wires turn chips inside out
Technology Research News
Wires a few hundred atoms thick that look
something like a raccoon's striped tail could solve the problem of what
to do when today's computer chip technology hits the wall in the next
decade or two.
Computer chips, lasers and many other modern electronic devices are made
by etching circuits into microscopically thin layers of semiconductor
materials using narrow beams of light. Narrower beams make smaller circuits,
and smaller circuits make faster devices. There is a limit to how narrowly
a light beam can be focused, however.
The electrical properties of the devices, which result from layering different
semiconductor materials together, have much smaller size limitations --
the layers can be as thin as a few atoms and still conduct electricity.
One hope for getting around the light limit is using nanowires grown by
condensing hot vapors of semiconductor atoms. In order to make useful
devices from individual nanowires, however, different types of the microscopic
wires must be organized and connected.
A technique for growing semiconductor nanowires from multiple semiconductor
materials -- developed independently by three different research teams
-- has turned the problem inside out.
The independent research teams from Harvard University, Lund University
in Sweden and the University of California at Berkeley have shown that
is possible to make entire devices out of individual nanowires rather
than using the wires as building blocks. "Now one can think about creating
devices... on single nanowires instead of using crossing nanowires," said
Peidong Yang, an assistant professor of chemistry at the University of
California at Berkeley.
The researchers made the multiple-semiconductor nanowires by starting
to grow nanowires using one semiconductor material, then abruptly switching
to another. Alternating between two or more materials produced segmented
nanowires that have different electrical and optical properties from those
of nanowires made from a single semiconductor material. A specific combination
of materials can result in nanowires that turn heat into electricity,
The researchers also showed that the nanowires can be doped, or chemically
altered, to make traditional electronic components like transistors and
diodes in a single nanowire.
These multiple-semiconductor nanowires have the potential to open up many
opportunities in nanotechnology, said Charles Lieber, a professor of chemistry
at Harvard University. "New materials enable revolutionary versus evolutionary
advances in technology," he said.
The nanowires' shape and small size means they can be used to make much
faster versions of conventional computers. There is also the potential
for "completely new and different" kinds of electronic devices, like "novel
circuit architectures and devices that we have so far only dreamt of,"
said Lars Samuelson, a professor of physics at Lund University in Sweden.
These could include quantum cryptography devices that emit a single photon
at a time, connected quantum dots that trap electrons for quantum computing,
or devices that emit light that can be modulated trillions of times per
second, making for faster optical communications, he said.
Quantum cryptography holds the promise of perfectly secure communications.
Quantum computers, which use atoms to process and store information, are
potentially much faster than conventional computers at cracking codes
and searching large databases.
As important as the nanowires' properties is how they are made. The multiple-semiconductor
nanowires are made millions at a time, making them relatively inexpensive.
The size of the nanowire devices can be closely controlled and can be
made as small as five nanometers in diameter, which is hundreds of times
smaller than a bacterium, and more than a dozen times smaller than today's
smallest transistors. A nanometer is a millionth of a millimeter.
All three versions of the technique use a microscopic droplet of liquid
gold as the catalyst for growing each nanowire. The semiconductor vapor
condenses on one side of the gold droplet and grows into a solid, crystalline
wire one atomic layer at a time and with about the same diameter as the
gold droplet. By removing the vapor of one semiconductor material and
replacing it with the vapor of another, the researchers made single nanowires
that contained layers of different materials.
The Harvard researchers made nanowires 20 nanometers in diameter and about
3,000 nanometers long that have segments of gallium arsenide and gallium
phosphide. The transition zones between the two semiconductor materials
range from 15 to 20 nanometers long. The researchers also used the technique
to change the chemical dopant during the growth of a silicon nanowire
in order to make a diode.
The Lund University researchers made 40-nanometer wide nanowires that
have alternating segments of indium arsenide and indium phosphide. The
segments of indium phosphide ranged from 100 nanometers to 1.5 nanometers
long depending on the growth rate, and the thinnest indium phosphide segments
had atomically perfect boundaries. "The growth rate... in our technique
can be kept on the level of one atomic monolayer per second for optimal
heterostructure control," said Samuelson.
The UC Berkeley researchers made 50- to 300-nanometer wide nanowires that
have alternating segments of silicon and silicon germanium.
The Harvard nanowire devices could be used for nanoscale bar codes, biological
and chemical sensors and polarized LEDs in two years, said Lieber. Using
the nanowires for logic circuits, photonic and electronic waveguide and
lasers is likely to take at least five years, he said.
The Lund University nanowire devices could be used as scanning probe tips
within a few years, said Samuelson. They could be used as single photon
sources in five years, he said.
The UC Berkeley nanowire devices could be used as light sources and in
thermoelectric applications like keeping electronic devices cool and turning
heat into electricity in five to ten years, said Yang.
Lieber's research colleagues were Mark Gudiksen, Lincoln Lauhon, Jianfang
Wang and David Smith of Harvard University. They published their results
in the February 7, 2002 issue of the journal Nature. Their research was
funded by the Air Force Office of Scientific Research (AFOSR), the Defense
Advanced Research Projects Agency (DARPA) and the Office of Naval Research
Samuelson's research colleagues were Mikael Björk, Jonas Ohlsson, Torsten
Sass, Ann Persson, Claes Thelander, Martin Magnusson, Knut Deppert and
Reine Wallenberg of Lund University. They published their results in the
February issue of the journal Nano Letters and the February 11, 2002 issue
of the journal Applied Physics Letters. Their work was funded by the Swedish
Foundation for Strategic Research, the Swedish Research Council, VR and
the European Union.
Yang's research colleagues were Yiying Wu and Rong Fan of the University
of California at Berkeley. They published their results in the February
issue of Nano Letters. Their work was funded by the National Science Foundation
(NSF), U.S. Department of Energy (DoE) and the University of California
Timeline: 2 years; 5 years; 5-10 years
Funding: Government; University
TRN Categories: Materials Science and Engineering; Semiconductors;
Nanotechnology; Integrated Circuits
Story Type: News
Related Elements: Technical paper "Growth of nanowire superlattice
structures for nanoscale photonics and electronics," Nature, February
7, 2002; Technical paper "One-dimensional Steeplechase for Electrons Realized,"
Nano Letters, February, 2002; Technical paper "One-dimensional heterostructures
in semiconductor nanowhiskers," Applied Physics Letters, February 11,
2002; Technical paper "Block-by-Block Growth of Single-Crystalline Si/SiGe
Superlattice Nanowires," Nano Letters, February, 2002
Tiny wires turn
chips inside out
share the load
Nanotubes take tiny
envisions DNA origami
Electric switch flips
Research News Roundup
Research Watch blog
View from the High Ground Q&A
How It Works
News | Blog
Buy an ad link