often, the presence of heavy atoms on the emitter molecule or using a heavy
atom-containing solvent or additive can significantly enhance phosphorescence signal

73
intensity (Turro 1991). However, even when methyl iodide was added to a DPAVB
solution in 2MeTHF before cooling to 77K, the frozen solution showed no emission from
the triplet (Figure 22).
BCzVBi is a commercially available relatively high-efficiency blue emitter with a
fluorescence emission peak maximum at ~470 nm (Hosokawa, Higashi et al. 1995). It has
been used in mixed fluorescent/phosphorescent white OLEDs as a blue fluorescent
dopant (Sun, Giebnic et al. 2006).
N
N

The triplet energy of BCzVBi has never been reported. However, due to the
presence of stilbene-like moieties its triplet energy is not expected to be high.
Interesting photophysical properties were reported for oxadiazole derivatives
(Liang, Wang et al. 2002). Two emission peaks in the photoluminescence spectrum of 2-
(2-hydroxyphenyl)-5-phenyl-1,3,4-oxadiazole (HOXD) at 365 and 489 nm were
simultaneously observed in CH
2
Cl
2
solution at room temperature. Solid state samples of

74
HOXD showed only one emission peak at 481 nm. It was claimed that this lower-energy
emission peak corresponds specifically to phosphorescence.

N
N
O
O
H
enol form
N
N
O
O
H
keto form

2-(2-hydroxyphenyl)-5-phenyl-1,3,4-oxadiazole (HOXD)

HOXD was employed as a neat blue emitter in an OLED with the structure
ITO/NPB/HOXD/BCP/Alq
3
/Mg:Ag. The electroluminescence spectrum of this OLED
showed a maximum at 451 nm. The device had a turn-on voltage of 7V and brightness of
650Cd/m
2
at 25V. The device quantum efficiency was not listed.
A synthesis of another blue-emitting oxadiazole derivative containing
phenanthroline units was described (Bing, Leung et al. 2004).

75
N N
O
N
N
N
N
N

This compound shows a solution quantum yield of 0.78 and presumably possesses
high charge conducting capabilities for both holes and electrons. Its emission λ
max
is 461
nm in THF solution.
Organo-silicon compounds
Silicon-containing compounds have also been investigated as blue fluorescent
materials (Palilisa, Ma¨kinen et al. 2003; (Lo, Sellinger 2006). A silole derivative 2,5-di-
(3-biphenyl)-1,1-dimethyl-3,4-diphenylsilacyclopentadiene (PPSPP) has been reported to
have a solid state quantum efficiency of 0.85.

76
Si
N N
N N
Si
PyPysPyPy PPSPP

Structures of PyPySPyPy and PPSPP (Palilisa, Ma¨kinen et al. 2003)

For the device employing PPSPP as an emitter and another silole derivative 2,5-
bis-(2’,2”-bipyridin-6-yl)-1,1-dimethyl-3,4-diphenylsilacyclopentadiene (PyPySPyPy) as
an electron-transport layer (see Figure 23, device structure on the right) a very efficient
exciplex emission was observed. The exciplex formed between PyPySPyPy and PPSPP at
1:1 ratio exhibits a red-shifted fluorescence with λ
max
at 495 nm, and makes an OLED
with an E.Q.E. as high as 3.4% at 100Cd/m
2
.

77

Figure 23 Device structures (Palilisa, Ma¨kinen et al. 2003)

The electroluminescence spectra of the devices with silole emitters are depicted
on Figure 24 along with the PL spectra of neat materials. An NPD EL spectrum is shown
for comparison.

Figure 24 Photoluminescence spectra of the films of NPD (closed circles), PPSPP
(closed squares), NPB:PPSPP (molar ratio 1:1) (closed triangles), and EL spectra of
bilayer and three-layer devices based on PPSPP as the emissive/electron transport layer
(open circles) and PPSPP and PyPySPyPy as the emissive and electron transport layers
(open squares), respectively (Palilisa, Ma¨kinen et al. 2003).

78
The solid state PL quantum yield of the exciplex emission was reported as 0.62.
Even though the PL efficiency of PPSPP as a neat solid is significantly higher than that of
the PPSPP:PyPySPyPy exciplex (0.86 vs. 0.62), the quantum efficiency of the bilayer
OLED with PPSPP as a neat emitter where the exciplex is not formed (the structure on
Figure 24 on the left) was only 1%.
The anticipated weak link in silanes as blue organic emitters (the same may apply
to most other heteroatoms) is the Si-C bond. Thermodynamical instability towards
irradiation with blue light is now thought to be one of the major causes of limited
operational life times for the organic blue electroluminescent devices, so introducing
weaker bonds into the emitter molecules should be considered a major disadvantage.
Polymers
Besides small-molecule devices, polymer-based OLEDs attract a lot of interest
due to simplification of the device manufacturing process: spin-coating and inkjet
printing which polymers are amenable for, is less costly and technologically simpler than
vacuum vapor-deposition.
One class of polymers that draws most attention in the field of organic polymer
devices is polyfluorenes and their derivatives (Grice, Bradley et al. 1998; (Setayesh,
Grimsdale et al. 2001; (Hung, Liao et al. 2005; (Liu, Min et al. 2006). They retain high
fluorescence efficiency of small molecule fluorenes and processability advantages of

79
polymers. At the same time, as pointed out, polyfluorenes have some intrinsic drawbacks
as OLED materials (Kobayashi, Kijima 2007). A significant part of the emission lies in
the UV-zone due to existence of a higher-energy shoulder in the emission spectrum. This
leads to brightness and efficiency losses; also, polyfluorenes frequently emit undesirable
green colors brought about by electrooxidation processes that are triggered during device
operation. These green bands do not appear in the PL spectra of polyfluorenes. It was
claimed that some of the derivatized polycarbazoles developed by Kobayashi et al.
(Kobayashi, Kijima 2007) are free from these drawbacks. The devices with PmDPAC
emitting polymer (see PmDPAC structure below) displayed blue electroluminescence
with the emission λ
max
of 450 nm, and the CIE coordinates of (0.159, 0.173), which had
no difference from the corresponding photoluminescence spectrum.

N
N
O
O
n

80
The turn-on voltage for this polymer device was 4.2 Volts; the device showed
fairly high brightness of 4000 Cd/m2 at 12 Volts; the E.Q.E. values were not reported.
The photoluminescence quantum yield of PmDPAC in a neat thin film measured using an
integrating sphere was 0.32.
Mixed fl/ph WOLEDs
The concept and principles of operation
A concept of using a blue fluorescent component in conjunction with red and blue
phosphorescent dopants in WOLEDs was mentioned above. The motivation to use
fluorescence instead of blue phosphors in white OLEDs is low operational stability of
blue phosphors and, particularly, materials that serve as their hosts. As one of the earlier
examples of realization of the fl/ph WOLED idea, a device containing a red phosphor
tris(1-phenylisoquinoline) iridium (III) (Ir(pic)
3
) doped into BCP, and a blue-emitting
fluorescent dopant 2,5,7,10-tetra-phenylpyrene (TPP) doped into a separate layer of 4, 4’,
4”-tris(carbazol-9-yl)-triphenylamine (TCTA) can be mentioned (Qin, Tao 2005).
However, it wasn’t until 2006 the idea of using fluorescent blue in combination with
phosphorescent red and green emitters to harvest singlet and triplet excitons through
completely separate channels was described (Sun, Giebnic et al. 2006).

81
Separate channels of exciton harvesting for singlets and triplets
In a device a general schematic of which is given on Figure 25 below, the singlet
excitons first formed on the host molecules are transferred onto the blue fluorophor.
According to spin statistics, these singlets amount for ¼ of all excitons formed. The rest
of the host excitons are triplets. If these triplet excitons diffuse to the phosphor-doped
region and are harvested by efficient red- and green-emitting phosphorescent dopants,
then the combined emission from such an OLED will be white.

phosphor
phosphor
fluorophor
HTL
ETL
anode
cathode
phosphor
phosphor
fluorophor
HTL
ETL
anode
cathode

Figure 25 A schematic of a white OLED that consists of blue fluorescent and red
and green phosphorescent emitters doped into a common host (shown in gray color).

82
In addition to addressing the problems of operational stability of blue
phosphorescent devices and availability of hosts for blue phosphors, this architecture
takes advantage of resonant injection into the blue emitter HOMO level that helps avoid
approx. 20% of power losses due to intersystem crossing present in all-phosphorescent
WOLEDs.
However, if the energy level diagram of the reported white device (Sun, Giebnic
et al. 2006) is examined closely, it can be noted that, since the triplet level of the blue
dopant is significantly lower than the host triplet level, the triplet excitons formed on the
host can potentially transfer onto the triplet level of the blue dopant. This defies the idea
of separate exciton harvesting by blue fluorophor and red and green phosphors. With that
in mind, we began looking to design a new blue-emitting material that will have high
emission efficiency, and will not capture triplet excitons.
Experimental
MeIC synthesis: 0.5 g of indolocarbazole, 0.03 g of triethylbenzylammonium
chloride, and 10 mL of DMSO were placed into a 100 mL round-bottom flask. To the
suspensia, 0.6 g of 50% aqueous NaOH solution was added, after which the color turned
green immediately. This was followed by addition of 0.4 mL of methyl iodide (color
changed to brown). After 48 hour stirring at ambient conditions, the reaction was stopped
by adding of 0.5 mL of water. Benzene (100 mL), ethyl acetate (100 mL), ether (100 mL)
were subsequently added for extraction. The combined organic phase was washed with

83
water and dried over drierite. After removing the solvents under vacuum, the solid was
passed through a hexane:CH
2
Cl
2
(50:50) column. The compound was further purified by
sublimation (T
subl
~ 195-205°C). Yield after sublimation 38%. The obtained product is
soluble in common organic solvents, and crystallizes easily in a form of yellow needles.
Aromatic IC derivatives: Substitution of hydrogen with phenyl, naphtyl, and 2,6-
dimethoxyphenyl groups at both N-atoms in indolocarbazole was done under Ullmann-
type coupling reactions conditions using Cu powder or Cu
2
SO
4
*5H
2
O catalysts in 1,2-
Dichlorobenzene or n-Tridecane.
Sample procedure 1 (Cu-powder catalyst/1,2-Dichlorobenzene solvent): 1.5 g of
indolocarbazole (1 eq.), 2.42 g of iodobenzene (2 eq.), 0.75 g of Cu powder (2 eq.), 2.61
g of K
2
CO
3
(3.2 eq.), and 0.156 g of 18-crown-6 (0.1 eq) was pre-mixed in a 3-neck
flask. 40 mL of 1,2-Dichlorobenzene was added, and the resulting highly-heterogeneous
suspensia was deaerated by bubbling dry N
2
for 20 min. followed by raising the
temperature of the sand bath to 190°C. After refluxing for 48 hours, the reaction mixture
was cooled to 80°C and passed through dry silica-gel to remove the crown ether and
copper catalyst. Solvent was removed by vacuum-distillation. The grayish-yellow solid
was purified by sublimation (T
subl
~ 250°C). Mass of the target product after two
sublimations: 0.83 g. Yield after two sublimations 35%. NPIC (using 1-Iodonaphtalene as
a starting material), and TBIC (using 3,5-Ditretbutylbromobenzene) were obtained in
satisfactory yields according to similar procedures.

84
Sample procedure 2 (CuSO
4
*5H
2
O catalyst/ tridecane solvent): 0.38 g of
indolocarbazole, 1.07 g of 2,6-Dimethoxyiodobenzene, 0.04 g of CuSO
4
*5H
2
O, 0.77 g of
K
2
CO
3
, and 5 mL of tridecane was placed into a 30neck 100-mL flask. Nitrogen was
bubbled through the suspension for 30 min before heating started. Reflux (sand bath
temperature 250°C) continued for 48 hours. After cooling to room temperature, the
reaction mixture was washed with water (3x100 mL), and the solvent from the organic
phase was removed by vacuum distillation using an oil pump and a vacuum line equipped
with a liquid nitrogen trap (T
distill.
(tridecane) ~120-130°C). The solid residue was passed
through a silica-gel column (eluent dichloromethane). Yield 36% (before sublimation).
Sublimation under reduced pressure (diffusion pump) resulted in partial decomposition of
the material in the boat. A pure DMIC fraction was collected at the oven temperature of
290°C.
Halo-arenes for Ullmann coupling with indolocarbazole:
2,6-Diisopropyl -1-Iodobenzene was synthesized from 2,6-Diisopropylamine in
two steps through diazonium salt.
NH
2 NI
HCl, -10C
N
Cl
NaNO
2
KI

85
The starting material 2,6-Diisopropylamine was purified by recrystallization of its
hydrochloride salt from methanol/toluene.
1.1 mL of purified 2,6-Diisopropylamine was placed into 100 mL immersed into
and ice-salt bath (bath temp. = -10°C). 5 mL of 37% aqueous HCl was added followed by
addition of ~5 mL of DI water. After 20 min, a cooled aqueous solution of 0.50 g of
NaNO2 was added dropwise to the mixture. Then a cooled NaI
(aq.)
was added dropwise.
Stirring continued until warming up to room temperature. Na
2
CO
3
and water added to
stop the reaction. After extraction with chloroform, the organic phase was treated with
starch to remove iodine, filtered, and dried over MgSO
4
. The dark oil was passed through
a hexane column to afford 0.47 g of colorless oil. The product was characterized by NMR
and mass spectrometry (paired signal of M
+
= 288 and demethylated fragment M = 273
of equal intensities).
9-Iodoanthracene was synthesized according to published procedure (Suzuki,
Kondo et al. 1986). The solvent hexamethylphosphoramide (HMPT) was distilled over
CaH
2
under vacuum before use in the reaction. A mixture of 9-Bromoanthracene (1.91 g),
CuI (7.31 g), and KI (13.2 g) in fresh-distilled HMPT (30 mL) was bubble-degassed with
N
2
followed by stirring under positive N
2
pressure for 40 hours at 160-170°C using a
sand bath. After cooling to room temperature, a dilute (1:1) HCl was added to quench the
reaction. 100 mL of toluene was added for extraction. After separation from water/HMPT
phase, the organic layer was dried over MgSO
4
, and solvent removed in vacuo. The

