Unusually long and aligned ’buckytubes’ grown at Duke

Duke University chemists have developed a method of growing one-atom-thick cylinders of carbon, called “nanotubes,” 100 times longer than usual, while maintaining a soda-straw straightness with controllable orientation. Their achievement solves a major barrier to the nanotubes’ use in ultra-small “nanoelectronic” devices, said the team’s leader.

The researchers have also grown checkerboard-like grids of the tubes which could form the basis of nanoscale electronic devices.

The accomplishment involved sprouting the infinitesimally thin structures, also called “single walled carbon nanotubes,” or “buckytubes,” from tiny catalytic clusters of iron and molybdenum atoms dotted onto a small rectangle of silicon inside a quartz tube.

These growing nanotubes continue to lengthen along the silicon’s surface in the direction of the flow of a feeding gas of carbon monoxide and hydrogen that had been quick-heated to a temperature hot enough to melt normal glass. Atoms from the feeding gas are used as molecular building blocks.

The process was described by Duke assistant chemistry professor Jie Liu, his senior research associate Shaoming Huang and his graduate student Xinyu Cai in an article posted Tuesday, April 22, 2003 in the on-line edition of the Journal of the American Chemical Society (JACS). Their research was funded by NASA, the Army Research Office and Dupont.

“To the best of my knowledge these are the longest individual single-walled carbon nanotubes ever recorded, although we removed that ’longest’ statement from our paper because you can never claim longest forever,” Liu said.

“In our paper, we claimed lengths of more than 2 millimeters, but in our own lab we are now growing 4 millimeter long nanotubes,” he added in an interview. “We may get even longer nanotubes later on.”

Nanotube lengths are normally less than 20 millionths of a meter, their JACS report said — about 100 times shorter than the ones Liu’s team is making. If its girth could somehow be bloated to a 1-inch diameter, then a 2-millimeter-long nanotube’s length would extend proportionally to more than 31 miles, Liu estimated.

After learning they could grow the very long and straight nanotubes, the researchers then discovered they could form cross-connecting nanotube grids as well. They formed the grids by growing additional nanotubes in perpendicular directions under the guidance of a reoriented feeding gas flow.

Such grid patterning could form the basis for billionths-of-a-meter scale electronic circuitry, Liu said. Exceptionally lengthy nanotubes could also be cut up into smaller lengths for splicing into electronic nanoarrays, he added.

Moreover, “such long nanotubes make the evaporation of multiple metal electrodes on a single nanotube a relatively easy task,” the authors wrote in JACS. “Thus, multiple devices can be created on the same nanotube along its length.”

Nanotubes, so named because their smallest dimensions measure just billionths of a meter, were first studied in the 1990s. They are sometimes called buckytubes because their ends, when closed, take the form of soccer ball-shaped carbon molecules known as buckminsterfullerenes, or “buckyballs.” Scientists are avidly studying nanotubes because of the cylindrical molecules’ exceptional lightness and strength as well as their intriguing electronic properties, Liu said.

Depending on their specific architectures, nanotubes of sufficient purity can behave either like semiconductors or like metals and could thus form the circuitry for molecular-scale nanoelectrical components of the future, Liu said.

Since coming to Duke from the Rice University laboratory of Nobel Laureate Richard Smalley, a leading researcher in the field, Liu has made a number of advances toward the goal of mass-producing electronically reliable nanotubes.

Last year, his group reported on the advantages of sprouting the nanotubes from catalytic iron and molybdenum seeds, and using a mixture of gaseous carbon monoxide and hydrogen to supply building materials for their growth.

This combination of advances allowed the Duke chemists to grow groups of nanotubes with diameters that were close to uniform. It also let them sprout tubes at locations of their choice on a surface.

But locational control still wasn’t pinpoint, said Liu. The scientists also needed to learn how to steer the direction of tube growth. An illustration in the JACS article shows such “normally”-prepared nanotubes bending in all directions like straw in a walked-on field.

Liu said that “For future electronics applications there are two major barriers in nanotube related research,” he added. “One of them is control of location and orientation.” The other major obstacle, he said, is tailoring nanotubes to behave consistently as pure metals or pure semiconductors.

Duke’s team has now achieved directional and orientational control by heating samples much more quickly and maintaining a growth temperature of 900 degrees centigrade, it announced in the JACS paper.

“Clearly, the fast-heating is favored for the growth of long and well-aligned nanotubes,” the authors wrote. “We believe that the extremely quick growth at the initial stage is the key factor,” they added.

Another illustration in that article shows long-straight nanotubes in some cases completely crossing the field of the scanning electron-microscope used to view them. Because the tubes themselves were too thin to be easily seen, the scientists traced in white parallel lines the same length as visual aids.

In their article, the authors also acknowledged that other research groups reported controlling nanotube orientation and location on a flat surface by using an electric field. “However, the introduction of a strong electrical field during the growth of nanotubes is not an easy task,” they wrote.

“Furthermore, orienting (nanotube) arrays into multidimensional crossed-network structures in a controllable manner by direct growth was not demonstrated.”