Write your abstract here.
Hybrid nano-wires provide link to silicon.
18:44 30 June 2004
Nanoscale
electronic
components that could be plugged into conventional computer
circuits have been developed for the first time by US chemists.
Nanoscopic
electronic wires have already been made in the laboratory, often using
exotic materials such as carbon nanotubes. However, to be of practical
use, such components will need to be connected to larger electronic
components. And existing
electronics are usually made with silicon,
which cannot easily be connected to nanotubes.
But
now, Charles Lieber and colleagues at Harvard University in Cambridge,
Massachusetts, have developed highly conductive nanowires by blending
silicon and nickel together. These could provide a way to connect
nanoscale components with existing electronic components, which are
hundreds or thousands of times larger.
A
huge amount of research is being done on nanoscale electronics, because
such miniaturization would enable computer makers to pack in far more
processing power.
High conductivity
Lieber''s
group coated tiny silicon wires measuring 20 nanometres in diameter
with the metal nickel before heating them to 550°C to blend the two
elements together. After etching away excess metal, they then tested
the electronic properties of the resulting nickel silicide wires.
They
were found to have an extremely high conductivity, making them very
promising for use as electronic components. Furthermore, by coating
only part of the wire, the process could be used to make wires that
were part silicon and part nickel-silicon, providing a way to interface
with existing silicon electronics.
"I
find this result quite amazing," says Silvano De Franceschi of Delft
University of Technology in the Netherlands. "It implies that such
heterostructures could be shrunk down to the nanometre scale. This is
relevant for the high-scale integration of electronic circuits."
The
Harvard researchers have also shown that the method can be used to
create simple nanoscale electronic components, e.g. field-effect
transistors. "By extending our approach to crossed nanowires it should
become possible to assemble large and dense arrays ... of transistors
and other devices that could enable hybrid integrated circuits," the
team writes in Nature.
But
De Fanseschi notes that significant obstacles still need to be overcome
before chemically created nanowires can be used as an alternative, or a
complement, to existing technology. "A major challenge is to devise
effective techniques to control their position to a sufficient degree
of accuracy," he says.
Journal reference: Nature (vol 430, p 61)