86
obtained brown oil solidified quickly after cooled to room temperature. The product was
recrystallized twice from methanol/ethylacetate. Yield 50%.
Photophysical characterization: Quantum yields for solutions, thin films, and
solids were determined using the Hamamatsu integrating sphere. Toluene was used as a
solvent for all solution quantum yield measurements except where stated otherwise. For
the solution quantum yield measurements the concentration was chosen such that the
absorption peak intensity falls within 0.08-0.25 units range since absorption significantly
below 0.1 produces unacceptably large errors, whereas it is necessary to minimize the
absorption to reduce concentration quenching. Where applicable, the toluene solutions
were deaerated by bubbling nitrogen through for up to 5 min.
Doped films were prepared by spin casting CH
2
Cl
2
solutions of appropriate
concentrations for 40s at 3000rpm.
Spectra and quantum yields of solids were recorded using solution-grown crystals
or powders of sublimed compounds.
Toluene solutions of 5-10mg/L (2-5*10
-5
M) IC concentrations were used for
extinction coefficient measurements. Absorption spectra were recorded using Hewlett-
Packard 4853 diode array spectrophotometer; Steady state emission spectra at room
temperature and 77 K were determined using a Photon Technology International
QuantaMaster model C-60SE spectrofluorimeter; 2-Methyltetrahydrofuran was used for

87
most 77K experiments to provide optically transparent glasses; Luminescence life times
were measured on the HBH Fluorocube lifetime instrument by a time correlated single
photon counting method using either a 373 nm LED or 405 nm laser excitation source.
Electrochemisty: Cyclic voltammetry was performed using an EG&G
potentiostat/galvanostat model 283. Anhydrous DMF was used as the solvent under an
inert atmosphere, and 0.1 M tetra-n-butylammonium hexafluorophosphate was used as
the supporting electrolyte. A glassy carbon rod was used as the working electrode, a
platinum wire was used as the counter electrode, and a silver wire was used as a
pseudoreference electrode. The redox potentials are reported relative to a
ferrocenium/ferrocene (Cp
2
Fe
+
/Cp
2
Fe) redox couple used as an internal reference.
OLEDs: fabrication and characterization of the OLEDs was performed according
to procedures described in the Experimental section of Chapter I.
Results and discussion
As pointed out in Introduction, the number of organic small molecules emitting in
the blue region is large. Our search for a material suitable as a blue fluorophor in the fl/ph
WOLED will be narrowed by several constraints: First, the material has to have a high
quantum yield of fluorescence to balance highly efficient emission from red and green
phosphors. If the efficiency of the blue component is not high enough, the resulting EL
spectrum of the device will not be pure white but yellowish instead, depending on the

88
amount of losses associated with a blue component. Second, the triplet level of the blue
dopant needs to be higher than that of the host to prevent triplet exciton transfer onto the
blue dopant molecules (triplets are to be diffusion-transferred to the phosphors). Third,
the singlet level of the host should be higher energy than the singlet of the dopant to
ensure that the singlet excitons from the host can be transferred to the blue fluorophor.
Lastly, the triplet levels of both host and blue dopant should be higher than triplet
energies of the green and red phosphors. To satisfy these energy level alignment
requirements, the blue-emitting dopant will have to have a small energy gap between its
singlet and triplet levels.
The next question to be asked is how to look for materials with a given range of
excited singlet and triplet states separation – in other words – what governs a singlet-
triplet gap in a molecule. To answer this question, J. Michl and E. W. Thulstrup (Michl,
Thulstrup 1976) considered the molecular orbital picture for two isomeric compounds –
naphthalene and azulene which have similar ionization potentials and electron affinities,
but quite different singlet-singlet excitation energies. A decrease in the energy of the first
singlet excited state for azulene was explained by taking into account the electron-
electron repulsion energies, and these differ considerably for alternant (such as
naphthalene) and non-alternant hydrocarbons. Non-alternant structures, i.e. those that
possess odd-member rings, are characterized by much larger degree of charge transfer
upon excitation, weaker electron-electron repulsion, and lower first singlet excited state

89
energies. Considering that the triplet energies of naphthalene and azulene are close, the
smaller electron repulsion integral in azulene explains its smaller singlet-triplet gap.
Therefore we can use the degree of charge transfer and whether or not a given structure is
non-alternant to predict the relative value of the singlet-triplet gap of the compound.
Charge-transfer transitions are characterized by broad and featureless spectral
lines. Opposed to OLEDs producing individual red, green or blue colors that are required
to have as narrow spectral lineshape as possible to ensure color purity, the components of
a white-emitting device can be broadband emitters. Even more so, broad spectra for
WOLED components are preferential to better match the white solar spectrum.
Potentially, a narrow singlet-triplet gap can lead to more facile intersystem
crossing to the triplet excited state, which means fluorescent efficiency losses. However,
it will be demonstrated that for an example of indolocarbazoles it is not a problem: less
than 2% of all excited molecules of a blue fluorophor undergo intersystem crossing to the
triplet state.
Fluoranthenes
The volatility of benzo[jk]fluorene, the simplest of fluoranthenes (Scheme aa) is
too high as far as vacuum thermal deposition is concerned: its sublimation temperature
was found to be 105°C under 10
-7
torr

90

Adding one fused benzene ring to the parent fluoranthene compound results in
50°C increase in sublimation temperature: under high vacuum, benzo[b]fluoranthene
(BbF) sublimed at 155°C. The choice between two available benzofluoranthene
derivatives benzo[k]fluoranthene (BkF) and benzo[b]fluoranthene (BbF) was based on
the predicted degree of charge transfer upon electronic excitation. BbF possesses more
asymmetric structure which provides stronger charge-transfer character to the transition
and, hence, possibly a smaller singlet-triplet energy difference. This prediction was
confirmed by semiempirical PM3 calculations that showed the electron density
distributions on HOMO and LUMO orbitals (see Figure 26)

91
BkF BbF

Figure 26 Chemical structures and valence orbital pictures for two
benzofluoranthene isomers BbF and BkF according to PM3 calculation.

The room temperature emission spectrum of BbF dissolved in 2-
Methyltetrahydrofuran shows a rather broad line with a peak maximum at 432 nm
(Figure 27).

92
200 300 400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Intensity (normalized)
Wavelength, nm
BbF emission, RT
excitation at 450nm
absorption

Figure 27 Room temperature excitation (empty circles), absorption (filled circles),
and emission (filled squares) spectra of BbF

Line broadening, a lack of features, and a large Stokes shift are indicative of
charge-transfer character of the S
1
-S
0
electronic transition in BbF.
When cooled to 77 K, BbF embedded in a 2-MeTHF frozen solution shows a
fluorescence peak that is 10 nm blue-shifted compared to the room temperature spectra.
At 77 K the vibrational structure is more pronounced, and a weak phosphorescence signal
is also visible (Figure 28).

93
300 400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Intensity (normalized)
Wavelength, nm
emission 77K, in 2MeTHF
excitation @525nm, 77K, CH
3
I
emission 77K, 2MeTHF + CH
3
I

Figure 28 Excitation (empty circles), absorption (filled circles), and emission
(filled squares) spectra of BbF recorded at 77 K

The phosphorescence signal (enhanced by adding methyl iodide due to the heavy
atom effect, as seen on Figure 28) has a peak maximum at 524 nm. Calculated as a
difference between fluorescence and phosphorescence peak maxima of the 77 K
spectrum, the singlet-triplet gap of BbF is 0.58 eV, which is at least 2 times smaller than
that of alternant aromatic compounds (for alternant hydrocarbons such as anthracene or
naphthalene the value of ΔE(S-T) lies in the range of 1.3-1.6 EV (Michl, Thulstrup
1976)).

94
OLED devices of the structure ITO/NPD/CBP:BbF(7%)/BCP/AlQ3/LiF/Al were
fabricated. The device data is shown on Figures 29 and 30
0.01 0.1 1 10
1E-4
1E-3
0.01
0.1
1
10
100
1000
Voltage, V
Current density, mA/cm
2
1E-4
1E-3
0.01
0.1
1
10
100
1000
10000
Brightness, Cd/m
2

Figure 29 I-V characteristics (open squares), and luminance-voltage plot (filled
diamonds) of the OLED with BbF-doped emissive layer (see text for the device structure)

As seen from the luminance-voltage plot, the BbF device showed a moderate
brightness, and had a turn-on voltage higher than 7 V.
The quantum efficiency plot for the BbF device along with the
electroluminescence spectrum as an inset is shown on Fig.rrr. The maximum quantum
efficiency of the device is 1.0% which corresponds to the BbF efficiency in solution ~0.3

95
or below if the light collection setup and losses due to triplet exciton formation are taken
into account.
1 10 100
1E-3
0.01
0.1
1
E. Q. E., %
Current Density, mA/cm
2
300 400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Intensity (normalized)
Wavelength, nm
8V
9V
10V
11V
12V
14V

Figure 30 External quantum efficiency (open circles), and electroluminescence
spectra taken at different voltages (inset) for the OLED with BbF as an emissive dopant

Introducing a 100 Å-thick layer of mCP as an electron-blocking layer to a device
with the architecture described above only reduces the current passing through the device
(Figure 31)

96
110
1E-3
0.01
0.1
1
E. Q. E. %
Current density, mA/cm
2
300 400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Signal (normalized)
Wavelength, nm
9V
10V
11V
12V
14V
110
1E-3
0.01
0.1
1
E. Q. E. %
Current density, mA/cm
2
300 400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Signal (normalized)
Wavelength, nm
9V
10V
11V
12V
14V
0.1 1 10
1E-3
0.01
0.1
1
10
Voltage, V
Current density, mA/cm
2
1E-4
1E-3
0.01
0.1
1
10
100
Brightness, Cd/m
2

Figure 31 BbF devices with 100 Å layer of mCP between the hole-transporting
(NPD) and emissive (CBP:BbF) layers.

Compared to the device without mCP, this device has slightly lower efficiency
and brightness.
According to the report by Chiechi et al, fluoranthenes can show high quantum
yields in solution, reaching unity for some sterically hindered derivatives (Chiechi, Tseng
et al. 2006). The fluorescent quantum yield of BbF in a deaerated toluene solution that we
measured using an integrating sphere was 0.55 ± 0.1. Inferior efficiency of the BbF
OLED may result from various deactivation pathways available in CBP matrix including
concentration quenching, or inefficient energy transfer from the CBP host.

97
Indolocarbazoles
N-derivatized indolocarbazoles were reported and patented by Xerox Corp. as
hole-transporting materials (Hu, Xie et al. 2000). Other research groups studied
applications of indolocarbazoles as HTLs (Zhao, Tao et al. 2007; (Zhao, Tao et al. 2007),
hosts for phosphorescent OLEDs (Asari, Yamamoto et al. 2007), and field-effect
transistor semiconductors (Boudreault, Wakim et al. 2007). X-ray structure of 5,11-
diphenylindolo[3,2-b]carbazole (DTIC) and its solution photophysical properties were
also described (Kawaguchi, Nakano et al. 2007). The electroluminescence of ICs and
their use in OLEDs has not been reported in literature, nor has it been realized that the
specific features of the indolocarbazole electronic structure such as a very small ΔE(S-T)
(much smaller than for fluoranthenes described above) make them interesting as blue
emitters for the devices with separate singlet- and triplet exciton-harvesting pathways.
Synthesis of indolocarbazole derivatives
Substitution of hydrogen with phenyl, naphtyl, and 2,6-dimethoxyphenyl groups
at both N-atoms in indolocarbazole was successfully done for certain iodoarenes under
Ullmann-type coupling reactions conditions using Cu powder or Cu
2
SO
4
*5H
2
O catalysts
in 1,2-Dichlorobenzene or n-tridecane according to the scheme on Figure 15:

98
N
N
NH
HN
I
R R
R
R
R
R
2 +
Cu-catalyst
base, heat
R=H, OCH
3
, 9-naphtyl

Figure 32 Synthesis of N-substituted indolocarbazoles.

To introduce more steric bulk to the N-substituents, we attempted to use the
commercially available 2,4,6-Triisopropylbromobenzene in Pd-catalyzed coupling
reactions for N-derivatization of indolocarbazole.
Br

Indolocarbazole was proven to be unreactive in the reaction with 2,4,6-
Triisopropylbromobenzene. No N-substitution was detected.

99
It was stated in literature that some sterically non-hindered bromoarenes can
efficiently couple with carbazole on palladium catalysts in presence of K
2
CO
3
base
(Watanabe, Nishiyama et al. 2000). At that point, we suspected that unlike ArI-s, the
arylbromides were not reactive enough, and that the reason for the above reaction to fail
was the bromoarene: Cu-catalyzed coupling did not show any reaction between 2,4,6-
Triisopropylbromobenzene and indolocarbazole either. Thus, we concluded that only
iodoarenes on Cu-catalysts are suitable for N-aromatic substitution in indolocarbazoles.
Since we needed a more sterically crowded substituent at the nitrogen atoms of
indolocarbazole, we focused on search for suitable iodoarene precursors.
Replacing the amino-group with iodine in 2,6-Diisopropylaniline via diazo-
reaction is not easy. Multiple attempts to make 2,6-Diisopropyliodobenzene by diazo-
synthesis resulted in formation of the corresponding phenol and xylene. However, some
amount of colorless oil was isolated which appeared as one spot on a hexane/DCM TLC.
The HNMR was satisfactory for the target product, and the mass spectrum showed two
peaks consistent with 2,6-Diisopropyliodobenzene: M
+
-ion (mass 288) and the fragment
with a mass of 275, that may correspond to the detachment of one CH
3
-fragment. The
obtained product was used as a reagent in the Ullmann-type reaction with
indolocarbazole. No formation of the N-substituted indolocarbazole was detected: after 3
days of refluxing in 1,2-Dichlorobenzene, the starting indolocarbazole was recovered.

