Flat Panel Displays

FLAT PANEL DISPLAYS
Advanced Organic Materials

RSC Materials Monographs
Series Editor: J.A. Connor, Department of Chemistry,

University of Kent, Canterbury, UK
Advisory Panel: G.C. Allen (Bristol, UK), D.J. Cole-Hamilton (St Andrews,
UK),W.J. Feast (Durham, UK),P. Hodge (Manchester, UK),M. Ichikawa
(Sapporo, Japan), B.F.G. Johnson (Cambridge, UK), G.A. Ozin (Toronto,
Canada),W.S. Rees (Georgia, USA)
The chemistry of materials will be the central theITie of this Series which aims to
assist graduates and others in the course of their work. The coverage will be
wide-ranging, encompassing both established and new, developing areas.
Although focusing on the chemistry of materials, the monographs will not be
restricted to this aspect alone.
Polymer Electrolytes
by Fiona M. Gray, School of Chemistry, University of St. Andrews, UK
Flat Panel Displays: Advanced Organic Materials

by S.M. Kelly, Department of Chemistry, University of Hull, UK
How to obtainfuture titles on publication
A standing order plan is available for this series. A standing order will bring
delivery of each new volume immediately upon publication. For further
information, please write to:
Sales and Customer Care

Royal Society of Chemistry
Thomas Graham House
Science Park
Milton Road
Cambridge
CB4 OWF
Telephone: + 44(0) 1223 420066
RSC
MATERIALS
MONOGRAPHS
Flat Panel Displays

Advanced Organic Materials
S.M. Kelly
Department of Chemistry, University of Hull, UK
RSeC
ROYAL SOCIETV OF CHEMISTRY
ISBN 0-85404-567-8
A catalogue record for this book is available from the British Library.
0The Royal Society of Chemistry 2000

All rights reserved.
Apart from any fair dealing for the purposes of research or private study, or criticism or
review as permitted under the terms of the UK Copyright, Designs and Patents Act, 1988,
this publication may not be reproduced,stored or transmitted, in anyform or by any means,
without the prior permission in writing of The Royal Society of Chemistry, in the case of
reprographic reproduction only in accordance with the terms of the licences issued by the
Copyright Licensing Agency in the UK,or in accordance with the terms of the licences
issued by the appropriate Reproduction Rights Organization outside the UK. Enquiries
concerning reproduction outside the terms stated here should be sent to The Royal Society
of Chemistry at the addressprinted on this page.
Published by The Royal Society of Chemistry,

Thomas Graham House, Science Park, Milton Road
Cambridge CB4 OWF, UK
For further information see our web site at www.rsc.org
Typeset by Paston PrePress Ltd, Beccles, Suffolk

Printed by MPG Books Ltd, Bodmin, Cornwall
Preface
Liquid crystals have found wide commercial application over the last 25-30
years in electro-optical flat panel displays (FPDs) for consumer audio-visual
and office equipment such as watches, clocks, stereos, calculators, portable
telephones, personal organisers, note books and laptop computers. There are
many other applications for liquid crystal displays (LCDs) such as information
displays in technical instruments and in vehicles initially as clocks, then
speedometers to a lesser extent and now increasingly as navigation and
positional aids or entertainment consoles. They are also used in low-volume
niche products such as spatial light modulators and generally as very fast light
shutters. More importantly they have come to dominate the displays market in
portable instruments due to their slim shape, low-weight, low voltage operation
and low power consumption, see Table 1.1. LCDs are now starting to win
market share from cathode ray tubes (CRTs) in the computer monitor market.
The market share of LCDs of the total market for displays is expected to
significantly increase over the next decade. There are a number of existing
competing flat-panel display technologies, such as plasma display panels
(PDPs), vacuum fluorescence displays (VFDs), inorganic semiconductor light-
emitting diodes (LEDs), digital micromirror devices (DMDs) and field emission
displays (FEDs). However, these have relatively small shares of the overall
displays market, see Table 1.1. The most promising technology for FPDs being
developed at the moment is represented by organic light-emitting diodes
(OLEDs) using either low-molar-mass (LMM) materials (small molecules) or
light-emitting polymers (LEPs). The first production lines for LEP technology
have recently been commissioned and products are commercially available.
However, in spite of these competing FPDs, especially OLEDs which are
expected to exhibit significant growth, the value of LCDs is still expected to
exceed that of CRTs in the near future. Manufacturing facilities for flat-panel
displays are very capital intensive, e.g. a plant for TN (twisted nematic)-LCDs
with active matrix addressing can cost upwards of $ 1 billion. As a consequence
of a combination of factors, such as the capital already invested in LCD plants,
which have to be depreciated, the steadily decreasing unit cost of LCDs and the
expanding market requirement for them in existing products, it will take many
years for competing technologies to gain a significant market share in the
V
vi Preface
displays market in general. In particular, LCDs can be expected to maintain a
dominant position in the market for portable displays.
The successful development of LCD technology was dependent on parallel
developments and progress in an unusual combination of scientific disciplines
such as synthetic organic chemistry, physics, electronics and device engineering.
These include improvements in batteries, polarisers, electrodes, CMOS drivers,
spacers, alignment layers and nematic liquid crystals. These developments were
made in response to a clear market requirement for a low-voltage, low-power-
consuming flat-panel display screen for portable, battery-operated instruments
in order to display graphic and digital information of ever more increasing
volume, speed and complexity. In this monograph we will attempt to illustrate
this development using the most important types of LCD currently in large-
scale manufacture. We will not describe LCDs using smectic liquid crystals,
although they may have the potential to become a major commercial product,
since they have failed to make a commercial breakthrough after more than two
decades of research and development. All of these types of LCDs can be
modified to use nematic gels or polymer-dispersed liquid crystals. Therefore,
these will not be dealt with here in any detail as the general principles of
operation and electro-optical effects are essentially the same.
OLEDs using electroluminescent small molecules have been in continuous
development for over 30 years and intensive development for at least the last
decade. Only now are OLEDs using low-molar-mass materials being manufac-
tured on any significant scale. The development of OLEDs using conjugated
organic polymers has been able to profit from the know-how and technology
developed for low-molar-mass OLEDs. However, the time from discovery to
manufacture has been much shorter, i.e. about 10 years. The first pilot plants for
OLEDs using LEPs have just been commissioned at Philips and UNIAX and
commercial products incorporating LEPs from Dow Chemicals, Clariant or
Covion are to expected on the market in 2001 based on technology under
licence, at least in part, from CDT.
The development and successful commercialisation of LCDs and OLEDs as
flat panel displays in consumer and industrial products and instruments is
illustrative of the general dependence of consistent improvement in the
performance of flat panel displays as a consequence of the invention, laboratory
preparation, evaluation, optimisation, scale-up and then large-scale manufac-
ture of organic compounds and their mixtures in high purity and at acceptable
cost with the required spectrum of physical properties. This development in
materials chemistry will be described in detail in the following chapters. The
interdependence of technologies in electro-optics, .electronics and organic
chemistry is illustrated by the imaginative use of liquid crystals in OLEDs as
charge-carrier transport layers and as electroluminescent materials as the source
of plane polarised light for hybrid OLED-LCDs with intrinsically higher
brightness.
The fundamental electro-optical principles of LCDs and OLEDs with their
relative advantages and disadvantages for a diverse range of specific applica-
tions are described in this monograph. These specifications then prescribe the
Preface vii
relative and absolute magnitude of a spectrum of physical properties for
nematic liquid crystals and electroluminescent organic materials to fulfil in
order for these types of flat panel displays to function as efficiently as possible
and attain their maximum potential for a particular application. The nature of
the nematic liquid crystalline state and the origin of electroluminescence in
small organic molecules and organic conjugated polymers is described suffi-
ciently to understand the correlation between molecular structure and those
physical properties of relevance to LCDs and OLEDs, at least where such
correlations are understood. This monograph concentrates primarily on the
developments in the design and synthesis of these two different classes of
organic materials specifically for use in the two most important types of FPDs
using organic compounds. Therefore, the theoretical, especially mathematical,
background to the phenomenon of liquid crystallinity and organic electrolumi-
nescence are kept to a necessary minimum.
Contents
Chapter 1 Flat Panel Displays 1

1 Flat Panel Displays 1
Flat Panel Cathode Ray Tubes 2
Plasma Display Panels 3
Vacuum Fluorescence Displays 3
Field Emission Displays 4
Digital Micromirror Devices 4
Inorganic Semiconductor Light-Emitting Diodes 5
Organic Light-Emitting Diodes 5
Liquid Crystal Displays 6
2 Conclusions 7
3 References 8
Chapter 2 Liquid Crystals and Liquid Crystal Displays (LCDs) 9

1 Physical Properties of Nematic Liquid Crystals 20
2 Physical Properties of Liquid Crystals 20
Optical Anisotropy (Birefringence) 20
Elastic Constants 22
Viscosity 23
Dielectric Anisotropy 24
3 Liquid Crystal Displays 25
4 Cell Construction of LCDs 27
5 Addressing Methods for LCDs 30
Direct Addressing 30
Multiplex Addressing 30
Active Matrix Addressing 32
6 Organic Polymer Alignment Layers 33
7 Organic Compensation Films for LCDs 38
8 References 40
ix
X Contents
Chapter 3 Liquid Crystal Displays Using Nematic Liquid
Crystals 45
1 Introduction 45
2 Dynamic Scattering Mode LCDs 45
Nematic Materials 48
3 Cholesteric-Nematic Phase Change (CNPC) LCDs 51
Chiral Nematic Materials 52
4 Electrically Controlled Birefringence (DAP/HN/ECB) LCDs 53
Nematic Materials of Negative Dielectric Anisotropy 56
5 Twisted Nematic LCDs 60
Nematic Materials of Positive Dielectric Anisotropy 66
Nematic Materials for Direct Addressing 66
Nematic Materials for Multiplex Addressing 74
Nematic Materials for Active Matrix Addressing 81
6 Super Twisted Nematic LCDs 85
Super Birefringent Effect LCDs 88
Electro-optical Performance of STN-LCDs 91
Temperature Dependence of Electro-optical
Performance of STN-LCDs 92
Black-and-white STN-LCDs 93
Nematic Materials for STN-LCDs 93
Polar Nematic Materials for STN-LCDs 94
Apolar Nematic Materials for STN-LCDs 99
7 Guest-Jost LCDs 103
Negative Contrast Heilmeier and Zanoni GH-LCDs 110
White and Taylor GH-LCDs 112
Negative Contrast White and Taylor GH-LCDs 113
Positive Contrast White and Taylor GH-LCDs 114
Super Twisted Nematic (STN) GH-LCDs 115
Dichroic Dyes-Guests 117
Positive Contrast Dyes 117
Negative Contrast Dyes 121
Nematic Liquid Crystals-Hosts 122
Nematic Liquid Crystal Hosts of Positive Dielectric
Anisotr opy 122
Nematic Liquid Crystal Hosts of Negative Dielectric
Anisotropy 123
8 In-Plane Switching (IPS) LCDs 124
Nematic Materials 126
9 References 127
Chapter 4 Photoluminescence and Electroluminescence from

Organic Materials 134
1 Introduction 134
2 Photoluminescence from Organic Materials 136
Contents xi
3 Electroluminescence from Organic Materials 138
4 References 144
Chapter 5 Organic Light-Emitting Diodes Using Low-Molar-

Mass Materials (LMMMs) 147
1 Introduction 147
2 Monolayer Organic Light-Emitting Diodes Using LMMMs 150
3 Bilayer OLEDs Using LMMMs 151
4 Trilayer OLEDs Using LMMMs 154
5 Low-Molar-Mass Organic Materials for OLEDs 155
Non-Emissive Electron Transport Layers (ETLs) 156
Non-Emissive Hole Transport Layers (HTLs) 156
Liquid Crystals as Charge-Carrier Transport Layers 158
Columnar Liquid Crystals as HTL 160
Smectic Liquid Crystals as HTL and ETL 163
Liquid Crystals as Electroluminescent Materials 165
Green Electroluminescent, Low-Molar-Mass Organic
Materials 166
Red Electroluminescent, Low-Molar-Mass Organic
Materials 168
Blue Electroluminescent, Low-Molar-Mass Organic
Materials 170
6 Performance of OLEDs Using LMMMs 173
Stability of OLEDs Using LMMMs 173
7 References 175
Chapter 6 Organic Light-Emitting Diodes Using Light-

Emitting Polymers 179
1 Introduction 179
2 Monolayer OLEDs Using LEPs 179
Light-Emit ting Polymers (LEPs) 184
3 Bilayer OLEDs Using LEPs 196
Polymers as ETLs 199
4 Trilayer OLEDs Using LEPs 206
5 Polarised Light Emission from OLEDs 208
6 Performance of OLEDs Using LEPs 212
Stability of OLEDs Using LEPs 215
7 References 216
Conclusions and Outlook 222
Subject Index 229

Abbreviations, Acronyms and
Symbols
AA Active Addressing
AC Alternating Current
Ah3 Aluminium tris(2-hydroxyquinolate)
AM Active Matrix
c (F) capacitance (farads)
CCH-5 trans- 1-(trans-4-Cyanocyclohexyl)-4-pen
tylcyclohexane
CDT Cambridge Display Technology
Ch Cholestolic
CNPC Cholesteric Nematic Phase Change
Cr Crystal
CRT Cathode Ray Tube
d(Pm) cell gap (micrometers)
DAP Deformation of Vertically Aligned
DC Direct Current
DMD Digital Micromirror Devices
DSM Dynamic Scattering Mode
DSTN Double Super Twisted Nematic
Eabs (eV) energy of absorption (electron-volts)
Eem (eV) energy of emission (electron-volts)
ECB Electrically Controlled Birefringence
EL Electroluminescence
E, (ev) phonon energy (electron-volts)
ETL Electron-Transport Layer
FDP Flat Panel Display
FED Field Emission Display
g Kirkwood Foehlich factor
GaAsP Gallium Arsenide Phosphorous
GH Guest-Host
h (J s) Planck's constant (Joule seconds)
HN Homeotropic Nematic
HOMO Highest Occupied Molecular Orbital
...
XI11
x1v Abbreviations, Acronyms and Symbols
HTL Hole-Transport Layer
I Isotropic
I (mA cmv2) current density (milliamp per centimetre squared)
IPS In-Plane Switching
IT0 Indium-Tin Oxide
I- v Current-Voltage
kl1, k22, k33 (N) elastic constants (newton)
L(cd m-2) brightness (candelas per meter squared)
LCD Liquid Crystal Display
LED Light Emitting Diode
LEP Light Emitting Polymer
LMMM Low Molar Mass
LUMO Lowest Unoccupied Molecular Orbital
MBBA N-(4-methoxybenzy1idene)-4’-butylaniline
MEH-PPV vinylene
Poly[2-methoxy-5-(2-ethylhexyloxy)]-4-phenylene
N Nematic
n average refractive index
N* Chiral Nematic
ne refractive index of the extraordinary ray
n0 refractive index of the ordinary ray
NW normally white
OLED Organic Light-Emitting Diode
OM1 Optical Mode Interference
P (P-4 pitch (micrometres)
PANi Poly(ani1ine)
PBD 1,3,4-0xadiazole
2-(4-tert.-butylphenyl)-5-(biphenyl-4-y1)-
PCH-5 trans-1-(4-cyanophenyl)-4-pentylcyclohexane
PD Polymer Dispersed
PDP Plasma Display Panel
PECH-5 trans-1-[2-(4-cyanophenyl)ethyl]-4-pentylcyclohexane
PEDOT Poly(3,4-ethylenedioxythiophene)
PET Poly(ethy1ene terephthalate)
Pixel Picture Element
PL Photoluminescence
PMMA Poly(methy1methacrylate)
PPP
PPV Poly(p-phenylene vinylene)
PVC Poly(viny1cinnamate)
PVK Poly(N-vinylcarbazole)
RGB Red Green Blue
RMS Root Mean Square
S order parameter
SAM Self-AssembledMonolayer
SBE Super Birefringent Effect
SCL Space-Charge Limited
SmA Smectic A
Abbreviations, Acronyms and Symbols xv
Chiral Smectic A
Smectic B
Smectic X
solubility (weight percent)
Surface Stabilised Ferroelectric
Super Twisted Nematic
temperature (degrees centigrade)
Thin-Film Transistor
glass transition temperature (degrees centigrade)
Twisted Nematic
nematic clearing point (degrees centigrade)
switch-off time (milliseconds)
switch-on time (milliseconds)
reduced temperature (degrees centigrade)
Ultra Violet
voltage at 10% light transmission (volts)
Vertically Aligned Nematic
capacitive threshold voltage (volts)
Vacuum Fluorescence Display
non-select voltage (volts)
operating voltage (volts)
select voltage (volts)
threshold voltage (volts)
birefringence
optical retardation (radians)
dielectric anisotropy
twist angle (degrees)
optical path difference (micrometres)
dielectric permittivity measured parallel to the director (fara-
days per metre)
E ~ (m-')
F dielectric permittivity measured perpendicular to the director
(faradays per metre)
quantum efficiency (percent)
quantum efficiency of fluorescence (percent)
double charge injection factor
rotational viscosity (millemetre squared per second)
power efficiency (lumens per watt)
external quantum efficiency (percent)
internal quantum efficiency (percent)
singlet formation efficiency (percent)
flow viscosity (centipoise)
wavelength of maximum absorption (nanometers)
wavelength of maximum emission (nanometers)
(cm2 V-' s-l) electron mobility (centimetre squared per volt per second)
pLe
xvi Abbreviations, Acronyms and Symbols
p h (cm2V-' s-l) hole mobility (centimetre squared per volt per second)
lu (D) dipole moment (debye)
8 ("1 tilt angle (degrees)
Q (0crn-') resistivity (ohm per centimetre)
Dabs (S-l) frequency of absorption (per second)
Dem (s- ') frequency of emission (per second)
CHAPTER 1
Flat Panel Displays
1 Flat Panel Displays

The cathode ray tube (CRT) is still the dominant electro-optical display device
today, although this is expected to change in the next few years. The CRT is still
the benchmark display in terms of cost and performance. There are many areas
of the market for electro-optic displays where one or more of the competing flat-
panel display technologies offers a superior technological performance to a
CRT, see Table 1.1.' Perhaps the most important are portable applications
where the combination of physical properties, such as low power consumption,
low operating voltage and light-weight of liquid crystal displays (LCDs) is
clearly superior to that of CRTs. Most flat panel displays are emissive displays,
i.e. they emit light without requiring absorbing polarisers like LCDs. Therefore,
their brightness and viewing angle dependence are fundamentally superior to
those of LCDs, which modulate the intensity of transmitted light from some
independent internal or external light source. Therefore, they must be used with
a back-light where insufficient ambient light is present. Light-emitting flat panel
displays (FPDs) offer superior performance in poor ambient light conditions or
in the dark whereas reflective FPDs are clearly superior in a bright light
Table 1.1 Estimated world-wide market share offlat panel displays in the year
2000'
Type offlat panel display (FPD) Number of units
Liquid crystal displays (LCDs) with segmented characters 1 470 000 000
Super-twisted nematic liquid crystal displays (STN-LCDs) 45 000 000
Liquid crystal displays (AM-TFT-LCDs) with active matrix 48 000 000
thin film transistor addressing
Organic electroluminescent displays (OLEDs) 300 000
Plasma display panels (PDPs) 630 000
Field emission displays (FEDs) 540 000
Inorganic semiconductor light-emitting diodes (LEDs) 181 000000
Vacuum fluorescent displays (VFDs) 166 000 000
Total 1 900 000 000
1
2 Chapter I
environment. The former are not visible in the dark and the latter are washed
out in bright light.
A flat panel display may be several millimetres or several centimetres thick.
There are many technologies capable of being used to create a flat panel display.
The most important flat panel displays are described briefly below; the two most
important are LCDs and OLEDs, which are the subject of this monograph.
Both require organic materials in order to function. Therefore, these are
described in much more detail.
A high-information-content display must be capable of displaying an equiva-
lent amount of information as a CRT of comparable size. The major segments
of the displays market in general for CRTs are as television screens and static,
i.e. non-portable computer monitors.
Emissive displays are intrinsically brighter than commercial LCDs currently
available, even those with a strong back-light. The use of crossed, absorbing
polarisers limits the maximum intensity of incident light transmitted to 25%.
Therefore, a large amount of research and development effort is being devoted
to optimising internal reflectors, which replace one polariser, optical retarders
and different types of LCDs, which use either one polariser or no polarisers.
Advances in optimising the physical properties in organic materials such as
nematic liquid crystals, electroluminescent small molecules and polymers are
the topic of this monograph. Oligomers are intermediate compounds between
low-molar-mass materials (small molecules) and polymers and serve as model
compounds for studying polymers without the polydispersity of the latter.
However, they are not used commercially, and probably will not be in the
foreseeable future. Therefore, they will not form part of this monograph.
Parallel developments in device peripherals such as organic polymer alignment
layers, organic optical retarders and polarisers are also important. These are
also described briefly. However, a satisfactory electro-optic performance of a
particular display type is not always a sufficient criterium for commercialisa-
tion. The properties of other electro-optic components, such as the cost of
drivers can play a decisive role in deciding whether a particular display
technology is manufactured at all, occupies a niche in the displays market or is
manufactured in large volumes. However, these parameters often depend on the
fundamental mode of operation of a particular display technology. These are
described and compared briefly in this chapter for FPDs in general and in much
more detail in Chapters 2-6 for LCDs and organic light-emitting diodes
(OLEDs).
Flat-Panel Cathode Ray Tubes2

The production of flat-panel cathode ray tubes (CRTs) is essentially a fabrica-
tion issue. The basic principle of operation is the same as a standard CRT.
Electrons are emitted from a hot cathode. These are guided by a magnetic field
to the glass screen coated in a layer of phosphorescent material. Upon impact
the energy of the electron is transferred to the phosphor and light is emitted. A
regular pattern of red, green and blue phosphors creates a dense pattern of
Flat Panel Displays 3
pixels, which allows the generation of full colour. A gas plasma discharge may
also be used as a source of electrons.
The high voltage requirement, i.e. <200 V, and power consumption are the
main restrictions to the utilisation of flat CRTs due to their incompatibility with
battery operation over an extended period of time due to the high voltages and
power consumption required. Other flat panel displays are more suitable and
are usually preferred for portable, hand-held applications. The difficulty
associated with manufacturing flat, rectangular large-area cathode ray tubes is
an added problem preventing their use as screens for portable instruments. Such
large CRTs would still be relatively heavy despite their relatively flat, thin
construction due to the weight of the thick-walled glass vacuum tube required
for mechanical stability.
Plasma Display Panels3

Plasma display panels (PDPs) based on an emissive gas discharge phenomenon
were invented over 30 years ago. Indeed large-area plasma panel displays have
been commercially available since 1970. Monochrome PDPs use visible light
emitted under the action of a small electric current flowing between the
electrodes. Full colour displays use UV emission at 150 nm or 173 nm to
address an alternating array of red, green and blue phosphorescent strips. Short
response times and steep electro-optic transmission curves facilitate the fabrica-
tion of very large-area, high-information-content plasma display panels ( > 60”
diagonal). However, their high cost and substantial size and weight has
restricted their acceptance for the consumer market. Moreover, flat-panel
plasma displays require a large number of expensive, high-voltage, alternating
current (AC) or direct current (DC) drivers. Furthermore, the high operating
voltages and power consumption prohibit their use in portable, battery-
operated applications. Therefore, PDPs have traditionally been used for non-
portable, high-cost, low-volume display applications, which are far less cost-
sensitive, such as industrial, commercial or military applications. LCDs with a
very large area and high information content, e.g. for TVs with a 40” diagonal
and above, are still very expensive and not competitive with PDPs. However,
the unit-cost of large-area, high-information-content PDPs is also steadily
decreasing. Consequently, the acceptance of PDPs as very large televisions and
monitors in the consumer market is gradually increasing. Unfortunately the
large pixel size (= 1 mm) gives rise to relatively low resolution and a grainy
appearance for short viewing distances.
Vacuum Fluorescence Displays4

Vacuum fluorescent displays (VFDs) are strongly related to flat-panel CRTs.
Electrons are ejected from a cathode source, traverse a vacuum and then strike a
pattern of triodes with individual anodes covered in red, green and blue
phosphorescent material. However, the operating voltages, e.g. 12 V, and
power consumption are much lower than those found for CRTs and PDPs.
4 Chapter I
The fabrication costs of VFDs are also relatively low. They are rugged with long
operating lifetimes. Therefore, small VFDs have been manufactured in large
volume for several decades for a variety of applications, e.g. as part of car
dashboards or orientation and navigation systems.
Once again the major problems associated with the commercialisation of
large VPDs is their manufacture. These include increasing weight of the glass
tubes, which are necessarily thick walled. Precise spatial matching of the
cathode and anode matrices is also problematical at large display size. Multiplex
addressing of larger displays results in unacceptably high operating voltages,
e.g. 100V, for battery-operated devices. VFDs with active matrix addressing use
much lower operating voltages, but are correspondingly more expensive.
Field Emission Displays596

Field emission displays (FEDs) utilise a very similar technology to the CRT
tube, i.e. electron-impact induced light emission from a flat screen coated with
alternating strips of red, green and blue organometallic phosphors. However,
the main difference is that the electrons are not generated as a beam from a hot
cathode, which is then directed by a magnetic field towards the screen, as in a
CRT, but are emitted individually from a dense matrix of pointed pixel
electrodes covering the active cathode area of the display. The narrow gap
between the flat phosphor screen on top of the anode and the planar emission
cathode layer and substrate is small, e.g. 2 mm. Therefore, considerably lower
voltages are required for FEDs than for CRTs. However, the current density is
significantly higher. This mode of operation allows light-weight flat panels to be
constructed with a relatively low power consumption, wide viewing angle, high
brightness, video-rate addressing and ruggedness. The contrast is generally
relatively low ( > 20: 1). Flat-panel FEDs are available as monochrome and full-
colour commercial products, although with a relatively small screen size (5”
diagonal) for the moment. Larger prototypes have been demonstrated (12”
diagonal). However, the most important factor holding back the wide-scale
adoption of FEDs as a flat-panel display is the high operating voltage ( > 20 V).
This inhibits their use in portable device applications due to short battery life-
times.
Digital Micromirror Devices798

Digital light processing devices use micro-electromechanical systems referred to
as a digital micromirror device (DMD). An array of rectangular polished
aluminium mirrors, e.g. 640 x 480 pixels, each individual mirror situated above
a CMOS memory chip, can be addressed by an applied voltage to reflect light
through a microlense in the on-state or deflect light in the off-state. This is a
bistable, black-on-white memory effect compatible with video-rate addressing
with high contrast ( > 1OO:l) and high brightness (=30O-400 lumens). The
mirrors are fabricated in a series of lithographic steps on a single substrate. Grey
scale can be realised using pulsed applied voltages with full colour achieved
using colour filters. Therefore, DMDs are used as high-information-content
front or rear projection devices, especially for home cinema and commercial
cinema or stadia applications. However, they are essentially projection devices
and the size and weight of the projector and light source are too large for
portable applications.
Inorganic Semiconductor Light-Emitting Diodesg

Light-emitting diodes (LEDs) are flat panel displays which emit light under the
action of an electric current passing through the emissive layer. Electrolumines-
cence in inorganic semiconductors was discovered before the corresponding
effect in organic materials was found. Consequently the first commercial alpha
numeric display devices fabricated in the early 1960s used electroluminescence
inorganic semiconductor materials, such as GaAs/P or ZnS/Mn on a glass
substrate sandwiched between two dielectric layers. These separate the emissive
material from the electrodes and limit the amount of current flowing through
the display. Pulses of alternating current result in light emission. Monochrome
semiconductor inorganic LEDs are manufactured on a large scale and are found
in many electronic instruments.
High-information-content LEDs using inorganic semiconductors have been
produced with .active matrix addressing using thin film transistors on a silicon
substrate. However, the size of the displays is limited by the amount of power
consumed by the large number of pixels due to the high capacitance at each
individual pixel. The power consumption of a large-area LED, such as a
notebook computer screen, would be considerable, e.g. 100 W. Other addres-
sing problems, such as non-uniform grey scale due to the steep curve of
brightness against voltage, also become disproportionately acute with increas-
ing display size.
Organic Light-Emitting Diodes"

There remains an enormous potential for light-emitting diodes using organic
materials (OLEDs) due to their advantageous combination of physical proper-
ties, such as ease of processing, robustness and an almost infinite possibility for
modification, e.g. wavelength of emission, by suitable materials chemistry
design and synthesis. The process of electroluminescencefrom organic materials
is essentially the same as that from inorganic materials except that the emission
takes place from a molecular excited state and not from an atomic excited state
(energy level). Therefore, the bandwidth of emission is broader due to molecular
vibrations. OLEDs are characterised by low operating voltages and power
consumption, wide viewing angles and high brightness and contrast ratios.
Thus, they are compatible with portable applications. High-information-
content OLEDs using organic materials can be addressed using direct addres-
sing, multiplex addressing or active matrix addressing. They are currently
fabricated using electroluminescent low-molar-mass materials or aromatic
conjugated electroluminescent polymers with a high glass transition tempera-
6 Chapter I
ture. Both of these classes of organic materials require a high Tgvalue in order
to avoid crystallisation, which can degrade device performance severely. Indeed,
life-time has been the major obstacle to commercialisation of this otherwise
attractive technology. However, commercial OLEDs using low molecular
weight materials and polymer are starting to appear on the flat-panel displays
market in significant volumes.
Liquid Crystal Displays' ' '*

A variety of liquid crystal displays (LCDs) dominate the market for flat panel
displays, especially for portable applications, see Table 1.1. The common
features of these devices are low weight, thin planar construction, low operating
voltages and power consumption, and acceptable contrast and viewing angles.
LCDs can be operated in a reflection mode using ambient light, a transmission
mode using a backlight and transflection combining both possibilities. LCDs
invented more than 30 years ago were not the first successful portable flat panel
display. Digital watches and calculators with segmented electrodes were first
produced using LEDs incorporating electroluminescent inorganic semiconduc-
tors. However, LCDs soon displaced LEDs from these products and then
enabled the fabrication, in the first instance, of digital watches and calculators,
then notebooks and laptop computers, camcorder viewers, portable telephones,
personal digital assistants, hand-held games, car navigation and orientation
systems and many more applications. Surprisingly most of the original LCD
prototypes were realised using nematic liquid crystals at temperatures over
100°C. The design and synthesis of new organic compounds, which exhibit a
nematic phase with a specific spectrum of properties, have been essential
contributions in establishing and expanding the multibillion dollar LCD
industry over the last 30 years as well as other multibillion dollar industries,
e.g. mobile telephones and hand-held games, such as Gameboy, would have
been much more difficult to bring to market without an LCD display. The
optical performance of LCDs has been improved to such an extent that LCDs
are starting to displace CRTs from applications where ergonomics and
footprint and not power consumption, operating voltage and even cost are the
deciding factors. However, the presence of absorbing polarisers in most types of
LCDs and in all commercial LCDs manufactured at the moment, means that
intrinsically, brighter emissive displays with Lambertian emission, such as
OLEDs, are potential competitors in the FDP market for a spectrum of
applications compatible with use in poor ambient light, if their performance
can be improved to fulfil the specifications already met by LCDs. However, the
memory effect of certain types of LCD, especially those with active matrix
addressing, means that their power consumption can be very low. Emissive flat
panel displays such as OLEDs continually draw power in the on-state and have
to be driven continually from frame to frame like a CRT.
2 Conclusions
The flat-panel displays market is characterised by its diversity and increasing
fragmentation. Therefore, many different display types dominate particular
segments of the market where their particular combination of performance,
system compatibility and cost represent the most appropriate choice. However,
over the last decade the different types of LCD have become the dominant flat
panel display for most applications. The market value of LCDs alone is
estimated at approximately $13 000 000 in the year 2000. The market share by
value of LCDs should overtake that of CRTs in the near future. However, the
market for flat panel displays is expanding rapidly with the popularity of digital
watches, calculators, notebook computers, personal digital organisers, palm-
top computers, hand-held computer games and toys, mobile telephones,
camcorders, digital cameras, etc. The rapid progress in mobile communications,
especially those providing access to the internet and e-mail, will only serve to
accelerate the growth in the market share and production volume of LCDs. The
electro-optical performance of LCDs is continually being improved and the unit
cost is steadily decreasing despite fluctuations due to economic dislocations and
natural disasters in the Far East, e.g. in countries such as Japan, Korea, Hong
Kong, Thailand and Malaysia, where most of the LCD manufacturing industry
is based. Indeed the current low cost of many types of LCD is threatening the
profitability of large parts of the LCD industry, which is particularly capital
intensive. However, LCDs are starting to steadily displace the bulky CRT from
crowded desktops, especially in countries such as Japan where space is at a
premium. This trend can confidently be expected to continue and indeed
accelerate in the near future.
The increased market share of LCDs is only slowing the volume growth of
CRTs and competing flat-panel display technologies rather than reducing their
overall volume or even displacing them from the displays’ market place. Even
the market for LCDs is remarkably diverse, although the falling cost of LCDs
with active addressing (e.g. active matrix thin-film transistor (AM-TFT)-LCDs)
is starting to reduce the market share of LCDs with multiplex addressing (super
twisted nematic (STN)-LCDs) even from relatively low-cost applications.
However, the number of display types of LCDs with active matrix addressing
is increasing to meet the growing needs for light, large-area, high-information-
content displays with video rate addressing, uniformly wide viewing angle and
relatively high brightness.
The only flat-panel technology with the potential to pose a realistic challenge
to LCDs in the medium term is OLED technology. The first factories for
OLEDs using either small molecules or polymers have started production, if in
relatively low volumes, see Table 1.1. Higher production volumes can be
confidently expected as the market acceptability and awareness of the capability
of OLEDs increases. A combination of the modulation of plane polarised light
provided by an OLED back-light by an LCD to create a hybrid OLED-LCD
may become a major commercial product in the near future. Oriented main-
chain polymers or anisotropic polymer networks in the nematic liquid crystal-
8 Chapter I
line state have established themselves recently as the most attractive source of
polarised electroluminescence. Moreover, liquid crystals in the smectic and
columnar states have great potential as non-dispersive charge-carrier transport
layers in multilayer OLEDs. Thus, the dominant flat-panel technology and its
main competitor share many common elements, which could be combined to
mutual benefit. Hybrid OLED-LCDs could expand the flat-panel displays
market and the market share of OLEDs and LCDs simultaneously, e.g. for
monochrome portable telephones with internet capability. The reasons for these
prognoses will be illustrated in the following chapters with the emphasis on the
dependence of these technologies on progress in research and development in
organic materials chemistry.
3 References
1 C. Williams, Electronic Materials for Displays Workshop, London, UK, 1997.
2 J. Smith, Proc. SID Euro Display ‘96, 1996,5.
3 J. Deschamps, Proc. SID Euro Display ‘96, 1996,31.
4 K. Kasano and T. Nakamura, Proc. SID Euro Display ‘81, 198 1,156.
5 C. Williams, Electronic Materials for Displays Workshop, London, UK, 1997.
6 J. F. Peyre, Proc. SID Euro Display ‘96, 1996, 169.
7 L. J. Hornbeck, SPIE Crit. Rev., 1989,1150,86.
8 L. J. Hornbeck, In?.Electron. Dev. Techn. Digest., 1993,381.
9 R. H. Mauch, Proc. SID Euro Display ‘96,1996,601.
10 L. J. Rothberg and A. J. Lovinger, J. Mater. Res., 1996, 11,3174.
11 T. Geelhaar, Liq. Cryst., 1998,24,91.
12 E. P. Raynes, Proc. SID Euro Display ‘96,1996,7.
CHAPTER 2
Liquid Crystals and Liquid

Crystal Displays (LCDs)
1 Physical Properties of Nematic Liquid Crystals

The vast majority of LCDs incorporate a thin film of a mixture of organic
compounds in the nematic phase, whose optical properties are modulated by an
applied electric field.' These form the subject of this chapter and of Chapter 3."'*
An LCD based on an electroclinic effect in achiral smectic A* phase,
constituted of optically active rod-like molecules, was reported in the late
1970s. However, such LCDs have not been manufactured commercially
despite intermittent interest and development, especially as spatial light mod-
''
ulators in applications for optical computing or telecommunications. A very
small number of commercial LCDs, which make use of the ferroelectric
properties of liquid crystalline mixtures in the optically active chiral smectic C*
phase to generate an electro-optical effect, are used as spatial light modulators,
e.g. in optical computing or telecommunication applications or projection
displays, e.g. for pilots' helmets.20921Surface stabilised ferroelectric liquid
crystal displays (SSF-LCDs) or other display types based on ferroelectric
liquid crystals may eventually have the potential to overcome some of the
limitations of nematic LCDS.~'*~' Analogous LCDs using antiferroelectric
liquid crystals may also have the potential to break through into the FDP
market.22 However, LCDs using smectic liquid crystals are still manufactured
on a very small scale, despite several decades of intense research and develop-
ment. Therefore, LCDs using smectic liquid crystals will not be described
further in this monograph. Laser-addressed LCDs for high-information-
content applications, such as very detailed maps, which are based upon light
scattering from side-chain liquid crystal polymers in the smectic A phase, have
been manufactured on a very limited scale at a very high unit cost, especially for
military and avionic applications. LCDs using ferroelectric side-chain liquid
crystal polymers for lightweight displays, such as head-up displays for pilots,
have also been developed. However, this type of LCDs utilising liquid crystal-
line polymers have yet to find commercial acceptance in the consumer displays
marketplace. Liquid crystalline polymers are described in depth
9
10 Chapter 2
Therefore, their applications in LCDs will not be discussed further in this
monograph. They may find application as electroluminescent materials as a
source of polarised light for LCDs, see Chapter 6. Columnar liquid crystals are
also not used as the active switching element in LCDs, although they are used as
passive optical compensation layers of negative birefringence for TN-LCDs and
STN-LCDs in order to compensate for interference colours and to increase the
effective viewing angle, see Section 7. They may also find application as organic
charge-carrier transport layers in multilayer OLEDs, see Chapter 5.
In order to understand the basic principles of operation of the many different
kinds of LCDs being developed and/or manufactured at the present time, it is
necessary to briefly describe the liquid crystalline state and then define the
physical properties of direct relevance to LCDs. First, the nematic, smectic and
columnar liquid crystalline states will be described briefly. However, the rest of
the monograph dealing with liquid crystals will concentrate on nematic liquid
crystals and their physical properties, since the vast majority of LCDs
manufactured operate using mixtures of thermo tropic, non-amphiphilic rod-
like organic compounds in the nematic state.
The unusual optical properties of liquid crystals had been remarked upon and
described for several centuries before their uniqueness as a state of matter was
r e c o g n i ~ e dTheir
. ~ ~ early
~ ~ reports described the strange melting behaviour and
appearance of some naturally occurring materials, either as pure compounds or
as gels in water, which have now been shown to be thermotropic or lyotropic
liquid crystals. Thermotropic liquid crystalline phases are formed under the
action of heat, see Figures 2.1 and 2.2, and the lyotropic liquid crystallinephases
are formed by the action of a solvent, such as water, usually with an amphiphilic
compound.28y29 However, the nature of these materials, or indeed their exact
- 145.5'C
Cr-N*
-178.5"C
N*-I
CRYSTAL, Cr CHIRAL NEMATIC PHASE, N* ISOTROPIC LIQUID, I

(CHOLESTERICPHASE, Ch*)
Figure 2.1 A schematic representation of the melting of crystalline cholesteryl benzoate

( 1 ) at 145.5 "C to form a chiral nematic (cholesteric) hase, which in turn
forms the isotropic liquid on further heating to 178.5 g,31
Liquid Crystals and Liquid Crystal Displays (LCDs) 11
-
Cr-Srn -
Sm-N
f--
N-Sm
N-I
___c
t--
I-N
CRYSTAL, Cr SMECTIC NEMATIC ISOTROPIC

PHASE, SrnA PHASE, N LIQUID, I
Figure 2.2 Schematic representation of the structures of a solid, a smectic phase (SmA),
the nematic phase and the isotropic liquid state formed by calamitic organic
molecules with a large length-to-breadth ratio.
chemical structure, was not known at the time. Therefore, the discovery of the
liquid crystalline state is usually attributed to the botanist Reinitzer from the
Institute for Plant Biology of the German University of Prague in 1888.30When
determining the melting points in capillary tubes of cholesteryl benzoate (l),he
noticed that it seemed to melt at 145.5"C to form a cloudy liquid with unusual
reflection colours. This opaque liquid then appeared to melt again at 178.5"C to
form a clear transparent liquid and then reform reversibly within a few degrees,
see Figure 2.1. This liquid crystalline state is now known to be the cholesteric
state (Ch), now referred to as the chiral nematic state (N*), i.e. the helical
equivalent, formed by some optically active materials, of the usual nematic state
(N), see Figure 2.3. The term 'flieBende Kristalle' flowing crystals or liquid
crystals was introduced31 to describe this unusual state of matter with the
fluidity of liquids and the optical properties of some crystals such as birefrin-
gence, see below. Liquid crystals are also often referred to as being meso-
morphic, exhibiting mesophases, i.e. nematic, smectic or columnar mesophases
or states 'in between' those of a crystal and a l i q ~ i d . A
~ ~liquid
- ~ ~crystal may
exhibit only one of these states or several of them at different temperatures, see
Figure 2.2. If the melting point of a particular compound is higher than the
temperature of the transition from the isotropic liquid to the liquid crystalline
phase then this phase is referred to as being monotropic. An enantiotropic phase
is observed above the melting point (Cr-Sm, Cr-N, Cr-Col).
In the nematic state there exists a local parallel orientation of the molecular
long axis of calamitic or rod-like molecules with a large length-to-breadth ratio,
see Figures 2.1 and 2.2 and Table 2.1.3843 This orientation may extend over
several micrometres. This parallel orientation of the molecular long axes is
absent in the liquid state, which may be considered as completely isotropic even
for rod-like molecules, see Figures 2.1 and 2.2. This is described by the
continuum theory, which postulates that the director of the nematic state
changes continually and gradually throughout the bulk of the nematic mate-
The average orientation of the molecular long axes at any one point is
defined as the director, n. There is an inversion symmetry axis at any one point
along the director and, therefore, n is identical to -n. The order parameter, S, is
a macroscopic scalar quantity, which represents the average orientation of the
molecules in the liquid crystal relative to the director?
12 Chapter 2
Figure 2.3 Schematic representation of the periodical helical structures of the chiral
nematic (cholesteric) phase. The pitch of the helix corresponds to the rotation
of the director through 360". There is no layered structure in a chiral nematic,
N*, phase.
s=
6 e
- (3 C O S ~ - 1)
where 8 is the angle between the long molecular axis of an individual molecule
and the director. In a completely disordered liquid made up of rod-like
molecules
COS* e = -31 (2)
and S is 0. In an ideal macroscopically ordered nematic liquid crystalline state
COS~ e=I (3)
and S = 1. For a typical compound in the nematic state at a temperature

Table 2.1 Transition temperatures (“C) for the nematic liquid crystals 2-16
Molecular structure Cr N I Ref
2 a 69 a 38
3 a < 25 a 38
4 a 49 37) 31,38
5 a 145 a 38
6 a 61 25) 0 39
7 a 40 25) 0 39
8 a 118 121 a 38,39
9 a 32 47 a 39,41
10 a 108 70) a 38
11 a 113 53) a 38,42
12 a 42 77 a 38,42
13 a 41 74 a 38
(continued)
14 Chapter 2
Table 2.1 continued
Parentheses represents a monotropic transition temperature.
relatively far away from the clearing point, TN-1, i.e. the temperature at which
the compound ceases to exhibit the nematic state and forms the liquid state,
usually S lies between 0.5 and 0.7. The Maier-Saupe theory postulates that the
order parameter, S, at a temperature T depends on the reduced temperature
Tred, (see Figure 2.4):
T
Tred =-
TN-I
The direction of the director is random in space and time. However, the director
Figure 2.4 The dependence of the order parameter, S, on the reduced temperature, T N - ~
T, where TN-1 is the clearing point (i.e. the temperature at which the transition
from the nematic phase into the isotropic liquid takes place) and the measure-
ment temperature, T.5'
of a bulk sample of a material in the nematic phase can be macroscopically and

almost uniformly oriented in one direction by a relatively weak external force
such as an electric or magnetic field, see Section 2, or an aligning surface, see
Section 6. The bulk realignment of the director, and thus the optic axis, by
surface forces and electric fields represents the underlying electro-optical effects
of liquid crystal displays using nematic liquid crystals, see Sections 3-7 and
Chapter 3. If a compound exhibits a nematic phase and one or more smectic
phases, the nematic phase is nearly always exhibited at a temperature above that
of the smectic phases due to the higher degree of order in smectic mesophases,
see Figure 2.2.
In the twelve smectic phases identified to date the molecules are arranged in
layers, see Figure 2.2.33-37The physical properties of these smectic phases
depend on the ability of the layers to slide over one another and to bend, the
average orientation of the director within the layers and the degree of order
within and between the layers. The director is orthogonal to the layer plane in
several smectic phases (SmA, SmB [hexatic] and SmE) and is tilted with respect
to the layer normal in some other smectic phases (SmC, SmF and SmI). The
more ordered phases (SmB, SmF and SmI) with hexagonal repeat units within
the layers are exhibited at lower temperatures. Some smectic phases with long-
range correlations between molecules in individual layers are more properly
referred to as plastic crystals (Crystals B, E, J, G, K and H). There are also a
large and increasing number of antiferroelectric and ferrielectric smectic phases.
Optically active compounds can also exhibit the chiral versions of the smectic
phases described above, denoted by an asterix, e.g. SmC*. The layer structure of
smectic phases results in a large bulk flow viscosity. This restricts their use in
those electro-optical display device configurations, which now use the nematic
phase, due to the long response times. However, a number of different types of
LCD using compounds in various smectic phases have been developed. These
are referred to above and do not form part of this monograph due to space
considerations and their very low share of the commercial LCD market.
Columnar liquid crystalline phases are formed generally by disc-shaped
molecules or self-assembled aggregates, see Figure 2.5. These are organised in
a supramolecular structure of nearly parallel columns with different degrees of
order in a two-dimensional lattice, see Section 4 of Chapter 5.36*37There is no
regular columnar structure in the nematic columnar phase formed at higher
temperatures. In columnar mesophases the molecular cores are organised above
each other in columns separated by the terminal chains oqthe outside of the
columns. Therefore, the intercolumnar distance ( z 15-40 A) is much greater
than the intracolumnar distance ( < 4.5 A) depending on the length and
conformation of the aliphatic chains. There are columnar phases with hexago-
nal and rectangular lattices (i.e. columnar hexagonal, colh; rectangular, col,;
oblique, c01,b) which are either ordered (e.g. colho) or disordered (e.g. colhd).
The nature of columnar phases, in which the aromatic cores of the molecules are
arranged above each other in columns, allows efficient charge-carrier migration
due to the overlap of the n-electron orbitals in the conjugated aromatic core of
16 Chapter 2
tn
Figure 2.5 Schematic representation of the Dho columnar phase with an ordered hexago-
nal arrangement of the columns of disc-shaped molecules with a regular period
of the discs within the columns. The director is parallel to the columns and
normal to the plane of the discs.
neighbouring molecules. This can be made use of in organic light-emitting

diodes (OLEDs), see Chapter 5 , Section 4 and photovoltaic effects.
Many new synthetic liquid crystals, which were found to exhibit smectic and/
or nematic mesophases, were prepared from the beginning of the 20th century,
especially at the Martin Luther University in Halle in Germany, and in the
United Kingdom at the University of Hull from the early 1950s to the present
day. These systematic studies served to establish the relationships between
molecular structure and the type of liquid crystalline state (mesophase)
exhibited, i.e. nematic, smectic or columnar, and at what temperature^.^^-^^
Some examples of the molecular structures of typical nematic liquid crystals (2-
16) synthesised up to the late 1960s are shown in Table 2.1 .3843 A typical liquid
crystal of this time possessed a linear structure with a central core containing
several collinear rings, a linear unsaturated linkage and two terminal chains.
The combination of one short alkyl or alkoxy chain on one ring and a polar
substituent on a second aromatic ring was found to promote nematic phase
formation. All of the compounds reported in Table 2.1 are aromatic with two
phenyl rings joined by an unsaturated central linkage. This seems to suggest that
the formation of the liquid crystalline state is promoted by a high degree of
conjugation and a large anisotropy of molecular shape and polarisability. These
findings are consistent with the basic theories describing the factors responsible
for the formation of the liquid crystalline state and especially the nematic
state .44-55
Calamitic compounds which exhibit a smectic and/or nematic phase usually
consist of a relatively rigid central core containing co-linear six-membered rings,
either aromatic rings, such as 1,4-disubstituted-phenylene, 2,5-disubstit u ted-
pyridine, 2,5-disubstituted-pyrimidine,3,6-disubstituted-pyridazine,and alicyc-
lic rings, such as trans- 1,4-disubstituted-cyclohexane, 1,4-disubstituted-bicy-
clo[2.2.2]octane, 2,5-disubstituted-dioxane.Heteroaromatic rings tend to lead
to the formation of smectic phases rather than the nematic phase unless
combined with a polar terminal function, such as a cyano group. The
dependence of the liquid crystalline transition temperatures on the nature of
Table 2.2 Transition temperatures ("C) of the nitriles 17-23
Molecular structure Cr N I
17 c 5 H 1 1 ~ c N 0 22.5 0 35
18 c 5 H 1 1 w c N 31 0 55
19 CN 0 62 0 100
20
21
22 c 5 H 1 1 w c N 62 0 85
23 c 5 H t 1 m C N 104 0 129
Parentheses represent a monotropictransition temperature.
the six-membered rings is shown by the data for the two-ring compounds (17-
23) collated in Table 2.2. The sequential replacement of the phenyl rings in the
fully aromatic compound (17) by the saturated alicyclic cyclohexane and
bicyclo[2.2.2]octane rings to yield the compounds 18-23 leads to a systematic
increase in the nematic clearing point. It is evident that the presence of a phenyl
ring and a non-conjugated cyano group in compound 19 gives rise to a low
clearing point. This illustrates the fact that it is very difficult to extrapolate from
one compound to another. This is dealt with in much more detail in Chapter 3.
Compounds containing 2,5-disubstituted five-membered rings, such as furan,
thiophene, 1,3,4-0xadiazoleand 1,3,4-thiadiazole, usually exhibit lower nematic
clearing points than those of similar compounds containing six-membered
rings. This is due to the non co-linear nature of the bonds of these five-
membered rings. Materials containing only two rings in the core are not
usually liquid crystalline if one of them is a five-membered ring. Condensed
2,6-disubstituted rings, such as naphthalene, quinoline, quinoxoline, tetralene,
chromane, dioxynaphthalene and trans,trans-decalin, do not give rise to
mesophase formation in the absence of another ring in the molecular core.
This is due to the small length-to-breadth ratio of such fused rings. Compounds
containing alicyclic or aromatic rings with more than six units, such as
cycloheptane, cycloheptatrienone, diazazulene and tropone, are generally not
mesomorphic, unless two other six-membered rings are also present. Two six-
membered rings are usually required in a molecule for the formation of a
mesophase. Carboxylic acids with only one ring, such as 4-methoxycinnamic
18 Chapter 2
Table 2.3 Typical values' and symbolsfor some physical properties of 4-cyano-4'-
pentylbiphenyl ( I 7)36p37
and 4- (trans-4-pentylcyclohexyl)benzonitrile
(18)588,59
Value/ Units
Property and symbol Value/ Units K15 ( I 7) PCHS (18)
Melting point (Cr-N) 22.5 "C 30 "C

Clearing point (N-I) 35 "C 55 "C
Melting point enthalpy 4.1 kcal mol- 21.35 kJ mol-'
Clearing point enthalpy 0.96 kJ mol-'
Density (p) 0.9706 g cm-3
Bulk viscosity (q) 22.5 mm s-'
Rotational viscosity (yl) 0.1507 pa s b
Refractive index of the extraordinary 1.702" 1.6173b,c
ray (4
Refractive index of the ordinary ray (no) 1.53ga 1.4924b3c
Birefringence (An) 0.163" 0.1249
Parallel dielectric permitivity (ell) 17.1" 17.5b*d
Orthogonal dielectric permitivity ( E ~ ) 7.2" 4.8 b9d
Dielectric anisotropy (AE) 9.9 a 12.7
Splay elastic constant (kl1) 9.6 10- l 2 N
Twist elastic constant (k22) 6.5 10-12Nb
Bend elastic constant (k33) 19.4 N
Bend/splay elastic constant ratio (k33/kl 1.46 2.03
Bend/twist elastic constant ratio (k33/k22) 3.0
Diamagnetic anisotropy ( x ) 1.51 lo-' m3 kg-3 0.46 m3 kg-3
a Measured at 29 "C;bmeasured at 20 "C;

'measured at 546 nm; dmeasured at 1 kHz.
acid, can exhibit a liquid crystalline phase. However, this is due to the formation
of molecular dimers due to hydrogen bonding between the carboxylic acid
groups. This results in a dynamic equilibrium of a mixture of one-ring and quasi
three-ring dimers, which possesses a nematic phase.
The rings in the molecular core may be linked directly or by a conjugated
linkage with two units, such as carboxy (-C02-), ethenyl (-CH-CH-), ethynyl
( - E C - ) , azo (-N=N-), azoxy (-N(O)=N-), azomethine (-CH-N-), or an
aliphatic linkage with two units, such as ethyl (-CH2CH2-) or methylenoxy
(-CH20-), see Table 2.1. Linkages with four units, such as butyl, (a-butenyl,
propyloxy and (a-propenyloxy have also been used but much more rarely.
Linkages with an odd number of units, such as methyl or propyl, lead to a non-
linear structure. Consequently, the clearing point of these compounds is usually
very low.
At least one terminal chain, such as an alkyl (R) or alkoxy chain (RO) is
attached to the molecular core. The second group may be a similar group or a
small substituent such as a halogen atom (F, C1, Br, I) or a cyano (CN), isothio-
cyanato (NCS), dimethylamino (N(CH3)2)or nitro (NO2) group or a fluorinated
group such as trifluoromethyl (CF3), trifluoromethoxy (CF30), difluoro-
methoxy (CF2HO), etc. Compounds with two short terminal alkyl or alkoxy
chains tend to exhibit a nematic phase. Mesomorphic compounds with short
Z
Figure 2.6 The molecular structure of a typical calamitic, thermotropic nematic liquid
crystal. The diameter of the molecule measured in the x-y plane is symmetrical
about the z-axis due to very rapid rotation about the molecular long axis.
terminal chains usually exhibit a nematic phase. However, most liquid crystal-
line materials with long terminal chains, e.g. 7-12 methylene units, often exhibit
a smectic phase. Mesomorphic compounds with longer chains generally only
possess smectic phases. A general order of increasing tendency of these terminal
groups to induce nematic phase formation has been established empirically, as
shown below, although these do vary for different molecular structures:
R O > R > C N > OCH3 > NO2 > C1> Br > N(CH3)2>CH3 > I > CF3 > H
Compounds with an odd number of methylene (CH2) or oxygen atoms in the

chain usually exhibit a higher clearing point than that of homologues with an
even number of units in the chain, i.e. an odd-even effect is often observed for
the nematic clearing point of a homologous series. The shapes of the plots of
nematic clearing point against number of units in the terminal shape are regular,
but may exhibit a variety of patterns. The same groups in a lateral, i.e. non-
terminal, position lead to a lower melting point and clearing point due to a
broadening of the molecular rotation volume, i.e. the length-to-breadth ratio is
20 Chapter 2
reduced. Smectic phases are also suppressed. This can lead to a broader nematic
phase at lower temperatures than those of corresponding materials without a
lateral substituent. Chain-branching can give rise to lower transition tempera-
tures for similar reasons.
The correlation between molecular structure, liquid crystal transition tem-
peratures and physical properties of the nematic phase of these materials of
relevance to individual types LCDs is dealt with extensively in Chapter 3.
2 Physical Properties of Liquid Crystals 1,15,25,56-58

The rod-like shape of liquid crystals means that their physical properties are
anisotropic, i.e. they are of a different magnitude when measured parallel or
perpendicular to the director,' see Figure 2.6 for a molecular model of a typical
nematic liquid crystal, i.e. 4-cyan04-pentylbiphenyl. Free rotation about the
molecular long axis gives an axis of symmetry parallel to the director so that the
values of the physical properties measured normal to the director, i.e. along the
y- and z-axes are identical. However, they differ from those measured parallel to
the director, i.e. along the x-axis. It is the anisotropic nature of the physical
properties of liquid crystals combined with the ability of magnetic and electric
fields to quickly influence the spatial orientation of the director, and thus the
optic axis, which enables electro-optical display devices to be constructed. The
fast reorientation of the director under the influence of a moderate electric field
is a result of their fluid nature and low viscosity. The physical properties are
temperature and pressure dependent as well as depending on the type of liquid
crystal state, e.g. nematic, smectic, columnar and the degree of order in that
state. The most important anisotropic properties of the nematic state of
relevance to LCDs are described below (see Table 2.3). 1*15725756-58
Optical Anisotropy (Birefringence)

The nematic phase of macroscopically aligned calamitic liquid crystals is
uniaxial due to the anisotropy of shape and polarisation. They are, therefore,
also optically anisotropic, i.e. birefringent, in that they exhibit different
refractive indices for light travelling parallel and perpendicular to the director
(optic axis). The physical property of birefringence is manifested by certain
crystalline solids, i.e. transparent crystals with a non-centrosymmetrical'lattice
structure, such as calcite. The complete freedom of rotation in liquids averages
out to zero any anisotropic molecular properties and renders the bulk liquid
optically isotropic. Incident plane polarised light entering a birefringent
medium, such as a non-centrosymmetrical transparent crystal or a thin
transparent film of a macroscopically aligned nematic phase, is split into two
mutually perpendicular components called the ordinary (0)and extraordinary
(e) rays. The electric field of the o-ray is perpendicular to the optic axis.
Therefore, the refractive index, no, of the medium for the o-ray is a constant
independent of propagation direction of the ray. The electric field of the e-ray
lies in a plane containing the optic axis of the medium. Therefore, the effective
Z
Figure 2.7 The refractive index ellipsoid of a uniaxial liquid crystalphase with the optical
axis parallel to ,the x-axis. The refractive index, no, of the ordinary ray is
independent of the direction of propagation. The refractive index, n,, of the
extraordinary ray is larger than no if the liquid crystalline phase is of positive
birefringence.'
refractive index, n,(0), of the medium for the e-ray depends on the angle, 0, the
ray makes with respect to the optic axis (see Figure 2.7):
A consequence of these different refractive indices for the ordinary ray and the
extraordinary ray is that the effective birefringence of the medium, An(@, also
depends on the propagation direction:
An(th) = n,(O) -no

22 Chapter 2
whereby the maximum value of the birefringence, An, is found for 8 = 90', i.e.
when the electric field of the e-ray is parallel to the optic axis. In this case
An = ne-no (7)
where n, = n,(O = 90'). Therefore, An is the difference between the refractive

indices for the o-ray and the e-ray of a macroscopically aligned nematic phase
propagating parallel and orthogonal, respectively, to the optical axis of the
nematic medium. Most nematic liquid crystals have positive birefringence,
An > 0, meaning that the e-ray is delayed with respect to the o-ray on passage
through the material. 1959*60
Interference between the e-ray and the o-ray, which have travelled with
different velocity through the nematic medium, gives rise to the coloured
appearance of LCDs operating with plane polarised light. For a wave at
normal incidence, the phase difference in radians between the o-ray and the e-
ray caused by traversing a birefringent film of thickness d and birefringence An,
is referred to as the optical retardation, S:
where 1, is the wavelength of light in a vacuum. The amount of optical

retardation is dependent on the wavelength so that interference occurs at
different frequencies resulting in the suppression of some of the visible spectrum
and, therefore, a coloured appearance. The optical retardation depends on the
path length, d, through the display. Therefore, the interference colours observed
are also dependent on the viewing angle, since d will vary with 8. The
birefringence and the optical retardation are also dependent on the wavelength
of light and temperature, since the magnitude of the refractive indices also vary
with these parameters. The material is no longer birefringent in the isotropic
liquid above the nematic clearing point (n, = no). Consequently, an isotropic
refractive index, ni, is observed, see Figure 2.8.
Elastic Constants'746
The spatial and temporal response of a nematic phase to a distorting force, such
as an electric (or magnetic) field is determined in part by three elastic constants,
k l l , k22 and k33, associated with splay, twist and bend deformations, respec-
tively, see Figure 2.9. The elastic constants describe the restoring forces on a
molecule within the nematic phase on removal of some external force which had
distorted the nematic medium from its equilibrium, i.e. lowest energy con-
formation. The configuration of the nematic director within an LCD in the
absence of an applied field is determined by the interaction of very thin layers of
molecules with an orientation layer coating the surface of the substrates above
the electrodes. The direction imposed on the director at the surface is then
1.64 -I
1.56
- "i
1
1.52
1.48 "0
1.40:
0
1 I I I I I I I
5 10 15 20 25 30 35 40 45 50 55 60 65 70
I I I I I I I
TT
Figure2.8 The dependence of the refractive indices, no and n,, of the ordinary and
extraordinary rays, respectively, on the temperature, T,for a typical nematic
liquid crystal. Above the clearing point, T,, there is no birefringence and only
one refractive index, nL is observed?8
Z Z Z
X X
X
SPLAY k i i TWIST k22 BENfi k33
Figure 2.9 Schematic representation of the elastic constantsfor splay, twist and bend, k l l ,
k22 and k33, respectively, of a nematic phase.',''
transmitted to the bulk of the nematic phase by elastic forces, see Section 6 and
Chapter 3.
Viscosity'sjl
The flow viscosity of a nematic phase also determines the spatial and temporal
response of the director to an applied field. The bulk viscosity of a nematic
phase depends on the direction of flow of each molecule with respect to the
director, averaged out over the whole of the sample. Therefore, bulk viscosity is
24 Chapter 2
dependent on the order parameter of the nematic phase and, consequently, its
magnitude is higher at lower temperatures. However, due to the anisotropic
molecular shape of calamitic liquid crystals, three viscosity coefficients are
required to characterise the viscosity of a nematic phase: 9 1 perpendicular to the
direction of flow, but parallel to the velocity gradient; 92 parallel to the direction
of flow, but perpendicular to the velocity gradient; and 93 perpendicular to the
direction of flow and to the velocity gradient.6' The response times and
operating voltages of the various types of LCDs depend on the individual
viscosity coefficients due to the spatial asymmetry imposed by the boundary
conditions and the applied electric field. The rotational viscosity, y1, of the
nematic phase is representative of the movement of a molecule from a
homogeneous planar conformation parallel to the substrate surfaces to a
homeotropic conformation with the molecular long axis (director) normal to
the substrate surfaces. The magnitude of the rotational viscosity of a nematic
medium often correlates well with the observed response times in LCDs, i.e. the
higher the viscosity the longer the response time under a given set of driving
conditions.
Dielectric Anisotropy
The response of the director of a nematic phase to an applied electric field is
dependent upon the magnitude of the dielectric permitivity (dielectricconstants)
measured parallel and perpendicular, ell and cl, respectively, to the director and
to the sign and magnitude of the difference between them, i.e. the dielectric
anisotropy, A&,see Equation 9 and Figure 2.10. Since the dielectric permitivity
measured along the x-axis is unique and the values of the dielectric permitivity
measured parallel to the y- and z-axes are the same,
24 'I
27
0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
TT
Figure 2.10. The dependence of the dielectric constants, ELand ~ ~ measured

