Why does a muddy puddle crack as it dries? American chemist Samuel Kistler
could not solve the puzzle, but he was able to overcome it and so produce
a whole range of unusual solids with a unique structure. Sixty years on,
his discovery is just coming into its own. Light enough to float on whisked
egg white, aerogels comprise at least 90 per cent air and are exceptional
in the world of materials science. They are special because their microstructure
is extremely porous, giving them unusual optical, thermal, electrical and
acoustic properties. Now that faster and safer methods of producing aerogels
have been devised, they are finding applications in areas ranging from solar
panels and replacements for CFC-filled insulating foam to subatomic particle
detectors. They are even used in space, to catchmicrometeorites for study.
In the early 1930s, Kistler, working at Stanford University, wanted
to see what would result if he could dry wet silica gels without shrinkage.
These gels are solids comprising a branching skeleton of silica – silicon
dioxide, Si02 – in a liquid. Kistler’s aim was to eliminate surface tensions
that normally arise in the drying process. The effect that he had to overcome
can be seen when mud in a puddle dries up in the sun – cracks develop due
to shrinkage and the fractured surface starts warping. This is caused by
the evaporation of water and the formation of billions of tiny concave menisci
where liquid and solid surfaces meet. Tensions in the menisci cause the
solid mud particles in the puddle to contract, resulting in shrinkage, cracking
and warping. If he could prevent this effect in drying silica gels, Kistler
hoped to be able to substitute the liquid within these solids for air,
without collapsing their delicate silica skeletons.
Silica is the major component of ordinary glass. In silica gels, chains
of tiny silica beads with many branches are immersed in a liquid. If this
is water, the wet gels are known as aquagels; if alcohol, they are called
alcogels. The solid silica typically makes up less than 10 per cent of the
total volume of such gels, so they shrink dramatically and crack if dried
in air. To prevent the build-up of surface tension, Kistler placed alcogels
in an autoclave, a pressure vessel. By increasing the temperature and pressure
above a critical value – 240 degree C and 81 bars for methanol-saturated
gels – the liquid would be transformed to a supercritical fluid. In such
fluids, molecules move about freely. There is no liquid-gas interface, so
the tensions that would normally destroy the structure of the solid on drying
are absent. After slowly draining the supercritical fluid from the bottom
of the vessel and allowing the remaining gel body to cool, Kistler was left
with new, highly porous gels that retained the shape of the originals: he
called them aerogels.
Advertisement
LABORIOUS WORK
Kistler produced many transparent silica aerogel samples with densities
between 30 and 300 kilograms per cubic metre, depending on the silica content
of his starting solutions. He also made other aerogels from alumina, tungsten,
iron and tin oxide. He worked for more than 10 years using the same laborious
and – because of the flammable nature of his alcogels – potentially dangerous
technique, amassing a huge body of knowledge about the unusual properties
of aerogels.
Twenty years later Stanislas Teichner and one of his students at the
Claude Bernard University in Lyons were asked by the French government to
develop porous storage materials for liquid fuels and tried to repeat Kistler’s
pioneering work. But it took weeks to prepare just two aerogel samples;
at this rate, the student would never finish his PhD thesis. So the French
group decided to look for a faster way to make aerogels. Kistler had mixed
water glass (sodium silicate) with hydrochloric acid to produce a gel which
he then saturated with methanol; this was the most time-consuming part of
the process. Teichner’s group mixed tetramethoxysilane Si(OCH3)4
with alcohol, water and a catalyst. In this mixture the -OH groups from
the water replace -OCH3, and methanol is released in the process.
The resulting Si-OH groups then pair up, giving Si-O-Si bonds and water.
Finally, clusters of Si-O-Si aggregate to give the characteristic silica
skeleton, immersed in a mixture of water and methanol.