100
Halogen exchange at 9-position of anthracene allows for the iodo-derivative in
satisfactory yields. 9-Bromoanthracene was heated with large excess of KI and CuI in
HMPT at 160°C overnight.
Br
I
CuI, KI
HMPT, 160C

The obtained 9-Iodoanthracene was recrystallized from toluene/methanol and
DCM/methanol. Bromo- and iodo-derivatives have identical retention times on hexane
TLC. However, one can unambiguously tell them apart by the mass spectra and
13
C-
NMR.
Refluxing 9-Iodoanthracene with indolocarbazole in 1,2-Dichlorobenzene in
presence of Cu-powder, K
2
CO
3
and 18-Crown-6 for 7 days resulted in formation of
anthracene in high yield. The sky-blue emitting anthracene dimer (M
+
mass 354) also
came out off the chromatographic column in small quantities. No substitution of
hydrogen at N-positions in indolocarbazole was detected.
One of the available iodoarenes is 2,6-Dimethyliodobenzene.
H
3
C
I
CH
3

101
Despite expectations, indolocarbazole was inert towards Ullmann coupling with
2,6-Dimethyliodobenzene: after 3-day refluxing with Cu-catalyst and base, the starting
materials remained intact.
Therefore, looking at the array of iodoarene reagents used in N-derivatizing
indolocarbazole, a trend is noticeable: if a halo-arene precursor contains fully substituted
carbon atoms in both ortho-positions to the halogen, then the arylation of indolocarbazole
does not occur (Figure 33).

102
N
N
N
N
N
N
N
N
N
N
O
O
O
O
N
N

Figure 33 The indolocarbazoles on the left were successfully prepared using
iodoarenes with sterically non-hindered ortho-positions relative to iodine. Iodoarenes
with fully-substituted carbons at 2,6-positions showed no reactivity towards coupling
with indolocarbazole resulting in failed syntheses of the derivatives shown on the right.

Non-substituted iodobenzene, 9-iodonaphtalene with one empty orto-position, or
2,6-Dimethoxyiodobenzene having only a lone pair on the oxygen atoms, react with
indolocarbazole to form N-disubstituted species, whereas 9-Iodoanthracene, 2,6-
Dimethyliodobenzene, 2,6-Diisopropyliodobenzene, and 2,4,6-

103
Triisopropylbromobenzene (all have both ortho-positions crowded) do not show any
reactivity in N-substitution, regardless of the nature of halogen or reaction conditions.
The fact that 3,5-Bromobenzene reacts with indolocarbazole to give the corresponding N-
disubstituted derivative (TBIC) proves that the 2,6-haloarenes did not react with IC due
to steric reasons instead of bromoarenes being non reactive towards Ullmann or
Buchwald coupling (Figure 34)
N
N
NH
HN
Br
2 +
Cu, 18-crown-6
K
2
CO
3
, DCB

Figure 34 TBIC synthesis

TBIC was obtained with 50% yield (measured after hexane column purification
and before sublimation). It sublimed at 305-310°C giving a bright-yellow solid with
polycrystalline and amorphous glassy islands.
Photophysical characterization of ICs
The room temperature and 77 K emission spectra of four indolocarbazole
derivatives are shown on Figure 35. Weak phosphorescence is detectable for all of these
compounds at 77 K. The phosphorescence signal is greatly enhanced by addition of

104
methyl iodide which provides heavy atom effect to facilitate intersystem crossing to the
triplet level.

350 400 450 500 550
0.0
0.2
0.4
0.6
0.8
1.0
Intensity (normalized)
Wavelength, nm
MeIC
DTIC
NPIC
DMIC

Figure 35 Solution spectra of three indolocarbazole derivatives at room
temperature

105
350 400 450 500 550 600 650
0.0
0.2
0.4
0.6
0.8
1.0
MeIC
MeIC + CH
3
I
DTIC
NPIC + CH
3
I
DMIC + CH
3
I
Intensity (normalized)
Wavelength, nm

Figure 36 Solution spectra of three indolocarbazole derivatives at 77 K

The solution excitation spectra match the absorption reasonably for all three
systems. MeIC emission, absorption and excitation are shown on Figure 37 as an
example

106
300 350 400 450 500 550
0.0
0.2
0.4
0.6
0.8
1.0
Signal (normalized)
Wavelength, nm
RT excitation
RT emission
77K emission
RT absorption

Figure 37 Absorption, emission (RT and 77 K) and excitation spectra of MeIC in
solution

The fluorescence and phosphorescence emission peak maxima and the
corresponding singlet-triplet energy differences for DTIC, MeIC, NPIC, and DMIC are
shown in Table 1.

107
Table 1 Fluorescence and phosphorescence peak maxima (measured at 77 K
temperature), and singlet-triplet gaps in electronvolts for four indolocarbazole derivatives
DTIC MeIC DMIC NPIC
Fluorescence
λ
max
, nm
411 421 419 417
Phosphorescence
λ
max
, nm
476 489 476 496
ΔE(S-T), eV 0.33 0.31 0.36 0.48

The singlet-triplet gaps for IC-s lie within 0.3-0.4 eV with the maximum value for
NPIC of 0.48 eV, which in case of MeIC is nearly two times smaller than ΔE(S-T) for
benzo[b]fluoranthene (0.58 eV) discussed earlier.

Blue OLEDs with IC dopants as emitters
To evaluate the electroluminescence efficiency of indolocarbazole derivatives as
individual emitters, the OLEDs with the following structures were fabricated:
ITO/NPD(400 Å)/CBP:IC, 5%(250 Å)/BCP(150 Å)/Alq3(250 Å)/LiF(10 Å)/Al(1200 Å).
The data collected for OLEDs with IC = NPIC, MeIC or DTIC is shown on Figure 38.

108
0.1 1 10 100 1000
1E-4
1E-3
0.01
0.1
1
Q. E., %
C. D., mA/cm
2
024 68 10 12
1E-6
1E-4
0.01
1
100
10000
Brightness, Cd/m
2
Voltage, V
0.01 0.1 1 10
1E-6
1E-4
0.01
1
100
NPIC
MeIC
DTIC
Current Density, mA/cm
2
Voltage, V
350 400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Signal (normalized)
Wavelength, nm

Figure 38 Current-voltage characteristics, external quantum efficiencies,
luminance, and electroluminescence spectra (recorded at 10 Volts) for the devices with
indolocarbazole-doped emissive layers (see text for device structures).

The electroluminescence spectra on bottom-right of Figure 38 show that the
emission from the methylated indolocarbazole is more than 15 nm red-shifted compared
to the aromatic derivatives NPIC and DTIC. The external quantum efficiencies of the
devices for all three emitters are similar and do not exceed 1.2%. If we recall that the

109
external quantum efficiency measurements account for 20% of all emitted photons, and
that for a fluorescent emitter 75% of all formed excitons is lost because they are triplets,
the predicted value for solution quantum yields of indolocarbazoles will be close to 0.3.
This is significantly lower than the reported values (Nijegorodov, Downey 1994; (Merlet,
Birau et al. 2002; (Hadizad, Zhang et al. 2005) for indenofluorenes that are compounds
with similar molecular skeleton to indolocarbazole except for the NH-groups replaced by
sp3-carbons.
Efficiency losses in the electroluminescent device can result from various factors
such as concentration quenching or inefficient host-to-dopant energy transfer for devices
with doped emissive layers. That means that the inherent efficiency of the emitter in an
inefficient device is not necessarily low. To the best of our knowledge, the solution
quantum efficiency studies of indolocarbazoles have not been published. Therefore we
found it interesting to measure IC quantum efficiencies in the solution because these
measurements can answer questions about inherent limitations of the indolocarbazole
system as an emitter.
Fluorescence quenching by oxygen in solution
Comparison of the IC quantum yields in degassed and aerated solutions measured
using integrated sphere clearly showed that, despite previous assumptions, there is a
significant amount of oxygen quenching of fluorescence. For the same solution of MeIC
in toluene, the quantum yield measured using identical experimental parameters

110
increased from 0.34 to 0.56 after deaerating the solution with nitrogen. The degassing
effect is also reflected in the respective lifetime values for aerated and deaerated
solutions. Measuring the quantum yield as a function of degassing time showed that most
of the oxygen quenching is eliminated within first minute of treatment with nitrogen gas.
The reverse process of oxygen dissolution in the prior degassed sample is slow allowing
for easy sample handling.
Along with the quantum yields the emission lifetimes were also monitored for
various oxygen levels, and the resulting quantum yield-lifetime plots always showed
linear behavior (see Figure 39 and 40).

111
6789 10 11 12
50
60
70
80
90
100
QE, %
τ, ns
QY_DMIC, in tolene, RT
Linear Fit of Data1_B
6789 10 11 12
50
60
70
80
90
100
QE, %
τ, ns
QY_DMIC, in tolene, RT
Linear Fit of Data1_B

Figure 39 Change in fluorescent lifetime and quantum yield upon bubble-aerating
the degassed solution of DMIC in toluene; time between 1
st
and last recorded points is
approx. 1 hour

112
6 7 8 9 10 11 12
50
60
70
80
90
100
QE, %
τ, ns
DMIC in toluene, exc. WL 340nm, abs. 0.12
Linear Fit of Data1_B
67 89 10 11
35
40
45
50
55
60
65
70
75
QE, %
τ, ns
DMIC, exc WL 383nm (abs. 0.09)
Linear Fit of Data1_D
fresh
N
2
, 3 sec
N
2
, 2 min
fresh
N
2
, 2 min
6 7 8 9 10 11 12
50
60
70
80
90
100
QE, %
τ, ns
DMIC in toluene, exc. WL 340nm, abs. 0.12
Linear Fit of Data1_B
67 89 10 11
35
40
45
50
55
60
65
70
75
QE, %
τ, ns
DMIC, exc WL 383nm (abs. 0.09)
Linear Fit of Data1_D
fresh
N
2
, 3 sec
N
2
, 2 min
fresh
N
2
, 2 min

Figure 40 Change in fluorescent lifetime and quantum yield for DMIC upon N
2
-
degassing the toluene solutions. The graph on the left was recorded for excitation
wavelength of 340nm, the one on the right – 383nm
Quantum yield dependence on the excitation wavelength
Another feature that can be noted when the two plots on Fig 1a are compared is
that the quantum yields for 340 and 383nm excitation are different. The excitation
wavelengths were chosen to match the two typical absorption regions of the ICs (the
MeIC absorption spectrum is shown on Fig. 41 as a typical example for IC absorption):

113
300 350 400 450 500
0.0
0.2
0.4
0.6
Absorbance, a.u.
Wavelength, nm
MeIC absorption in toluene

Figure 41 MeIC absorption spectrum in toluene

Shown in Table 2 is the data summary on the solution quantum yields, fluorescent
lifetimes, and calculated radiative rates for all four IC derivatives for two excitation
wavelengths mentioned above.

114
Table 2 Quantum yields, fluorescent life times, and radiative rate constants (k
R
=
QE/ τ) for IC derivatives measured for degassed toluene solutions at two different
wavelengths using glass cells
NPIC MeIC DTIC DMIC
λ
exc.
, nm 340
383
343
392
340
380
340
383
QE
degassed
0.76
0.48
0.81
0.56
1.00
0.68
0.94
0.73
τ, ns 10.9
10.7
13.1
12.9
12.5
12.6
11.2
11.4
k
R
7.0x10
7
4.5x10
7
6.2x10
7
4.3x10
7
8.0x10
7

5.9x10
7

8.4x10
7
6.4x10
7

In order to rationalize the quantum yield dependence on the excitation
wavelength, several control experiments were run:
Sacrificing the quantum efficiency losses due to self-absorption, the
measurements were taken for the solution of higher concentration to allow for recording

115
and comparison of quantum yields for the exact same sample at two different
wavelengths at which the absorption intensity is 10-fold different;
Glass quvettes were replaced by quartz to minimize the possible effect of glass
absorption at shorter UV wavelengths;
The attempts were made to record the quantum yields of the well known
standards such as BASF-083, AlQ3, and quinine sulfate. The absorption and excitation
spectra for the same IC solution were quantitatively compared. The outcomes of these
control tests are as follows.
Quantum yields measured for the same (relatively concentrated) solution using
different excitation wavelengths:

Table 3 Quantum yield wavelength dependence for NPIC. Same sample in
toluene
Exc. WL Abs. QY(NPIC)
403 0.12 0.52
383 0.09 0.48
340 1.2 0.58

116
Table 4 Quantum yield wavelength dependence for MeIC. Same sample in
toluene
Exc. WL Abs. QY(MeIC)
414 0.1 0.63
392 0.09 0.60
342 1.1 0.70

Tables 3 and 4 show that the change of the quantum yield with the wavelength is
systematic in a sense that a ~10% overestimation is always observed at higher energy
(340nm) excitations.
2) Quartz vs. glass:
Table 5 The effect of replacing glass quvettes with quartz ones. 340nm excitation
wavelength was used in all 4 measurements
QY(DTIC) QY(DMIC)
Glass 1.00 1.00
Quartz 0.81 0.93

117
This test showed that using quartz quvettes minimizes the overestimations of the
quantum yields when 340nm excitation wavelength is used. This is due to the glass
absorption offset in the near-UV region.
3.a) BASF-083 standard emitter:
Table 6 Excitation wavelength dependence of the quantum yield of the BASF-083
fluorescent standard
Exc. wavelength, nm QY(BASF-083)
401 0.90
355 0.50
343 0.46

Measuring the quantum yields for the BASF-083 standard as a function of
excitation wavelength turned out to be not informative: the emitter is doped into a
polymer matrix that strongly absorbs at 340nm. The two-fold decrease in quantum yield
for 340nm compared to 401nm excitation is due to the host absorption and consequent
inefficient energy transfer to the dopant (Table 6).