1 . perpendicular
and parallel, respectively, to the nematic director, on the temperature, T, for a typical
nematic liquid crystal. Above the clearing point, T,, the dielectric anisotropy, AE = €\!-€A,
disappears and only one dielectric constant, E ~ ,the permitivity of the isotropic liquid, is
observed.'*'*
A&= E~I-EL (9)
The dielectric permitivity, E, of a dielectric, i.e. insulating,material is the ratio of

the capacitance, &at, of a parallel plate capacitor containing that material to
the capacitance, Gat, of the same capacitor containing a vacuum:
G a t
-- -&
CVaC
The dielectric constants of an aligned nematic phase are dependent upon both
the temperature and the frequency of the applied field at temperatures below the
clearing point. The dielectric permitivity, G, measured parallel to all three axes
above the clearing point in the isotropic liquid is the same. Therefore, the
dielectricanisotropy of the same compound in the liquid state is zero, see Figure
2.10. The sign and magnitude of the dielectric constants and, therefore, the
dielectric anisotropy are dependent upon the anisotropy of the induced
molecular polarisability, Aa, as well as the anisotropy and direction of the
resultant permanent molecular polarisation determined by permanent dipole
moments.
3 Liquid Crystal Displays2-18

Use has been made of three distinct physical phenomena associated with the
nematic state in order to modulate the passage of light through a thin film of a
material in the nematic state by an electric field."'* These electro-optical effects
have only been used since the late 1960s to create a wide variety of different
types of LCDs, see Chapter 3, although the interaction between the nematic
state and magnetic and electric fields had been investigated since the beginning
of the 19th
The first LCD was reported by Heilmeier, Zanoni and Barton at the RCA
Laboratories in Princeton, New Jersey in 196862*63and made use of an
electrohydrodynamic effect in a nematic medium under the action of an
applied electric field, although changes in the orientation of the director due
the stirring effect of an electric field on the nematic state had already been
observed in the 1 9 3 0 The~ ~ formation of visible domains, known as Williams
domains, at low voltages and light scattering at higher threshold voltages had
also been discovered at the RCA Corporation several years earlier.'*@ The
optical effect is due to electrohydrodynamic instabilities generated by the
movement under the action of an electric field of ions in the nematic
The movement of the ions under the action of an electric field results in the
circular movement of adjacent liquid crystal domains in different directions to
produce visible stripes referred to above as Williams domains.'*@ The resultant
shear is balanced out by the elastic and dielectric torques under steady state
conditions at low applied voltage. However, at higher fields these visible black-
and-white stripes are replaced by a bright white appearance as incident visible
light is scattered homogeneously. This effect is referred to as dynamic scattering
26 Chapter 2
and formed the underlying electro-optical effect of the first LCD to be
manufactured on a commercial scale as a flat panel display for portable
instruments. LCDs based on dynamic scattering are no longer manufactured
due to various reasons, see Chapter 3.
The second physical effect made use of to fabricate an LCD was a field effect,
i.e. the reorientation of the nematic director and, therefore, the optic axis, under
the action of an electric field. Organic compounds of high resistivity are
insulating dielectrics in the nematic state. Therefore, the application of an
electric field results in a reorientation of the nematic director either parallel or
perpendicular to the direction of the applied field due to the dielectric coupling
of the field with the induced and permanent dipole moments of the mole-
cules. 1756757The molecules align themselves with the molecular axis of greatest
resultant polarisation parallel with the field. The dielectric realignment of the
director of a nematic phase by an electric field can be described essentially by the
same equations as those used to define the effect of magnetic field on a nematic
liquid crystal with appropriate modifications due to the different nature of the
applied field. This corresponds to the Kerr effect observed in liquids, but is of a
much higher magnitude in the nematic state. This is the basic electro-optical
effect used in all commercial LCDs using nematic liquid crystals being
manufactured at the moment.2-'8
The third electro-optical effect using calamitic nematic liquid crystals makes
use of a flexoelectric effect manifested by a curved asymmetrical nematic
medium. This corresponds to piezoelectricity in crystals. The existence of
flexoelectricity in a nematic phase under certain boundary conditions was
predicted in the late 1 9 6 0 and ~ ~ then
~ confirmed experimentally several years
later.68However, LCDs using this effect, such as bistable nematic display^^"^'
are only in the development stage and as such they will not be discussed in this
monograph.
If polymerisation of a reactive monomer takes place in a mixture of the
monomer and a non-polymerisable liquid crystalline medium, then very many
small droplets of the unreactive nematic mixture are formed by phase separa-
tion from the polymer as the polymerisation progresses. These droplets can then
be addressed by an external electric field. Such mixtures can be polymerised to
form thin sheets. If the refractive indices of the polymer and the nematic phase
are well matched, then the device appears transparent in the on-state and
opaque in the off-state due to light scattering, if the size of the droplets is larger
than the wavelength of visible light, e.g. 1 pm. This is the physical basis of
polymer dispersed ( P D ) - L C D S . ~ ~These - ~ ~ LCDs are commercially available
and are used as coatings in privacy screens and many other applications. Gel-
stabilised LCDs are strongly related to PD-LCDs. Anisotropic gels are also
formed from a reactive monomer and a liquid crystalline mixture. However, the
monomer is usually also a liquid crystal. This forms a liquid crystalline polymer
network on polymerisation. The liquid Crystalline mixture occupies the space
within the n e t ~ o r k . ~ ' - ~
These
' liquid crystalline gels are much more stable to
external shocks since coupling between the immobile network and the low-
molar-mass liquid crystal molecules stabilises their orientation. In principle all
POLARISER
OPTICAL
RETARDER
FRONT SUBSTRATE
AND ELECTRODE
PASSIVATION LAYER
ALIGNMENT LAYER
NEMATIC
MIXTURE
ALIGNMENT LAYER
PASSIVATION LAYER
REAR SUBSTRATE
AND ELECTRODE
OPTICAL
RETARDER
ANALYSER
MIRROR
Figure 2.11 A schematic representation of the elements of a generalised liquid crystal

display (LCD).
the commercial LCD types described in Chapter 3 can be produced as a gel-

stabilised LCDs or polymer-dispersed LCDs. However, the electro-optical
effect and the fundamental modes of operation are very ~imilar.~~-'l
Therefore,
they will not be discussed further in this monograph.
4 Cell Construction of Liquid Crystal Displays

The basic modes of construction of the various commercial types LCD
described in detail in Chapter 3 share many common elements. They consist of
a very thin layer of a nematic liquid crystal mixture enclosed between two
transparent parallel glass substrates held apart by solid spacers and glued
together around the edges, see Figure 2.11 for a schematic representation of the
optical elements, some or all of which can be combined to construct an LCD.
The cell gap (d E 2-10 pm) should be as uniform as possible in order to
minimise variations in the optics of the display. Thick cells are turbid, due to
28 Chapter 2
Figure 2.12 Orientationof the nematic director at substrate. 40 is the preferred azimuthal
angle, 90, is the preferred polar angle and 9 is the tilt angle commonly used.
light scattering, and exhibit unacceptably long response times. Before the LCD
is assembled the inner surfaces of the glass substrates are often coated with a
passivation or barrier layer, e.g. silica, in order to prevent diffusion of ions from
the substrates into the nematic mixture. LCDs with full colour usually
incorporate a regular alternating pattern of red, green and blue picture elements
(pixels) formed using dyed colour filters. However, the absorption of two thirds
of the light passing through each pixel renders LCDs driven in reflection
insufficiently bright for commercial acceptance. Therefore, LCDs with colour
filters are operated either in a transmission mode or a transflection (combined
transmission and reflection) mode with a powerful back-light. This results in a
much higher power consumption for colour LCDs compared to that of
comparable monochrome LCDs with the same brightness. Electrodes made of
a transparent conducting material, usually indium-tin oxide (ITO), are depos-
ited on top of the these layers. Another thin barrier layer intended to prevent
diffusion of ions into the nematic mixture may be deposited on top of the
electrodes. This is followed by an alignment layer, which will be in direct contact
with the nematic mixture in order to induce a homogeneous orientation of the
director in the azimuthal and zenithal plane of the device, see Figure 2.12. The
two glass substrates are then assembled and glued together leaving a hole so that
the evacuated cell can be filled with a nematic liquid crystal under positive
pressure, cleaned and then sealed, e.g. with epoxy resin or gold. The glass
substratesmay be offset in order to allow the drive electronics to be connected to
LCDs with direct addressing or multiplexed addressing. The contacts are
usually sheets of plastic coated alternating strips of a conducting polymer and
an insulating polymer. The diameter of the strips correspond to the width of the
row or column electrodes on LCDs with multiplex addressing.
Direct, multiplex and active matrix addressing are the three electronic drive
methods used to generate the appropriate voltage at a particular pixel of an
LCD, see Figures 2.1 1-2.14. The size, shape and pattern of electrodes on LCD
substrates are fashioned to be compatible with the chosen method of addres-
sing. In directly addressed LCDs the desired pattern of pixel electrodes is
created by etching on one surface. A non-patterned back electrode on the
second surface provides the electrical contact. LCDs with multiplexed addres-
- FRONT
SUBSTRATE
REAR -
SUBSTRATE
- SEGMENT
ELECTRODES
BACK
ELECTRODE - SEGMENT
mECTRODES
ALIGNMENT [
LAYER
Figure 2.13 Schematic representation of a simple, low-information-content, alpha-

numeric TN-LCD with direct addressing and segmented electrodes.
- FRONT
SUBSTRATE
REAR -
SUBSTRATE
- COLUMN
EI.,EClXODE
ROW - - PIXEL
ELECTRODE
Figure 2.14 Schematic representation of the pixels, made up of a pattern of orthogonal

rows and columns of electrodes, of an LCD with multiplexed addressing.”
sing require a series of uniform rows of electrodes on one surface and an

equivalent, sometimes identical, pattern of electrode columns on the other
surface. The rows and columns are arranged orthogonal to each other to form a
pattern of rectangular or square pixels. LCDs with active matrix addressing use
30 Chapter 2
one silicon substrate covered with a regular array of thin film transistors and a
back electrode on the second substrate surface.
A plastic sheet polariser may then be attached, usually by contact bonding, to
the outer surface of one or both substrates depending on the type of LCD, see
Chapter 3. The polarisation axis of a sheet polariser makes an angle, a or p to
the alignment direction and, therefore, the optical axis (director) of the nematic
mixture, at each substrate surface. An optical retardation layer may also be
present between the glass substrate and the polariser. Another passive optical
element, which scatters transmitted light through a wider viewing angle cone
may also be attached to the outside of the display.
5 Addressing Methods for Liquid Crystal Displayd2

The electronics of an LCD convert the information to be displayed into a series
of applied voltages of a given sign, amplitude and pulse width in order to
activate the desired pixels at the appropriate time. This results in the modulation
of the light intensity at the addressed pixels, thus creating the image or
information to be displayed. The main methods by which this can be achieved
are described below.
Direct Addressing
LCDs with direct addressing for each pixel are driven directly with a dedicated
electrical contact and a driver for each segment of the digit, see Figure 2.13. The
first commercial LCDs, which were designed for operation in simple, low-
information-content displays such as digital clocks and calculators, were
constructed with segmented electrodes for direct addressing. Direct addressing
allows the off-state voltage to be zero and the on-state voltage to be several
times larger than the threshold voltage. Therefore, a good contrast can be
attained as well as low power consumption, e.g. in a twisted nematic (TN)-LCD
with low information content. However, the market demand for flat panel
displays with a higher information content, i.e. personal organisers, notebooks
and portable computers, meant that more sophisticated addressing schemes and
electrode patterns had to be developed. High-information-content LCDs are
almost impossible to realise with direct addressing due to the high cost of using
many individual drivers for each pixel or digit segment and the absence of
sufficient space for the electrical contacts.
Multiplex Addressing * 2382

Multiplex addressing with M electrode columns and N electrode rows allows
M x N pixels to be created driven by M + N connections made at the end of
each row and column, see Figure 2.14. This type of addressing allows large-area,
high-information-content LCDs to be produced with acceptable contrast and
viewing angles, see Chapter 3.
In an LCD with multiplex addressing, parallel rows and columns of elec-
trodes arranged perpendicular to one another create a matrix of rows and

columns. A scan or select voltage pulse (Vs) is applied to each row sequentially,
whereas the columns are addressed by data voltage pulses ( V D ) . These
incorporate the information to be displayed. When the amplitude of the
resultant voltage of the row and column voltages in phase is larger than the
threshold voltage, the nematic director is reoriented by the electric field and the
pixel addressed is in the on-state. If a non-select voltage ( V N S )below the
threshold voltage is applied to the addressed pixel, the elastic forces restore the
original alignment configuration of the nematic director in the off-state. This is
the case for the commercial LCDs using nematic mixtures described in Chapter
3, although other LCD types using nematic liquid crystals may use voltage
pulses of opposite sign to switch a pixel on and off. Root mean square (RMS)
voltage is usually applied in most commercial LCDs, since the reorientation of
the nematic director is a result of the coupling between the electric field with the
induced polarisation of the nematic medium, which is proportional to the
square of the applied electric field.
The number of addressable lines in a multiplexed LCD with acceptable
contrast and viewing angle dependency is limited, although to a much lesser
extent than using direct addressing. The maximum number of addressable lines
(N) is given by (see Figure 2.15):'*
Therefore, the more lines there are to be addressed, the smaller the difference
between the select and non-select voltages becomes, i.e. there is only a difference
of about 11% between Vs and VNS for 64 addressable lines; An undesirable
consequence of this small voltage difference is the inadvertent activation of
adjacent pixels (cross-talk), which reduces the contrast. In order to address a
VNS Vs V0LTAGEa.u.
Figure 2.15 Schematic representation of a typical plot of the transmission versus applied
voltage for an LCD with multiplexed addressing. VS and VNSare the select
and the non-select voltages, respectively.12
32 Chapter 2
large number of lines with good contrast, the electro-optical response curve, i.e.
transmission versus voltage, should exhibit as steep a slope as possible, see
Figure 2.15, although this inadvertently reduces the degree of grey-scale. This
complicates the generation of full colour, since grey-scale allows the intensity of
light transmitted to be controlled by modulating the amplitude and pulse width
of the applied voltage, see Figure 2.15. Consequently, LCDs with multiplexed
addressing are only capable of displaying a limited amount of information with
good contrast over a relatively wide range of viewing angles, unless the electro-
optical transmission curve is very steep, e.g. for STN-LCDs, see Chapter 3.
LCDs without a steep electro-optical transmission curve, such as TN-LCDs, see
Chapter 3, cannot display high information-content with good contrast and
wide viewing angles.
Active Matrix Addressing

LCDs with high-information-content are presently realised commercially by
using versions of passively addressed TN-LCDs and STN-LCDs with multi-
plexed addressing, or ECB-LCDs, TN-LCDs or in-plane switching (IPS)-LCDs
with active matrix addressing, which are all discussed in Chapter 3.83-87Most
commercial LCDs with active matrix addressing use a discrete thin-film
transistor ( T W g 4or much less frequently a diodes5 at each pixel of an array
of many tens of thousands or even hundreds of thousands of pixels fabricated
on an amorphous silicon see Figure 2.16 for the most common
commercial LCD with active matrix addressing, i.e. a TN-TFT-LCD, see
SIGNAL ELECI'RODES COUNTER ELECTRODE
I
I I
1.
THIN FILM
TRANSISTOR
NON-ACTIVE
AREA
SILICON
BACKPLATE
I
I
I
I ' I
Figure 2.16 Schematic representation of a high-information-content LCD with active

matrix addressing provided by thin j l m transistors (TFTs).i2
Chapter 3. Active matrix addressing using cadmium selenide (CdSe) thin
film-transistors was developed in the early 1960s to address several kinds of
display devices.’* High-information-content, large-area LCDs with active
matrix addressing generally exhibit high-contrast, wide-viewing angles,
absence of cross-talk, fast response times and adequate grey-scale for full
colour. A compromise between brightness and high power consumption must
be made due to the low luminosity of such LCDs with two polarisers. The cost
of manufacture of LCDs with active matrix addressing is high due to the
presence of the active substrate. However, the cost is falling steadily with
improved manufacturing methods and machinery and pixel repair techniques.
LCDs with active matrix addressing are switched on by a voltage pulse at the
thin film transistor or diode. The charge at the pixel, which corresponds to a
capacitor, should remain constant until the pixel is addressed again in the next
frame. Leakage of charge away from the pixel into the nematic mixture reduces
the effective voltage at the pixel. This results in lower contrast. This effect can be
counteracted by using very pure nematic materials with a very high resistivity
value, see Chapter 3.
6. Organic Polymer Alignment Layers

The LCDs described in Chapter 3 are all based on field effects. Either the on-
state or the off-state are stabilised or regenerated in the absence of an applied
field by an alignment layer of some kind covering the inner cell surface in some
way or A nematic phase deposited on an aligned substrate
spontaneously develops an equilibrium, or easy orientation, at the interface
between the alignment layer and the nematic material. The easy orientation is
characterised by the polar angle 8 0 and the azimuthal angle 4 0 , see Figure 2.12.
The corresponding polar and azimuthal anchoring energies, Fa(@ and Fa(4),
respectively, represent the energy required to deform the director from Oo and
4 0 , respectively. These anchoring energies can be expressed in the Rapini-
Papoular approximations
where We is the polar part of the anchoring energy and W4 is the azimuthal part
of the anchoring energy with 40(Oo)fixed. The polar anchoring energy may be
an order of magnitude larger than the azimuthal anchoring energy, e.g.
W, > J rn-’ and W4 > J m-’ for strongly rubbed p~lyimide.’~ The
interaction between the substrate and the liquid crystalline phase is described
phenomenologically by the ratio of the surface tension of the surface (7s) and
the surface tension of the liquid crystal ( y ~ c ) The
. alignment is homeotropic if
the intermolecular forces in the bulk of the liquid crystalline phase are larger
than the forces between the surface and the liquid crystal (7s < yLc). Homo-
geneous alignment occurs when the surface anchoring forces are greater. The
nature of the interaction between the surface and the liquid crystal molecules is
34 Chapter 2
complex and includes dispersive (Van der Waals), polar, steric, order-electric
and ionic terms. The topology of the surface may also be an important factor as
indicated by the alignment of the nematic director parallel to the grooves of
gratings in order to minimise the elastic strain.93 A unique azimuthal orienta-
tion usually requires surface treatment of the alignment layer, for example, by
mechanical rubbing or illumination with polarised ultra violet (UV) light.
A self-assembled monolayer (SAM) of long aliphatic chains fixed to the
substrate surface through reactive siloxane endgroups is the common method
used to generate homeotropic alignment.94 Aliphatic SAMs usually generate
homeotropic alignment, perhaps with a small pretilt angle measured from the
normal to the plane of the cell (8, w lo), due to the low value of the surface
tension attributable to the planar surface of apolar methyl groups in a terminal
position. However, most LCD types require homogeneous alignment with the
director inclined at a small angle to the plane of the cell. Originally inorganic
oxides such as SiOz were deposited from an oblique angle under high vacuum to
generate homogeneous alignment.95Nematic liquid crystals wet these inorganic
surfaces producing high ( w 10") pretilt angles, 8, where 8 = 90" - 8,. However,
the production technique to produce these alignment layers is intrinsically
expensive and not compatible with continuous production processes. The most
common materials used over the last two decades at least to create an
orientation effect are polyimide derivatives with a low value for the surface
tension, such as those (24-29) shown in Table 2.4.96'00 A soluble precursor
polyamide is first deposited and then cured at high temperature to form a
uniform layer of insoluble and intractable polyimide with very high glass
transition temperaturis and decomposition temperatures. In commercial LCD
fabrication these layers are rubbed (buffed) mechanically in one direction to
create a unique orientation direction in the azimuthal and zenithal plane^.^^*^'
The buffing of organic polymers is the standard alignment technique for
homogeneous alignment with a defined pretilt angle in commercial L C D S . ~ " ' ~ ~
Mechanical rubbing can cause pixel damage and generate static electricity
leading to dielectric breakdown or higher conductivity of the nematic layer. This
is especially important for LCDs with active matrix addressing, where the
presence of even very small amounts of dust or charged particles can lead to
lower production yields and higher manufacturing costs. Non-contact align-
ment layers use plane polarised UV light to generate a surface anis~tropy.~'
This, in turn, can lead to a uniform alignment of the nematic director. There-
fore, they have considerable potential to replace buffed polymer orientation
layers especially for LCDs with active matrix addressing. A diverse range of
related non-contact alignment methods making use of the same effect, i.e. the
cis-trans photoisomerisation of azo-dyes, see Figure 2.17, to produce an
orientation effect of the nematic director by co-operative effects, has been
developed. 101-109 The azo dye may be located in the orientation layer,"' the
nematic mixture, '02~'03 or covalently bonded to polymers deposited on the
substrate surface '04-'06 or in monolayers deposited by Langmuir-Blodgett
technique^.'^^-'^^ Non-contact alignment layers using azo-chromophores
require the use of laser light for relatively long exposure times to induce the
Table 2.4 Polyimides (24-29) used as alignment layers, especially as non-contact
alignment layers, for LCDs.
Molecular structure Ref
96
25 96
26 97
n
27 98
(continued)
desired isomerisation of the dye molecules, which are present in high concentra-
tion, e.g. 2 wt%. This may result in an undesirable colouration of the display.
Unfortunately, the thermal and UV stability of the orientation layers may be
insufficient for commercial LCDs.
Poly(viny1cinnamate) (PVC) films exposed to linearly polarised ultraviolet
36 Chapter 2
Table 2.4 continued
100
hu (ultra violet)
hu (visible)
Figure 2.17 Cis (2)and trans ( E ) isomericforms of azobenzene.
light can also generate homogeneous alignment of the nematic dire~tor."~'lo-' l6

The resultant orientation is perpendicular to the polarisation direction of the
plane polarised UV light. The alignment is caused by an anisotropic depletion of
the cinnamate sidechains as a consequence of a 2 + 2 cycloaddition reaction,
although cis-trans isomerisation of the carbon-carbon double bond may also
be present at low fluences, see Figure 2.18.'12 A low azimuthal surface
anchoring energy (z4 x J m-') has been found for PVC it~e1f.l'~
However, stronger anchoring of the nematic director is found when derivatives
of PVC are used."6116 Small pretilt angles (0 x 0.3") normal to the surface,
which are required to avoid areas of reverse tilt (see Chapter 3) have been
generated by a double exposure of the PVC layer.'14 However, larger pretilt
angles have been achieved by exposure at non-normal incidence of a coumarin
side-chain polymer,' 17,' l8 see Figure 2.19. The alignment direction is parallel to
the polarisation direction of the UV light. LCDs with improved viewing-angle-
dependence of the contrast can be realised using this technique to create sub-
2$2L+
POLY(V1NYLCI"AMATE)
PHOTOREACIlVE \'
n
n
1. HEAD-TO-HEAD
PHOTOPRODUCT
&
TRANs/CIS-ISOME@SATION
Figure 2.18 Photochemical reactions of PVC on irradiation of UV light: (a) 2 + 2

cycladdition and ( b ) translcis-(E1Z)-isomerisation. E indicates the polarisa-
tion direction of the incident beam.
J
n
EV hu
PHOTOREACTIVE HEAD-TO-HEAD
COUMARIN POLYMER PHOTOPRODUCT
Figure 2.19 Schematic representation of the anisotropic crosslinking of a photoreactive

coumarin polymer by the action of polarised UV light to produce an
anisotropic network as a non-contact ali nment layer. E indicates the
polarisation direction of the incident beam."-'"
domain pixels, each with a different microscopic alignment direction.I 199120

Anchoring energies comparable to those achieved using standard rubbed
polyimide have been determined for non-contact alignment layers. 121-126
The anisotropic photodegradation of polyimide layers, such as those poly-
imides shown in Table 2.4, using plane polarised UV light can also be used to
induce uniform alignment of the nematic director by a non-contact
method. 125~126However, the ablation process proceeds by breaking chemical
bonds and removing the debris thermally. This generates substantial amounts of
chemical decomposition products and a rough surface. These can pose a serious
problem for LCDs with active matrix addressing,,if the debris is not completely
removed from the substrate surface.
38 Chapter 2
TRANSMITTED
LIGHT WAVE FRONT
INCIDENT
LIGHT WAVE FRONT
Figure2. Schematic representation of viewing angL;s, 00 and 4, in the zenithal and

azimuthal planes for light traversing an LCD.”
7 Organic Polymer Compensation Films for Liquid

Crystal Displays127-141
Optical retarders are used as compensation films for some types of LCD in
order to correct for the very strong viewing angle dependence of the contrast,
luminosity, grey-scale and colour by matching the optical symmetry of the
compensation layer to the optical pattern of the nematic layer, but with the
opposite sign of birefringence. The viewing-angle dependence of these optical
properties is caused by the anisotropic nature of the nematic layer in the LCD
and polarisers. These optical properties are optimised in commercial LCDs
for viewing directly from above. Therefore, the optical performance of an LCD
usually degrades as the angle of view increases from the normal to the plane of
the cell, see Figure 2.20. This viewing angle dependence is also asymmetric, i.e. it
depends on the direction of view and not just the angle of view. These problems
are particularly acute for LCDs based on wave-guiding, such as TN-LCDs, and
LCDs based on interference effects, such as ECB-LCDs and STN-LCDs (see
Chapter 3). The optical properties of LCDs with the optical axis in the plane of
the cell, such as IPS-LCDs, are much less viewing-angledependent (see Chapter
3)-
There is an almost infinite potential to combine different kinds of compensa-
tion films in many different ways to achieve a given effect, e.g. compensation
layers of negative birefringence with the optical axes in the plane of the cell
situated on either side of a TN-LCD improve the viewing angle dependence in
the horizontal plane, but not in the vertical However, if the optical
axis of the compensation layer is tilted with respect to the plane of the cell, then
superior optical properties in the vertical direction are also achieved. Compen-
n, c no and An c 0
DIRECTION OF STRETCHING
DIRECTION OF STRETCHING
Figure 2.21 Schematic representation of the linear structure of a stretched polycarbonate

and a stretched polystyrene component of an optical retardation sheet.
sation films may be organic layers of positive or negative birefringence, whose

optical axis makes an angle to the plane of the cell. This angle may be fixed or
increase in a splayed configuration of the nematic director. The optical axes of
individual layers may be parallel to the nematic director at the same substrate
surface or the major axis of absorption of the polariser attached to this
substrate, or may make some angle to either one or both of them. A number of
stacked compensation layers may be fixed either above and/or below the cell
situated between the polariser and the substrate. Computer programs, such as
the Jones or Berreman programs,139are often used to design compensation films
by predicting the effect of a compensation layer of a given birefringence,
thickness and orientation of the optical axis on the optics of the LCD in
question.
Simple compensation films consist of oriented films of mainchain polymers
incorporating polarisable aromatic units, e.g. a polystyrene or polycarbonate
polymer, see Figure 2.21. The films are mechanically stretched to align the
polymer chains parallel to each other in the direction of mechanical stress.127
The stretching process creates a polarisation anisotropy, since the polarisability
parallel to the long polymer axis is greater than that orthogonal to it for the
polycarbonate. Therefore, the retardation film is of positive birefringence, since
ne is greater than no (see Equation 7). A stretched styrene polymer sheet128
exhibits a negative birefringence, since the refractive index orthogonal to the
oriented polymer chain will be larger than that parallel to the chain, since ne is
less than no (see Figure 2.21).
More sophisticated optical compensation layers can be produced as aniso-
tropic networks from reactive mesogens (liquid crystals) in a macroscopically
40 Chapter 2
aligned liquid crystalline state. 129-141 These films are generally prepared by spin
casting a reactive mesogen onto a substrate, evaporating off the solvent and
then polymerising and crosslinking the reactive compound in the nematic state
using a small amount of a photoinitiator with UV light. This process can be
used to produce a solid three-dimensional network. The nematic liquid state is
usually preferred to any of the smectic mesophases due to the lower viscosity of
the former. However, columnar networks are produced commercially, despite
their very high viscosity. Some commercial compensation films, e.g. Fuji WV
Film Wide View A, using columnar liquid crystals, are a hybrid of multiple
compensation layers of positive and negative birefringence with the optical axis
in the azimuthal plane of the cell as well as configured with the optical axis tilted
and splayed in the zenithal plane. They are prepared from photopolymerisable
columnar liquid crystals on a carrier film of cellulose triacetate, which also
contributes to the compensation effect. Other products incorporate a nematic
anisotropic network with positive birefringence and a tilted optical axis'35 or
holographic form birefringence.136
There are many other processing methods which can reduce the viewing angle
dependence of the optical properties of LCDs, e.g. pixel-divided cells'40 and
multidomain cells.'41 However, these are beyond the scope of this monograph
and will not be discussed further. Optical retardation sheets are often preferred
by LCD manufacturers and systems integrators of FPDs, because commercially
available compensation films can simply be laminated to the LCD using contact
bonding.
Optically anisotropic non-contact alignment layers could also be used to
generate photo-patterned, high-resolution optical retarders, but within the cell
rather than as an external This could offer practical advantages, since
there would be no need to fix compensation layers to the LCD by contact
bonding. However, the birefringence of such alignment layers is generally low.
Consequently, very thick films would have to used in order to obtain the desired
retardation. However, they could be used in hybrid compensation layers in
combination with anisotropic networks of high birefringence formed from
reactive m e ~ o g e n s .l9~ ~Macroscopically
~' oriented films can be formed by
orientation of the reactive mesogen in the nematic state on the photoalignment
layer, which is then cross-linked to form a patterned optical retarder as an
anisotropic network.
8 References
1 D. Dunmur and K. Toriyama, in 'Physical Properties of Liquid Crystals', Eds.
D. Demus, J. W. Goodby, G. W. Gray, H.-W. Spiess and V. Vill, Wiley-VCH
Verlagsgesellschaft GmbH, Weinheim, Germany, 1999, p. 85.
2 V. Freederieks and V. Zolina, Trans. Faraday SOC.,1933,29,919.
3 V. Zwetkoff, Acta Physico. Chem. USSR,1937,6,885.
4 P. Chatelain, Bull. SOC.Fr. Mine'ral. Cristallogr., 1943,66, 105.
5 R. Williams, J. Chem. Phys., 1963,39, 384.
6 W. Helfrich, Mol. Cryst. Liq.Cryst., 1973,21, 187.
7 M. Schadt and W. Helfrich, Appl. Phys. Lett., 1971, 18, 127.
8 G. Baur, Mol. Cryst. Liq. Cryst., 1981,63,45.
9 T. J. Scheffer and J. Nehring, Appl. Phys. Lett., 1984,45, 1021.
10 C. Waters, V. Brimmel and E. P. Raynes, Proc. Japan Display ’83, 1983,396.
11 L. M. Blinov, in ‘Handbook of Liquid Crystal Research’, Eds. P. J. Collings and
J. S.Patel, Oxford University Press, Oxford, UK, 1997.
12 T. J. Scheffer and J. Nehring, Ann. Rev. Muter. Sci., 1997,27, 555.
13 M. Schadt, Ann. Rev. Muter. Sci., 1997,27,305.
14 V. G. Chigrinov, in ‘The Physics of Liquid Crystal Display Devices’, Springer, New
York, USA, 1998.
15 I. C. Sage, in ‘Handbook of Liquid Crystals’, Eds. D. Demus, J. W. Goodby, G. W.
Gray, H.-W. Spiess and V. Vill, Wiley-ITCH Verlagsgesellschaft GmbH, Weinheim,
Germany, 1998, Vol. 1, p. 731.
16 L. M. Blinov, in ‘Handbook of Liquid Crystals’ Eds. D. Demus, J. W. Goodby,
G. W. Gray, H.-W. Spiess and V. Vill, Wiley-VCH Verlagsgesellschaft GmbH,
Weinheim, Germany, 1998, Vol. 1, p. 477.
17 L. M. Blinov and V. G. Chigrinov, in ‘Electro-optic Effects in Liquid Crystal
Materials’, Springer, New York, USA, 1994.
18 E. P. Raynes, in ‘The Optics of Thermotropic Liquid Crystals’, Eds. S. Elston and
R. Sambles, Taylor and Francis, London, UK, 1998, p. 289.
19 S.Garoff and R. B. Meyer, Phys. Rev. Lett., 1997,38, 848.
20 R. B. Meyer, Mol. Cryst. Liq. Cryst., 1977,40, 33.
21 S . T. Lagerwall, in ‘Handbook of Liquid Crystals’, Eds. D. Demus, J. W. Goodby,
G. W. Gray, H.-W. Spiess and V. Vill, Wiley-VCH Verlagsgesellschaft GmbH,
Weinheim, Germany, 1998, Vol. 2B, p. 515.
22 A. Fukuda, J. Muter. Chem., 1994,4,997.
23 V. Percec and C. Pugh, in ‘Side Chain Liquid Crystalline Polymers’, Ed.
C. B. McArdle, Blackie, Glasgow and London, UK, 1989, p. 30.
24 J. C. Dubois, P. le Barny, M. Mauzac and C. Noel, in ‘Handbook of Liquid
Crystals’, Eds. D. Demus, J. W. Goodby, G. W. Gray, H.-W. Spiess and V. Vill,
Wiley-VCH Verlagsgesellschaft GmbH, Weinheim, Germany, 1998, Vol. 3, p. 207.
25 ‘Handbook of Liquid Crystals’, Eds. D. Demus, J. W. Goodby, G. W. Gray, H.-W.
Spiess and V. Vill, Wiley-VCH Verlagsgesellschaft GmbH, Weinheim, Germany,
1997.
26 V. Vill, Liq. Cryst., 1998,24,21.
27 V. Vill, ‘LiqCryst. 1 - Database of Liquid Crystalline Compounds for Personal
Computers’, LCI Publisher, Hamburg, Germany, 1996.
28 D. Blunk, K. Praefcke and V. Vill, in ‘Handbook of Liquid Crystals’, Eds.
D. Demus, J. W. Goodby, G. W. Gray, H.-W. Spiess and V. Vill, Wiley-VCH
Verlagsgesellschaft GmbH, Weinheim, Germany, 1998, Vol. 3, p. 305.
29 C. Fairhurst, S.Fuller, J. Gray, M. C. Holmes and G. J. T. Tiddy, in ‘Handbook of
Liquid Crystals’, Eds. D. Demus, J. W. Goodby, G. W. Gray, H.-W. Spiess and V.
Vill, Wiley-VCH Verlagsgesellschaft GmbH, Weinheim, Germany, 1998, Vol. 3, p.
341.
30 F. Reinitzer, Monatsch. Wiener Chem. Ges., 1888,9,421.
31 0.Lehmann, Z. Phys. Chem., Leipzig, 1889,4,462.
32 D. Vorlander, in ‘Kristalline FliiDigikeiten und FluDige Kristalle’, Engelmann,
Leipzig, Germany, 1905.
33 D. Demus, in ‘Handbook of Liquid Crystals’, Eds. D. Demus, J. W. Goodby, G. W.
Gray, H.-W. Spiess and V. Vill, Wiley-VCH Verlagsgesellschaft GmbH, Weinheim,
Germany, 1998, Vol. 1, 134.
34 H. Sackmann, Liq. Cryst., 1989,5,43.
35 G. W. Gray, in ‘Molecular Structure and the Properties of Liquid Crystals’,
Academic Press, London, UK, 1962.
36 G. W. Gray, in ‘Advances in Liquid Crystals’, Academic Press, New York, USA,
1976, Vol. 2, p. 1.
42 Chapter 2
37 G. W. Gray, Phil. Trans. R. SOC.Lond., 1983, A309,77; 1990, A330,73.
38 R. E. Rondeau, M. A. Berwick, R. N. Steppe1 and M. P. ServC, J. Am. Chem. SOC.,
1972,94,1096.
39 R. Steinstrasser, 2. Naturforsch., 1972,27b, 774.
40 J. A. Castellano, J. E. Goldmacher, L. A. Barton and J. S . Kane, J. Org. Chem.,
1968,33,3501.
41 R. Steinstrasser and L. Pohl, 2. Naturforsch., 1971,26b, 577.
42 M. Rosenberg and R. A. Champa, Mol. Cryst. Liq. Cryst., 1970,11, 191.
43 H. Kelker and B. Scheurle, Angew. Chem. Int. Ed. Eng., 1969,8,884.
44 H. Zocher, Trans. Faraday Soc., 1933,29,931 and 945.
45 C. W. Oseen, Trans. Faraday SOC.,1933,29,883.
46 F. C. Frank, Discuss. Faraday Soc., 1958,25,19.
47 J. L. Eriksen, Art. Rat. Mech. Anal., 1962,9,371.
48 J. L. Eriksen, Trans. SOC.Rheof., 1961,523.
49 F. M. Leslie, Art. Rat. Mech. Anal., 1968,28,265.
50 F. M. Leslie, in ‘Physical Properties of Liquid Crystals’, Eds. D. Demus, J. W.
Goodby, G. W. Gray, H.-W. Spiess and V. Vill, Wiley-VCH Verlagsgesellschaft
GmbH, Weinheim, Germany, 1999, p. 25.
51 W. Maier and A. Saupe, 2. Naturforsch., 1958,13a, 564; 1959,14a, 882; 1960,15a,
287.
52 R. Alben, Mol. Cryst. Liq. Cryst., 1971,13, 193.
53 P. J. Flory, and G. Ronca, Mof. Cryst. Liq. Cryst., 1979,54,311.
54 M. A. Cotter, Phil. Trans. R. SOC.,London, 1983, A309, 127.
55 M. A. Osipov, in ‘Physical Properties of Liquid Crystals’, Eds. D. Demus, J. W.
GmbH, Weinheim, Germany, 1999, p. 40.
56 P. G. de Gennes, in ‘The Physics of Liquid Crystals’, Oxford University Press, UK
1974.
57 P. G. de Gennes ayd J. Prost, in ‘The Physics of Liquid Crystals’, Oxford University
Press, Oxford, UK, 1993.
58 U. Finkenzeller, Kontakte (Darmstadt), 1988,2, 7.
59 G. W. Gray, K. J. Harrison and J. A. Nash, Electron. Lett., 1973,9, 130.
60 P. Chatelain, Bull. SOC. Franc. Mineral Crist., 1954,77, 353.
61 M. Miezowicz, Nature, 1946,158,27.
62 G. H. Heilmeier and L. A. Zanoni, Appl. Phys. Lett., 1968,13,91.
63 G. H. Heilmeier, L. A. Zanoni and L. A. Barton, Proc. IEEE, 1968,56,1162.
64 P. A. Penz, Phys. Rev. Lett., 1970,24,1405.
65 E. F. Carr, Mol. Cryst. Liq. Cryst., 1969,7,253.
66 W. Helfrich, J. Chem. Phys., 1969,51,4092.
67 R. B. Meyer, Phys. Rev. Lett., 1969,22,918.
68 D. Schmidt, M. Schadt and W. Helfrich, Z. Naturforsch., 1972,27a, 277.
69 J. S. Patel and R. B. Meyer, Phys. Rev. Lett., 1987,58, 1538.
70 I. Dozov, M. Nobili and G. Durand, Appl. Phys. Lett., 1997,70,1179.
71 G . P. Bryan-Brown, C. V. Brown, J. C. Jones, E. L. Wood, I. C. Sage, P. Brett and
J. Rudin, Proc. SID ’97Digest, 1997, 37.
72 J. L. Ferguson, SID Digest Tech. Papers., 1985,16,68.
73 D. Coates, J. Mater. Chem., 1995,5,2063.
74 G. P. Crawford, J. W. Doane and S. Zumer, in ‘Handbook of Liquid Crystal
Research’, Eds. P. J. Collings and J. S. Patel, Oxford University Press, Oxford, UK,
1997, p. 347.
75 D. J. Broer, R. A. M. Hikmet and G. Challa, Makromol. Chem., 1989,190,3201.
76 J. Lub, D. J. Broer, R. A. M. Hikmet and K. G. J. Nierop, Liq. Cryst., 1995,18,319.
77 S . M. Kelly, J. Mater. Chem., 1995,5,2047.
78 D.-K. Yang and J. W. Doane, SID ’92 Digest, 1992,759.
79 U. Behrens and H.-S. Kitzerow, Polyrn. Adv. Tech., 1994,5,433.
80 R. A. M. Hikmet, H. M. J. Boots and M. Michielson, Liq. Cryst., 1995,19,65.
81 R. A. M. Hikmet, Adv. Mater., 1992,4,679.
82 P. M. Alt and P. Pleshko, IEEE Trans. Electron. Devices, 1974,21, 146.
83 S. Kobayashi, H. Hori and Y. Tanaka, in ‘Handbook of Liquid Crystal Research’,
Eds. P. J. Collings and J. S. Patel, Oxford University Press, Oxford, UK, 1997.
84 B. J. Lechner, Proc. IEEE, 1971,59,1566.
85 D. E. Castlebury, IEEE Trans. Elec. Dev., 1979, ED-26, 1123.
86 D. Dixon, IEEE Trans. Elec. Dev., 1973, ED-20,995.
87 E. Kaneko in, ‘Handbook of Liquid Crystals’, Eds. D. Demus, J. W. Goodby, G. W.
Gray, H.-W. Spiess and V. Vill, Wiley-VCH Verlagsgesellschaft GmbH, Weinheim,
Germany, Vol. 2A, 1998,230.
88 T. P. Brody, J. T. Asars and G. D. Dixon, IEEE Trans. Elec. Dev., 1973, ED-20,995.
89 J. Cognard, Mol. Cryst. Liq. Cryst., 1982,51, 1.
90 D. Berreman, Phys. Rev., 1972,28, 1683.
91 M. O’Neill and S. M. Kelly, J. Phys. D: Appl. Phys.. 2000,33, R67.
92 M. G. Tomlin, J. Opt. Technol., 1997,64,458
93 D. W. Berreman, Mol. Cryst. Liq. Cryst., 1973,23,215.
94 F. J. Kahn, Appl. Phys. Lett., 1973,22,386.
95 L. Janning, Appl. Phys. Lett., 1972,21, 173.
96 M. Nishikawi, K. Tamas and J. L. West, Jpn. J, Appl. Phys., 1999,38, L334.
97 J-H. Kim, B. R. Acharya and S. Kumar, Appl. Phys. Lett., 1998,73,3372.
98 M. Nishikawi, B. Taheri and J. L. West, Appl. Phys. Lett., 1998,72,2403.
99 B. Park, Y. Jung, H. Choi, H. Hwang, M. Kakimoto and H. Takezoe, Jpn. J. Appl.
Phys., I, 1998,37, 5663.
100 Y. Wang, C. Xu, A. Kanazawa, T. Shiono, T. Ikeda, Y. Matsuki and Y. Takeuchi,
J. Appl. Phys., 1998,84, 181.
101 W. M. Gibbons, P. J. Shannon, S.-T. Sun and B. J. Swetlin, Nature, 1991,351,49.
102 S.-T. Sun, W. M. Gibbons and P. J. Shannon, Liq. Cryst., 1992,12,869.
103 W. M. Gibbons, P. J. Shannon and S.-T. Sun, Mol. Cryst. Liq. Cryst., 1994, 251,
191.
104 K. Ichimura, N. Hamada, S. Kato and I. Shinohara, J. Polym. Sci., Polym. Chem.
Ed., 1983,21, 1551.
105 Y. Imura, J. Kusano, S. Kobayashi and Y. Aoyagi, Jpn. J. Appl. Phys., 1993, 32,
L93.
106 K. Ichimura, Y. Suzuki, T. Seki, A. Hosoki and N. Aoki, Langmuir, 1988,4, 1214.
107 K. Ichimura, in ‘Photochemical Processes in Organized Molecular Systems’, Ed.
K. Honda, Elsevier, Amsterdam, The Netherlands, 1991, p. 343.
108 K. Ichimura, Y. Hayashi, Y. Kawanishi, T. Seki, T. Tamaki and N. Ishizuki,
Langmuir, 1993,9, 857.
109 K. Ichimura, Y. Hayashi and N. Ishizuki, Chem. Lett., 1992, 1063.
110 A. Dyadyusha, V. Kozenkov, T. Marusii, Y. Reznikov, V. Reshetnyak,
A. Khizhnyak, Ukr. Fiz. Zh., 1991,36, 1059.
11 1 M. Schadt, K. Schmitt, V. Kozinkov and V. G. Chigrinov, Jpn. J. Appl. Phys., 1992,
31,2135.
112 K. Ichimura, Y. Akita, K. Akiyama, K. Kudo and Y. Hayashi, Macromolecules,
1997,30,903.
113 G. P. Bryan-Brown and I. C. Sage, Liq. Cryst., 1996,20,825.
114 M. Schadt, H. Seiberle, A. Schuster and S. M. Kelly, Jpn. J. Appl. Phys., 1995, 34,
L764.
115 M. Schadt, H. Seiberle, A. Schuster and S. M. Kelly, Jpn. J. Appl. Phys., 1995, 34,
3240.
116 Y. Imura, S. Kobayashi, T. Hashimoto, T. Sugiyama and K. Katoh, I H C E Trans.
Electron, 1996, E79-C, 1040.
117 R.-P. Herr, F. Herzog and A. Schuster, Patent Application WO 96/10049, 1996.
118 M. Schadt, H. Seiberle and A. Schuster, Nature, 1996,381,212.
44 Chapter 2
119 M. Schadt, Mol. Cryst. Liq. Cryst., 1997,292,235.
120 M. Schadt and H. Seiberle, Proc. SID, 5/4,1997,367.
121 S . M. Kelly, P. Hindmarsh, G. J. Owen, P. 0.Jackson, P. J. Taylor and M. O'Neill,
SID EID '98 Digest, 1998, 5.1.
122 P. Hindmarsh, G. J. Owen, S. M. Kelly, P. 0. Jackson, M. ONeill and
R. Karapinar, Mol. Cryst. Liq. Cryst., 1999,332,439.
123 P. 0. Jackson, R. Karapinar, M. O'Neill, P. Hindmarsh, G. J. Owen and S. M.
Kelly, Proc. SPIE, 1999,3635,48.
124 M. Hasegawa and Y. Taira, IRDC '94 Digest, 1994,213.
125 J. L. West, X. Wang, Y. Ji and J. R.Kelly, SID '95 Digest, 1995,703.
126 C . J. Newsome, M. O'Neill and G. P. Bryan Brown, Proc. SPIE, 1999,3618,132.
127 P. J. BOS,Proc. SID '94, 1994,118.
128 Y . Fujimura, SID Digest, 1992,397.
129 S . Kaneko, Y. Hirai and K. Sumiyoshi, SID Digest '93, 1993,265.
130 H. L. Ong, Japan Display '92, 1992,247.
131 J. P. Eblen, W. J. Gunning, J. Beedy, D. Taber, L. Hale, P. Yeh and M.
Khoshnevisan, SID '94 Digest, 1994,245.
132 A. Shimizu, Asia Display '98, 1998,207.
133 S. Stallinga, J. M. van den Eerenbeemd and J. A. M. M. van Haaren, Jpn. J. Appl.
Phys., 1998,37, 560.
134 H. Mori, Y. Itoh, Y. Nisiura, T. Nakamura and Y. Shinagawa, AM-LCD '96/IDW
'96 Digest, 1996, 189.
135 P. van de Witte, J. van Haaren, J. Tuijtelaars, S. Stallinga and J. Lub, Jpn. J. Appl.
Phys., 1999,38,748.
136 C . Joubert and J.-C. Lehureau, Asia Dispfay '98, 1998, 1119.
137 R. Brinkley, G. Xu, A. Abileah and J. Vanderploeg, SID '98 Digest, 1998,471.
138 A. Lien, Appl. Phys. Lett., 1990,57,2767.
139 M. Jones and G. Xu, SID '98 Digest, 1998,475.
140 K. R. Sarma, H. Franklin, M. Johnson, K. Frost and A. Bernot, SID '89 Digest,
1989, 148.
141 K. H. Yang, IDRC '91 Digest, 1991,68.
CHAPTER 3
Liquid Crystal Displays Using

Nematic Liquid Crystals
1 Introduction
The liquid crystals displays (LCDs) currently manufactured in significant
volume for the wide variety of instruments that incorporate them in order to
display numeric and graphical data all make use of the reorientation of the
optical axis of nematic liquid crystals under the influence of an applied electric
field. The principles of operation of the most important types of LCDs, which
have either been manufactured or are still being manufactured in large numbers,
are illustrated in this chapter. Their capacity to display information, their
intrinsic limitations as well as those imposed by the nematic materials used are
characterised. The development in the design and synthesis of the nematic
materials made use of in the first prototypes of each kind of LCD through to the
components of complex nematic mixtures used today is also described. The
correlation between the molecular structure of these nematic materials and the
magnitude of the physical properties required to fulfil the widely differing
specifications of the various types of LCD is poorly understood. However, an
attempt is made to summarise some of this often empirical knowledge.
Representative examples of the most important classes of nematic liquid
crystals used in LCDs are described.
2 Dynamic Scattering Mode Liquid Crystal Displays

The first practical electro-optical flat panel display utilising any kind of liquid
crystal as the switchable light valve device, was reported by Heilmeier, Zanoni
and Barton at the RCA Laboratories in Princeton, New Jersey in 1968.' In this
particular case the nematic director does not always align itself parallel with the
electric field vector for voltages below a certain threshold voltage for dielectric
coupling, although this voltage can be very high. The anomalous behaviour is
due to the movement under the action of an electric field of a small number of
ions within the nematic liquid crystal, which gives rise to electrohydrodynamic
in~tabilities.~-~It is the anisotropy of conductivity of liquid crystals which leads
45
46 Chapter 3
-I -Fx-
Figure 3.1 Schematic representation of the turbulence in a nematic medium caused by the
movement of ions under the action of an electricjieid.2
to charge segregation and the generation of space charges. The stirring effect of
an electric field on nematic liquid crystals had already been observed in the
1 9 3 0 ~This
. ~ effect leads to the formation of visible domains, known as Williams
domain^,^ at low voltages and light scattering above a higher threshold voltage.'
The interaction of the electric field with the ions gives rise to transverse fields
orthogonal to the applied field if the conductivity is greater parallel to the
nematic director than perpendicular to it.2 This in turn leads to turbulence in the
liquid crystalline medium and produces a wavy structure, see Figure 3.1 .2 This
results in the movement of areas of liquid crystal in alternate directions to
produce Williams domain^.^ The elastic and dielectric torque under steady state
conditions counterbalances the shear produced by this electrohydrodynamic
effect. At higher fields these stripes are replaced by a homogeneous scattering of
incident visible light. Although liquid crystals are generally regarded as
dielectrics, i.e. non-conductive, insulating organic materials, the conductivity
required for the formation of electrohydrodynamic instabilities is relatively low
(lo-' IR- cm- '). However, the high operating voltages and limited lifetime of
many commercial dynamic scattering mode liquid crystal displays (DSM-
LCDs) were limiting factors with regard to their initial acceptance in the
broader electro-optic market. The advent of competing LCD types with an
improved property profile essentially sealed the fate of dynamic scattering mode
(DSM)-LCDs.
Display Configuration
The DSM-LCD is a sandwich cell consisting of two parallel glass substrates
separated (6 pm < d > 25 pm) by spacers (Teflon/Mylar), which usually incor-
porates a nematic liquid crystal of negative dielectric anisotropy aligned either
homeotropically or homogeneously (planar alignment). However, most com-
mercial DSM-LCDs were based on homeotropically aligned cells, see Figure
Liquid Crystal Displays Using Nematic Liquid Crystals 47
UNPOLARISED
mT
E /
\ \ 'SUBSTRATE' //
SORBED \ABSORBANT / REFXECTEDIN

F-STATE BACKGROUND THE ON-STATE
Figure 3.2 Schematic representation of a dynamic scattering mode (DSM) LCD.'
3.2. The substrates are coated with transparent electrodes (nesa or ITO) and
often, but not always, with an appropriate alignment layer, usually a surfactant
such as lecithin. A specularly reflecting mirror with a black background is fixed
to the second substrate. No polarisers are required. Therefore, DSM-LCDs are
intrinsically bright display devices with high contrast.
Off-State
Incident light striking a non-addressed part of a DSM-LCD with homeotropic
alignment is transmitted unchanged by the homeotropically aligned nematic
mixture, which is essentially transparent, and is then absorbed by the black
background. Therefore, the cell appears dark in the non-addressed state (off-
state). If the incident light is said to propagate along the z-axis then the
refractive indices of the homeotropically aligned nematic medium in the x- and
y-axes are equal. Therefore, the ordinary and the extraordinary rays of light are
slowed down equally and no directional optical effect is apparent.
On-State
The application of a voltage with a magnitude above the threshold voltage, e.g.
',
5 V pm- across an activated pixel results in the scattering of incident light at
that pixel. This is dl;e to a circular stirring movement caused by the movement
of positive and negative ions in opposite directions within the liquid crystal layer
upon the action of the applied electric field. Momentum is transferred by
friction to the liquid crystal, which results in turbulence and alignment
48 Chapter 3
discontinuities. The alignment discontinuities which cause changes in the
effective refractive indices of adjacent nematic domains are designated as
electrohydrodynamic instabilities.2 Since the scattering centres are five to 10
times larger than the wavelength of incident visible light, the scattering is
independent of wavelength and the resultant image is uniformly white. Thus,
bright white information is displayed against a dark black background. The
contrast ratio between the on-state and the off-state is often large (100: 1). The
threshold voltage (Vth) is dependent upon a complex combination of elastic
constants, viscosity coefficients (xl and ql), dielectric constants and the
anisotropy of the electrical conduction, which is normally positive (all/al x
4/3),394although independent of the cell gap.3
I
vt
Below this threshold voltage, Williams domains are observed as visible

birefringent stripes. As the voltage is increased further, more light is scattered
and consequently the pixel appears brighter. This gives rise to grey-scale and the
possibility of full colour, although typical operating voltages are relatively high
(20-3OV). The switch on time, ton, is inversely proportional to the current ( I )
and is generally short ( x 1-5ms), even at high operating voltages (50-
lOOV x 5-10Vpm-').
to, cx I - '
On removal of the electric field, the ionic movement decreases dramatically and
the initial alignment is recreated by elastic forces propagated from the surface
boundary layers as well as by conduction. Therefore, switch-off times, t,R, are
an order of magnitude longer ( x 20-100 ms) due to the high degree of disorder
caused by the flow of liquid crystal and the absence of a restoring field effect.
The life-times of such displays depend on the type of applied potential: for AC
operation, values of in excess of 10000 operating hours have been reported,
whereas DC potentials often lead to deposition of ions, with a consequent
decrease, often dramatic, in life-time. The stability of the nematic material to
electrochemical effects is also of critical importance.
Nematic Materials
The DSM effect was found' at the RCA Laboratories using individual Schiff's
bases, see Table 3.1, which had been originally synthesised at the beginning of
Table 3.1 Transition temperatures ("C),dielectric anisotropy (Ae) and conduc-
tivity (Q-' em-') for nematic liquid crystals (1-11) and a commercial
mixture (Merck) designed for DSM-LCDs
Molecular structure Cr N I A& Q Ref

0 83 0 100 0 -3.5 5 x 8
0 116 0 121 0 10
0 108 (0 70) 0 10
0 49 (0 37) 0 11
41 (0 25) 0 -0.4 12
0 32 0 47 0 +0.2 13,14
0 20 0 47 0 15
0 74 0 124* 0 0.0 lo-'' 16
0 14 0 28 0 +0.2 17
il 6
4
0 16 0
Parentheses represent a monotropic transition temperature.

