In a report posted Oct. 28, 2002, in the online version of the Journal of the American Chemical Society, the Duke group described a method to synthesize starting catalytic "nanocluster" particles of identical size that, in turn, can foster the growth of carbon nanotubes that vary in size far less than those produced previously.
"This is really a first step toward a big future," said Jie Liu, a Duke associate professor of chemistry and the group's leader, of the unprecedented nanotube uniformity they achieved using this process.
Sometimes called "buckytubes," carbon nanotubes' properties were first studied by Japanese researchers in the early 1990s. The nanotubes, measuring just billionths of a meter in diameter (nano means "billionths"), were found to be lightweight but exceptionally strong, with unusual electronic properties.
Depending upon their atomic arrangements, nanotubes can act like conducting metals or like semiconductors, Liu said.
Since microelectronic devices such as computer chips use both semiconductors and metals, researchers foresee nanotubes as the building blocks for even smaller electronic circuitry than the millionths-of-a-meter scale resolutions of today's microchips.
However, "controlling the electronic properties of the nanotubes is becoming the biggest bottleneck that limits the development of nanotube research," Liu said in an interview.
The control problem arises because those electronic properties vary with the way nanotubes' atoms are arranged. And how their atoms are arranged is directly tied to the nanotubes' diameters -- which, until the fabrication advance by Liu and his colleagues, could vary considerably.
In their journal report, Liu's graduate student Lei An, Liu and two University of North Carolina at Chapel Hill researchers describe a technique for growing nanotubes with diameters that varied by about 17 percent.
Using a technique called chemical vapor deposition, An and Liu sprouted the nanotubes from tiny catalyst particles called "nanoclusters." The researchers were able to make each of the nanoclusters completely identical.
"We have shown quite convincingly that by controlling the size of the starting catalyst we can control the diameter of the nanotubes," Liu said. "This is the first time that an identical catalyst has been used.
"The ultimate goal of the research is to produce multiple identical nanotubes using the same kind of catalyst particle," said Liu. "We're still pretty far from there. But it really represents a step forward to show that we have a collection of identical catalyst particles to start with."
The specific nanocluster made in An and Liu's Duke laboratory is one of a large family of catalytic molecules based on molybdenum oxide, he said.
Their nanoclusters contain 30 iron and 84 molybdenum atoms, plus carbon, hydrogen and oxygen atoms. While such clusters are not available from chemical supply houses, they are quite easy to make, Liu said. "And because it's so easy to make these clusters, it should also be easy to scale up to make large amounts of catalyst and large amounts of nanotubes," he said.
The researchers credited the use of a growth-regulating chemical called 3-aminopropyltriethoxysilane (APTES) for achieving more-uniform nanotubes diameters. The APTES kept the nanocluster particles confined to separate islands of discrete size as the nanotubes budded from a silicon dioxide surface.
If researchers can precisely control the nanotubes' diameters, said Liu, the researchers hope in the near future to make pure semiconducting and pure metallic nanotubes. "All the samples we are able to make now are a mix of metallic and semiconducting tubes," he said.
Carbon nanotubes are sometimes called buckytubes because of their structural similarities to carbon-based polyhedral molecules called buckminsterfullerenes, or "buckyballs." Pioneering work with buckyballs won a Nobel Prize for Richard Smalley's research group at Rice University, where Liu did postdoctoral work before coming to Duke.
The problems controlling nanotubes' electronic properties were recently noted in a news feature in the Oct. 10, 2002, issue of the journal Nature. "These difficulties may not be insurmountable," that article said, "but they have persuaded some scientists to turn their attention elsewehere."
In 2001, IBM researchers announced a "constructive destruction" method for separating semiconducting from metallic nanotubes by destroying the metallic ones with bursts of electricity.
An IBM news release said that other researchers have found semiconducting carbon nanotubes should be able to perform as well as silicon when configured into transistors. But nanotubes' molecular-scale sizes could result in computers that are smaller and operate faster using less power than today's silicon-based technology.