Jan Stange and Steffen Mitzner have devised a way to remove toxins, such as mercaptans and phenols, that bind to proteins in blood. These are normal by-products of metabolism but, says Mitzner, if they accumulate in the bloodstream they contribute to a build-up of fluid in the brain that eventually leads to death.

In the blood of a healthy person, the poisons bind to the protein albumin and are converted in the liver into less toxic or water-soluble compounds that are excreted in bile or by the kidneys. But if the liver is damaged – by alcohol abuse or hepatitis, for example – it can lose its ability to separate the toxins from the albumin.

The conventional way to remove poisons from blood – when a person’s kidneys fail, for example – is dialysis. In this process, blood is pumped past a semipermeable membrane, on the other side of which flows a dialysing solution containing concentrations of salts and sugars similar to those found in normal blood. While the concentrations of these useful substances are thus preserved or restored to normal, poisons with a small enough molecular size are drawn out of the blood, by osmosis.

Unfortunately, protein-bound molecules are too big to pass through the pores of conventional dialysis membranes. But Stange and Mitzner have used a membrane with pores wide enough to allow the toxins to pass through, but too small for the proteins. In addition, they use a dialysing solution of albumin. The toxins detach themselves from the albumin in blood and bind to the albumin in the dialysing solution. This solution can then be recycled, while the blood is passed through a conventional dialysis machine to remove other toxins.

The precise mechanism of how the protein-bound toxins detach themselves from the albumin and transfer across the membrane is not fully understood. “It just works,” says Stange. The technique has been tried on 11 seriously ill patients at the Rostock University Clinic who were not expected to survive without a new liver. Seven survived without a transplant. Stange says that if the technique proves successful in other clinical trials, he expects it will be further tested in patients with chronic liver disease.

The Rostock technique is “interesting from a technological point of view”, says Robin Hughes, a biochemist in the liver unit at King’s College Hospital, London. “But its clinical value is hard to gauge.”

Competition to create artificial livers is fierce and a number of methods are being tested. The ultimate goal is to make the organs out of cultured liver cells, says Hughes. He expects the German technique to be an “interim step”.

But Stange expects the technique to be used in combination with cell cultures, so that the cells receive detoxified blood. To judge by present results, he says, the membrane technique is “at least as good”, as well as being “safer and cheaper” than any of the methods based on cell cultures.

In synthetic polymers, short sections of identical composition can align to
form crystals. In a polymer fibre, these crystals will typically be surrounded
by unaligned sections of chains, which form an amorphous (unstructured)
material. In spider silk, sections of the chain consisting of between four and
ten consecutive molecules of the amino acid alanine line up in parallel to
form the crystalline regions.

Now nuclear magnetic resonance (NMR) spectroscopy has revealed that the
material surrounding these alanine regions is more ordered than in the
amorphous regions of synthetic polymers.

Alexandra Simmons, a member of the Cornell team, explains that the strength
of a synthetic polymer is dictated by the crystalline areas, while its
elasticity depends on the properties of the amorphous material. A better
understanding of silk’s structure in the partially crystalline amorphous areas
could help to explain its remarkable elasticity, she says. A fibre of spider
silk has at least twice the tensile strength of steel, yet it can be stretched
by more than one-third and still recover its original length (Technology, 7
May). As spider silk can absorb more energy before breaking than any metal or
synthetic polymer fibre, it would be ideal for making bulletproof jackets or
ultralight parachutes. Most researchers are concentrating on the dragline silk
produced by golden orb weaver spiders, since it is easy to extract by
“milking” the spiders.

The Cornell team now plans to produce fibres containing the hydrogen
isotope deuterium by feeding the spiders deuterium-rich food. NMR measurements
on the fibres will then give more information about the bond angles and the
molecular structure of both the crystalline and amorphous regions.

But Chris Viney, who is working on the same material at the University of
Washington, says the structure is even more complicated than the Cornell team
believes. “The crystalline regions are less ordered than you would expect,
while the amorphous regions are more ordered,” he says. He attributes this to
the protein chain’s composition of up to 20 different amino acids. By
contrast, a synthetic polymer is made up of only one or two different
molecular building blocks.