*SmC-N transition at 95 "C.
76 0 +0.2 18
the 20th century.' The first report of the DSM-LCD described dynamic
scattering of incident light at elevated temperatures within the nematic phase
range of N-(4-methoxybenzylidine)-4'-acetoxyaniline(l),8*' see Table 3.1, above
the melting point (83 "C), but below the clearing point (100 "C). However, this
publication soon stimulated attempts to synthesise new nematic materials with a
high clearing point, but also with a much lower melting point in order to enable
50 Chapter 3
the fabrication of commercial DSM-LCDs functional over a wide temperature
range, including room temperature. The molecular structure of these new
chemical entities exhibits a remarkable similarity which is characterised by two
1,4-disubstituted phenyl rings joined by an unsaturated linking unit and with an
alkyl or alkoxy chain at each end of the molecule. Typical examples8-17 (1-11)
reported in 1971 or 1972 are collated in Table 3.1. The combination of a butyl
and a methoxy terminal substituent were found generally to give rise to the
lowest melting point of a wide series of homologues, see also Table 2.1 in
Chapter 2. The nematic phase of these materials is generally of very slightly
negative or positive dielectric anisotropy. However, most of these classes of
compounds possess at least one seriously disadvantageous property for electro-
optical applications: azobenzenes and azoxybenzenes are coloured and undergo
trans-cis isomerisation; SchiFs bases are susceptible to decomposition by
atmospheric moisture; trans-stilbenes undergo photocatalysed isomerisation to
non liquid-crystalline cis-stilbenes which are subject to oxidation by atmo-
spheric oxygen; phenyl benzoates are susceptible to photochemical instability
due to photo-Fries reactions.
The material N-(4-methoxybenzylidine)-4'-butylaniline(MBBA; 7)' was the
first compound synthesised with an enantiotropic nematic phase at room
temperature. Compounds such as MBBA and 4,4'-dibutylazoxybenzene
(9),16*17which also exhibits a nematic phase at room temperature, greatly
facilitated the development of DSM-LCD prototypes as well as a better
understanding of the mechanism of the DSM effect. However, the nematic
clearing point of both compounds is too low for commercial applications.
Therefore, nematic mixtures consisting of several components with a low
melting point and a high clearing point, e.g. Licristal N4, Merck, a (6:4) binary
mixture of 4-butyl-4'-methoxyazoxybenzene (10) and 4-methoxy-4'-butylaz-
oxybenzene (11),18 were then made commercially available, see Table 3.1.
These mixtures exhibited a wide temperature range, high ionic conductivity, a
positive electrical conduction anisotropy and weakly negative dielectric aniso-
tropy. However, the first commercial DSM-LCDs commonly used mixtures
containing ionic impurities, resulting from insufficient purification procedures
or chemical decomposition. This often resulted in an inconsistent electro-optical
performance, electrochemical decomposition of the nematic mixture itself and
finally dielectric breakdown in the cell.19 More reliable displays with longer
lifetimes were produced by the use of improved commercial nematic mixtures
with lower intrinsic conductivity (5 x lo-'' IR-' cm-'> doped with an appro-
priate amount of ionic material in order to produce the higher conductivity
required for satisfactory DSM-LCD performance. The unwanted injection of
additional ions from the electrodes could then be minimised by the use of AC
instead of DC voltages.
Commercial DSM-LCDs generally utilised nematic liquid crystals of negative
dielectric anisotropy. The exact role of the magnitude and sign of the dielectric
anisotropy is somewhat unclear, although many investigations of this relation-
ship were carried out.*' Cells using nematic liquid crystals of positive dielectric
anisotropy doped with ions and with a planar alignment give rise to a limited
amount of dynamic scattering under certain conditions, but the major effect is
dielectric realignment in electric f i e l d ~ . ~ l - ~ ~
3 Cholesteric-Nematic Phase Change Effect (CNPC)

LCDs24-33
The effect of magnetic and electric fields on chiral nematic structure was
calculated24925first and then confirmed by experiment for the magnetic
case.26,27The confirmation of the effect of an electric field on the chiral
nematic state led indirectly to the invention of a novel electro-optical display
device based on the field induced cholesteric-nematic phase change effect, which
was in~ented~'-~' at the Xerox Corporation Research Laboratories, Webster,
New York, USA, at almost the same time as the DSM-LCD and GH-LCD, see
Sections 2 and 7, at the RCA Corporation. Although based on light scattering in
the same way as the DSM-LCD, the cholesteric-nematic phase change LCD is
intrinsically less bright and exhibits a lower contrast ratio than corresponding
DSM-LCDs due to the absorption of light of the first polariser. However, since
the light scattering is caused by reflection from the helix of the chiral nematic
phase as a kind of Bragg reflection and not by the electrohydrodynamic moment
of ions, device lifetimes are intrinsically longer. There are several modifications
of this basic operating p r i n ~ i p l e . ~This
~ - ~device
~ is used most commonly in
conjunction with dichroic dyes in order to improve the contrast, see Section 7.
Display Conjguration
A chiral nematic (cholesteric) liquid crystal mixture is contained in a liquid
crystal cell as shown in Figure 3.3.
Off-State
The helix of a chiral nematic liquid crystal of positive dielectric anisotropy is
aligned parallel to the cell walls by application of a small electric field strong
enough to overcome orientation effects due to anchoring at the cell boundaries.
As the voltage is increased the pitch of the helix becomes larger, as more
molecules are aligned parallel to the electric field. Eventually the helix is
completely unwound and a pseudonematic phase of infinite pitch with the
director parallel to the electric field vector is produced. Thus, application of an
electric field across the cell has effected a phase change from the chiral nematic
state to the nematic state. On removal of the field the optically active molecules
in the chiral nematic mixture regenerate the helix in chiral nematic domains.
However, there is no longer any macroscopic alignment of the helix axis in the
cell and linearly polarised light passing through the first polariser is strongly
scattered. Some of this scattered light is then transmitted through the second
polariser and the display appears white in the off-state after the initial alignment
treatment.
52 Chapter 3
REFucmDIN
THE OFF-STATE ANALYSER STATE
Figure 3.3 Schematic representation of a cholesteric-nematic phase-change eflect

(CNPC)LCD.28-30
On-State
When the helix is completely unwound a clear state is produced as a conse-
quence of the formation of a pseudonematic phase with no helical structure
from the corresponding chiral nematic phase. The analyser then absorbs
linearly polarised light. Therefore, black information can be displayed at the
appropriate pixels against a white background to produce an image with
positive contrast. The threshold voltage for the unwinding of the helix is
dependent on a number of physical parameters:
wherepo is the pitch length at zero field. It is found that relatively large electric
fields (z4-10Vpm-') are required in order to completely unwind the helix.
However, if a pitch is used that is longer than originally utilised, e.g. 3 pm, much
lower operating voltages can be achieved. Moreover, switching times are short,
especially t , ~ e.g.
, 60 ps, since spontaneous formation of the chiral nematic helix
takes place on removal of the electric field.*
Chiral Nematic Materials

The first prototype chiral nematic phase change devices were demonstrated with
a tertiary mixture of cholesteryl chloride (12), nonanoate (13) and oleyl
Table 3.2 Transition temperatures (“C)for chiral nematic derivatives of choles-

terol (12-14) and a tertiary mixture (30%/56%/14% respectively)I
for a CNPC-LCD
X Cr SmA* N* I
12 c1 0 97 - (0 62) 0
13 C8H17C02 0 80.5 (0 78) 0 92 a
14 (Z)-CgH17C=CC7H150C02 50.5 (0 42) (0 48)
Mixture 0 -c 25 - 0 57
carbonate (14). These three optically active derivatives of naturally occurring

cholesterol, see Table 3.2, were of doubtful purity (98-99%), hydroscopic, of
weak dielectric anisotropy and probably highly viscous, although no data were
given. However, sample breakdown was reported above the clearing point,
which is indicative of the presence of substantial amounts of chemical impu-
rities.
Chemically, photochemically and electrochemically stable chiral nematic
mixtures of low viscosity and strong positive dielectric anisotropy are required
in order to produce CNPC-LCDs with fast switching times at low threshold and
operating voltages and acceptable device lifetimes. This was soon achieved by
doping nematic liquid crystals of positive dielectric anisotropy, see Section 5,
with a suitable amount of optically active (chiral) dopant in order to induce the
desired chiral nematic state with an appropriate pitch length. A chiral dopant
with a very large helical twisting power allows small amounts to be used. This
limits the increase in viscosity usually observed on addition of chiral dopants
caused by chain branching at the optically active centre. It also keeps the change
in the other physical properties of the host nematic mixture to a minimum.
4 Electrically Controlled Birefringence (DAP/HN/

ECB) LCDs34-39
The many different kinds of L C D S ~ ~based ~ ’ on electrically controlled
birefringence (ECB) are variants of the Frkedericksz effect first reported in the
1 9 3 0 ~These
. ~ related display types based on the same electro-optical effect, but
a slightly different display configuration are denominated by a multitude of
names. However, current versions of this type of display are now often referred
54 Chapter 3
to as deformation of vertically aligned phase DAP-LCDs or VAN-LCDs. They

are nearly always combined with active matrix addressing.
The first LCDs based on electrically controlled birefringence reported by
Schiekel and Fahrenschon of the Siemens Corporation in Germany were simple
alpha-numeric displays with direct addressing of each However, these
early prototypes were soon superseded by ECB-LCDs with passive, multiplexed
addressing for higher-information-content applications. Unfortunately, they
were not particularly suited for this kind of application, since they generally
exhibit low brightness (z20%), relatively low contrast (25: I), limited grey-scale
and the observed colour varies strongly with viewing angle. This combination
prohibited the large-scale commercialisation of ECB-LCDs with multiplex
addressing. However, the advent of active matrix addressing moderated some
of these disadvantages and ECB-LCDs with active matrix addressing are
starting to appear on the market due to their capacity to generate full colour
using grey-scale and relatively rapid response times compared to those of TN-
LCDs with active matrix addressing. The development of stable nematic
mixtures of negative dielectric anisotropy, high birefringence and low viscosity
over recent years has made a significant contribution to the increasing
atractivity of this type of LCD.
Display Conjgura t ion

A nematic liquid crystal of negative dielectric anisotropy is aligned with the
director aligned orthogonal to the cell wa_llsby means of a surfactant orienta-
tion layer, see Figure 3.4. One or two linear, elliptical or circular polarisers are
Figure 3.4 Schematic representation of an electrically controlled birefringence (ECB)

LCD.34-39
required depending on the particular configuration of the device and whether
the cell is operated in the transmissive or reflective mode.39
Off-State
If the cell is positioned between crossed polarisers then, when in the off-state,
polarised light produced by the first polariser travels through the homeotropi-
cally aligned nematic medium unchanged, is absorbed by the analyser and the
cell appears dark, see Figure 3.4. There is no effective birefringence (An,s = 0)
since ne = no, see Chapter 2.
On-State
The application of a voltage above the threshold voltage, Vth,
where E~ - clI < 0, causes the nematic director to tilt away from the normal due
to coupling of the electric field with the induced dielectric a n i ~ o t r o p y This
.~~
induces an effective birefringence, AneK, which is proportional to the magnitude
of the applied field up to the maximum value of An = ne - no, see Chapter 1.
Therefore, a proportion of the polarised light traverses the cell in the on-state.
The proportion increases with increasing magnitude of the induced effective
birefringence. However, since the optical effect is produced by interference
between the extraordinary and the ordinary rays, white incident light is usually
converted into coloured light due to interference. The optical retardation, 6, is
proportional to the induced effective birefringen~e:~~
where, d is the cell gap, when operated in the transmission mode for a given
wavelength of light A. The refractive index of the ordinary ray (no) passing
through the cell is not affected by the applied field. However, the magnitude of
the refractive index of the extraordinary ray (n,) increases with voltage due to
the dielectric coupling of the nematic director with the field, if the liquid crystal
mixture is of negative dielectric anisotropy. Thus, the effective birefringence
increases with applied electric field strength. In theory, this should allow the
production of coloured displays by manipulation of the degree of optical
retardation. However, in practise there is virtually no change in colour at
voltages near the threshold voltages. Therefore, high-information-content
ECB-LCDs with full colour are usually operated at voltages near the threshold
voltage in combination with standard RGB colour filters.
56 Chapter 3
The intensity, Z, of light traversing the cell also depends on the optical
retardation (phase difference), 6,as well as the angle of incidence of light (4):
where I. is the intensity of the incident plane polarised light. The intensity of
transmitted light is a maximum when
h d = (2m + 1)TR
where m is an integer. Usually, Equation 8 is satisfied in the operating voltage
region with m = 0 and R z 550nm, which corresponds to the central wave-
length of the visible spectrum and the wavelength of maximum sensitivity of the
human eye. These conditions result in a black-and-white display with good
uniform transmission across the visible spectrum. Since the induced birefrin-
gence depends on the strength of the electric field, the amount of light
transmitted can be modulated in order to produce tuneable birefringence, i.e.
grey-scale. Since the magnitude of the displacement of the director is not large
for an optical effect to be observed, the switch-on times of ECB-LCDs are very
short (ton = 2 0 ~ s ) .However, such short switch-on times are exhibited at
relatively high operating voltages. The switch-off time, t0R, is a more complex
parameter and depends on a viscosity coefficient, q, the cell gap, d and a
curvature elastic modulus, K : ~ ~
Therefore, switch-off times are independent of the field strength and directly
dependent on material parameters, such as viscosity coefficients and elastic
constants, and the cell configuration. Therefore, they are often three or four
orders of magnitude larger than the switch-on times. However, sophisticated
addressing techniques can produce much shorter combined response times
(lo,., + ton = 50ps).43*44
The nematic director should be inclined, e.g. 1" pretilt,
in one specific direction in the zenithal axis in order to produce a uniform
optical appearance which, however, may reduce the sharpness of the threshold
v01tage.4~
Nematic Materials of Negative Dielectric Anisotropy

High-information-content ECB-LCDs require a sharp threshold voltage, low
operating voltages and a steep electro-optical contrast curve.46 Therefore, a
high negative value of A& and high positive values of k33/kll, An and 6 are
required. However, high values of An, which allow thin cells and, therefore,
shorter response times, see Equation 9, also give rise to a residual birefringence
in the off-state, due to the finite pretilt, and interference colours. This is
responsible for the low contrast observed for thin ECB-LCDs. Unfortunately,
a large cell gap generates long response times according to Equation 9. A E / ~
should also be low, i.e. both and ql should be high. However if A E / is~ too
low, high operating voltages, which are already relatively high for operation
with standard, low-cost CMOS drivers, are the result.
The first ECB-LCDs used nematic Schiffs bases of negative dielectric
anisotropy, such as N-(4methoxybenzylidene)-4'-butylaniline (7), see Table
3.1, or mixtures of homologues of this class of compound, which generally
exhibit a nematic phase, a high clearing point and a melting point at or below
room temperature. However, the dielectric anisotropy of these Schiff s bases is
low (0 > AE > - 1) due to the absence of any strong dipole moments orthogo-
nal to the molecular axis. Consequently, the operating voltages of prototypes
ECB-LCDS~ were relatively high (= 7-8 V). The dependence of the threshold
voltage on the elastic constants of these compounds could not be determined,
since they were not known at the time.
Subsequent LCD types using electrically controlled birefringence, such as
homeotropic nematic (HN-LCD)39 or electrically controlled birefringence
(ECB-LCDS)?~~~' used liquid crystal mixtures with a greater magnitude of the
dielectric anisotropy with a negative sign. These mixtures incorporated compo-
nents, such as compound (15) shown in Table 3.3, possessing a cyano group
(CN) in a lateral p0sition.4~In this way the vector component of the dipole
moment (4 D) of cyano group perpendicular to the long molecular axis was
greater than that parallel to the long molecular axis.47-55However, substantial
steric effects due to the large van der Waals volume of the cyano group induces
substantial clearing point depressions as well as a high viscosity, while the
induced negative dielectric anisotropy is only moderate (-2 > AE > -4).49
This value is too low for practical ECB-LCDs, since it results in high operating
voltages (see Equation 5).
Materials incorporating two cyano groups in adjacent lateral positions,S2p5s
e.g. compound 16;' exhibit a much greater negative value for the dielectric
anisotropy (- 12 > A& > -20); see Table 3.3. This is due to the large resultant
dipole moment (= 7 D) orthogonal to the molecular long axis of the molecule,
since the dipole vectors parallel to the long molecular axis cancel each other out.
The viscosity does not increase proportionately for dicyano-substitutedderiva-
tives compared to that of corresponding mono-cyano-substituted derivative,
although the resulting viscosity is very high. The clearing point may be even
higher than that of the corresponding mono-cyano-substituted material. This
may well be due to the shielding effect of the first nitrile group, since the
molecular rotation volume of the mono- and dicyano-substituted compounds
are almost identical. Hence, the degree of molecular separation due to steric
effectsof the lateral-substituent is very similar.
In the case of two-ring compounds the steric effects due to the presence of
large lateral substituents induce a very low clearing point and a very high
viscosity. The corresponding three-ring materials are also viscous. However, the
presence of another ring often leads to observable mesomorphism in three-ring
58 Chapter 3
Table 3.3 Molecular structures, transition temperatures ("C) and dielectric
anisotropy (A&measured at 20°C)for nematic liquid crystals (15-27)
Molecular structure Cr Sm N I Ae Ref
15 0 59 - 113 -3.8 47
16 0 138 - 148 -19.0 48
17 0 106 - (a 102) 0 -11.5 52-55
18 0 88 0 154" 169 0 50
19 0 230 - - 0 50
20 0 224 - - 0 56
21 0 63 (0 43) 79.5 0 57
22 0 71.5 0 162 0 173 0 58
23 59.5 - (0 0) 0 -6.0 59-61
24 0 25 0 30b 0 66 -8.0 62,63
25 0 74 0 86b 171 -4.1 64
26 0 112 (0 105)b 0 190 a -5.3 65
27 cn - 135 -5.1 66
a SmA. SmB.
compounds incorporating the nitrile function in a lateral position, e.g. com-

pound 15,47 due to a higher degree of anisotropy in the molecular length/
breadth ratio and polarisability. Nematic mixtures of high negative dielectric
anisotropy for LCD applications were made available in the late 1970s by the
Chisso Corporation of Japan. Liquid crystalline ester derivatives of 4-n-alkoxy-
2,3-dicyanophenol, such as compound 16, were used to induce a negative value
of the dielectric a n i ~ o t r o p y .Unfortunately,
~~ these mixtures exhibited an
unacceptable degree of photochemical instability, which severely limited the
lifetime of ECB-LCDs incorporating them. The high viscosity of these mixtures
also gave rise to long response times, see Equation 9. The replacement of the
terminal oxygen atom of 4-n-alkoxy-2,3-dicyanophenolby a methylene (CH2)
group to produce analogous 4-n-alkyl-2,3-dicyanophenyl ester derivatives, such
as compound 17, generated nematic materials of strong negative dielectric
anisotropy which were sufficiently stable for commercial applications.52953
However, mixtures incorporating these substances were still very viscous with
long response times in ECB-LCDs. Attempts to produce lower viscosity
compounds by incorporating two cyclohexane rings, e.g. compound 18,50or
replacing the carboxy function by an ether link, e.g. compound 19,50or by an
ethyl group, e.g. compound 20,56created almost completely insoluble, non-
mesomorphic compounds with very high melting points. The incorporation of
only one cyano group in non-ester, three-ring compounds via sp2 bonding on
aromatic rings (21)57and sp3 bond angles in central linkages or terminal chains
(22)58generated materials with nematic and smectic phases, but unfortunately
with low negative values of dielectric anisotropy.
A diverse range of liquid crystalline 3,6-disubstituted pyridazines, such as
compound 23,whose lone-pair electrons situated at the electronegative nitrogen
atoms exhibit a large resultant dipole moment perpendicular to the molecular
long axis, see Figure 3.5, have also been prepared.5g41 The lone-pair electrons
do not increase the breadth of the molecule. Therefore, due to the absence of
steric effects associated with lateral substituents, two-ring 3,6-disubstituted
pyridazine derivatives exhibit nematic phases of high negative dielectric aniso-
tropy (- 6 > Ae > - 10) and much lower viscosity than dicyano-substituted
compounds with a dielectric anisotropy of similar magnitude. Unfortunately,
they also suffer from the serious disadvantages for commercial ECB-LCDs of
photochemical instability and significant dark-current conductivity.
The trans,trans-4,4'-dialkyl-
1,l '-dicyclohexyl-4'-carboni triles, e.g. compound
24, incorporate only one cyano group like the mono-nitriles (15,21 and 22).
However, the sp3 bonding and the axial position of this nitrile function on one of
the carbon atoms of one of the cyclohexane rings results in the dipole moment
being almost exactly perpendicular to the molecular long axis, see Figure
3.6.62*63The axial hydrogen atoms on the same side of the cyclohexane ring
shield the axial cyano group to some degree. Consequently, these chemically,
thermally and photochemically stable materials exhibit an enantiotropic
nematic phase over a wide temperature range. The viscosity of the bicyclo-
hexyl-nitriles is much lower than that of the aromatic, monocyano-substituted
Figure 3.5 Schematic representation of the resultant transverse dipole moment across the
molecular long axis of a 3,6-disubstitutedpyridazinederivative.
60 Chapter 3
PI
I PIES = PI
Figure 3.6 Schematic representation of the resultant transverse dipole moment across the
molecular long axis of an axially 4'-Jluoro-substituted or 4'-cyano-substituted
trans,trans4,4'-dialkyl-I,I'-dicyclohexanederivative.
compounds. Unfortunately, the very low birefringence of axially substituted

bicyclohexanes renders them unsuitable as ECB-LCDs. However, related
derivatives incorporating two benzene rings and one axially cyano-substituted
cyclohexane ring exhibit a birefringence of sufficient magnitude compatiblewith
ECB-LCD applications.
2,3-Difluorobenzene derivatives such as compound 25,@ see Table 3.3,
embody the best combination of physical properties of nematogens of negative
dielectric anisotropy and low viscosity for ECB-LCDs. The small van der Waals
A)
volume of the fluorine atom (5.8 and its relatively small dipole moment (1.47
D) lead to materials of relatively low viscosity and weak to moderately strong
negative dielectric anisotropy in the nematic phase. In diether derivatives of 23-
difluoroquinone the electronegative fluorine atoms polarise the additional n-
electron density provided by the conjugated lone-pair electrons on the oxygen
atoms. The large resultant dipole moment is perpendicular to the molecular
long axis. Thus, relatively large values of the dielectric anisotropy
(-4 > A& > -6) are observed. A similar effect is observed for the fluoro-
substituted pyridines, such as compound 26, which also exhibit a strongly
negative dielectric anisotropy due to the resultant dipole moment orthogonal to
the long molecular axis.65 However, these synthetically more complex com-
pounds appear to exhibit comparable properties to those of the analogous 2,3-
difluorophenyl derivatives. Some attempts to improve the properties of 2,3-
difluorophenyl derivatives by synthesising compounds with different central
linkages, e.g. compound 27, were relatively successful.66 A range of diether
derivatives of 2,3-difluoroquinone are manufactured commercially (E. Merck)
for a several types of LCDs, which require nematic mixtures of negative
dielectric anisotropy, especially ECB-LCDs.
5 Twisted Nematic L C D S ~ ~
The twisted nematic liquid crystal display (TN-LCD) was reported by Schadt
and Helfrich of F. Hoffman-La Roche in Basle, Switzerland in 1970.67This was
part of a tripartite collaboration between F. Hoffman-La Roche in Basle,
Brown Boveri of Baden and Ebauche in Neuchatel, all in Switzerland. The
intention was to design and develop flat panel displays, e.g. for digital watches.
The first LCD factory was constructed in Lenzburg, Switzerland in the mid-
1970s by Videlec, a subsidiary of Brown Boveri. Since then the TN-LCD has
established itself as the dominant electro-optical display by far in portable

instruments and is slowly gaining market share in the computer monitor
market. TN-LCDs with direct or multiplexed addressing are to be found in
low-information-content electro-optical devices such as digital clocks, watches,
computer games and calculators. TN-LCDs with active-matrix addressing are
steadily displacing multiplexed STN-LCDs, see Section 6, from the high-
information-content displays market, e.g. in notebooks, personal organisers or
laptop computers. The dominance of the TN-LCD with different modes of
addressing is due to an advantageous combination of properties for a given
application selected from low power consumption, low threshold and operating
voltages, acceptable contrast (at least for narrow viewing angles), relatively
efficient multiplexing (ateleast at low ratios, e.g. a 1:32 duty cycle), a long
operating lifetime and exceptionally low cost. The main drawbacks of the TN-
LCD are long response times, especially at low temperatures, poor brightness,
limited contrast and a large variation of contrast with viewing angle, especially
in applications with even a moderate degree of multiplexed addressing.
A standard TN-LCD consists of a nematic liquid crystal mixture of positive
dielectric anisotropy contained in a cell with an alignment layer on both
substrate surfaces, usually rubbed polyimide, crossed polarisers and a cell gap
of 5-10pm, see Figure 3.7. The nematic director is aligned parallel to the
direction of rubbing in the azimuthal plane of the device. The alignment layer
induces a small pretilt angle (0 M 1-3") of the director in the zenithal plane. The
WOLARISISED UNPOLARISED
RISED LIGHT
NSMITIED
+-G
SUBSTRATE
/ LIGICI'ABSORBED
INmE ON-STATE
Figure 3.7 Schematic representation of a twisted nematic liquid crystal displaf7 (TN-
LCD).169
62 Chapter 3
direction of alignment at the upper glass substrate is orthogonal to that at the
lower glass substrate, i.e. there is an overall twist of 90" in the director from the
top to the bottom of the display maintained by elastic forces in the off-state. The
polarisation axes of the polarisers are usually arranged parallel to the director of
the nematic medium at each substrate surface. Occasionally the polarisers are
fixed parallel to one another in order to generate a dark off-state and a bright
on-state with a wide viewing angle cone.
Off-State
The plane of polarised light generated by passage through the first polariser is
rotated by 90" by the twisted nematic medium due to a wave-guiding action as
long as the wavelength of light (380-780nm) transmitted is much less than the
propagation depth (d z 5-10pm) of the nematic layer.68 This waveguiding
action allows the plane polarised light to pass through the second polariser
(analyser), since the plane of linear polarised light and the transmission axis of
the analyser are now parallel. Consequently, the off-state, or transition-state, is
white and is often referred to as the white- or open-mode. In the reflective TN-
LCD configuration, transmitted, wave-guided light is reflected back along the
same pathway by a reflective surface.
In order to suppress the formation of elliptically polarised light rather than
plane polarised light in a TN-LCD, it is necessary to control the cell thickness, d,
and the birefringence of the nematic medium, An, with respect to the wavelength
of light in a vacuum, 1,according to the equation:
TN-LCDs are often constructed with as small a cell gap as practically possible,
since the response time is proportional to d 2 . The first commercial TN-LCDs
operated with large cell gaps (d !z 8 pm) due to an inability to accurately control
the cell gap with an acceptable tolerance below this critical value. However, the
developments in spacer technology over the last 25 years now allow much
smaller cell gaps ( d !z 5 pm) to be produced uniformly and reproducibly.
However, an undesirable consequence of smaller cell gaps is that transmitted
light has a significant component of elliptically polarised light caused by the
tight pitch of the twisted nematic medium (p = 0.25 x 4. Thus, the desired
white appearance of TN-LCDs can be contaminated with undesired interfer-
ence colours, especially for non-orthogonal viewing angles, due to interference
between the extraordinary and ordinary rays. However, interference at the
analyser can inhibit the production of elliptically polarised light to a significant
degree for a limited range of wavelengths of light, according to the equation
2dAn
u=-
3,
An=0.05 0.10 0.17 0.22
Figure3.8 Gooch and Tarry plot of transmission versus the coeficient, u, and the
corresponding birefringence values, An,for a Twisted Nematic Liquid Crystal
Display TN-LCD of thickness 6.55 pm operating between parallel polari-
sers!9.169
where the coefficient u must be larger than unity, which is the so-called Maugin
limit.68Thus, for the centre of the spectral range of visible light (A z 550nm),
the wavelength at which the eye is most sensitive, a series of maxima and minima
in transmission are produced for certain values of u (i.e. J3, 415, J35, 463,
etc.), see Figure 3.8. These correspond to the first, second, etc., Maugin, or
Gooch and Tarry, Therefore, for a TN-LCD assuming ideal
parallel polarisers and neglecting light absorbed by the first polariser, the
transmission:
I cx sin2
rhere the amplitude of transmission decreases with increasing vali e of u.

Therefore, TN-LCDs should be operated in the first or second minimum in
order to optimise brightness, contrast and the dependence of the contrast on
viewing angle. The minima become maxima and vice versa for crossed
polarisers. A TN-LCD operated in the first minimum with a moderately small
cell gap (d = 6.55pm) and a nematic liquid crystal mixture of a low birefrin-
gence (An < 0.10) exhibits a bright appearance, see Figure 3X70 A nematic
mixture of high positive dielectric anisotropy and low viscosity is used to
achieve fast response times. Higher values of the birefringence at constant cell
gap correspond to higher Gooch and Tarry minima with a consequent lower
transmission in the off-state. A certain amount of light still leaks through all
TN-LCDs in the off-state due to non-ideal polarisers, non-uniform nematic
alignment, cell thickness variation, temperature dependence and dispersity of
An and the polychromaticity of light. A combination of all of these factors
results in a lower contrast between the on- and off-state than calculations would
suggest.
64 Chapter 3
On-State
Upon application of a voltage above the threshold voltage, Vth, in a TN-LCD
the molecules in the centre of the cell and, therefore, the optical axis, which is
coincident with the nematic director, start to tilt under the influence of the
electric field. They realign themselves in the direction of maximum polarisation,
i.e. parallel to the electric field, if the nematic phase is of positive dielectric
anisotropy. This realignment of the optical axis results in a decrease in the
effective average birefringence. A very high voltage would lead to a tilt angle of
90" at the centre of the cell, where the twist would be completely unwound. The
effective birefringence is still very low at moderately high applied voltages and
the plane of the polarised light is no longer rotated. The plane polarised light
strikes the analyser in the crossed polariser configuration and is absorbed. A
TN-LCD in this configuration exhibits positive contrast, i.e. displays black
information on a white background. The threshold voltage for a TN-LCD is
defined
Vth ="[&hl + k33-4 2k22)11'2

where kl 1, k22 and k33 are the Frank elasticconstants, EO is the dielectric constant
of a vacuum and A& is the dielectric anisotropy of the nematic medium. The
switch-on time, to, is defined by:67
ton
?la
a AEE- un2
In the absence of a distorting applied field, surface and elastic forces regenerate
the original twist in the nematic director and plane polarised light is once again
transmitted. Several layers of the nematic medium several molecules thick do
not realign at all under the action of an electric field due to strong anchoring
with the orientation layer. Although these residual layers of bend and splay at
the cell surfaces lower the contrast, they transmit the anchoring of the director
at the substrate surfaces and are responsible for the regeneration of the twisted
nematic configuration in the absence of an applied electric field. Therefore, the
switch-off time, t0R, is defined as:67
where
.=[kilt( k33 -4 2k22 )]

and q is the rotational viscosity, 71, and E is the applied field. The response of
the director to the applied electric field is clearly not identical throughout the
cell and depends on the strength of the anchoring at the substrate surfaces.
Thus, the director at the centre of the cell, i.e. furthest from the substrate
surfaces, responds first.
The contrast of a TN-LCD is highest parallel to the optical axis in the middle
of the cell and lowest at angles normal to this direction. Therefore, the contrast
of a TN-LCD is not symmetrical and depends not only on the viewing angle, but
also on the direction of view with respect to the director, see Chapter 2. In the
absence of a chiral dopant, there are also two degenerate, but optically non-
equivalent, directions of 90" twist of the director in the off-state (+ 90" and
-270"). This gives rise to domains of opposite (reverse) twist upon removal of
the applied electric field. Disclination lines between adjacent domains of
opposite twist reduce the contrast. This can be eliminated by the addition of a
small amount of an optically active compound (chiral dopant),72 which induces
a uniform direction, or handedness of twist, in the nematic medium (+ 90').
However, if the ratio of the pitch and cell gap (p/d) is too small then areas of
reverse twist can still be formed. Therefore, the pitch, p , induced in the nematic
medium by the chiral dopant should be long corresponding to a relatively large
ratio of the chiral nematic pitch to the cell gap (p/d = 7). An upper limit of the
pitch is set by the fact that too large values for the pitch will fail to inhibit the
formation of reverse-twist domains. An unintended consequence of the addition
of a chiral dopant is that the medium is also no longer a twisted nematic phase,
but a chiral twisted nematic phase, which exhibits a higher threshold voltage:
where the pitch of the chiral nematic mixture represents an additional term.67
Initially derivatives of cholesterol, such as cholesteryl nonanoate (13),see Table
3.273 were added to induce the desired handedness of twist. However, ester
derivatives of cholesterol are hygroscopic, relatively unstable and induce a high
nematic viscosity at low concentration. Therefore, they were soon replaced by
optically active versions of the components of the nematic mixture, e.g. 4-
cyano-4'-[( S)-2-methylbutyl]biphenyl .74p75
Non-ideal birefringence effects are enhanced during grey-scale operation, e.g.
for full-colour displays, since the optical axis of the nematic layer makes an
oblique angle to the normal of the cell and the applied electric field. This results
in a pronounced asymmetric viewing angle dependence of contrast, luminance
(brightness) and chromaticity, due to interference effects, and even to inversion
of contrast at large viewing angles. Contrast ratios above 1O:l are limited for a
typical TN-LCD to a maximum viewing angle of k45"in the horizontal plane
and + 60" and - 15" in the vertical plane, see Chapter 2. This is clearly
unacceptable for large area displays, e.g. for computer monitors and portable
computers.
66 Chapter 3
The large viewing angle dependency of TN-LCDs attributable to the positive
birefringence of the nematic layer can be compensated to a large degree by using
one or more optical retarders of positive or negative birefringence in combina-
tion with the normally white (NW) mode, i.e. with crossed polarisers, see
Chapter 2. The NW mode of TN-LCDs is used in combination with active
matrix addressing to generate near video-rate addressing for high-information-
content, full-colour applications. The suppression of unwanted leakage of light
in the on-state leads to higher contrast at wider viewing angles and suppresses
grey-scale inversion. The asymmetric twist of the nematic medium is compen-
sated for by aligning the optical axis of the two compensation layers on either
side of the cell at right angles to each other, which results in no retardation at
normal incidence. Therefore, optical compensation layers can significantly
improve the contrast and the viewing-angle dependence of contrast in the
vertical direction as well as in the horizontal direction for non normal viewing
angles, while maintaining the good optical properties at normal viewing.
However, these layers of organic material also absorb a proportion of the
transmitted light and, therefore, they contribute to the intrinsically low bright-
ness of LCDs. A higher initial intensity of incident light leads to a higher power
consumption. This is undesirable for portable applications due to the shorter
time between battery recharging times.
Nematic Materials of Positive Dielectric Anisotropy

Over the last 30years since the invention of the TN-LCD67 there has been a
gradual, but steady increase in the size, complexity and information-content of
TN-LCDs, represented in the number of pixels. A full-colour TN-LCD with
active matrix addressing for a portable computer or computer monitor may
contain three million pixels or more. The changes in the methods of electronic
addressing, from direct addressing of individual pixels in watches and simple
calculators, to multiplex addressing used in complex calculators, personal
digital assistants and cellular telephones, to active matrix addressing essential
for portable computers and computer monitors, resulted in a continual need to
synthesise new classes of nematic liquid crystals to create nematic mixtures with
the desired spectrum of physical properties. The major developments in the
synthesis of nematic materials for TN-LCDs with the three different types of
addressing are described below.
Nematic Materials for Direct Addressing

The operating principle of the TN-LCD was first d e m o n ~ t r a t e dat~a~ tempera-
ture above 100°C between the melting and clearing point of N-(4-ethoxybenzy-
lidene)-4-aminobenzonitrile (28), see Table 3.4. This Schiff s base was first
prepared at the beginning of the 20th century in Vorlander’s research group in
Halle, Germany.76 It was synthesised again later and was found to exhibit a
positive dielectric anisotropy (An z + 14).77978The researchers at Roche then
prepared a tertiary mixture of two 4-n-alkoxybenzylidene-4-aminobenzoni-
Table 3.4 Transition temperatures ("C)and dielectric anisotropy for the aro-
matic nitriles (28-40)
Molecular structure Cr N I A& Ref
28 0 106 0 128 + 14 77,78
29 0 63 0 111 79
30 0 55 0 101 79
31 0 54 0 98 80
32 0 66 (0 41) 80
33 0 46 0 55 80
34 0 65.5 0 77.5 0 80
35 0 33 0 64.5 0 80
36 a 22.5 35 0 +8.5 82
37 C5H1- 1 CN 0 33.5 0 43.5 0 83
38 0 73 [o 311 0 +9.4 84
39 0 71 (0 52) 0 +21 85
40 0 96 0 109" 0 +3.4 85
Parentheses represent a monotropic transition temperature; brackets represent a virtual, extrapo-

lated transition temperature; a SmA phase at 93.5"C.
68 Chapter 3
t r i l e ~(29
~ ~and 30) and 4-heptanoylbenzylidene-4-aminobenzonitrile80 (31).
This simple nematic mixture of high birefringence and strong positive dielectric
anisotropy enabled prototype TN-LCDs to be operated and evaluated at room
temperature.*' Mixtures of liquid crystalline compounds usually exhibit a much
lower melting point than that of the individual components of the mixture,
whereas the clearing point is usually an average, although some deviation from
ideal behaviour is nearly always observed. Mixtures with the lowest melting
point using the components chosen are referred to as eutectic mixtures. The
composition of these mixtures can be predicted with a good degree of accuracy,
if the melting point and the enthalpy of fusion of the individual components are
known. The binary mixtures consisting of the esters (32and 33) and the Schiff s
bases (34and 35) are also both nematic at room temperature.81 However, the
threshold voltage of both of these mixtures is much lower than that of the
tertiary mixture of 29,30 and 31. This is a direct consequence of a favourable
combination of elastic constants (see Equation 13) as the dielectric anisotropy
of all three mixtures is almost identical and the threshold voltage is independent
of the cell gap. Furthermore, the shorter switching times observed for the binary
mixture of the Schiffs bases (34 and 35) is also due to a lower rotational
viscosity, y1, of these compounds compared to that of the esters 32 and 33 (see
Equations 14 and 15) since the value of the elastic constant ratio K of each of the
binary mixtures is very similar. The viscosity of the binary mixture of alkyl-
substituted SchifYs bases (34 and 35) is clearly lower than that of the tertiary
mixture of the corresponding alkoxy-substituted or alkanoyloxy-substituted
Schiff s bases (29,30and 31).
The first commercially available TN-LCDs were manufactured in the mid-
1970s using more sophisticated mixtures of Schiff s bases and phenyl benzoate
esters supplied by F. Hoffmann-La Roche. The Schiff s bases had to be handled
with some care due to the hydrolysis of the central linkage by atmospheric
moisture. Researchers in the research group of G. W. Gray at the University of
Hull in the United Kingdom realised in the early 1970s that the chemical,
photochemical or electrochemical instability encountered for many liquid
crystals known at that time (see Table 2.1) could be directly attributed to the
conjugated, unsaturated central linkage between the aromatic rings. Therefore,
they designed and synthesised the class of compounds known usually as the
cyanobiphenyls, see Table 3.4 for a representative example of this important
class of liquid crystalline compounds, i.e. 4-cyano-4'-pentylbiphenyl(36),which
is nematic at room temperature. The 4-n-alkyl-4'-cyanobiphenylsdo not
contain a central linkage at all between the two phenyl rings.82 It is not
surprising, therefore, that they were found to be sufficiently chemically,
photochemically and electrochemically stable for use in TN-LCDs with direct
addressing. Furthermore, and perhaps critically, the absence of the central
linkage induced a lower viscosity and melting point than those of the corre-
sponding aromatic cyanophenyl benzoate esters and cyano-substituted Schiff s
bases available at that time. As a consequence of this advantageous combina-
tion of physical properties, mixtures such as E7 and E8 (BDH Chemicals, now a
subsidiary E. Merck, Darmstadt, Germany) of 4-n-alkyl-4'-cyanobiphenyls,4-
n-alkoxy-4'-cyanobiphenyls and 4-n-alkyl4-cyano-p-terphenyls gradually dis-
placed the mixtures of Schiffs bases and phenyl benzoate esters from most
commercial TN-LCDs.
Nematic mixtures composed of these new classes of materials were used
successfully in simple TN-LCDs, e.g. digital watches, clocks and calculators,
where the information content is limited and each pixel can be driven directly
with a dedicated electrode contact and individual driver, see Chapter 2. With
this type of addressing the off-voltage is zero and the on-voltage may be several
times larger than the threshold voltage. Therefore, in a TN-LCD, where the
electro-optical characteristic is relatively flat, a good contrast can be attained as
well as low power consumption. Nematic materials for this kind of application
should, in the first instance, be chemically, electrochemically and photochemi-
cally stable. Furthermore, they should exhibit a low melting point, a high
clearing point, a wide nematic phase, no smectic mesomorphism, especially no
ordered smectic phases such as smectic B or E, a low viscosity, a high
birefringence and a high positive value of the dielectric anisotropy. The
compounds described below represent attempts to design and synthesise new
classes of nematic compounds with an optimal combination of these physical
properties for TN-LCDs.
The classes of nematic compounds represented in the Tables 3.4 and 3.5 by
their pentyl homologues (36-59)82-'01 in order to facilitate comparison, are
essentially modifications of the 4-n-alkyl-4'-cyanobiphenyls.82The dielectric
anisotropy of the 2-(4-cyanophenyl)-5-pentylpyridine (37),83 2-cyano-5-(4-pen-
tylpheny1)pyridine (38)84and the 2-(4-cyanophenyl)-5-pentylpyrimidine (39)85
is higher than that of the 4-cyano4-pentylbiphenyl (36) without a nitrogen
heterocycle. The higher value of the dielectric anisotropy of the heterocyclic
compounds (37-39) is due to the resultant dipole moments associated with the
lone pair electrons on the nitrogen atoms being directed along the long
molecular axis, see Table 3.4. If the dipole moments are directed in the opposite
direction, e.g. in the 5-cyano-2-(4-pentylphenyl)pyrimidines(40),85the magni-
tude of A& is correspondingly lower. The 2-(4-cyanophenyl)-5-n-alkylpyrimi-
dines85 were synthesised on a commercial scale as components of nematic
mixtures for TN-LCDs for many years by F. Hoffmann-La Roche.
The chemically, photochemically and electrochemically stable class of com-
pounds referred to generally as the PCHs (pentylcyclohexanes) is exemplified by
the trans- 1-(4-cyanophenyl)-4-pentylcyclohexane (41), see Table 3.5.87 This
compound possesses a trans-1,4-disubstituted cyclohexane ring in place of the
phenyl ring attached to the alkyl group of the structurally related 4-cyano-4'-
pentylbiphenyl (36). However, PCH-5 (41) exhibits a lower birefringence, a
lower viscosity and a comparable value of the dielectric anisotropy to those of
the corresponding cyanobiphenyl(36). This particularly advantageous combi-
nation of physical properties inspired the synthesis of a wide range of related
materials, e.g. compounds (41-53),8c96 with one alicyclic ring bearing a
terminal alkyl group and one aromatic ring bearing a cyano group. Several
classes of the nitriles prepared, such as homologues of the dioxane (49)93have
also been manufactured as components of nematic mixtures for TN-LCDs. In
70 Chapter 3
Table 3.5 Transition temperatures ("C) and some values for the dielectric
anisotropy of the nitriles (41-59)
MoIecular structure Cr N I A& Ref
41 0 31 55 0 + 9.9a 86
42 0 45.5 55.5 0 87
43 0 48 0 88
44 0 84 0 89
45 0 70 0 89
46 0 40 0 90
47 0 48 0 91
48 0 35 0 92
49 0 56 0 +llb 93
50 0 48 0 94
51 0 74 0 95
52 0 98 0 95
53 0 62 0 + 10" 96
54 0 < 25 a 97
55 0 113 0 98
56 0 62 0 +3-4 99
(continued)
Table 3.5 continued.

Molecular structure Cr N I A& Ref
59 104 0 129 0 98
Parentheses represents a monotropic transition temperature; * SmB phase at 94°C; BEmeasured at:
a0.98 x TNI; b42"C;'25°C.
the case of the dioxanes this is partly due to the large value of the dielectric
anisotropy and low birefringence. The high A& is due in part to the dipole
moments associated with the electronegative oxygen atoms. The high dielectric
anisotropy leads to lower threshold and operating voltages. The other classes of
aromatic nitriles collated in Table 3.5 exhibit at least one unfavourable
property, which inhibited their manufacture, such as the photochemical
instability of conjugated cyclohexenes, e.g. compound 47, low resistivity values
due to ionic impurities'of the borinane (50) or high viscosity, e.g. the
bicyclo[2,2,2]octane (53).
If the cyano group is attached to a non-conjugated alicyclic ring, as in the
aliphatic nitriles (54-59) also collated in Table 3.5Y7-'O' then the dipole moment
of the cyano group is much lower than of an analogous benzonitrile due to the
absence of aromatic conjugation with the cyano group. Consequently, the value
of the dielectric anisotropy is lower for aliphatic nitriles compared to that of the
related conjugated benzonitriles also shown in Table 3.5. However, the trans-1-
(trans-4-cyanocyclohexyl)-4-pentylcyclohexane(56) with two cyclohexane rings
still possesses a moderately high value for the dielectric anisotropy in spite of the
absence of polarisable aromatic n - e l e c t r ~ n sSince
. ~ ~ the nematic clearing point
is high and the birefringence and viscosity values are low, homologues of the
aliphatic nitrile (56), often referred to as CCH-5, are also used in nematic
mixtures (E. Merck). They are used in particular as important components of
nematic mixtures of low An for TN-LCDs with a moderate cell gap
( d z 6.55pm) operated in the first minimum,70 see Figure 3.8. The other
aliphatic nitriles collated in Table 3.5 generally exhibit an unattractive combi-
nation of a high melting point, a low nematic clearing point and a high viscosity.
The trans-1-n-alkyl-4-(4-cyanophenyl)cyclohexanes86exhibited an unparal-
leled advantageous combination of physical properties for TN-LCDs at this
time. Therefore, a central linkage was incorporated between the cyclohexane
and the phenyl ring of PCH-5 (41) to form the structurally related compounds
(60-69) collated in Table 3.6 in attempts to improve on these properties.86i'02-'06
However, only the trans- 1-n-alkyl-4-[2-(4-cyanophenyl)ethyl]cyclohexanes
(PECHs), e.g. PECH-5 (62),'04 were almost equivalent from the point of view of
72 Chapter 3
Table 3.6 Transition temperatures ("C)
for the benzonitriles (41 and 6 M 9 )
60
-s 0 55 a 81 0 102
61
41
-0 .;r 0 74
30 59
103
86
62 45 55 104
63 74 49) 104
64 0 54 107 0 105
65 a 41 73 0 105
66 W 51 39) 106
67 W 40 39) 106
68 W 63 54) 106
69 W 64 43) 0 106
Parentheses represent a monotropic transition temperature; 'monotropic SmX-SmB transition at
display devices. None of these materials are currently being manufactured, apart
from the ester (a), which was prepared before the PCHs. lo2
Variations in the nature of the terminal chains was then also investigated, see
Table 3.7, where a series of trans-4-substituted cyclohexylbenzonitriles (41and
70-85)bearing a chain of fixed length (five units) is s h o ~ n . * ~ *lo~Of~ the
~ - series
'
of ethers (70-73)differing only in the position of the oxygen atom, only the
methoxypropyl-substituted benzonitrile (73) with a non-conjugated oxygen
atom at some distance from the core exhibits a nematic phase.lo7However, the
melting point is higher than that of the corresponding trans4(kyanophenyl)-
4pentylcyclohexane (41). Therefore, the temperature range of the nematic
phase is much narrower. The angle made by the carbon-oxygen-carbon bond
(CH20CH2)is very similar to that with a methylene unit (CH2) in place of the
oxygen atom (CH2CH2CH2). Therefore, it may be assumed that the effect on
the clearing point of the oxygen atom in a non-conjugated position in a terminal
chain of a particular compound is of a polar and not a steric nature.
Significant variations in the magnitude of the physical properties, such as the
liquid crystal transition temperatures, viscosity, birefringence and the elastic
and dielectric constants, of alkenyl-substituted compounds with a carbon-
for the cornpoundr (41 and 70-85)

Table 3.7 Transition temperatures ("C)
R Cr N r Ref
41 w 0 30 55 0 86
70 we 0 60 - 0 107
0 23 - 0 107
0 42 - 0 107
0 52 55 0 107
74 (4 ---Lq-- 0 16 59 0 108
75 (a w 0 - - 1441 0 108
76 (E) \=/ 0 16 -671 0 108
0 -8 -541 0 108
0 60 74 0 108
0 33 - 14) 0 108
0 30 10) 0 108
81 \b 0 86 1.51 0 107
0 84 0 109
0 66 59) 0 109
0 56 - 0 110
85 (9 0 128 119) 0 110

0

Brackets represent a virtual (extrapolated) transition temperature.
carbon double bond (C+C) in the terminal chain, as exemplified by the trans-4-
alkenylcyclohexyl-4'-benzonitriles (74-80), have generally been found, depend-
ing on the position and configuration of the carbon-carbon double bond."*
This discovery was of significant importance for the successful development of
74 Chapter 3
nematic mixtures for super twisted nematic liquid crystal displays (STN-LCDs),
see Section 6.' "-' l 3 The results of molecular modelling of ensembles of
molecules suggest that steric effects due to the different configuration and,
therefore, conformation, of the terminal chain of these alkenyl-substituted
compounds are primarily responsible for the very large differences in their
'
transition temperatures and other physical properties.' The corresponding
acetylene (81) derivati~e''~exhibits a virtual nematic clearing point just above
0°C. This may be due to the non-axial position of the acetylene group, which
would serve to increase the breadth of the molecule, thereby reducing the
effective length-to-breadt h ratio.
The unexpected and non-linear effect of combining two functional groups, i.e.
one exerting polar, the other steric effects, on the mesomorphic behaviour of
compounds'09 is illustrated to some extent by the transition temperatures of the
ether isomers (82 and 83) differing in the position of the oxygen atom and the
carbon-carbon double bond, see Table 3.7. The ether (82) is not liquid crystal-
line, whereas the other ether (83) exhibits a monotropic nematic phase at a
temperature higher than that of trans-1 -(4-cyanophenyl)-4-pentylcyclohexane
(41). The ester (84)'" is also not liquid crystalline. However, the related
compound (85) containing an ester group (COO) and an additional carbon-
carbon double bond exhibits a very high nematic clearing point, especially for a
'*'
compound with only two six-membered rings. This is probably a function of
an advantageous combination of molecular rigidity and extended conjugation in
spite of the presence of the flexible trans-1,4-disubstituted cyclohexane ring. * l4
Nematic Materials for Multiplex Addressing

The requirement for large-area, flat panel displays for high-information-content
portable instruments, such as computers, personal organisers and notebooks,
could only be met at the time by TN-LCDs with more sophisticated addressing
schemes than direct addressing due to the high cost of the many drivers required
and the physical impossibility of contacting each and every pixel directly. The
way chosen initially to resolve this problem was to adopt TN-LCDs with
multiplex addressing, see Chapter 2.
Multiplex addressing involves a series of rows of electrodes on one substrate
surface at right angles to a similar series of electrode columns on the second
substrate surface, see Chapter 2. The overlapping rows and columns create a
pattern of rectangular pixels. A combination of a voltage applied to a row with a
voltage applied to a column electrode crossing at a given pixel will activate that
pixel, if the two voltages are in phase and they create a voltage larger than the
threshold voltage of the nematic mixture in the cell formed by the substrates.
There is an upper limit to the number of addressable lines (N) with acceptable
contrast in a TN-LCD with multiplex addressing given by the equation:93
where V, and V,, are the select and non-select voltages, respectively. Therefore,
as the number of lines to be addressed increases, the more the ratio of the select
voltage and the non-select voltages approaches unity and the smaller the
acceptable difference between them becomes. Even for a relatively modest
degree of multiplexing, e.g. for a high-information calculator with 64 addres-
sable lines, there is only a small difference ( x 11%) between the select and non-
select voltages. The contrast also decreases with increasing multiplex rates due
to the unintentional activation of adjacent pixels (cross-talk), partly due to
fringing field effects. Therefore, the need to address a large number of lines with
an acceptable contrast in a TN-LCD with multiplex addressing imposes severe
limitations on a nematic mixture compatible with these requirements. It is
essential that in the electro-optical response curve, transmission against applied
voltage is very steep, i.e. a small increase in voltage just above the threshold
voltage should lead to a large increase in the transmission. Unfortunately, a
consequence of a steep electro-optical characteristic is that the capacity for grey-
scale and, therefore, full colour with an array of RGB filters, is severely reduced,
since a small voltage change results in a large change in transmission. Therefore,
TN-LCDs with multiplex addressing are only capable of displaying a limited
amount of black-on-white information with acceptable contrast and brightness.
In practical devices this sets an upper limit for the duty cycle of 1:64, i.e. only 64
rows can be addressed in one frame time. A duty cycle of 1:32 is common for
large, sophisticated calculators.
Nematic mixtures designed for use in TN-LCDs with multiplex addressing
had to satisfy a stringent set of specifications including a high Ae for a low
threshold voltage, see Equation 13, a low viscosity for short response times, see
Equation 14, and a combination of a low k33/k11,high k33/k22 and low A&/&*for
a steep electro-optical response curve. A series of nematic mixtures with these
initial properties and a range of values for the birefringence, An, were required
for displays with different cell gaps driven in the first or second transmission
minimum, see Equations 11 and 12. Furthermore, since the transmission curve
also depends on temperature, a nematic mixture with a low rate of change in the
threshold voltage with temperature (dV/dT < 0.01 V'C-') would eliminate the
need to compensate for this change electronically. However, in practice the
variation of operating voltages with temperature is often compensated for
electronically or by using an optically active dopant whose pitch decreases with
increasing temperature. Therefore, the physical properties of existing nematic
materials were re-evaluated to assess their suitability for TN-LCDs with multi-
plex addressing. However, the design and synthesis of new classes of nematic
liquid crystals for use in these applications was required in order to optimise the
performance of multiplexed TN-LCDs with moderately high-information-
content.
The nematic liquid crystals collated in Tables 3.4-3.7 exhibit strong antipar-
allel molecular correlation, '"-'' probably due to dipole-dipole interactions of
neighbouring cyano groups attached to the phenyl ring which have a large
dipole moment (4.0 D).'15-'19 Thus, the nematic phase formed by rod-like
compounds with a strongly polar group in a terminal position can be envisaged
76 Chapter 3
as being comprised of a binary mixture of uncorrelated molecules (monomers)
and pairs of associated molccules (dimer~)''~ in a dynamic, temperature-
dependent equilibrium. The effective resultant dipole moment and dielectric
anisotropy of the molecular dimers will be much lower than those of the non-
correlated monomers due to the antiparallel correlation of the terminal dipoles
in the effective molecular dimer. Consequently, the effective molecular length
(1.4 x I) of the molecular dimer is greater than the molecular length (I) of the
''
monomer for the 4-n-alkyl-4-cyanobiphenyls.' This larger anisotropy in the
effective length-to-breadth ratio of the molecular dimer than that of unasso-
ciated molecules may well be a decisive factor in the formation of a nematic
phase at or just above room temperature for the relatively short 4-n-alkyl-4'-
cyanobiphenyl molecules. The degree of the molecular association of polar
nematic liquid crystals, described by the Kirkwood Froehlich factor g, can be
calculated from macroscopic physical data, such as the dielectric constants,
refractive index and molecular dipole moments, determined for the bulk
nematic There is no molecular correlation when g = 1 and
there is complete correlation of the molecular dipoles when g = 0. Thus,
knowledge of the Kirkwood Froehlich factor can provide important informa-
tion on some bulk physical properties of a nematic phase of relevance for the
design of new materials and the formulation of nematic mixtures for TN-LCDs
with multiplex addressing.
It was soon found that standard nematic mixtures used in TN-LCDs with
direct addressing, such as E7 (E. Merck, Darmstadt, Germany), consisting
completely of polar compounds with a cyano group in a terminal position, i.e. 4-
n-alkyl-4'-cyanobiphenyls, 4-n-alkoxy4-cyanobiphenyls and 4-n-alkyl-4"-
cyano-p-terphenyls, could not be multiplexed to any significant degree.
However, it was found'20 that mixtures of the nitriles (36-69) shown in Tables
3.4-3.7 and nematic compounds of low dielectric anisotropy, exemplified by the
pentyl homologues of the esters (86-95)49*'21-'26shown in Table 3.8, consist of a
much lower proportion of associated molecular dimers of polar compounds
than that found in mixtures consisting purely of nitriles. The higher proportion
of non-associated polar compounds not only gives rise to a high value for the
dielectric anisotropy and, therefore, a lower threshold voltage, but also a low
value of k33/k11. Thus, it was found that mixtures of apolar and polar nematic
liquid crystals were much more suited to TN-LCDs with multiplexed addressing
than mixtures consisting only of polar liquid crystals with terminal substituents
with large dipole moments, e.g. the cyano group.
The aromatic phenyl benzoate esters (86-88) exhibit a high-viscosity nematic
phase with either a low nematic clearing point or a narrow temperature
range. 121-123 The phenyl cyclohexyl esters (89-91) possess a nematic phase with
a relatively wide temperature range and an acceptable vis~osity.'~'*'~~ Although
the ester (92) with two cyclohexane rings only exhibits an enantiotropic smectic
B phase, homologues with shorter chains also possess a nematic phase with a
relatively high clearing point, low viscosity and a very low birefringence. 124 The
bicyclo[2.2.2]octane esters (93-95) exhibit a nematic phase with a high clearing
point, a low k33/kli ratio, but unfortunately also a high v i s ~ o s i t y . ~ ~ * The
'~~,'*~
Table 3.8 Transition temperatures ("C) for the esters (8695)
Molecular structure Cr SmB N I Ref
36 - (0 26) 121
87 0 50 - 0 58 0 122
0 2 - 0 5 0 123
0 36 (0 29) 0 48 0 121
0 49 - 0 81 0 121
0 18 - 37 0 123
92 c 5 H 1 1 5 1~
1 0 ~ 51 c0 71 - 1 24
0 31 - 0 64.5 0 49,125
0 42 - 100 0 125
0 26 - 0 65 0 49,126
presence of a fluorine atom in a lateral position in the esters (88,91 and 95) leads
to a lower melting point, sometimes below room temperature, and the elimina-
tion of smectic phases. However, the viscosity of these fluoro-substituted esters
is considerably higher than that of the corresponding non-fluoro-substituted
esters (86, 89 and 93). Therefore, homologues of the 4-n-alkylphenyl trans-4-n-
alkylcyclohexanoates and the corresponding 4-n-alkoxyphenyl trans-4-n-alkyl-
78 Chapter 3
cyclohexanoates'21 were used in combination with nitriles, such as the cyanobi-
phenyls, to create nematic mixtures for multiplexed TN-LCDs. However, these
mixtures exhibited unacceptably long response times, even in low-information-
content TN-LCDs, due to the high viscosity of the mixtures, see Equations 14
and 15. Therefore, in attempts to produce apolar nematogens of lower viscosity,
a series of new materials (96--105)'27-'36incorporating one or two trans-1,4-
disubstituted cyclohexane rings with a variety of different central linkages, were
synthesised, see Table 3.9.
Most of the compounds (96-105)'27-'36 collated in Table 3.9 exhibit a
nematic phase as well as an unwanted smectic B phase. However, only the
compounds linked directly, e.g. (96-99, 102 and 103),12891293'34 or linked by the
ethane linkage, e.g. (104 and 105),'35were found to exhibit the desired low
values of viscosity. Apolar two-ring compounds such as (96), which are not
mesomorphic at room temperature but exhibit a very low viscosity, and the
corresponding three- and four-ring apolar materials, such as compounds (97-
99), with very high nematic clearing points are used in commercial mixtures with
Table 3.9 Transition temperatures ("C)for the apolar compounds (96-105)
Molecular structure Cr SmB N I Ref
96 c 3 H 7 w C2HS
0 1 - [0-70] 128,129
- -
97 C d l , *c541 0 13 0 164 0 170 0 128,129
98 CS", 1 0 37 - 0 117 0 128,129
99 C6Hl, 0 45 0 275* 0 305 128,129
100
101
102 0 64 0 82t - 132
103 0 49 - 0 50 0 133,134
104 0 52 0 72 - 135
CSHll
105 0 23 0 135 0 143 0 136
CSHll
Brackets represent an extrapolated transition temperature; *several ordered smectic phases present.
polar compounds to form multiplexable nematic mixtures with the desired
spectrum of physical properties.'2g9129The presence of a fluorine atom in a
lateral position in apolar materials, such as (98), leads to the suppression of
smectic phases and a more pronounced resultant tendency to form the nematic
phase due to steric The suppression of the highly viscous
smectic B phase by mixture formulation is usually difficult, especially at low
temperatures. Therefore, compounds such as homologues of the cyclohexane
ester (100) with longer chains, which only possess a smectic B phase, are not
very suitable components of nematic mixtures with a wide nematic temperature
range. The presence of a smectic B phase at high temperatures for such
compounds can also leads to a limited solubility at low temperatures in
nematic mixtures. However, homologues of the cyclohexane ester (100) with
shorter chains were used for some time in nematic mixtures of low viscosity and
low birefringen~e.'~~ The ether (101)13'was synthesised in attempts to lower the
viscosity even further. However, although this compound exhibits a nematic
phase, it is unfortunately of high viscosity. Homologues with longer terminal
chains exhibit predominantly smectic B phases. 13' The trans,trans-4-n-alkyl-4'-
n-alkylbicyclohexane (102) with no oxygen atom only exhibits smectic B phase,
although it was synthesised in the hope that a nematic phase would exhibit a
lower viscosity than that of the corresponding ester (100) and ether (101). The
corresponding trans,trans-4-n-alkyl-4'-n-alkoxybicyclohexane(103) with an
additional oxygen atom exhibits an enantiotropic nematic phase. The two ring
ether (104) and the three-ring compound (105) exhibit nematic phases of a
viscosity of a somewhat higher magnitude of analogous compounds without an
ethyl linking unit.'359136However, they can be used to create nematic mixtures
with short response times. Homologues with longer alkyl chains are purely
smectic. The compounds 96-99 and 103-105 are representatives of important
classes of commercial nematic liquid crystals manufactured for use in nematic
mixtures for TN-LCDs with multiplex addressing.
Aromatic liquid crystals, such as compounds 106-109 collated in Table 3.10,
have been investigated to a much lesser extent since they were perceived to be
intrinsically more viscous than the analogous cyclohexane derivatives. This is
indeed often the case. It was hoped that the incorporation of a fluorine atom in a
lateral position of a polar compound, such as the ester (106)'37,would create a
nematic material, such as the esters (107 and 108),'17 of positive dielectric
anisotropy and low ratio of k33/kll and A E / E ~Although
. this was partially
achieved, see the relatively low value for the elastic constant ratio k33/k11
obtained for the ester (107), this does not improve the multiplexability of
mixtures containing such components. Furthermore, the viscosity of these fully
aromatic esters is very high. However, fully aromatic apolar compounds, such
'
as 5-n-alkyl-2(4-n-alkoxyphenyl)pyrimidines,e.g. ( m ) , 389 39 were found to
possess very low k33/kll elastic constant ratios, a high birefringence and a low
viscosity. Mixed with polar nematics, such as the compounds shown in Tables
3.4-3.7, e.g. homologues of compound (41), and related three-ring materials,
fully aromatic apolar compounds are important components of nematic
mixtures of high birefringence for TN-LCDs with multiplexed addressing.
00
0
Table 3.10 Transition temperatures ("C),elastic constant ratio* (k33/kll), birefringence* (An), dipole moment ( p D ) , Kirkwood-
Froehlichfactor ( 8 ) and dielectric anisotropy (AE)for the compounds (106-109)
~~
Molecular structure Cr SmA N I k33/klI All g h Ref
44 - 0 57 0 1.52 0.15 5.6 0.7 19.9 117
107 47 0 54 0 1.41 4.8 0.8 9.8 117
108 c 7 H 1 5 w 4 0 28 0 28.5 0 1.71" 0.14" 6.1 1.0 48.9 117

\ / CN
109 CsH1l < m 0 c 8 H 1 7 0 34 0 56 0 64 0 1.0 1.0 xo 139

~~
*Measured at 0.95 x TN- 1.