118
The quantum yield of BASF-083 obtained for the excitation wavelength of 401
nm was 0.90, which is in excellent agreement with the value of 0.91 provided by BASF
for this fluorophor.
3.b) AlQ3:
Table 7 Wavelength dependence of quantum yield for AlQ
3

Exc. WL, nm Absorption (a.u.) QE(AlQ
3
)
340 0.08 0.23
397 0.12 0.19

Measurements of the AlQ
3
quantum yields generally confirmed the conclusions
made earlier for the ICs: the quantum yield values are ~10% overestimated if an
excitation at 340nm is used (Table 7).
The published value for the quantum yield of quinine sulfate is 0.55. Our numbers
for quinine sulfate solution in 0.1M H
2
SO
4
when excitation at 340nm was used range
from 0.60 to 0.65, confirming the same trend as for AlQ3.
4) Interestingly, the same 10% difference was observed between the intensities of
absorption and excitation peaks for the NPIC solution if one of the two absorption peaks
is normalized (see Figure 42)

119
300 320 340 360 380 400 420
0.0
0.2
0.4
0.6
0.8
1.0
Signal (normalized)
Wavelength, nm
NPIC in toluene, dilute
In air
excitation (mon. @420nm)
absorption
370 380 390 400 410 420
-0.02
0.00
0.02
0.04
0.06
0.08
0.10
0.12
Signal (normalized)
Wavelength, nm

Figure 42 Differences in absorption and excitation for the same solution of NPIC
in toluene. The two spectra were normalized to the height of the 340 nm peak.

It should be mentioned here that the quantum yield dependence on the excitation
wavelength is not unique for the integrating sphere setup. Realizing that such wavelength
dependence exists helped reconcile the data recorded with the traditional method used
earlier to estimate IC quantum yields. Inconsistencies in previously recorded data can
now be explained by the use of different excitation wavelengths without examining the
quantum yield dependence on the excitation wavelength.

120
The “final” values of the solution quantum yields of the compounds discussed
here have been measured for degassed toluene solutions with absorptions below 0.2 with
excitations at 340nm. These values are: 0.70, 0.81, 0.84, and 0.96 for NPIC, MeIC,
DTIC, and DMIC, respectively. If these numbers are corrected to match the quinine
sulfate quantum yield (15% overestimation upon 340nm excitation), then the following
data table for IC quantum yields, lifetimes, and radiation rates can be constructed (Table
8):

Table 8 Quinine-corrected quantum yields (QY
degas
), fluorescent life times ( τ),
and radiative rate constants (k
R
= QY/ τ) for IC derivatives measured for degassed toluene
solutions at room temperature using quartz cells
NPIC MeIC DTIC DMIC
QY
degas
,
corrected
0.60 0.72 0.73 0.82
τ, ns 10.9 13.1 12.5 11.2
k
R
, *10
-7
5.5 5.5 5.8 7.3

When available, the excitation wave for the quantum yield measurements should
be chosen within visible part of spectrum to avoid higher error bars in the UV region. For
indolocarbazoles, however, the absorption bands at 400 nm are much weaker than those

121
at 340 nm requiring the solutions to be made concentrated. Also, these weaker absorption
bands of the 400 nm region overlap significantly with fluorescence emission, making the
quantum yield measurements less accurate. These two factors dictated the choice of 340
nm excitation wavelengths for this study, with the subsequent correction for the identified
systematic error using quantum yield data for the well established standards.
Extinction coefficients
If we assume that the high-energy absorption corresponding to π- π* transitions is
approximately the same for all four compounds, and normalize the intensity to the 340nm
peak absorption of NPIC, then the trend in extinction coefficients will follow the
radiative rate constants more closely (Figure 43)

122
300 350 400 450
0
1x10
4
2x10
4
3x10
4
4x10
4
5x10
4
6x10
4
7x10
4
ε, M
-1
cm
-1
Wavelength, nm
DTIC
DMIC
MeDTIC
NPIC
in toluene
360 380 400 420 440
0.0
2.0x10
3
4.0x10
3
6.0x10
3
8.0x10
3
1.0x10
4
ε, M
-1
cm
-1
Wavelength, nm

Figure 43 Extinction coefficients of ICs normalized to the high energy absorption
peak of NPIC.

The experimental errors associated with weighing and multiple diluting are
estimated as reaching 15% in ε values.
S
1
→T
1
intersystem crossing in indolocarbazoles
The degree of intersystem crossing to the triplet state was estimated as the relative
area of the phosphorescence peak to fluorescence. The emission spectrum of NPIC in
2Me-THF at 77K showing fluorescence and phosphorescence to scale is given on Figure
44.

123
350 400 450 500 550 600 650
0
500000
1000000
1500000
2000000
2500000
3000000
3500000
Intensity, # of counts
Wavelength, nm
NPIC in 2MeTHF, 77K emission
exc. WL = 340nm
480 500 520 540 560 580
0
2000
4000
6000
8000
10000
12000
Intensity, # of counts
Wavelength, nm
4.17*10
5
4.20*10
7
408nm
488nm

Figure 44 Emission spectrum of NPIC in 2MeTHF recorded at 77 K. The
phosphorescence peak with the λ
max
= 488 nm is shown on the inset.

For NPIC, the percent of triplet emission calculated as a ratio of triplet-to-singlet
peak areas amounts for 1.0%. The corresponding values for MeIC and DTIC estimated
earlier using the same approach are 1.6% and 1.9%, respectively. Thus, the intersystem
crossing to the triplet does not appear to be more favorable for NPIC than for other IC
derivatives.
Concentration quenching; Solution
In the control experiment mentioned above for MeIC it was found that
concentration affects the quantum yield significantly. Upon diluting the solution with the

124
absorption value of 1.1 (342nm) to 0.12 absorption units, the quantum yield increases
from 0.70 to 0.81
Concentration quenching; Doped films
Quantum yield data for the CBP films doped with ICs at three different doping
concentrations is summarized in Table 9.

Table 9 Quantum yields for the films of DMIC and DTIC doped into CBP at
different doping concentrations measured under N
2
positive pressure and in air (numbers
in parentheses)
QY(DTIC) QY(DMIC)
0.7% in CBP 0.57 (0.53 in air) 0.63
2% in CBP 0.51 --
5% in CBP 0.43 (0.35 in air) 0.44

Table 9 shows that the quantum efficiencies for the IC doped films decrease with
the increased doping levels. Notably, low enough doping concentrations to eliminate
quenching lie well below the capability of our vacuum deposition setup.

125
Solids
When we began to examine the properties of DMIC, an interesting feature was
observed for its solid state room temperature emission: at first, seemingly spontaneous
change of emission color attracted attention: the color that the DMIC solution in CH
2
Cl
2

showed when illuminated by a UV-lamp changed from violet to sky-blue. The sky-blue
color was attributed to the emission from a neat solid film of DMIC that formed on the
flask walls upon evaporation of the solvent. The recorded spectrum of a neat thin film of
DMIC is show as a sky-blue line on Figure 45.
400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Intensity, normailzed
Wavelength, nm
RT DCM excitation 420nm
RT DCM emission
RT film emission
RT film excitation 470nm
77K DCM emission, CH
3
I added
77K DCM excitation 475nm

Figure 45 Absorption, emission (RT and 77 K) and excitation spectra of DMIC in
solution

126

The broad peak of the thin film emission had a λ
max
of 464 nm. The fluorescent
decay recorded for this emission had significantly non-linear character consisting of two
regions with the decay life times of 10 ns and 170 µs components (Figure 46).
0 200 400 600 800 1000
10
-1
10
1
10
3
10
5
10
7
# of counts
τ, ns
DMIC thin film on quartz; 405nm source

Figure 46 Emission decay plot for DMIC neat thin film recorded at room
temperature

Room temperature emission spectra for neat solid samples and films of DTIC,
MeIC, and NPIC are shown on Figure 47.

127
350 400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Emission intensity (normalized)
Wavelength, nm
NPIC emission
spincast film
solid
solution in toluene
350 400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Emission intensity (normalized)
Wavelength, nm
DTIC emission
spincast film
solution in toluene
neat solid in a tube
350 400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Emission intensity (normalized)
Wavelength, nm
MeIC emission
neat solid
spincast film
solution in toluene
dropcast film
350 400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Intensity (normalized)
Wavelength, nm
DTIC neat solid
emission (exc. @ 340nm)
excitation (mon. @490nm)

Figure 47 Emission spectra of neat solids and thin films of DTIC, MeIC, and
NPIC. The red-shifted broad emission and the corresponding excitation spectra for DTIC
are shown on the bottom-right.

A strongly red-shifted broad emission band was at first only observed for DMIC.
Large variations in emission peak positions and line shapes with film or solid sample
morphology are the reason why the sky-blue emitting morphs do not always appear in the
spectra of solids or films.

128
The bottom-right graph on Figure 47 demonstrates that the DTIC morph that
shows red-shifted emission also exhibits the corresponding changes in the excitation
spectrum, the latter being a mirror image of the emission spectrum.
The origin of these red-shifted bands is not yet fully understood. Since the
absorption and fluorescence spectra of ICs overlap considerably, at some point it was
assumed that the reason for the appearance of these spectral features is intensity reduction
of the 0,0-transition line due to strong self-absorption in the solid. However, this
explanation seems less likely when the excitation spectra are taken into account.
Quantum yields for the neat films and solids of DTIC and DMIC were measured
using an integrating sphere. The quantum yields appeared independent on the variation in
the spectral line shapes and emission maxima, and the corresponding QY values were
recorded as 0.25 and 0.29 for DTIC and DMIC neat solids, respectively.
3,5-substituted aromatic IC derivatives
Since the coupling reaction between 2,6-substituted haloarenes and
indolocarbazole at N-atom appears to be problematic, we thought that using haloarene
reagents having bulky groups in the aromatic ring at positions than ortho might affect the
tendency of IC towards emission self-quenching.
3,5-Ditretbutylbenzene reacted with indolocarbazole to give the corresponding
N,N-substituted derivative abbreviated as TBIC (see Figure 34 above for the TBIC

129
molecular structure) with acceptable yield. In dilute solution in toluene at room
temperature, TBIC showed the emission 418 nm, and a quantum efficiency of 0.78
(uncorrected).
The photoluminescence spectra of TBIC in neat and doped films are similar to
those for TBIC solution in both lineshapes and peak positions: TBIC doped into CBP and
in neat film show similar emission spectra with peak maxima at 417 and 415 nm,
respectively (Figure 48).

130
350 400 450 500 550 600 650 700
0.0
0.2
0.4
0.6
0.8
1.0
Emission signal (norm.)
Wavelength, nm
neat film spincast from DCM
TBIC doped into CBP, spincast film
neat crystalline solid from sublimator
0 102030405060
10
0
10
1
10
2
10
3
10
4
10
5
# of counts
τ, ns

Figure 48 Normalized room temperature emission spectra of TBIC in the neat
film, doped CBP film, and neat crystalline solid. A life-time plot for the neat crystalline
sample is shown as an inset

As seen from Figure 48, the TBIC spun-coated neat film emission doesn’t show
any red-shifted features unlike the films of the other four derivatives discussed above.
The crystalline TBIC does show weak emission with a peak at 446 nm, however, unlike
DTIC, no excitation spectrum could be recorded for the crystalline solid. The quantum

131
efficiencies for the neat films and crystals were at the error bar limits of what can be
measured with the integrating sphere, and amount for 0.1 or less.
Doped films of TBIC in CBP of different doping levels were prepared to evaluate
how the butyl groups in 3,5-positions of the phenyl affect the self-quenching of TBIC
compared to DTIC. Table 10 shows the quantum yields of 0.7 and 5% DTIC and TBIC-
doped CBP films prepared on the same day under identical conditions.

Table 10 Comparison of quantum efficiency decrease with increasing the doping
concentration for DTIC and TBIC (fresh spin-cast from CH
2
Cl
2
films held under positive
pressure of nitrogen during QE measurements)
QE (DTIC) QE (TBIC)
0.7% in CBP 0.63 0.61
5% in CBP 0.53 0.59

The film quantum yield data on Table 10 indicate that the amount of
concentration quenching for TBIC in the range of doping levels of 0.7 to 5% wt. is
smaller than for DTIC. With this information in mind, the devices with the emissive layer
of TBIC doped into CBP (5% wt) were fabricated (device data presented on Figure 49).

132
0.1 1 10
10
-8
10
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
10
1
10
2
10
3
I-V
Voltage, V
Current Density, mA/cm
2
10
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
10
1
10
2
10
3
brightness
Brightness, Cd/m
2
350 400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Intensity, normalized
Wavelength, nm
9 V
10 V
11 V
12 V
13 V
110 100
0.1
1
NPD400/CBP:TBIC(5%)250/BCP150/Alq250
E.Q.E. %
C. D., mA/cm
2

Figure 49 Current-Voltage and Brightness-Voltage plots (left), and external
quantum efficiency and EL spectrum (right) for the device with the structure
ITO/NPD(400Å)/CBP:TBIC 5%(250Å)/BCP(150Å)/AlQ3(250Å)/LiF(10Å)/Al(1200Å)

The external quantum efficiency of the TBIC:CBP OLED didn’t exceed that of
the best of other IC devices: the E.Q.E. of an CBP:MeIC device was 1.23%, and, for
comparison, the E.Q.E. of a CBP:TBIC device was 1.24%.

