91av

One of the first things you learn in school science lessons is that
there are three states of matter-solid, liquid and gas. This is, however,
not completely true because there is a fourth state that lies between a
liquid and a solid. It is called the liquid crystalline state.

Most people associate liquid crystals with digital watches and laptop
computer screens. Liquid crystals, however, are much more common than that.
Most biological systems, ourselves included, rely on the liquid crystalline
state to hold them together-cell membranes are, in effect, liquid crystals.
Many kinds of modern synthetic liquid crystals have unusual mechanical and
electrical properties that engineers and technologists are only just beginning
to exploit.

The discovery of liquid crystals resulted from microscope techniques
developed in the 19th century. Doctors were already using microscopes for
research, and it was a 19th-century German physician, Rudolf Virchow, who
was the first person to observe a liquid crystalline phase through a microscope.
In 1853 he discovered myelin, the material that sheathes nerves, although
at the time he did not realise it was a liquid crystal. It was not until
the second part of the century that the German physicist Otto Lehmann, who
was interested in studying melting, identified the liquid crystalline state.

By 1888 Lehmann was already well known for his studies of crystallisation,
which he observed using a microscope, and he had just been made a professor
at the Polytechnic in Aachen. On 14 March, he received a letter from the
Austrian botanist Friedrich Reinitzer, who was working in Prague on a compound
he had recently synthesised called cholesteryl benzoate. Like any good organic
chemist, Reinitzer had begun to characterise his new material by measuring
its melting point. To do this, he used a microscope on which he could heat
his new compound while he observed its appearance. Reinitzer knew that a
pure substance should change from being a solid to a liquid at a precise
temperature, which is characteristic of the compound.

When Reinitzer tried to measure the melting point of cholesteryl benzoate,
he noticed something very unusual: the compound appeared to have two melting
points. At 145.5 °C, the white crystals first melted to form a cloudy
liquid. Then at 178.5 °C the liquid suddenly cleared. On cooling the
clear liquid, Reinitzer observed that the reverse happened.

Reinitzer had enclosed some of his sample in the letter, and Lehmann
set about examining the cloudy, ‘in-between’ phase, or mesophase, as it
is now called. Lehmann looked at the cholesteryl benzoate under his microscope
and found that the mesophase did unexpected things to polarised light. Unlike
a normal liquid, which would appear black when viewed through a simple device
called ‘crossed polarisers’, the mesophase appeared to be brightly coloured.

To understand what this means, you need to appreciate that ordinary
sources of light produce a mixture of electromagnetic waves that oscillate
in all directions. If the light wave oscillates in just one plane, it is
said to be polarised. You can select a particular plane of polarisation
from a beam of light by interposing a polariser-a piece of Polaroid, for
example. This is how Polaroid sunglasses cut down the amount of sunlight
reaching the eyes. If the plane-polarised light passes through a second
polariser with its plane of polarisation at right angles to the first-a
‘crossed polariser’-no light can get through. A transparent material inserted
between the first and the second polariser should not make any difference-it,
too, ought to appear black, but this is not always the case. Lehmann already
knew that crystalline solids can rotate the plane of polarisation of light,
thereby allowing light through the crossed polarisers.

It works like this: because light consists of an oscillating electromagnetic
field, when it travels through a material, it will cause the electrons in
atoms and molecules to respond by vibrating backwards and forwards. They
cannot, however, respond instantaneously, which slows the light wave down.
This is the phenomenon of refraction. In some materials, depending on how
their atoms or molecules are arranged, the retarding effect of the electrons
is different in different directions. The speed of light then has two possible
values depending on the polarisation of the light relative to the crystal.

This is what gives rise to the effect of double refraction, which you
may have seen in the crystal calcite. The imbalance in refractive indices
also has the effect of rotating plane-polarised light and thereby allowing
light through a crossed polariser, the result being beautifully coloured
‘birefringent’ patterns (see the photographs above).

Crystals have a definite structure-a repeating three-dimensional pattern
of atoms or molecules-so you might expect some of them to be birefringent.
Liquids, on the other hand, have no particular arrangement of their atoms
or molecules, which are free to move about randomly. Therefore liquids have
just one refractive index and normally appear black through crossed polarisers.
So Lehmann and Reinitzer were surprised to see birefringence in the liquid
cholesteryl benzoate mesophase.