Nematic Materials for Active Matrix Addressing

The low contrast and brightness, long response times as well as the strong
viewing angle dependency of TN-LCDs with even a moderate degree of multi-
plexing renders them unsuitable for applications where fast, high-information-
content displays are required, such as portable computers and computer
monitors. These applications presently use STN-LCDs with a high degree of
multiplexed addressing, see Section 6, or TN-LCDs with active matrix addres-
sing, see Chapter 2. TN-LCDs with active-matrix addressing exhibit a relatively
high contrast ratio, minimal cross-ta1k;grey-scale, fast response times and can
be fabricated over a relatively large-area (12"-14" diagonal). They also suffer
from either low luminosity or high power consumption, which are a direct
consequence of the low intrinsic brightness of the TN-LCD. The cost of the
silicon substrate and low production yield due to pixel damage and subsequent
repair led initially to high production costs, which retarded the market
acceptance of active-matrix TN-LCDs. However, the price differential with
comparative STN-LCDs has been reduced substantially by the development of
improved production equipment and processes. The market share of TN-LCDs
with active matrix addressing of the portable computer displays market is
dominant ( > 90%).
Most TN-LCDs with active matrix addressing use a silicon substrate with
thin-film transistors (TFT) as the active drive element at each pixel. The TFT in
a TFT-TN-LCD maintains the charge on the pixel capacitor until the pixel is
addressed again in the following frame, i.e. it is a memory effect dependent upon
the maintenance of the pixel capacitance over time. The loss of charge from the
pixel effectively decreases the voltage applied across the pixel, which degrades
the contrast. Therefore, it is essential that the resistivity of the nematic liquid
crystal mixture should be as constant as possible over time. This requires
nematic mixtures with an initial resistivity (a = 10'2-10'4 l2 cm) several orders
of magnitude higher than that of commercial nematic mixtures for standard
TN-LCDs with direct or multiplex addressing. The holding ratio, HR, is an
important factor representing the change in the initial resistivity of the liquid
crystal mixture over time:
where T is the frame time and z is the time constant for the pixel and storage
capacitor.
The requirement of a very high and constant resistivity over time of nematic
mixtures for TN-LCDs with active matrix addressing meant that new liquid
crystals which met this and the other specifications, such as low viscosity and
high positive dielectric anisotropy, were needed. The nitriles used in nematic
mixtures for TN-LCDs with direct or multiplex addressing were soon found to
be unsuitable for TN-LCDs with active matrix addressing. The polar nature of
the cyano group leads to the solvation of ions from some of the layers on the
82 Chapter 3
LCD substrate. The presence of these mobile ions reduces the resistivity and,
consequently, the holding ratio. Therefore, liquid crystals containing substitu-
ents with a much smaller permanent dipole moment than that of a cyano group,
e.g. halogen atoms, especially fluorine atoms, were synthesised, see Tables 3.1 1
and 3.12, for some representative examples (11&128)'4s'54 of nematogens
synthesised for TN-LCDs with active matrix addressing, although very many
related derivatives have been reported.889lo6, log*129y1 55-157 The corresponding
two-ring analogues, which are usually non-mesomorphic, to the three-ring
compounds collated in Tables 3.1 1 and 3.12 are also used in the same mixtures.
The compounds 111-113 with a halogen atom in a terminal position, collated
in Table 3.11, exhibit a moderately high value of A&,which increases with the
polarisability of the halogen-carbon bond.14* The largest value of A&exhibited
by the halogenated compounds, i.e. 113,is almost half that of the corresponding
nitrile (114).14' The viscosity also increases with the size of the terminal halogen
atom. 140 Therefore, taking these factors into account, terminally fluorinated
compounds appeared to offer the best combination of a moderately high
positive value of A&and a moderately low viscosity. However, the low value of
the dielectric anisotropy of mono-fluorinated nematic liquid crystals, e.g.
compound 111in Table 3.1 1, resulted in unacceptable high threshold voltages.
Significantly higher values of dielectric anisotropy were achieved for com-
pounds incorporating terminal groups containing several fluorine atoms
bonded to the same carbon atom, e.g. compounds 115-118, incorporating
trifluoromethoxy, trifluoromethyl or difluoromethoxy groups, see Table
3.12.142-147 The nematic clearing point and A&generally increase with increasing
polarity of the terminal substituent, e.g. the trifluoromethyl-substituted com-
pound (117) exhibits a larger A& value than that of the corresponding nitrile
(114). A very large A&is observed for the sulfonyl fluoride (118) due to the high
polarisability of the sulfonyl group. Therefore, these and related compounds are
found in commercial nematic mixtures for TN-LCDs with active matrix
addressing.
Compounds containing two or more fluorine atoms spread around the
molecule can exhibit high values of the dielectric anisotropy if the resultant
dipole moments of the carbon-fluorine bonds are additive, c.J the dielectric
data for the poly-fluorinated compounds (119-124) in Table 3. 12.1433'48-'51
However, the viscosity is not proportional to the number of fluorine atoms in a
lateral position, since the molecular rotation volume is not increased by the
presence of each fluorine atom. However, the nematic clearing point does
decrease with increasing number of lateral fluorine atoms, e.g. compounds 123
and 124 with four and five fluorine atoms in terminal and lateral positions,
respectively, are not mesomorphic. However, such poly-fluorinated compounds
are used in small amounts as components of nematic mixtures in order to
increase the positive value of the dielectric anisotropy. Related compounds,
which combine a terminal group containing halogen atoms with additional
fluorine atoms in lateral positions, e.g. compounds 125 and 126 also listed in
Table 3.12, can also exhibit values of A&comparable to that of the analogous
'
aromatic nitriles. 52 The presence of other electronegative heteroatoms, such as
n
Table3.11 Transition temperatures ("C), C-X bond length (A), dipole moment of C-X bond (Debye) birefringence (20"C),
dielectric anisotropy (extrapolated to 100% at 20"C),and viscosity (cP at 20°C) for the compounds (110-114)
Molecular structure Cr SmA N I C-X p An A& q Ref @

n
- 0 82 0 1.10 ZO 0.15 ZO 22.7 140 s

g'
111 0 125 0 1.39 1.47 0.17 +4.9 25.1 140 2.
112 "" 0 100 0 158 0 1.78 1.57 0.22 +7.5 46.6 140
5s
2;
n
0 163 0 1.93 1.59 0.24 +7.6 63.0 140
114 c5H \ / \ / CN 0 79 0 87 0 184 0 1.14 4.0 0.23 +13 141
w
00
84 Chapter 3
Table 3.12 Transition temperatures (“C) and dielectric anisotropy (AEmeasured
at 20°C; *measuredat T N - I - IOOC) for compounds (115-128)
Molecular structure Cr SmB N I Ae Ref
0 43 0 128 0 147 0 +9 143
0 123 - 124 +13 145
0 67 0 120 0 162 0 +9 144
0 156 - [O 601 +27.1 147
0 90 - 158 +7.3 142
0 46 - 0 124 0 +9.3 143
0 48 0 85 0 159 0 +3.2* 151
122 C& -#+QF

0 42 - (0 33) + 12.6 49
F
f IF
0 6 4 - - 0 +15.2 50
F
124 C #\ / \ / I 0 ~123 - - 0 +20.5 150

F F
125 w 7 = (\ 3
/ &
\ /- OCHzCF3 0 84 - 105 + 12.2 152
F
F
126 c 3 H 7 m 7 4 0 52 - 0 104 0 +12.5 152
F F
0 124 - 0 156 0 +17.7 153
0 99 - 0 157 0 +20.8* 154
~~~ ~
Parentheses represent a monotropic transition temperature; brackets represent an extrapolated

‘virtual’ transition temperature; *Sma-N transition at
oxygen and nitrogen, incorporated in aromatic or aliphatic rings can also give
rise to compounds of strong positive dielectric anisotropy, e.g. the pyrimidine
(127)lS3and the dioxane (l28).lS4
Many of the compounds of strong positive dielectric anisotropy shown in
Tables 3.1 1 and 3.12 exhibit very high stable resistivity values and solvate ions
to a much lesser extent than the analogous nitriles. Therefore, nematic mixtures
containing them often exhibit high stable holding ratios as well as low threshold
voltages and viscosity. Therefore, due to this advantageous combination of
physical properties, many of the polar poly-fluorinated compounds collated in
the Tables 3.1 1 and 3.12 are used in commercial mixtures for TN-LCDs with
active addressing.
6 Super Twisted Nematic LCDs158-173

It was clear in the early 1980s that the optical performance of highly multiplexed
TN-LCDs was completely inadequate, if high-information-content displays for
portable instruments, such as notebooks, personal digital assistants and
portable computers, were to be realised. TN-LCDs with more than a limited
amount of multiplex addressing (1:32 or 1:64), e.g. in the first handheld
televisions and portable computers, exhibited poor contrast, low brightness
and a very strong angle dependency of the viewing angle. Therefore, research
was focused on alternative types of LCD. Fortunately, it was found that a range
of super twisted nematic (STN) LCDs invented in the mid-1980s had the
potential to display large amounts of information with a satisfactory appear-
ance and ~ p e e d . " ~ -This
' ~ ~ was due to a very steep increase in the transmission
for a small change in voltage above the threshold voltage of LCDs with a highly
twisted nematic structure. This allows a very high degree of multiplexing and,
therefore, a high information content, according to the formalism of Alt and
P l e ~ h k o , see
' ~ ~Chapter 2. The term STN-LCD is often used to describe the
class of LCDs with a highly twisted director configuration as well as one
particular type of STN-LCD. Competing technologies based on TN-LCDs with
active-matrix addressing, see Section 5, and ferroelectric liquid crystals, which
are not discussed here due to space considerations, were still in a development
stage.
The steepness of the electro-optical contrast curve of applied voltage against
optical transmission increases dramatically for twist angles, @, > 90" up to
values of = 180". Hysteresis and even bistability can be observed for more
highly twisted structures (180" < @ > 360"). However, a stripe texture can also
disrupt the appearance of the display under certain condition^.'^^ Fortunately,
the stripe textures can be suppressed under well defined conditions and bistable
displays with very steep electro-optical transmission curves can be produced.
The steepness of the electro-optical transmission curve increases with increasing
twist angle until a value with infinite slope is found for @ = 270°,see Figure 3.9.
This allows the very high degree of multiplexing essential for high-information-
content displays. The first commercial STN-LCDs were addressed using RMS
voltages and standard Alt and P l e ~ h k odriving
' ~ ~ schemes. However, although
86 Chapter 3
90
60
4MAX
(DEGREES)
30
0
0.5 1.o 1.5 2.0 2.5
VOLTAGE (VOLTS)
Figure 3.9 A plot of the twist angle at the centre o a cell against applied voltage for a
series of twist angles of nematic LCDs.16
such displays are adequate for high-information-content applications such as

notebooks, personal computers, calculators, etc., the relatively long response
times of STN-LCDs generate a low contrast at the video addressing rate ( < 50
ms) required for television screens or computer monitors. In order to address
this problem the so-called Active AddressingTMof STN-LCDs was devel-
oped. 175 This electronic drive scheme can produce significantly shorter response
times than standard addressing schemes. Therefore, an acceptable contrast for
displays at an addressing rate approaching video rate can be realised. STN-
LCDs were manufactured in volume shortly after their invention due to the
requirement for acceptable flat panel displays with low operating voltages and
power consumption for portable instruments, such as calculators and laptop
computers. The market share of STN-LCDs in the high-information-content
part of the FPD market is being eroded by TN-LCDs with active addressing, see
Section 5. However, STN-LCDs are, in turn, steadily displacing TN-LCDs with
multiplexed addressing from medium and low-information-content applica-
tions, such as calculators, games and personal organisers. STN-LCDs dominate
the displays for the personal telephone market.
The first super twisted nematic STN-LCD was reported by researchers from
Bell Laboratories, Murray Hill, New Jersey, USA. 158-160 These prototypes and
the computer programs used to simulate their optical performance contributed
to the discovery of the other commercially successful STN-LCDs, although
LCDs based on this concept were not manufactured due to several disadvanta-
geous features described below. A nematic mixture of positive dielectric
anisotropy containing a small amount of chiral dopant is homogeneously
aligned with a very high pretilt angle (0 5 45") in a glass cell with a large gap
(d < 20 pm) between crossed polarisers. The front polariser is aligned parallel to
the nematic director in the off-state, which is tilted at a high angle from the
substrate surface at both substrates, whereby the molecules of the chiral nematic
phase are essentially parallel from the top to the bottom of the cell. There is no
twisted structure in the cell. There is extinction between crossed polarisers and
the non-addressed parts of the display appear black. If a short pulse of voltage,
e.g. 2 V, is applied above the threshold voltage, the addressed pixel reverts to the
original off-state orientation. However, if a higher voltage pulse, e.g. 3V, is
applied the nematic director twists through 360" at various angles to the cell
normal to form a second stable state, the on-state. In this orientation two
elliptically polarised modes are formed and some light traverses both the
polariser and analyser. Thus, a bistable nematic LCD is produced with negative
contrast, i.e. a bright image against a black background. The chiral dopant
increases the energy barriers between the non-twisted and fully twisted state.
Other metastable states, which are also possible, can be eliminated by an
appropriate combination of device configuration, addressing scheme and
properties of the chiral nematic mixture. There are no visible disclinations in
the activated areas.
In principle, this mode of operation creates the possibility of producing high-
information-content displays due to the short frame times associated with
bistable displays, since they are basically a memory effect and only new
information must be changed. Unfortunately, metastable twist states of inter-
mediate twist, which degrade the optical performance of the device, form
around dust particles in cells with a cell gap below a certain value (d < 20pm).
Therefore, since the response time is proportional to d 2 , very long response
times are observed (z1 s) for LCDs with a cell gap above this critical value.
These optically disruptive metastable twist states also form at the interface with
spacers used to generate a uniform cell gap.
A super twisted nematic guest-host (STN-GH) LCD containing a dichroic
dye in a nematic mixture with a high degree of twist of the nematic director from
the bottom to the top of the cell was then found to exhibit a comparably steep
electro-optical response curve, but without the formation of undesirable twist
states, see also Section 7 for guest-host L C D S . ' ~ ' - 'The
~ ~ dichroic dye was used
to generate the electro-optical effect and no polarisers were used. The amount of
light absorption by the dye depends on the direction of its polarisation axis,
which is parallel to the LC director in this type of STN-LCD. In the off-state
half the light is transmitted, whereas both the elliptically polarised rays are
absorbed in the twisted on-state. This produces a positive contrast, i.e. black
information on a coloured background. Unfortunately, the contrast between
the on-state and the off-state transmission is low, since the thick cells
(d > 20mm) used in this type of STN-LCD results in a high degree of optical
retardation (And), see Chapter 2.
It was then found at the Royal Signals and Radar Establishment, Malvern,
UK (now DERA) that STN-GH-LCDs using one polariser and a nematic liquid
crystal of high birefringence or a STN-GH-LCD with no polarisers and a chiral
nematic liquid crystal of low birefringence both exhibit a very steep electro-
optical transmission curve over a very short range of voltages, see Section 7 for a
detailed description of this type of guest-host LCD.
88 Chapter 3
The STN-GH-LCD using one polariser displays white information against a
coloured background for reflective displays with a very high degree of multi-
plexing (1500). However, practical displays operate at lower duty cycles (1:200),
since the operating voltages are very sensitive to small inhomogeneities in the
cell gap (e.g. f 0.1 pm over a gap of 9.0 pm). The temperature dependence of the
threshold and operating voltages must be compensated for either through the
drive electronics or through the use of optically active dopants, which also
reduces the maximum degree of multiplexing possible. However, although a
polariser is used in this type of STN-GH-LCD, the viewing angle cone is still
very wide and the information content is high. Unfortunately, the switching
times of this type of display are long partly as a consequence of the high
viscosity of the nematic mixture due to the presence of large amounts of chiral
dopant used to induce the high degree of twist in the nematic director, but more
importantly the presence of the dichroic dye used to generate the electro-optical
effect. These problems were resolved by removing the dichroic dye and by
adding two polarisers with unusual configurations to generate an optical
contrast by the use of interference in the super birefringent effect (SBE) LCD.
Super Birefringent Effect L C D S ' ~ ~ - ' ~ '

The super birefringent effect (SBE-LCD)'67-'69 reported by Scheffer and
Nehring from Brown Boveri in Baden, Switzerland, uses the optical interference
of two normal, elliptically polarised modes of transmitted light generated by a
high-tilt, highly-twisted nematic structure viewed between two polarisers set in
an unusual way, i.e. the input polarisation direction is not parallel to the
nematic director at either substrate surface and the polarisers are not crossed at
go", see Figure 3.10.
Device Configuration
Prototype SBE-LCDs combined a chiral nematic mixture with large twist angle,
Q, = 240"-270" with a low ratio of the cell gap to pitch (d/p x 0.75), a positive
DIRECTOR
REAR SUBS1
FRONT SUBSTRATE
Figure 3.10 Absorption axes of thepolariser andanalyse with respect to the director at the
front and rear substrate of a STN-LCD.'69
INCIDENT
WOLARISISED POLARISER D
+ALIGNMENT
POIARISED LIGHT LIGHT ABSORBED

TRANSMIlTED IN IN THE ON-STATE
THE OFF-STATE
Figure 3.11 Schematic representation of a super birefringent efect (SBE) LCD.167-'69
dielectric anisotropy (A&x + 12), a moderately high birefringence (An x 1.4)

and a very large pretilt angle (20" < 8 < 30°), see Figure 3.1 1. The high pretilt
angle was produced using vacuum-deposited SiO/Si02. The polarisation axes of
the polariser and analyser are set at an angle of a and p, respectively, to the
nematic director at the front substrate surfaces, see Figure 3.10.
Off-State
The highly twisted structure of the chiral nematic phase generates two
elliptically polarised modes normal to each other. The difference in the optical
retardation induced by the chiral nematic phase causes the two rays to be out of
phase. The two rays interfere at the analyser; when they are recombined a
coloured appearance (blue or yellow) for the static off-state is the result. The
highest contrast between the transmission intensity of the off-state and the on-
state is produced when the polarisers were positioned at 30" and 60" to the
direction of alignment (nematic director) at the front and the rear surfaces,
respectively. This combination gives rise to an appearance of positive contrast
of black information on a yellow background often referred to as the yellow
mode. Rotating one of the polarisers by 90" inverted the contrast to yield white
figures on a blue background, referred to as the blue mode.
On-State
Upon the application of an operating voltage just above the threshold voltage
results in a modest realignment ( x 10%) of the nematic director in the middle of
90 Chapter 3
the cell since the chiral nematic phase is of positive dielectric anisotropy. This
realignment of the director reduces the effective birefringence of the nematic
medium. Therefore, the optical path difference and, consequently, the degree of
interference and, therefore, the intensity of transmitted light, are changed. Thus,
a large optical effect is generated by a small change in the applied voltage and a
very steep slope of the transmission versus voltage curve is produced as a direct
consequence of the highly twisted structure of the nematic phase. In practical
displays, a twist angle, 4, is chosen so that the display can be addressed at either
side of a narrow bistable range exhibited by the distortion-voltage curve.
A very high degree of multiplexing, e.g. 1:600, using RMS addressing and
standard Alt and Pleshko driving schemes is theoretically possible using this
device configuration. 174 However, practical SBE-LCDs exhibit a lower effective
multiplexing rate, e.g. a duty cycle of 1:120, for a contrast ratio of 1O:l for the
yellow mode and 8: 1 for the blue mode (direct viewing). This is due to a number
of factors, such as the lowering of the effective applied voltage as a result of the
resistance of the electrode layers and the internal resistance of the drive
electronics, or small variations in the homogeneity of the cell gap. The contrast
ratio is modest, e.g. 4: 1, for wide viewing angles, e.g. 45". However, it is still far
superior to that observed for TN-LCDs and GH-STN-LCDs (see Section 7) at a
comparable multiplexing ratio.
The response time of prototype SBE-LCDs with normal RMS addressing
were long, e.g. x 300ms at 20°C for 90% transmission. Therefore, they were
not suitable for applications, such as television, where video-rate addressing is
essential (response time < 40-50ms) in spite of a degree of bistability. The
highly twisted and tilted configuration of the nematic director in the cell is
stabilised by the high tilt angle in both the off-state and in the on-state. The high
tilt angle also helps to suppress a coqpeting distorsional structure as well as
light-scattering stripe instabilities.'71 It also contributes to a high transmission,
I, when the condition
Andcos2O,, = 0.8 pm (20)
is satisfied, where O,, is the average tilt angle in the non-select state. The
temperature dependence of the select and non-select voltages must be compen-
sated for either by using the device electronics or by the presence of appropriate
optically active dopants in the chiral nematic mixture. 177
The STN-LCD, 170 independently reported by Kando, Nakagomi and Hasa-
gawa in Japan, is very similar to the SBE-LCD, but with a different polariser
combination and lower twist and tilt angle (0 x 5"). Tilt angles of this
magnitude can be induced using rubbed polyimide, which is a much cheaper
and more practical alignment technique than using vacuum-deposited S O /
SiO2, which requires special deposition chambers. This type of STN-LCD
contained a chiral nematic mixture of positive dielectric anisotropy with a twist
of 180" in the nematic director from the top to the bottom of the cell. The
polariser and analyser were parallel with their polarisation axes at angles
or,P = 45" from the director at the substrates. Thus, the two elliptically polarised
rays create a bright yellow background in the off-state. The transmission, I, is
given by
where the transmission is optimised for yellow light, when the optical path
difference
6=And=lpm (22)
satisfies this requirement. The application of an appropriate voltage above the

threshold voltage gives rise to optical extinction, which produces a positive
contrast of black information on a yellow background.
In 1987 researchers at F. Hoffmann-La Roche in Basle, Switzerland reported
an optical mode interference (OMI) LCD, whose degree of multiplexability was
comparable with that observed for other S T N - L C D S . ' ~ ~However,
-'~~ it had the
significant advantage of being an intrinsically black-and-white display of
positive contrast with black figures on a white background. Furthermore, its
response times were also found to be much shorter than those of comparable
STN-LCDs. Moreover, its optical performance was much less sensitive to
variations in the cell gap. However, despite this combination of advantageous
features, OMI-LCDs suffer from the major disadvantage of poor brightness.
Although this can be improved to some extent by increasing the twist angle, the
degree of brightness remains substantially lower than that of a comparable
STN-LCDs or even TN-LCDs. As a consequence OMI-LCDs have yet to be
manufactured on a commercially significant scale.
The major difference between the configuration of the OM1 sandwich cell and
other STN-LCDs is that the optical path difference (6 = And << 1 pm) is much
lower. There is no requirement for a significant pretilt (0 < 9 < 5"), the twist
angle of the chiral nematic layer is lower (1 SO"), the front polariser is parallel to
the nematic director (a = 0") and the polariser and analyser are crossed
(/3 = 90"). The 180" twist gives rise to strong interference between the two
elliptically polarised rays. If the optical path difference is small, e.g. 0.4pm, a
bright, white, non-dispersive off-state is produced. The chiral nematic mixture
should be of positive dielectric anisotropy, low birefringence and exhibit a low
cell gap to pitch ratio (d/p = 0.3).
Electro-optical Performance of STN-LCDs169,178

This general description is typical for STN-LCDs with a twist angle
240" < @ < 270", pretilt angle 5" < 9 < 10" and a cell gap, d = 5pm. The
steepness of the electro-optical curve, y, should be as low as possible in order
to optimise the number of lines to be a d d r e ~ s e d : ' ~ ~
92 Chapter 3
therefore, k33/kll and E~ should be as large as possible. A large value of AE

would also lower the threshold voltage. However, since the overall response
time is given by
(24)
a high value of k33/k11 and E~ result in long response times. Therefore, the
easiest way to obtain short response times is to use a cell gap, d, as low as is
practically feasible and nematic mixtures with a low viscosity and compromise
values for k33/k11, AEand ELfor a steep electro-optical response curve and low
threshold voltage. The product And is fixed for STN-LCDs operating in the first
minimum, e.g. 0.85 for a 240" twist cell. Therefore, An should be as high as
possible. Unfortunately, the elastic constant and dielectric ratios also determine
the upper limit of the ratio of cell thickness to the chiral nematic pitch, before
the optically scattering stripe texture is manifested:176.178
A large k33/k11 ratio also permits a greater tolerance of pitch variations for a
given cell gap.
Temperature Dependence of Electro-optical Performance of STN-

LCDs
The electro-optical characteristics of multiplexed STN-LCDs exhibit a signifi-
cant dependence on temperature. This has to be compensated in order to avoid
variations of the optical performance of the display with temperatures. This can
be achieved electronically. However, this problem can also be solved by the use
of optically active, chiral dopants. The capacitive threshold voltage of a chiral
nematic mixture depends on the pitch of the mixture:'79
where p is the pitch, 4 0 the twist angle and d the cell gap. The temperature
dependence of the first term of Equation 26 is given by:
q?(T) = v322"C) +$!- 22°C)
It is reasonable to assume that the optical threshold voltage is proportional to

the capacitive threshold voltage. The twist angle and elastic constant ratios are
almost temperature independent. Therefore, since decreases with tempera-
ture, then the pitch of the chiral nematic mixture should also decrease
proportionately with temperature according to Equation 26, so that
becomes independent of temperature as possible. Therefore, optically active
dopants are used in order to induce the desired twist in the STN-LCD cell, as
well as a temperature decreasing pitch. The structure of the chiral dopants
should influence,other properties of the nematic mixture, e.g. the viscosity and
the nematic clearing point, as little as possible. Therefore, the chiral dopants
should both possess as high a helical twisting power as possible so that only
small amounts have to be used. The temperature dependence of STN-LCDs can
be successfully compensated for by using two chiral dopants of opposite twist
sense.177
Black-and-white STN-LCDs
In order to produce black-and-white as well as full-colour STN-LCDs, the
monochrome interference colours must first be eliminated. This was achieved
initially by using two STN-LCDs in a combined double-layer (DSTN) LCD
configuration.180*181 This involves the use of another non-addressed, passive
STN cell in addition to the active display STN-LCD. However, the non-
addressed cell has an opposite sense of twist of the nematic director in the cell
to that of the addressed STN-LCD. The second STN-LCD, which is identical to
the first, but not addressed at all, acts as a retardation compensation layer. The
use of an identical second STN-LCD in combination with the active STN-LCD
has the advantage that both displays exhibit exactly the same temperature
dependence of the birefringence with the same dispersion, assuming that both
cells are filled with the same liquid crystal mixture. The second STN-LCD is not
addressed and, therefore, there is no increase in power consumption. However,
the use of two identical STN-LCDs instead of only one clearly increases the cost
and weight of the final product significantly.
Simpler, cheaper and more practical approaches adopted subsequently to
compensate for the interference colours of STN-LCDs involve the use of a
passive optical element, such as selective polarisers and colour filters, see
Chapter 2.'82 A black-on-white appearance is produced most efficiently by
using optical retardation layers with a high birefringence and an opposite twist
sense to the STN-LCD. However, this also reduces the brightness of the display
due to additional light absorption.
Nematic Materials for STN-LCDs

Nematic liquid crystals with a combination of a terminal cyano group and a
94 Chapter 3
short alkyl chain were found not only to possess a positive value of the dielectric
anisotropy, but also a relatively high ratio of the twist and bend elastic
constants, k33/k11.183-185 This is illustrated by the data collated in Table 3.13
for the nitriles (36, 41, 53, 39, 49 and ~ 0 ) . ~ These nematic
~ 9 ~ ~ * ~ ~ * ~ ~ * ~ ~ 9 ~ ~
compounds combine a positive value of the dielectric anisotropy

'
(+4.5 < AE < +34) with a strong variation in the dielectric ratio (0.5 <
A E / E<~ 4.3) and a high bend-to-splay elastic constant ratio (1.4 < k33/
kll .c 2.4), see Table 3.13.'86*187This combination of physical properties can
be used to produce a steep electro-optical characteristic curve and a low
threshold voltage in nematic mixtures for use in STN-LCDs. However, these
nitriles also induce high values of the rotational and bulk viscosities. 188-'90
Comparisons of values quoted in the literature for the physical properties of
liquid crystals are often of dubious validity due to differences in the methods of
assessment often carried out at different absolute temperatures (e.g. 22°C or
25°C) or reduced temperatures (e.g. T N - I - 10°C or 0.95 x TN-1). The use of
extrapolated data from a wide variety of nematic mixtures of different
composition and properties at various concentrations is also common. Unfor-
tunately non-ideal behaviour is common for such mixtures and non-linear
behaviour is not unusual, i.e. the values extrapolated to 100% are more often
than not dependent on the matrix used and the concentration of the compound
to be evaluated. However, although the absolute values of the data collated in
Table 3.13, measured in the same way at the same reduced temperature
(0.96 x TN-I),are lower than those reported for the same compounds in the
literature," '91917192
usually measured at 22°C; the trends and relative values are
very similar.
Polar Nematic Materials for STN-LCDs

Mean-field theories,1939194 and the results of molecular modelling,' l 3 were used
in the design and synthesis of new nematic materials with a high positive value
for the dielectric anisotropy and the transverse dielectric constant, low rota-
tional and bulk viscosity and a high k33/kll ratio for use as components in
improved nematic mixtures for STN-LCDs. These studies seemed to suggest
that the elastic constant ratio k33/k11should be proportional to the length/
breadth ratio of individual molecules, i.e. long molecules with a narrow rotation
volume about the molecular long axis should exhibit a high k33/k11ratio. There
is some degree of validity to this theoretical treatment, if ensembles of molecular
dimers are taken into account,' l3 since molecular dimerisation effectively gives
rise to a large length-to-breadth ratio and, hence, in many cases a high ratio of
k33/ki1 -
This interpretation was confirmed to some extent by experimental results,
since it had been found that nematic mixtures composed of apolar nematic
compounds and polar nematic compounds (nitriles) consist of a number of
unassociated polar and non-polar monomers as well as associated polar
dimers.'13-'20This not only gives rise to a high value for the observed dielectric
anisotropy of the mixture due to the non-associated polar compounds, but also
h
3.
E
n
Table 3.13 Transition temperatures ("C) and some values187for the dielectric anisotropy (AE), the ratio of the dielectric anisotropy 2
and dielectric constant measured parallel to the director ( A E / E ~and
constants for the nitriles 36,41,53,39,49 and 50
) the ratio of the bend (k33) and splay (kll) elastic '
!?
S'
~~
Molecular structure Cr Sm2 Sml N I AE k33/kll Ref 3

rc1
s.
36 C 5 H l l w ~0 ~22.5 - - 0 35 0 + 13.3" 2.0 1.43" 82 09
41 C5H
*C
,,N 0 31 - - 0 55 0 +11.2" 2.1 1.77" 86 .:
$
t-
a
- 0 100 0 + 10.0" 2.35" 96
n
- (0 52) 0 +34.0b 4.3 1.13" 85 2
2
49 C s H l , < ~ C N 56 - - (0 52) 0 + 17.4b 2.2 1.36" 93
50 C s H 1 l w C N 0 62 0 43 0 52 0 85 0 +4,5" 0.5 1.59" 99

aMeasured at a reduced temperature of 0.96 x TN-~.
bExtrapolated value from a 10% concentration in a nematic mixture.
96 Chapter 3
to a low value of k33/k11, since the proportion of molecular dimers with a large
effective length-to-breadth ratio is low. Thus, mixing apolar materials of low
viscosity, weakly positive or negative dielectric anisotropy and a low k33/kl
elastic constant ratio with polar nematic liquid crystals of high positive
dielectric anisotropy and a high k33/kl ratio creates nematic mixtures suitable
for TN-LCDs with multiplexed addressing, since these require a low ratio of
k33/kl1, but which are not suitable for STN-LCDs.
Initial attempts to synthesise nematic compounds specifically for STN-LCDs
involved the preparation of molecules with polar substituents in lateral
positions. The dipole moment of the lateral substituents should have given rise
to a high E~ value, a low ratio and, hopefully, also a high k33/k11 elastic
constant ratio due to molecular dimerisation, see Table 3.10. However, this was
not found to be the case for the phenylbenzoate esters (106-108). l 7 The degree
of molecular association of the 4-cyano-2-fluorophenyl-4-heptylbenzoate (107)
and 4-cyano-3-fluorophenyl 4-heptylbenzoate (lo€!), which both possess a
fluorine atom in a lateral position, was found to be much lower than that of
the related non-fluoro-substituted 4-cyanophenyl 4-heptylbenzoate (106). The
complete absence of molecular dimerisation in the nematic phase of the ester
(108) (g = 1) means that the effective length is the molecular length. Therefore,
this results in a much lower clearing point for this ester compared to that of the
non-laterally substituted ester (106) or 2-fluoro-substituted ester (107), where
dimerisation is present. However, the value found for the k33/kl1 ratio for the 3-
fluoro-substituted benzoate (108) is the highest of all three closely structurally
related esters. The absence of association combined with the large dipole
moment of 6.1 D of the ester (108) results in the largest value for the dielectric
anisotropy observed for a liquid crystal (= + 50). This fortuitous combination
of properties led to the commercial production of the ester (108) and other
homologues as components of nematic mixtures with a low threshold voltage
and operating voltage and a steep electro-optical characteristic for use in STN-
LCDs.
A real breakthrough occurred with the synthesis of nematic compounds with
an alkenyl substituent, i.e. compounds with a non-conjugated carbon-carbon
bond in the terminal alkyl or alkoxy chain. A non-conjugated carbon-carbon
double bond absorbs in the near UV part of the spectrum well below the cut-off
frequency of the glass substrates used in LCDs with non-flexible substrates.
Thus, alkenyl analogues of commercial nematic materials with alkyl and alkoxy
terminal chains are just as chemically, electrochemically and photochemically
stable in the absence of oxygen and moisture. It was found that the physical
properties of these alkenyl-substituted material, such as the elastic constants
and the nematic clearing point, varied greatly for (otherwise identical) isomers,
depending on the position and configuration, i.e. trans and cis or ( E ) and (Z),
respectively, of the double bond, see Table 3.14.'"J '39192~195
Compounds with a trans carbon-carbon double bond at an odd number (1
and 3) of carbon atoms away from the molecular core (e.g. 74 and 78),see Table
3.14, exhibit a high nematic clearing point even higher than that of the reference
material (41)86with no double bond in the terminal chain.lg5The analogous cis-
Liquid CrystaI Displays Using Nematic Liquid Crystals 97
Table 3.14 Transition temperatures ("C), elastic constants (kll, k22,. k33, 10- l2
N ) , dielectric anisotropy (AE), dielectric constant measured perpen-
dicular to the molecular long axis ( E J , birefringence (An), refractive
index measured perpendicular to the director (no), rotational viscosity
(yl, Poise) and bulk viscosity ( q , Poise) for trans-1-(4-cyanophe-
nyl)-4-pentylcyclohexane (41), trans-l-(4-cyanophenyl)-4-[(E)-
pent-I-enyl]cyclohexane (74) and trans-l-(4-cyanophenyl)-4-[(E)-
pent-3-enyl]cyclohexane (78) extrapolated to 100% at
41 74 78
Cr-N 30 16 60
N-I 55 59 74
kll 8.98 9.40 9.41
k22 4.73 6.15 5.74
k33 18.3 22.8 24.1
k33h1 2.03 2.42 2.56
AE 12.22 13.03 13.13
E l 4.85 5.02 5.02
AE/El 2.52 2.60 2.62
An 0.119 0.136 0.130
no 1.487 1.493 1.483
Y1 1.28 1.34
v 215 22 1
isomers (75 and 79) exhibit a very low (extrapolated) clearing point. This has
been discussed in detail in Section 5 and will not be dealt with further here.
However, it was also found' 11-'13 that the values of most of the other physical
properties of the nitriles (74 and 78)195are also equal or superior to those of the
reference material (41)86with respect to use as components in nematic mixtures
for LCDs, especially STN-LCDs, see Table 3.14. The values of the elastic
',
constant ratio k33/kl birefringence, dielectric anisotropy of the alkenyl
compounds (74 and 78) are higher than those of the standard compound (41),
whereas the viscosity is comparable. Thus, alkenyl-substituted compounds are
used widely as components of nematic mixtures for use in STN-LCDs due to a
combination of very steep electro-optical curve, high nematic clearing point,
low viscosity and fast response times.
The thermal data collated in Table 3.15 for the compounds (74, 129-133)195
illustrate the effect of length of the alkenyl chain on the liquid crystal transition
temperatures. The nematic clearing point exhibits a clear odd-even effect
depending on the number of carbon atoms in the alkenyl chain. The highest
clearing points are those of the compounds (130, 74 and 133) with an odd
number of units in the chain. This is probably due to a higher degree of
anisotropy of the molecular polarisability along the molecular long axis for the
compounds 130,74 and 133 compared to the corresponding materials (129,131
and 132) with an even number of carbon atoms in the alkenyl chain. The
98 Chapter 3
Table 3.15 Transition temperatures ("C) for some trans-1-alkenyl-4-(I-cyano-
pheny1)cyclohexanes (74 and 129439)
~ ~~
129 56 (0 29) 0 108
130 (E) 0 66 0 73 0 108
131 (E) 0 45 55 0 108
74 (E) 0 16 0 59 0 108
132 (E) 0 14 0 39 0 108
133 (E) 0 18 0 49 0 108
134 0 29 0 31 0 108
135 0 50 0 53 0 108
136 0 30 (0 10) 0 108
137 0 46 0 53 0 107,108
138 0 19 0 32 0 107,108
139 0 38 0 53 0 107,108
transition temperatures for the compounds 129 and 134-139, 107~1089195also

collated in Table 3.15, reveal an odd-even effect for the clearing point and at
least one low melting point for a homologue (138) with a long alkenyl chain.
Alkenyl chains with a terminal carbon-carbon double bond are cheaper to
manufacture than alkenyl compounds with a trans carbon-carbon double bond
due to the absence of stereoisomers. Chemical reactions to form carbon-carbon
double bonds normally give rise to a mixture of trans- and cis-isomers. The cis-
isomer can be converted by chemical or catalytic procedures into the energeti-
cally more favoured trans-isomer. However, this is costly and time consuming.
Residual amounts of the unwanted isomer still have to be removed, usually by
recrystallisation from an appropriate organic solvent. Alkenyl chains with a
terminal double bond do not possess trans- or cis-isomers, since the terminal
hydrogen atoms are equivalent. Therefore, they are advantageous from a cost/
performance point of view, although their physical properties are often not as
good as those of the corresponding trans-isomers with respect to their potential
use in mixtures for LCDs.
Apolar
+
Nematic Materials for STN-LCDs
Apolar nematic compounds usually possess a low k33/kl1 ratio (e.g. 1.O < k33/
kl < 1S), partially due to the absence of molecular dimers. However, they are
still essential components of nematic mixtures for STN-LCDs, since they are
used to lower the viscosity and melting point of a nematic mixture of polar
components as well as improve the multiplexability of the mixture due in part to
the reduction in the proportion of molecular dimers of associated polar
molecules. Therefore, the synthesis of the first polar alkenyl liquid crystals with
high k33/kll ratios led to the synthesis of a series of apolar alkenyl-substituted
compounds with the carbon+arbon double bond in various positions in the
terminal chain. Some typical compounds (140-149) are shown in Table
3.16.107,196
Based on the evaluation of the results of physical measurements on polar
alkenyl compounds described above, the position and configuration of the
carbon-carbon double bond found in the apolar compounds is similar to that in
the polar compounds, essentially for the same reasons. However, the combina-
tion of a low melting point and a high nematic clearing point of the
bicyclohexane compounds (142, 143 and 145) is remarkable. The birefringence
and viscosity of these compounds is remarkably low. lo7
The corresponding three-ring compounds, e.g. 140-149, see Table 3.16,
exhibit higher melting and clearing points than those of the analogous two-
ring materials, as would be expected.1o79196 The low melting points of most of
these kinds of alkenyl compounds and the absence of ordered smectic phases,
such as the smectic B phase, renders them very useful components of nematic
mixtures. They can be used to increase the temperature range of the nematic
phase by increasing the clearing point and lowering the melting point of the
mixture. The presence of two cyclohexane rings and one phenyl ring gives rise to
a low viscosity and a low birefringence for these three-ring materials.'07*1089196
Fluoro- and methyl-substituted compounds are attractive from a synthetic
point of view due to the availability of cheap starting materials in an otherwise
complex synthetic procedure. The short methyl group is attractive, as a terminal
alkyl chain also contributes to a low viscosity and the suppression of smectic
phases.
The data collated in the Tables 3.14-3.16 reveals that the presence of a trans
carbon-carbon double bond in a limited number of specific positions gives rise
to a higher clearing point and a higher k33/kll ratio than those of the
corresponding material without a double bond in the terminal chain. Smectic
phases, especially the ordered smectic B phase, are strongly suppressed. This is
100 Chapter 3
Table 3.16 Transition temperatures ("C) for the alkenyl-substitutedcyclohexane
derivatives (14&149)"
140
141
*-
Molecular structure
m e
Cr
0
42
49
SmB
-
N
0
58
62
I
142 0 12 0 22 0 62 0
143 0 13 - 0 45 0
144 0 -8 52 0 64 0
145 0 19 - 0 32 0
146 0 74 - 0 214 0
147 0 105 - 0 194 0
148 0 87 - 0 186 0
149 0 52 0 104 0 177 0
particularly advantageous for the formulation of nematic mixtures with a low

melting point and a high clearing point as well as a steep electro-optical
transmission curve. The effect on the nematic clearing point and the suppression
of smectic phases is additive to some degree, see Table 3.17.'97*'98Compound
150197,198 only exhibits a smectic B phase. Compound 151 with one carbon-
carbon double bond possesses a smectic B phase at a lower temperature as well
as a nematic phase. Compound 152 with two double bonds exhibits a nematic
phase with a higher clearing point only than that of compounds 150 and 151. No
smectic phase could be observed. The compound 153 with three trans carbon-
carbon double bonds exhibits the highest clearing point of all and is purely
nematic.'98 However, the ratio of k33/kll is only slightly higher for compound
152 with two double bonds than that of the corresponding compound (151) with
only one double bond. Thus, the effect of the presence of carbon-carbon double
bonds on the elastic constants appears not to be additive, at least not in a linear
Table 3.17 Transition temperatures ("C) andsome valuesfor the viscosity (mPas)
and the ratio of the bend (k33) and splay (kll) elastic constants for
the apolar cyclohexane derivatives (15&155)197*'98
150 1
Molecular structure Cr
0 3 6 0
SmB
83
N
-
I Y1
0
k33h
151 fi 0 37 0 53 0 69 0 27 1.60
152 2 0 56 - 0 90 0 37 1.67
153 1 0 86 - 0 121 0 49
154 0 39 0 207 0 232 0 381
155 0 22 0 189 0 253 0 442 1.57
or consistent way. The ratio of k33/kll for the compound with two carbon-
carbon double bonds is still much lower than that exhibited by polar materials,
such as the phenyl benzoates (106-108), shown in Table 3.10, and the
benzonitriles (36,41,53,39,49 and 50), without a double bond in the terminal
chain, listed in Table 3.13.They are very much lower than those of the alkenyl-
substituted benzonitriles (74 and 78) shown in Table 3.14.
The viscosity of the three compounds 151-153 increases with the number of
carbon-carbon double bonds in the molecule. This is a general phenomenon for
alkenyl compounds. The three-ring alkenyl compounds 154 and 155198shown in
Table 3.17 exhibit higher clearing points than those of the analogous two-ring
materials. Although the melting points are low, compounds 154 and 155 exhibit
a smectic B phase at very high temperatures. The viscosity is also much higher
than that of the analogous two-ring compounds, as expected. The k33/k11 ratio
is slightly lower than that of the corresponding two-ring materials (151 and
152), perhaps due to a dilution effect attributable to the third cyclohexane ring.
Similar effects have been found for trans-4-n-alkylcyclohexyltrans-4-n-
alkylcyclohexanoates and the corresponding esters with one or two trans
carbon-carbon double bonds in the terminal chain. 199-201 However, the ester
with two terminal alkenyl chains only possesses a nematic phase over a
temperature range of 2 " ~199-201
. Related bicyclohexanes with an additional
non-conjugated heteroatom in the terminal chain show similar effects, however,
they are clearly not additive. This highlights the danger of extrapolating from
one mesogenic system to a n ~ t h e r . ~ ' ~However,
-~'~ some synergetic effects, due
102 Chapter 3
Table 3.18 Transition temperatures ("C),elastic constants (kll, k22, k33,
N) , dielectric anisotropy (A&), dielectric constant measured perpen-
dicular to the director ( E L ) , birefringence (An), refractive index
measured parallel and perpendicular to the director (n, and no),
viscosity (yl, mm2s-'), threshold voltage ( V ) , rate of change of
voltage with temperature (%"C-') and twist angle (") measured at
20°C unless otherwise stated for three typical nematic mixtures for
STN-LCDS'~~~~~~
Low twist STN-LCD206 STN-LCDiT8 High twist STN-LCD206
< -30 <40 < -30

79 100 71.5
13.1 13.1
6.9 6.9
21.1 21.1
1.61 1.61
3.1 3.1
+ 24.3 + 8.2 + 7.39
6.5 3.9 3.95
3.73 2.10 1.9
0.15 0.12 0.16
1.65 1.61 1.67
1S O 1.49 1.51
28 23 16.5
128 80
7100
1.04 2.15 2.14
0.33
180 220 270
to the presence of both an oxygen atom and a carbon-carbon double bond, can
give rise to a wide nematic temperature range and a low melting p ~ i n t . ~ ' ~ - ~ ' ~
These effects and the relationship between molecular structure and elastic
constants are not well understood.
Alkenyl compounds, such as those shown in Tables 3.14-3.17 are used as
major components of nematic mixtures for STN-LCDs with high information
content, see Table 3. 18.151*178*205*206
The data are quoted in the table to give
some indication of the general magnitude of the physical parameters of mixtures
designed for STN-LCDs with different twist angles. Nematic mixtures with a
sufficiently high k33/k11ratio can be prepared by adding components with very
high k33/kl1 ratio, such as those shown in Table 3.17, to other, much cheaper,
components in order to minimise the final cost of the m i ~ t u r e . ~ ' ~ Chemical
-~~'
companies offer a whole range of nematic mixtures for STN-LCDs. However,
these are usually prepared specifically for individual customers and data quoted
in catalogues generally only serve as an indication of general properties, which
are generally far superior to those quoted.
2 AXIS
(lj"" ROTATION
Figure 3.12 Molecular structure and absorption spectrum of an organic dye with apositive
dichroic ratio.
7 Guest-Host LCDS21 1-224

Guest-host (GH) LCDs are coloured displays,211-224 whereby the colour and
changes in colour are attributable to the absorption of incident light by a
dichroic guest dye dissolved in a liquid crystalline host material, usually
nematic, and the co-operative reorientation of these dichroic dyes in an electric
field, i.e. electro-optical devices.
A dichroic dye is a coloured organic compound whose optical absorption
depends strongly on the angle of incidence plane polarised light makes with
respect to the molecular long axis of the m ~ l e c u l esee , ~Figures
~ ~ ~ 3.12
~ ~ and
3.13. The absorption and emission of light is usually generated by a linear array
of conjugated n bonds giving rise to a molecular orbital n-n* or n-n* transition
from the ground state to the excited state. These transitions are usually
associated with the formation of charge transfer states with large dipole
moments. Therefore, the molecular shape of a dichroic dye is usually cylindrical
and often strongly resembles that of a calamitic liquid crystal. If the electrical
vector of incident polarised light is parallel to the transition moment of the
dichroic dye then some of the incident light is absorbed with a typical
absorption peak and bandwidth. Should the electric vector of incident polarised
light be orthogonal to the transition moment of the dichroic dye then little of the
light is absorbed. If small amounts of such a dichroic dye (guest) is dissolved in a
nematic material (host) and the nematic liquid crystal can then be reoriented by
an electric or magnetic field, then the dichroic dye molecules will also be
realigned co-operatively by nematic host under the influence of the electric
field. Thus, a light shutter for coloured light or a coloured liquid crystal display
device can be realised.
104 Chapter 3
A compound to be used as a dye in a coloured LCD must fulfil a number of
specifications. The dispersion forces of attraction between a guest dye and the
host nematic molecules are at a maximum for a dye molecule with a high length-
to-breadth ratio.235 Furthermore, the dye should be non-ionic as well as
chemically, photochemically and electrochemically stable in order to exhibit an
acceptable device lifetime. The dyes should also be available in a range of
colours. The colour may differ slightly from one nematic host to another due to
interaction between the chromophore and the surrounding liquid crystal. Since
dyes normally increase the viscosity of the host nematic mixture, a low
concentration is desirable in order to maintain short response times. Therefore,
chromophores with a high absorption coefficient, which is sometimes referred to
as the optical density, and a good solubility (z1-3%) in the liquid crystal host
are required in order to create the deep colouration of an effective display.
Solubility is especially critical at low temperatures, where the long aromatic dye
molecules generally have a pronounced tendency to crystallise out of the
nematic mixture. Therefore, only the azo and quinone dyes seemed capable of
satisfying these requirements. Typical structures of dichroic dyes developed for
GH-LCDs are listed in Tables 3.19-3.23. Furthermore, the transition moment
responsible for the visible colour, i.e. the charge transfer absorption band
excitation, should be parallel to the molecular long axis for most guest-host
LCDs. In order to maximise the contrast between the on-state and the off-state,
the transition moment of the dyes of positive dichroism should be well aligned
to the nematic director, i.e. they should exhibit a high order parameter Sdye:
where 8 is the angle between the nematic director and the transition moment of
the dye, All is the absorption of light polarised parallel to the director and A 1 is
the absorption of light polarised perpendicular to the director. Clearly A l l
should be as large as possible and Al should be as small as possible for a dye
with the transition moment parallel to the long molecular axis. Such a dye is said
to be of positive dichroism. If the transition moment is orthogonal to the
molecular long axis A l will be larger than All and the dye order parameter will
be negative. The dye is then defined as being of negative dichroism, see Figure
3.13. A slightly different order parameter is defined for dyes of this type:
where
and Al, A; are the optical densities of the homeotropically aligned sample and
the isotropic liquid, respectively, pn and pi are the density of the nematic phase
Table 3.19 Molecular structure, peak absorption (lmax,abs nm) , colour and order
parameter ( S ) for the dichroic dyes (156165)
Colour
Molecular structure (&nax,abs) S Ref
156 Red (482) 0.17 21 1,212
157 Blue (585) 0.55 21 1-213
158 Blue(595) 0.71 213
159 (450) 0.37 231
160 (359) 0.62 231
161 QYL@t +I q (400) 0.64 231
162 @Q
,- \ / Purple (560) 0.71 225,231
163 Orange (505) 0.46

(n = 1)
Violet (578) 0.51 222
(n = 2)
165 Blue (613) 0.59
(n = 3)
and the isotropic liquid, respectively and no, ni are the refractive indices of the
ordinary ray and the isotropic liquid, respectively.
It was postulated that Sdye should be higher than 0.73 in order that a good
contrast between the on-state and the off-state could be observed.23' Although
this is higher than that found for the nematic host far from the nematic clearing
point (Shost = 0.65), the order parameter of the dye itself is usually higher,
probably as a consequence of the greater shape anisotropy of the dye molecules.
This may also be attributable in part to a lower degree of thermal fluctuation of
the dye molecules or a higher local order induced in the nematic host by the
dye.213A limiting factor for Sdye may be that the non-linear shape could give rise
106 Chapter 3
Table 3.20 Colour, absorption maximum (3Lmax,abs, nm) , order parameter ( S )
and solubility (Sol., wt%) for the anthraquinone dyes (166-169)
Molecular structure
166
167 Blue O.Ma 0.6 231,242
168 Blue 0.66b < 1.0 240,242
169 Yellow (557) 0.66b 240,242
C4H 9
aMeasured in E43;bmeasured in E7.
to a small transition moment orthogonal to the major transition moment. This

would result in a lower contrast or at least a coloration of both the on-and off-
states. A related and often encountered measure of the order of the guest dye in
the nematic host is the dichroic ratio (R):
Table 3.21 Colour, absorption maximum (;lmax,abs, nm), order parameter ( S )
and solubility (Sol., wt%) for the anthraquinone dyes (170476)
Colour
Molecular structure (&nax,abs) S Sol. Ref
170 Blue 0.73" 0.8 247,248
d H H'O
171 Blue 0.72a 1.6 247,248
Blue 0.73" 247,248
Blue 0.74a 249
Blue (640/598) 0.72a 249
175 Blue 0.63a < 1.0 249

OCsHt9
d"H do
176 c7H'/ / Red (618) 0.66a 6.0 249
H' H o d o ' WYS
aMeasured in E7.
108 Chapter 3
Table 3.22 Colour, absorption maximum (Amax,abs, nm), order parameter ( S )
and solubility (Sol., wt%) for the anthraquinonedyes (177-180)
Molecular structure
177 Yellow (465) 0.79a 1.7 247,248
178 Yellow 0.79a 9.2 247,248
179
Q& /
Orange 0.80a 247,248
180 Red (520) 0.80" 15 247,248
'Measured in E7.
Table 3.23 Colour, absorption maximum (Amax,abs, nm), order parameter ( S ) ,
solubility (Sol., wt%) and transition temperatures ("C) for the
tetrazine dyes (18Ia, 182' and 183') and the azulene dyes (184dand
185")
Colour
Molecular structure (Amax,abs) S Sol. Ref
Red/violet (550) 0.7 15 252
254
255
'Cr-SmC, 49°C; SmC-N, 52.5"C;N-I, 63°C; bCr-SmC, 127°C;SmC-N, 132.5"C;N-I, 180°C;'Cr-

N, 107°C;N-I, 21 1°C;dCr-I, 124°C;N-I, 116°C;eCr-S, 190°C;S-I, 250°C.
Z AXIS
AXIS ABS
(IJ E
ROTATION E
Figure 3.13 Molecular structure and absorption spectrum of an organic dye of negative
dichroic ratio.
110 Chapter 3
A value of R > 6 is generally regarded as necessary for a GH-LCD with an
acceptable contrast.
There are many variants of this type of LCD. However, there are two main
types: the first uses a dichroic dye, one polariser and a nematic host; the second
type utilises a dichroic dye dissolved in a chiral nematic host and does not
require any polarisers. Both types can be used with nematic liquid crystals of
positive or negative dielectric anisotropy and dichroic dyes with positive or
negative dichroism. The first nematic guest-host effect was developed in the late
1960s at the RCA Laboratories2119212 at the same time as the DSM-LCD, see
Section 2. The second kind of GH-LCD was reported213six years later by White
and Taylor of Bell Laboratories, New Jersey, USA. The original versions of
both types of GH-LCDs were of negative contrast, i.e. white information
against a coloured background. Human factor studies suggest that positive
contrast displays, i.e. coloured information against a white background, exhibit
a higher degree of readability compared to equivalent negative contrast displays
with the same intrinsic contrast ratio between addressed and non-addressed
areas. The contrast of a given type of LCD can generally be inverted by driving
the background instead of the pixellated areas. However, this involves much
larger active electrode areas with the associated higher power consumption and
the possibility of electric field fringe effects.214Heilmeier and White and Taylor
GH-LCDs with positive contrast have been developed. However, only White
and Taylor GH-LCDs are of any real commercial interest due to the absence of
polarisers which are costly, absorb light and limit the already restricted viewing
angle further. There is a resurgence of interest in this type of display after more
than 25years due to these very reasons, although a limited number of GH-
LCDs have been manufactured during this time for the consumer market, e.g.
coloured digital watches.
Negative Contrast Heilmeier and Zanoni GH-LCDs211,212

The first nematic guest-host prototype nematic guest-host display device
contained a nematic liquid crystal (4-butoxybenzoicacid) and a pleochroic dye
(methyl red or indophenol blue) sandwiched between two (Nesa) electrodes
(d z 12pm) rubbed uniaxially, but with no additional orientation layer, see
Figure 3.14. One polariser was fixed to the front substrate surface with its
direction of maximum absorption parallel to the rubbing direction and, there-
fore, the nematic director.
Off-State
The nematic liquid crystal mixture containing the pleochroic dye is of positive
dielectric anisotropy and is aligned parallel with the director parallel to the
substrate surfaces. Therefore, plane polarised light is absorbed by the dye
molecules in non-addressed areas of the display and they appear coloured.
On-State
The application of an electric field between the electrodes results in a realign-
ment of the nematic liquid crystal mixture and the dichroic dye molecules
parallel to the electric field resulting in a lower optical density (absorption) and,
theoretically, the disappearance of colour assuming an ideal order parameter
( S = 1) of the nematic liquid crystal director and the dye molecules. A residual
absorption in this state gives rise to a display with a strongly coloured
background and weakly coloured information.
A Heilmeier and Zanoni GH-LCD with positive contrast has a very similar
construction. A suitable surface treatment results in a homeotropic orientation
for a host nematic mixture of negative dielectric anisotropy incorporating a
dichroic dye of positive dichroic ratio.2169217 The electric vector of incident
polarised light is perpendicular to the absorption moment of the dissolved dye.
No light is absorbed and the background appears colourless. An applied electric
field above a threshold value realigns the nematic host of negative dielectric
anisotropy and the guest dye parallel with the cell substrates. The transition
moment of the dye is now parallel with the electric vector of the transmitted
polarised light, which is consequently absorbed. Therefore, activated pixels
appear coloured against a white background.
Another version of the Heilmeier and Zanoni G H - L C D S ~ 'with ~ ? ~positive
~~
contrast is essentially the inverse of the Heilmeier and Zanoni GH-LCDs*l 1*212
with negative contrast described above. A nematic mixture of negative dielectric
anisotropy incorporating a dichroic dye of negative contrast is aligned parallel
INCIDENT AI INCIDENT
ur POLARISER ID
ELECTRODE b /
y@-+ALIG"T'llu
LIGHT TRANSMI'ITED LIGHT TRANSMITTED

IN THE OFF-STATE IN THE ON-STATE
Figure 3.14 Schematic representation of a Heilmeier and Zanoni guest-host liquid crystal
display (GH-LCD).162*163
112 Chapter 3
to the substrate surface in a standard cell with one polariser. Since the transition
moment of a dye of negative dichroism is perpendicular to its long molecular
axis,217see Figure 3.13, and, therefore, also to the direction of propagation of
the linearly polarised light, absorption should not take place. Therefore, the
non-addressed background pixels appear colourless or weakly coloured. An
applied voltage above the threshold value results in the dielectric realignment of
the host nematic mixture and the guest dye parallel to the field. The molecular
axes of both now lie in the plane of the cell orthogonal to the field. However, the
transition moment of the dichroic dye is now parallel with the electric vector of
the incident polarised light. The light is absorbed and activated pixels appear
coloured against a white background. The contrast ratio of this display mode is
lower than other configurations due to the low dichroic ratio ( x 5) of the
known dyes of negative dichroism and the fact that the transition moment
rotates rapidly round the long molecular axis and still absorbs some light in the
off-state.
White and Taylor GH-LCDs2'

The GH-LCD reported by White and Taylor in 1974 possesses a number of
advantageous features for simple alpha-numeric displays (two electrical con-
tacts per pixel segment) compared to the related GH-LCDs originally reported
by Heilmeier and Zanoni as well as TN- and STN-LCDs, which utilise
p ~ l a r i s e r s .It~ displays
~~ shadow-free information with a much wider viewing
angle than comparable GH-LCDs with one polariser and TN-LCDs (with two
polarisers) with similar low operating voltages, but at lower cost and higher
brightness. This is mainly due to the absence of polarisers, but also partially due
to a higher tolerance of cell gap inhomogeneities. The absence of polarisers also
allows the reflector to be integrated into the device itself, i.e. in front of the rear
glass substrate, in reflective rather than transmissive displays. This eliminates
the problems of parallax found for TN-LCDs, which degrades image quality.
The White and Taylor GH-LCD is based upon a change of phase type, i.e. from
a twisted (chiral) nematic phase to an oriented, untwisted (pseudo) nematic
phase, induced by the interaction with an applied electric field, see Figure 3.15.
In a chiral nematic phase incident unpolarised light is transmitted as two
elliptically polarised rays for any direction of propagation other than perpendi-
cular to the helical axis. The major axis of the ellipse of one mode and the minor
axis of the other mode are parallel to the nematic director. Therefore, either the
major or minor part of the elliptically polarised ordinary and extraordinary rays
is absorbed, depending on which one is parallel with the absorption moment of
the dye. There are several versions of the White and Taylor GH-LCDs with
positive and negative contrast, which exhibit important differences, especially
with respect to boundary conditions, sign and magnitude of the dielectric
anisotropy of the nematic mixture and the transition moment of the guest dye.
\ SUBSTRATE
COLOURED WHITE UNPOLARISED

LIGHT REFLECTED LJGHTREFLECED
IN THE OFF-STATE IN THE ON-STATE
Figure 3.15 Schematic representation of a White and Taylor guest-host liquid crystal
display (GH-LCD) ,213*214
Negative Contrast White and Taylor GH-LCDs*'

Display ConJiguration
The electrode surfaces of a normal LCD sandwich cell (d z 8-10 pm) are coated
with an alignment layer in order to induce a planar alignment of a host (chiral)
nematic mixture containing the dichroic dye of positive dichroism and a chiral
dopant. Due to the absence of polarisers a very thin mirror can be incorporated
within the cell on top of the rear glass plate electrode in direct contact with the
guest-host mixture, see Figure 3.15.
Off-State
The planar boundary conditions align the molecules of the chiral nematic
mixture and the dye molecules in the azimuthal plane, i.e. parallel to the device
substrates. The twisting power of the chiral dopant then gives rise to the
spontaneous formation of a helical structure of pitch, p , with an axis perpendi-
cular to the substrate surfaces (Grandjean texture), i.e. from the rear substrate
to the front substrate. Thus, incoming unpolarised light is absorbed efficiently
by the dye, which is distributed through an angle of 360"in the plane of the cell.
Thus, non-activated areas of the display appear strongly coloured.
On-State
The application of an electric field above a threshold value orients the director
114 Chapter 3
of the nematic guest-host mixture parallel to the field applied between the
electrodes, thereby leading to the unwinding of the helix, if the nematic mixture
is of strongly positive dielectric anisotropy. The director of the nematic phase
and, consequently, the transition moment of the dissolved dye is now perpendi-
cular to the electric vector of propagating light which is, therefore, no longer
absorbed. Therefore, transmitted light is reflected back along its path by the
internal mirror and the activated pixels appear white (or slightly coloured)
against a strongly coloured background. The threshold voltage (Vth) of actual
displays correlates well with the theoretical value for the unwinding of a chiral
nematic helix under these boundary conditions calculated much earlier.218
where d is the cell gap, p the pitch, k22 the twist elastic constant and A& the
dielectric anisotropy. Although the threshold voltage is inversely proportional
to the pitch, p , the degree of ellipticity decreases with increasing pitch, i.e. Alp
should be small. Therefore, since the contrast, C, is highest for almost circularly
polarised light, a compromise between threshold voltage and on-state transmis-
sion, I,,, must be made:
where the background brightness, I0K, is a constant for a given display. Since the
contrast ratio is found empirically to depend on the product Anp, a small An
value allows a larger value for p and, therefore, a lower threshold voltage for a
given cell thickness (dlp = 3 for optimum contrast). Similarly, although increas-
ing the dye concentration also increases the contrast ratio, the brightness is
found to decrease proportionately. However, commercial displays exhibit
moderate contrast at low threshold voltages in relatively thick cells, e.g. 5:l at
3 V in a 10 pm thick cell, where standard cell gap variations are inconsequential.
A turbid, cloudy appearance is produced for significant cell gap inhomogeneity.
High brightness can also be achieved with a suitable internal mirror.214
However, a major disadvantage of this configuration is that high operating
voltages are required to align the optic axis parallel to the electric field in order
to obtain good contrast. Furthermore, metastable disclination lines formed
after a given pixel has been deactivated can give rise to a visible after image.
Positive Contrast White and Taylor G H - L C D S ~69219-221

~~,~~
D isplay Configuration
A suitable surface treatment results in a homeotropic orientation for a nematic
mixture incorporating a dichroic dye of positive contrast and an amount of a
chiral dopant insufficient to overcome the surface forces and generate a twisted
structure in the nematic phase.
0ff-State
The static homeotropically aligned guest-host nematic mixture (plus guest dye
and chiral dopant) is optically transparent and, therefore, the display appears
colourless in the non-activated state. The lower limit of the pitch for a given cell
gap, before a twisted nematic structure becomes energetically more favoured
than the homeotropic nematic structure, is determined by the d/p ratio:221
where k33is the bend elastic constant. The homeotropic boundary conditions of
this GH-LCD should result in the helix axis lying in the plane of the cell.
However, the observed texture, referred to as the scroll texture is made up of a
multitude of left- and right-handed helices. Indeed the helix axis is found to still
lie predominantly perpendicular to the cell walls. However, this texture does not
scatter light and replaces the scattering texture responsible for the formation of
after-images in cells with parallel boundary conditions, resulting in their
elimination for this display configuration.
On-State
The application of an electric field above the threshold value results in a
reorientation of the nematic liquid crystal mixture, if the nematic phase is of
negative dielectric anisotropy. The optically active dopant then applies a torque
to the nematic phase and causes a helical structure to be formed in the plane of
the display. The guest dye molecules are also reoriented and, therefore, the
display appears coloured in the activated pixels. Thus, a positive contrast
display is produced of coloured information against a white background. The
threshold voltage is dependent upon the elastic constants, the magnitude of the
dielectric anisotropy, and the ratio of the cell gap to the chiral nematic pitch:
Super Twisted Nematic (STN) GH-LCDS'~'-'~'

A super-twisted version of the Heilmeier and Zanoni GH-LCD'67-'69 with one
polariser and a nematic liquid crystal of low birefringence exhibits very steep
electro-optical transmission curves and a high degree of multiplexability. 161-166
This is of great advantage compared to both the standard Heilmeier and Zanoni
and White and Taylor GH-LCDs due to the ability to display high information
content at high contrast and low voltage. This is especially advantageous using
achromatic guest-host mixtures in order to produce black-on-white displays.
Chapter 3
CoWuRED UNPOLARISED
LIGHT REFLECTED LIGHTREFLECTED
IN THE OFF-STATE SUBSTRATE IN THE ON-STATE
Figure 3.16 Schematic representation of a super twisted White and Taylor guest-host
liquid crystal display (STN-GH-LCD).16'-16'
The cell contains a nematic mixture with a twist of 270" and homogeneous
alignment with a high pretilt angle (O), see Figure 3.16. The nematic mixture is
composed of one or several dichroic dyes, a chiral dopant and a nematic host of
low birefringence.
0ff-St a te
Incident plane polarised light after traversing the polariser is absorbed by the
dichroic dye, whose transition moment is parallel to the electric vector of the
light. The internal reflector reflects the coloured light back through the
polariser. Thus, the non-addressed background appears coloured.
On-State
The application of a surprisingly low voltage, e.g. 1 V, results in the reorienta-
tion of the nematic host and the dissolved dyes parallel to the applied field and
orthogonal to the plane of the cell as the chiral nematic twist is unwound.
Therefore, the transition moment of the dye is aligned perpendicular to the
electric vector of the plane polarised light and little or no absorption takes place.
Thus, activated pixels appear white against a coloured background. The
difference in voltage between the threshold voltage, Vth, and the voltage, Yo,,
at which the cell can be regarded as activated and light transmission is high, is
Liquid Crystal Displays Using 117
50 %
40 Yo
z
P
g 30%
z 20%
dF
10 Yo
0%
0 1 2 3 4
VOLTAGE (V)
Figure 3.17 Schematic representation of the transmission against voltagefor a guest-host

LCD.
small. This results in a very steep electro-optical characteristic, see Figure 3.17.
Therefore, since the number of addressable lines according to Alt and Pleshko is
proportional to the ratio Von/Vth, a large number of lines can be multiplexed,
e.g. 200 for Von/Vth = 1.07 and 8 = 30". The ratio Von/Vth is a constant for
0.5 < d/p < 1.0. However, since there is a h e a r dependence of Von/Vth on d/p,
then p should be as large as possible within the above limit for a given cell gap.
The temperature dependence of the operating voltages can be compensated
either electronically or by using a chiral dopant whose pitch decreases with
temperature.
Dichroic Dyes - Guests

Dichroic dyes of positive and negative dichroism are required for various
configurations of the Heilmeier and Zanoni as well as the White and Taylor
GH-LCDs with positive and negative contrast. They should be chemically,
photochemically and electrochemically stable, as well as exhibit a high order
parameter, a high dichroic ratio and good solubility in the host nematic matrix.
Positive Contrast Dyes

The dye used in the original guest-host LCDs with one polariser reported by
Heilmeier and Zanoni2' 1-212 was a standard commercially available dye called
methyl red (156), shown in Table 3.19. Methyl red is an azo dye with a
cylindrical shape, a relatively high length-to-breadth ratio and a transition
moment parallel to the molecular long axis. Therefore, it exhibits a relatively
high positive dichroic ratio. A number of other commercially available dyes
with a less linear molecular structure, such the indophenol dye (157), were also
investigated,
The first prototype GH-LCDs based on the chiral nematic phase change
effect investigated hitially by White and Taylor also used indophenol dyes.
118 Chapter 3
Transition Moment
..-
Nematic Director
Figure 3.18 Canonical structures of an azo dye in the ground state and the excited state.
However, the order parameter of these dyes was found to be too low to generate
a high contrast ratio.213Therefore, a blue thiazole azo dye (158), with a quasi
liquid crystalline molecular structure, also shown in Table 3.19, was synthesised
specifically for improved chiral nematic phase change GH-LCDs. The presence
of an electron-donating and an electron-withdrawing group at each end of the
molecule leads to a large energy difference between the n ground state and the z*
excited state. It is this process of absorption, transition from the n to n* states
and subsequent radiative decay, which is generally regarded as responsible for
light emission in azo dyes. The excited state is a charge transfer complex, whose
transition moment is parallel to the director of the nematic host, see Figure 3.18.
The polar nature of asymmetrically subsiituted azo dyes, such as 158, in the
ground and excited states gives rise to a large dipole moment in the direction of
the molecular long axis. This facilitates solvation and orientation in a nematic
host of positive dielectric anisotropy and co-operative reorientation an electric
field. Such asymmetrical azo dyes with liquid crystalline properties had been
synthesised many decades earlier.238The use of long thin azo dyes, such as 158,
resulted in a higher brightness and contrast, due to the higher order parameter,
than similar prototypes using the indophenol dye (157).2' 1-213 Thiazole azo
dyes, such as 158, had been developed originally for colouring man-made fibres.
It was postulated that the order parameter should increase with increasing
length-to-breadth ratio, i.e. anisotropy of molecular shape.229-2329239 This was
indeed found to be case for the four azo dyes (159-162) and the three
merocyanine dyes (163-165; n = 1, 2 and 3) collated in Table 3.19.225 The
dichroic ratio of the tri-azo dyes (161 and 162) is significantly higher than that of
the related mono-azo and di-azo dyes (159 and 160). The mono-azo dye (159)
exhibits a particularly low order parameter. Unfortunately the solubility also
decreases with increasing molecular length. Moreover, even with very long thin
dyes, such as compound 161, there is still some absorption in the supposedly
colourless state of GH-LCDs. This lowers the observed contrast since both the
on-state and the off-state are coloured, strongly and weakly, respectively.
However, azo dyes such as those shown in Table 19 appeared more-or-less to
satisfy most of the requirements of the guest-host chiral nematic phase change
LCDs. However, accelerated life-time tests revealed that photochemical degra-

dation of many azo dyes in strong sunshine would lead to unacceptably short
device lifetimes. This was especially true of azo dyes absorbing in the blue region
of the visible spectrum. The fading of the dye due to photochemical degradation
in intense sunlight was found to follow first order kinetics, i.e. to be linearly
dependent on time at constant illumination.2339234 Since the optical density is
proportional to the concentration of the dye according to the Beer-Lambert
law, then
is found to be valid,233where A. is the optical density before illumination and At

is the optical density at time t , and k is a rate constant. Some of the
decomposition products collect at the substrate surface and gave rise to an
homeotropic alignment of the nematic director and the dissolved dye. This can
lead to an apparent acceleration of the fading of the dye. Although the glass
used as substrates in GH-LCDs actually removes most UV light up to the glass
cut-off point (z300nm), azo dyes still absorb in the UV in this region.
However, the use of UV filters was found to significantly improve practical
display lifetimes by absorbing most of the residual UV radiation above the glass
cut-off point. It is this residual UV which is responsible for the photochemical
degradation of most dyes with colours from yellow to red. However, GH-LCDs
incorporating azo dyes were manufactured for devices which are not exposed to
continual sunlight, such as digital watches.234However, UV filters do not
significantly inhibit the photochemical degradation of blue azo dyes. The rate of
decomposition of blue azo dyes is much higher than that of azo dyes of any
other colour. This not only prohibits the creation of blue GH-LCDs, it prevents
the formulation of achromatic black dye mixtures. Unfortunately, although the
incorporation of lateral substituents in azo dyes gives rise to a substantial
increase in colour fastness for textile applications, attempts to improve the
stability of blue or other azo dyes designed for GH-LCDs by incorporating
lateral substituents did not result in longer lifetimes in nematic mixtures.234
Dyes incorporating lateral substituents were also found to exhibit lower order
parameters, although greater solubility, than equivalent non-substituted dyes.
This is probably due to the lower degree of molecular shape anisotropy. The
effect on the order parameter of multiple lateral substitution was not additive as
this effect is probably due to the steric effect of broadening the molecule of the
largest lateral s ~ b s t i t u e n t . ~ ~ ~
Similar lifetime issues were encountered for a large variety of dyes with an
elongated molecular structure with an extended conjugated core similar to that
of the azo dyes, such as the merocyanine dyes (163-165; n = 1-3), shown in
Table 3.19, as well as azomethines and methine dyes.2279235 However, most of
these exhibit much lower order parameters and dichroic ratios than conven-
tional azo dyes prepared earlier.
Anthraquinone dyes were already known to exhibit good thermal stability
120 Chapter 3
Figure 3.19 Isomeric structures of an anthraquinone dye used in guest-host LCDs.
Transition Moment
Nematic Director
Figure 3.20 Isomeric structures of a diamino-substituted anthraquinone dye.
and light fastness, e.g. for dyeing textiles. This is due in part to hydrogen
bonding between the hydrogen atoms on the amino and hydroxyl functions with
the adjacent carbonyl groups, see Figure 3.19. They also possess a transition
moment parallel to the director of the host nematic mixture, see Figure 3.20.
The excited state formed by a n-n* transition is a charge transfer state with a
large dipole moment. Therefore, many anthraquinone dyes24c251 used in
plastics and textiles, such as some of those collated in Tables 3.20-3.22, were
investigated as potential dichroic dyes for GH-LCDs. Accelerated sun tests
established that anthraquinone dyes,240such as 166-169, were much more
resistant to photochemical degradation in nematic host mixtures than the azo
dyes investigated previously. The rate constants for fading of the dye under
identical illumination conditions are at least an order of magnitude lower for the
anthraquinone dyes than those of comparable azo or merocyanine dyes.2339234
Unfortunately, the order parameter (Sdye = 0.64) and solubility ( < 1wt%) of
simple non-substituted anthraquinones, such as 166 and 167, were found to be
too low to obtain a satisfactory ~ o n t r a s t . ~Since
~ ~ -the
~ ~optical
~ density of
anthraquinone dyes is three or four times lower than that of azo dyes of the
same colour, correspondingly higher amounts of the dichroic dye must be
dissolved in the nematic host in order to obtain the same deepness of colour.
Although the presence of one or two aromatic substitutents on amino functions
in positions 1,4,5 or 8 in modified anthraquinone dyes such as 168 and 169 led
to higher order parameters and s o l ~ b i l i t these y ~ ~were~ ~still
~ too
~ ~low
~ ~ ~ ~ ~ ~
for satisfactory operation of GH-LCDs. Higher order parameters were
obtained for dyes such as 170-176 shown in Table 3.21, with one alkyl or aryl
substituent in position 2 or two alkyi or aryl groups in positions 6 or 7 adjacent
to the amino and hydroxy functional groups in the anthraquinone
core.225,236,247-249 Unfortunately, the solubility and range of colour of these
dyes were also found to be insufficient for satisfactory display performance,
especially for black-and-white displays. However, anthraquinone dyes incor-
porating sulfur atoms as the linkage between the anthraquinone core and the
pendent aliphatic and/or aromatic groups in both the 2,6 and 2,7 positions were
found to be surprisingly soluble (515%) in host nematic mixtures of positive
and negative dielectric anisotropy, see Table 3.22. Many of these thio-substi-
tuted anthraquinone dyes exhibit high order parameters (Sdye M 0.8) and
absorption coefficients in nematic mixtures, as well as exhibiting a full range of
colours.22592473248 The colours exhibited by anthraquinone dyes cover some part
of all of visible spectrum depending on the number, position and nature of the
substituents. For example, some compounds with two pendent groups in
position 2,6 or 2,7 such as 177 and 178, exhibit a yellow colour, some dyes with
three substituents, such as 179, are orange and some dyes with four substituents
such as 180 are red. Dyes with cyclohexyl substituents are very soluble in
various nematic host mixtures, e.g. compare compounds 177 and 178. The high
order parameter of these anthraquinone dyes is surprising to some degree.
However, the long molecular axis of suitably substituted anthraquinone dyes
may still be more-or-less parallel to the director of the host nematic mixture.
This appears to be valid for 1,5-disubstituted anthraquinones and 2,6-disub-
stituted anthraquinones shown in Tables 3.20-3.22. However, the high order
parameter of the tri- and tetra-thiosubstituted anthraquinone dyes, such as 179
and 180, is still surprising.
Positive contrast GH-LCDs with black figures against a white background
exhibit a higher contrast than similar GH-LCDs with coloured figures against a
white background. However, this requires black dye This can be
achieved by mixing several of the anthraquinone dyes collated in Tables 3.19-
3.22, but especially the thio-substituted dyes shown in Table 3.22, to cover the
whole of the visible These dyes have absorption bands between
50-150nm broad. Therefore, three dyes are sufficient to absorb almost all
wavelengths in the visible spectrum (380-780 nm).246251
Negative Contrast Dyes

The tetralins (181-183) collated in Table 3.23 are red liquid ~ r y s t a l swith
~ ~ a~ * ~ ~ ~
transition moment orthogonal to the molecular long axis, see Figure 3.13. Since
they are liquid crystalline themselves they are very soluble in nematic host
mixtures. Such mixtures exhibit high negative dichroic ratios and can be used in
White and Taylor GH-LCDs with positive contrast. Furthermore, mixtures
containing an additional anthraquinone dye of positive dichroic ratio can be
used to switch from one colour to another under the action of an electric field.
The azulenes (184 and 185) also shown in Table 3.23 are red liquid crystalline
dyes with a negative dichroic ratio. The compound 184 exhibits a low order
parameter, probably due to its strongly non-linear shape.254 However, the
longer, thinner compound 185 results in the formation of an unidentified
122 Chapter 3
smectic phase.255 Both of these classes of azulene dyes are very soluble in
nematic host mixtures suitable for GH-LCDs.
Nematic Liquid Crystals -Hosts

Analysis of the electro-optical characteristics of GH-LCDs showed214that for
all types of guest-host chiral nematic phase change effects the contrast is
greatest when
AnP
-<<
R 1 (37)
where p is the pitch of the chiral nematic host. However, since p should be as
large as possible in order to minimise the value of the voltage required to unwind
the helix, then An of the chiral nematic host should be as low as possible. This is
fortunate since nematic hosts are required with a low UV absorption above the
glass cut-off point in order to minimise photochemical degradation of the
chromophore. Both physical properties are lower for compounds incorporating
aliphatic rings, such as cyclohexane, than for analogous materials with aromatic
phenyl rings. In the absence of polarisers and dyes, liquid crystal molecules with
an aromatic core absorb UV light and are degraded photochemically. However,
in nematic mixtures containing organic dye molecules, energy transfer from the
liquid crystal molecules in the excited state to the chromophore in the ground
state leads to accelerated photodegradation of the photosensitised dye as well as
decomposition of the host nematic mixture. Conversely, the degradation of the
nematic host is retarded.23' This degradation can be minimised by using nematic
hosts with no aromatic rings. Furthermore, in order to maximise the contrast
between the on-state and the off-state the order parameter of the chiral nematic
mixture and dyes should be as high as possible. This implies a high nematic
clearing point, see Chapter 2.237As always the viscosity should be as low as
possible in order to minimise switching times. They should be of high positive
dielectric anisotropy for GH-LCDs of negative contrast and of negative
dielectric anisotropy for GH-LCDs of positive contrast.
Nematic Liquid Crystal Hosts of Positive Dielectric Anisotropy

The first nematic guest-host prototype nematic GH-LCD reported by Heilme-
ier and Zanoni2' contained methyl red (157) as the dichroic dye dissolved in 4-
butoxybenzoic acid256as the nematic liquid crystal host. Other hosts investi-
gated later212included 4-methoxycinnamic and 4-ethoxy-4-aminoben-
~ o n i t r i l e ~(28),
~ ' ~see
~ Table 3.4. The melting point of these three single
components is very high. Therefore, prototype GH-LCDs had to be operated
and evaluated at very high temperatures. Thermal decomposition of the
mixtures led sequentially to lower contrast, homeotropic orientation due to
decomposition products and finally device breakdown. However, these initial
experiments were sufficient to demonstrate the feasibility of this display type.
The advent of the apparently superior White and Taylor device retarded the
development of better guest-host materials and mixtures for this device type.
White and Taylor found it was much more practical to use a room
temperature nematic mixture containing MBBA (7), the corresponding ethyl
homologue N-(ethoxybenzylidene)-4'-butylaniline (EBBA) and compound (28),
see Table 3.4, reported earlier for use in TN-LCDs. This mixture was doped
with an optically active Schiffs base as chiral dopant and various dichroic
The azo-thiazole dye (158) was found to generate the highest order
parameter (0.71 at in a related mixture consisting of cyano-substituted
Schiff s bases and another optically active Schiff s base as chiral dopant. This
mixture exhibits a chiral nematic phase with a short pitch (3pm) at room
temperature due to the helical twisting effect of the optically active material.
This mixture has a relatively high threshold voltage ( x 6 V) in a 12pm test cell
with an internal mirror. The contrast ratio between the blue or blue-green off-
state to the white on-state was about 5: 1. Brightness was high due to the absence
of polarisers and the internal mirror.
Most of the dyes developed for GH-LCDs were evaluated in commercial
nematic host mixtures such as E7 and E8 from the British Drug Houses
company (now part of E. Merck) and related mixtures from F. Hoffmann-La
Roche. Although the birefringence of these mixtures is high (An x 2.2) and,
therefore, not very suitable for commercial GH-LCDs, they allowed compar-
isons to be made between the data obtained for different dyes in very similar
host mixtures. The merocyanines and a wide variety of related dichroic dyes, see
Table 3.19, were evaluated and compared in a simple tertiary mixture of Schiff s
bases, although this mixture was of no real commercial relevance, it allowed
dyes to be compared in the same matrix and promising dyes identified. This is
also true of a nematic mixture composed of 4-cyanophenyl esters and 4-n-
alkoxyphenyl esters of trans-4-n-alkylcyclohexaneacids. 12' Red guest-host
nematic mixtures containing dyes of negative dichroism possess attractive
response times in prototype GH-LCDs.
Commercial nematic mixtures of positive dielectric anisotropy now contain
nematics of low birefringence such as 50, see Table 3.13. The birefringence of
nematic liquid crystals containing cyclohexane rings, such as 41 and 50, in place
of aromatic rings is lower than that of the fully aromatic compound such as 36
due to the absence of polarisable n electrons. The dioxane derivative (49)
combines a relatively low value of birefringence with a high positive dielectric
anisotropy. This promotes a high dye order parameter under the action of an
electric field.
Nematic Liquid Crystals Hosts of Negative Dielectric Anisotropy

A binary mixture of 4-pentylphenyl4-pentylbicyclo[2.2.2]octaneand 4-heptyl-
pheny14-pentylbicyclo[2.2.2]octane,49.'26see Table 3.8, exhibits a wide nematic
phase at room temperature with a high clearing point, a low birefringence
(0.08), a moderate viscosity (46 cP) and a weakly negative dielectric anisotropy
(A& - 1.1) measured at 20°C.247 It is a surprisingly good solvent for
124 Chapter 3
anthraquinone dyes such as the blue anthraquinone dye (171). A 12pm cell with
the Heilmeier and Zanoni configuration with one polariser and homeotropic
alignment containing a mixture of 1 wt% of the dye 171 and the binary mixture
and the binary ester mixture exhibits positive contrast (blue figures on a white
background), a low threshold voltage (4.5V), short response times (ton =
tog = 0.5 s at 9 V) and acceptable contrast (5: 1) at 20°C.
8 In-Plane Switching (IPS) LCDs258-261

A novel electro-optical display device based on the response of a nematic liquid
crystal to electric fields in the plane of the display was reported by researchers at
the Fraunhofer-Institut fur Festkorper Physik in Freiburg, Germany in the
early 1990s.258-261 This type of LCD does not exhibit the strong viewing angle
dependency and asymmetry of optical contrast observed for ECB-LCDs, TN-
LCDs and STN-LCDs, since the optical axis of the nematic liquid crystal lies in
the plane of the cell. Therefore, it aroused considerable interest as a possibility
to overcome the limitations of commercial LCDs, although field effects utilising
interdigitated electrodes had been reported There are several
variants of IPS-LCDs. Some of these, however, display a pattern of stripe
defects at the electrodes due to fringing effects. Therefore, the most promising
type utilises a nematic liquid crystal of negative dielectric anisotropy in the
original display c o n f i g u r a t i ~ n . ~The
~ ~ -electro-optical
~~~ characteristic allows
grey-scale, i.e. full colour. In-plane switching (IPS) LCDs utilising this config-
uration with active matrix addressing are currently being manufactured,
especially as computer monitors due to their wide viewing angles and good
contrast. 264-268 The dark state is black with the absence of interference colours
even without a compensation (optical retarder) film.
A nematic liquid crystal mixture is contained in a liquid crystal cell with non-
transparent interdigitated finger electrodes on one substrate of the cell as shown
in Figure 3.21. The second substrate does not carry an electrode. The narrow
stripe electrodes, e.g. = 5pm, are metallic, e.g. I T 0 or chromium, and are
separated, e.g. 10 pm, to create the active pixel area. Therefore, the aperture
ratio is low. Consequently the display brightness may also be low in the absence
of strong backlighting. A zig-zag configuration of the electrodes can further
suppress interference colours. A homogeneous azimuthal alignment of the
nematic liquid crystal director with a non-zero zenithal pretilt angle (0 x 3-5")
is obtained using rubbed polyimide at both surfaces. The relatively high pretilt
angle is in order to avoid reverse tilt. The nematic director is parallel at both
surfaces and throughout the cell. The polarisers are crossed with the first
polariser set parallel to the optic axis at the first substrate surface.
INCIDENT
I POLARISER iD
POLARISEDLIGHT
TRANSMITI'ED
IN THE OFF-STATE
/ LIGHT ABSORBED
INTHEON-STATE
Figure 3.21 Schematic re resentation of an in-plane switching liquid crystal display (IPS-
,~CD).258-26?
Off-State
The homogeneously aligned nematic liquid crystal mixture does not rotate the
plane polarised light generated by the first polariser. Therefore, the light is
absorbed by the analyser. Very efficient absorption of light takes place.
On-State
The application of an electric field between the electrodes causes the nematic
liquid crystal mixture to reorient itself with the molecular long axis parallel to
the applied field. The strong anchoring at the rubbed polyimide surfaces
maintains the original parallel alignment at the surfaces. This results in two
areas of twist as shown in Figure 3.21. The direction of twist of the two regions
of twist are opposite. However, the upper area of twist dominates the optics of
the cell in the on-state as long as the ratio of the pitch of the upper area of twist
(PT)to the lower area of reverse twist (PR)is greater than 5: 1. A twist angle of
up to 90"can be produced. The amplitude of transmission in the first minimum
(u = J 3 ) is then relatively high. However, the influence of the area of reverse
twist increases for the other Gooch and Tarry minima (i.e. Jl5, $5, J63, etc.).
The switch-on time, ton, and the switch-off time, t,E, are both proportional to
the square of the cell gap, d, as for the TN-LCD?"
126 Chapter 3
(39)
where
Therefore, thin cell gaps also give rise to short rise times. Thus, thin cells are
required for short response times, e.g. to, = 20 ms and ton = 30 ms for the IPS-
LCD prototype described above with a small cell gap (d = 3 pm). Switching
times of this order of magnitude are compatible with active matrix addressing at
video rate. However, special electrode configurations2s266 had to be developed
in order to shield the pixel areas from electric fields caused by the bus electrodes.
These fringing effects generated crosstalk and seriously degraded the optical
response of the first prototype IPS-TFT-LCDs. Large IPS-TFT-LCDs (13"
diagonal) are now commercially available with XGA resolution, see Table
3.24.2673268The data for a CRT and a TN-TFT-LCD are also collated in Table
3.24 for comparison purposes. In the IPS-TFT-LCDs the polariser orientation
is inverted in order to produce black information on a white background as
opposed to the white information on a black background described above.
Nematic Materials
Nematic liquid crystals of negative dielectric anisotropy with a low rotational
viscosity and a high value of the twist elastic constant k22 are required, see
Equations 3840. Furthermore, a high value for the resistivity and short
Table3.24 Specijications for a 15" C R T , a 13.3" IPS-TFT-LCD with XGA