Conclusions
Fluorescence in solution is quenched by oxygen. Degassing leads to more than
50% solution fluorescence quantum yield increase. Fluorescence quenching by oxygen
was also confirmed by emission life time measurements.

133
Indolocarbazoles have high quantum efficiencies (>0.8) in dilute solutions and
films with low doping concentration. IC quantum yields are comparable or higher than
those reported for indenofluorenes. Indolocarbazoles have small singlet-triplet energy
level splitting (0.3-0.5eV). This small Δ(S-T) gap does not lead to high rates of
intersystem crossing and fluorescence efficiency losses.
Concentration quenching has shown to be a factor limiting the efficiency of blue
devices with ICs as fluorescent emitters. The measured quantum yield values for the
films range from ~0.3 for the neat solids to 0.7 for the films with low doping levels
(0.7%).
Some indolocarbazoles in their solid state show strongly red-shifted aggregate
emission bands ( λ
max
~ 460-470nm), the position and the spectral line shape of which is
sensitive to the morphology of the solid. The quantum yields of this aggregate emission is
in the 0.25~0.30 range.
Small Δ(S-T) and high quantum efficiencies will allow indolocarbazoles to be
used as blue fluorescent dopants in mixed fluorescent/phosphorescent WOLEDs. To
further reduce self-quenching, derivatization of ICs at positions other than N-atoms
should be examined. Also, some red-shifting of the fluorescence peak of the IC emission
is desirable. The applicability of ICs as neat emitters, and their aggregate behavior in neat
solids should be further investigated.

134
Chapter 5. Spectral line shapes of iridium cylometalated
complexes doped into ultrahigh bandgap hosts (UGHs)
Introduction
One of the limitations in the use of organic LEDs in display applications is their
often broad spectral lines. Line broadening in the electroluminescence (EL) spectra
originates from the presence of vibronic progressions in the emission line, and
inhomogeneous broadening. Narrowing of the line by external filters is a potential
solution, but causes unwanted decrease of the device efficiency. Therefore, finding the
general method to minimize spectral line broadening is very desirable. In this Chapter,
we examine the effect the host materials may have on the line widths of phosphorescent
dopants in OLED EL spectra.
Much of the phosphorescent device research to date has focused on finding and
developing new emitting materials, whose emission spectra span the visible spectrum
(Lamansky, Djurovich et al. 2001; (Lamansky, Djurovich et al. 2001; (Ranjan, Lin et al.
2003; (Tsuboyama, Iwawaki et al. 2003; (Cocchi, Virgili et al. 2004; (Holder, Langeveld
et al. 2005; (Jia, McCormick et al. 2005; (Sotoyama, Satoh et al. 2005). However, it has
been shown that varying the host environment can also produce significant changes in the
electroluminescence spectrum of the devices (Bulovic, Deshpande et al. 1999). Changing
the polarity of the host affects the intermolecular host-dopant interactions for both ground

135
and excited states of the dopant molecule. These host related energy shifts are due to
conventional solvatochromic shifts, commonly observed for emissive dyes. Changes in
the solvent (or host material) may result in marked shifts of the emission maxima. If the
excited sate posses a greater dipole moment than the ground state, an increased polarity
of the solvent/host will stabilize the excited state relative to the ground state, leading to a
red shift of the emission maximum. Conversely, for a dopant with a lower dipole
moment in its excited states, the blue shift will be observed. The principal difference
between the doped host (solid solution) and a fluid solution is that the orientations of the
host molecules are fixed and cannot reorient in response to the excited dye molecule.
Each dopant sees a unique electronic environment, leading to a potentially broader
distribution of sites in the local environment of the dopant compared to a non-polar one.
Thus, the same emitter doped into a host with the larger dipole moment may produce a
broadened emission spectrum.
A family of host materials have been reported which have very low polarities.
These materials were not chosen based on their low polarity, but were chosen for their
high singlet and triplet energies. Tetra-aryl-silane derivatives were shown to be excellent
host materials for deep-blue phosphorescent dopants, giving pure dopant emission and
high efficiency (Holmes, D'Andrade et al. 2003; (Ren, Li et al. 2004). High triplet
energy of the host is essential for minimizing the dopant-host phosphorescence
quenching. A potential benefit for the present study is the low dipole moments of these

136
aromatic materials, which we expect to minimize the affects of inhomogeneous
broadening in these host materials. Efficient red emission from phosphorescent iridium
complexes doped into the carbazol-based hosts has been extensively characterized
(Lamansky, Djurovich et al. 2001; (Lamansky, Djurovich et al. 2001; (Adamovich,
Brooks et al. 2002)
, and refs. therein
; however, there have been no reports of the same dopants
being incorporated into aryl-silane based OLEDs. Since narrowing of the site
distribution can potentially be achieved when non-polar hosts are used, one can expect to
obtain a less broadened emission spectrum from the red dopant. In order to explore this
hypothesis, here we use the wide gap, low polarity materials, tetraphenyladamantane
(4pa), and (p-bis(triphenylsilyl)benzene) (UGH2) as hosts for the red emitter bis(2-(2’-
benzo[4,5-a]-thienyl)pyridinato-N,C
3’
) iridium(acetylacetonate) (BtpIr) in organic light-
emitting devices.
Experimental
All emissive and host materials were purified by zone sublimation before use.
Solution-processed films were obtained by applying dichloromethane solutions of
corresponding host and dopant mixtures onto quartz substrates with spin rates of 3000
rpm and spinning time of 30 seconds. Vapor-deposited doped films were prepared when
film fabrication by spincoating was not feasible due to insolubility of host materials.
Vapor-deposited film fabrication was done under high vacuum according to procedure
described in Chapter 2.

137
Results and discussion
The structures of the phosphorescent emitters bis(2-(2’-benzo[4,5-a]-
thienyl)pyridinato-N,C
3’
) iridium(acetylacetonate) (BtpIr) and bis(2-phenylquinolyl-
N,C
2’
) iridium(acetylacetonate) (PQIr), and ultra-high-band-gap host materials
tetraphenyladamantane (4pa) and p-bis(triphenylsilyl)benzene (UGH2) are shown on
Figure 50.

Si Si
N
S
Ir
O
O
2
N
Ir
O
O
2
4pa UGH2
BtpIr PQIr

Figure 50 Chemical structures of the ultra-high band gap hosts and
phosphorescent emitters discussed in this Chapter

138
PQIr and BtpIr (Figure 50, bottom structures) are efficient phosphorescent iridium
complexes that emit at 600 and 615 nm, respectively (emission peak maxima for dilute
solutions in CH
2
Cl
2
). The structure of these compounds was optimized using DFT
calculations (B3LYP/6-31G*), and for the optimized geometries the dipole moments
were calculated as 1.4 D for PQIr, and 3.3 D for BtpIr. The host materials depicted on
Scheme 1 are characterized by very high HOMO-LUMO energy gaps, very low polarity,
and propensity to π-staking in the solid films.
Once the BtpIr-doped thin films were fabricated and their photoluminescence
spectra recorded, it became obvious that the line shape of the spectra of BtpIr varies
significantly with changing the host material. Figure 51 demonstrates that the emission
from the dopant exhibits the narrowest spectral line and has the lowest Huang-Rhys
parameter values when BtpIr is doped into the 4PA host.

139
450 500 550 600 650 700 750 800
0.0
0.2
0.4
0.6
0.8
1.0
Intensity, normalized
Wavelength, nm
solution in CH
2
Cl
2
in 4PA, vapor-deposited
in CBP, spin-coated

Figure 51 Photoluminescence spectra from thin films with BtpIr as an emitter
doped into 4PA and CBP hosts, and the solution photoluminescence of BtpIr at room
temperature

OLED devices containing the emissive layer of similar composition to the doped
films were fabricated. The devices (see Figure 52) consisted of an indium tin oxide (ITO)
anode, a 400-Å-thick 4,4’-bis[N-(1-naphthyl)-N-phenyl-amino]biphenyl (NPD) hole
transport layer (HTL), a 250-Å-thick light-emitting layer (EML, consisting of a host
doped with 7% of the guest phosphor); an electron blocking layer; a 300-Å-thick 2,9-
dimethyl-4,7-diphenyl-phenanthroline (BCP) electron transport layer (ETL), and a
cathode comprised of a 10-Å-thick layer of LiF and 1200-Å-thick aluminum film.

140

BCP 30nm
NPD 40nm
LiF 1nm
Al 120nm
ITO on glass
(EBL 10-15nm)
UGH:Ir-dopant(7%) 25nm

Figure 52 The device architecture. UGH is either tetraphenyladamantane (4pa) or
(p-bis(triphenylsilyl)benzene) (UGH2).

Once the electroluminescence spectra of the devices with 4PA and UHG2 hosts
were recorded, it was realized that they contained a strong blue component which was the
electroluminescence from the hole-transporting material NPD. It was concluded that an
electron-blocking layer (EBL) will be necessary to eliminate the NPD emission. Figures
53-57 illustrate the effect of adding electron-blocking layers to the 4PA:BtpIr devices.

141
110
10
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
10
1
10
2
10
3
Current density, mA/cm
-2
Voltage, V
NPD/4pa:BtpIr(7%)/BCP
NPD/mCP/4pa:BtpIr(7%)/BCP
NPD/Irppz/4pa:BtpIr(7%)/BCP

Figure 53 Current-voltage plots for the devices containing emissive layers of
BtpIr doped into 4PA. Device structures: ITO/NPD(40nm)/EBL(x)/4PA:BtpIr 7%
(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm), EBL = mCP(10nm) (red), Irppz(10nm) (blue),
or no blocking layer (black).

142
10
-3
10
-2
10
-1
10
0
10
1
10
2
10
3
0.1
1
E. Q. E., %
C. d., mA/cm
2
110
10
-4
10
-3
10
-2
10
-1
10
0
10
1
10
2
10
3
Brightness, Cd/m
2
Voltage, V

Figure 54 Brightness-voltage data for the devices containing emissive layers of
BtpIr doped into 4PA. Device structures: ITO/NPD(40nm)/EBL(x)/4PA:BtpIr 7%
(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm), EBL = mCP(10nm) (red), Irppz(10nm) (blue),
or no blocking layer (black). External quantum efficiencies are shown as an inset.

143
300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
no EBL
intensity, normalized
Wavelength, nm
8V
9V
10V
11V

Figure 55 Electroluminescence spectra of BtpIr/4PA devices with no electron-
blocking layer. Device structure: ITO/NPD(40nm)/4PA:BtpIr
7%(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm)
300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
EBL = mCP
intensity, normalized
Wavelength, nm
8V
9V
10V
11V

Figure 56 Electroluminescence spectra of BtpIr/4PA devices with an mCP
electron-blocking layer. Device structure: ITO/NPD(40nm)/mCP(10nm)/4PA:BtpIr
7%(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm)

144
300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
EBL = Irppz
Intensity, normalized
Wavelength, nm
8V
9V
10V
11V

Figure 57 Electroluminescence spectra of BtpIr/4PA devices with an Irppz
electron-blocking layer. Device structure: ITO/NPD(40nm)/mCP(10nm)/4PA:BtpIr
7%(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm)

The electroluminescence spectra of the device with no EBL show strong voltage
dependence (note the increasing amount of the NPD emission with rising voltage). A 100
Å-thick layer of Irppz placed between NPD and the BtpIr-doped emissive layer was
sufficient to block the blue emission from NPD. An mCP EBL did not introduce
significant changes to the 4PA:BtpIr device. The lack of hole-blocking ability of mCP
compared to Irppz may be due to lower LUMO energy of mCP (-2.3 eV vs -2.1 eV for
Irppz (Adamovich, Brooks et al. 2002)) which does not impose a sufficiently high barrier

145
for electrons entering the HTL. The boundary orbital energy level arrangement for the
device materials is visualized by Figure 58

4.99
2.4
5.98
2.3
5.03
2.1
5.3
2.7
7.0
(?)
6.5
2.9
4.1
4.7
ITO
NPD
Irppz/
mCP
BtpIr
in
4PA
BCP
Al

Figure 58 Energy levels (in electronvolts) for the 4PA:BtpIr device. The dashed
box corresponds to the energy levels of mCP. The HOMO values were obtained using
UPS (D'Andrade, Datta et al. 2005); the available LUMO values are taken from the work
of Adamovich et al (Adamovich, Brooks et al. 2002)

146
The devices of similar structures were fabricated using the UGH2 host matrix
instead of 4PA (Figure 59)

0.1 1 10
1E-7
1E-6
1E-5
1E-4
1E-3
0.01
0.1
1
10
C. D., mA/cm
2
Voltage, V
1E-3 0.01 0.1 1 10
0.1
1
Q. E., %
C. D., mA/cm
2
0.1 1 10
10
-5
10
-3
10
-1
10
1
10
3
Brightness, Cd/m
2
Voltage, V
300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
Intensity, normalized
Wavelength, nm

Figure 59 I-V, brightness-voltage, electroluminescence spectrum, and quantum
efficiency plots for the ITO/NPD(40nm)/Irppz(10nm)/UGH2:BtpIr,
7%(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm) devices

147
For UGH2 devices too, Irppz proved to be an efficient EBL material: the EL
spectrum on Figure 7 recorded at 12 Volts shows no NPD component to the emission. It
is worth pointing out that, other parameters being equal, the quantum efficiency of the
UGH2:BtpIr device was 3 times higher than that of a similar device with the 4PA host
(inset on the right graph on Figure 59).
To complete the picture, 4PA:PQIr devices were also fabricated (Figure 60), and
the electroluminescence spectra were compared to the emission from the PQIr fluid
solution.
1E-4 0.01 1 100
0.01
0.1
1
Q. E., %
C. D., mA/cm
2
02 46 8 10 12
1E-7
1E-5
1E-3
0.1
10
1000
Voltage, V
Current density, mA/cm
2
0.01
0.1
1
10
100
Brightness, Cd/m
2

Figure 60 Current-voltage and brightness-voltage plots (on the right), and the E.
Q. E. plot (on the right) for the devices with the structure ITO/NPD(40nm)4PA:PQIr, 7%
(25nm)/BCP(30nm)/LiF(1nm)/Al(120nm).