With this new technique, alcogels could be made within a few hours,
but aerogel production remained difficult because of the delicate drying
procedure. Not until the early 1980s did aerogels find their first useful
application, in the field of high-energy physics to detect fast subatomic
particles. Gunter Poelz at DESY, Germany’s national accelerator laboratory
in Hamburg and Sten Henning at a Swedish company, Airglass, used the French
method to produce hundreds of highly transparent flawless silica aerogel
tiles, each 20 centimetres square by 2 centimetres thick.
At DESY and at CERN (the European Laboratory for Particle Physics) in
Geneva the tiles were used by particle physicists as an alternative to compressed
gases or low-density liquids in so-called Cherenkov detectors, to detect
pions, muons and protons moving at close to the speed of light. Such particles
give off light in a cone-shaped electromagnetic shock front which surrounds
their path and resembles the acoustic shock wave emitted by a supersonic
airplane. Just as a sound shock wave can be detected as it moves through
air by the bang it makes, so a shock wave created by subatomic particles
can be detected by the light it emits when it moves through a compressed
gas, low-density liquid or aerogel. A particle’s velocity can then be deduced
from the angle between the shock front and the particle’s flight path. Aerogels
bridge the gap between compressed gases and low-density liquids in the spectrum
of materials that can be used to detect fast-moving subatomic particles.
With this complete spectrum, these experiments can now be performed with
even greater accuracy.
Cherenkov detectors require only relatively small amounts of the rather
expensive aerogel tiles. If they could be mass-produced cheaply, aerogels
would have enormous potential. Beginning with Kistler, scientists working
with aerogels recognised that their unique structure gives them unusual
thermal properties, making them ideal insulators. Their main advantage over
other insulators is their high thermal resistance and their stability –
they do not melt at temperatures as high as 600 degreeC, and they do not
burn. Following the international oil crisis in the 1970s, a new impetus
to save energy led to research at the German company BASF. Scientists there
developed a way to produce aerogel pellets instead of tiles. Their aim
was to make aerogels suitable for use as insulating spacer material in windows
and in solar panels.
Until now, drying alcogels at high temperature and pressure had been
a necessary but risky procedure. In 1984 a gasket failed in the 3000 litre
autoclave at the laboratory of Airglass, ejecting 1000 litres of methanol
and causing an explosion that destroyed the entire facility.
Scientists at BASF aimed to modify this part of the process to make
the production of aerogels safer, and to devise a cheaper means of producing
the wet gels. Liquid carbon dioxide (C02), which is nonflammable and nontoxic,
seemed an ideal substitute for methanol, and the wet gels would be produced
as pellets rather than tiles. The pellets are made by spraying waterglass
and acid from a mixing jet to give droplets that gel while falling into
a flask. After washing, the water in the aquagel pellets is exchanged for
acetone, then liquid C02. The resulting gels can be dried at around 31 degreeC
and at a pressure of 74 bars.
Aerogel pellets have diameters of between 1 and 8 millimetres and densities
of around 200 kilograms per cubic metre. They are not as perfectly transparent
and scatter more light than the high-quality tiles, but they are a lot simpler
to produce. Most importantly, they are excellent insulators. Heat is transported
in these materials in three ways: via gaseous conduction within the air-filled
pores, via solid conduction along the tiny chains and via infrared radiation.
Air within any porous solid conducts heat. In aerogels the pores measure
between 1 and 100 nanometres, which is similar to the average distance travelled
unimpeded by air molecules at normal pressure and temperature before they
collide with each other. So in aerogels, collisions among air molecules
occur about as often as collisions of air molecules with the pore walls.
As a result, gaseous thermal conductivity in aerogels is only about one-fifth
the level that it is in polystyrene, which is itself a good thermal insulator.
Conduction of heat in the solid skeleton of aerogels is limited because
of its highly branched structure. In a silica aerogel the conductivity in
the silica part is about 200 times less than in vitreous silica, its nonporous
counterpart.