resolution and a 12.1" TN-TFT-LCD with XGA resolution
(1024 x 768 x 3pixels)267.268
Property CRT TN- TFT-LCD IPS-TFT-LCD
Approximate display weight 13kg 0.5 kg 1 kg

Display volume 48 858 cm3 468 cm3 1080cm3
External dimensions 36.2 x 35.8 x 37.7cm 27.6 x 20.0 x 0.9cm 33.8 x 26.4 x 1.3cm
(width x height x depth)
Effective display area 270 x 200mm 246 x 184mm 270 x 202mm
Pixel size 0.28 x 0.28mm 0.26 x 0.26mm 0.26 x 0.09mm
Number of pixels 6 888 000 (RGB) 2 359 296 (RGB) 2 359 296 (RGB)
Number of colours > 260000 > 260000 > 260000
Vertical viewing angle & 90" + 10" and -30"
Horizontal viewing angle
Brightness
f 90"
100cd m-2
* 45"
70 cd m-*
70"
rt 70"
120 cd m-'
Power consumption 85W 4w 18W
Response time (ran + tom) 80 ms 70 ms
Contrast ratio > 1OO:l > 1OO:l > 1OO:l
response times, i.e. to, + toff5 50 ms, are necessary for active matrix addressing,
e.g. on a TFT substrate. Appropriate mixtures are commercially available. They
contain difluoro-substituted liquid crystals of negative dielectric anisotropy, see
Table 3.3.
9 References
1 G. H. Heilmeier, L. A. Zanoni and L. Barton, Proc. IEEE, 1968,56,1162.
2 W. Helfrich, J. Chem. Phys., 1969,51,4092.
3 W. Helfrich, Mol. Cryst. Liq. Cryst., 1973,21,187.
4 G. H. Heilmeier and W. Helfrich, Mol. Cryst. Liq. Cryst., 1970,16, 155.
5 G. Durand, M. VeyssiC, F. Rondelez and L. Leger, C. R. Acad. Sci. Paris, 1970,
270B,97.
6 V.Zwetkoff, Acta Physiochim. USSR, 1937,6,885.
7 R.Williams, J. Chern. Phys., 1963,39,384.
8 G. Hansen, unpublished dissertation, 1907,Halle, Germany.
9 J. A. Castellano, J. E. Goldmacher, L. A. Barton and J. S. Kane, J. Org. Chem.,
1968,33,3501.
10 A. Aviram, R.J. Cox and W. R. Young, Angew. Chem. Int. Ed. Eng., 1971,10,410.
1 1 J. Billard, J. Jacques, M. Leclerq and J. Malthete, Comptes Rendus. Acad. Sci. C
Paris, 1971,273,291.
12 A. Aviram, I. Haller and W. R. Young, Mol. Cryst. Liq. Cryst., 1971,13,357.
13 R. Steinstrasser, 2. Naturforsch., 1972,276,774.
14 R. Steinstrasser and L. Pohl, 2. Naturforsch., 1971,266,577.
15 H. Kelker and B. Scheurle, Angew. Chem. Int. Ed. Eng., 1969,8,884.
16 C. Weygand and R. Gabler, Ber. Dtsch. Chem. Ges., 1938,71,2399.
17 J. van der Veen, W. H. de Jeu, A. H. Grobben and J. Boven, Mol. Cryst. Liq. Cryst.,
1972,17,291.
18 W. Greubel and U. Wolff, Appl. Phys. Lett., 1971,19,213.
19 D. P.McLemore and E. F. Carr, J. Chem. Phys., 1972,37,3245.
20 E. F. Carr, Mol. Cryst. Liq. Cryst., 1969,7,253.
21 W. H. de Jeu, C. J. Gerritsma and A. M. van Boxtel, Phys. Lett., 1971,34A,203.
22 W. J. A. Goossens, Phys. Lett., 1972,40A,95.
23 W. H. de Jeu and C. J. Gerritsma, J. Chem. Phys., 1972,56,4752.
24 R. B. Meyer, Appl. Phys. Lett., 1968,20,1024.
25 P.G. de Gennes, Solid State Cornrnun., 1968,6,163.
26 R. B. Meyer, Appl. Phys. Lett., 1969,14,208.
27 F. J. Kahn, Phys. Rev. Lett., 1970,24,209.
28 J. J. Wysocki, J. Adams and W. Haas, Phys. Rev. Lett., 1968,20,1024.
29 J. Adams, W. Haas and J. J. Wysocki, Mol. Cryst. Liq. Cryst., 1969,8,9.
30 J. Adams, W. Haas and J. J. Wysocki, Mol. Cryst. Liq. Cryst., 1969,8,471.
31 H . Baessler and M. M. Labes, Phys. Rev. Lett., 1968,21,1791.
32 E.Jakeman and E. P. Raynes, Phys. Lett., 1972,39A,69.
33 P.R. Gerber, 2. Naturforsch., 1982,38A,407.
34 M. F. Schiekel and K. Fahrenschon, Appl. Phys. Lett., 1971,15,391.
35 H. Mailer, K. L. Likins, T. R. Taylor and J. L. Ferguson, Appl. Phys. Lett., 1971,
118, 105.
36 M.Hareng, G.Assouline and F. Leiba, Appl. Opt., 1972,11,2920.
37 F. J. Kahn, Appl. Phys. Lett., 1972,20,199.
38 G.Labrunie and J. Robert, J. Appl. Phys., 1973,44,4869.
39 H-p. Schad, SID '82 Digest, 1982,244.
40 V . FrCedericksz and V. Zolina, Trans. Faraday Soc., 1933,29,919.
41 W.Helfrich, Mol. Cryst. Liq. Cryst., 1973,21,187.
128 Chapter 3
42 L. M. Blinov, in ‘Handbook of Liquid Crystal Research’, Eds. P. J. Collings and
J. S. Patel, Oxford University Press, Oxford, 1997, p. 125.
43 S. T. Wu, Appl. Phys. Lett., 1990,57,986.
44 S. T. Wu, Mol. Cryst. Liq. Cryst., 1991,207, 1.
45 A. Rapini and M. Papoular, J. de Phys. (Paris), 1969,30, C4-54.
46 F. Gharadjedaghi, Disp. Technol., 1986,1,95.
47 R. Steinstrasserand F. Del Pino, Ger. Pat., DE-24 500 088, 1974.
48 T. Inukai, H. Inone, K. Furukawa, S. Saito, S. Sugimori and K. Yokohama, DE-2
937 700,1979.
49 G. W. Gray and S. M. Kelly, Mol. Cryst. Liq. Cryst., 1981,75, 109.
50 M. A. Osman, Mol. Cryst. Liq. Cryst., 1982,82,295.
51 S . M. Kelly and H-p. Schad, Mol. Cryst. Liq. Cryst., 1984,110,239.
52 S . M. Kelly and T. Huynh-Ba, Helv. Chim. Acta, 1983,66, 1850.
53 M. A. Osman and T. Huynh-Ba, Mof. Cryst. Liq. Cryst. Lett., 1983,92, 57.
54 S. M. Kelly and H-p. Schad, Helv. Chim. Acta, 1985,68,813.
55 Hp. Schad and S. M. Kelly, Mof. Cryst. Liq. Cryst., 1986,75, 133.
56 M. A. Osman and T. Huynh-Ba, Mol. Cryst. Liq. Cryst. Lett., 1983,82, 339.
58 M. A. Osman, Helv. Chim. Acta, 1985,68,606.
59 M. Schadt, Proc. Japan Display ’83,Kobe, 1983,220.
60 M. Schadt, M. Petrzilka, P. R. Gerber, A. Villiger and G. Trickes, Mol. Cryst. Liq.
Cryst., 1983,94, 139.
61 M. P. Burrow, G. W. Gray, D. Lacey and K. J. Toyne, 2. Chem., 1986,26,21.
62 R. Eidenschink, 0. Haas, M. Romer and B. S. Scheuble, Angew. Chem., 1984, 96,
151.
63 R. Eidenschink and B. S . Scheuble, Mol. Cryst. Liq. Cryst. Lett., 1986,3,33.
64 V. Reiffenrath, J. Krause, H. J. Plach and G. Weber, Liq. Cryst., 1989,5, 159.
65 V. Reiffenrath, and M. Bremer, Angew. Chem. Int. Ed. Eng., 1994,33,87.
66 S . M. Kelly, Liq. Cryst., 1991,10,261.
67 M. Schadt and W. Helfrich, Appl. Phys. Lett., 1971,18,127.
68 Ch. Maugin, Bull. Soc. Fr. Miniral. Cristallogr. Phys. Z., 191 1,12, 101 1.
69 C. H. Gooch and H. A. Tarry, Electron. Lett., 1974,10,2.
70 L. Pohl, G. Weber, R. Eidenschink, G. Baur and W. Fehrenbach, Appl. Phys. Lett.,
1981,38,497.
71 J. van der Veen, W. H. de Jeu, A. H. Grobben and J. Boven, Mol. Cryst. Liq. Cryst.,
1972, 17,291.
72 E. Jakeman and E. P. Raynes, Phys. Lett., 1972,39A, 69.
73 F. M. Jaeger, Recueil Trav. Chim. Pays-Bas, 1907,26,3 11.
74 G. W. Gray and D. G. McDonnell, Mol. Cryst. Liq. Cryst. Lett., 1977,34,211.
75 G. W. Gray and D. G. McDonnell, Mol. Cryst. Liq. Cryst., 1976,37, 189.
76 E. Froehlich, Dissertation, Halle, 1910.
77 G. H. Heilmeier, L. A. Zanoni and J. E. Goldmacher, in ‘Liquid Crystals and
Ordered Fluids’, Eds. J. F. Johnson and R. S. Porter, Plenum Press, New York,
1970.
78 J. A. Castellano, Ger. Pat. Appl., DE-OS 1 928 003, 1971.
79 L. Pohl and R. Steinstriisser, Ger. Pat. Appl., DE-OS 2 024 269, 1971.
80 R. Steinstrasser, Z. Narurforsch., 1972,27b, 774.
81 A. Boller, H. Scherrer M. Schadt and P. Wild, Proc. IEEE, 1972,60, 1002.
82 G. W.Gray, K. J. Harrison and J. A. Nash, Electron. Lett., 1973,9, 130.
83 A. 1. Pavluchenko, V. V.Titov and N. I. Smirnova, in ‘Advances in Liquid Crystal
Research and Applications’, Ed. L. Bata, Pergamon Press, Oxford, 1980, 1007.
84 L. A. Karamysheva, E. I. Kovshev, A. I. Pavluchenko, K. V. Roitman, V. V. Titov,
S. I. Torgova and M. F. Grebyonkin, Mol. Cryst. Liq. Cryst., 1981,67,241.
85 A. Boller. M. Cereghetti, M. Schadt and H. Scherrer, Mol. Cryst. Liq. Cryst., 1977,
42, 1225.
86 R. Eidenschink, D. Erdmann, J. Krause and L. Pohl, Angew. Chem., 1977,89, 103.
87 A. I. Pavluchenko, N. I. Smirnova, V. F. Petrov, M. F. GrebyonkinandV. V. Titov,
Mol. Cryst. Liq. Cryst., 1991,209, 155.
88 M. F. Grenyonkin, V. F. Petrov and B. I. Ostrovsky, Liq. Cryst., 1990,7,367.
89 V. Petrov, S. I. Torgova, L. A. Karamysheva and S. Tanenaka, Liq. Cryst., 1999,26,
1141.
90 J. Bartulin, C. Zuniga, H. Muller, T. R. Taylor and W. Haase, Mol. Cryst. Liq.
Cryst., 1987,15OB, 237.
91 T. Inukai, H. Inoue and H. Sato, USA Patent, 4 21 1 666, 1978.
92 M. A. Osman and L. Revesz, Mol. Cryst. Liq. Cryst., 1982,82,41.
93 H. Sorkin, Mol. Cryst. Liq. Cryst., 198C, 56,279.
94 V. S. Bezborodov, Zh. Org. Chim., 1989,25,2168.
95 Y. Haramoto and H. Kamogawa, Chem. Lett., 1985,79.
96 G. W. Gray and S. M. Kelly, Angew. Chem. In?. Ed., 1981,20,393.
97 A. Villiger, A. Boller and M. Schadt, 2. Naturforsch., 1979, MB, 1535.
98 G. W. Gray, Abstract, 12‘hArbeitstagung Flussigkristalle, Freiburg, Germany, 1978.
99 R. Eidenschink, D. Erdmann, J. Krause and L. Pohl, Angew. Chem., 1978,90, 133.
100 W. Sucrow and W. Schatull, 2. Naturforsch., 1982,37b, 1336.
101 M. Petrzilka and M. Schadt, US Patent, 4 565 425, 1986.
102 H.-J. Deutscher, F. Kuschel, S. Konig, H. Kresse, D. Pfeiffer, A. Wiegeleben,
J. Wulf and D. Demus, 2. Chem., 1977,17,64.
103 S. Sugimori, H. Sato, T. Inukai and K. Furakawa, Ger. Pat. Appl., DE-OS 3 029
378, 1981.
104 N. Carr, G. W. Gray and D. G. McDonnell, Mul. Cryst. Liq. Cryst., 1983,97, 13.
105 M. Petrzilka, Mol. Cryst. Liq. Cryst., 1984,111, 329.
106 S. M. Kelly, Liq. Cryst., 1991, 10,273.
107 R. Buchecker and M. Schadt, Mol. Cryst. Liq. Cryst., 1987,149, 359.
108 M. Petrzilka and M. Schadt, Mol. Cryst. Liq. Cryst., 1985,131, 109.
109 S. M. Kelly, A. Germann, R. Buchecker and M. Schadt, Liq. Cryst., 1994,16,67.
110 S . M. Kelly, M. Schadt and H. Seiberle, Liq. Cryst., 1995,18, 581.
1 1 1 M. Schadt, R. Buchecker, A. Villiger, F. Leenhouts and J. Fromm, Trans. IEEE,
1986, ED-33, 1187.
112 M. Schadt, M. Petrzilka, P. R. Gerber and A. Villiger, Mol. Cryst. Liq. Cryst., 1985,
122, 241.
113 M. Schadt, R. Buckecker and K. Miiller, Liq. Cryst., 1989,5,293.
114 G. W. Gray and S. M. Kelly, J. Mater. Chem., 1999,9,2037.
115 A. J. Leadbetter, R. M.Richardson and C. N. Collings, J. de Phys., 1975,36,37.
116 Hp. Schad and M. A. Osman, J. Chem. Phys., 1981,75,880; 1983,79,5710.
117 Hp. Schad and S. M. Kelly, J. Chem. Phys., 1984,81,1514.
118 C. J. F. Boettcher and P. Bordwijk, in ‘Theory of Electric Polarisation’, Elsevier,
Amsterdam, Vol. 11, 1978.
119 P. Bordwijk, J. Chem. Phys., 1980,73,595.
120 M. J. Bradshaw and E. P. Raynes, Mol. Cryst. Liq. Cryst., 1983,91, 145.
121 H.-J. Deutscher, B. Laaser, W. Dolling and H. Schubert, J. Prakt. Chem., 1978,320,
194.
122 M. E. Neubert, L. T. Carlino, D. L. Fishel and R. M. D’Sidocky, Mol. Cryst. Liq.
Cryst., 1980,59,253.
123 G. W. Gray, C. Hogg and D. Lacey, Mol. Cryst. Liq. Cryst., 1981,67, 1.
124 M. A. Osman and L. Revesz, Mul. Cryst. Liq. Cryst., 1980,56, 157.
125 G. W. Gray and S. M. Kelly, Mol. Cryst. Liq. Cryst., 1981,75,95.
126 N. Carr, G. W. Gray and S. M. Kelly, Mol. Cryst. Liq. Cryst. Lett., 1985, 1, 53.
128 R. Eidenschink, Kontakte (Darmstadt), 1979,1, 15.
129 R. Eidenschink, M. Romer and F. V.Allen, in ‘Liquid Crystals and Ordered Fluids’,
Eds. A. C. Griffin and J. F. Johnson, Plenum Press, New York, 1984, p. 737.
130 Chapter 3
130 M. A. Osman, 2. Naturforsch., 1983,38a, 693.
131 M. A. Osman, Mol. Cryst. Liq. Cryst., 1982,72, 291.
132 K. Praefcke, D. Schmidt and G. Heppke, Chem. Zeit., 1980,104,269; Is. J. Chem.,
1979,18, 195.
133 R. Eidenschink, Mol. Cryst. Liq. Cryst., 1985,113,57.
134 R. Eidenschink, M. Romer, G. Weber, G. W. Gray and K. J. Toyne, German Patent,
DE 3321373,1983.
135 N. Carr and G. W. Gray, Mol. Cryst. Liq. Cryst., 1985,124,27.
136 H . Takatsu, K. Takeuchi and H. Sato, Japan Display '83,1983,228.
137 H.-J. Deutscher, M. Korber and H. Schubert, 'Advances in Liquid Crystal Research
and Application', Ed. L. Bata, Pergamon Press, Oxford, 1980, p. 1075.
138 E. Poetsch, V. Reiffenrath and B. Rieger, 24'h Arbeitstagung Flt;issigkristalle,
Freiburg, Germany, 1995.
139 B. S . Scheuble, Proc. 161h Freiburger Arbeitstagung Flussigkristalle, Freiburg,
Germany, 1987.
140 H . Takatsu, K. Takeuchi and H. Sato, Mol. Cryst. Liq. Cryst., 1983,100, 345.
141 G. W. Gray and D. G. McDonnell, Mol. Cryst. Liq. Cryst., 1979,53, 147.
142 R. Eidenschink and M. Romer, Proc. 131h Freiburger Arbeitstagung Flussigkristalle,
Freiburg, Germany, 1983.
143 U . Finkenzeller, A. Kunneier and E. Poetsch, Proc. 18IhFreiburger Arbeitstagung
Flussigkristalle, Freiburg, Germany, 1989.
144 E. Bartmann, D. Dorsch and U. Finkenzeller, U. Kurmeier and E. Poetsch, Proc.
191hFreiburger Arbeitstagung Flussigkristalle,Freiburg, Germany, 1990.
145 B. Rieger, E. Poetsch and V. Reifenrath, Proc. 19'h Freiburger Arbeitstagung
Flussigkristalle,Freiburg, Germany, 1990.
146 A. I . Pavluchenko, N. I. Smirnova, V. F. Petrov, Y. A. Fialkov, S. V. Shelyazhenko
and L. M. Yagupolsky, Mol. Cryst. Liq. Cryst., 1991,209,225.
147 E. Bartmann, D. Dorsch and U. Finkenzeller, Mol. Cryst. Liq. Cryst., 1991,204,77.
148 Y. Goto, T. Ogawa, S. Sawada and S. Sugimori, Mol. Cryst. Liq. Cryst., 1991,209,
1.
149 Y . Onji, M. Ushioda, S. Matsui, T. Kondo and Y. Goto, Eur. Pat. Appl., EPA 0 647
696 Al, 1993.
150 K. Tarumi, M. Bremer and T. Geelhaar, Ann. Rev. Sci., 1997,27,423.
151 H. Takatsu, K. Takeuchi, M. Sasaki and H. Onishi, Mol. Cryst. Liq. Cryst., 1991,
206, 159.
152 E. Bartmann, J. Krause and K. Tarumi, Proc. 23'd Freiburger Arbeitstagung
Flussigkristalle,Freiburg, Germany, 1994.
153 M . Bremer, Adv. Muter., 1995,7, 867.
154 R. Buchecker, G. Marck and M. Schadt, Mol. Cryst. Liq. Cryst., 1995,260,93.
155 E. Bartmann, Adv. Muter., 1996,8,570.
156 M . Goulding, S. Greenfield, 0. Parri and D. Coates, Mol. Cryst. Liq. Cryst., 1995,
265,27.
157 S. Greenfield, D. Coates, M. Goulding and R. Clemitson, Liq. Cryst., 1995,18,665.
158 D. W. Berreman and W. R. Heffner, J. Appl. Phys., 1985,122,l.
159 D. W. Berreman and W. R. Heffner, J. Appl. Phys., 1981,52,3032.
160 R. N. Thurston, Mol. Cryst. Liq. Cryst., 1985, 122, 1.
161 C. M. Waters, E. P. Raynes and V. Brimmell, Japan Display '83, 1983,396.
162 C. M. Waters, E. P. Raynes and V. Brimmell, Proc. SID '84, 1984,25,261.
163 C. M. Waters, E. P. Raynes and V. Brimmell, Mol. Cryst. Liq. Cryst., 1985,123,303.
164 E. P. Raynes, Mol. Cryst. Liq. Cryst. Lett., 1986,4, 1.
165 E. P. Raynes and C. M. Waters, Displays, 1987,59.
166 E. P. Raynes, in 'The Optics of Thermotropic Liquid Crystals', Eds. S. Elston and
R. Sambles, Taylor and Francis, London, 1998, p. 289.
167 T. J. Scheffer and J. Nehring, Appl. Phys. Lett., 1984,45, 1021.
168 T. J. Scheffer, J. Nehring, M. Kaufmann, H. Amstutz, D. Heimgartner and P. Eglin,
SID '85 Digest, 1985, 120.
169 T. J. Scheffer and J. Nehring, Ann. Rev. Sci., 1997,27,555.
170 N. Kando, T. Nakagomi and S . Hasagawa, Ger. Pat. Appl., DE 3403259,1985.
171 M. Schadt and F. Leenhouts, J. Appl. Phys., 1987,50,236.
172 F. Leenhouts and M. Schadt, Mol. Cryst. Liq. Cryst., 1988,158B, 241.
173 F. Moia, M. Schadt and H. Seiberle, Proc. SID '93, 1993, 1.
174 P. M. Alt and P. Pleshko, IEEE Trans. Electron. Dev., 1974,21, 146.
175 T. J. Scheffer, B. Clifton, D. Prince and A. R. Connor, Displays, 1993,14,74.
176 V. G. Chigrinov, V. V. Belyaev, S. V. Belyaev and M. F. Grebenkin, Sov. Phys.
4JETP, 1979,50,994.
177 A. Gobl-Wunsch, G. Heppke and F. Oestereicher, J. Phys. (Paris), 1979,40,773.
178 S . Naemura, T. Oyama, H. Plach, G. Weber and B. Scheuble, E. Merck, Liquid
Crystal Seminar Data.
179 D. W. Berreman, Appl. Phys. Lett., 1974,25, 12.
180 N. Kimura, T. Shinomiya, K. Yamamoto, Y. Ichimura, K. Nakegawa, Y. Ishi and
M. Matsuura, Proc. SID '88, 1988,49.
181 H. Watanabe, 0. Okumura, H. Wada, A. Ito, M. Yazaki, M. Nagata, H. Takeshita
and S . Morozumi, Proc. SID '88, 1988,416.
182 P. J. Bos, Proc. SID '94, 1994, 118.
183 M. Schadt and P. R. Gerber, 2. Naturforsch., 1982,37a, 165.
184 W. H. de Jeu and W. A. P. Claasen, J. Chem. Phys., 1977,67,3705.
185 Hp. Schad, G. Baur and G. Meier, J. Chem. Phys., 1979,70,2770.
186 M. Schadt, Liq. Cryst., 1993,14,73.
187 V. Petrov, S. I. Torgova, L. A. Karamysheva and S . Tanenaka, Liq. Cryst., 1999,26,
1141.
188 M. Schadt and F. Muller, Rev. Phys. Appliquke, 1979,14,265.
189 J. Constant and E. P. Raynes, Mol. Cryst. Liq. Cryst., 1980,62, 115.
190 M. Schadt and P. R. Gerber, Z. Naturforsch., 1982,37a, 179.
191 J. Constant and E. P. Raynes, Mol. Cryst. Liq. Cryst., 1980,62, 115.
192 M. Schadt and M. Petrzilka, Japan Display '84, 1984,53.
193 R. G. Priest, Phys. Rev., 1973, A7, 720.
194 J. P. Straley, Phys. Rev., 1973, AS, 2181.
195 M. Petrzilka, Mol. Cryst. Liq. Cryst., 1985,131, 109.
196 H. Ohnishi, H. Takatsu, K. Takeuchi, F. Moia and M. Schadt, SID '92 Digest, 1982,
17.
197 K. Praefcke, D. Schmidt and G. Heppke, Chem. Zeit., 1980,104,269.
198 E. Poetsch, V. Reiffenrath and B. Rieger, 24'h Arbeitstagung Flussigkristalle,
Freiburg, Germany, 1995, PO 1.
199 M. Petrzilka, R. Buckecker, S.Lee-Schmiederer, M. Schadt and A. Germann, Mol.
Cryst. Liq. Cryst., 1987, 148,123.
200 M. Petrzilka, USA Patent, 4676604, 1987.
201 M. Schadt, R. Buckecker and A. Villiger, Liq. Cryst., 1990,79,519.
202 S. M. Kelly, A. Germann and M. Schadt, Liq. Cryst., 1994, 16,491.
203 S. M. Kelly, Liq. Cryst., 1994,17,211.
204 R. Eidenschink, G. W. Gray, K. J. Toyne and A. E. F. Waechtler, Mol. Cryst. Liq.
Cryst. Lett., 1988,5, 177.
205 S. Kawakami, K. Kotani, S. Shirokura, H. Ohnishi, Y. Fujita, T. Ohtsuka,
H. Takatsu and F. Moia, Proc. SID '92, Japan Display, 1992,487.
206 H. Seiberle, F. Moia,, M. Schadt, H. Takatsu and K. Kotani, 2OrhFreiburger
Arbeitstagung Flussigkristalle, Freiburg, Germany, 1992.
207 L. M. Blinov and V. G. Chigrinov, in 'Electro-optic Effects in Liquid Crystal
Materials', Springer, New York, 1994.
208 M. Schadt, Ann. Rev. Mater. Sci., 1997,27,305.
209 D. Coates, E. Merck, private communication.
132 Chapter 3
210 D. Dunmur, in ‘Physical Properties of Liquid Crystals’, Eds. D. Demus, J. W.
GmbH, Weinheim, Germany, 1999.
21 1 G. H. Heilmeier and L. A. Zanoni, Appl. Phys. Lett., 1968,13,91.
212 G. H. Heilmeier, J. A. Castellano and L. A. Zanoni, Mol. Cryst. Liq. Cryst., 1969,8,
293.
213 D. L. White and G. N. Taylor, J. Appl. Phys., 1974,45,4718.
214 T. J. Scheffer and J. Nehring, in ‘The Physics and Chemistry of Liquid Crystal
Devices’, Ed. G. J. Sprokel, Plenum Publishing, New York, 1980, p. 173.
215 T. Uchida, H. Seki, C. Shishido and M. Wada, IEEE, Trans. Elec. Dev., 1979, ED-
26, 1373.
216 T. Uchida, H. Seki, C. Shishido and M. Wada, Mol. Cryst. Liq. Cryst., 1979, 54,
161.
217 D. Demus, B. Krucke, F. Kuschel, H. U. Nathninck, G. Pelzl and H. Zaschke, Mol.
Cryst. Liq. Cryst., 1979,56, 1 15.
218 P. G. de Gennes, Solid State Commun., 1968,6, 163.
219 M. Morita, S. Imamura and J. K. Yatabe, J. Appl. Phys., 1975,14,315.
220 F. Gharadjedaghi, Mol. Cryst. Liq. Cryst., 1981,68, 127.
221 F. Fischer, Z. Naturforsch., 1976,31a, 41.
222 T. J. Scheffer, in ‘Nonemissive Electro-optic Displays’, Eds. A. R. Kmetz and F. K.
von Willisen, Plenum Press, New York, 1976, p. 45.
223 F. Gharadjedaghi, J. Appl. Phys., 1983,54,4989.
224 I. C. Sage, in ‘Handbook of Liquid Crystals’, Eds. D. Demus, J. W. Goodby, G. W.
Gray, H.-W. Spiess and V. Vill, Wiley-VCH VerlagsgesellschaftGmbH, Weinheim,
Germany, 1998, Vol. 1, p. 731.
225 G. W. Gray, Dyes Pigments, 1982,3,203.
226 A. Bloom and P. L. K. Hung, Mol. Cryst. Liq. Cryst., 1977,40,213.
227 R. J. Cox, Mol. Cryst. Liq. Cryst., 1979,55, 1.
228 M. A. Osman, T. J. Scheffer and H. R. Zeller, ntz Archiv., 1979,8,185.
229 A. C. Lowe, Mol. Cryst. Liq. Cryst., 1980,59,63.
230 L. M. Blinov, V. A. Kizel, V. G. Rumyantsev and V. V. Titov, J. de Phys., Coll, C2,
1975,36,69.
231 F. Jones and T. J. Reeve, Mol. Cryst. Liq. Cryst., 1980,60,99.
232 A. C. Lowe and R. J. Cox, Mol. Cryst. Liq. Cryst., 1980,59, 63,
233 F. Jones and T. J. Reeve, Mol. Cryst. Liq. Cryst., 1980,60,99.
234 J. Cognard and T. Hieu Phan, Mol. Cryst. Liq. Cryst., 1981,68,207.
235 T. Uchida, C. Shishido, H. Seki and M. Wada, Mol. Cryst. Liq. Cryst. Lett., 1977,
34, 153; Mol. Cryst. Liq. Cryst., 1977,39, 39.
236 R. Eidenschink, Kontakte (Darmstadt), 1984,2,25.
237 W. Maier and A. Saupe, 2. Naturforsch., 1958, 13a, 564; 1959, 14a, 882; 1960, Ha,
287.
238 D. Vorlander, Trans. Faraday SOC.,1933,29,907.
239 J. Constant, E. P. Raynes, I. A. Schanks, D. Coates, G. W. Gray and D. G.
McDonnell, J. Phys. D: Appl. Phys., 1978, 11,479.
240 J. Constant, M. G. Pellatt and I. H. C. Roe, Mol. Cryst. Liq. Cryst., 1980,59,299,
241 M. Schadt and P. Gerber, Mol. Cryst. Liq. Cryst., 1981,65,241.
242 J. Cognard and T. Hieu Phan, Mol. Cryst. Liq. Cryst., 1981,70, 1.
243 S. Aftergut and H. S. Cole, Appl. Phys. Lett., 1981,38, 599.
244 G. Heppke, B. Kippenberg, A. Moller and G. Scherowsky, Proc. Euro Display ’82,
1981,25.
245 N. Bastiirk, J. Cognard and T. Hieu Phan, Mol. Cryst. Liq. Cryst., 1983,95,71.
246 G. Heppke, B. Kippenberg, A. Moller and G. Scherowsky, Mol. Cryst. Liq. Cryst.,
1983,94, 191.
247 F. C . Saunders, K. J. Harrison, E. P. Raynes and D. J. Thompson, Proc. SID ’82,
1982,121.
248 F. C. Saunders, D. J. Thompson K. J. Harrison and E. P. Raynes, R S R E News Lett.
Res. Rev., 1982,44, 1.
249 J. Cognard ,T. Hieu Phan and N. Basturk, Mol. Cryst. Liq. Cryst., 1983,91,327.
250 T. J. Scheffer, J. Appl. Phys., 1982,53,257.
251 B. S. Scheuble, G. Weber, L. Pohl and R. E. Jubb, SID '83 Digest, 1983,176.
252 D. Demus, B. Krucke, F. Kuschel, H. U. Nothnick, G. Pelzl and H. Zaschke, Mol.
Cryst. Liq. Cryst. Lett., 1979, 56, 115; D. Demus and H. Zaschke, Mol. Cryst. Liq.
Cryst., 1981,63, 129.
253 A. Isenberg, B. Krucke, G. Pelzl, H. Zaschke and D. Demus, Cryst. Res. Technol.,
1983,18,1059.
254 K. Praefcke and D. Schmidt, 2. Naturforsch., 1981,36b, 375.
255 H. Enzenberg, D. Schadler and H. Zaschke, 2. Chem., 1982,22,59.
256 A. E. Bradfield and B. Jones, J. Chem. SOC.[London], 1929,2660.
257 G. W. Gray and B. Jones, J. Chem. SOC. [London], 1959,1467.
258 R. Kiefer, B. Weber, F. Windscheid and G. Baur, Japan Display '92, 1992,547.
259 G. Baur, Proc. 2YdFreiburger Arbeistagung Flussigkristalle, Freiburg, Germany,
1993.
260 G. Baur, Proc. 23rd Freiburger Arbeistagung Flussigkristalle, Freiburg, Germany,
1994.
261 G. Baur, M. Kamm, R. Kiefer, H. Klausmann, B. Weber, B. Wieber and F.
Windscheid, 231d Freiburger: Arbeistagung Flussigkristalle, Freiburg, Germany, 1994.
262 R. A. Soref, J. Appl. Phys., 1974,45,5466.
263 S. Kobayashi, SID Digest, 1972,68.
264 M . 0-he, M. Ohta and K. Kondo, Asia Display '95, 1995,707.
265 M. Ohta, K. Kondo and M. 0-he, IEICE Trans. Electron., 1996, E79-C.
266 M. Ohta, M. Yoneya and K. Kondo, EuroDisplay '96, 1996,49.
267 M. Ohta, M. Yoneya and K. Kondo, Proc. SID '96, San Diego, 1996.
268 G. Baur, Proc. SID-MID Eur. Meeting, Darmstadt, Germany, 1996.
CHAPTER 4
Photoluminescence and
Electroluminescencefrom
Organic Materials
1 Introduction
Light-emitting diodes (LEDs) are flat-panel display devices, which emit light
under the action of an electric current passing through an electroluminescent
emissive layer. Electroluminescence in inorganic semiconductors was dis-
covered before the corresponding effect in organic materials was found. 1*2
Consequently the first commercial alpha numeric display devices using electro-
luminescence used inorganic semiconductor materials, such as GaAsP, in the
early 1960s. Simple monochrome inorganic semiconductor LEDs with low
information-content are manufactured on a large scale and are found in many
electronic instruments, see Chapter 7.
Each pixel is driven directly with two electrodes or by active matrix
addressing. This renders high-information-content displays, such as portable
computers with up to 3 000 000 pixels, either impractical or very expensive to
realise. Even LEDs with a relatively low-information-content are difficult to
fabricate cheaply due to the requirement for individual dedicated pixel elec-
trodes. Furthermore, full colour displays would require arrays of alternating
pixels of the three primary colours. These would have to be fabricated
individually. There remains, therefore, an enormous potential for light-emitting
diodes using organic materials due to their advantageous combination of
physical properties, above all their ease of processing for displays of almost
any size and an almost infinite possibility for modification of the electro-optical
properties by suitable design and chemical synthesis. Furthermore, high-
information-content LEDs using organic materials can be addressed using
either direct addressing, multiplex addressing or active matrix addressing as
for LCDs, see Chapter 2.
Electroluminescence from organic materials was first discovered using solid
anthracene crystals immersed in a liquid Very high voltages, e.g.
400-2000 V, were necessary in order to observe electroluminescence partly due
134
Photoluminescence and Electroluminescencefrom Organic Materials 135
to the thickness of the crystals used ( ~ mm). 5 Considerably lower operating

voltages were found using solid electrodes.' However, the decisive break-
through was the use of thin, vapour-deposited films and solid
Unfortunately, the efficiency of electroluminescent devices using one layer of
light-emitting material sandwiched between two electrodes was very low.
Consequently there was little activity in the area of organic electroluminescence
for many years until the report by Tang and Van Slyke of the Eastman Kodak
Corporation of two-layer devices with a much higher efficiency and lower
operating voltage. 12?' Even higher efficiencies were found for three-layer
devices.14-22 However, despite development over three decades organic light-
emitting diodes (OLEDs) are only set now to make a decisive commercial
breakthrough.20 The critical factors inhibiting mass production of OLEDs
using small molecules have been too short a display lifetime and insufficient
reliability, rather than performance.
Conjugated organic polymers have been prepared and their physical proper-
ties studied for several decades. They may exhibit a range of physical properties,
such as high electrical conductivity21922and photocond~ctivity,~~ charge
photolumine~cence~~ and nonlinear optical response,26 of interest
for a variety of electronic and optoelectronic applications, such as semiconduc-
tors, sensors and telecommunications.27They have been studied intensively due
to their ability to combine the optoelectronic properties of inorganic semicon-
ductors with some typical properties of organic polymers, such as ease of
processing and patterning, mechanical flexibility and, to some extent, rugged-
ness. However, it is the property of electroluminescence that is of the highest
interest at the present time.28-35Electroluminescent organic polymers possess a
number of physical properties compared to inorganic or low-molar-mass
electroluminescent materials. Electroluminescent inorganic semiconductors
and low-molar-mass organic electroluminescent materials have to be deposited
by sublimation or vapour deposition under high vacuum. This is an expensive
process and not ideally suited to the mass production of large-area displays. In
contrast, uniform thin films of conjugated organic polymers can be deposited
from dilute solutions in suitable solvents by the process of spin coating or
Doctor blade deposition. Organic polymers form stable glasses, which freeze in
the desired structural order without the formation of defects at grain boundaries
formed in the process of crystallisation. Furthermore, the production of
copolymers from a mixture of different monomers allows the fine-tuning of the
physical properties of interest.
The first reports of electroluminescence from an organic polymer failed to
attract much However, the report of electroluminescence from
poly(pphenyleneviny1ene) (PPV) by researchers from the Cavendish laboratory
in Cambridge, UK,38y39 and later from a related derivative by researchers from
Santa Barbara, USA:o34' stimulated a great deal of interest and activity in
OLEDs using light-emitting polymers (LEPs). The semiconductor properties of
organic conjugated polymers stem from the formation of a n bonding molecular
orbital, which corresponds to the highest occupied molecular orbital (HOMO),
and a n* antibonding molecular orbital, which is equivalent to the lowest
136 Chapter 4
unoccupied molecular orbital (LUMO). These energy levels are often referred
to as a n valence band and a n* conduction band in analogy with the
nomenclature used to describe inorganic semiconductors, such as silicon.
The overlap of the p-atomic orbitals of sp2 hybridised atoms, such as carbon
or nitrogen, making up the conjugated polymer backbone, gives rise to a
substantial degree of delocalisation of electron density. The alternating con-
figuration of conjugated single and double bonds in conjugated polymers gives
rise to their semiconductor properties, but also results in a rigid, linear and
planar molecular conformation. This tends to give a very high melting point and
a very limited solubility in all organic solvents.
Low-molar-mass materials and oligomers with a high degree of conjugation
and a propensity to self-assembly are of increasing interest2* for such applica-
tions due to their monodisperse character, flexibility of synthesis, good
processability as thin solid films, etc. An additional advantage is the improved
properties observed due to the propensity of certain classes of these organic
materials to self-organise in a supramolecular structure on the desired substrate
surface. A serious drawback associated with some low-molar-mass materials
and to a lesser extent oligomers is the tendency to form defects and traps at
crystal grain boundaries over time.
2 Photoluminescence from Organic Materials

Electroluminescence and photoluminescence are related physical phenomena.
Electroluminescence from organic semiconductor materials is the process
whereby light is emitted from a layer of an organic material on passage of a
current of electricity through it. In order to understand electroluminescence a
brief description of photoluminescence first may be of help. Photoluminescence
is the process whereby light is absorbed by the organic compound and then re-
emitted, usually at a longer wavelength, see Figure 4.1. These materials are often
referred to as being fluorescent or phosphorescent. The former is a faster
process than the latter. Absorption of light by a conjugated organic molecule
can give rise to an n-n* or n-n* transition and promote an electron to the
LUMO. The energy levels are broadened by vibration and rapid vibronic
relaxation to the bottom of the LUMO is followed by radiative decay to the
ground state, see Figure 4.1. Hence, the photoluminescence absorption and
emission spectra are broad bands, as shown in Figure 4.2, with the latter red-
shifted to longer wavelengths. The electroluminescence and photoluminescence
emission spectra of a particular organic material are almost always identical,
since they originate from the same energy levels. The only difference is the
method of excitation of an electron from the HOMO to the LUMO. In a
molecular solid, intermolecular interactions and disorder also broaden the
transitions. In conjugated organic molecules and polymers, emission is often in
the visible region of the electromagnetic spectrum. The excited carriers are
mobile: intermolecular excitation transfer occurs via neutral excitons or
polarons in conjugated polymers. Excitons consist of bound electron-hole
Photoluminescence and Electroluminescence from Organic Materials 137
EXCITED STATE
Eabs Eem = huetn Eabs
GROUND STATE
Figure 4.1 Schematic representation of the mechanism of photoluminescence from a

conjugated organic material. Absorption of a photon of a given energy can
lead to promotion of an electron into one of the energy levels of the excited
state. Relaxation from the lowest excited energy state with the highest
population gives rise to emission at a lower frequency than that, Dabs, of the
absorbed photon. Thefrequency, u,, of light emitted of maximum intensity is
related to the band gap, E,,, between the lowest energy levels of the ground
state and excited state ( E = hu) where h is Plank’s constant). This is
representative from the Franck-Condon principle.
0.4
0.2
0.0 1
300 350 400 450 500 550 600 700
WAVELENGTH nm
Figure 4.2 Absorption and emission spectra of aluminium tris (quinolate) [Alqj].
138 Chapter 4
pairs in the LUMO and HOMO, respectively, whereas polarons are single
carriers screened by lattice relaxation.
3 Electroluminescencefrom Organic Materials

Organic light-emitting diodes (OLEDs) have been constructed using a variety of
classes of electroluminescent organic materials, such as low-molar-mass organic
compounds, oligomers, side-chain polymers, polymer blends, conjugated main-
chain polymers and cross-linked polymer networks. These OLEDs have a
similar configurations, see Chapters 5 and 6. However, some of the common
features and characteristics of OLEDs using small molecules and polymers will
be discussed here in order to avoid repetition in the following two chapters. One
or more layers of organic material are situated between two electrodes of
different materials and work functions, see Figure 4.3. Under a forward bias
(voltage) radical anions are generated by injection of electrons into the organic
material from a low-work-function cathode, e.g. an electrode made from
aluminium or calcium. At the same time holes are injected as radical cations
into the organic material from the anode, i.e. an electrode with a high work
function, such as indium/tin oxide. The electrons and holes travel through the
organic material in opposite directions towards the electrode of opposite
polarity under the influence of the electric field. This construction creates a
diode because the current flowing between the cathode and the anode under
positive bias (voltage) is orders of magnitude higher than that flowing in the
opposite direction under negative bias, i.e. if the direction of the electric field is
reversed. The rectification ratio of monolayer and multilayer OLEDs using
either electroluminescent small (low-molar-mass) molecules or LEPs is usually
very high. Although very little current flows under negative bias and no light is
emitted, AC driving, rather than DC driving, can lead to longer device lifetimes
due to the reduction in the build-up of space charge at the interface between one
of the electrodes and the adjacent organic material.
The electrons and holes flowing between the cathode and the anode in
opposite directions may collide on route and then form an excited state, i.e. an
exciton. Upon relaxation of the excited state to the ground state, light is emitted,
if the excited state is a singlet. In the singlet excited state the spin states of the
CATHODE
EMITTER
TRANSPARENT
ANODE I I
TRANSPARENT
SUBSTRATE
hu hu hu
Figure 4.3 Schematic representation of a simple monolayer organic light-emitting diode

(OLED) incorporating an electroluminescent material between a transparent
anode and a cathode.
two electrons in the LUMO and HOMO energy levels are of opposite sign,
whereas in the triplet excited state the spin states of the two electrons are of
equal sign. Triplets usually relax to the ground state by a non-radiative
pathway. However, since the probability of triplet formation to singlet forma-
tion for excitons produced by the recombination of a hole and an electron is 3: l,
the maximum efficiency of electroluminescence from the excited states of
organic materials is equal to 25%.
Some authors claim42that in polymers, recombination occurs between free
charged polarons, so that the maximum efficiency corresponds to the internal
quantum efficiency, “lint (i.e. the number of photons emitted for every electron
injected), as given by
where $F is the quantum efficiency of fluorescence, V R is the efficiency of singlet

formation and y is a double charge injection factor. The latter parameter will
also depend on the electrodes and the match of the electrode work functions
with the energy, levels of the adjacent organic material. However, a more
relevant parameter is the external quantum efficiency, qeXt, as given by the
following relationship:4143
“lint
“lext =3 (2)
where n is the refractive index of the organic material. Therefore, the external
quantum efficiency is substantially lower than the internal quantum efficiency,
since the refractive index of organic compounds is often relatively high, e.g.
n = 1.4 for a conjugated organic polymer. This suggests that a lot of the light
generated is lost within the device itself due to internal reflection and re-
absorption leading primarily to non-radiative decay. Indeed most of the light
is lost ( 285%). The power efficiency, qE, is the ratio of light output power to
electrical input power and is related44945to the operating voltage, Vop, and the
energy of the emitted phonons, Ep in eV, by the equation
If the power efficiency, qE, in lumens per watt (Im W-’) is high, then the OLED
is bright at low current density and applied low voltage. A high power efficiency
generally corresponds to a more extended device lifetime than that of compar-
able devices with the same brightness, but higher operating voltages and power
consumption. Low voltage and power consumption are also pre-requisites for
battery-operated, portable devices. The luminous efficiency in lumens per watt
(Im W-’) is the product of the power efficiency, VE, and the luminous efficacy,
which compensates for the wavelength dependency of human eye ~ e n s i t i v i t y . ~ ~
This is at a maximum for yellow-green light in the centre of the visible part of
the electromagnetic spectrum (ca. 550 nm). The brightness of a display is also
140 Chapter 4
10-2
10-3
10-4
h 10-5
i
3
u
10-6
10-7
10-8
10-9
0.01 0.10 1.oo 10.00

APPLIED VOLTAGE (V)
Figure 4.4 Typical plot of the current, I, against voltage, V , of an organic light-emitting
diode (OLED).
often expressed in terms of the observed luminance in candelas per unit area
(cd m-2).
Figure 4.4 depicts a typical plot of the current, I, against voltage, V, for an
OLED. A threshold voltage can be seen by a sharp change in the steepness of the
I-V curve. The flow of electric current and, therefore, electroluminescence, takes
place above this threshold voltage. The barriers to charge injection into the
organic material must first be overcome by an appropriate electric field strength.
The current density and intensity of light emitted then increase rapidly with
increasing applied voltage. Plots of electric current versus voltage (I-V) curves
are characteristic of a given OLED. It was found that multilayer OLEDs could
significantly reduce operating voltage and power consumption, while, at the
same time, increasing the luminous efficiency. In a bilayer OLED consisting of
two layers of different organic materials, the LUMO of an electron transport
layer (ETL) is matched to the work function of the cathode, while the HOMO of
a hole transport layer (HTL) is matched to the work function of the anode. This
leads to a more balanced injection of holes and electrons. Recombination then
takes place at the interface between the two organic layers away from the
electrodes where quenching can take place, see Figures 4.5 and 4.6 for the energy
diagrams of monolayer and bilayer OLEDs, respectively. Recombination and
light emission then takes place in either one of the layers near the organic layer
interfa~e.~~-’l If a third layer of emissive material is incorporated between the
ETL and HTL layers to produce a three-layer OLED, then the power efficiency
may be even higher and operating voltages lower. Electroluminescence then
takes place in the central emission layer where the holes and electrons recombine.
Figure 4.5 Schematic energy kvel diagram of a generalised monolayer organic light-
emitting diode (OLED) .34
---- - _ - - -VACUUM LEVEL