The two scientists worked busily through March and April of 1888, discussing
their observations and trying to determine what the mesophase was: liquid
or crystal? By the end of April, Lehmann’s conclusion was simply that the
substance looked strange, and nothing more was done for 16 months. Then,
towards the end of August 1889, Lehmann started work again and became convinced
that he was dealing with ‘very soft crystals’, which he christened fliessende
Kristalle, liquid crystals. The name has stuck.

How do liquid crystals arise? It was not until 1924 that Daniel Vorlander
of the University of Halle in Germany showed that materials with liquid
crystalline phases consist of molecules that are shaped like rods rather
than the expected spheres. In crystals of roughly spherical molecules, the
molecules are held together in a definite place-the molecules have ‘positional
order’. Rod-shaped molecules also have positional order, but in addition
they all point in the same direction. This is called ‘orientational order’
(see Figure 1).

When a normal crystal melts, thermal energy overcomes the molecular
forces, breaking up the structure and destroying the positional order. The
molecules are free to move randomly. If the molecules are rod-shaped, something
different can happen. At a certain temperature, the thermal energy may be
enough to destroy the positional order but insufficient to disrupt the molecular
forces responsible for the orientational order. This is what Reinitzer saw
when his crystals melted into a cloudy liquid. The resulting phase consists
of molecules lined up, more or less parallel, to each other but distributed
randomly in space. The orientational order extends over millions of molecules.
The direction of alignment is called the director.

What should be the properties of such a phase? The absence of positional
order means that it must be a fluid. The orientational order means that
certain physical properties-for example, refractive index-will vary, depending
on the direction at which the measurements are made relative to the director.
The phase, therefore, appears birefringent when viewed through crossed polarisers.

On further heating, the phase reaches a temperature at which the orientational
order is destroyed to give a normal liquid. This temperature corresponds
to the transition from the cloudy liquid to the clear liquid, often called
the clearing temperature. On cooling, the reverse happens as the rod-like
molecules assemble themselves into an ordered fluid structure.

This is the simplest liquid crystalline structure and is called the
nematic phase (see Figure 1c). Cholesteryl benzoate represents a particular
type called a chiral nematic phase; the name ‘chiral’ means that the rod-like
molecules have a handedness-rather like a screw, which is usually right-handed
but could be left-handed. When any neighbouring screws are touching, they
impart a small rotation to each other as the threads mesh. In a similar
way, chiral molecules in a nematic phase can impart a gentle rotation to
their neighbours. The chirality of the molecules reveals itself as a helical
twist of the director in the nematic phase: that is, the average molecular
direction twists around as we travel perpendicularly to it, like the rungs
of a spiral staircase.

The complete twist of the helix is often as long as the wavelength of
visible light. This means that the particular wavelength that is reflected
by this phase depends on the number of twists in a certain length, in much
the same way as the number of lines in a given length of an optical diffraction
grating dictates which wavelengths are reflected. Because Reinitzer observed
the first chiral phase in his cholesteryl benzoate, chiral nematic phases
are frequently called cholesteric phases.

Chiral nematic phases produced one of the first commercial applications
of liquid crystals. The selective reflection of light produces highly iridescent
colours. An increase in temperature diminishes the degree of twist, and
this changes the colours reflected. They are used in thermometers, and more
recently to make fashion fabrics that change colour with body heat.

There are other more complex types of liquid crystalline phase. For
example, on heating a crystal, its three-dimensional positional order may
not be lost all in one go. Instead, layers of molecules could form. One
layer of molecules would not interact with those in another layer, but molecules
would move randomly within each layer. These kinds of arrangement, where
some positional order is retained, are called smectic phases (see Figure
1b).

In fact, a rich diversity of liquid crystalline phases exist, involving
many molecular arrangements intermediate between the fully ordered crystal
and fully disordered liquid. These complex arrangements of molecules represent
a kind of ‘supramolecular’ architecture. We are only just beginning to understand
how important this kind of molecular organisation is to Nature. DNA, which
carries the genetic code, will, for example, form a nematic phase.

One of the easiest ways of recognising these structures is to look at
their birefringence patterns under a polarising microscope. This is how
Georges Friedel at the University of Strasbourg first classified liquid
crystals in the 1920s. Different liquid crystal structures produce different
textures. These arise because although the direction of alignment of the
molecules stays the same over short distances, in areas called ‘domains’,
it varies randomly over larger distances. Where domains with different alignments
of directors meet, defects are formed (like those in a magnetic material).
The distribution and type of defects depends on the degree and type of molecular
organisation of a particular liquid crystal phase.

When the liquid crystal is placed between crossed polarisers and viewed
under a polarising microscope, the variation in refractive index splits
up light to produce bright colours while the defects appear as dark lines,
points or brushes separating domains. In the photographs opposite, we show
some beautiful examples of the classic textures that nematic and smectic
phases form.