IMPROVING PERFORMANCE
The weak link as far as the insulating capacity of silica aerogels is
concerned is that they transfer too much heat in the form of infrared radiation
at temperatures exceeding 20 degreeC. Below and around this temperature,
radiation is absorbed effectively by molecules in the silica skeleton. This
selective absorption of radiation is not uncommon. My colleagues and I
at the University of Wurzburg looked for a way to reduce the passage of
infrared heat through aerogels even at high temperatures. Experiments adding
small quantities of carbon black, a powerful infrared absorber, to silica
aerogels, decreased optical transparency but greatly improved thermal insulation.
Such aerogels are much better insulators than foams filled with CFCs,
which have been widely used as insulators until quite recently. The Montreal
Protocol has banned CFCs after January 1996, making alternatives essential.
If insulating foams are filled with carbon dioxide or water instead of CFCs,
their insulating capability is drastically reduced. Our aerogels with carbon
black insulate nearly twice as well as such foams. Their insulation capability
can be improved still further if they are evacuated. With a vacuum instead
of air in the pores there is no gaseous conduction at all. Such aerogel
superinsulators are currently being tested. As pumping out the air is a
laborious process these superinsulators seem unlikely to find mass applications
in the near future.
Aerogel pellets are now undergoing trials in windows and transparently
insulated walls. As windows, sandwiched between two layers of glass, they
make excellent insulators – minimising heat loss. Because the granular surfaces
scatter light, they are a good alternative to frosted glass, but can also
be used to improve illumination, for example in libraries and factories.
Known as ‘daylighting’, this works because the aerogel pellets scatter light
passing through them, allowing it to penetrate to areas of a room not normally
illuminated by direct sunlight.
Silica aerogels have also been used in the construction of solar panels
which can dramatically reduce the heating demand of houses in cold but sunny
climates. Like the windows, the panels comprise an aerogel layer between
two panes of glass. This is then mounted on the outside of a building in
front of a thick absorptive wall. When sunlight shines onto the building,
between 50 and 80 per cent of the visible light passes through the aerogel
layer, which is a better thermal insulator than the wall. The transmitted
light is absorbed by the blackened surface on the wall, which heats up.
Most of the heat travels through the wall to warm the house, while a smaller
amount escapes back to the environment through the aerogel layer. In summer,
overheating can be avoided by drawing a blind between the aerogel layer
and the wall, thus blocking off the blackened absorptive surface. This
research into transparent insulators is attracting a lot of interest worldwide.
Aerogels also have special acoustic properties. As my colleagues and
I discovered about eight years ago, sound travels even more slowly in aerogels
than in air: about 100 metres per second. In nonporous vitreous silica the
sound velocity is about 5000 metres per second. Recently, Joachim Grobeta,
at that time one of my PhD students, measured sound velocities of about
20 metres per second in evacuated ultralow density aerogels. Such aerogels
with densities of around 5 kilograms per cubic metre (just four times the
density of air) were first prepared in 1990 by Lawrence Hrubesh at the Lawrence
Livermore National Laboratories in the US, using a two-step gelation technique.
The combination of low sound velocity and low density means that the
product of these two, which is known as acoustic impedance, is also small.
The impedance of aerogels is low enough for them to be used as acoustic
anti-reflection layers – the acoustic equivalent of the optical coatings
on camera lenses. Experiments funded by the German government and industry
are testing aerogels as coatings for piezoceramic ultrasonic transducers.
These measure distances by emitting ultrasonic pulses and then measuring
the time it takes for them to be reflected back off a solid object – rather
like a bat does. They are widely used in robotic systems that measure distances
automatically. Piezoceramics have a very high impedance – they are acoustically
hard – and air has a very low impedance. A coating of aerogel, which has
a value somewhere between the two, helps match the acoustic impedances,
so boosting the strength of ultrasonic pulses emitted by the piezoceramic
by a factor of about 500.