- - -1---------
1
C-@CATHODE
%node
ANODE @-
Figure 4.6 Schematic energ level diagram of a generalised bilayer organic light-emitting
diode (OLED).y4
It is evident from the above discussion that the threshold voltage, current
density, power efficiency, luminous efficiency and, to some extent, device
lifetime of OLEDs using organic low-molar-mass compounds, oligomers and
polymers depends on intrinsic molecular properties, such as HOMO and
LUMO energy levels, efficiency of hole and electron injection and subsequent
transport, efficiency of singlet formation and fluorescence efficiency. The
142 Chapter 4
wavelength and bandwidth of the emitted light is also of critical importance. All
of these bulk physical material properties can be influenced by the design and
synthesis of suitable organic molecules and their subsequent combination in
single and multilayer OLEDs.
The efficiency of injection of holes and electrons into the organic material in
an OLED is dependent on the nature of the cathode and anode and how closely
they match their work functions with the LUMO and HOMO, respectively, of
the organic material in the centre of the device. There is some degree of
flexibility of device design with respect to matching these parameters by
judicious choice of electrode materials and organic compounds.
The cathode is usually made of a reactive metal such as calcium or aluminium
with a low work function. Calcium has a lower work function than aluminium,
but aluminium is less reactive in the presence of moisture and air. Metal
electrodes and conjugated organic material, such as conjugated polymers, are
both rapidly oxidised by traces of oxygen and water. The result is a rapid decay
in device efficiency and eventually complete device failure after a relatively short
time. 952-54 Therefore, effective encapsulation of OLEDs is generally required in
order to exclude oxygen and water and produce acceptable device operational
lifetimes. Other metals, such as gold and mixed metal/metal oxides, such as
aluminium/aluminium oxide, have also been investigated for use as the cathode
~~OLEDS.~'
The anode may be a standard electrode material such as indium/tin oxide
(ITO) or a conducting organic material, such as sulfonated poly(ani1ine)
(PANi), or poly(3,4-ethylenedioxythiophene) (PEDOT) doped with sulfonic
acid, see Figure 4.7.5560Organic electrodes, such as PEDOT, may be mechani-
cally rubbed to form an alignment layer for liquid crystalline materials,
although the alignment quality is generally poor. The uniform parallel align-
ment of the director in the azimuthal plane can then result in polarised
photoluminescence and electroluminescence. A combination of an IT0 elec-
trode and PEDOT can result in a much higher power efficiency than that using
I T 0 alone as the electrode.
poly(3,4-ethylenedioxythiophene): PEDOT
poly(aniline) salt: PANi with counterion

Figure 4.7 Molecular structure of two typical organic electrode materials.
Photoluminescence and Electroluminescence from Organic Materials 143
OLEDs can be addressed in a similar fashion to LCDs, see Chapter 2, i.e.

directly with segmented electrodes, see Figure 4.8, by multiplexed addressing
with rows and columns of electrode strips, see Figure 4.9 and by active matrix
addressing with one transistor at each pixel, see Figure 4.10. The major
TRANSPARENT SEGMENTED
SUBSTRATE METAL
/
PLATE CONTACT+ HOLE EMISSION ELECTRON
ANODE PROTECT~ONTRANSPORT LAYER TRANSPORT
LAYER LAYER-HTL LAYER - ETL
Figure 4.8 Schematic representation some of the elements of a generalised multilayer

organic light-emitting diode (OLED) with direct addressing. The thin metallic
cathode segments are connected directly to the electron-transport layer
( E T L ) . The impermeable encapsulation is not shown.
TRANSPARENT METAL
SUBSTRATE CATHODE
IT0 CONTACT+ HOLE EMISSION ELECTRON

ANODE PROTECTION TRANSPORT LAYER TRANSPORT
ROWS LAYER LAYER-HTL LAYER - ETL

organic light-emitting diode (OLED) with multiplex addressing. The thin
metallic cathode strips are connected directly to the electron-transport layer
( E T L ) . The impermeable encapsulation is not shown.
144 Chapter 4
TRANSPARENT
SUBSTRATE
/
ANODE CONTACT+ HTL EMISSION ETL METAL
TFr PROTECTION LAYER PLATE
PIXELS LAYER CATHODE
organic light-emitting diode (OLED) with active matrix addressing using
thinfilm transistors. The impermeable encapsulation is not shown.
difference between the addressing of LCDs and OLEDs is that a current of

several flows between the electrodes in an OLED. Therefore, the
power consumption is significantly higher for OLEDs than that of a comparable
LCD of similar size and complexity with the same method of addressing. Active
matrix addressing of OLEDs is very similar to that used for LCDs. Indeed
active matrix addressing using thin-film transistors was originally designed to
drive LEDs using inorganic semiconductors, before being adapted for use with
large-area, high-information-content LCDs. However, OLEDs require high-
power current drivers in order to provide sufficient current to generate electro-
luminescence.
4 References
1 J. R. Sheats, H. Antoniadis, M. Hueschen, W. Leonard, J. Miller, R. Moon, D.
Roitman and A. Stocking, Science, 1996,273,884.
2 E. W. Williams and R. Hall, in ‘Luminescence and the Light Emitting Diode’,
Pergamon Press, Oxford, UK, 1978.
3 W. Helfrich and W. G. Schneider, Phys. Rev. Lett., 1965,14,229.
4 W. Helfrich and W. G. Schneider, J. Chem. Phys., 1966’44’2902.
5 W. Mehl and W. Bucher, 2. Phys. Chem., 1965,47,76.
6 D. F. Williams and M. Schadt, Proc. IEEE, 1970,58,476.
7 J. Dresner, RCA Rev., 1969,30, 322.
8 R. E. Kellogg, J. Chem., Phys., 1966,44,411.
9 P. S. Vincett, W. A. Barlow, R. A. Hann and G. G, Roberts, Thin SolidFilms, 1982,94,
171.
Photolurninescence and Electrolurninescencefrom Organic Materials 145
10 F. J. Campas and M. Goutermann, Chem. Phys. Lett., 1977,48,233.
11 J. Kalinowski, J. Godlewski and Z. Dreger, Appl. Phys. Lett. A., 1985,37, 179.
12 C. W. Tang and S. A. Van Slyke, Appl. Phys. Lett., 1987,51,913.
13 C. W. Tang, S. A. Van Slyke and C. H. Chen, Appl. Phys. Lett., 1989,65,3610.
14 M. Era, C. Adachi, T. Tsutsui and S . Saito, Chem. Phys. Lett., 1991,178,488.
15 C. Adachi, S. Tokito, T. Tsutsui and S . Saito, Jpn. J. Appl. Phys., 1988,27, L713.
16 C. Adachi, T. Tsutsui and S . Saito, Appl. Phys. Lett., 1989,55,1489.
17 C. Adachi, T. Tsutsui and S . Saito, Appl. Phys. Lett., 1990,56,799 and 1990,57,531.
18 C. Adachi, T. Tsutsui and S . Saito, Acta Polytech. Scand. Appl. Phys., 1990,170, 145
and 215.
19 Y. Hamada, C. Adachi, T. Tsutsui and S-Saito, Jpn. J. Appl. Phys., 1992,31,1812.
20 A. Dodabalapur, L. J. Rothberg and T. M. Miller, Appl. Phys. Lett., 1994,65,2308.
21 J. Kido, M. Kimura and K. Nagaim, Science, 1995,67,1332.
22 S . Miyata and H. S. Nalwa, in ‘Organic Electroluminescent Materials and Devices’,
Gordon and Breach, New York, USA, 1997.
23 K. Y.Law, Chem. Rev., 1993,93,449.
24 H. Naarmann, in ‘Conjugated Polymer Materials: Opportunities in Electronics,
Optoelectronics and Molecular Electronics’, Eds. J. L. Bredas, and R. R. Chance,
Kluwer, Dordrecht, Netherlands, 1990.
25 A. Bohnen, H. J. Rader and K. Mullen, Synth. Met., 1992,47,37.
26 N. F. Colaneri, D. D. C. Bradley, R. H. Friend, P. L. Bum, A. B. Holmes and C. W.
Spangler, Phys. Rev., 1990, B42, 11670.
27 T. J. Marks and M. A. Ratner, Angew. Chem., 1995,107, 167.
28 M. Baumgarten and K. Mullen, in ‘Topics in Current Chemistry’, Springer Verlag,
Berlin, Germany, 1994, Vol. 169.
29 R. H. Friend, D. D. C. Bradley and A. B. Holmes, Phys. World, 1992,5/(11),42.
30 D. D. C. Bradley, Synth. Met., 1993,54,401.
31 R. H. Friend and N. C. Greenham, Solid State Phys., 1995,49,1.
32 D. D. C. Bradley, Curr. Opin. Solid State Mater. Sci., 1996,1,789.
33 R. W. Gymer, Endeavor, 1996,20,115.
34 J. Salbeck, Ber. Bunsenges. Phys. Chem., 1996,100, 1667.
35 L. J. Rothberg and A. J. Lovinger, J. Mater. Res., 1996,11,3174.
36 K. Kaneto, K. Yoshino, K. Koa and Y. Inuishi, Jpn. J. Appl. Phys., 1974,18, 1023.
37 R. H. Partridge, Polymer, 1983,24,755.
38 J. H. Burroughs, D. D. C. Bradley, A. R. Brown, N. Marks, K. Mackay, R. H.
Friend, P. L. Burn and A. B. Holmes, Nature, 1990,347,539.
39 J. H. Burroughes, D. D. C. Bradley, A. R.Brown, R. N. Marks, R. H. Friend, P. L.
Burn and A. B. Holmes, Nature, 1990,347,539.
40 F. Wudl, P. M. Allemand, G. Srdanov, Z. Ni and D. McBranch, ACS Symp. Ser.,
1991,455.
41 D. Braun and A. J. Heeger, Appl. Phys. Lett., 1991,58,1982.
42 A. J. Heeger, in ‘Primary Photoexcitation in Conjugated Polymers’, Ed. N. S.
Sacricifiri,World ScientificPress, New York, USA, 1987, Chapter 2.
43 N. C. Greenham, R. H. Friend and D. C. C. Bradley, Adv. Muter., 1994,6,491.
44 T. Tsutsui and S. Saito, NATO ASI Series, 1993,246.
45 S . Tasch, W. Graupner, G. Leising, L. Pu, W.Wagner and R. H. Grubbs, Adv.
Muter., 1995,7,903.
46 S . A. Carter, M,Angelapoulos, S. Karg, P. J. Brock and J. C. Scott, Appl. Phys. Lett.,
1997,70,2067.
47 J. C . Carter, I. Grizzi, S. K. Heeks, D. J. Lacey, S. G. Latham, P. G. May, 0. R. de 10s
Panos, K. Pichler, C. R. Towns and H. F. Wittmann, Appl. Phys. Lett., 1997,71,34.
48 N. C. Greenham, R. H. Friend, A. R. Brown, D. C. C. Bradley, K. Pichler, P. L.
Burn, A. Kraft and A. B. Holmes, Proc. S P I . Int. SOC.Opt. Eng., 1993,1910,84.
49 A. R. Brown, D. C. C. Bradley, J. H. Burroughes, R. H. Friend, N. C. Greenham,
P. L. Burn and A. B. Holmes, Appl. Phys. Lett., 1992,61,2793.
146 Chapter 4
50 Y. Yang and Q. Pei, J. Appl. Phys., 1995,77,4807.
51 E. Buchwald, M. Meier, S. Karg, P. Posch, H.-W. Schmidt, P. Strohriegel, W.Riess
and M. Schwoerer, Adv. Mater., 1995,7,839.
52 R. D. Scurlock, B. Wang, P. R. Ogilby, J. R. Sheats and R. L. Clough, J. Am. Chem.
SOC., 1995,117, 10194.
53 T. Zyung and J.-J. Kim, Appl. Phys. Lett., 1995,67,3420.
54 K. Z. Xing, N. Johansson, G. Beamson, D. T. Clark, J.-L. BrCdas and W. R.
Salaneck, Adv. Mater., 1997,9, 1027.
55 A. C.FOU,0. Onitsuka, M. Ferreira, D. Howie and M. F. Rubner, Pofym. Muter.
Sci. Eng., 1995,72, 160.
56 A. C . Fou, 0.Onitsuka, M. Ferreira, M. F. Rubner and B. R. Hsieh, Muter. Res. SOC.
Symp. Proc., 1995,369,575.
57 M. Onoda and K. Yoshino, Jpn. J. Appl. Phys., 1995,34, L260.
58 M. Onoda and K. Yoshino, Jpn. J. Appl. Phys., 1995,78,4456.
59 A. C. Fou, 0. Onitsuka, M. Ferreira, M. F. Rubner and B. R. Hsieh, J . Appl. Phys.,
1996,79,7501.
60 M. Ferreira, 0.Onitsuka, A. C. Fou, B. R. Hsieh and M. F. Rubner, Mater. Res. SOC.
Symp. Proc., 1996,413,49.
CHAPTER 5
Organic Light-Emitting Diodes

Using Lo w-Molar-Mass
Materials (LMMMs)
1 Introduction
The first prototype organic light-emitting diodes (OLEDs) using low-molar-
mass organic and organometallic compounds, such as metal chelates, were
constructed over 30 years ago. However, despite intense research and
development and a range of attractive features, OLEDs have only started to
penetrate the market for flat panel displays at the beginning of the 21st
~entury.~
Electroluminescence from organic materials, as opposed to doped inorganic
semiconductors, was first discovered using single crystals of anthracene (1)
(Table 5.1) immersed in an electrochemical cell.'** The cell contained an
electrolytic solution of negatively charged anthracene ions, prepared from a
solution of anthracene and sodium in tetrahyrofuran next to the cathode and a
solution of positively charged anthracene ions, prepared from a solution of
anthracene and aluminium trichloride in nitromethane, next to the anode. The
two liquid electrolytic solutions were physically separated from each other by
the solid anthracene crystal. The two electrodes were connected to a source of
potential. The electrolytic solutions were responsible for charge transport to the
electroluminescent anthracene crystal of a current of electrons injected from the
cathode and holes injected from the anode, re~pectively.'-~A visible emission of
blue light (ho < 3.1 eV) was observed above a certain threshold voltage from the
side of the device under the action of an electric current through the cell. This
was the first unambiguous report of electroluminescence from an organic
material caused by the recombination of holes and electrons to form an emissive
singlet exciton. Since anthracene is an insulating organic material with a high
resistivity and a low permitivity ( E = 3.4), the maximum electron mobility
through anthracene is still low (pe x 0.4cm2V-' sec-') compared with that
through metals or doped graphite, although relatively high for an organic
material of this type. Consequently, the electric current through anthracene is
147
148 Chapter 5
Table 5.1 Glass transition temperature ("C), emission maximum (nm) and
colourfor the compounds (1-3)
Molecular structure Acronym Tg A,, Colour Ref
430 Blue 192
2 TPAC 60 13
175 530 green 14
very low or zero at low field strengths. High voltages (Vop M 400-2000 V) had to
be applied in order to create the high electric fields (E M lo4- lo5Vcm-')
necessary to induce an electric current density of a significant magnitude
( I z lo-' A cm-2) to generate visible blue electroluminescence from the crys-
tals. This was due to the thickness of the crystals used ( d M 5mm), since in the
high current regime
V
Em-
L
The electroluminescence was located in the part of the anthracene crystal next
to the hole-injecting anode, which suggests an imbalance of charge-carrier
injection and transport. The intensity of light was linearly proportional to the
Organic Light-Emitting Diodes Using Low-Molar-Mass Materials (LMMMs) 149
magnitude of the current above the threshold voltage for electroluminescence.
This appeared to correspond to a space-charge limited (SCL) model of charge-
carrier injection and transport. The external quantum efficiency of these
simple, prototype OLEDs was reported as being remarkably high
(qext w 35%). The mobility of both holes and electrons in crystalline anthra-
cene is high for an organic material. Therefore, recombination may also occur
in the bulk of the material. This may partially explain the high quantum
efficiency of this device, which corresponds to a photoluminescence efficiency
of 99%.6
Considerably lower operating voltages were found using solid, metallic
electrodes instead of the two liquid electrolytic solutions used for charge
transport and injection to thin planar wafers of cleaved anthracene crystal^.^
This prototype, solid-state OLED was glued to a semi-transparent solid
substrate for mechanical stability and planarity. The cathode was a negatively-
charged alkali metal salt [sodium (Na), potassium (K), or a sodium-potassium
(Na-K) alloy] of anthracene with a reasonably low work function (4 = 2.05 eV
for the Na-K alloy and 4 = 2.3eV for K). The cathode was prepared by
reaction of the metal with the interface with the anthracene wafer. The anode
was made from gold (Au) or an alloy of equal amounts of selenium and
tellurium (Se50-Te50),both of which possess a high work function (4 = 5.2eV
for gold (Au) and 4 = 4.8 eV for Se50-Te50).Relatively low threshold voltages
( V t h > 35 V) and operating voltages (Vop > 100 V) for visible electrolumines-
cence of blue light (Amax = 430nm w 3eV) were obtained using the very thin
wafers of anthracene (30pm to 200pm) compared to those (400-2000 V)
observed using thick (5000 pm) anthracene crystals.lP2This is a clear function
of the difference in the thickness, d, of the organic electroluminescent material.
The electroluminescence intensity was linearly proportional to the current
and almost independent of the sample thickness above the threshold voltage.
The external quantum efficiency (1% < qext < 8%) of these OLEDs using thin
wafers of anthracene appeared to be lower than that observed for the OLEDs
using anthracene crystals (qext w 35%). This may be attributable in part to the
mismatch of the work functions of the electrodes and the LUMO and HOMO
energy levels of anthracene. Remarkably high current densities ( I = lo-' A
cm-2) were observed for high voltages (Vop = 600 V) without the onset of
dielectric breakdown of the diode or apparent electrochemical degradation of
the organic anthracene layer.
Very thin layers ( 3 pm < d < 30 pm) of anthracene were produced by deposi-
tion of anthracene powder followed by melting and subsequent recrystallisation
of the However, the presence of pinholes in extremely thin layers meant
that thicker layers (d > 14pm) had to be used. However, very thin layers
(d w 0.1 pm) without pinholes were finally produced by physical vapour-
deposition of anthracene under high vacuum conditions.*-" OLEDs produced
using these layers exhibited low threshold voltages ( Vth w 3.5 V) and operating
voltages (V,, < 30 V) using solid electrodes. However, the quantum efficiency
of these devices was low (qext w 0.050/,), probably due to the poor quality of the
films deposited.
150 Chapter 5
The configuration and construction of monolayer and multilayer OLEDs
have undergone substantial changes and modifications since these first reports
of organic electroluminescence from low-molar-mass materials. Several types of
OLEDs using small organic and organometallic molecules are described
schematically below.
2 Monolayer Organic Light-Emitting Diodes Using

LMMMsl-’
D ispIay Configuration
A thin film (z100nm) of the organic electroluminescent material is sandwiched
between two electrodes supported on a solid substrate, such as glass, see Figure
5.1. The device substrate providing the mechanical support to the anode,
electroluminescent polymer layer and the cathode is transparent or semi-
transparent in order to allow the light generated by electroluminescence to
actually leave the display and be observed. An appropriate voltage can be
applied between the electrodes. The cathode, e.g. a thin layer ( z7-15 nm) of
aluminium, calcium or a magnesium/silver alloy, has a low work function
(ionisation potential), whereas the anode, e.g. a thin metallic layer (1 5 nm) of
indium-tin oxide (ITO), has a high work function. No light is emitted in the
absence of an electric current and so the non-activated pixels appear dark. The
application of a voltage between the two electrodes above the threshold voltage,
Vth, leads to the injection of electrons from the cathode and holes from the
anode into the organic electroluminescent material and the emission of light.
The threshold voltages and operating voltages for monolayer OLEDs using
small molecules are primarily determined by the thickness of the organic layer,
the compatibility of the work functions of the electrodes and the LUMO and
HOMO of the organic material.
METALLIC
-
4
CATHODE
mTLAND
EMI?TER
TRANSPARENT
ANODE
TRANSPARENT
-I 000000 0 000000
00 00000 000000
ti t t t t t t t t t t t
SUBSTRATE
Figure 5.1 Schematic representation of a monolayer OLED using low-molar-mass

materials incorporating an electroluminescent material between a transparent
anode and a cathode.
Organic Light-Emitting Diodes Using Low-Molar-Mass Materials (LMMMs) 15 1
3 Bilayer OLEDs Using LMMMs

The external quantum efficiency of electroluminescentdevices using one layer of
organic light-emitting material sandwiched between two metallic electrodes was
generally found to be very low, whatever the organic material used. This
phenomenon has several causes intrinsic to the device configuration. If there is
a difference between the injection barriers of charged species at the electrodes
then one type of charge-carrier will be preferentially injected. Furthermore, a
given organic material will transport one type of charge-carrier preferentially,
i.e. the mobility of holes and electrons through the organic monolayer will be
different, often substantially so. This will lead to a concentration of both charge-
carriers near the electrode injecting the minority charge-carrier. Therefore,
recombination of holes and electrons, consequent exciton formation and
subsequent radiative decay of singlets takes place in a narrow planar region
near the interface between one electrode and the organic material. However, the
metal atoms of the electrode generally induce non-radiative quenching of the
excitons. Non-radiative transfer of energy from the excitons to the metal, e.g. by
Forster transfer, can also occur near to the interface. If the ratio between the
two charge-carrier currents is large then a number of one of the charge carriers
can pass from one electrode to the other without recombination with the
oppositely charged species being able to take place. A combination of all these
phenomena generally leads to a low external quantum efficiency for single-layer
OLEDs.
Consequently, interest in organic electroluminescence waned for many years
until the report by Tang and Van Slyke (Eastman Kodak Corporation) of
bilayer OLEDs with a much higher efficiency and lower operating voltage due,
at least in part, to a more balanced charge injection from the anode and
cathode. 1 1 . 1 2 The emission is also concentrated at the interface between the two
organic layers away from the metallic electrodes, which reduces quenching by
the metal atoms. There are several possibilities for constructing bilayer OLEDs.
These include an electron-transport layer (ETL) and a combined hole-transport
(HTL) and emission layer. Conversely a hole-transport layer and a combined
electron-transport and emission layer is also effective.
Display Con$gura t ion

A thin film (z100nm) of an organic hole-transport layer (HTL) supports a
second thin film of an organic electron-transport layer (ETL) situated between
two electrodes supported on a substrate, see Figures 5.2 and 5.3. One or both of
the layers form the emission layer. The anode is transparent in order to allow the
passage of the light generated. The electrodes are connected to a source of
potential. The metal cathode has a low work function, which is matched as
closely as possible to the LUMO of the ETL in order to facilitate efficient
electron injection. The HOMO of the HTL is also matched as near as possible to
the high work function of the anode, e.g. indium-tin oxide (ITO) or an organic
polymer electrode such as poly(ani1ine) salt (PANi). No light is emitted in the
152 Chapter 5
METALLIC
CATHODE 1
ETL
HTLAND
EMITTER
TRANSPARENT
ANODE
TRANSPARENT
SUBSTRATE
Figure 5.2 Schematic representation of a bilayer OLED using low-molar-mass materials

incorporating a combined hole-transport and emission layer and an electron-
transport layer situated between a transparent anode and a cathode.
METALLIC
CATHODE 1
ETLAND
EMITTER 0000000000000
HTL
TRANSPARENT
ANODE 4 t-'
TRANSPARENT
SUBSTRATE
Figure 5.3 Schematic representation of a bilayer OLED using low-molar-mass materials

incorporating a hole-transport layer and a combined electron-transport and
emission layer situated between a transparent anode and a cathode.
absence of an electric current and so the non-activated pixels appear dark.

Application of a voltage above the threshold voltage leads to injection of
electrons from the cathode into the ETL and holes from the anode into the
HTL. The close matching of the work functions of the individual electrode
materials to those of the ETL and the HTL lowers the barriers to charge
injection. Thus, the threshold voltage is lower than that of analogous monolayer
OLEDs. The resultant electric current passing through the organic layers is
much more balanced than that found in monolayer OLEDs, since each layer
preferentially transmits one type of charge-carrier and blocks the transport of
the oppositely charged species. Thus, a high percentage of the injected electrons
and holes congregate at the interface between the ETL and HTL, where charge
recombination and exciton formation can take place. This may further lower
the barrier to charge injection at the electrodes. The concentration of the
charge-carriers at the interface between the HTL and ETL gives rise to the
efficient emission of light from the excitons in a thin strip in one or both of the
layers, either side of the interface, deRending on the presence or absence of
emitters in the individual layers, see Figures 5.2 and 5.3. Thus, the activated
pixels emit light and appear bright.
This bilayer OLED configuration generally results in a relatively high
quantum efficiency (1% < qext< 4%) and high luminance at low operating
voltages. Tang and van Slyke reported" a green monochrome, bilayer OLED
with high brightness at low operating voltages using a positive bias across a
bilayer of the diamine (2)13 as the HTL and Alq3 (3)14 as the ETL and emission
layer, see Tables 5.1 and 5.2. The metal chelate Alq3 (3) has a relatively high
electron-transport mobility (pe = lo-' cm2V- s- ') for an organic material,
whereas the diamine (2) has a high hole-transport mobility (ph = cm2 V-
s- '). The cathode was made from an alloy (Mglo-Agl) of magnesium (Mg) and
silver (Ag) with a relatively high work function (4 = 3.6eV) and the anode
consisted of a layer of indium-tin oxide (ITO) with a high work function
(4 = 4.6 eV). The advantage of a Mg-Ag alloy as the cathode is that, despite the
high work function, it is stable with respect to electrochemical oxidation in air.
Very high values for the magnitude of the brightness ( L > 1000 cd m-2) can be
achieved using higher operating voltages ( Vop > 10 V). However, the half-life of
Table 5.2 ConJguration and typical electro-optical performance of an

early bilayer OLED11p12 using the low-molar-mass diamine
(2) as HTL" and Alq3 ( 3 ) as the ETL and emission layer14
Property Value
Cathode Mglo-Agl
Anode IT0
Hole transport layer (HTL) (2) 75 nm
Electron transport layer (ETL) and emission layer (3) 60 nm
Threshold voltage, Vth 2.5 V
Operating voltage, Vop 5.5 v
Quantum efficiency, 4 1%
Power conversion efficiency 0.46%
Luminous efficiency 1.5 lm W-'
Current density, I 10-'-102 mA cm-2
Colour Green
Emission maximum, A,, 550 nm
Bandwidth 200 nm
Brightness, L 100 cd m-'
154 Chapter 5
emission was found to be short (100 h) under normal drive conditions ( L =
50 cdm-* and Z = 5 mA cm-2). This was attributed to the degradation of the
electrodes as manifested by the formation of black spots, although the exact
mechanism of degradation was not completely understood at the time. All the
organic and metallic layers are created by vacuum deposition at high vacuum
( lod5 Torr).
4 Trilayer OLEDs Using LMMMs15-22

'
Even higher external quantum efficiencieswere found for three-layer devices. 5-22
There are several reports of trilayer OLEDs using three distinct layers for
electron transport, hole transport and emission.15-20 This device configuration
has the advantage that each layer can be optimised for one particular function,
i.e. hole-transport, electron-transport or charge capture, recombination,
exciton formation and subsequent emission of light. Simultaneous emission of
red, green and blue light from a trilayer OLED can also be used to generate
white light However, the fabrication of OLEDs with three distinct
organic layers of controlled thickness, integrity and homogeneity with high
production yield and low overall cost is a challenging task.
A thin layer (= 100 nm) of a hole-transporting material with a low ionisation
potential is situated next to the anode. An electron-transporting layer with a
high electron affinity is situated next to the cathode. An organic emission layer is
sandwiched between the HTL and the ETL, see Figure 5.4. At least one of the
two electrodes is transparent in order to allow the passage of the light generated.
A potential can be applied between the electrodes. The low work function of the
metal cathode is matched as closely as possible to the LUMO of the ETL in
order to facilitate efficient electron injection. The high work function of the
anode is also matched as near as possible to the HOMO of the HTL. The
electroluminescent material in the centre of the display is chosen to exhibit a
high quantum efficiency of emission at the desired wavelength. No light is
emitted in the absence of an electric current and so the non-activated pixels
appear dark. Upon application of a voltage above the threshold voltage
electrons are injected from the cathode into the ETL and holes are injected
from the anode into the HTL. Therefore, a current of electrons passes through
the ETL and a current of holes travels in the opposite direction through the
HTL. This creates a concentration of oppositely charged species in the emissive
layer. The electrons and holes then recombine in the emission layer or at the
interface between the emission layer and the HTL and ETL. This leads to
efficient charge recombination in the middle of the device with consequent
generation of singlet excitons, from which light is then emitted, as well as
formation of excitons in the generally non-emissive triplet state.
Very thin emitter layers (5nm) are sufficient to produce bright electrolumi-
nescence. 19321However, even one or two layers of an organic electroluminescent
METALLIC
CATHODE
ETL
EMISSION
LAYER
HTL
TRANSPARENT
ANODE
TRANSPARENT
SUBSTRATE
Figure 5.4 Schematic representation of a trilayer OLED using low-molar-mass materials

incorporating a hole-transport layer (HTL), an electron-transport layer
(ETL) and a central emission layer situated between a transparent anode and
a cathode.
material deposited by Langmuir-Blodgett techniques situated between a HTL

and ETL in a trilayer OLED were found to generate electroluminescence of a
sufficient intensity to be visible to the eye.15
5 Low-Molar-Mass Organic Materials for OLEDs23-25

The low-molar-mass materials (LMMMs) designed and synthesised for use in
OLEDs must satisfy a number of essential requirements in order to be of use in
practical commercial devices. They must be capable of being deposited as a
pure, uniform, thin solid film by vapour deposition under high vacuum. The
individual layers should exhibit a high glass transition temperature in order to
inhibit crystallisation during the lifetime of the device. They must be thermally,
chemically, photochemically and electrochemicallystable. They should function
as a hole-transport layer, an electron-transport layer, a highly efficient fluor-
escent/phosphorescent emitter or a combined transport and emission layer
depending on the device combination. Materials, which satisfy some or all of
these requirements, are described in detail below. Where available the colour
and maximum wavelength (Amax) of electroluminescent emission of these
materials are given in the Tables 5.1-5.12. Unfortunately, these values are not
always readily available, especially from sources reporting the synthesis of new
materials. The photoluminescence and electroluminescence spectra of organic
or organometallic molecules are usually broad due to molecular vibrations, see
156 Chapter 5
Chapter 4. Therefore, the observed colour does not always correspond exactly
to the A,, as is the case for atomic emission spectra.
Non-Emissive Electron-Transport Layers (ETLs)23,24

The majority of electron-transport materials are also good emitters. These will
be described in detail in the section on emissive materials. However, several
classes of low-molar-mass organic materials are used as non-emissive, combined
electron-transport and hole-blocking It is important that a charge-
transport layer conducts predominantly one type of charge-carrier and blocks
the transport of the other charged-species, i.e. the ratio between the mobility of
the holes and electrons should be as high as possible for each individual layer.
The HTL should predominantly transport holes and block the passage of
electrons. The ETL should transport electrons preferentially and minimise the
transport of holes.
Some typical examples of low-molar-mass, electron-transporting and hole-
blocking materials for multilayer OLEDs are collated in Table 5.3.A bis(benz-
imidazoly1)perylenedicarboximide derivative (4; R = H) was used in the first
prototype OLED2’ using an emissive HTL and a non-emissive ETL. Electron-
deficient nitrogen heterocycles are used as the ETL in OLEDs using LMMMs
due to their high electron affinity attributable to the presence of the electro-
negative nitrogen atoms in the five-membered heterocyclic ring. These include
2,5-disubsituted-1,3,4-0xadiazoles (57)26327 and triazoles (8). The 2-(4-tert-
butylpheny1)-5-(biphenyl-4-y1)-1,3,4-0xadiazole (5), often referred to by its
acronym of PBD, is used very often as a non-emissive ETL. However,
compounds incorporating two oxadiazole rings, such as 728are also used due
to the lower tendency of thin layers containing them to crystallise in devices.
2,5-Disubsituted-1,3,4-triazoles such as compound 829 also contain a five-
membered heterocyclic ring. However, triazoles contain three nitrogen atoms
rather than the two nitrogen atoms of oxadiazoles, and consequently, triazoles
generally possess a greater electron affinity than that exhibited by the
analogous 2,5-disubsituted- 1,3,4-0xadiazoles. The use of bis(benzimidazoly1)-
naphthalenedicarboximides (9)24 and thiopyran sulfones ( as ETLs has
also been reported. The spiro-PBD (11) forms a stable glass with a high glass
transition temperature due to its twisted structure as a result of the sp3
bonding at the point of linking.30-32Although the melting points of these
spiro-linked compounds are generally high due to their high molecular weight,
they form the glassy state on quenching to temperatures below the melting
point .31932
Non-Emissive Hole-Transport Layers ( H T L S ) ~ ~ , ~ ~

Since a large number of low-molar-mass emitters are also efficient electron-
transport materials, there is a clear requirement for efficient, non-emissive, hole-
transporting, electron-blocking LMMMs. Fortunately, a large range of hole-
transporting, low-molar-mass organic materials had already been developed as
Table 5.3 Molecular structure of the low-molar-mass, non-emissive ETL
materials (4-11)
Molecular structure Acronym Ref
PV 25
PBD 26-28
BND 26-28
OXD 1 2%
TAZ 29
PD 24
10 TPS 24
11 Spiro-PDP 30,31
158 Chapter 5
photovoltaic materials for use in xerography and laser printing.24A whole series
of low-molecular-weight, aromatic diamines have been synthesised and used as
an HTL in OLEDs due to their low ionisation potential and high hole mobility
-'
( p h = I O - ~ C ~ ~ Vs - ),
' see Table 5.4.23*24 A further advantageous feature of
such aromatic diamines is their capability for forming uniform, amorphous
layers free of pinholes upon vapour deposition under high vacuum. The glass
transition temperature, Tg,of arylamines, such as the arylamine (2) shown in
Table 5.1 and those arylamines (12-15)24933-34shown in Table 5.4,is often
significantly above room temperature. However, their Tg is often not high
enough to avoid posing problems during high-temperature fabrication steps or
to completely inhibit crystallisation at moderately high operating temperatures,
where crystallisation is a ~ c e l e r a t e dNonetheless,
.~~ they do permit the fabrica-
tion of multilayer OLEDs with much longer device lifetimes and a much higher
efficiency ( x 10) than those of monolayer OLEDs using low-molar-mass
electroluminescent materials. For example, bilayer OLEDs utilising compound
15 as the HTL and Alq3 (3) as the ETL and emissive layer have been found to
operate at high temperatures (> 140 "C) over many hours without dielectric
breakd~wn.~' The N,N'-diphenyl-N,N'-bis(3-methylphenyl)biphenyl-4,4'-yldia-
mine (12), often referred to by its acronym (TPD), is frequently used as a non-
emissive HTL in multilayer O L E D S . ~ Pyrazoline
~ derivatives, such as com-
pound 16, where the nitrogen atoms are incorporated in an aromatic ring, also
exhibit good electron-transport and hole-blocking proper tie^.^^ Spiro-linked
versions of the compounds shown in Table 5.4can also be used to form stable
ETL with high glass temperatures.30931
Liquid Crystals as Charge-Carrier Transport Layers

The order present in liquid crystalline phases offers the possibility of attaining
the high mobility found in single crystals, but without the traps and defects
present at crystal grain boundaries, i.e. uniform thin layers with high mobility
can be prepared over the area of a display and even lithographically patterned,
should this be required. The nature and transport of ions present in nematic
liquid crystals had been extensively studied in order to understand their
contribution to the electro-optical response of LCDs, especially with regard
to the DSM-LCD and TN-LCDs and IPS-LCDs with active matrix addres-
sing, see Chapter 3, well before the use of liquid crystalline phases as transport
layers in OLEDs had been proposed.374' However, the charge transport of
photogenerated ions (and ionic impurities) through nematic liquid crystals
under the action of an applied electric field was also found to be primarily ionic
in nature and not electronic.4246 This was shown by the viscosity-dependent
mobility of ions with a size much larger than electronic charge-carriers in
accordance with Walden's However, the photogeneration of charge-
carriers and subsequent transport under the action of an applied electric field
through the ordered columnar liquid crystalline phases and smectic phases has
been shown clearly to be ele~tronic.4~"~ Furthermore, the charge-carrier
mobility can be high (lo-' cm2 V-' s-' > p > cm2 V-' s-'), almost
Table 5.4 Molecular structure and glass transition temperature ("C) of the low-
molar-mass, non-emissive HTL materials (12-16)
Molecular structure Acronym Tg Ref
13
' p TTB 82 24
14 NPD 98 24
15 TPTE 130 24
comparable to that through single crystals, for certain compounds in these

ordered liquid crystalline phases. Therefore, they may well be suitable as
efficient organic transport layers, certainly for holes and perhaps also for
electrons, in multilayer OLEDs.
160 Chapter 5
Columnar Liquid Crystals as HTL
Improvements in conductivity could be expected, if the one-dimensionality of
the generally amorphous or partially ordered LMMMs used as a HTL could be
extended to two dimensions with an increase in order and homogeneity in an
organised structure, if crystallisation could be inhibited over the storage and
operating temperature range of the A thin layer of copper phthalo-
cyanine situated between the anode and the HTL is often used in OLEDs in
order to reduce the injection barrier of holes into the organic layers. It was
recently discovered that t r i p h e n y l e n e ~and
~ ~phthalocyanines
~~ (17)68-70could
exhibit large charge mobility in the columnar (discotic) liquid crystalline state,
see Table 5.5 and Figure 5.5. The values observed, e.g. in the ordered hexagonal
columnar phase of hexakis(n-a1koxy)triphenylenes (18) and in the helical
columnar phase of the hexakis(n-alky1thio)triphenylenes ( 1 9 p for the charge
carrier mobility (lo-' cm2 V-' s-l > p h > 1 0 - ~cm2 V-' s-'> are inter-
mediate between those observed for organic single crystals and inorganic
semiconductors (p z 1 cm2V-'s-') and those cm2 V-' s-l > p >
cm2 V- s- ') of amorphous conjugated polymers, c.f poly(N-vinylcarbazole)
(PVK), see Chapter 6. Holes represent the majority charge carrier, rather than
electrons, due to the very limited electron affinity and low ionisation potential of
the electron-rich aromatic cores of these columnar liquid crystals. The charge
transport is non-dispersive and the charge carrier mobility parallel to the
director, i.e. parallel to the core of the columns of self-assembled discs, is
orders of magnitude higher than that measured orthogonal to the director, i.e.
from column to column. This is partly attributable to the insulating nature of
the lateral aliphatic chains surrounding the aromatic cores.
Hence, columnar (discotic) liquid crystals with an aromatic central core60-82
are of significant interest for application in OLEDs as an HTL, if the columnar
phase can be aligned with the columns orthogonal to the substrate surface, i.e.
the flow of holes from the anode to the ETL and/or emission layer is
Figure 5.5 Schematic representation of the Dho columnar phase with an ordered hexago-
nal arrangement of the columns of disc-shaped molecules with a regular period
of the discs within the columns. The director is parallel to the columns and
normal to the plane of the discs.
Organic Light-Emitting Diodes Using Low-Molar-Mass Materials ( L M M M s ) 161
Table 5.5 Transition temperatures (“C)for the columnar liquid crystals (17-21)
Molecular structure Cr 0 2 Dho I Ref
0 193 - 0 197 0 62-64
>-5 65 178 0 60,61
>-5 65 178 60,61
21 0 18* 0 83 0 >400 0 98
* D3-Dz transition.
Columnar liquid crystals generally consist of disc-shaped mole-

cules or self-assembled aggregates organised in a supramolecular structure of
nearly parallel columns of varying degrees of order in a two-dimensional lattice,
see Figure 5.5.69-73In the related nematic discotic phase formed by similar or
even the same compounds at higher temperatures there is no regular columnar
structure.7G82In columnar mesophases the molecular cores are organised above
162 Chapter 5
each other in columns separated by tbe peripheral aliphatic chains. Therefore,
the intercolumnar distance ( z 15-40 A) is much greater than the intracolumnar
distance (<4.5 A), depending to some extent on the length and degree of
conformational mobility of the aliphatic chains. The classification depends on
the degree and type of order in the lattice (columnar hexagonal, colh; rectan-
gular, col,; oblique, c01,b; which are either ordered, e.g. colho, or disordered, e.g.
~ ~ l h ~The ) . nature
~ ~ -of~columnar
~ phases, especially those with hexagonal
symmetry, C d h , in which the aromatic cores of the molecules are arranged above
each other in columns allows charge carrier-migration due to the overlap of the
n-electron orbitals in the conjugated aromatic core of neighbouring mole-
. ~ ~ ester (20) exhibits a plastic columnar phase, which does not
c u l e ~ The
crystallise at room temperature.60961 The hole mobility in this plastic columnar
phase is high cm2 V-' s-' > ph > cm2 V-3 s ).-'
Most compounds exhibiting columnar phases for electrical and electro-
optical applications incorporate triphenylene derivatives due to their high
tendency to form the required columnar phases and to the presence of the
large conjugated aromatic molecular core with a large delocalised n-electron
system. In order to be able to synthesise them efficiently in large quantities in
high purity and expand the scope for potential columnar liquid crystals with
modified structures a series of new synthetic methodologies have had to be
developed and o p t i m i ~ e dThey
. ~ ~ are
~ ~usually based on the use of palladium-
catalysed cross-coupling reactions involving arylboronic acids.9s97 Such cross-
coupling reactions are very tolerant of many functional moieties and enable the
synthesis of a wide range of materials not otherwise accessible or with additional
protection/deprotection steps.
The largest values for mobility (p x 4 x lo-' cm2 V-' s-') have been
foundg8-lm for columnar liquid crystals with large polyaromatic cores incor-
porating many phenyl rings, such as the coronene (2l)." However, the very high
clearing point of these columnar liquid crystals renders them very difficult to
align by cooling from the isotropic liquid into the column phase followed by
annealing in the columnar state. This is a common problem for the classes of
columnar liquid crystal collated in Table 5.5.
Polycatenar compounds, with more than one terminal chain at each end of a
rod-like molecule can also exhibit columnar phases, although they possess a
linear molecular structure. lo' Three linear molecules aggregate together to form
a self-assembled disc. These discs then stack up to form a fluid columnar
structure. Non-dispersive charge-carrier mobility in such polycatenar com-
pounds has also been found.lo2
A further potential problem associated with the practical fabrication of
multilayer OLEDs with low-molar-mass calamitic and columnar liquid crystals
with a charge-transport layer, especially as the HTL, is the fluid nature of the
liquid crystalline state. This could permit contamination of the columnar liquid
crystal during the deposition of subsequent layers by the process of vapour
deposition or spin-coating. The columnar materials could well be soluble in the
solvent used to deposit the next layer. This would lead to chemical contamina-
tion of the individual layers due to mixing of one component into another, the
formation of layers of uneven thickness and even to the formation of pinholes.
Therefore, the Langmuir-Blodgett technique was used to fabricate bilayer and
trilayer OLEDs using columnar liquid crystals as the HTL. However, this is not
really compatible with low-cost, high-volume device fabrication. Multilayer
OLEDs using liquid crystals as the HTL and ETL and/or emission layers
created by physical vapour deposition have yet to be reported. Side or main-
chain polymers or cross-linked polymer networks could potentially alleviate this
fabrication problem, see Chapter 6. However, the polymer structure tends to
disrupt the columnar order and lead to significantly lower mobility.60 Mono-
layer OLEDs utilising the ester 20 were found to emit in the blue part of the
spectrum (Amax = 500 nm) with a large
Smectic Liquid Crystals as HTL and ETL

The compounds (22-26) collated in Table 5.6 represent the small number of
electroluminescent calamitic liquid ~ r y s t a l s . ~ ' -Compounds
~~ in the liquid
crystalline state generally exhibit a higher charge-carrier mobility than in
the crystalline state or in the liquid state. This is due to the higher degree
of molecular order in the liquid crystalline state without grain-boundary
defects and traps associated with polycrystalline compounds and amorphous
solid^!^-^^ The mobility in the smectic A phase of compound 22 has been shown
to be two orders of magnitude higher than that in the isotropic liquid state of the
same compound.47748 In smectic mesophases the calamic, rod-like molecules are
aligned in layers. In the smectic A phase the director is parallel to the molecular
long axis and orthogonal to the layer plane, see Figure 5.6. The smectic A phase
represents a fluid two-dimensional density wave. The charge-carrier mobility in
the ordered lamellar smectic phases of compounds with a rod-like molecular
structure, such as compounds (22-24) shown in Table 5.6, has been found to be
CHARGE MOBILITY
Figure 5.6 Schematic representation of the smectic A phase (SmA) formed by calamitc,
rod-like molecules. The director is parallel to the molecular long axis and
normal to the plane of the layers.
164 Chapter 5
% 0
IA ri
VI
r--
-3
oc
d
QI
VI
0 0 0
r-
I 1 I 53
0
0 Q\
VI 00
El s! I
0 0 0
I I I I I
*
m
I r-
00 s! I I
0
m
m
0 I: d
00
53
0 0 0 0 0
d" F
+
Q,
5
N m
N
d.
b
N
v) N
\o
surprisingly high cm2 V-' s-' > p > cm2 V-' s-') as well as
electronic in n a t ~ r e . ~However,
~ - ~ ~ the melting point of all three of these
compounds is well above room temperature and, therefore, time-of-flight
measurements of electric current had to be made at high temperatures. The
compounds 22-24 are clearly not suitable for practical applications due to their
crystalline nature at room temperature.
The transport of photogenerated current through the smectic A phase of the
organic conjugated heterocyclic thiobenzazole (22) is nondispersive and pri-
marily of holes (ph < 5 x cm2V-' s-') due to the low ionisation potential
of this electron-rich c o m p o ~ n d .A~ high
~ * ~degree
~ of rectification indicated that
the majority charge-carrier was holes. The charge-carrier transport through the
smectic A phase and the. unknown smectic X phase (probably smectic B or
smectic E) of the aromatic oxadiazole (23) is also unipolar in nature. However,
the majority charge-carriers are electron^.^^'^' The mobility in the unknown, but
certainly more ordered, smectic X phase ( p e < 0.8 x cm2 V-' s-l) is
higher than that ( p e = cm2 V-' s-') found in the less ordered smectic A
phase. The charge-carrier transport through the aromatic phenylnaphthalene
(24) is ambipolar in nature, transporting holes and electrons almost equally
We11.5738
Liquid Crystals as Electroluminescent Materials

The ability to macroscopically align the director in the nematic or smectic state
formed by rod-like, low-molar-mass liquid crystals on conducting alignment
layers offers the possibility of polarised electroluminescent emission.49750953-59
The first report of polarised electroluminescence from a low-molar-mass liquid
crystal described emission from an aligned sample of compound 23 in the
smectic X phase at temperatures above the recrystallisation temperature well
above room temperature.49950Thin layers (15-20nm) of Alq3 (3) and CuPc
(copper phthalocyamine) were used as the ETL and HTL, respectively. A thin
layer (3 pm)of the compound 23 was the emission layer in a trilayer device, see
Figure 5.4. The HTL was deposited by epitaxy on a conductive I T 0 layer in
order to generate the required alignment in the zenithal plane. Polarised
emission was observed at 70°C from the smectic X phase at 90 V pm- and 70 '
mAcm-2. However, the brightness (0.8 cdm-2) and quantum efficiency
(2.8 x %) were very low. The polarisation order parameter was also low
( S = 0.32), probably due to the poor alignment treatment which would give rise
to a random pattern of aligned smectic domains.
The nematic compound 25 exhibits a nematic phase above a low melting
However, on quenching this material forms the glassy state with a T,
well above room temperature. Therefore, this material can be uniformly aligned
on rubbed polyimide in the relatively fluid nematic state and then quenched by
cooling to room temperature in order to fix this order and orientation in the
stable glassy state. Nematic glasses of compound 25 have been found to
generate polarised photoluminescence and electroluminescence with a high
polarisation ratio ( x 12:l) and moderate brightness ( L = 100 cdm-2). This
166 Chapter 5
material is unique in that it represents the first low-molar-mass, organic
electroluminescent material to form a nematic glass above room temperature
and then emit plane polarised light with a high polarisation under the action of
an applied electric field from the vitrified state. This compound can also be
cross-linked to form an insoluble electroluminescent macromolecule, see
Chapter 6. Therefore, this compound and related electroluminescent glasses
and networks have great potential in combination with a low-cost, clean-up
polariser as a source of polarised backlighting for bright LCDs.
Compound 26 exhibits a nematic phase over a narrow temperature range.
However, it has been transferred to an OLED substrate by laser ablation under
high vacuum without significant photochemical or thermal degradati~n.'~ If the
substrate is an alignment layer, then oriented samples could theoretically be
obtained by depositing the electroluminescent material on top of it, which
would be another possibility for generating polarised light emission without the
need for spin-coating of organic solutions of electroluminescent compounds.
This may facilitate the preparation of multilayer OLEDs without the problems
associated with layer deposition spin-coating, such as layer corruption and
resultant inhomogeneity and non-reproducibility.
Green Electroluminescent, Low-Molar-Mass Organic Materials

Typical examples of low-molar-mass electroluminescent materials, which emit
green light are collated in Tables 5.7 and 5.8. The metal chelates (27-31) are
thermally stable, efficient emitters of green light. They are also capable of being
transferred onto a device substrate as uniform thin layers by physical vapour
deposition. Most electroluminescent materials incorporate metal atoms from
groups I1 or I11 from the periodic table. Alq3 (3) contains aluminium from
group I11 and, therefore, has a co-ordination number of 6, i.e. the aluminium
atom requires six covalent or quasi covalent bonds. The nature of the
substituent, R, in the aluminium quinolates (27)'03p'04 can be chosen to
modulate the emission maximum depending on the electron-donating or
electron-withdrawing effect (+ I or -I) of the substituent R. Many other
metal chelates with quinolinolato ligands have been synthesised.lo5*'06
However, Alq3 (3) is still the most widely used green emitter.' '-12 One reason is
the high quantum yield observed for thin layers of pure Alq3 (3) in multilayer
OLEDs and another is the high T,. Multilayer OLEDs using Alq3 (3) as a
combined ETL and emission layer can give rise to very bright displays
( L > 15000 ~ d m - ~ ) .
Chelates 28 and 29107*'08 containing the metal atoms beryllium (BE) and zinc
(Zn) from group I1 in the periodic table have the co-ordination number 4.
Therefore, their complexes require two bidentate ligands instead of three for the
chelates incorporating metals from group I11 such as Alq3 (3). The metal
chelates BeBqz and Zn(BTZ)2 have been found to be stable electroluminescent
materials of high brightness and long life-times, especially as dopant emitters in
an HTL. '07 Zn(BTZI2 (29)exhibits a very broad electroluminescent spectrum
almost corresponding to white light emission. This would eliminate the problem
Table 5.7 Emission maximum (nm) and colourfor the metal chelates (27-31)
Molecular structure Acronym L.zx Colour Ref
27
6F.O 0 \
0 R
530 f 35 Green 103,104
28 516 Green 107
29 Zn(BTZ)2 486,524 Greenish 108

white
30 Tb(acac), 544 Green 110
31 Tb(acac)3phen 543 Green 111

168 Chapter 5
of controlling the light intensity of three individual materials emitting either red,
green or blue light in attempts to produce white light emission from an OLED
using LMMMs. 12,2122,109 However, very few materials24emit broad-band light
covering most of the visible spectrum and at least two emitters are usually
required for white light emission. 12*219229109Therefore, white-light emitters are
not dealt with as a separate section in this Chapter.
Lanthanide-containing metal chelates such as compounds 30 and 31' '
generally possess sharp emission spectra with narrow band widths. Thus, the
terbidium chelates (30 and 31) emit almost monochromatic green light. Electro-
luminescent materials containing lanthanide metal atoms may exhibit very high
quantum efficiencies. Terbidium and europium are often used as dopant
emitters in inorganic LEDs. The emission results from intramolecular transfer
of energy from the excited triplet state to the 4f energy levels of the metal ion.
Thus, the normal maximum (25%) for the quantum efficiency due to emission
from the singlet state may be overcome by this intramolecular energy transfer.
Thus, theoretically, the quantum efficiency of metal chelates containing lan-
thanides is four times higher than that exhibited for standard electroluminescent
organic materials, where only the singlet state decays by radiative emission.
The coumarin (32), the coronene (33), the triarylamine (M), the quinacroline
(35), the dimethylamino-substituted oxadiazole (36) and the naphthalimide (37)
listed in Table 5.8 are organic electroluminescent materials rather than organo-
metallic chelates. However, all of these materials have been found to exhibit
light in the green region of the electromagnetic spectrum, despite a wide
variation in molecular structure.23p24 Coumarin 6 (32) is a commercial dye used
'
in some laser applications. ',12 Coronene 33 is the conjugated polyaromatic core
unit of the columnar liquid crystals referred to in Table 5.5."' The aromatic,
heterocyclic compounds (34-36) can also be used as transport layers of either
positive or negative charge-carriers. 17*19~1 1291l 3 Naphthalimide 37 also emits
green light in spite of the small aromatic core of this particular class of organic
The materials collated in Tables 5.7 and 5.8 may be used either as discrete,
combined electron-transport-and-emission layers or as electroluminescent
dopants at low concentration in a non-emissive, electron-transporting matrix.
Many of these materials are self-quenching, so that the luminescent efficiency of
doped layers of these materials in an inert matrix, even at very low concentra-
tions ( ~ 0 . 5 %< c < 5 %) can exhibit higher quantum efficiencies than those
exhibited by thin layers of the pure material itself.
Red Electroluminescent, Low-Molar-Mass Materials

The molecular structures 38-43 shown in Table 5.9 are typical of red, low-
molar-mass electroluminescent materials. The number of efficient low-molar-
mass electroluminescent materials emitting in the red is relatively limited. The
''
europium complexes (38 and 39)' 14,' exhibit narrow emission spectra in the
red part of the visible spectrum. The first europium complexes synthesised, such
as compound 38, were not stable enough to be deposited by physical vapour
Table 5.9 Emission maximum (nm) and colourfor the compounds (38-43)
~~ ~
Molecular structure Acronym A,, Colour Ref
38 Eu(TT'FA)3 620 Red 114
39 Eu(lTFA)3phen 613 Red 115
DCM-1 570-620 Red 12
Rhodamine Orange 24
metal
complex
42 %=:-$ \ / \ / TBPD Red 24,105
Perylene 600 Orange 105
deposition. However, modified complexes, such as compound (39), which could

be deposited by this method were subsequently developed. The coumarin dye
-
(40) was used at very low concentrations ( ~ 0 . 5 % as
) a red chromophore with
high quantum efficiency (Q, = 2 3%) in an ETL of Alq3 (3)12 and or
na~hthalimide"~ as part of a bilayer OLEDs. The efficiency of these OLEDs
with a doped emission layer was found to be higher than that of the
170 Chapter 5
corresponding OLEDs with non-doped emission layers. However, even moder-
ately low dopant concentrations (z2%) led to a much lower quantum efficiency
due to self-quenching. It was speculated that charge recombination and, there-
fore, emission takes place on the coumarin molecules. The emission spectrum of
the rhodamine metal complexes (41)24can be modulated by varying the nature
of the counter ion, X. More importantly the thermal stability of these
compounds depends crucially on the nature of the counter ion. Only a limited
number of modified rhodamine metal complexes (41; X = GaCl;, TaClTand
InCl;), known originally as red laser dyes, are stable enough to be deposited by
physical vapour d e p ~ s i t i o n .The
~ ~ perylene derivative (42),24*'05and even
perylene (43)'" itself, have been used as red or orange dyes, respectively, in
combined electron-transporting and emission layers in multilayer OLEDs.
The colour of perylene varies from blue in dilute solution to yellow in the pure
crystalline state due to excimer formation at relatively modest concentrations. A
shift in the emitted colour or indeed complete quenching of electroluminescence
is a problem for many compounds used as efficient chromophores or dye-
dopants in OLEDs. Thus, the external quantum efficiency of an OLED with a
doped emission layer can be higher for lower concentrations of the dopant. The
large t-butyl substituents on the perylene derivative 42 serve to inhibit such
molecular association by steric hindrance. This, in turn, limits the degree of self-
quenching.
Blue Electroluminescent, Low-Molar-Mass Materials

Some typical low-molar-mass, organic electroluminescent materials (44-49),
which emit in the blue region of the electromagnetic spectrum are listed in Table
5.10. The first blue emitter used in OLEDs was anthracene (1) shown in Table
5.I. However, this is a highly crystalline material. Therefore, other organic
materials were required, which form the glassy state at room temperature with a
high T,. The oxazoline (44)exhibits blue photoluminescence and electrolumi-
nescence in the pure state. However, the formation of exiplexes in the presence
of a HTL, such as TPD, in a bilayer device gives rise to emission of green rather
than blue light.18 The oxadiazole (45) is a good electron-transport material and
a highly efficient blue emitter. The non-linear structure tends to inhibit crystal-
lisation.22The tetraphenylbutadiene (46),'8*23*249'the triarylamine (47),18 the
pentaphenyl-cyclopentadiene (48)18 and the arylvinylene (49)' all emit blue
light. The triarylamine (47)l8 functions well as an efficient hole-transport and
blue-emission layer.
The molecular structures (5&54) collated in Table 5.11 are typical of blue
electroluminescent metal chelates. The azomethin-zinc compound (50) exhibits
bright blue electroluminescence. l7 The aluminium metal chelates (51-53)
illustrate attempts to blue-shift the green emission of Alq3 (3) by modifying the
number and nature of the l i g a n d ~However,
.~~ the unfilled co-odination sites of
the aluminium compound (52) render this material intrinsically unstable. The
presence of the methyl substituents on compound 52 shields this material from
nucleophilic attack. This substantially increases the stability of this compound.
Organic Light-Emitting Diodes Using Low-Molar-Mass Materials (LMMMs) 17 1
Table 5.10 Emission maximum (nm) and colour for the aromatic compounds
(44-49)
Molecular structure Acronym Amax Colour Ref
450 Blue 18
45 OXD-4 470-480 Blue 22
46 TPB 430 Blue 18
47 TPA Blue 18
48 PPCP 465 Blue 18
49 DPVB 440490 Blue 116

172 Chapter 5
Molecular structure Acronym A,, Colour Ref
50 Blue 117
51 490 Blue 25
52 51 1 Blue 25
53
$rp 0 \
440 Blue 25
54 500 Blue 107
The presence of the electron-withdrawing nitrogen atom at position 5 of the

compound (53)results in a significant hypsochromic shift to the blue (90nm)
compared to the emission maximum of Alq3 (3). The metal chelate (54) with the
group I1 magnesium ion emits bright blue
6 Performance of OLEDs Using LMMMs

The technical data for a commercial prototype OLED using low-molar-mass
materials is collated in Table 5.12.'18s'19The brightness and the contrast ratio of
a green monochrome OLED with a 256 x 64 pixel dot matrix is comparable
with that of a standard commercially available TN-TFT-LCD with active
matrix addressing. However, emission of light from an OLED is Lambertian
and, therefore, the viewing-angle dependence of contrast of OLEDs is greatly
superior to that of most commercially available LCDs. Only LCDs with the
optical axis in the plane of the device, such as IPS-TFT-LCDs with active matrix
addressing (see Chapter 3) exhibit equally broad viewing angles. Although the
power consumption of this OLED panel is higher than that of a comparable
LCD with backlighting, the power consumption, external quantum efficiency
and luminous efficiency of OLEDs using low-molar-mass materials are compar-
able or superior to that of inorganic LEDs of comparable size, complexity and
colour. Suitably encapsulated and packaged OLEDs exhibit adequate life-times
under normal driving conditions. Therefore, the electro-optical performance of
OLEDs using LMMMs is clearly sufficient for commercial applications. Indeed,
monochrome OLEDs of this type are already commercially available from the
Pioneer Corporation for car dashboard applications. Active matrix OLED
panels with full-colour and high-information content, being developed jointly
by Sanyo and Kodak, are expected in 2001.
Stability of OLEDs Using LMMMs

