These incredibly tiny gears aren't real--yet. They exist only as supercomputer simulations of molecule-sized machine parts produced by a team of researchers at the NASA Ames Research Center. But judging from the modellers' results, such nanostructures are certainly plausible and could have enormous potential.

"Hope is growing that products made of thousands of tiny machines that could self-repair or adapt to the environment can ultimately be constructed," says Al Globus, an Ames computer scientist who is co-author of a paper describing the work, which will be published soon in the journal Nanotechnology.

Small and Large Gear

Globus and his colleagues at Ames's Numerical Aerospace Simulation Systems Division are among a growing number of investigators who now believe that atom-scale factories will one day produce new structural materials and advanced computer components, and may even act as tiny repairmen.

The idea of nanotechnology dates back at least to Richard P. Feynman's 1959 talk, "There is Plenty of Room at the Bottom." Although most researchers were deeply skeptical of the feasibility of molecular machines, some key developments seem to be bringing them much closer to reality. Recent technologies such as the scanning tunneling microscope allow scientists to "see" single atoms--and to move them at will.

Nor did the doubters anticipate the discovery of a remarkable class of molecules called fullerenes. These structures consist of 60 carbon atoms arranged in a ball-like lattice, rather like a geodesic dome conceived by inventor Buckminster Fuller (hence the nickname for these molecules, Buckyballs). Fullerenes are among the strongest materials known. And, in 1991, researchers discovered that such interlocked carbon atoms could also be formed into immensely strong, hollow tubes.

The Ames team envisions using these "Buckytubes" as the structural components and machine parts in molecular-scale factories. In the images shown here, the researchers have constructed hypothetical gears by attaching benzene molecules to the outside of the tubes. The bulging benzene molecules form the intermeshing "teeth" of the gear. The individual gears would be about a nanometer (billionth of a meter) across.

Helium Cooling

Gears are no good unless they rotate, of course. So the researchers turned back to their supercomputer and ran more simulations to produce videos of spinning gears. To drive the gears, Globus and his colleagues simulated a laser that could serve as a motor. "The laser creates an electric field around the nanotube. We put a positively charged atom on one side of the nanotube, and a negatively charged atom on the other side. The electric field drags the nanotube around like a shaft turning," says Globus.

Assembling New Materials

The computer simulations predict that the gears would rotate best at about 100 billion turns per second, or six trillion rotations per minute. And these gears are virtually unbreakable. The models predict the gears could fail if they begin slipping but the bonds that form them would not be severed. Simply slowing the rotation or cooling the gear would return it to normal operation.

Creon Levit, another Ames scientist, envisions that these molecular structures could form the basis of a "matter compiler." Given a raw material, such as natural gas and directed by a computer program, the compiler would arrange atoms into macro scale parts or machines. "A matter compiler is not just science fiction," Levit says. "In the biotechnology industry, there are already 'peptide synthesizers' in use. You give them a sequence of amino acids you want produced, and the machine will create those peptides. But you can't make rockets out of peptides."

Another possibility is making tiny machines that form the components of super-tough materials. All they would do is copy themselves, unit by unit, to fabricate the materials in bulk, "just as a living cell can duplicate itself," says Levit. Such assembler-replicators might be used in space to make materials for assembling space stations. They could also make active, or "smart," materials. "There is absolutely no question that active materials can be made," Globus claims. "Look at your skin. It repairs itself. It sweats to cool itself. It stretches as it grows."

There is no question that real nanomachines are probably decades away. But more and more, research is demonstrating that such things are possible--possibly sooner than most of us think.