Another aim of the Cornell team is to use NMR to find out how the spider
makes the silk. The technique should produce images of the internal anatomy of
the spider, including the gland where the protein is produced and any ducts
leading to the spinnerets.

Alex Hamza and Mehdi Balooch, two engineers at the Lawrence Livermore
National Laboratory in California, have used buckyballs to produce intricately
shaped areas of silicon carbide film on silicon wafers. Silicon carbide is
very useful, as it can be used where silicon is unsuitable. For example,
electronic devices made from silicon carbide can function at up to 600 °C,
while those made from silicon fail above 75 °C. Similarly, mechanical
devices made from silicon carbide can function at higher temperatures than the
600 °C at which silicon softens, and are harder and stronger. Silicon
carbide is also very resistant to corrosion. But silicon carbide devices are
difficult to fabricate, since the material is difficult to grow and etch into
the patterns.

Buckyballs’ almost spherical shape gives them unusual properties. The
best-known version consists of 60 carbon atoms arranged in a football shape.
But so far there has only been one other proposed practical use for
buckyballs: researchers from Argonne National Laboratory in Illinois used them
to grow diamond films (Technology, 30 July).

The new technique relies on the interaction between buckyballs and silicon.
If a stream of buckyballs is fired at a silicon wafer heated to 930 °C,
the carbon bonds break, so that the molecules “open up like tulips”. These are
then highly reactive, and combine with the raw silicon to cover the surface in
patterns of silicon carbide.

More importantly, the buckyballs do not react with silicon dioxide. This
means that the standard lithographic technique of “masking” the circuit
pattern on a wafer of raw silicon can be used to produce a device. After the
bombardment with the carbon molecules, producing the same pattern of silicon
carbide, the oxide can be dissolved with hydrofluoric acid to leave the
silicon carbide microstructure.

The size of the silicon carbide areas is only limited by the lithographical
method itself. “The smallest structures we have made are 1 micrometre in
diameter, and range in thickness from a few hundred angstroms to greater than
a micrometre,” says Hamza.

The applications for this technique are not limited to semiconductors. “One
of the most promising is micro-electromechanical devices,” says Hamza. These
could include pressure sensors in car engines to control fuel injection,
flame-out detectors in aircraft engines, accelerometers in airbags or tiny
engine parts built on microchips.

Tests by scientists at NASA and at a chemicals company in Germany have
found that polyaniline, a polymer with the basic formula
(C₆H₅ąˇąá−)_n, can stop iron and steel
rusting better than conventional methods can. The polymer would be applied to
the surface of shuttle launch pads, which are presently repainted after every
launch to stop rust developing and weakening the structure.

Rust is a hydrated form of iron oxide, and is produced in the presence of
air and water. Both can permeate through a layer of paint, and because one
cubic centimetre of iron produces around three cubic centimetres of rust, it
will eventually burst any protective layer of paint. This exposes more iron to
the air and prolongs the cycle, further undermining the strength of the
structure. Air trapped between flakes of rust can continue to cause corrosion
The steel launch structures at the Kennedy State Center at Cape Canaveral,
Florida are presently protected with a primer consisting of an expoxy base and
about 85 per cent zinc. The zinc oxidises before the steel because it is more
reactive, but once it has all reacted the steel will still rust. In addition,
the zinc cannot protect scratched areas not covered by the epoxy.

Although the coating can protect the steel from the salty, humid air at the
space centre, the fumes produced by the booster rocket on liftoff are highly
acidic, producing hydrochloric acid which eats into the coating, dissolves the
zinc and washes it away. So after each shuttle launch the whole launch pad
must be cleaned to remove any rust and repainted with the zinc primer. This
means treating an area of 9300 square metres, which takes almost two days. On
average there are four launches from each of two launch pads every year,
making rust an expensive enemy.