Liquid crystals are not just attractive oddities, however. Their ability
to organise themselves in a mobile way has given them some useful properties.
A remarkable feature of liquid crystalline materials is that applying an
electric or magnetic field can change the direction of the director. It
will reorient so that it lies parallel to the field or perpendicular to
it, depending on the chemical structure of the constituent molecules. This
means that the refractive index also changes, thus altering the optical
properties of the liquid crystal.

In the early 1930s, physicists suggested that liquid crystals could
be used to produce display devices that would use up much less power than
cathode ray tubes. The first attempts to produce a commercially viable device
failed, however, because the liquid crystalline compounds known then were
not stable when exposed to heat or light.

This situation changed in the 1960s, when George Gray and his research
group at the University of Hull discovered a new family of stable liquid
crystals, the alkylcyanobiphenyls. Gray worked with the Ministry of Defence
and the chemical company BDH to design, synthesise and exploit the materials
for displays. This collaboration is one of the first examples of molecular
engineering of new materials based on a thorough understanding of the structural
principles underlying the behaviour of liquid crystals.

Gray’s liquid crystal molecules behave like stiff rods forming nematic
phases. They are cheap and robust, and so seemed ideal for liquid crystal
display devices. A typical device consists of a thin layer of liquid crystal
material between two glass plates whose surfaces have been coated with a
transparent conducting layer of indium tin oxide etched into the required
patterns. The conducting surfaces are coated with a polyimide-a material
that, when rubbed in a certain direction, induces surface forces which align
liquid crystal molecules. If the direction of rubbing of the top plate is
perpendicular to that of the bottom plate then the liquid crystal is forced
to adopt a 90 degree twist across the cell. For this reason the displays
are called twisted nematic devices.

To form a display, the cell containing the liquid crystal is placed
between crossed polarisers. Light entering the cell is plane polarised,
and the 90 degree twist of the liquid crystal rotates the plane of polarisation
by 90 degrees so that the light passes unhindered through the second polariser.
If, however, you apply a voltage to the conducting areas of the plates an
electrical field is produced. This is sufficient to overcome the effect
of the polyimide surface and align the liquid crystal molecules in these
regions so that they lie parallel with the electrical field. This means
that the plane of polarisation of the light passing through is not changed,
so it does not pass through the second polariser. The region affected by
the electrical field therefore appears black (see Figure 2).

Liquid crystal twisted nematic displays have worked extremely well for
small displays such as digital watches. But so far, electronics companies
have not been able to produce large, high resolution displays suitable for
flat television screens on a commercial scale. This is because you need
a large number of cells or pixels, each with separate electrical connections,
which is difficult to do. Also as the pixels get smaller and more closely
crammed together there is a tendency for the electrical field from one pixel
to affect its neighbours, causing ‘crosstalk’.

However, there is a prospect of new and better displays that work slightly
differently. These displays depend on smectic phases that are also chiral,
and in which, the molecules in each layer are tilted. In these smectics,
the molecules in each layer are tilted with respect to the layers. In 1975,
Robert Meyer and his colleagues at the University of Paris South predicted
and showed that using chiral molecules in such a mesophase would give rise
to chirality in the liquid crystal which, in certain cases, would make the
material ‘ferroelectric’.

As with chiral smectic phases, each layer has its average tilt direction
rotated with respect to the layer below it. You can design the molecules
so that they have a positive electrical charge on one side and a negative
charge on the other-in other words, give them a dipole moment. Electrostatic
forces between the dipoles mean that they all line up within a layer. However,
because each smectic layer is rotated with respect to its neighbour, the
material has no net dipole moment because every layer has a twin at a distance
of half the periodicity of the helix above or below it whose dipole is pointing
in the opposite direction. Such a material is called ferroelectric, or more
precisely helielectric.

In 1984, Noel Clark and Sven Lagerwall from the University of Colorado
in Boulder showed that such phases could be useful for displays. Their device
is similar to the twisted nematic display, but this time the polyimide coating
on the glass plate is rubbed in a direction that completely unwinds the
helix. The dipole moments then no longer cancel each other, resulting in
a net dipole moment between the two glass surfaces of the cell. Applying
an electric field to the conducting layer causes the dipoles to flip, reversing
the average molecular tilt. As with the twisted nematic displays, the etched
patterns in the conducting layers of the cell can be observed when the cell
is placed between crossed polarisers.