Once a proper electron-blocking layer is introduced between the HTL and the
emissive layer, the unwanted emission coming from the HTL material is eliminated, and

148
thus obtained EL spectra from the blocked BtpIr and PQIr devices confirmed the
conclusions drawn for the photoluminescence spectral line shape dependences in the
doped films. The comparison of the electroluminescence spectra of the devices with
emissive layers of BtpIr doped into two UHG hosts - 4PA and UGH2 - and CBP is shown
on Figure 61.
500 550 600 650 700 750
0.0
0.2
0.4
0.6
0.8
1.0
intensity, normalized
Wavelength, nm
in CBP
a
in UGH2, OLED EL
in 4PA, OLED EL

Figure 61 Electroluminescence spectra of the OLEDs with emissive layers
consisting of BtpIr doped into 4PA, UGH2, and CBP (7% doping by weight).
a
CBP:BtpIr
EL data taken from Lamansky et al (Lamansky, Djurovich et al. 2001). The devices
structures correspond to the one shown on Figure 3.

The electroluminescent spectra of the OLEDs confirm the trends described above
for the photoluminescence of doped thin films: the EL spectra show similar line shape

149
changes for BtpIr doped into the three hosts; the decrease in the relative intensity of the
vibrational band for the UGH hosts compared to the conventional CBP host was
confirmed.
To see how the choice of host material influences the emission lines of other
iridium phosphors, we fabricated OLEDs with bis(2-phenylquinolyl-N,C
2’
)
acetylacetonate (PQIr) as an emissive dopant. For the hosts we chose the same ones that
gave the maximal and minimal Huang-Rys parameters for the described above BtpIr
emission spectra. Figure 62 represents the emission spectra of PQIr doped into CBP and
4PA hosts, and the PQIr solution spectrum in dichloromethane.

150
400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
intensity, normalized
Wavelength, nm
NPD/CBP:PQIr(7%)/BCP/AlQ
3
NPD/4PA:PQIr(7%)/BCP
PQIr in DCM solution at RT

Figure 62 Electroluminescence spectra of the devices with PQIr doped into 4PA
and CBP hosts (device structures ITO/NPD(40nm)/4PA: PQIr 7%
(25nm)/BCP(30nm)/LiF(1nm)/(120nm); ITO/NPD(40nm)/CBP: PQIr 7%
(25nm)/BCP(15nm)/AlQ
3
(25nm)LiF(1nm)/(120nm) and in fluid dichloromethane
solution.

Comparison of Figures 61 and 62 indicates that the emission spectra of these two
iridium phosphors are different in several aspects: the BtpIr spectra show vibrational
progression the intensity of which changes with host variation; PQIr, on the other hand,
shows broadened featureless lines similar in all environments. The electronic transition
corresponding to the observed red emission of PQIr possesses more charge-transfer

151
character than BtpIr hence broader spectral line, and the absence of vibrational features.
Also, the dipole moment of BtpIr calculated for the DFT-optimized structure is more than
twice of that of PQIr. Higher dipole moment means stronger intermolecular interactions
and more non-homogeneous broadening. As a result of these factors, the spectral line
shapes show no dependence on the host material for PQIr emission, whereas using the
non-polar UGH hosts in case of BtpIr allowed to observe the decrease in unwanted
vibrational satellites in the emission spectrum also with some decrease in the main
spectral line broadening.

Conclusions
Low-polar ultra-high band gap host materials UGH2 and 4PA have been shown to
decrease Huang-Rys parameters of the BtpIr emission at least two-fold compared to a
conventional host material CBP. The effect of decreasing vibrational satellites in the
BtpIr spectrum was comparable in both thin film photoluminescence and OLED
electroluminescence. The line-shape dependence on the nature of the host material was
not observed for another iridium phosphorescent dopant PQIr. This difference in behavior
of these two emitters was attributed to host-dopant intermolecular interaction which is
stronger for the more polar BtpIr dopant. Irppz was shown to be an appropriate electron-
blocking material for UGH-based phosphorescent devices with BtpIr and PQIr emissive
dopants.

152
References
Adamovich, V., Brooks, J., Tamayo, A., Alexander, A., Djurovich, P. I., D’Andrade, B.
W., C.;, A., Forrest, S. R., Thompson, M. E. (2002). "High efficiency single dopant white
electrophosphorescent light emitting diodes." New Journal of Chemistry 26(9): 1171-
1178.

Anderson, J. D., Paulsen, E. S., Dearden, D. V. (2003). "Alkali metal binding energies of
dibenzo-18-crown-6: Experimental and computational results." International Journal of
Mass Spectrometry 227(1): 63-76.

Asari, T., Yamamoto, T., Komori, M., Kai, T. (2007). Organic electroluminescent device.
PCT Int. Appl. P. I. Appl. Japan. .

Baldo, M. A., Forrest, S. R. (2001). "Interface-limited injection in amorphous organic
semiconductors." Phys. Rev. B 64: 085201-17.

Baldo, M. A., Lamansky, S., Burrows, P. E., Thompson, M. E., Forrest, S. R. (1999).
"Very high-efficiency green organic light-emitting devices based on electro-
phosphorescence." Appl. Phys. Lett. 75(1): 4-6.

Baldo, M. A., O'Brien, D. F., Thompson, M. E., Forrest, S. R. (1999). "Excitonic singlet-
triplet ratio in a semiconducting organic thin film." Physical Review B: Condensed
Matter and Materials Physics 60(20): 14422-14428.

Bhattacharjee, A., Mallik, B. (1990). "Electrical conduction in ferrocene and its
derivatives." Bulletin of Electrochemistry 6(9): 780-784.

Bhaumik, M. L., El-Sayed, M. A. (1965). "Studies on the triplet-triplet energy transfer to
rare-earth chelates." J. Phys. Chem. 69(1): 275-280.

Bing, Y. J., Leung, L. M., Menglian, G. (2004). "Synthesis of effcient blue and red light
emitting phenanthroline derivatives containing both hole and electron transporting
properties." Tetrahedron Lett. 45: 6361–6363.

Blochwitz, J., Pfeiffer, M., Fritz, T., Leo, K. (1998). "Low voltage organic light emitting
diodes featuring doped phthalocyanine as hole transport material." Appl. Phys. Lett.
73(6): 729-731.

153
Borek, C., Hanson, K., Djurovich, P. I., Thompson, M. E., Aznavour, K., Bau, R., Sun,
Y., Forrest, S. R., Brooks, J., Michalski, L., Brown, J. (2007). "Highly efficient, near-
infrared electrophosphorescence from a pt-metalloporphyrin complex." Angewandte
Chemie, International Edition 46(7): 1109-1112.

Boudreault, P.-L. T., Wakim, S., Blouin, N., Simard, M., Tessier, C., Tao, Y., Leclerc, M.
(2007). "Synthesis, characterization, and application of indolo[3,2-b]carbazole
semiconductors." J. Am. Chem. Soc. 129(29): 9125-9136.

Bradley, A., Hammes, J. P. (1963). "Electrical properties of thin organic films." Journal
of the Electrochemical Society 110(1): 15-22.

Brisset, H., Navarro, A.-E., Moggia, F., Jousselme, B., Blanchard, P., Roncali, J. (2007).
"Electrosynthesis of a functional conducting polymer incorporating ferrocene unit from
an edot-based bithiophenic precursor." Journal of Electroanalytical Chemistry 603(1):
149-154.

Brooks, J., Babayan, Y., Lamansky, S., Djurovich, P. I., Tsyba, I., Bau, R., Thompson,
M. E. (2002). "Synthesis and characterization of phosphorescent cyclometalated platinum
complexes." Inorg. Chem. 41(12): 3055-3066.

Bulovic, V., Deshpande, R., Thompson, M. E., Forrest, S. R. (1999). "Tuning the color
emission of thin film molecular organic light emitting devices by the solid state solvation
effect " Chem. Phys. Lett. 308(3-4): 317-322.

Carolan, N., Forster, R. J., O'Fagain, C. (2007). "Covalent attachment of ferrocene to
soybean peroxidase glycans: Electron transfer mediation to redox enzymes."
Bioconjugate Chemistry 18(2): 524-529.

Chao, T.-C., Lin, Y. T., Yang, C.-Y., Hung, T. S., Chou, H.-C., Wu, C.-C., Wong, K.-T.
(2005). "Highly efficient uv organic light-emitting devices based on bi(9,9-
diarylfluorene)s." Adv. Mater., Weinheim (Germany) 8: 992-996.

Chiechi, R. C., Tseng, R. J., Marchioni, F., Yang, Y., Wudl, F. (2006). "Efficient blue-
light-emitting electroluminescent devices with a robust fluorophore: 7,8,10-
triphenylfluoranthene." Adv. Mater., Weinheim (Germany) 18: 325–328.

Cocchi, M., Kalinowski, J., Virgili, D., Fattori, V., Develay, S., Williams, J. A. G.
(2007). "Single-dopant organic white electrophosphorescent diodes with very high
efficiency and its reduced current density roll-off." Appl. Phys. Lett. 90(16): 163508/1-
163508/3.

154

Cocchi, M., Virgili, D., Sabatini, C., Fattori, V., Di Marco, P., Maestri, M., Kalinowski,
J. (2004). "Highly efficient organic electroluminescent devices based on cyclometallated
platinum complexes as new phosphorescent emitters." Synth. Met. 147(1-3): 253-256.

Crabtree, R. H., Mingos, D. M., Eds. (2007). Comprehensive organometallic chemistry
iii. Oxford, Elsevier.

D'Andrade, B. W., Datta, S., Forrest, S. R., Djurovich, P., Polikarpov, E., Thompson, M.
E. (2005). "Relationship between the ionization and oxidation potentials of molecular
organic semiconductors." Organic Electronics 6(1): 11-20.

Dexter, D. L. (1953). "A theory of sensitized luminescence in solids." J. Chem. Phys.
21(5): 836-850.

Dye, J. L., Ellaboudy, A. (1984). "Crystalline alkalides and electrides." Chem. Ber. 20:
210-215.

Farre, C., Spinelli, N., Bouchet, A., Marquette, C., Mandrand, B., Garnier, F., Chaix, C.
(2007). "Synthesis and electropolymerization studies of water-soluble pyrrole-ferrocene
derivatives towards biochip device application." Synth. Met. 157(2-3): 125-133.

Forrest, S. R., Bradley, D. D. L., Thompson, M. E. (2003). "Measuring the efficiency of
organic light-emitting devices." Adv. Mater., Weinheim (Germany) 15(13): 1043-1048.

Fromhold Jr., A. T. (1963). "Space charge in growing oxide films." J. Chem. Phys. 38(1):
282-283.

Fry, A. J., Liu, R. S. H., Hammond, G. S. (1996). "Mechanisms of photochemical
reactions in solution. Xli. Comparison of rates of fast triplet quenching reactions." J. Am.
Chem. Soc. 88(21): 4781-4782.

Gebeyehu, D., Walzer, K., He, G., Pfeiffer, M., Leo, K., Brandt, J., Gerhard, A., St¨oßel,
P., Vestweber, H. (2005). "Highly efficient deep-blue organic light-emitting diodes with
doped transport layers." Synthetic Metals 148: 205-211.

Getty, S. A., Engtrakul, C., Wang, L., Liu, R., Ke, S.-H., Baranger, H. U., Yang, W.,
Fuhrer, M. S., Sita, L. R. (2005). "Near-perfect conduction through a ferrocene-based
molecular wire." Physical Review B: Condensed Matter and Materials Physics 71(24):
241401/1-241401/4.

155
Gilbert, A., Bagot, J. (1991). "Essentials of molecular photochemistry." 1st ed. Oxford,
UK, Blackwell Scientific: 173.

Grice, A. W., Bradley, D. D. C., Bernius, M. T., Inbasekaran, M., Wu, W. W., Woo, E. P.
(1998). "High brightness and efficiency blue light-emitting polymer diodes." Appl. Phys.
Lett. 73(5): 629-631.

Grozea, D., Turak, A., Feng, X. D., Lu, Z. H., Johnson, D., Wood, R. (2002). "Chemical
structure of al/lif/alq interfaces in organic light-emitting diodes." Appl. Phys. Lett.
81(17): 3173-3175.

Hadizad, T., Zhang, J., Wang, Z. Y., Gorjanc, T. C., Py, C. (2005). "A general synthetic
route to indenofluorene derivatives as new organic semiconductors " Org. Lett. 6(5): 795-
797.

Herkstroeter, W. G. (1975). "Triplet energies of azulene, beta-carotene, and ferrocene." J.
Am. Chem. Soc. 97(15): 4161-4167.

Holder, E., Langeveld, B. M. W., Schubert, U. S. (2005). "New trends in the use of
transition metal-ligand complexes for applications in electroluminescent devices."
Advanced Materials (Weinheim, Germany) 17(9): 1109-1121.