The temporal stability of OLEDs using low-molar-mass organic compounds
Table 5.12 Conjiguration and typical electro-optical performance of a

typical bilayer 0LED1l8 using the low-molar-mass, aromatic
diamine TPD (12)23*34as HTL and Alq3 (3)14 doped with a
quinacridone dervative (35)Il2 as the ETL and emission layer
~
Property Value
Cathode Algg-Li
Anode IT0
Hole transport layer (HTL) (12)
Electron transport layer (ETL) and emission layer (35)
Threshold voltage, V,, 2.5 V
Operating voltage, V,, 5.5 v
Quantum efficiency, 4 1%
Power conversion efficiency 0.46%
Luminous efficiency 7.4 Im W-'
Current density, I lo-'- lo2 mA cm-2
Colour Green
Emission maximum, Amax 520 nm
Bandwidth 200 nm
Brightness 300 cd m-'
I74 Chapter 5
has been the prime factor over the last 30 years in inhibiting the successful
commercialisation of OLEDs using small molecules, i.e. the performance of the
devices is degraded significantly during the projected life-span of a normal FPD
(> 10 000 h). Although most of the organic materials used as an ETL, HTL and
emission layer form an amorphous glassy state at room temperature, they still
tend to crystallise slowly over time, especially at temperatures above room
temperature where the rate of recrystallisation is a c ~ e l e r a t e d . The
~ ~ ~small
~~~’~~
crystals formed initially grow gradually as smaller crystals coalesce into larger
crystals. This leads to the formation of discrete grain boundaries between
individual crystals. These defects and inhomogenities act as traps and quench-
ing sites and slowly reduce the quantum efficiency of the display. The crystal-
lisation also disrupts the interface between the electrodes and adjacent organic
material, where covalent bonding between the electrode metal and the organic
material may exist. 3 3 ~ 1 2 1This mechanical stress can even lead to delamination
of the electrode surface. The heterojunction in multilayer OLEDs may also be
corrupted by the process of crystallisation of one or both of the layers either side
of the junction. This can also lead to the diffusion of one material into the
other.’22Since the layers are very thin, this can have severe consequences for the
life-time of the display. The electrodes may also degrade over time in the
presence of oxygen or moisture giving rise to dark spots due to oxidation.’23
Electrochemical degradation and excited-state chemical reactions of the organic
material may also be contributory factors to the degradation of device
performance and eventual failure, although their exact nature may be difficult
to identify and monitor.24 These reactions deplete the number of electroactive
species. Furthermore, the reaction products then may act as quenching sites,
which then reduce the device efficiency even further. An AC drive form instead
of a DC driving voltage may also extend the device lifetime.’24Therefore, it is
evident that the stability of OLEDs using low-molar-mass organic compounds
is dependent upon a whole range of interrelated factors associated not only with
the organic materials used, but also the device configuration, the nature of the
electrodes and the device packaging.337’07”25”26.L27
The use of insoluble, highly cross-linked anisotropic networks created by the
polymerisation of photoreactive monomers, eliminates the problem of crystal-
lisation, at least for organic materials, since polymer networks are macromole-
cular structures incapable of crystallising, see Chapter 6 . Furthermore, the
fabrication of multilayer devices would be facilitated by the use of a cross-linked
stable HTL next to the anode on the solid substrate surface, onto which
subsequent layers can be deposited by vapour deposition. Multilayer OLEDs
are intrinsically more stable than monolayer devices due to a better balance of
charge-carriers and concentration of the charged species away from the
electrodes. The synthesis and cross-linking of a suitable aromatic triarylamine
derivative with a polymerisable oxetane group at each end of the molecule for
use as a HTL has been reported recently.12*
An additional advantage of the use of reactive, photopolymerisable liquid
crystalline monomers with charge transfer or electroluminescent properties is
the ability to generate circularly and linearly polarised light.53-56This possibility
has many potential applications in flat panel display device applications,
especially as a source of polarised backlighting for LCDs with improved
brightne~s.’~
7 References
1 W. Helfrich and W. G. Schneider, Phys. Rev. Lett., 1965,14,229.
2 W. Helfrich and W. G. Schneider, J. Chem. Phys., 1966,44,2902.
3 S . Miyata and H. S. Nalwa, in ‘Organic Electroluminescent Materials and Devices’
Gordon and Breach, New York, USA, 1997.
4 W. Mehl and W. Bucher, 2. Phys. Chem., 1965,47,76.
5 D. F. Williams and M. Schadt, Proc. IEEE, 1970,58,476.
6 J. Dresner, RCA Rev., 1969,30, 322.
7 R. E. Kellog, J. Chem. Phys., 1966,44,411.
8 P. S. Vincett, W. A. Barlow, R. A. Hann and G. G. Roberts, Thin Solid Films, 1982,
94, 171.
9 F. J. Campas and M. Goutermann, Chem. Phys. Lett., 1977,48,233.
10 J. Kalinowski, J. Godlewski and Z. Dreger, Appl. Phys. Lett. A., 1985,37, 179.
1 1 C. W. Tang and S. A. Van Slyke, Appl. Phys. Lett., 1987,51,913.
12 C. W. Tang, S. A. Van Slyke and C. H. Chen, Appl. Phys. Lett., 1989,65,3610.
13 M. Abkowitz and D. M. Pai, Philos. Mag., B., 1986,53, 193.
14 D. C . Freeman and C. E. White, J. Am. Chem. SOC.,1956,78,2678.
15 M. Era, C . Adachi, T. Tsutsui and S. Saito, Chem. Phys. Lett., 1991,178,488.
16 C. Adachi, S. Tokito, T. Tsutsui and S. Saito, Jpn. J. Appl. Phys., 1988,27, L713.
17 C . Adachi, T. Tsutsui and S. Saito, Appl. Phys. Lett., 1989,55, 1489.
18 C. Adachi, T. Tsutsui and S. Saito, Appl. Phys. Lett., 1990, 56, 799 and 1990, 57,
531.
19 C. Adachi, T. Tsutsui and S. Saito, Acta Polytech. Scand. Appl. Phys., 1990,170,145
and 215.
20 Y. Hamada, C. Adachi, T. Tsutsui and S. Saito, Jpn. J. Appl. Phys., 1992,31, 1812.
21 A. Dodabalapur, L. J. Rothberg and T. M. Miller, Appl. Phys. Lett., 1994,65,2308.
22 J. Kido, M. Kimura and K. Nagaim, Science, 1995,67, 1332.
23 J. Kido, Bull. Electrochem., 1994,10, 1.
24 C. H. Chen, J. Shi and C. W. Tang, Macromol. Symp., 1997,125,l.
25 C. H. Chen, J. Shi and C. W. Tang, Coord. Chem. Rev., 1998,171,161.
26 Y. Hamada, C. Adachi, T. Tsutsui and S. Saito, Optoelectronics, 1992,7,83.
27 M. Ohta, Y. Sakon, T. Takahashi, C. Adachi and K. Nagai, USA Pat., 5,420,288,
1995.
28 S . Saito, T. Tsutsui ,C. Adachi and Y. Hamada, USA Par., 5,382,477, 1995.
29 J. Kido, C. Ohtaki, K. Hongawa and K. Nagai, Jpn. J. Appl. Phys. Part 2, 1993,32,
L9 17.
30 J. Salbeck, Ber. Bunsenges. Phys. Chem., 1996,100, 1667.
31 J. Salbeck, N. Yu, J. Bauer, F. Weissortel and H. Bestgen, Synth. Met., 1997, 91,
209.
32 J. Salbeck, F. Weissortel and J Bauer, Macromol. Symp., 1997,125, 121.
33 J. R. Sheats, H. Antoniadis, M. Hueschen, W. Leonard, J. Miller, R. Moon,
D. Roitman and A. Stocking, Science, 1996,273,884.
34 M. Stolka, J. F. Yanus and D. M. Pai, J. Phys. Chem., 1984,88,4707.
35 S . Tokito, H. Tanaka, A. Okada and Y. Taga, Appl. PAYS. Lett., 1996,696,878.
36 T. Sano, T. Fuji, Y. Nishio, Y. Hamada, K. Shibata and K. Kuroki, Jpn. J. Appl.
Phys., 1995,34,3124.
37 G. H. Heilmeier and P. M. Heyman, Phys. Rev. Lett., 1967,18, 583.
38 K. Yoshino, S. Hisamatsu and Y. Inuishi, J. Phys, SOC.Jpn., 1972,32,867.
176 Chapter 5
39 T. Yanagisawa, H. Matsumoto and K. Yahagi, Jpn. J. Appl. Phys., 1977,16,45.
40 M. Yamashita and Y .Ameniya, Jpn. J. Appl. Phys., 1978,17, 1513.
41 A. Sugimura, N. Matsui, H. Takahashi, H. Sonomura, H. Naito and M. Okuda,
Phys. Rev. B., 1991,43,8272.
42 K. Okamoto, S. Nakajima, M. Ueda, A. Itaya and S. Kusabayashi, Bull. Chem. SOC.
Jpn., 1983,56,3545.
43 K. Okamoto, S. Nakajima, M. Ueda, A. Itaya and S. Kusabayashi, Bull. Chem. SOC.
Jpn., 1983,56, 3830.
44 H. Naito, M. Okuda and A. Sugimura, Phys. Rev. A., 1991,44,3434.
45 H. Naito, K. Yoshida, M. Okuda and A. Sugimura, Jpn. J. Appl. Phys., 1993, 73,
1119.
46 S. Murikami, H. Naito, M. Okuda and A. Sugimura, J. Appl. Phys., 1995, 78,
4533.
47 M. Funahashi and J.-I. Hanna, Jpn. J. Appl. Phys., 1996,35, L703.
48 M. Funahashi and J.-I. Hanna, Phys. Rev. Lett., 1997,78,2184.
49 H. Tokuhisa, M. Era and T. Tsutsui, Appl. Phys. Lett., 1998,72,2639.
50 H. Tokuhisa, M. Era and T. Tsutsui, Adv. Muter., 1998,8,2639.
51 M. Funahashi and J.-I. Hanna, Appl. Phys. Lett., 1997,71,602.
52 M. Funahashi and J.-I. Hanna, Mol. Cryst. Liq. Cryst., 1999,331, 516.
53 M. Grell and D. D. C. Bradley, Adv. Muter., 1999,11,895.
54 S . H. Chen, H. Shi, B. M. Conger, J. C. Mastrangelo and T. Tsutsui, Adv. Mater.,
1996,8,998.
55 B. M. Conger, H. Shi, S. H. Chen and T. Tsutsui, Mat. Res. SOC.Symp. Proc., 1996,
425,239.
56 S . H. Chen, D. Katsis, A. W. Schmidt, J. C. Mastrangelo, T. Tsutsui and T. N.
Blanton, Nature, 1999,397,506.
57 S. Farrar, A. E. A. Contoret, J. E. Nicholls, M. O’Neill, S. M. Kelly and A. J.
Eastwood, Appl. Phys. Lett., 2000,76,2553.
58 A. E. A. Contoret, S. Farrar, L. May, J. E. Nicholls, M.O’Neill, S. M. Kelly and
G. J. Richards, Adv. Mater., 2000, 12,971.
59 A. E. A. Contoret, S. Farrar, S. M. Kelly, J. E. Nicholls, M. O’Neill and G. J.
Richards, J. Chem. SOC.Chem. Commun., in press.
60 G. Liissem and J. H. Wendorff, Pofym.Adv. Technol., 1998,9,443.
61 B. Glusen, W. Heitz, A. Kettner and J. H. Wendorff, Liq. Cryst., 1996,20,627.
62 D. Adam, D. Haarer, F. Closs, T. Frey, D. Funhoff, K. Siemensmeier, P.
Schumacher, and H. Ringsdorf, Ber. Buns. Phys. Chem., 1993,97, 1366.
63 D. Adam, F. Closs, T. Frey, D. Funhoff, D. Haarer, P. Schumacher, H. Ringsdorf
and K. Siemensmeier, Phys. Rev. Lett., 1993,70,457.
64 C. Catry, M. van der Auweraer, F. C. de Schryver, H. Bengs, K. Haussling, 0.
Karthaus and H. Ringsdorf, Makromol. Chem., 1993,194,2985.
65 D.,Adam, P. Schumacher, J. Simmerer, K. Haussling, K. Siemensmeier, K. H.
Etzbach, H. Ringsdorf and D. Haarer, Nature, 1994,371, 141.
66 E. 0. Arikainen, N. Boden, R. J. Bushby, J. Clemments, B. Movaghar and W.
Wood, J. Mater. Chem., 1995,5,2165.
67 J. Simmerer, D. Adam, P. Schumacher, W. Paulus, K. H. Etzbach, K. Siemensme-
ier, H. Ringsdorf and D. Haarer, Proceedings of the 26thArbeitstagung Flussigkris-
tulle, Freiburg, Germany, 1996.
68 P. G. Schouten, J. M.Warman, M. P. de Haas, C. F. van Nostrum, G. H. Gelinck,
R. J. M. Nolte, M, J. Copyn, J. W. Zwikker, M. K. Engel, M. Hanack, Y. H. Chang
and W. T. Ford, J. Am. Chem. Soc., 1994,116,6880.
69 P. G. Schouten, J. M.Warman, M. P. de Haas, J. F. van der Pol and J. W. Zwikker,
J. Am. Chem. SOC.,1992,114,9028.
70 B. Blanzat, C. Bartou, N. Tercier, J.-J. And& and J. Simon, J. Am. Chem. SOC.,
1987,109,6193.
71 S. Chandrasekhar, B. K. Sadashiva, and K. A. Suresh, Pramana, 1977,9,471.
72 S. Chandrasekhar, in ‘Advances in Liquid Crystals’, Acadademic Press, New York,
USA, 1982, Vol. 5.
73 S. Chandrasekhar, Mol. Cryst. Liq. Cryst., 1993,14,3.
74 J. Billard, J. C. Dubois, C. Vaucher and A. M. Levelut, Mot. Cryst. Liq. Cryst.,
1981,66, 115.
75 C . Destrade, J. Malthete, H. T. Nguyen and H. Gasparoux, Phys. Lett., 1980,78A,
82.
76 R. Fugnitto, H. Strzelecka, J. C. Zann, J. C. Dubois and J. Billard, J. Chem. SOC.
Chem. Commun., 1980,271.
77 J. W. Goodby, P. S. Robinson, B. K.Teo and P. E. Cladis, Mol. Cryst. Liq. Cryst.,
Lett., 1980,56, 303.
78 J. D. Bunning, J. W. Goodby, G. W. Gray and J. B. Lydon, Proc. Conf. Liq. Cryst.
Two-Dimensional Order, Garmisch-Partenkirchen,1980,397.
79 C. Destrade, M. C. Mondon and J. J. Malthete, J. de Phys. (Paris), Coll. C. 3, 1979,
40,17.
80 C. Destrade, N. H. Tinh, H. Gasparoux, J. Malthete and A. M. Levelut, Mot. Cryst.
Liq. Cryst., 1981,71, 111.
81 N. H. Tinh, M. C. Mondon-Bernaud, G. Sigaud and C. Destrade, Mol. Cryst. Liq.
Cryst., 1981,65, 307.
82 C. Destrade, P. Foucher, H. Gasparoux, H. T. Nguyen, A. M. Levelut and
J. Malthete, Mot. Cryst. Liq. Cryst., 1984, 106, 121.
83 C. Mertesdorf, H. Ringsdorf and J. Stumpe, Liq. Cryst., 1991,9, 337.
84 I. M. Matheson, 0.C. Musgrave and C. J. Webster, J. Chem. SOC.Chem. Commun.,
1968,278.
85 N. Boden, R. C. Borner, R. J. Bushby, A. N. Cammidge and M. V. Jesudason, Liq.
Cryst., 1993,15, 851.
86 N. Boden, R. J. Bushby and A. N. Cammidge, J. Chem. SOC.Chem. Commun., 1994,
465.
87 R. C. Borner and R. F. W. Jackson, J. Chem. SOC.Chem. Commun., 1994,845.
88 J. W. Goodby, M. Hird, K. J. Toyne and T., Watson, J. Chem. SOC.Chem.
Commun., 1994,1701.
89 N. Boden, R. J. Bushby, A. N. Cammidge and G. Headdock, J. Mater. Chem., 1995,
5,2275.
90 P. Henderson, H. Ringsdorf and P. Schumacher, Liq. Cryst., 1995,18, 191.
91 F. Closs, L. Haussling, P. Henderson, H. Ringsdorf and P. Schumacher, J. Chem.
SOC.Perkin Trans. I , 1995,829.
92 P. Henderson, S. Kumar, J. A. Rego, H. Ringsdorf and P. Schumacher, J. Chem.
SOC.Chem. Commun., 1995,1059.
93 N. Boden, R. J. Bushby, A. N. Cammidge and G. Headdock, Synthesis, 1995,31.
94 P. Henderson, H. Ringsdorf and P. Schumacher, Liq. Cryst., 1995,18, 191.
95 N. Miyaura T. Yanagi and A. Suzuki, Synth. Commun., 1981,11,513.
96 M. Hird, G. W.,Gray and K. J. Toyne, Mot. Cryst. Liq. Cryst., 1991,206, 187.
97 S. J. Cross, M. Hird, A. W. Hall, J. W. Goodby, S. M. Kelly, K. J. Toyne and
C. Wu, Liq. Cryst., 1998,25, 1 .
98 A. M. van de Craats, J. M. Warman, K. Mullen, Y. Geerts and J. D. Brand, Adv.
Muter., 1998,10, 36.
99 A. M. van de Craats, J. M. Warman, A. Fechtenkotter, J. D. Brand, M. A. Harbison
and K. Mullen, Adv. Mater., 1999,11, 1469.
100 S. Kumar, D. S. S. Roa and S. K. Prasad, J. Mater. Chem., 1999,9,2751.
101 H.-T. Nguyen, C. Destrade and J. Malthite, Adv. Mater., 1997,9, 375.
102 S. Farrar, J. E. Nicholls, M. O’Neill, S. M. Kelly, G. J. Richards and C. Wu, Liq.
Cryst., in press.
103 M. Matsumura and T. Akai, Jpn. J. Appl. Phys., 1996,35, 5357.
104 P. E. Burrows, 2. Shen, V. Bulovic, D. M. McCarthy, S. R. Forrest, J. A. Cronin
and M. E. Thompson, J. Appl. Phys., 1996,79,79991.
178 Chapter 5
105 C. Adachi, S. Tokito, T. Tsutsui and S. Saito, Jpn. J. Appf. Phys., 1988,27, L269.
106 Y. Hamada, T. Sano, M. Fujita, T. Fujii, Y. Nishio and K. Shibata, Jpn. J. Appf.
Phys., 1993,32, L514.
107 Y. Hamada, T. Sano, K. Shibata and K. Kuroki, Jpn. J. Appf.Phys., 1995,34, L824.
108 Y. Hamada, T. Sano, H. Fujii, Y. Nishio, H. Takahashi and K. Shibata, Jpn. J.
Appl. Phys., 1996,35, L1339.
109 M. Strukelj, R. H. Jordanand A. Dodabalapur, J. Am. Chem. SOC., 1996,118,1213.
110 J. Kido K. Nagai and Y. Ohashi, Chem. Lett., 1990,657.
11 1 S. Dirr, H.-H. Johannes, J. Schobel, D. Ammermann, A. Bohler, W. Kowalsky and
W. Grahn, SID’97 Digest, 1997,778.
112 J. Shi and C. W. Tang, Appl. Phys. Lett., 1997,70,1665.
113 K. Utsugi and S. Takano, J. Electrochem. SOC.,1992,139,3610
114 J. Kido, N. Nagi, Y. Okamoto and T. Skotheim, Chem. Lett., 1991,1267.
115 R. A. Campos, I. P. Kovalev, Y. Guo, N. Wakali and T. Skotheim, J. Appf. Phys.,
1996,80,7144.
116 H. Tokolin, M. Matsura, H. Higashi, C. Hosokawa and T. Kusumoto, Proc SPIE,
1993,38,1910.
117 Y. Hamada, T. Sano, M. Fujita, T. Fujii, Y. Nishio and K. Shibata, Jpn. J. Appf.
Phys., 1993,32, L511.
118 T. Wakimoto, S. Kawami, K. Nagayama, Y. Yonemot, R. Murayama, J. Funaki,
H. Sato, N. Nakada and K. Imai, Asia Dispfay ‘95, 1995,77.
119 T. Wakimoto, R. Murayama, K. Nagayama, Y. Okuda, H. Nakada and T. Thoma,
SID ‘96Digest, 1996,849.
120 E.-M. Han, L.-M. Do, Y. Niidome and M. Fujihara, Chem. Lett., 1994,969.
121 L.-M. Do, E.-M. Han, Y. Niidome, M. Fujihara, T. Kanno, S. Yoshida, A. Maeda
and A. J. Ikushima, J. Appl. Phys., 1994,76,5118.
122 L.-M. Do, E.-M. Han, Y. Niidome and M. Fujihara, Mol. Cryst. Liq. Cryst., 1995,
267,411.
123 P. E. Burrows, V. Bulovic, S. R. Forrest, L. S. Sapochak, D. M. McCarty and M. E.
Thompson, Appf. Phys. Lett., 1994,65,2922.
124 S. A. Van Slyke, C. H. Chen and C. W. Tang, Appf. Phys. Lett., 1996,69,2160.
125 C. Adachi, K. Nagi and N. Tomato, Appf. Phys. Lett., 1995,66,2679.
126 Y. Hamada, T. Sano, M. Fujita, Y. Nishio and K. Shibata, Chem. Lett., 1993,905.
127 H. Antoniadis, M. R. Hueschen, J. N. Miller, R. L. Moon, D. B. Roitman and J. R.
Sheats, Macromol. Symp., 1997,125,59.
128 M. S. Bayerl, T. Braig, 0. Nuyken, D. C. Miiller, M. GroIj and K. Meerholz,
Macromof. Rapid Commun., 1999,20,224.
CHAPTER 6
Organic Light-Emitting Diodes

Using Light-Emitting Polymers
1 Introduction
There are many configurations of electroluminescent display devices'-14 using
very many different organic polymer^'^'^, which correspond to the analogous
configurations using low-molar-mass materials, see Chapter 5. Single-layer
devices demonstrated the principle of LEDs incorporating organic conjugated
semiconductor polymers. '-' However, bilayer and trilayer devices usually
exhibit higher efficiency, lower threshold and operating voltage as well as lower
current. This is due to a more balanced charge injection and transport with
recombination of the holes and electrons taking place at the interface between
layers and not at the interface between the organic material and one of the
electrodes. Therefore, the lifetime of multilayer devices tends to be significantly
longer than those of comparable single-layer OLEDs.
2 Monolayer Organic Light-Emitting Diodes (OLEDs)

Using LEPs
During evaluation of the physical properties of a conjugated organic polymer
consisting of a regular alternating pattern of a 1,4-disubstituted phenylene ring
and a trans-carbon-carbon double (vinylene) bond, i.e. poly(p-phenyleneviny-
lene) [PPV], compound 1, see Table 6.1, at the Cavendish Laboratories in
Cambridge, UK, it was discovered that this semiconducting polymer emitted
yellow-green light, if a sufficiently large electric current passed through it.
A thin film ( ~ 1 0 0nm) of a PPV derivative is sandwiched between two
electrodes supported on a solid substrate, such as glass, a flexible plastic
substrate, such as poly(propylene), or even a thin film of aluminium, see
Figure 6.1. The device substrate provides the mechanical support to the anode,
electroluminescent polymer layer and the cathode. The anode is transparent in
179
180 Chapter 6
Table 6.1 Emission maximum (nm) and colourfor the PPV derivatives (Id)
~~
Molecular structure A,, Colour Ref
550 Green 1
626 Red 2-5
-0 J n
523 Green 3640
585 Red 41,42
625 Red 43
6 570 Yellow 44
Organic Light-Emitting Diodes Using Light-Emitting Polymers 181
METALLIC
Q
CATHODE
iimf I
POLYMER
0 0 00000 0 0
HTL/ETLAND
EMITTER
1
TRANSPARENT
ANODE
TRANSPARENT
SUBSTRATE
Figure 6.1 Schematic representation of a monolayer OLED incorporating an electrolu-

minescent conjugatedpolymer between a transparent anode and a cathode.
order to allow the light generated by electroluminescenceto actually leave the

display and be observed. The electrodes are connected to a source of potential.
Practical displays will also be encapsulated within an inert polymer coating,
especially next to the metallic cathode, in order to prevent the intrusion of
oxygen and moisture into the display. The cathode, e.g. a thin layer (x7-1 5 nm)
of aluminium or calcium, has a low work function, whereas the anode, e.g. a
thin metallic layer ( 15 nm) of indium-tin oxide (ITO) or an organic layer such as
poly(ani1ine) salt (e.g. PANi-camphor sulfonate) or a sulfonic acid derivative of
polyethylenedioxythiophene (PEDOT), has a high work function. No light is
emitted in the absence of an electric current and so the non-activated pixels
appear dark. The application of a voltage between the two electrodes above the
threshold voltage, Vth, leads to the injection of electrons from the cathode and
holes from the anode. The high work function of the electrode should match as
closely as possible the highest occupied molecular orbital (HOMO) of the
electroluminescent polymer. Most electroluminescent polymers are organic
semiconductors with a high intrinsic resistivity, e.g. 1OI2 SZ cm for PPV (similar
in value to that of typical nematic liquid crystals, which are usually regarded as
dielectrics). Although PPV preferentially transports holes, their mobility is still
low cm2 V-' s-' > p h > lo-* cm2 V-' s-I). This means that high
',
electric fields, e.g. lo5 V cm- are required in order to induce an electric current
in the organic layer. However, the polymer layer required for bright, visible
electroluminescence is relatively thin, e.g. x 100 nm (even monolayers will emit
easily detectable visible light). Consequently, the threshold voltages and
operating voltages for OLEDs using LEPs are primarily determined by the
thickness of the organic layer, the compatibility of the work functions of the
electrodes and the LUMO and HOMO of the conjugated polymer. Therefore,
threshold voltages and operating voltages for monolayer OLEDs using LEPs
are usually relatively low, e.g. 2-10 V for a layer of PPV of standard thickness,
e.g. 100 nm.
182 Chapter 6
Upon application of an electric field to an OLED very little current flows until
a threshold field is reached. Above this threshold value light is emitted from
activated pixels due to the fast radiative decay ( Z 1 ns) of excitons formed by the
recombination of a hole and an electron for voltages, V > Vth. Current flows
according to the relationship:
Ia ~2 exp($)
assuming carrier injection takes place by a tunnelling mechani~rn.~-~ Different

current-voltage characteristics are observed if the current is limited by the
mobility of the polymer.' The rectification ratio of these diodes, i.e. the ratio of
current with forward bias to that with reversed bias, is often high, e.g. lo4. The
current flowing through an OLED is low, e.g. ~ 1 0 - 5 0mA cm-2. Thus the
operating voltages and power consumption of OLED using LEPs are compa-
tible with use in battery-operated instruments.
The external quantum efficiency, qext, of monolayer OLEDs, i.e. the number
of photons actually seen by the observer, is related to the number of photons
emitted for every electron injected and is usually low, e.g. 0.05% for the first
OLEDs using PPV.' The observed external efficiency, qext, is much lower than
the internal efficiency, qjnty as given by the following r e l a t i ~ n s h i p : ~ ~ ~
where n is the refractive index of the polymer. The external quantum efficiency is
substantially lower than the internal quantum efficiency, since the refractive
index of conjugated aromatic main-chain polymers is high, e.g. n = 1.4 for PPV.
This suggests that a significant fraction of the light generated is lost within the
device itself due to internal reflection and re-absorption leading primarily to
non-radiative decay, since, generally, only the singlet excited state relaxes to the
ground state with the emission of visible light. The excitons generated by
electroluminescence on conjugated polymers can be represented as a quantum
mechanical wave, which is localised on one part of the extended polymer chain.
The energy of the triplet state is much lower than that of the singlet state and the
triplet state usually relaxes to the ground state by a non-radiative pathway.
Since symmetry considerations stipulate that three triplets are formed for every
singlet, then the maximum luminance efficiency, i.e. the percentage of excitons
which decay by a radiative pathway, is postulated to be 25%.' Quenching at
defects and traps can reduce the quantum efficiency significantly. The formation
of inter-chain excimers or exciplexes, consisting of one part of a polymer chain
in the ground state and one in the excited state, also decay non-radiatively. A
substantial fraction of radiative energy is also subsequently lost as heat rather
than as visibly emitted light. Therefore, it is not surprising that external
quantum efficiencies for OLEDs are intrinsically low. However, acceptable
luminance efficiency can still be achieved for monolayer OLEDs, e.g. 3 lm W -
for PPV.
Organic Light- Emitting Diodes Using Light-Emitting Polymers 183
Different energy barriers for the injection of holes or electrons exist for an
organic polymer with a band gap corresponding to the visible spectrum. This is
due to the difference in the work functions (energy required) for the injection of
an electron to an organic material (electron affinity) and the injection of a hole,
i.e. removal of an electron, from the same material (ionisation potential).
Therefore, either holes or electrons will preferentially be injected and the
charge transport will be unbalanced in favour of one species of charge carrier.
This will result in the build up of one charge carrier species at the interface
between the organic polymer and either the cathode or anode, where quenching
of electroluminescenceoccurs. Most conjugated organic polymers preferentially
transport holes rather than electrons due to their electron-rich aromatic nature,
which tends to trap electrons. Therefore, charge capture and recombination
with subsequent exciton formation takes place near the cathode, see Figure 6.1.
This generally results in a low ( < 1-2 %) quantum efficiency, i.e. the
-
percentage of photons generated per electron injected, for monolayer OLEDs
at low drive voltage (z2 3 V). This may be partly attributable to substantial
quenching at the interface between the polymer and the metallic electrode. This
interface is not distinct due to a significant degree of covalent or ionic bonding
between the metal of the cathode and the organic polymer depending on the
nature of both. Indeed this composite metal-polymer interfacial layer between
the pure metallic and polymer layers may even lower the energy barrier for
charge injection and lead to higher currents at lower voltages. However, in spite
of all these possible pathways for loss of observable light, the luminance of
monolayer OLEDs may still be comparable with that (100 cd m-2) of a CRT or
a large-area, high-information-content LCD, such as a lap-top computer. The
latter instruments possess low light-transmission coefficients ( < 3%) and are
sometimes regarded as not particularly bright.6*'
The emission spectra of OLEDs are often broad due to the coupling of the
exciton energy with the vibrational structure of the polymer in the excited state.'
The shape of the polymer in the ground (n)state is different from that in the
slightly rearranged excited (n*) state. Therefore, vibrational modes lead to a
variation in the effective band gap, which leads to a broadening of the emission
spectrum. Therefore, PPV is observed to emit yellow-green light with two
maxima (Amax = 520 and 551 nm).' The observed colour of the polymers
collated in Tables 6.1-6.16 do not always correspond to Amax due to broad-
band emission. Much of the research on improving the properties of electro-
luminescent polymers has been devoted to manipulating the band gap between
the LUMO and HOMO energy levels in order to determine the colour of light
emitted and its bandwidth, see the following section. However, these values are
not always quoted, especially in those papers describing the synthesis of new
polymers for OLED applications and, therefore, some values are absent from
the tables. The absolute values of the LUMO and HOMO are designed to match
as closely as possible those of the electrode materials for a particular type of
OLED. These can be predicted from the molecular structure using standard
theories. l4
184 Chapter 6
Light-Emitting Polymers (LEPs)
The first reports of electroluminescence from an organic polymer matrix doped
with an organic electroluminescent dye in a 'Guest-Host' system failed to
attract a great deal of attention.'5916However, the report of electroluminescence
from poly(p-phenylenevinylene)(PPV) by researchers from the Cavendish
laboratory in Cambridge, UK,' followed by similar reports using a PPV
derivative by researchers from Santa Barbara, USA,24 stimulated enormous
interest and activity in organic light-emitting diodes (OLEDs) using light-
emitting polymers (LEPs) and the synthesis of new electroluminescentpolymers
with improved properties.l7-I9
The PPV used in the discovery of electroluminescence was prepared by a
soluble precursor route. However, the first synthesis of PPV predates the
discovery of its electroluminescent properties by several decades. It was first
synthesised using symmetrical phenyl phosphorous Wittig-Horner20921and
Wittig reagents22and 174-phthaldehyde.The resultant polymer was an infusible,
intractable solid with a high melting point, which was insoluble in organic
solvents. Related PPV derivatives, which were found to be soluble in certain
organic solvents, were also prepared using this However, the
molecular weight of these polymers, or indeed oligomers, was low on average.
Alternative routes to PPV and soluble PPV derivatives using a soluble
sulfonium precursor polymer were developed subsequently.2s27 These
methods were then modified24 in order to produce thin films so that their
semiconductor properties could be investigated. Thin transparent films were
prepared by depositing a soluble ionic precursor polymer on the desired
substrate by spin casting from solution followed by transformation into PPV
by heating the film at a high temperature (180"C-3OO0C) under vacuum for
several hours. This procedure has the considerable disadvantage of requiring
chemistry to be carried out on the device substrate at high temperatures for a
long time. This imposes limitations on the other device components commensu-
rate with these demanding processing conditions. The complete removal of side-
products and residual ionic impurities produced during the reaction is not a
trivial task. Traces of oxygen lead to the oxidation of parts of the polymer giving
rise to carbonyl groups such as aldehyde and acid derivatives. These may act as
traps and lead to quenching of electroluminescence, reducing the amount of
light emitted.28-30Therefore, the processing of the precursor polymer and heat
treatment to produce the fully conjugated PPV derivative by elimination are
carried out under a dry atmosphere of an inert gas, e.g. nitrogen. However, a
series of modified PPV polymers and copolymers with a wide spectrum of
physical properties has been successfully produced using this precursor method.
Processing temperatures below 160"C3155 allow plastic substrates, such as
poly(ethy1ene terephthalate) [PET] or poly(propy1ene) to be used.35
These problems were eliminated by the use of modified PPV
such as poly[2-methoxy-5-(2-ethylhexyloxy)-4-phenylene vinylene] (MEH-PPV)
compound 22, see Table 6.1, which are soluble in organic solvents. The presence
of lateral substituents in the polymers 3-6 shown in Table 6.1 induces a lower
melting point or glass transition temperature for these polymers compared to
that of PPV. Consequently these polymers are sufficiently soluble in organic
solvents. This removes the necessity of carrying out chemistry and thermal
processing on the substrate. The lateral substituents also represent a means of
modulating the LUMO and HOMO levels and, therefore, the wavelength of
emission of these PPV derivatives. For example, dialkoxy-substituted PPV
derivatives, such as compounds 2-62-5*3- exhibit a shifted emission compared
to that of PPV (l).’*45-47Branched chains such as that in MEH-PPV give rise to
the largest effects. However, if the supstituent is very large, such as in the
cholesteryl-substituted PPV derivative (6) then a dilution effect leads to lower
quantum efficiencies. Therefore, a compromise must be made between solubility
and quantum efficiency.
The ability to control the wavelength of emission and quantum efficiency,
&E, by synthesis is demonstrated by the structures of the modified PPV
derivatives 7-11 listed in Table 6.2. The quantum efficiency of polymer 9 is less
than half (24%) that (65%) of the similar polymer 11. The presence of one to
four additional phenylene rings in the compounds 7-1149-55compared to PPV
(1) shifts the wavelength of maximum emission by over 130 nm changing the
colour from blue-green to red-orange. The presence of the alkyl or alkoxy chain
renders most of these materials (7-10) soluble in organic solvents. Compound
11,” with two phenylene groups in a lateral position, must be prepared by the
precursor route for practical applications, since it is insoluble in organic
solvents. However, this lack of solubility in solvents may be of use in the
fabrication of multilayer devices, see Section 3.4.
The synthesis of copolymers based on PPV incorporating varying amounts of
different structural elements to yield compounds, such as those (12-14)5095’*56
collated in Table 6.3, allows fine-tuning of their physical properties by choice
and number of the structural elements to be combined and the molar ratio of
these elements. Once again the presence of long alkyl and alkoxy chains renders
high-molecular-weight copolymers, such as 12 and 13, soluble in certain organic
solvents. It can be seen that changing the molar ratio (n:m:p) of the three
constituent parts of the copolymer (13) allows the colour of emission to be
varied from green to yellow to orange. The use of a 1,3-disubstituted phenylene
ring in copolymer (13) contributes to the good solubility of this copolymer by
maintaining the axis of the lateral alkyl group parallel to the molecular long axis
of the copolymer.
The presence of a carbon-carbon triple bond in the poly(pheny1ene ethyny-
lenes) (15-17) shown in Table 6.4 gives rise to a fully conjugated structure
similar to that of the corresponding PPV derivatives with a carbon-carbon
double bond instead of the triple b ~ n d . ’ ~ -However,
~l the polymer backbone is
somewhat more linear. This results in a blue-shift for poly(pheny1ene ethyny-
lene) derivatives compared to the corresponding PPVs. Poly(pheny1ene ethyny-
lenes) such as those (15-17) listed in Table 6.4 are usually prepared by transition
metal-catalysed cross-coupling However, despite their ready
synthesis they do not appear to offer any appreciable advantages over the
corresponding PPV analogues in standard types of OLEDs. However, several of
186 Chapter 6
Table 6.2 Emission maximum (nm) and colour for the substituted PPV deriva-
tives (7-11)
Molecular structure ,A Colour Ref
7 620 Red-orange 49
8 515 Green 50,51
Q
9 560 Green 52
10 510 Blue-green 53
11 490 Blue-green 54
Organic Light- Emitting Diodes Using Light-Emit t ing Polymers 187
188 Chapter 6
Table 6.4 Emission maximum (nm) and colour for the copolymer PPV deriva-
tives (15-1 7 )
Molecular structure Amax Colour Ref
15 490 Blue 57
495 Blue 58-60
480 Blue-green 61
these polymers, e.g. 15 and 16, exhibit a thermotropic, enantiotropic nematic

phase and have been used to generate polarised light emission, see Section 4.5.
The polymers 18-26 containing the thiophene ring shown in Table 6.5 all
exhibit some common features, such as long carbon chains in lateral positions to
generate solubility in organic solvents and the 2,5-disubstitution att tern.^^-^^
The presence of the different lateral substituents determines their relative and
absolute solubility in organic solvents, melting point and glass transition
Molecular structure A,,- Colour Ref
18
an- 640 Yellow-orange 63-67
19
&- n
640 Yellow-orange 68,69
20 460 Blue 70
21 Green 70
22 555 Green 7 1-73
23 460 Blue 7 1-73
24 ) 490 Blue 75
(continued)
190 Chapter 6
Table 6.5 continued
Molecular structure A,,, Colour Ref
25 620 Red 76
26 590,605 Orange-red 77
temperature. The size of the lateral substituent affects the degree of rotation out
of the plane of conjugation of the substituted aromatic rings. The larger and
bulkier lateral substituents have the largest effects. A rotation out of the plane of
the molecule reduces the degree of conjugation, changes the band gap and,
consequently, shifts the wavelength of emission. Thus, the whole of the visible
spectrum can be covered by suitably substituted poly(thi0phenes).
The conjugated aromatic poly(p-phenylene) backbone linked by an aliphatic
methylene group in fluorene derivatives constrains the phenylene rings to lie in
one plane. This generates a large band gap and, consequently, blue photo-
luminescent and electroluminescent emission for the poly(fluorenes) (27-34)
collated in Table 6.6.78-89The first blue-emitting OLED using LEPs utilised
polymer (27) as the emission The solubility in organic solvents,
melting point and glass transition temperature can all be modulated by the
length and nature (branched or unbranched) of the alkyl substituents attached
to the fluorene at position 9. Since the substituents at position 9 do not
contribute to the degree of conjugation of the poly(p-phenylene) backbone the
band gap of the polymers (27-29) is remarkably uniform. The nature of the
monomer components of the copolymers (29-34)determines the melting point,
glass transition temperature and to some extent the solubility of the resultant
c ~ p o l y m e r . The
~ ~ -colour
~ ~ of emission can also be tuned by varying the relative
amounts of the monomers in he copolymers, e.g. polymer 34; m = n;
A,,, = 453 nm.89
The poly(p-phenylene) derivatives (35-42)collated in Table 6.7 all exhibit
high band gaps and, consequently, most of them emit in the blue part of the
visible s p e c t r ~ m . Poly(p-phenylene)
~ ~ ~ ~ ~ ~ ~ ’ itself (35)was first synthesised by a
complex precursor route, which generates an insoluble, intractable polymer
with a high melting point like PPV (l).9@92 However, subsequent syntheses of
poly(pheny1ene) derivatives has been accomplished much more simply by using
transition metal catalysed aryl-aryl cross-coupling r e a c t i ~ n s . ~The
~ - ’presence
~~
of an alkoxy substituent in polymer 3693-96or two alkoxy substituents in
polymer 3797renders these polymers soluble in organic solvents. However, the
Organic Ligh t-Emitting Diodes Using Light-Emitting Polymers 191
Table 6.6 Emission maximum (nm) and colour for the poly(j?uorenes) (27-34)
27 470 Blue 78-8 1
477 Blue 82
29 Blue 83,84
30 434 Blue 85
31 570 Yellow 86,87
32
33 410 Blue 88
(cont inued)
192 Chapter 6
Table 6.6 continued
Blue 89
presence of the lateral substituents results in rotation of the phenyl ring out of
the plane of neighbouring phenyl rings. This interruption of the conjugation
results in a further shift in the emission spectrum and a lower quantum yield
than that of poly(pheny1ene). The poly(pyridine) (38), with an additional
nitrogen atom in the aromatic ring," and the copolymers (39 and 40) also emit
in the b l ~ e . ~The
~ *presence
'~ of the lateral substituents results in a significant
shift due to rotation of the substituted phenylene rings out of the plane of the
non-substituted rings. This effect is even more pronounced for the disubstituted
poly(pheny1enes) (41 and 42), where a more significant shift is observed. lo'
The ladder polymers (43-45) collated in Table 6.8 are essentially soluble
pphenylene derivatives. However, the phenyl rings cannot rotate out of the
plane of the aromatic core of the polymer due to the presence of the methylene
linking unit on the fluorene This generally gives rise to a broad
emission spectrum. Aggregation of the polymer chain due to the planar nature
of the aromatic core unit, e.g. for polymer (43) can give rise to quenching and
'
time-dependent electroluminescent behaviour. l o Lateral substituents, such as
the methyl groups present in the polymers (44), reduce this problem and
consistent blue emission with a relatively narrow band width is observed. l 1
The presence of random phenyl rings in the stepladder copolymer (45) leads to a
degree of twisting of the aromatic core sufficient to suppress aggregation, but
this is not too prevalent to reduce the quantum efficiency of emission. l 2
It was found that the use of a PPV derivative, e.g. compound 46, see Table
6.9, containing fully conjugated segments separated by non-conjugated lin-
kages, which had not been converted to the conjugated carbonxarbon double
bond by elimination of the leaving group from the precursor polymer by
annealing actually exhibited a significantly higher quantum efficiency. 13-' l 6
This is a general observation found for a range of conjugated organic polymers
exemplified by the typical structures (46-52) collected in Table 6.9.113-131
OLEDs containing partially conjugated PPV were found in some instances to
exhibit an external efficiency up to two orders of magnitude higher than that
determined for a similar device using fully conjugated PPV.' l 6 The wavelength
of emission (508 nm) is also blue shifted relative to PPV (551 nm) due to the
higher band gap of the short conjugated PPV fragments. These polymers
illustrate the point that diluting the active chromophore by chemically binding
~~
Molecular structure Anax Colour Ref
459 Blue 90-92
Blue 93-96
37 400,460 Blue 97
Blue 98
39 470 Blue 99
485 Blue 100
41
Green 101
42 Black 101
194 Chapter 6
Table 6.8 Emission maximum (nm) and colour for the combined polymers (43-
45)
Molecular structure ,A CoZour Ref
43 Yellow 110
44 420 Blue 111
45 460 Blue 112

Table 6.9 Emission maximum (nm) and colourfor the main-chainpolymers (46-
52)
Molecular structure A,,, Colour Ref
508 Blue-green 113, 114
470 Blue 117-1 19
48 tD(u-4hgp \ / 498 Blue-green 120
4
460-550 Blue-green 121-124
452 Blue 125-127
51 475 Blue 128
52
it to a polymer backbone can actually lead to an increase in quantum yield. A

short chromophore may maximise efficient emission due to the absence of
defects acting as quenching sites often found in a long polymer chain. The
absence of exciplex and eximer formation may result in a higher quantum yield
per chromophore than that observed for a corresponding fully conjugated
polymer. This is counterbalanced to some extent by a lower concentration of
emitting parts of the polymer due to the presence of spacers or inert polymer
backbone. However, it is difficult to control the degree of conversion and,
196 Chapter 6
therefore, the number of non-conjugated linkages and the length of the
conjugated PPV segments, since the elimination of leaving groups occurs
randomly along the polymer backbone. Furthermore, elimination of some of
the linkage leaving groups, e.g. sulfonium groups, may occur spontaneously
over time in the OLED itself. This will lead to a significant decrease in efficiency
over the lifetime of the display.
The problems of reproducibility and stability of partially converted PPV
derivatives were solved by the synthesis of a series of main-chain polymers
incorporating sections of PPV of defined length separated by inert, non-
aromatic, non-conjugated spacer units. The short identical chromophores emit
blue light with a narrower bandwidth. This is illustrated by the structures
collated in Table 6.9 for some typical main-chain polymers (47-50).' 17-13' The
transition temperatures and efficiency of emission of the polymers (47-49) may
be controlled by varying the relative concentrations of the inert spacer and the
PPV segments as well as the length of the segments themselves, e.g. n and m for
polymer 49.121-124 It was also found subsequently that a small inert group, such
as a silane moiety in polymers 50 and 51,125-127served to interrupt the
conjugation at well-defined intervals without greatly diluting the active chro-
mophore. The four ethyl groups attached to the silicon atoms, rather than the
aromatic rings of polymer 51 containing p-quarterterphenyl segments separated
by a silane spacer, also act as lateral substituents. Thus, they serve to render this
polymer soluble in organic solvents in spite of the high melting point and
insolubility of molecular p-quarterphenyl. 12* The polymer 52 exhibits an
enantiotropic nematic phase. 129-131 This can be made use of for polarised light
emission of blue light, see Section 5.
An alternative approach is to dissolve an efficient electroluminescent chro-
mophore in an inert polymer host in a guest-host system or polymer
blend. 132-135 Unfortunately, phase separation and demixing often limits the
amount of low-molar-mass electroluminescent chromophores that can be
dissolved in polymer hosts. Therefore, the relative brightness is lower for such
guest-host systems. This problem can be overcome by fixing the chromophore
chemically to the polymer itself as a pendent group on a side-chain polymer
separated by spacer units. Bulk phase separation is then impossible, although
microphase separation may still take place. This is illustrated by the structures
collated in Table 6.10 for some typical side-chain polymers 53-55.136'41
3 Bilayer OLEDs Using LEPs

There are several possibilities for constructing bilayer OLEDs with a more
balanced charge injection. These include an electron-transport layer (ETL) and
a combined hole-transport (HTL) and emission layer. Conversely a hole-
transport layer and a combined electron-transport and emission layer is also
effective.
Organic Light-Emitting Diodes Using Ligh t-Emit t ing Polymers 197
Table 6.10 Emission maximum (nm) and colour for the side-chainpolymers (53-
55)
53 620-630 Red 137,138
54 475 Blue 139
55 141
/ \
Display ConJigurat ion

A thin film ( x 100 nm) of a polymer hole-transport layer (HTL) supports a
second thin film of a polymer electron-transport layer (ETL) sandwiched
between two electrodes supported on a substrate, see Figures 6.2 and 6.3. The
anode is transparent in order to allow the passage of the light generated. A
potential can be applied between the electrodes. The metal cathode has a low
198 Chapter 6
METALLIC
CATHODE 4 h
POLYMERETL
POLYMER
HTLAND
EMlTrER
TRANSPARENT
ANODE 4 P
TRANSPARENT
SUBSTRATE
Figure 6.2 Schematic representation of a bilayer OLED incorporating a combined

polymer hole-transport and emission layer and a polymer electron-transport
layer situated between a transparent anode and a cathode.
METUIC
CATHODE
POLYMER
ETL AND
E m 0000000000000
,0000000000000
POLYMER c
HTL
TRANSPARENT
ANODE + -I
TRANSPARENT
SUBSTRATE
Figure 6.3 Schematic representation of a bilayer OLED incorporating a polymer hole-

transport layer and a combined polymer electron-transport and emission layer
situated between a transparent anode and a cathode.
work function, which is matched as closely as possible to the LUMO of the ETL
in order to promote the efficient injection of electrons. The HOMO of the HTL
is also matched as near as possible to the high work function of the anode, e.g.
indium-tin oxide (ITO) or an organic polymer electrode such as poly(aniline)
salt (PANi). The application of a voltage above the threshold voltage leads to
injection of electrons from the cathode into the ETL and holes from the anode
into the HTL. The matching of the work functions of the electrode materials to
those of the ETL and the HTL minimises the barriers to the injection of holes
from the anode and electrons from the cathode. The electrons and holes
recombine preferentially at the interface between the ETL and HTL. This
results in efficient emission of light from the resultant excitons in one or both of
the layers depending on the difference in the band gap of the polymers, see
Figures 6.2 and 6.3. Thus, the activated pixels emit light and appear bright.
This configuration generally results in a higher quantum efficiency
(qeXt< 4%) than that observed for comparable monolayer OLEDs due to
balanced charge-injection, charge-transport and recombination of holes and
electrons away from the electrodes. Very high luminance (= 1000 cd m-2) can
be achieved at reasonable operating voltages ( x 6 V).
Polymers as Electron Transport Layers (ETLs)

The use of low-molecular-weight 2,5-disubsituted-1,3,4-0xadiazolesas an ETL
in OLEDs using small molecules had been extensively reported, although these
often suffer from problems caused by partial crystallisation, see Chapter 5.
Therefore, in the first reported bilayer OLEDl4I using two polymer layers, a
thin layer (30 nm) of 2-(4'-biphenylyl)-5-(4+butylphenyl)- 1,3,4-0xadiazole,
PBD (56)142*143, see Table 6.1 1, was also used as the electron-transport material.
However, it was dispersed in an amorphous polymer, i.e. poly(methy1 metha-
crylate), in order to inhibit crystallisation. This polymer blend was deposited by
spin coating from solution onto an insoluble layer of PPV (1).The layer of PPV
functioned simultaneously as a hole transport layer and an emission layer. The
observed external efficiency of the resultant bilayer device was higher than that
of a comparable monolayer d e v i ~ e . ' ~The * ' ~holes
~ are injected into the PPV
layer, travel to the interface between the PPV and PBD/PMMA layer and are
blocked by the PBD/PMMA layer at this interface. The field associated with
this positive space charge at the layer interface lowers the energy barrier for
electron-injection at the cathode. This results in a flow of electrons to the other
side of the interface between the PBD/PMMA and PPV layers. The electrons
cross over into the PPV layer, combine with holes and the resultant singlet
excitons emit visible yellow-green light from the lower band gap PPV layer.
However, the concentration of PDP in the host PMMA matrix had to be low
in order to avoid phase separation. This imposes an upper limit on the
improvement in external efficiency in using bilayer OLEDs instead of mono-
layer OLEDs. Therefore, there was a clear requirement for conjugated organic
polymers with a stable, high glass transition temperature and a high affinity for
electrons for use as an amorphous, non-crystalline ETL.
200 Chapter 6
Table 6.1 1 Molecular structure of the electron-deficient heterocyclic compounds
(54-61)
56 142,143
57 146-149
58
59 151,152
F n
60 153
61 154,155
n
Organic Light-Emitting Diodes Using Light-Emitting PoZymers 201
Most conjugated aromatic polymers promote the transport of holes rather
than electrons. This is a direct consequence of the high density of delocalised
electrons attributable to the aromatic rings and carbon-carbon double bonds of
a conjugated poly(ary1ene vinylene). It is much easier to remove an electron
than add an electron to an electron-rich organic material. Therefore, most PPV
derivatives can function as an HTL or as a combined HTL and emission layer,
e.g. PPV (l), which preferentially transports holes and blocks the movement of
electrons. Polymers for use as electron-transporting and hole-blocking layers in
multilayer OLEDs should exhibit a kigh electron affinity, i.e. be electron
deficient. Consequently, electronegative atoms and atomic units, such as the
strongly electronegative cyano group, were incorporated in modified PPV
derivatives, see below, in order to create new strongly polarised, electron-
deficient conjugated polymers for use as an ETL.19 The presence of the
electronegative atoms lowers the energy of the LUMO and HOMO levels. This
increases the electron affinity of the polymer and promotes electron injection
from the cathode.
Main-chain polymers, such as polymer 57 shown in Table 6.1 1, incorporating
a higher concentration the 2,5-disubsituted-lY3,4-oxadiazolemoiety directly
bound to the polymer backbone were prepared as E T L S . ' ~The ~ ~glass
~
transition temperature can be controlled by varying the relative concentration
of the monomer units in copolymers.146-149 Polymers incorporating other
electron-deficient heterocyclic rings containing at least one nitrogen atom were
found generally to exhibit superior properties.150-158 Some typical examples are
collated in Table 6. I 1 such as poly(quino1ine) (58),150poly( 1,3,5-trisubstituted
triazole) (59),1519152poly( 1,3,5-trisubstituted triazine) and poly(pheny1-
quinoxaline) (61).154-158 Many of these main-chain polymers also incorporate a
number of fluorine atoms attached to an aromatic ring or a methylene spacer
unit. The electronegative fluorine atoms contribute to the electron-transporting
properties of these polymers. The presence of an oxygen atom between two
aromatic rings in the polymers (58-60) gives them a non-linear conformation,
which renders them soluble in organic solvents for spincasting. The phenyl rings
in a lateral position on polymer 61 serve the same purpose.
Poly(N-vinylcarbazole) [PVK] (62), Table 6.12, is a commercial side-chain
polymer, which had been found to be a very useful hole-transporting material,
although its main application is in photovoltaic devices such as photocopiers.
PVK was also used to form an effective HTL as thin uniform film on a device
substrate by spincoating, see Table I2.ls9 Therefore, an analogous method was
used to synthesise sidechain polymers incorporating nitrogen heterocyclic rings
as part of pendent groups separated and decoupled from the polymer backbone
by inert spacer units for use as ETLs in OLEDs. This allowed the use of
standard repeat units, such as acrylates, methacrylates and styrenes, to be used
as the polymer backbone. This has many synthetic advantages. A typical
example of such a side-chain polymer (63) incorporates the standard (2,5-
dipheny1)-1,3,4-0xadiazolering as the electron-deficient moiety. The glass
transition temperature of side-chain polymers can be controlled by varying the
relative concentration of the monomer units in copolymers such as Side-
202 Chapter 6
Table 6.12 Molecular structure of some sidechain polymers (62-64)
62 159
w
63 160-162
64 163
chain polymers tend to form good quality films. However, the operating
voltages of OLEDs using side-chain polymers as transport layers have been
found to be high, perhaps due to the dilution of the conjugated aromatic core by
the aliphatic spacer and the polymer backbone.
The incorporation of the electron-withdrawing cyano group into the central
vinyl linkage between the phenyl rings of PPV derivatives could be relatively
easily achieved by the condensation of an aromatic dialdehyde with a diacetoni-
trile in a Knoevenagel reaction to effectively produce a series of regularly
alternating electron-deficient copolymers, such as those (65-69) collated in
Table 6.13, as an ETL. The electron-donating effects of the lateral alkyl and
alkoxy substituents attached to the phenyl rings of the CN-PPV derivatives 65-
67 and para- and rneta-substitution patterns result in emission at any part of the
visible The presence of the thiophene ring in the polymer (68)
results in a small band gap. Thus, it emits in the infrared.'67 The cyano group
attached to the carbon-carbon double bond lowers the LUMO level by about
0.6 eV compared to that of corresponding PPV derivatives. There appears to be
Table 6.13 Emission maximum (nm) and colour for the copolymer PPV deriva-
tives (6569)
Molecular structure A,, Colour Ref
65 590 Red 164
66 510 Green 165,166
67 Blue 165, 166
68 740 IR 167
69 168
no steric effect on the planarity of the polymer attributable to the presence of the
bulky cyano group. The flexible nature of the alkyl and alkoxy lateral
substituents also confers a good degree of solubility in organic solvents on the
polymers so that they can be deposited as thin uniform layers by spincoating.
The polymer 69 incorporates an electronegative nitrogen atom in one of the
phenyl rings, instead of on the carbonsarbon double bond, in order to give rise
to a large value for the electron-affinity of the p ~ l y r n e r . ' ~ ~ - ' ~ '
204 Chapter 6
Table 6.14 Emission maximum (nm) and colour for the electron-deficient PPV
derivatives (7&74).
Molecular structure Anax Colour Ref
70 620-630 Red 63-69
71 620-630 Red 63-69
72
+Jy
F F
540-570 Yellow-orange 172,173
73
+&+" 605 red 168
74 630 red 169-1 7 1
The PPV derivatives (7&74) collated in Table 6.14 incorporate an electron-

withdrawing group on the aromatic ring, either as a lateral substituent on
the ring or as an heteroatom (N) within the ring, instead of on the vinyl
linkage.6M9.'68-'73 All of these polymers exhibit a higher electron affinity than
unsubstituted PPV (1). The band gaps of all of these PPV derivatives are low
and emission is in the yellow-orange or red region. However, the high electro-
negativity of the trifluoromethyl group in the polymer (72)'729'73blue-shifts the
emission compared to that of the polymers 70 and 7163"9 with either a bromo or
chloro substituent. The solubility of the pyridine polymer (73)'719'72
is less than
that of the corresponding alkyl-substituted pyridine polymer (74). 168 The
external quantum efficiency of emission of these substituted polymers is not
very high and, therefore, they are used purely as an ETL rather than as an
emission layer with a high electron affinity.
The polymers (75-77) listed in Table 6.15 combine hole-transporting,
Organic Light- Em it t ing Diodes Using Light-Emit t ing Polymers 205
Table 6.15 Emission maximum (nm)and colourfor the combinedpoiymers (75-

77)
Molecular structure It,, COIOUl Ref
75 580 Yellow-orange 175
76 620-630 Blue 149,163
176
electron-transporting and emission functional units in the same

~ ' ~PPV
p ~ l y r n e r . ' ~ ~The ~ ~backbone ~ polymer 75 acts as an HTL and
' ~ ~ ' ~ of
emission layer at the same time.'75 The oxadiazole side-chain acts as an
electron-transport functional group. The polymer 76 combined hole-transport-
ing and electron-transporting functional entities on the same poly(methacry-
late) The relative amounts (m and n) of the monomer units
206 Chapter 6
determine the final properties of the copolymer. The short PPV fragment results
in blue emission. The oxadiazole ring in the polymer 77 assists electron
transport through the hole-transporting PPV fragment in the polymer back-
bone.'76 It also serves to blue-shift the emission. Unfortunately the stability of
such combined polymers is sometimes insufficient for practical devices.149~163
Bilayer OLEDs using insoluble PPV derivatives as the hole-transporting layer
(HTL) are particularly attractive from a fabrication point of view. They can be
readily prepared by depositing a soluble precursor and then converting it into an
intractable and insoluble film. The second ETL can then be deposited from
solution by spin casting. The intractable nature and insolubility of the PPV
derivative eliminates problems of dissolution of underlying layers during the
deposition and drying of subsequent layers sometimes encountered with soluble
polymers. In this case polymers must be chosen which are soluble in one solvent
for deposition and insoluble in the solvent solution of the next layer to be
deposited. Swelling of polymer layers by solvent absorption can be a serious
problem using the technique of polymer deposition by spin-coating from
solution.
A bilayer OLED consisting of a layer of the poly(thiophene) (25)76 as ETL
next to the cathode and the low-molar-mass PBD (56) as HTL'42-144next to the
anode emits white light.76Simultaneous emission of red and green light from the
poly(thiophene) (25) and blue light from PBD (56) leads to the mixing of the
three primary colours and to the production of white light.
4 Trilayer OLEDs Using LEPs

There are several reports of trilayer OLEDs using three distinct layers for
electron transport, hole transport and emission. This device configuration has
the advantage that each layer can be optimised for one distinct function, i.e.
hole-transport, electron-transport and light emission. The problems associated
with the fabrication of OLEDs with three distinct polymer layers of controlled
thickness, integrity and homogeneity by the technique of deposition from
solution by spin-coating are not inconsiderable.
A thin layer ( M 100 nm) of a hole-transporting polymer is next to the anode such
as ITO. An electron-transporting layer lies next to the cathode. An emission
polymer layer is situated between the HTL and the ETL, see Figure 6.4. At least
one of the two electrodes is transparent in order to allow the passage of the light
generated. An electric field can be applied between the electrodes. The low work
function of the metal cathode matches as closely as possible that of the LUMO
of the ETL in order to enable efficient electron injection to take place. The high
work function of the anode is also matched as closely as possible to the HOMO
of the HTL. The electroluminescent polymer in the centre of the display exhibits
a high quantum efficiency of emission at the desired wavelength. The applica-
tion of a voltage above the threshold voltage leads to injection of electrons from
207
TRANSPARENT
SUBSTRATE
Figure 6.4 Schematic representation of a trilayer OLED incorporating a polymeric hole-

transport layer (HTL), an electron-transport layer (ETL) and a central
emission layer situated between a transparent anode and a cathode.
the cathode into the ETL and holes from the anode into the HTL. The electric
current passing through the HTL and ETL polymer layers give rise to a
concentration of charge at the interface between them. The electrons and holes
then recombine in the emission layer, create excitons and light is emitted.
An early example of a trilayer OLED used PVK (62)15' as the HTL, a
polymer blend of PBD (56)142i'43in poly(methacry1ate) as the ETL and a
poly(quinoxa1ine) (61)154,'55 as the emission layer in the middle of the other two
1aye1-s.'~'A related device used the same materials as the HTL and ETL, but
used segmented PPV derivatives such as those shown in Table 6.9 as the
emission 1 a ~ e r .Thel ~ ~efficiency of these devices may be comparable to that of
single-layer OLEDs. However, the operating voltages and current are usually
significantly lower, which leads to longer effective device lifetimes.
Conjugated organic polymers such as those shown in the Tables have been
used in multilayer OLEDs as the HTL or combined HTL and emission layers or
as the ETL or combined ETL and emission layer. The combined polymers (75-
77)shown in Table 6.15 have been used as combined ETL, HTL and emission
layers in various OLED configurations. Blends of these polymers have also been
used to maximise OLED efficiency, although phase separation is always a
problem with mixtures (blends) of main-chain polymers. 1329133*179-184
Trilayer OLEDs using LEPs have been fabricated with external quantum
efficiencies ( q e x~ 410%) comparable to those obtained for analogous OLEDs
