The next big thing could be tiny indeed

Imagine a wire no more that a millionth the thickness of a penny, so small it can't be made - it has to be grown

Imagine a wire no more that a millionth the thickness of a penny, so small it can't be made - it has to be grown. A team at the Nanochemistry Group laboratory at NUI Dublin has achieved this remarkable feat, producing wires that have the potential to revolutionise miniaturised electronics.

But much will have to be done before the team, led by Dr Donald Fitzmaurice, finally knows what it has discovered. The wires are so small it is unclear how they will perform. They may be subject to quantum effects, Dr Fitzmaurice said. "We would expect to see unusual physical properties. Their physics and chemistry will change."

But there is no doubt the group has achieved something highly significant. Its work is detailed in the current edition of the international journal, Advanced Materials, and, indicating its significance, provides the magazine's cover story.

The Nanochemistry Group exists within the Department of Chemistry and involves 18 researchers. "We want to understand the chemistry of materials on a nano scale," said Dr Fitzmaurice. Nano refers to measurements down to a billionth of a metre or a millionth of a millimetre. They are often so small they are difficult to comprehend.

READ MORE

The group's nanowire discovery was fortuitous. They were trying to achieve a different objective but instead found a way to produce these minute wires. "We were trying to do something else," Dr Fitzmaurice said.

Forming nanowires is a cheap, simple process, he said, something that could be learned by a student in about a day. Actually making the wires only takes about half-an-hour.

The process involves the creation of tiny "nanoparticles" of conductive metal, in this case silver. The egg-shaped particles are anisotropic, that is, their electrical charge is not the same in all directions.

Placed in solution, the free-moving particles will tend to self-organise, achieving the lowest possible energy configuration. They align side by side, touching on their long sides because of attractive forces. But the "egg ends" have a repelling force allowing the rows built by these particles to form close-packed parallel lines that don't "short-circuit".

The solution is evaporated off a carbon substrate. Left behind are rows of nanowires less than seven nanometres in diameter, about 1,000 times smaller than the wires in current production microchips.

"We have chemically controlled the interactions between the particles so that they are stronger in one direction than another, with the result that they assemble themselves into wires with a distinct gap between wires," Dr Fitzmaurice said.

The reaction was not dependent on the use of silver; other conductors could be used. The key seems to be the selection of nanoparticles of a similar size. They are formed and then precipitated out in a way that causes them to form layers on the basis of particle size.

A major part of the nanowire work was carried out by a postdoctoral student, Dr Brian Korgel. He leaves later this summer for the post of Professor of Chemical Engineering at Austin University in Texas.

Analysis is now under way to characterise these new wires and to assess their potential in a new, even smaller generation of microprocessors. They may allow miniaturisation to proceed at an entirely new level. On the other hand they may perform in a unique new way, introducing wholly new methods of producing working circuits and transistor-like devices.

There may be some lower limit of miniaturisation beyond which a material cannot go and still retain its original properties, said Dr Fitzmaurice. A wire is designed to carry a current, but what happens when the wire is so small and the current so minute that interactions at the atomic level begin to influence its performance?

The next step, Dr Fitzmaurice said, will be to develop a method to assemble arrays of metal semiconductor nanoparticles consisting of a repeating unit that possess transistor-like properties. The result would mean literally growing a computer in solution.