Holmes, R. J., D'Andrade, B. W., Forrest, S. R., Ren, X., Li, J., Thompson, M. E. (2003).
"Efficient, deep-blue organic electrophosphorescence by guest charge trapping." Appl.
Phys. Lett. 83(18): 3818-3820.

Horwitz, C. P., Suhu, N. Y., Dailey, G. C. (1992). "Synthesis, characterization and
electropolymerization of ferrocene monomers with aniline and phenol substituents." J.
Electroanal. Chem. 324: 79-91.

Hosokawa, C., Higashi, H., Nakamura, H., Kusumoto, T. (1995). "Highly efficient blue
electroluminescence from a distyrylarylene emitting layer with a new dopant." Appl.
Phys. Lett. 67(26): 3853-3855.

Hu, N.-X., Xie, S., Popovic, Z. D., Ong, B., Hor, A.-M. (2000). "Novel high tg hole-
transport molecules based on indolo[3,2-b]carbazoles for organic light-emitting devices."
Synthetic Metals 111-112: 421-424.

Hung, L. S., Tang, C. W., Mason, M. G. (1997). "Enhanced electron injection in organic
electroluminescence devices using an al/lif electrode." Appl. Phys. Lett. 70: 152-154.

156
Hung, M.-C., Liao, J.-L., Chen, S.-A., Chen, S.-H., Su, A.-C. (2005). "Fine tuning the
purity of blue emission from polydioctylfluorene by end-capping with electron-deficient
moieties " J. Am. Chem. Soc. 127(42): 14576-14577.

Hunter, T. B., Tyler, P. S., Smyrl, W. H., White, H. S. (1987). "Impedance analysis of
poly(vinylferrocene) films. The dependence of diffusional charge transport and exchange
current density on polymer oxidation state." Journal of the Electrochemical Society
134(9): 2198-2204.

Jabbour G. E.; Kippelen, B. A., N. R.; Peyghambarian, N. (1998). "Aluminum based
cathode structure for enhanced electron injection in electroluminescent organic devices."
Appl. Phys. Lett. 73(9): 1185-1187.

Jia, W. L., McCormick, T., Tao, Y., Lu, J.-P., Wang, S. (2005). "New phosphorescent
polynuclear cu(i) compounds based on linear and star-shaped 2-(2'-
pyridyl)benzimidazolyl derivatives: Syntheses, structures, luminescence, and
electroluminescence." Inorg. Chem. 44(16): 5706-5712.

Jiang, X. Y., Zhang, Z. L., Zhao, W. M., Zhu, W. Q., Zhang, B. X., Xu, S. H. (2000).
"White-emitting organic diode with a doped blocking layer between hole- and electron-
transporting layers." Journal Of Physics D-Applied Physics 33(5): 473-476.

Johansson, N., Osada, T., Stafstrom, S., Salaneck, W. R., Parente, V., dos Santos, D. A.,
Crispin, X., Bredas, J. L. (1999). "Electron structure of tris(8-hydroxyquinoline)
aluminum thin films in the pristine and reduced states." J. Chem. Phys. 111(5): 2157-
2163.

Johansson, N., Osada, T., Stafstrom, S., Salaneck, W. R., Parente, V., dos Santos, D. A.,
Crispin, X., Bredas, J. L. (1999). "Electron structure of tris(8-hydroxyquinoline)
aluminum thin films in the pristine and reduced states." J. Chem. Phys. 111(5): 2157-
2163.

Kawaguchi, K., Nakano, K., Nozaki, K. (2007). "Synthesis of ladder-type pi-conjugated
heteroacenes via palladium-catalyzed double n-arylation and intramolecular o-arylation."
J. Org. Chem. 72(14): 5119-5128.

Kawamura, Y., Yanagida, S., Forrest, S. R. (2002). "Energy transfer in polymer
electrophosphorescent light emitting devices with single and multiple doped luminescent
layers." J. Appl. Phys. 92(1): 87-93.

157
Kealy, T. J., Pauson, P. L. (1951). "A new type of organo-iron compound." Nature 168:
1039 - 1040

Kido, J., Matsumoto, T. (1998). "Bright organic electroluminescent devices having a
metal-doped electron-injecting layer." Appl. Phys. Lett. 73(20): 2866-2868.

Kido, J., Shionoya, H., Nagai, K. (1995). "Single-layer white light-emitting organic
electroluminescent devices based on dye-dispersed poly(n-vinylcarbazole)." Appl. Phys.
Lett. 67(16): 2281-2283.

Kikuchi, M., Kikuchi, K., Kokubun, H. (1974). "A study of quenching of triplets by
ferrocene." Bull. Chem. Soc. Japan 47: 1331-1333.

Knapp, R., Rehahn, M. (1993). "Palladium-catalyzed arylation of ferrocene derivatives:
A convenient high yield route to 1,1’-bis(halophenyl)ferrocenes." J. Organomet. Chem.
452: 235-240.

Ko, C. W., Tao, Y. T. (2001). "Bright white organic light-emitting diode." Appl. Phys.
Lett. 79(25): 4234-4236.

Kobayashi, N., Kijima, M. (2007). "Blue electroluminescent properties of poly(n-
arylcarbazole-2,7-ylene) homopolymers." Appl. Phys. Lett. 91: 081113-1/081113-3.

Lamansky, S., Djurovich, P. I., Murphy, D., Abdel-Razzaq, F., Kwong, R., Tsyba, I.,
Bortz, M., Mui, B., Bau, R., Thompson, M. E. (2001). "Synthesis and characterization of
phosphorescent cyclometalated iridium complexes." Inorg. Chem. 40(7): 1704-1711.

Lamansky, S., Djurovich, P. I., Murphy, D., Abdel-Razzaq, F., Lee, H.-F., Adachi, C.,
Burrows, P. E., Forrest, S. R., Thompson, M. E. (2001). "Highly phosphorescent
bis'cyclometalated iridium complexes: Synthesis, photophysical characterization and use
in organic light emitting diodes." J. Am. Chem. Soc. 123(18): 4304-4312.

Lee, J., Park, Y., Kim, D. Y., Chu, H. Y., Lee, H., Do, L.-M. (2003). "High efficiency
organic light-emitting devices with al/naf cathode." Appl. Phys. Lett. 82(2): 173-175.

Lee, S. T., Wang, Y. M., Hou, X. Y., Tang, C. W. (1999). "Interfacial electronic
structures in an organic light-emitting diode." Appl. Phys. Lett. 74(5): 670-672.

Li, J., Djurovich, P. I., Alleyne, B. D., Yousufuddin, M., Ho, N. N., Thomas, J. C.,
Peters, J. C., Bau, R., E.;, T. M. (2005). "Synthetic control of excited-state properties in
cyclometalated ir(iii) complexes using ancillary ligands." Inorg. Chem. 44(6): 1713-1727.

158

Liang, F., Wang, L., Ma, D., Jing, X., Wang, F. (2002). "Oxadiazole-containing material
with intense blue phosphorescence emission for organic light-emitting diodes." Appl.
Phys. Lett. 81(1): 4-6.

Liao, C.-H., Lee, M.-T., Tsai, C.-H., Chen, C. H. (2005). "Highly efficient blue organic
light-emitting devices incorporating a composite hole transport layer." Appl. Phys. Lett.
86: 203507(1)-203507(3).

Liu, J., Min, C., Zhou, Q., Cheng, Y., Wang, L., Ma, D., Jing, X., Wang, F. (2006). "Blue
light-emitting polymer with polyfluorene as the host and highly fluorescent 4-
dimethylamino-1,8-naphthalimide as the dopant in the sidechain." Appl. Phys. Lett. 88:
083505.

Liu, Q.-D., Lu, J., Ding, J., Day, M., Tao, Y., Barrios, P., Stupak, J., Chan, K., Li, J., Chi,
Y. (2007). "Monodisperse starburst oligofluorene-functionalized 4,4',4''-tris(carbazol-9-
yl)-triphenylamines: Their synthesis and deep-blue fluorescence properties for organic
light-emitting diode applications." Adv. Funct. Mater. 17(6): 1028-1036.

Liu, Z., Pinto, J., Soares, J., Pereira, E. (2001). "Efficient multilayer organic light
emitting diode." Synthetic Metals 122(1): 177-179.

Lo, M. Y., Sellinger, A. (2006). "Highly fluorescent blue-emitting materials from the
heck reaction of triphenylvinylsilane with conjugated dibromo aromatics." Synlett 18:
3009-3012.

Mallik, B., Bhattacharjee, A. (1989). "Adsorption-induced unusual changes in the
electrical conductivity of ferrocene." J. Phys. Chem. Solids 50(11): 1113-1119.

Marcus, R. A. (1956). "On the theory of oxidation-reduction reactions involving electron
transfer." J. Chem. Phys. 24(5): 966-978.

Marcus, R. A. (1993). "Electron transfer reactions in chemistry: Theory and experiment."
Angewandte Chemie International Edition in English 32(8): 1111-1121.

Mariani, R. D., Abruna, H. D. (1987). "Transport properties of liquid crystal doped
films of polyvinylferrocene." Electrochimica Acta 32(2): 319-23.

Mason, M. G., Tang, C. W., Hung, L.-S., Raychaudhuri, P., Madathil, J., Yan, L., Le, Q.
T., Gao, Y., Lee, S.-T., Liao, L. S., Cheng, L. F., Salaneck, W. R., dos Santos, D. A.,

159
Bredas, J. L. (2001). "Interfacial chemistry of alq3 and lif with reactive metals." J. Appl.
Phys. 89(5): 2756-2765.

Meerheim, R., Walzer, K., Pfeiffer, M., Leo, K. (2006). "Ultrastable and efficient red
organic light emitting diodes with doped transport layers." Appl. Phys. Lett. 89(6):
061111/1-061111/3.

Merlet, S., Birau, M., Wang, Z. Y. (2002). "Synthesis and characterization of highly
fluorescent indenofluorenes." Org. Lett. 4(13): 2157-2159.

Michaux, G., Reisse, J. (1982). "Solution thermodynamic studies. Part 6. Enthalpy-
entropy compensation for the complexation reactions of some crown ethers with alkaline
cations: A quantitative interpretation of the complexing properties of 18-crown-6."
Journal of the American Chemical Society 104(25): 6895-6899.

Michl, J., Thulstrup, E. W. (1976). "Why is azulene blue and anthracene white? A
simple mo picture." Tetrahedron 32(2): 205-209.

Myers, R. T. (1980). "Coordination and the size of ions and ligands." Inorganic and
Nuclear Chemistry Letters 16(8, i): 329-30.

Nakayama, T., Hiyama, K., Furukawa, K., Ohtani, H. (2007). "Development of
phosphorescent white oled with extremely high power efficiency and long lifetime." SID
conference proceedings.

Nath Bera, R., Mallik, B. (2003). "Preparation and electrical conduction of octadecanoyl
ferrocene organized in langmuir-blodgett films." Molecular Crystals and Liquid Crystals
392: 59-67.

Nijegorodov, N. I., Downey, W. S. (1994). "The influence of planarity and rigidity on the
absorption and fluorescence parameters and intersystem crossing rate constant in
aromatic molecules." J. Phys. Chem. 98: 5639-5643.

Nijegorodov, N. I., Mabbs, R., Downey, W. S. (2001). "Evolution of absorption,
fluorescence, laser and chemical properties in the series of compounds perylene,
benzo(ghi)perylene and coronene." Spectrochimica Acta A 57(13): 2673-2685.

Nitzan, A. (2001). "A relationship between electron-transfer rates and molecular
conduction." J. Phys. Chem. A 105: 2677-2679.

160
Palilisa, L. C., Ma¨kinen, A. J., Uchida, M., Kafafi, Z. H. (2003). "Highly efficient
molecular organic light-emitting diodes based on exciplex emission." Appl. Phys. Lett.
82(14): 2209-2211.

Parthasarathy, G., Shen, C., Kahn, A., Forrest, S. R. (2001). "Lithium doping of
semiconducting organic charge transport materials." J. Appl. Phys. 89(9): 4986-4992.

Piromreun, P., Oh, H.-S., Shen, Y., Malliaras, G., Scott, J. C., Brock, P. J. (2000). "Role
of csf on electron injection into a conjugated polymer." Appl. Phys. Lett. 77(15): 2403-
2405.

Pittman, C. U., Jr.;, Surynarayanan, B., Sasaki, Y. (1976). "Mixed valence,
semiconducting ferrocene-containing polymers." Advances in Chemistry Series 150: 46-
55.

Pope, M., Kallmann, H. P., Magnante, P. J. (1963). "Electroluminescence in organic
crystals." J. Chem. Phys. 38(8): 2042-2043.

Qin, D., Tao, Y. (2005). "White organic light-emitting diode comprising of blue
fluorescence and red phosphorescence." Appl. Phys. Lett. 86: 113507(1)-113507(3).

Radhakrishnan, S., Paul, S. (2007). "Conducting polypyrrole modified with ferrocene for
applications in carbon monoxide sensors." Sensors and Actuators, B: Chemical B125(1):
60-65.

Ranjan, S., Lin, S.-Y., Hwang, K.-C., Chi, Y., Ching, W.-L., Liu, C.-S., Tao, Y.-T.,
Chien, C.-H., Peng, S.-M., Lee, G.-H. (2003). "Realizing green phosphorescent light-
emitting materials from rhenium(i) pyrazolato diimine complexes " Inorg. Chem. 42(4):
1248 -1255.

Ren, X., Li, J., Holmes, R. J., Djurovich, P. I., Forrest, S. R., Thompson, M. E. (2004).
"Ultrahigh energy gap hosts in deep blue organic electrophosphorescent devices."
Chemistry of Materials 16(23): 4743-4747.