using small molecules or LEDs." The low threshold and operating voltages
( w 5 V) of these multilayer polymer OLEDs means that they can be used as
displays in battery operated devices.
208 Chapter 6
5 Polarised Light Emission from OLEDs

If the chains of a main-chain, electroluminescent polymer are aligned parallel to
each other then the absorption and emission of polarised light can be observed.
The dipole transitions of the chromophore are aligned parallel to the polymer
chains. If these are aligned macroscopically in the azimuthal plane then the
absorption and emission of light will be anisotropic. The maximum of absorp-
tion or emission will be parallel to the polymer chain and the corresponding
minima orthogonal to the polymer chains. This has practical consequences, for
example, sheets of uniformly aligned electroluminescent polymers could be used
as a source of bright polarised light for use as efficient backlights for LCDs. The
requirement for two polarisers in standard LCDs, such as TN-LCDs with
passive or active matrix addressing and highly multiplexed STN-LCDs would
be reduced. This would theoretically double the amount of light transmitted by
the LCD and this would ameliorate one of the prime disadvantages of LCDs,
i.e. low brightness and poor contrast for battery-driven displays. Polarised
electroluminescencecan be achieved using several techniques under appropriate
conditions. However, the potential for polarised electroluminescence was often
investigated first by determining the degree of macroscopic alignment of the
chromophores by polarised absorption or photoluminescence measurements.
Anisotropic absorption and photoluminescence spectra for polymer blends
had been obtained soon after the initial discovery of polymer electrolumines-
cence. lS5 Gel-processed polymer blends of MEH-PPV (2) in poly(ethy1ene) were
mechanically stretched to uniformly align the long polymer chains parallel to
each other in the azimuthal plane. The stretching process unravels and unwinds
the coiled polymer chains found in solution and aligns the straightened chains
parallel to the direction of stress. This resulted in very high polarisation ratios
(60:1) of photoluminescence (PII/PI) for these blends. 185 More recently photo-
luminescence, but not electroluminescence, with high polarisation ratios was
achieved by essentially the same technique of mechanically stretching polymer
blends of poly(arylethynes), such as 15 and 16,5741 in polyethylene.18'
However, the polarisation ratio for electroluminescent devices realised using
stretched poly(thiophenes) on a polyethylene substrate was low (2.4:1).187
The macroscopic alignment of polymer chains was also achieved by deposit-
ing soluble poly(p-phenylenes), such as 3797or poly(3-thiophenes) such as 26,77
from solution by the Langmuir-Blodgett technique. However, low polarisation
ratios (EII/EI)were found for electroluminescence (3: 1).
Low polarisation ratios ( < 2:l) for absorption had also been found for
amorphous PPV (1) deposited from solution by spin-coating on rubbed
poly(tetrafluoroethy1ene)[FTFE].18' It is evident that this could be improved
on by making use of the high order parameter and self-organising properties of
the nematic phase of liquid crystalline electroluminescent polymers' 89 such as
those (16, 28 and 78-82) shown in Table 6.16.16~58-60~81~82~130~1g~192 This was
then found subsequently to be the using thermotropic liquid crystal-
line polyfluorenes, such as 28 and 80 shown in Table 6.6 and segmented PPV
derivatives, such as 81.129-'31 The nematic phase exhibits the lowest viscosity of
Table 6.16 Transition temperatures ("C)for the liquid crystalline polymers (Id,
28, 78-82)
Molecular structure Tg Cr LC I Ref
66 0 103 0 133 0 41
100 0 -165 0 -260
<93 0 170 0 280 0 81
28 - 0 167 0 >190 0 82
16 100 - 0 > 135 0 58-60
82 62 - 0 149 0 192
the liquid crystalline phases and, therefore, it is the easiest to macroscopically

align by elastic forces on rubbed substrates. Many of the other polymers
collated in Tables 6.1-6.15 exhibit thepnotropic liquid crystalline phases,
although they are often not well characterised. Thermal cross-linking hinders
the determination of liquid crystalline transition temperatures of polymers with
unsaturated linkages such as poly(p-phenylenevinylenes)and poly(ary1ethynes).
210 Chapter 6
Many of the poly(fluorenes) collated in Table 6.16, the poly(p-phenylenes)
collated in Table 6.7 and the segmented PPV derivatives shown in Table 6.9
exhibit an enantiotropic liquid crystalline phase above the melting point. After
being aligned in the nematic phase or smectic A phase the macroscopic structure
of the homogeneously aligned liquid crystalline phase can be fixed in the glassy
state by quenching the polymer. The liquid crystalline phase must be homo-
geneously aligned on a conducting orientation layer, such as rubbed polyimide
with the director parallel to the direction of rubbing. Since polyimide is an
insulator, it must be doped with a conducting although some
polyimides can also act as HTLs. The best polarised electroluminescence was
achieved with an electroluminescence ratio of 15:1 and a switch-on voltage of
18 V from a liquid crystalline poly(fluorene) aligned on rubbed doped poly-
imide.18’ An electroluminescence polarisation of 12:l with 200 cd m-’ was
achieved from PPV (l), which was rubbed when partially formed from its
p r e c u r ~ o r ? Alignment
~ of the polymer chains can also be achieved by rubbing
the polymer directly on the substrate surface.
The order must then be frozen in before crystallisation occurs, since this
would result in the formation of grain boundaries and a reduction in transport
or emission efficiency. Device breakdown is also a possibility. The most efficient
way to fix the liquid crystalline order is the formation of anisotropic networks
by the polymerisation of reactive mesogens in the liquid crystalline
state. 189,193-195 Anisotropic polymer networks formed from the thermal or
photoinitiated polymerisation of polymerisable, so-called photoreactive, liquid
crystalline monomers have been used in a wide variety of electrooptic applica-
tions, see Chapter Z.’9”’98 This is a more attractive approach than cross-linking
aligned sidechain polymers due to the greater ease of orienting low-molar-mass
organic materials as a consequence of their much (orders of magnitude) lower
viscosity compared to that of polymers. lg9 The synthesis of photoreactive
oxetanes, deposition on a substrate and subsequent cross-linking to form
intractable, insoluble HTL in multilayer OLEDs, has been reported recently,
although the monomers were not liquid crystalline.200
The first report’93of this concept did not use a pure photoreactive monomer.
However, the photoreactive arylethyne (83),lg3 which is not liquid crystalline
itself, see Table 6.17, was mixed at low concentration (5 wt%) with the
photoreactive liquid crystal 84196to form a polymerisable nematic mixture
with a high melting point (1 16°C) and clearing point (150°C).The director of
this nematic guest-host mixture, see Chapter 3, was aligned in a cell with rubbed
nylon 66 as the alignment layer. This alignment was imposed on the chromo-
phore (83). Photoinitiated polymerisation of the mixture at a high temperature
above the melting point of the mixture fixed the mixture in place. A high
polarisation ratio ( w 11:1) of blue photoluminescence was obtained. Electro-
luminescence was not reported. The photoreactive monomer (S5)lg4exhibits an
enantiotropic smectic and nematic phase at high temperature. 194 It was aligned
in the nematic state on rubbed polyimide and then polymerised and cross-linked
in a thermally initiated polymerisation reaction by raising the temperature of
the mixture to 180°C. A moderately high dichroic ratio (3.5) of polarised
Organic Light-Emitting Diodes Using Light-Emitting Polymers !I 1
m \D
s s
0 0
v,
v
&,
I
0
*n
I 00 I
0
W 0
00
00
0
00 2 d-
0 0 o
k
I
In \o
00 00
212 Chapter 6
absorption was observed. The diene derivative (86)19’ exhibits a low melting
point and a moderately high nematic clearing point. No other mesophases could
be observed. This allows the nematic director to be aligned on a non-contact
alignment layer at moderate temperatures ( < 8OoC)and then photochemically
cross-linked with ultra violet (UV) light using Irgacure (Ciba) as the initiator.
The photoalignment layer was doped with 4,4,4-tris(naphthyl)-N-(phenylami-
no)triphenylamine, a commercial hole-transporting material, to increase its
conductivity. The photoalignment layer was exposed to 0.2 J cm-2 of polarised
UV light prior to deposition of the photoreactive liquid crystalline diene
chromophore (2), which was subsequently cross-linked when in a nematic
glassy state by exposure to unpolarised UV light at room temperature using
very small amounts of photoinitiator to form a uniform, defect-free cross-linked
layer. Using a diene as the photoreactive group has the advantage that it
polymerises much more slowly, specifically and selectively than analogous
acrylates or methacrylates via a sterically controlled intramolecular cyclisation
reaction to form a cyclic polymer. The diene derivative (86) with a reactive
group at each end of the molecule forms a cross-linked network, see Figures 6.5
and 6.6. Polarised electroluminescent emission of about 100 cd m-2 was
achieved with a dichroic ratio (ELII/ELI) of lO:l, see Figure 6.7, and with a
threshold voltage of 8 V. This threshold can clearly be improved as a very low
threshold voltage of 2 V was achieved when the photoalignment layer was not
included in the device, see Figure 6.8. Furthermore, cross-linkingdoes not affect
either the photoluminescence or electroluminescenceintensity, or the threshold
voltage, and actually improves the polarisation ratio of the OLED.
6 Performance of OLEDs Using LEPs

The technical data for a prototype OLED using LEPs are collated in Table
6.18.201*202The brightness and the contrast ratio ( > 1OO:l) of monochrome
OLEDs with various colours is comparable with that of a standard commer-
cially available TN-LCD with active matrix addressing. The lifetime is clearly
Table 6.18 Configuration and electro-

optical performance of a pro-
totype OLED f o r backlighting
using a soluble PPV derivative
as the emission layer201*202
Property Value
Anode IT0
Operating voltage, V,, 3.5 v
Power consumption 5 mA cm-2
Colour Green-yellow
Brightness 100 cd m-’
Lifetime (8 cm2 display) 30000 h
214 Chapter 6
POLYMER
BACKBONE - CROSSL.INK
POLYMER
BACKBONE
MOLECULARCORE
--
n
SPACER
AND CHROMOPHORE
SPACER
Figure 6.6 Schematic representation of a cross-linked anisotropic network.
1.o
0.8
0.6
0.4
0.2
0.0
400 500 600 700

WAVELENGTH nm
Figure 6.7 Plot of the relative intensity ( A .U.)of polarised electroluminescence against
wavelength (nm) of a bilayer OLED consisting of an electron-transport and
emission layer represented by a nematic network formed by polymerising
compound (86) with isotropic UV light and a combined hole-transport and a
coumarin non-contact alignment layer doped with 4,4,4-tris(naphthylyl)-N-
(phenylamino)triphenylamine.
sufficient for commercial applications. The low operating voltage and power
consumption are also compatible with battery-operation for portable instru-
ments. Monochrome OLEDs using electroluminescent polymers are being
developed by Philips as backlights for LCDs in portable telephones. Chemical
suppliers of electroluminescent polymers, which emit across the visible spec-
trum, include Aventis, Cambridge Display Technology (CDT) and Dow
Chemicals. It is to be expected that OLEDs using light-emitting polymers will
enter the market for flat-panel displays and establish themselves as important
volume products within the next five years.
Organic 215
0.01 0.10 1.oo 10.00

APPLIED VOLTAGE (V)
Figure 6.8 Plot of the current ( A ) against applied voltage ( V ) of a bilayer<OLED

consisting of an electron-transport and emission layer represented by a nematic
network formed by polymerising compound (86) with isotropic UV light and a
combined hole-transport and a coumarin non-contact alignment layer doped
with 4,4,4- tr is (naphthy ly1) -N- (phenylamino) t r @henylamine.
Stability of OLEDs Using LEPs

The lifetime of simple, alpha-numeric monolayer OLEDs using light-emitting
polymers is satisfactory (5000 h < lifetime < 10000 h). The conjugated poly-
mers used in OLEDs form stable amorphous glasses with sufficiently high glass
transition temperatures. Therefore, conjugated polymers used in commercial
OLEDs do not crystallise over the lifetime of the display. The degradation over
time of the electrodes in the presence of oxygen or moisture is due to oxidation
and gives rise to visible dark spots in the di~play.”~However, the major
stability problem for OLEDs using light-emitting polymers is the oxidation of
carbon-carbon bonds between the aromatic rings of poly(p-phenylene viny-
lenes). l 3 These oxidation reactions reduce the conjugation length of the polymer
leading initially to a blue shift in the emission spectrum. Further oxidation can
completely deplete the number of electroactive species on a particular polymer
chain. In addition, the carbonyl functions generated by the oxidisation act as
highly efficient quenching sites. These reduce the external quantum efficiency
even further. An AC drive form instead of a DC driving voltage may also extend
the device’s lifetime. Therefore, the device packaging of OLEDs using electro-
luminescent polymers is a critical factor in determining the lifetime of OLEDs
containing them.’-’
216 Chapter 6
7 References
1 J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Marks, R. H. Friend, P. L.
Burn and A. B. Holmes, Nature, 1990,347,539.
2 D. Braun and A. J. Heeger, Appl. Phys. Lett., 1991,58,1982.
3 D. Braun, A. J. Heeger and H. Kroemer, J. Electron Mat., 1991,20,945.
4 F . Wudl, P. M. Allemand, G. Srdanov, Z. Ni and D. McBranch, ACS Symp. Ser.,
1991,455.
5 S. Doi, M. Kuwabara, T. Noguchi and T. Ohnishi, Synth. Met., 1993,5557,4174.
6 R. H. Friend, D. D. C. Bradley and A. B. Holmes, Phys. World, 1992,5/(11), 42.
7 D. D. C. Bradley, Synth. Met., 1993,54,401.
8 R. H. Friend and N. C. Greenham, Solid State Phys., 1995,49,1.
9 D. D. C. Bradley, Curr. Opin. Solid State Mater. Sci., 1996, 1,789.
10 R. W. Gymer, Endeavor, 1996,20,115.
11 J. Salbeck, Ber. Bunsenges. Phys. Chem., 1996,100,1667.
12 L. J. Rothberg and A. J. Lovinger, J. Mater. Res., 1996,11,3 174.
13 J. R. Sheats, H. Antoniadis, M. Hueschen, W. Leonard, J. Miller, R. Moon, D.
Roitman and A. Stocking, Science, 1996,273,884.
14 J. Cornil, D. Beljonne, D. A. dos Santos and J. L. BrCdas, Synth. Met., 1996,76,101.
15 K. Kaneto, K. Yoshino, K. Koa and Y . Inuishi, Jpn. J. Appl. Phys., 1974,18, 1023.
16 R. H. Partridge, Polymer, 1983,24,755.
17 A. B. Holmes, D. D. C. Bradley, A. R. Brown, P. L. Burn, J. H. Burroughes, R. H.
Friend, N. C. Greenham, R. W. Gymer, D. A. Halliday, R. W. Jackson, A. Kraft,
J. H. F. Martens, K. Pichler and I. D. W. Samuel, Synth. Met., 1993,5557,4031.
18 W. J. Feast, J. Tsibouklis, K. L. Pouwer, L. Groenendaal and E. W. Meijer,
Polymer, 1996,37,5017.
19 A. Kraft, A. C. Grimsdale and A. B. Holmes, Angew. Chem. Int. Ed. Eng., 1998,37,
402.
20 U. Scholkopf, Angew. Chem., 1959,71,260.
21 G. Drehfal, H. H. Horhold and H. Wildner, Nachr. Chem. Techn., 1965,13,451.
22 R. M. McDonald and T. W. Campbell, J. Am. Chem. SOC.,1960,82,4669.
23 H. H. Horhold, J. Gottschaldt and J. Opfermann, J. Prakt. Chem., 1977,319,611.
24 H. Martelock, A. Greiner and W. Heitz, Makromol. Chem., 1991,192,967.
25 R. A. Wessling, J. Polym. Sci., Polym. Symp., 1985,72, 55.
26 R. W. Len, C. C. Han and M. Lux,Polymer, 1989,30,1041.
27 T. Momii, S. Tokito, T. Tsutsui and S. Saito, Chem. Lett., 1988, 1201.
28 F. Papadimitrakopoulos, M. Yan, L. J. Rothberg, H. E. Katz, E. A. Chandross and
M. E. Galvin, Mol. Cryst. Liq. Cryst., 1994,256,663.
29 F. Papadimitrakopoulos, K. Konstadinidis, T. M. Miller, R. Opila, E. A.
Chandross and M. E. Galvin, Chem. Mater., 1994,6, 1563.
30 V. H. Tran, V. Massardier, T. P. Nguyen and J. Davenas, Polymer, 1996,37,2061.
31 A. Beerden, D. Vanderzande and J. Gelan, Synth. Met., 1992,52,387.
32 R.0. Garay, U. Baier, C. Bubeck and K. Mullen, Adv. Mater., 1993,5,561.
33 M. Herold, J. Gmeiner and M. Schwoerer, Acra Polym., 1994,45,392.
34 M. Herold, J. Gmeiner, W. Riess and M. Schwoerer, Synth. Met., 1996,76, 109.
35 G. Gustafsson, Y. Cao, G. M. Treacy, F. Klavetter, N. Colaneri and A. J. Heeger,
Nature, 1992,357,477.
36 S. Hoger, J. J. McNamara, S. Schricker, F. Wudl, Chem. Mater., 1994,6, 171.
37 D.-H. Huang, H.-K. Shim, .I.-I. Lee and K.-S. Lee, J. Chem. SOC.Chem. Commun.,
1994,246 1.
38 C. Zhang, S. Hoger, K. Pakbaz, F. Wudl and A. J. Heeger, J. Electron Mater., 1994,
23,453.
39 D.-H. Huang, I.-N. Kang, M.-S. Jang, H.-K. Shim and T. Zyung, Polym. Bull.,
1996,36,383.
Organic Light-Emit ring Diodes Using Light-Emitting Polymers 217
40 S. T. Kim, D.-H. Huang, X.-C. Li, J. Griiner, R. H. Friend, A. B. Holmes and H.-K.
Shim, Adv. Mater., 1996,8,979.
41 M. Hamaguchi and K. Yoshino, Jpn. J. Appl. Phys., 1994,33, L1478.
42 M. Hamaguchi and K. Yoshino, Jpn. J. Appl. Phys., 1995,34, L712.
43 C. Zhang, S. Hoger, K. Pakbaz, F. Wudl and A. J. Heeger, J. Electron Mater., 1993,
22,413.
44 S. Hoger and F. Wudl, Polym. Preprints, 1994,34,327.
45 I. Murase, T. Ohnishi, T. Noguchi and M. Hirooka, Synth. Met., 1987,17,639.
46 S . A. Askari, S. D. Rughooputh and F. Wudl, Synth. Met., 1989,29, 129.
47 D. A. Haliday, D. D. C. Bradley, P. L. Burn, R.H. Friend and A. B. Holmes, Synth.
Met., 1991,41,931.
48 P. L. Bum, D. D. C. Bradley, R. H. Friend, D. A. Halliday, A. B. Holmes, R. W.
Jackson and A. Kraft, J. Chem. SOC.Perkin Trans. I , 1992,3225.
49 S. Tasch, W. Graupner, G. Leising, L. Pu, W.. Wagner and R. H. Grubbs, Adv.
Mater., 1995,7,903.
50 Z. Yang, B. Hu and F. E. Karasz, Macromolecules., 1995,28,6151.
51 H. Spreitzer, H. Becker, E. Kluge, W. Kreuder, H. Schenk, R. Demandt and
H. Schoo, Adv. Mater., 1998, 10, 1340.
52 F. H. Boardman, A. W. Grice, M. R. Ruther, T. J. Sheldon, D. D. C. Bradley and
P. L. Bum, Macromolecules, 1999,32, 1 11.
53 S.-J. Chung, J.4. Jin and K.-K. Kim, Adv. Mater., 1997,9, 551.
54 S.-J. Chung, J.4. Jin, C.-H. Lee and C.-E. Lee, Adv. Mater., 1998, 10,684.
55 B. R. Hsieh, Y. Yu, E. W. Forsythe, G. M. Schaaf and W. A. Feld, J. Am. Chem.
Suc., 1998,120,231.
56 H.-P. Weitzel and K. Miillen, Makromol. Chem., 1990,191,2837.
57 A. P. Davey, S. Elliot, 0. OConnor and W. Blau, J. Chem. SOC.,Chem. Commun.,
1995,1433.
58 C. Weder, M. S. Wrighton, R. Spreiter, C. Bosshard and P. Gunter, J. Phys. Chem.,
1996,100,1893 1.
59 C. Weder, and M. S. Wrighton, Macromolecules, 1996,29,5157.
60 D. Steiger, P. Smith and C. Weder, Macromol. Rapid Commun., 1997,18,643.
61 K. Tada, M. Onada, M. Hirohata, T. Kawaii and K. Yoshino, Jpn. J. Appl. Phys.,
1996,35, L25 1.
62 R. Giesa, J. Macromol. Sci. Rev. Macromol. Chem. Phys., 1996,36,631.
63 Y. Ohmori, M. Uchida, K. Muro and K. Yoshino, Jpn. J. Appl. Phys., 1991, 30,
L1938.
64 Y. Ohmori, M. Uchida, K. Muro and K. Yoshino, Solid State Commun., 1991,80,
605.
65 D. Braun, G. Gustaffson, D. McBranch and A. J. Heeger, J. Appl. Phys., 1992,72,
564.
66 M. Uchida, Y. Ohmori, C. Morischima and K. Yoshino, Synth. Met., 1993, 57,
4168.
67 N. C. Greenham, A. R. Brown, D. D. C. Bradley and R. H. Friend, Synth. Met.,
1993,57,4134.
68 A. Bolegnesi, C. Botta, Z. Geng, C. Flores and L. Denti, Synth. Met., 1995,71,2191.
69 M. Pomeranz, H. Yang and Y. Cheng, Macromolecules, 1995,28,5706.
70 R. E. Gill, G. G. Malliaras, J. Wildeman and G. Hadziioannou, Adv. Mater., 1994,
6, 132.
71 M. Berggren, G. Gustafsson, 0. Inganas, M. R. Andersson, 0. Wennerstrom and
T. Hjertberg, Adv. Mater., 1994,6,488.
72 M. Berggren, 0. Inganas, G. Gustafsson, J. Ramusson, M. R. Andersson,
T. Hjertberg and 0. Wennerstrom, Nature, 1994,372,444.
73 0. Inganas, M. R. Andersson, G. Gustafsson, T. Hjertberg 0. Wennerstrom,
P. Dyreklev and M. Grandstrom, Synth. Met., 1995,71,2121.
218 Chapter 6
74 M. R. Anderson, M. Berggren, 0. Inganas, G. Gustafsson, J. C. Gustafsson-
Carlberg, D. Selse, T. Hjertberg and 0. Wennerstrom, Macromolecules, 1995, 28,
7525.
75 N. Tanigaki, H. Masuda and K. Kaeriyama, Polymer, 1997,38, 1221.
76 M. Berggren, G. Gustaffson, 0. Ingenas, M. R. Anderson, T. Hjertberg and
0.Wennerstrom, J. Appl. Phys., 1994,76,7530.
77 A. Bolegnesi, G. Bajo, J. Paloheimo, T. Ostergard and H. Stubb, Adv. Mater., 1997,
9. 121.
78 Y. Ohmori, M. Uchida, K. Muro and K. Yoshino, Jpn. J. Appl. Phys., 1991, 30,
L1941.
79 M. Fukuda, K. Sadawa and K. Yoshino, Jpn. J. Appl. Phys., 1989,28, L1433.
80 M. Fukuda, K. Sadawa and K. Yoshino, J. Polym. Sci. Polym. Chem., 1993, 31,
2465.
81 M. Grell, D. D. C. Bradley, M. Inbasekaran and E. P. Woo, Adv. Mater., 1997,9,
798.
82 M. Grell, W. Knoll, D. Lupo, A. Meisel, T. Miteva, D. Neher, H.-G. Nothofer,
U. Scherf and A. Yashuda, Adv. Mater., 1999,11,671.
83 H. N. Cho, D. Y. Kim, Y. C. Kim, J. Y. Lee and C. Y. Lee, Adv. Mater., 1997,9,
326.
84 H. N. Cho, D. Y. Kim, Y. C. Kim, J. Y. LeeandC. Y. Lee, ProcSPIE, Int. SOC. Opt.
Eng., 1997,3418, 151.
85 M. Inbasekaran, W. Wu and E. P. Woo, USA Pat., 1998,5 777 070.
86 M. Grell, M. Redeker, K. S. Whitehead, D. D. C. Bradley, M. Inbasekaran, W. Wu
and E. P. Woo, Liq. Cryst., 1999,26, 1403.
87 M. Ranger and M. Leclerc, Mat. Sci. Spec. Iss. Can. I. Chem., in press.
88 M. Ranger, D. Rondeau and M. Leclerc, Macromolecules, 1997,30,7686.
89 G. Klarner, M. H. Davey, W.-D. Chen, J. Campbell Scott and R. D. Miller, Adv.
Mater., 1998, 10, 993.
90 G. Grem, G. Leditzky, B. Ullrich and G. Leising, Adv. Mater., 1992,4, 36.
91 G. Grem, G. Leditzky, B. Ullrich and G. Leising, Synth. Met., 1992,51, 383.
92 D. L. Gin and V. P. Conticello, Trends Polym. Sci., 1996,4,217.
93 M. Hamaguchi, H. Sawada, J. Kyokane and K. Yoshino, Chem. Lett., 1996,527.
94 A.-D. Schluter and G. Wegner, Acta Polym., 1993,44,59.
95 Y. Yamamoto, Progr. Polym. Sci., 1992,17, 1153.
96 Y. Yang, Q. Pei and A. J. Heeger, J. Appl. Phys., 1996,79,934.
97 V. Cimrova, M. Remmers, D. Neher and G. Wegner, Adv. Mater., 1996,8,146.
98 T. Yaamoto, T. Ito and K. Kubota, Chem. Lett., 1988, 153.
99 M. Remmers, M. Schulze and G. Wegner, Macromol. Rapid Commun.,1996, 17,
239.
100 M. Hamaguchi, H. Sawada, J. Kyokane and K. Yoshino, Chem. Lett., 1996,527.
101 J. A. John-and J. M. Tour, J. Am. Chem. SOC., 1994,116,5011.
102 T. Yamamoto, Progr. Polym. Sci., 1992,17, 1153.
103 T. Vahlenkampf and G. Wegner, Macromol. Chem. Phys., 1994,195,1933.
104 T. F. McCarthy, H. Witteler, T. Pakula and G. Wegner, Macromolecules, 1995,28,
8350.
105 M . Remmers, M. Schulze and G. Wegner, Macromol. Rapid. Commun.,1996, 17,
239.
106 V. Percec, S. Okita and R. Weiss, Macromolecules, 1992,25, 1816.
107 V. Percec, J.-Y. Bae, M. Zhao and D. H. Hill, Macromolecules, 1995,28,6726.
108 V. Percec, M. Zhao, J.-Y. Bae and D. H. Hill, Macromolecules, 1996,29, 3727.
109 K. Miyashita and M. Kaneko, Synth. Met., 1995,68, 161.
110 U. Scherf and K. Mullen, Macromol. Chem. Rapid Commun.. 1991,12,489.
111 S . Tasch, A. Niko, G. Leising and U. Scherf, Mat. Res. SOC.Symp. Proc., 1996,413,
71.
112 S . Tasch. A. Niko. G. Leising and U. Scherf. A m l . Phvs. Lett.. 1996.68. 1090.
Organic Light-Emit tins Diodes Using Light-Emitting Polymers 219
113 P. L. Burn, A. B. Holmes, A. Kraft, D. D. C. Bradley, A. R. Brown and R. H.
Friend, J. Chem. SOC.,Chem. Commun., 1992,32.
114 P. L. Burn, A. B. Holmes, A. Kraft, D. D. C. Bradley, A. R. Brown, R. H. Friend
and R. W. Gymer, Nature, 1992,356,47.
115 D. Braun, E. G. J. Staring, R. C. J. E. Demandt, G. L. J. Rikken, Y. A. R. R.
Kessener and A. H. J. Venhuizen, Synth. Met., 1994,66,75.
116 C. Zhang, D. Braun and A. J. Heeger, J. Appl. Phys., 1993,73,5177.
117 I. Sokolik, Z. Yang, F. E. Karasz and D. C. Morton, J. Appl. Phys., 1993,74,3584.
118 Z. Yang, I. Sokolik and F. E. Karasz, Macromolecules, 1993,26, 1188.
119 Z. Yang, F. E. Karasz and H. J. Geise, Macromolecules, 1993,26,6570.
1201T. Zyung, D.-H. Hwang, I.-N. Kang, H.-K. Shim, W.-Y. Hwang and J.-J. Kim,
Chem. Mater., 1995,7, 1499.
121 M. Hay and F. L. Klavetter, J. Am. Chem. SOC.,1995,117,7112.
122 G. G. Malliaras, J. K. Herrema, J. Wildeman, R. E. Gill, R. H. Wieringa, S. S.
Lampoura and G. Hadzionnou, Proc. SPIE Int. SOC.Opt. Eng., 1993,2025,441.
123 G. G. Malliaras, J. K. Herrema, J. Wildeman, R. H. Wieringa, R. E. Gill, S. S.
Lampoura and G. Hadzionnou, Adv. Mater., 1993,5,721.
124 J. K. Herrema, P. F. van Hutten, R. E. Gill, J. Wildeman, R. H. Wieringa and
G. Hadzionnou, Macromolecules, 1998,28,8 102.
125 F. Garten, A. Hilberer, F. Cacialli, E. Esselink, Y. van Dam, B. Schlatmann, R. H.
Friend, T. M. Klapwijk and G. Hadziioannou, Adv. Muter., 1997,9, 127.
126 H.-J. Brouwer, V. V. Krasnikov, A. Hilberer and G. Hadziioannou, Adv. Mater.,
1996,8,935.
127 A. Hilberer, M. Moroni, R. E. Gill, H.-J. Brouwer, V. V. Krasnikov, T.-A. Pham,
G. G. Malliaras, S. Veenstra, M. P. L. Werts, P. F. van Hutten and G. Hadzionnou,
Macromol. Symp., 1997, 125,99.
128 K. Yoshino, K. Tada, M. Hirohata, R. Hidayat, S. Tatsuhara, M. Ozaki, A. Naka
and M. Ishikawa, Jpn. J. Appl. Phys., 1997,36, L1548.
129 J. Oberski, R. Festag, C. Schmidt, G. Lussem, J. H. Wendorff, A. Greiner,
M. Hopmeier and F. Motamedi, Macromolecules, 1995,28, 8676.
130 G. Lussem, R. Festag, A. Greiner, C. Schmidt, C. Unterlechner, W. Heitz, J. H.
Wendorff, M. Hopmeier and J. Feldmann, Adv. Mater., 1995,7,923.
131 G. Lussem, F. Geffarth, .A. Greiner, W. Heitz, M. Hopmeier, M. Oberski,
C. Unterlechner, and J. H. Wendorff, Liq. Cryst., 1996,21,903.
132 H. Vestweber, J. Oberski, A. Greiner, W. Heitz, R. F. Mahrt and H. Bassler, Adv.
Muter. Opt. El., 1993,2, 197.
133 H. Vestweber, R Sander, A. Greiner, W. Heitz, R. F. Mahrt and H. Bassler, Synth.
Met., 1994,64, 141.
134 J. Birgerson, K. Kaeriyama, P. Barta, P. Broms, M. Fahlman, T. Granlund and
W. R. Salaneck, Adv. Mater., 1996,8,982.
135 I.-N. Kang, D.-H. Hwang, H.-K. Shim, T. Zyung and J.-J. Kim, Macromolecules,
1996,29, 165.
136 J. Pommerehne, H. Vestweber, W. Guss, R. F. Mahrt, H. Bassler, M. Porsch and
J. Daub, Adv. Mater., 1995,7, 551.
137 X.-C. Li, F. Cacialli, M. Giles, J. Griiner, R. H. Friend, A. B. Holmes, S. C. Moratti
and T. M. Young, Adv. Mater., 1995,76, 153.
138 F. Cacialli, X.-C. Li, R. H. Friend, S. C. Moriatti and A. B. Holmes, Synth. Met.,
1995,75, 161.
139 J.-K. Lee, R. R. Schrock, D. R. Baignent and R. H. Friend, Macromolecules, 1995,
28, 1966.
140 D. R. Baignent, N. C. Greenham, J. Gruner, R. N. Marks, R. H. Friend, S. C.
Moratti and A. B. Holmes, Synth. Met., 1994,67, 3.
141 J. J. M. Halls, D. R. Baignent, F. Cacialli, N. C. Greenham, R. H. Friend, S. C.
Moratti and A. B. Holmes, Thin Solid Films, 1996,276, 13.
142 J.-L. Bredas and A. J. Heeger, Chem. Phys. Lett., 1994,217,507.
220 Chapter 6
143 C. W. Tang and S. A. Van Slyke, Appl. Phys. Lett., 1987,51,913.
144 C. Adaichi, S. Tokito, T. Tsutsui and S. Saito, Jpn. J. Appl. Phys., 1998,28, L269.
145 A. R. Brown, D. D. C. Bradley, J. H. Burroughes, R. H. Friend, N. C. Greenham,
P. L. Burn, A. B. Holmes and A. Kraft, Appl. Phys. Lett., 1992,61,2793.
146 Y. Yang and Q. Pei, J. Appl. Phys., 1995,77,4807.
147 E. Buchwald, M. Meier, S. Karg, P. Posch, H.-W. Schmitt, P. Strohriegel, W. Riess
and M. Schwoerer, Adv. Mater., 1995,7,839.
148 Q. Pei and Y. Yang, Chem. Mater., 1995,7,1568.
149 X.-C. Li, A. Kraft, R. Cervini, G. C. W. Spencer, F. Cacialli, R. H. Friend,
J. Gruner, A. B. Holmes, J. C. DeMello and S. C. Moratti, Mat. Res., Symp. Proc.,
1996,413, 13.
150 I. D. Parker, Q. Pei and M. Marrocco, Appl. Phys. Lett., 1994,65, 1272.
151 M. Strukelj, F. Papadimitrakopoulos, T. M. Miller, L. J. Rothberg, Science, 1995,
267,1969.
152 M. Strukelj, T. M. Miller, F. Papadimitrakopoulos and S. Son, J. Am. Chem. SOC.,
1995,117,11976.
153 R. Fink, C. Frenz, M. Thelekkat and H.-W. Schmitt, Macromof.Symp., 1997,125,
151.
154 D. O’Brien, M. S. Weaver, D. G. Lidzey and D. D. C. Bradley, Appl. Phys. Lett.,
1996,69,881.
155 M. Thelekkat, R. Fink, P. Posch, J. Ring and H.-W. Schmitt, Polym. Prepr., 1997,
38,323.
156 T. Kanabara and T. Yamamoto, Chem. Lett., 1993,1459.
157 T. Yamamoto, T. Inoue and T. Kanabara, Jpn. J . Appl. Phys., 1994,33, L250.
158 T. Fukada, T. Kanabara, T. Yamamoto, K. Ishikawa, H. Takezoe and A. Fukuda,
Appl. Phys. Lett., 1996,68,2346.
159 T. Fujii, M. Fujita, Y. Hamada, K. Shibata, Y. Tsujino and K. Kuroki, J.
Photopolym. Sci. Technol., 1991,4,135.
160 M. Meier, E. Buchwald, S. Karg, P. Posch, M. Greczmiel, P. Strohriegel and
W. Riess, Synth. Met., 1996,76,95.
161 M. Greczmiel, P. Posch, H.-W. Schmitt, P. Strohriegel, E. Buchwald, M. Meier,
W. Riess and M. Schwoerer, Makromol. Symp., 1996,102,371.
162 X.-C. Li, T. M. Yong, J. Gruner, A. B. Holmes, S. C. Moratti, F. Cacialli and R. H.
Friend, Synth. Met., 1997,84,437.
163 X.-C. Li, F. Cacialli, M. Giles, J. Gruner, R. H. Friend, A. B. Holmes, S. C. Moratti
and T. M. Yong, Adv. Mater., 1995,7,898.
164 N. C. Greenham, S. C. Moratti, D. D. C. Bradley, R. H. Friend and A. B. Holmes,
Nature, 1993,365,628.
165 E. G. J. Staring, R. C. J. E. Demandt, D. Braun, G. L. J. Rikken, Y. A. R. R.
Kessener, A. H. J. Verhuizen, M. M. F. van Knippenberg and M. Boumans, Synth.
Met., 1995,71,2179.
166 S. C. Moratti, R. Cervini, A. B. Holmes, D. R. Baigent, R. H. Friend, N. C.
Greenham, J. Griiner and P. J. Hamer, Synth. Met., 1995,71,2117.
167 D. R. Baigent, P. J. Hamer, R. H. Friend, S. C. Moratti and A. B. Holmes, Synth.
Met., 1995,71,2175.
168 M. J. Marsella, D.-K. Fu and T. M. Swager, Adv. Mater., 1995,7, 145.
169 M. J. Marsella and T. M. Swager, Polym. Prepr., 1992,33, 1196.
170 J. Tian, C.-C. Wu, M. E. Thompson, J. C. Sturm, R. A. Register, M. J. Marsella and
T. M. Swager, Adv. Mater., 1995,7,395.
171 J. Tian, C.-C. Wu, M. E. Thompson, J. C. Sturm and R. A. Register, Chem. Mater.,
1995,7,2190.
172 A. Lux, A. B. Holmes, R. Cervini, J. E. Davies, S. C. Moratti, J. Griiner, F. Cacialli
and R. H. Friend, Synth. Met., 1997,84,293.
173 A. C. Grimsdale, X.-C. Li, F. Cacialli, J. Gruner, A. B. Holmes, S. C. Moratti and
R. H. Friend, Synth. Met., 1996,76, 165.
Organic Light-Emitting Diodes Using Light-Emitting Polymers 22 1
174 2.Boa, 2.Peng, M. E. Galvin and E. A. Chandross, Chem. Mater., 1998,10,1201.
175 2.Peng and M. E. Galvin, Chem. Mater., 1998,10, 1785.
176 2.Peng, 2.Boa and M. E. Galvin, Chem. Mater., 1998, 10,2086.
177 G. Lussem and J. H. Wendorff, Polym. Adv. Technol., 1998,9,443.
178 M. Remmers, D. Neher, J. Gruner, R. H. Friend, G. H. Gelinck, J. M. Warman,
C. Quattrocchi, D. A. dos Santos and J.-L. Brkdas, Macromolecules, 1996,29,7432.
179 G. Yu, H. Nishino, A. J. Heeger, T.-A. Chen and R. D. Rieke, Synth. Met., 1995,72,
249.
180 J. Birgerson, M. Fahlman, P. Broms and W. R. Saleneck, Synth. Met., 1996,80,125.
181 L.-N. Kang, D.-H. Hwang, H.-K. Shim, T. Zyung and J.-J. Kim, Macromolecules,
1996,29, 165.
182 N. Tanigaki, H. Masuda and K. Kaeriyama, Polymer, 1997,38, 1221.
183 H. Westweber, A. Greiner, U. Lemmer, R. F. Mahrt, R. Richert, W. Heitz and
H. Bassler, Adv. Mater., 1992,4,661.
184 T. W. Hagler, K. Pakbaz, K. F. Voss and A. J. Heeger, Phys. Rev., 1991,44,8652.
185 C . Weder, C. Sarwa, C. Bastiaansen and P. Smith, Adv. Mater., 1997,9, 1035.
186 C. Weder, C. Sarwa, A. Montali, C. Bastiaansen and P. Smith, Science, 1998, 279,
837.
187 P. Dyreklev, M. Berggren, 0. Ingenas, M. R. Anderson, 0. Wennerstrom and
T. Hjertberg, Adv. Mater., 1995,7,43.
188 K. Pichler, R. H. Friend, P. L. Bum and A. B. Holmes, Synth. Met., 1993, 55-57,
454.
189 M. Grell and D. D. C. Bradley, Adv. Mater., 1999, 11, 895.
190 Z. Bao, Y. Chen, R. Cai and L. Yu, Macromolecules, 1993,26,5281.
191 2.Bao, Y. Chen and R. Cai, Angew. Chem. Int. Ed. Eng., 1993,32,1345.
192 Y. Watanabe, T. Mihara and N. Koide, Macromol. Chem. Phys., 1998,199,977.
193 A. P. Davey, R. G. Howard and W. J. Blau, J. Mater. Chem., 1997,7,417.
194 A. Bacher, P. G. Bentley, D. D. C. Bradley, L. K. Douglas, P. A. Glarvey, M. Grell,
K. S. Whitehead and M. L. Turner, J. Mater. Chem., 1999,9,2985.
195 A. E. A. Contoret, S. R. Farrar, P. 0. Jackson, L. May, M. O’Neill, J. E. Nicholls,
S . M. Kelly, and G. J. Richards, Adv. Mater., 2000,12,971.
196 D. J. Broer, Adv. Polym. Sci., 1994,118, 1.
197 S. M. Kelly, J. Mater. Chem., 1995,5,2047.
198 S . M. Kelly, Liq. Cryst., 1998,24, 71.
199 X.-C. Li, T.-M. Yong, J. Gruner, A. B. Holmes, S . C. Moratti, F. Cacialli and R. H.
Friend, Synth. Met., 1997,84,437.
200 M. S. Bayerl, T. Braig, 0. Nuyken, D. C. Miiller, M. GroD and K. Meerholz,
Makromol. Chem. Rapid Commun., 1999,20,224.
201 Philips Web Site.
202 CDT Web Site and the Cavendish Laboratory Web Site.
CHAPTER 7
Conclusions and Outlook
The various types of liquid crystal displays (LCDs) undeniably represent the
dominant flat panel display technology at the present time and probably for the
foreseeable future.'-' In many cases they have not replaced a display unit based
on another technology in an existing instrument or device, but have provided
the enabling technology to allow a new type of instrument requiring a visual
display unit to be manufactured or rendered it commercially viable. Small
monochrome TN-LCDs with directly driven segments are used in digital
watches, clocks, thermometers, measurement instruments, portable and static
music systems, etc. Medium-sized TN-LCDs and STN-LCDs with multiplex
addressing are used in calculators, handheld computer games, point-of-sale
registers and mobile telephones, to name but a few applications. Large-area
LCDs are used primarily in portable instruments, such as camcorders, personal
digital assistants, notebooks and laptop computers. The variety of instruments
and applications incorporating LCDs is reflected in the fact that 1.5 billion
LCDs worth over $13 billion were sold in 1999. The success of LCD technology
can be attributed, amongst several other essential factors, to continual progress
in the design and synthesis of new classes of stable nematic liquid crystals with
an improved combination of physical property values for a particular type of
LCD for a specific electro-optical device application.
1 Portable Computers
In the short to medium term the most important market segment for LCDs in
terms of value, volume growth and profitability will still be laptop computers
and notebooks. Eighteen million laptop computers with an LCD screen were
sold in 1999. These screens are relatively large (12-14) and bright (> 100 200 cd
m-2), with full colour (>250000) and high information content, e.g.
1024 x 768 x 3 (RGB) pixels at high contrast (> 1OO:l) with short response
times (z50-100 ms) and low power consumption (> 10 W). The great majority
(92%) are TFT-TN-LCDs with active matrix addressing provided by thin-film
transistors, while the rest incorporated an STN-LCD with a high degree of
multiplex addressing. The optical performance of TFT-TN-LCDs is markedly
superior to that of STN-LCDs of comparable size in spite of continual
222
improvements in STN-LCD performance. However, the power consumption of
STN-LCDs is much higher ( x 4) and response times significantly longer ( x 3)
than those values of a comparable TFT-TN-LCD with a much lower contrast,
e.g. 25:l compared to 1OO:l. The market share of TFT-TN-LCDs has risen
inexorably as the unit cost has decreased and the size has increased. New
nematic liquid crystals with a low viscosity and a high dielectric anisotropy
would allow lower voltages to be used, e.g. 3.3 V instead of 4.0 V. This would
extend the battery lifetime by lowering the power consumption. A lower
viscosity would also shorten response times towards video-rate addressing
times. This would reduce blurring and increase resolution. A higher resistivity
of the nematic mixture would increase the holding ratio of the device with a
resultant increase in contrast. Continued progress in developing improved
reflective polarisers using chiral nematic films (plus quarter-wave plate), multi-
layer sheets or anisotropically scattering films would improve the viewing angle
and brightness considerably.
2 Desktop Computers
Larger versions of the active-matrix addressed screens used in laptop computers
will become increasingly important for use in non-portable desktop computer
monitors (12"-21"). LCDs are set to steadily erode the market share of CRTs
due to the advantages of low power consumption, i.e. lower running costs,
absence of flicker and emission, portability and, most importantly, much
smaller footprint. The cost disadvantage of LCDs of this size and complexity
compared to that of CRTs is still more than a factor of three, but the price
differential is decreasing constantly and quickly. The size of commercial LCDs
at a realistic price for monitors is also increasing. However, the cost of a CRT
for the monitor market is still very low and has been more-or-less constant in
real terms for several years. The performance of CRTs is also being continually
improved. The cost of an LCD with active matrix addressing of comparable size
has fluctuated somewhat from year to year, but on average it has fallen
substantially in real terms over the last five years. The total computer monitor
market is large, i.e. 100 million units in 1999 and expanding rapidly. However,
the growth in the number of LCD monitors is three times greater than that of
CRTs, although from a relatively modest level ( > 3 million units). Three
competing types of LCD with active matrix addressing share the market for
LCD monitors. Four times as many TFT-TN-LCDs are still sold as LCDs
based on in-plane switching (TFT-IPS-LCD) or vertically aligned nematics
(VAN-TFT-LCD) technology, both with thin-film transistor active matrix
addressing. However, the market share of the latter two LCD types is increasing
steeply at the cost of the former. A lower viscosity of the nematic mixture would
clearly improve optical performance of all three display types by reducing
response times towards video rate. The inability of LCDs to operate at real
video-rate frame times is the greatest limitation of LCD technology at the
present time, more than size, cost, low brightness, insufficient contrast and
limited viewing angles at acceptable contrast.
224 Chapter 7
3 Mobile Telephones
In 1999 over 200 million mobile telephones were manufactured with an STN-
LCD display unit. Nematic components with a larger dielectric anisotropy
would allow lower drive voltages to be used. This would in turn reduce the
power consumption and prolong the lifetime of the battery. A steeper electro-
optical response curve would allow more information to be displayed. Mono-
chrome OLEDs are an attractive option for mobile telephones from a consumer
appeal aspect due to the ability for product differentiation and novelty.>'*
Hybrid OLED-LCDs with one polariser replaced by an electroluminescent
organic material emitting plane polarised light could also break into this market
segment. PolyLED, a business unit of the Dutch consumer electronics division
of Philips, intends to launch a monochrome OLED backlight using light-
emitting polymers in 200 1. Other contenders are surface-stabilised, chiral-
nematic displays or bistable nematic displays, although these are still in the
research and development stage. Larger LCDs with a higher information
content are required for mobile telephones with internet capabilities.
4 Televisions
The very low cost and high performance of CRTs with small to very large
screens (12"-42") means that alternative technologies have failed to break into
the market for non-portable TVs. The relative merits of CRT and LCD
technology are reflected in the high market share of the former and the small
market share of the latter for non-portable TVs and the inverse situation in the
market for small, battery-driven, portable TVs. Very large plasma flat panels
may capture a greater market share as their price continues to fall. The large
number of processing steps during manufacture lead to the high production
cost, rather than any other factors such as materials cost. However, lifetime,
luminous efficiency and image quality are still issues to be addressed. Another
possibility is a hybrid plasma-addressed LCD, where the plasma screen serves as
a replacement for active addressing. These are potentially large-area screens
with video-rate response times, full colour and high brightness. However, the
practical problems involved and the cost of fabricating large LCD screens above
1 4 is inhibiting the development of plasma addressed LCDs. Projection
displays using small LCD screens (1") with a very high pixel density have
enjoyed some success, although the higher intrinsic brightness of digital
micromirror displays (DMDs) suggests that they may occupy this niche in the
TV market.
5 Miscellaneous Consumer Products

A surprisingly large number of consumer appliances incorporate an LCD, e.g. 1
million personal digital assistants, 4 million digital cameras and 7 million video
cameras were produced in 1999. The requirement for short response times, low
operating voltages, low power consumption and high information content with
full colour of these devices means that competing flat panel displays will find it
very difficult to enter this part of the market. However, OLEDs could well find
application in car navigation systems due to the extreme temperature range
specifications and relatively low and unsophisticated information content.
Pioneer has already launched a range of monochrome OLEDs using small
organic molecules for automotive applications.
6 Future Prospects
The major limitations and unsatisfactory aspects of LCDs are generally low
brightness and restricted viewing angles with acceptable contrast. '-14 These
issues will have to be addressed in order for LCDs to maintain their position in
the marketplace. More efficient backlights will certainly play a role in reducing
power consumption and increasing brightness in standard LCDs. New types of
LCD under development, which reflect incident light rather than transmit light
from a backlight incorporated in the instrument, may offer a significant
improvement in display brightness. Reflective and transflective LCDs with one
or no absorptive polarisers may contribute to resolving these problems. Both
reflective and transflective LCDs consume less power due to the use of reflected
incident light. The absence of a backlight in a reflective LCD would lead to
thinner, lighter displays with much lower power consumption. However, only a
low proportion of incident light is reflected. A more promising kind of LCD is a
black guest-host LCD with colour filters and active matrix addressing.
However, the contrast is too low for practical applications. Therefore, the
most promising approach is the further improvement in the physical properties
of the components of nematic mixtures. Sub-pixellisationimproves the viewing
angle dependence of LCDs with the optical axis not in the plane of the cell, e.g.
TN-LCDs and STN-LCDs. Therefore, this may be introduced in the medium
term. This may require the introduction of non-contact alignment layers where
the alignment direction is determined by the use of plane polarised UV light.
Compensation layers will play a significant part in improving the optical
performance of LCDs. Significantly increased size at acceptable cost is a much
more intractable problem. However, this is a fabrication issue, rather than being
dependent on material development. The use of plastic substrates will reduce
weight and perhaps improve ruggedness for portable applications.
Of the emergent, competing flat-panel technologies large-area plasma display
panels (PDPs) occupy a niche for very large direct view TV and as commercial,
very large visual display units. This may expand into the consumer TV market
with decreasing unit cost and improved resolution. Digital micromirror devices
(DMDs) have great potential in projection displays for TVs and cinemas.
However, along with flat-panel cathode ray tubes (CRTs) and vacuum fluores-
cence displays (VFDs), their considerable weight, high operating voltages and
power consumption excludes this display type from portable applications.
Interest in field emission displays (FEDs) has already waned and research and
development is limited to a few specialist companies and universities.
Only organic light emitting diodes (OLEDs) using either organic low-molar-
226 Chapter 7
mass materials deposited by vacuum sublimation or polymers deposited by
spin-coating or doctor blade techniques exhibit a combination of properties
suitable for portable applications. The low-weight, power consumption and
operating voltages combined with high brightness and good viewing angle
dependence of contrast means that OLEDs are an attractive alternative
technology to LCDs. However, the issue of power consumption and high
information content will have to be successfully addressed, if they are to gain a
significant market share of the flat panel displays market. High-information-
content OLEDs with active matrix addressing exhibit twice the power con-
sumption of a comparable LCD with backlighting. This is a direct consequence
of their different modes of operation. OLEDs emit light on the passage of an
electrical current and, therefore, current drivers are required. LCDs operate
using a field effect where the optic axis is reoriented by the application of an
electric field and very little power is required. The high power consumption of
LCDs with backlighting is due to the backlight itself. This explains the interest
in developing reflective or transflective LCDs.
The first production lines for OLEDs using low-molar-mass small molecules
and conjugated polymers have been commissioned and products are appearing
on the flat panel displays market. Pioneer is already supplying OLEDs
(64 x 256 pixels) for automotive audio applications using technology licenced
from Kodak involving low-molar-mass materials with red, green and blue
elements. Sanyo and Kodak are collaborating on the development of a full-
colour, high-information-content OLED panel (852 x 222 pixels) with active
matrix addressing on polysilicon, rather than amorphous silicon, also using
low-molar-mass electroluminescent molecules. A commercial product designed
for camcorders and PDAs is expected in 2001. Philips has commissioned a
production line in Heerlen in the Netherlands for the fabrication of OLEDs
utilising light-emitting polymers. The initial use of these OLEDs produced
under licence from CDT in the UK will probably be as backlights for LCDs.
The Philips PolyLED business unit has also demonstrated a multiplexed OLED
with more than 7000 pixels and 256 grey levels. CDT, initially a spin-off from
the University of Cambridge, is a quoted company and currently has more than
50 employees. The UNIAX Corporation of Santa Barbara, USA, employs over
30 people and has also recently commissioned an OLED production line using
light-emitting polymers. The first products include small (2” diagonal) multi-
plexed OLEDs with over 600 pixels. Light-emitting polymers are manufactured
by Covion Organic Semiconductors. Some of therse are manufactured under
licence from CDT. A collaboration between CDT and Seiko-Epson has
generated a full-colour OLED using ink-jet printing of soluble LEPS onto a
poly-silicon active matrix substrate. OPSYS is a spin-off from Oxford Uni-
versity and employs more than 25 research staff. The prime interest of OPSYS is
the development of electroluminescent metal chelates based on the lanthanides.
This is also the case for ELAM-T, a start-up company linked to the South Bank
University .
Therefore, although the market share of OLEDs is now very small, i.e. $3
million in 1999, it is projected to increase significantly in the first years of the
new millennium, i.e. $500 million in 2005.5 This will probably be at the expense
of standard LEDs using inorganic semiconductors and vacuum fluorescent
displays (VFDs), whose production volumes may actually decrease even though
the overall market for flat panel displays is still expanding r a ~ i d l yIn
. ~the near
term OLEDs will probably be used as backlights for LCDs and simple
monochrome and colour displays with relatively low information content.
Over the medium term OLEDs will probably be integrated in mobile tele-
phones, PDAs camcorders and digital cameras. A potentially important market
for, OLED technology over the long term may well be light-weight FPDs
supported on plastic substrates for a variety of portable devices. OLED
technology on plastic substrates would have the added advantage of being
compatible with low-cost roll-to-roll fabrication processes. Furthermore, Uni-
versal Display Corporation (UDC) of the USA has demonstrated flexible
OLEDs on plastic substrates. OLEDs with a microcavity configuration offer
the possibility of tuning the emission characteristics of bright OLED displays
and, perhaps, the production of OLED lasers. However, the various research
teams in companies and universities developing OLED technology will have to
resolve a range of problems, such as device life-times, power efficiency, stable
chromophores with matched chromaticity over time, patterning and addressing
and, above all, large-scale, low-cost manufacture, before OLEDs are produced
in bulk and start to capture significant market share from LCDs.
The significant growth in the number and value of all kinds of flat panel
displays is set to continue. This is due, at least in part, to steady growth in the
market for consumer products with digital displays and information-technology
products such as mobile communications, calculators, personal digital assis-
tants and portable computers. It is also partially due to the gradual erosion of
the CRT share of the market for non-portable computer monitors. However,
the explosive growth in the use of mobile telephones, especially with internet
access in the future, is also making a very significant contribution to growth in
the flat panel displays market despite the small size of the display. Mobile
telephones with internet access will require larger, more sophisticated full-
colour displays in the near future. A review of the materials chemistry aspects of
LCDs and OLEDs, such as this monograph, suggests that new applications of
both technologies are to be expected. This will be reflected in the volume and
value of the high-value-added organic materials manufactured for use in LCDs
and OLEDs. In turn this will help finance research and development in efforts to
design and synthesise new liquid crystals and organic electroluminescent and
charge-transport materials, both low-molar-mass materials and polymers, with
an improved spectrum of physical properties.
7 References
1 D. Coates, E. Merck, UK, personal communication.
2 E. P. Raynes, Proc. SID Euro Display '96, 1996,7.
3 T. Geelhaar, Liq. Cryst., 1998,24,91.
4 M. Thompson, Proc. SID EID '98, 1998,6.1,
228 Chapter 7
5 Stanford Resources, California, USA.
6 P. May, Proc. SID Euro Display '96, 1996,613.
7 D. D. C. Bradley, Proc. EMD Workshop '97, 1997.
8 K. Pichler, Proc. EMD Workshop '97, 1997.
9 N. Bailey, Information Display, 2000,3, 12.
10 M. T. Johnson and A. Sempel, Information Display, 2000,2, 12.
11 L. J. Rothberg and A. J. Lovinger, J. Mater. Res., 1996,11,3174.
R. H. Friend, R. W. Gymer, A. B. Holmes, J. H. Burroughes, R. N. Marks, C. Talliani,
D. D. C. Bradley, D. A. Dos Santos, J. L. Brkdas, M. Logdlund and W. R. Salaneck,
Nature, 1999,397, 121.
13 J. R. Sheats and P. F. Barbara, Acc. Chem. Res., 1999,32, 191.
14 S. Forrest, P. E. Burrows and M. E. Thompson, Chem. Ind., 1998,12,1023.
Subject Index
Addressing, Compensation (retardation) films, 38

LCDs, 28 Fuji Film, 40
Active AddressingTM,86 polycarbonate, 39
active matrix addressing, 6,28,32,81 polystyrene, 39
direct addressing, 28, 30 Copolymers, 192, 195, 197,200,203,205
multiplex addressing, 28, 30,74 Coumarins, 168
plasma, 224 Current density, 140
OLEDs, 134,143 Cyanobicyclohexyl compounds,
Alkenyl compounds, 72,73,96,97,98,100 axially-substituted, 59
Alt and Pleshko, 85,90 equatorially-substituted, 70, 71,95
Aluminium tris(quinolate), Alq3, 137, 153, Cyanobiphenyls, 68,95
169 Cyanopyrimidines, 69,95
Anthracene, 134 Cyclohexane compounds, 58,70,73,77,
Arylamines, 153, 158, 168, 171 78, 83,84,95
Azo compounds, 13,34,36, 105,109, 117 Cyclohexenes, 70
Azoxy compounds, 13
DCM, 169
Band gap, 199 Deformation of vertically aligned (DAP)
Bicyclo[2.2.2]octanes, 17, 70,71,77,95, LCD, 54
123 Desktop computers, 223
Boroxinanes, 70 Dichroic dyes, 103
anthraquinone dyes, 106, 119
Calamitic liquid crystals, 16 azo-dyes, 105, 117
Carbonates, 14 azulene-dyes, 109, 122
Cathode ray tubes, CRTs, 1, 183,225 merocyanine-dyes, 105, 119
Chiral dopants, 65 tetraline-dyes, 109, 121
Chiral nematics, 12, 52 Dichroic ratio, 106
Cholesteric-nematic phase change Dicyanobenzene compounds, 58
(CNPC)-LCD, 5 1 Difluorophenyl compounds, 60
Cholesterol derivatives, 10, 53, 180 Digital micromirror devices, DMDs, 4,
Columnar liquid crystals, 15, 158, 160 225
coronenes, 162 Dioxanes, 7 1,95
phthalocyanines, 160 Dithianes, 70
polycatenars, 162 Dynamic scattering mode (DSM) LCD,
triphenylenes, 160 25,45
229
230 Subject Index
E7 and E8,68 Field effects, 26
EBBA, 66 Field emission displays, FEDs, 4,225
Electrically controlled birefringence (ECB) Flat panel cathode ray tube, 2
LCD, 53 Flexoelectric effects, 26
Electrodes for LCDs, 28 Fluorobicyclohexyl compounds,
Electrodes for OLEDs, axially-substituted, 59
anode, 138, Forster transfer, 151
gold, 149, Frkedericksz effect, 53
indiumltin oxide, ITO, 28, 142, 150
PEDOT, 142,181 Gooch and Tarry, 63, 125
polyaniline salts, PANi, 142, 151, Guest-host (GH) LCDs, 103
181 Heilmeier and Zanoni GH-LCD, 110
selenium/tellurium, 149 STN-GH-LCD, 87,115
cathode, 138 White and Taylor GH-LCD, 110,112
aluminium, 142
aluminium oxide, 142 Hairy-rod polymers, 180, 186
calcium, 142 Holding ratio, 8 1
gold, 142 Hole transport layer, HTL, 140, 196
magnesium/silver, 150
ZIVcurve, 140,215
sodium/potassium, 149
Indium/tin oxide, 28, 142
Electrohydrodynamic effects, 25
In-plane switching (IPS) LCD, 124
Electroluminescence, 134, 138
polarised electroluminescence, 165, 174, Kirkwood Froehlich factor, 76,96
208
Electroluminescent liquid crystals, Ladder polymers, 194
low-molar-mass nematic glasses, 165, Langmuir-Blodgett layers, 163
212 Layer deposition,
nematic networks, 166,210 Doctor blade, 135
polymers, 208 gel-processing, 208
Electron transport layer, ETL, 140, 196 Langmuir-Blodgett, 163
Electrons and holes, 138 laser ablation, 166
injection, 150 spin coating, 135
mobility, 156, 158 vapour deposition, 155
columnar liquid crystals, 160 Life-times of OLEDs,
nematic liquid crystals, 158 grain boundaries, 2 10
smectic liquid crystals, 163 ionic impurities, 184
recombination, 138, 179 moisture, 2 15
Emission layer, 140, 152 oxidation, 184,215
Emission spectra, 136, 183,214 Light emitting diodes, LEDs, 5, 134,227
Energy level diagram, 141 Light emitting polymers, LEPs, 135, 179,
Energy levels, HOMO/LUMO, 135, 141, 184
183 polarised emission, 208
Esters, 13,49,67,72, 73, 77, 78, 80, 109 Liquid crystal displays, LCDs, 6,25
Ethers, 13,49,73,78, 84, 100 cell construction, 27
Ethoxy-linked compounds, 72 colour filters, 28
Ethyl-linked compounds, 72,83 pixels, 28
Eutectic mixtures, 68 Low-molar-mass electroluminescent
Excimers, 170 materials, LMMMs, 135
Excited states, 139, 151, 182 Luminous efficiency, 139
Excitons, 138, 182 Lyotropic liquid crystals, 10
Subject Index 23 1
MBBA, 50 Oxadiazoles, 156, 168, 171,200
MEH-PPV, 180, 184
Metal chelates, 153, 166, 169, 172 PANi, 142,151,181
Mobile telephones, 224 PBD, 153,156,199
Multilayer organic light emitting diodes, PCHs, 70,71,73,95,97
OLEDs, 5, 135,140 PECHs, 71
bilayer, 135, 140 PEDOT, 142, 181
monolayer, 135, 138 Perylene, 169
trilayer, 135 Phosphors, 3
Photoluminescence, 134, 136
Naphthalenedicarboximides,156, 169 Photons, 139
Nematic liquid crystals, 9 Piperazines, 70
physical properties, 9, 20 Piperidines, 70,71
birefringence, 20 Pixels, 28, 134
dielectric anisotropy, 24 Plasma display panels, PDPs, 3,225
elastic constants, 22 Polarised luminescence,
optical anisotropy, 20 electroluminescence, 165, 174,208
typical values, 18 photoluminescence, 208
viscosity, 23 Poly(arylenevinylenes), 180
Nematic materials of negative dielectric copolymers,<185
anisotropy, 56, 123 MEH-PPV, 184
Nematic materials of positive dielectric soluble derivatives, 184
anisotropy, 58,66, 70, 80 Polycatenar columar liquid crystals, 162
Nitriles, 95 Poly(fluorenes), 190,210
aliphatic, 58, 70, 71 Polyfluorinated compounds, 82, 84,200,
aromatic, 58,67, 70, 72, 73, 80, 83,203 204
Normally white (NW) mode, 66 Polyimide, 34,210
Polymer blends, 196,207
Optical retardation, 22, 55 guest-host, 184
Order parameter, 11, 104 Polymer dispersed (PD) LCDs, 26
Organic light emitting diodes, OLEDs, Poly(methy1methacrylate), 199,207
134,226 Poly(norbornenes), 197
using low-molar-mass materials, 147 Poly(oxadiazoles), 205
bilayer, 151 Poly(p-phenyleneethynylenes), 185, 188
monolayer, 150 Polyb-phenylenes), 190,210
trilayer, 154 Poly(p-phenylenevinylene), PPV, 135
using light-emitting polymers, 179 monolayer OLED using PPV, 179
bilayer, 196 partially converted, 196
monolayer, 179 spacers, 195
trilayer, 206 synthesis, 184
Organic polymer alignment layers, 33 Poly(phenylquinoxaline), 201,207
anchoring energy, 33 Poly(pyridines), 193,203,204
inorganic alignment layers, 34 Poly(quinoline), 20 1
rubbed polyimide alignment layers, 34 Poly(thiophenes), 189
non-contact alignment layers, 34 Poly(N-vinylcarbazole), PVK, 20 1,207
ablated alignment, 37 Portable computers, 222
azo dyes, 34 Power efficiency, 139
cinnamate side-chain polymers, 35 Pyridazines, 58, 59
coumarin side-chain polymers, 36 Pyridines, 58,67,70
self-assembled monolayers, SAMs, 34 Pyrimidines, 67,80,84,95
232 Subject Index
Quantum efficiency, STN-LCD, 90
electroluminescence, 139, 183 STN-GH-LCD, 87
fluorescence, 139 Surface stabilised ferroelectric (SSF)
Quenching, aggregation, 192 LCDs, 9
Quinacroline, 168
Television, 224
Rare earth metal chelates, 169 Thermotropic liquid crystals, 10,209
Rectification ratio, 138 Thiobenzazole, 165
Rhodamine metal complexes, 169 Thiopyran sulphones, 156
Tolanes, 13,49, 72,84
Schiff’s bases, 48 TPD, 158
Side-chain polymers, 197,202, 205 Triazoles, 156, 200
Silanes, 180 Twisted nematic (TN) LCD, 60,225
Singlets and triplets, 139, 151
Smectic liquid crystals, 15, 69, 99, 158
Stilbenes, 13, 14,49,67,72 Vacuum fluorescence displays, VFDs, 3,
Super twisted nematic (STN) LCDs, 32, 225
85,88,225 Vertically aligned phase (VAN) LCD, 54
bistable STN-LCD, 86
DSTN-LCD, 93 Walden’s rule, 158
OMI-LCD, 91 Williams domains, 25,46
SBE-LCD, 88 Work function, 142, 152

Flat Panel Displays - Advanced Organic Materials

Uploaded by

Copyright:

Available Formats

Flat Panel Displays - Advanced Organic Materials

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Flat Panel Displays - Advanced Organic Materials

Uploaded by

Copyright:

Available Formats

What are flat panel displays and how do they work?

What are flat panel displays and how do they work?

What are some common organic materials used in flat panel displays?

What are some common organic materials used in flat panel displays?

Advanced Organic Materials

Series Editor: J.A. Connor, Department of Chemistry,

Flat Panel Displays: Advanced Organic Materials

How to obtainfuture titles on publication

Sales and Customer Care

Flat Panel Displays

0The Royal Society of Chemistry 2000

Published by The Royal Society of Chemistry,

For further information see our web site at www.rsc.org

Typeset by Paston PrePress Ltd, Beccles, Suffolk

Chapter 1 Flat Panel Displays 1

Chapter 2 Liquid Crystals and Liquid Crystal Displays (LCDs) 9

Chapter 4 Photoluminescence and Electroluminescence from

Chapter 5 Organic Light-Emitting Diodes Using Low-Molar-

Chapter 6 Organic Light-Emitting Diodes Using Light-

Conclusions and Outlook 222

Subject Index 229

Flat Panel Displays

1 Flat Panel Displays

Type offlat panel display (FPD) Number of units

Flat-Panel Cathode Ray Tubes2

Plasma Display Panels3

Vacuum Fluorescence Displays4

Field Emission Displays596

Digital Micromirror Devices798

Inorganic Semiconductor Light-Emitting Diodesg

Organic Light-Emitting Diodes"

Liquid Crystal Displays' ' '*

Liquid Crystals and Liquid

1 Physical Properties of Nematic Liquid Crystals

CRYSTAL, Cr CHIRAL NEMATIC PHASE, N* ISOTROPIC LIQUID, I

Figure 2.1 A schematic representation of the melting of crystalline cholesteryl benzoate

CRYSTAL, Cr SMECTIC NEMATIC ISOTROPIC

COS* e = -31 (2)

and S is 0. In an ideal macroscopically ordered nematic liquid crystalline state

COS~ e=I (3)

and S = 1. For a typical compound in the nematic state at a temperature

Molecular structure Cr N I Ref

8 a 118 121 a 38,39

11 a 113 53) a 38,42

Molecular structure Cr N I Ref

Parentheses represents a monotropic transition temperature.

of a bulk sample of a material in the nematic phase can be macroscopically and

neighbouring molecules. This can be made use of in organic light-emitting

Parentheses represent a monotropictransition temperature.

Melting point (Cr-N) 22.5 "C 30 "C

a Measured at 29 "C;bmeasured at 20 "C;

Compounds with an odd number of methylene (CH2) or oxygen atoms in the

2 Physical Properties of Liquid Crystals 1,15,25,56-58

Optical Anisotropy (Birefringence)

An(th) = n,(O) -no

where n, = n,(O = 90'). Therefore, An is the difference between the refractive

where 1, is the wavelength of light in a vacuum. The amount of optical

SPLAY k i i TWIST k22 BENfi k33

Figure 2.10. The dependence of the dielectric constants, ELand ~ ~ measured