The number of groups involved in aerogel research worldwide has mushroomed,
as have its applications since the mid-1980s. At Sandia National Laboratories
in Albuquerque, New Mexico, lithium borate aerogels made by Jeffrey Brinkerare
seem suitable as wick material for deuterium and tritium mixtures used in
laser fusion experiments. Laser fusion may become an important source of
energy in the next century. At the same laboratories, silica aerogels with
tiny amounts of phosphors added have been combined with radioactive tritium.
The system produces radioluminescence, and could have an application as
a light source that runs without electrical input.
Ultralow density aerogels can even be used in outer space to trap fast
micrometeorites. These particles of interstellar debris, a few microns across,
are found in their billions above the atmosphere but burn up as they fall
to Earth. They easily penetrate the tenuous skeleton of the aerogel and
are gradually decelerated. Once stopped, the meteorites can be inspected
in situ, due to the high transparency of the aerogel target.
More fancifully still, in 1987, the French aerogel group at Languedoc
University, Montpellier successfully made aerogels with pronounced fractal
properties. Fractals are known for their self-similarity: they look similar
when observed with increasing magnification. These aerogels are important
because they allow study of the dynamics and structure of fractal systems.
Experiments with aerogels made from alumina, titania, iron oxide and
other metal oxides show that they have potential in catalytic reactions
– gases can be made to react within them if tiny amounts of an appropriate
catalyst are added to the skeleton. They can also form precursors for ceramics
– when all the air-filled pores are removed and the solid fraction is compacted
down, very pure substances result. With silica aerogels the result is a
nonporous form of glass that could be used to make optical fibres. The work
done by Philips, however, suggests that this application may not be commercially
viable.
A breakthrough in aerogel research came in 1987, when Richard Pekala,
working at the Lawrence Livermore National Laboratories, succeeded in making
the first organic aerogels. He polymerised organic monomers in a formaldehyde
and water solution using sodium carbonate as a catalyst. With melamine which
is a ring molecule with carbon and nitrogen, he produced a clear, transparent
gel. Resorcinol, a carbon ring with side chains, gave a dark red product.
One advantage of organic aerogels is that they are less brittle, and therefore
a lot easier to handle than their inorganic counterparts, which shatter
more readily than the finest glass. Pekala also found that by heating a
resorcinol- formaldehyde aerogel to about 1000 degreeC, a black carbon aerogel
is formed which can conduct electricity. An obvious application for this
material would be in the construction of supercapacitors with high energy
density and extremely high power density. These could be used for energy
storage, for example in electric cars. Pekala’s work has enormous implications.
Polystyrene and polyethylene aerogels may be developed in the near future,
perhaps resulting in cheap and highly efficient thermal insulators.
There is, however, already plenty of scope for the commercialisation
of inorganic aerogels. Several groups are looking at ways to adapt these
to improve their potential in the mass market. One drawback of silica aerogels
is their sensitivity to environmental moisture. Contact with water causes
collapse of the tenuous skeleton of these aerogels, resulting in their destruction.
Recent research has demonstrated that the problem of excessive absorption
of water can be overcome by integrating methyl (CH3) groups into
the skeleton. This turns highly hydrophilic aerogels into hydrophobic ones.
Some are so water-resistant they will even float.
Now several research groups in Europe and the US are developing means
to produce aerogels without using high temperatures and pressures. The idea
is to reduce the forces which result from the surface tension in the pores
of the wet gels and to strengthen the skeleton at the same time. This could
be achieved in three ways: modification of the silica surface to reduce
the interaction between gel and liquid; use of liquids with low surface
tension; and stiffening of the gel, for example by increasing the degree
of connectivity between the chains and by making the chains more rigid.
The aim is to produce aerogels by drying wet gels in near-environmental
conditions. Whether this method will prove faster or cheaper remains to
be seen. But if this can be achieved without too much sacrifice of their
special properties, the future commercialisation of aerogels on a large
scale is assured.
Jochen Frick is Professor of Experimental Physics at the University
of Wurzburg in Germany.