However, separate work by scientists at NASA with the Los Alamos National
Laboratory, and at Zipperling Kessler in Hamburg, Germany, has shown that a
coating of polyaniline, doped with hydrogen chloride, stops iron rusting. At
Los Alamos, samples of mild steel coated in the polymer and then scratched
survived for around 25 weeks in dilute hydrochloric acid with no signs of rust
formation. In contrast, rust was visible in uncoated samples after around 4
weeks. Independent research in Germany showed that the polyaniline coating
also led to less rust formation than the coating of zinc.

Yet although the teams agree on the polymer’s effectiveness, they disagree
on exactly how it achieves it. According to Holger Melkle, a researcher at
Zipperling Kessler, the presence of the polymer forces the iron or steel in
contact with it to oxidise to a thin, uniform layer of iron oxide. Because
this is unhydrated, it acts as a barrier to the formation of rust. He also
suspects – though he has not yet shown – that the polymer can convert rust
back to its unhydrated form. But while the iron oxide layer has been observed
at the interface with the polymer, Melkle is uncertain how the polymer causes
its formation.

In contrast, Coleman Bryan, chief of failure analysis and physical testing
at Cape Canaveral, claims that the polyaniline protects the metal by
oxidising, just as zinc does, and so loses its conductivity. Using hydrogen
chloride as a dopant, Bryan says, should mean that the shuttle’s booster fumes
help to keep the coating effective. The pads would then only need repainting
every three to four years, he estimates.

The work at Los Alamos and Kennedy is still in the research stage, but
Zipperling Kessler has found a way to process the polyaniline commercially, by
dispersing fine particles of the polymer in a lacquer, which may be applied
with a paintbrush or sprayed on. They expect to begin selling it commercially
in Germany in the next six months.

Traditional welding methods need expensive equipment, such as metal inert
gas (MIG) welders or oxyacetylene torches to melt the alloys. In both cases,
the join may be filled with a mixture of aluminium alloys. Oxyacetylene
torches also require flux materials which are added during welding to prevent
the metal oxidising. But to prevent corrosion all the flux residues must be
removed.

At the start of the new joining method a small amount of a zinc alloy is
melted on the surfaces to be joined. Made by Techno-Weld of Aston, Birmingham,
it contains around 90 per cent zinc, 7 per cent aluminium, and small amounts
of 10 other elements, including magnesium and manganese, and melts at
approximately 380 °C.

The zinc alloy diffuses into the surface to a depth of around an eighth of
a millimetre, changing the composition and reducing the melting point of the
parent alloy.

Two pieces of aluminium treated in this way can be reheated to 380 °C
to melt the surface layer, and then joined together, fusing as the Techno-Weld
solidifies. The manufacturer admits, though, that it is unclear exactly how or
why the joining process works; nor do they know how the joint will behave in
all possible conditions.

In tests on pieces of aluminium/copper alloy joined in this way the shear
strength of the weld was around 90 per cent of the original material, compared
to around 70 per cent for a conventional weld. BOC Gases, which will be
marketing Techno-Weld from November, has also tested the bond and found it to
be as strong, or stronger, than the original aluminium parts.

Early versions of this technique were used to join pieces of aluminium
jewellery and to weld parts of aeroplanes from the US in the Second World War.
These simply used zinc as the joining material, but the bond formed was weak
and highly susceptible to corrosion. The new elements in Techno-Weld seem to
stabilise and strengthen the join.

This technique is highly attractive to the DIY and small repairs market,
where it could be used on aluminium radiators, gearboxes and transmission
casings, and on the bodywork of Land Rovers and Range Rovers. Many car repair
or home improvement enthusiasts can weld steel. But to weld aluminium needs
greater skill and more sophisticated equipment.

The principal saving is in initial outlay. While the materials for
conventional welding cost about the same, a Techno-Weld kit and blowtorch
could cost less than ÂŁ50, against ÂŁ1000 for a MIG welding kit.

A second advantage is that the same formulation of Techno-Weld may be used
to join almost any pair of aluminium alloys, since the join is effectively of
two layers of Techno-Weld.