Rumi, M., Ehrlich, J. E., Heikal, A. A., Perry, J. W., Barlow, S., Hu, Z., McCord-
Maughon, D., Parker, T. C., Rockel, H., Thayumanavan, S., Marder, S. R., Beljonne, D.,
Bredas, J.-L. (2000). "Structure-property relationships for two-photon absorbing
chromophores: Bis-donor diphenylpolyene and bis(styryl)benzene derivatives " J. Am.
Chem. Soc. 122(39): 9500-9510.

161
Segal, M., Baldo, M. A., Holmes, R. J., Forrest, S. R., Soos, Z. G. (2003). "Excitonic
singlet-triplet ratios in molecular and polymeric organic materials." Physical Review B:
Condensed Matter and Materials Physics 68(7): 075211/1-075211/14.

Setayesh, S., Grimsdale, A. C., Weil, T., Enkelmann, V., Mullen, K., Meghdadi, F., List,
E. J. W., Leising, G. (2001). "Polyfluorenes with polyphenylene dendron side chains:
Toward non-aggregating, light-emitting polymers " J. Am. Chem. Soc. 123(5): 946-953.

Shaheen, S. E., Jabbour, G. E., Morrell, M. M., Kawabe, Y., Kippelen, B., Nabor, M.-F.,
Schlaf, R., Mash, E. A., Armstrong, N. R. (1998). "Biright blue organic light-emitting
diode with improved color purity using a lif/al cathode." J. Appl. Phys. 84(4): 2324-2327.

Shi, J., Tang, C. W. (2002). "Anthracene derivatives for stable blue-emitting organic
electroluminescence devices." Appl. Phys. Lett. 80(17): 3201-3203.

Shirota, Y., Kakuta, T., Mikawa, H. (1984). "Electrochemical oxidation of
poly(vinylferrocene) with concurrent precipitation on the electrode; preparation of an
electrically conducting polymer." Makromolekulare Chemie, Rapid Communications
5(6): 337-40.

Shoustikov, A. A., You, Y., Thompson, M. E. (1998). "Electroluminescence color tuning
by dye doping in organic light-emitting diodes." IEEE Journal of Selected Topics in
Quantum Electronics 4(1): 3-13.

Sixl, H., Schwoerer, M. (1970). "Dynamics of optical electron spin-polarization in
naphthalene. Magnetic field effects on phosphorescence." Chem. Phys. Lett. 1(6): 21-25.

Sotoyama, W., Satoh, T., Sawatari, N., Inoue, H. (2005). "Efficient organic light-emitting
diodes with phosphorescent platinum complexes containing ncn-coordinating tridentate
ligand." Appl. Phys. Lett. 86(15): 153505(1-3).

Steed, J. W. (2001). "First- and second-sphere coordination chemistry of alkali metal
crown ether complexes." Coordination Chemistry Reviews 215: 171-221.

Sudhakar, M., Djurovich, P. I., Hogen-Esch, T. E., Thompson, M. E. (2003).
"Phosphorescence quenching by conjugated polymers." J. Am. Chem. Soc. 125(26):
7796-7797.

Sun, Y., Giebnic, N. C., Kanno, H., Ma, B., Thompson, M. E., Forrest, S. R. (2006).
"Management of singlet and triplet excitons for efficient white organic light-emitting
devices." Nature 440: 908-912.

162

Suzuki, H., Kondo, A., Inouye, M., Ogawa, T. (1986). "An alternative synthetic method
for polycyclic aromatic iodides." Synthesis 2: 121-122.

Tang, C., Liu, F., Xia, Y.-J., Lin, J., Xie, L.-H., Zhong, G.-Y., Fan, Q.-L., Huang, W.
(2006). "Fluorene-substituted pyrenes—novel pyrene derivatives as emitters in nondoped
blue oleds." Organic Electronics 7: 155–162.

Tang, C. W., VanSlyke, S. A. (1987). "Organic electroluminescent diodes." Appl. Phys.
Lett. 51(12): 913-915.

Tang, C. W., VanSlyke, S. A., Chen, C. H. (1989). "Electroluminescence of doped
organic thin films." J. Appl. Phys. 65(9): 3610-3616.

Tao, S., Peng, Z., Zhang, X., Wang, P., Lee, C.-S., Lee, S.-T. (2005). "Highly efficient
non-doped blue organic light-emitting diodes based on fluorene derivatives with high
thermal stability." Adv. Funct. Mater., Weinheim (Germany) 15: 1716-1721.

Tasch, S., List, E. J. W., Ekstrom, O., Graupner, W., Leising, G., Schlichting, P., Rohr,
U., Geerts, Y., Scherf, U., Mullen, K. (1997). "Efficient white light-emitting diodes
realized with new processable blends of conjugated polymers." Appl. Phys. Lett. 71(20):
2883-2885.

Togni, A. (1995). "Ferrocene-containing charge-transfer complexes. Conducting and
magnetic materials." Ferrocenes 433: 69.

Trasatti, S. (1986). "The absolute electrode potential: An explanatory note
(recommendations 1986)." Pure Appl. Chem. 58: 955-966.

Tsuboyama, A., Iwawaki, H., Furugori, M., Mukaide, T., Kamatani, J., Igawa, S.,
Moriyama, T., Miura, S., Takiguchi, T., Okada, S., Hoshino, M., Ueno, K. (2003).
"Homoleptic cyclometalated iridium complexes with highly efficient red
phosphorescence and application to organic light-emitting diode." J. Am. Chem. Soc.
125(42): 12971-12979.

Turro, N. J. (1991). "Modern molecular photochemistry." Sausalito, CA: University
Science Books.

Tyutnev, A. P., Saenko, V. S., Nikitenko, V. R., Kundina, Y. F., Pozhidaev, E. D.,
Vannikov, A. V. (2001). "Bipolar radiation-induced electroconductivity of polystyrene."
Zhurnal Nauchnoi i Prikladnoi Fotografii 46(6): 28-35.

163

Uehara, T., Belosludov, R. V., Farajian, A. A., Mizuseki, H., Kawazoe, Y. (2006).
"Electronic and transport properties of ferrocene: Theoretical study." Japanese Journal of
Applied Physics 45(4B): 3768-3771.

Uehara, T., Igarashi, N., Belosludov, R. V., Farajian, A. A., Mizuseki, H., Kawazoe, Y.
(2006). "Theoretical study of conductance properties of metallocene." Nippon Kinzoku
Gakkaishi 70(6): 478-482.

Wagner, M. J., Dye, J. L. (1993). "Alkalides, electrides, and expanded metals." Annu.
Rev. Mater. Sci. 23: 223-253.

Watanabe, M., Nishiyama, M., Yamamoto, T., Koie, Y. (2000). "Palladium/p(t-bu)3-
catalyzed synthesis of n-aryl azoles and application to the synthesis of 4,4',4''-tris(n-
azolyl)triphenylamines." Tetrahedron Lett.: 481-483.

Watanabe, S., Agata, Y., Tanaka, D., Kido, J. (2005). "High-efficiency phosphorescent
oleds using chemically doped layers." Journal of Photopolymer Science and Technology
18(1): 83-86.

Wei, Y., Chen, C.-T. (2007). "Doubly ortho-linked cis-4,4'-
bis(diarylamino)stilbene/fluorene hybrids as efficient nondoped, sky-blue fluorescent
materials for optoelectronic applications." J. Am. Chem. Soc. 129(24): 7478-7479.

Wen, S.-W., Lee, M.-T., Chen, C. H. (2005). "Recent development of blue fluorescent
oled materials and devices." Journal of Display Technology 1(1): 90-99.

Wheland, R. C., Gillson, J. L. (1976). "Synthesis of electrically conductive organic
solids." J. Am. Chem. Soc. 98(13): 3916-3925.

Wilkinson, G., Rosenblum, M., Whiting, M. C., Woodward, R. B. (1952). "The structure
of iron-bis-cyclopentadienyl." J. Am. Chem. Soc. 74(8): 2125 - 2126.

Williams, E. L., Haavisto, K., Li, J., Jabbour, G. E. (2007). "Excimer-based white
phosphorescent organic light emitting diodes with nearly 100% internal quantum
efficiency." Advanced Materials (Weinheim, Germany) 19(2): 197-202.

Winther-Jensen, B., Chen, J., West, K., Wallace, G. (2005). ""Stuffed" Conducting
polymers." Polymer 46(13): 4664-4669.

164
Xiao, X., Brune, D., He, J., Lindsay, S., Gorman, C. B., Tao, N. (2006). "Redox-gated
electron transport in electrically wired ferrocene molecules." Chem. Phys. 326(1): 138-
143.

Xie, Z., Yang, B., Liu, L., Li, M., Lin, D., Ma, Y., Cheng, G., Liu, S. (2005).
"Experimental and theoretical studies of 2,5-diphenyl-1,4-distyrylbenzenes with all-cis-
and all-trans double bonds: Chemical structure determination and optical properties." J.
Phys. Org. Chem. 18: 962–973.

Yang, J. P., Jin, Y. D., Heremans, P. L., Hoefnagels, R., Dieltiens, P., Blockhuys, F.,
Geise, H. J., Van der Auweraer, M., Borghs, G. (2000). "White light emission from a
single layer organic light emitting diode fabricated by spincoating." Chem. Phys. Lett.
325(1-3): 251-256.

Zhang, F.-F., Wan, Q., Wang, X.-L., Sun, Z.-D., Zhu, Z.-Q., Xian, Y.-Z., Jin, L.-T.,
Yamamoto, K. (2004). "Amperometric sensor based on ferrocene-doped silica
nanoparticles as an electron transfer mediator for the determination of glucose in rat brain
coupled to in vivo microdialysis." Journal of Electroanalytical Chemistry 571(2): 133-
138.

Zhao, H.-P., Tao, X.-T., Wang, F.-Z., Ren, Y., Sun, X.-Q., Yang, J.-X., Yan, Y.-X., Zou,
D.-C., Zhao, X., Jiang, M.-H. (2007). "Structure and electronic properties of
triphenylamine-substituted indolo[3,2-b]carbazole derivatives as hole-transporting
materials for organic light-emitting diodes." Chem. Phys. Lett. 439(1-3): 132-137.

Zhao, H.-P., Tao, X.-T., Wang, P., Ren, Y., Yang, J.-X., Yan, Y.-X., Yuan, C.-X., Liu,
H.-J., Zou, D.-C., Jiang, M.-H. (2007). "Effect of substituents on the properties of
indolo[3,2-b]carbazole-based hole-transporting materials." Organic Electronics online (in
press): doi:10.1016/j.orgel.2007.05.001.

Zhou, X., Pfeiffer, M., Blochwitz, J., Werner, A., Nollau, A., Fritz, T., Leo, K. (2001).
"Very-low-operating-voltage organic light-emitting diodes using a p-doped amorphous
hole injection layer." Appl. Phys. Lett. 78(4): 410-412.

Zhou, X., Pfeiffer, M., Huang, J. S., Blochwitz-Nimoth, J., Qin, D. S., Werner, A.,
Drechsel, J., Maennig, B., Leo, K. (2002). "Low-voltage inverted transparent vacuum
deposited organic light-emitting diodes using electrical doping." Appl. Phys. Lett. 81(5):
922-924.

Abstract (if available)

Abstract Organic light-emitting devices (OLEDs) with two-component layers consisting of a guest material (dopant) dispersed in a matrix (host) are examined in this work. Both emissive organic materials as dopants and non-emissive conductivity dopants are discussed. For the conductivity doping, lithium-coordinating agents including cryptands and crown ether derivatives have been used as dopants for electron-transport layers to improve the device operating voltages and luminance. Devices with oxidatively doped hole-transport layers consisting of novel hole-transporting ferrocene derivatives are also described. For the emissive doping, the spectral line shape dependence on the varied host materials for the emission of phosphorescent iridium complexes is examined. It has been shown that the value of Huang-Rhys parameters for the phosphorescence spectra decreases several times when non-polar ultra-high-band-gap hosts (UGHs) are used, compared to conventional carbazole-based host materials. Finally, a new class of efficient blue fluorescent dopants specifically designed as blue components for the white OLEDs with separate channels of harvesting blue singlet and red and green triplet excitons has been characterized.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

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

PDF

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

PDF

New materials and device structure for organic light-emitting diodes

PDF

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

PDF

New materials for organic light emitting diodes

PDF

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

PDF

Synthesis and photophysical characterization of phosphorescent cyclometalated iridium (III) complexes and their use in OLEDs

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

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

PDF

Molecular design strategies for blue organic light emitting diodes

PDF

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

PDF

Organic solar cells: molecular electronic processes and device development

PDF

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

PDF

Molecular design of pyridine and quinoline based compounds for organic optoelectronics

PDF

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

PDF

Theoretical, experimental, device fabrication, and degradation studies of materials for optoelectronic devices

PDF

Synthesis and photophysical study of phosphorescent hetero-cyclometalated organometallic complexes involving phosphino-carbon ligands

PDF

Synthesis, structural and photophysical characterization of phosphorescent three-coordinate Cu(I)-N-heterocyclic carbene complexes

PDF

Organic photovoltaics: from materials development to device application

Asset Metadata

Creator Polikarpov, Evgueni (author)

Core Title Dopants for organic light-emitting devices

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publication Date 04/10/2010

Defense Date 03/14/2008

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

Tag OAI-PMH Harvest,OLED

Language English

Advisor Thompson, Mark E. (committee chair), Bradforth, Stephen E. (committee member), Goo, Edward K. (committee member)

Creator Email polikarp@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1099

Unique identifier UC1188428

Identifier etd-Polikarpov-20080410 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-52938 (legacy record id),usctheses-m1099 (legacy record id)

Legacy Identifier etd-Polikarpov-20080410.pdf

Dmrecord 52938

Document Type Dissertation

Rights Polikarpov, Evgueni

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