Made of semiconducting metal oxides, these extremely thin and flat structures -- dubbed "nanobelts" -- offer significant advantages over the nanowires and carbon nanotubes that have been extensively studied.

The ribbon-like nanobelts are chemically pure, structurally uniform and largely defect-free, with clean surfaces not requiring protection against oxidation. Each is made up of a single crystal with specific surface planes and shape.

Described for the first time in the March 9 issue of the journal Science, nanobelts could provide the kind of uniform structure needed to make practical the mass-production of nanoscale electronic and optoelectronic devices.

"Current research in one-dimensional systems has largely been dominated by carbon nanotubes," said Zhong Lin Wang, professor of Materials Science and Engineering and director of the Center for Nanoscience and Nanotechnology at the Georgia Institute of Technology.

"It is now time to explore other one-dimensional systems that may have important applications for nanoscale functional and smart materials. These nanobelts are the next step in developing structures that may be useful in wider applications," he said.

Wang and his group members Zhengwei Pan and Zurong Dai have produced nanobelts from oxides of zinc, tin, indium, cadmium and gallium. This family of materials was chosen because they are transparent semiconductive oxides, which are the basis for many functional and smart devices being developed today. But Wang believes other semiconducting oxides may also be used to make the unique structures.

"The crystallographic structure varies a great deal from one oxide to another, but they all have a common characteristic as part of a family of materials that have ribbon-like structures with a narrow rectangular cross-section" Wang explained. "In comparison to the cylindrical symmetric nanowires and nanotubes reported in the literature, these are really a distinctive group of materials."

Nanobelts may not have the high structural strength of cylindrical carbon nanotubes, but make up for that with a uniformity that could make them useful in electronic and optoelectronic applications.

Processes for producing carbon nanotubes still cannot be controlled well enough to provide large volumes of high purity, defect-free structures with uniform properties. Nanobelts, however, can be produced with structural control, allowing large quantities to be made in pure form and mostly defect-free.

"Defects in any nanostructures strongly affect their electronic and mechanical properties and possibly cause heating when electrical current passes through them. This creates problems if you want to integrate them into smaller and smaller devices at a high density," Wang noted. "More importantly, defects can destroy quantum mechanical transport properties in nanowire-like structures, resulting in the failure of quantum devices fabricated using them."

Nanowires made of silicon and other materials have also generated interest, but these structures oxidize and require complex cleaning steps and handling in controlled environments. As oxides, nanobelts do not have to be cleaned or handled in special environments and their surfaces are atomically sharp and clean.

Based on known properties of the oxide nanobelts, Wang points to at least three significant applications.

* Zinc oxide and tin oxide nanobelts could be the basis for ultra-small sensors because the conductivity of these materials changes dramatically when gas or liquid molecules attach to their surfaces.

* Tin-doped indium oxide nanobelts provide high electrical conductivity and are optically transparent, making them candidates for use in flat-panel displays.

* And because of their response to infrared emissions, nanobelts of fluoride-doped tin oxide could find application in "smart" windows able to adjust their transmission of light as well as preservation of heat.

"This is a vitally important area of nanotechnology," Wang said. "If we are successful at these applications, it may lead to major technological advances in nano-size sensors and functional devices with low power consumption and high sensitivity."

Wang says production of the nanobelts is simple and should scale up easily for high-volume production at industrial quantities.

Researchers begin by placing commercially available metal oxide powders in the center of an alumina tube. As argon or nitrogen gas is flowed through it, the tube is heated in a furnace to temperatures just below the melting point of the powders, approximately 1,100 - 1,400 degrees Celsius, depending on materials.

The powders evaporate, then form the crystalline nanobelts as they return to solid phase on an alumina plate in a cooler part of the furnace. No purification is needed.

Though the temperature, pressure and processing times must be kept within bounds, Wang says the growth of the nanobelts does not appear sensitive to temperature fluctuations or variations in the processing time.

Finished nanobelts appear as clumps that resemble a wad of cotton. Under microscopic study, they appear like "shredded paper," Wang said. Despite their origin in normally brittle oxide compounds, the nanobelts are flexible and can be bent 180 degrees without breaking.

Typical width of the nanobelts is from 30 to 300 nanometers, with a thickness of 10-15 nanometers. Some have been produced in lengths of up to a few millimeters, though most are tens to hundreds of micrometers long.

Georgia Tech researchers have done preliminary studies of nanobelt properties, though they would still like to learn more about the optical, electrical and surface characteristics.

Wang expects the Science paper on nanobelts will spawn a new area of nanoscience research.

"I believe this area will expand very rapidly. Just like carbon nanotubes, these nanobelts provide a new nanomaterials system that allows people to study nano-scale physics and device fabrication using smart and functional oxide materials" he said.

"Anybody can make these, and there is a lot to measure. There is certainly enough to be discovered to occupy researchers for several years."

The research was sponsored by Georgia Tech, and a provisional patent application has been filed on the new structures. - By John Toon

[Contact: John Toon]

13-Mar-2001