This sort of device has a number of attractions over the conventional
twisted nematic display. First, you can switch it on and off much more quickly-in
microseconds rather than milliseconds. Secondly, the molecules are hindered
in reversing their tilt direction when a field is applied, because to reverse
the tilt of one molecule requires that we also move all the nearest neighbours
out of the way, and so on ad infinitum. This hindrance to changing the overall
tilt of the phase means that the device is bistable; in other words, once
the applied voltage has reversed the molecular tilt, it can be turned off
and the overall tilt will remain unaltered. Also, because we require only
a voltage ‘blip’ to flip the average molecular tilt, smaller and more closely
spaced pixels are possible without causing crosstalk. Before flat screen
televisions can be mass-produced, a reliable means of switching large arrays
of ferroelectric pixels will have to be developed. But, having said this,
researchers at GEC’s Hirst Laboratories in London have used this technology
to demonstrate a number of prototype screens of A4 size. Looking further
into the future, the bistability of the ferroelectric devices means the
devices could be used as switches in electro-optic circuits, and in optical
computers.

The development of liquid crystal displays over the past 10 years has
motivated us to learn more about the molecular forces that determine the
kind of liquid crystals rod-like molecules form. Chemists now know enough
to design new liquid crystalline molecules with a range of predetermined
physical characteristics.

More than 70 years ago, however, Vorlander showed that molecules with
shapes other than rods could form liquid crystalline phases as well. He
synthesised liquid crystalline molecules that were T-shaped and U-shaped.
His research group at the University of Halle was so prolific and innovative
that it is only in the past 15 years or so that the rest of the world has
caught up and started to investigate the liquid crystalline behaviour of
more exotic molecular forms. It turns out that our theories and models for
simple cigar-shaped molecules give us an incomplete picture of how mesophases
form for other molecular shapes. Studies of these exotic molecules should
give us a deeper insight into the forces between molecules which give rise
to liquid crystallinity.

Another reason for doing such work is, of course, to look at the technological
potential of these materials. Although we know little about the mesophases
formed by these new materials, we believe that their self-organisation into
complex one-and two-dimensional structures will produce unusual mechanical,
electrical and optical properties.

The first of these new liquid crystals to be studied were the ‘discotics’.
They were predicted and discovered in 1978 by the Indian physicist, S. Chandrasekhar
at the Raman Research Institute in Bangalore. As their name implies, discotics
are disc-shaped molecules. In the liquid crystalline state, they appear
to stack one on top of the other, rather like a pile of plates, and these
stacks then arrange themselves to lie side by side on a two-dimensional
hexagonal lattice. Of course, the phase is liquid-like so these molecular
stacks wiggle and slither around the whole time.

Since 1978, many other research groups have studied the discotics, and
have demonstrated some applications for them. For example, Neville Boden’s
group at the University of Leeds has shown that the molecular stacks of
some discotic molecules can conduct an electric current. Because of their
self-assembling properties, such liquid crystalline phases could be made
into molecular wires-a step towards the development of electronic devices
on a molecular scale.

At the University of Southampton and Imperial College, we have developed
a discotic liquid crystal based on a slight modification of a copper soap.
We would have expected it to form a classic discotic mesophase, but surprisingly
it forms a new structure. In this case, the molecular stacks have arranged
themselves to lie side by side in sheets, with the sheets themselves lying
one on top of each other.

This structural change at the molecular level leads to dramatic changes
in physical properties. Whereas the standard copper soap is difficult to
form into strands, the layered supra-molecular structure of the modified
version makes it easy to pull long molecularly aligned strands, which rapidly
cool to form glassy optical fibres. Since the copper soaps show ‘nonlinear’
optical effects, whereby light passing through the material can be changed
in wavelength, the ability to form optical-quality fibres from them is critically
important to their use in optical systems.

The search for new molecular shapes capable of forming liquid crystals
continues. Researchers have rediscovered Vorlander’s work and are looking
at his T-and U-shaped molecules. Researchers have recently discovered liquid
crystal molecules shaped like stick insects, called phasmids, and molecules
shaped like swallowtails and called hemiphasmids. They are also investigating
hybrid molecular shapes formed by attaching rod-like molecular units to
discotic cores using a flexible link. These exotic structures are bound
to show unusual physical characteristics. Research into liquid crystals
and the technologies they spawn should continue expanding into the next
century.

George Attard is Courtaulds Research Fellow in the chemistry department
at the University of Southampton. Richard Templer is Royal Society Research
Fellow in the chemistry department at Imperial College, London.