Slashdot Mirror
Killing Friction: Nanotube Springs And Bearings
leb writes: "Physicists at U.C Berkeley have 'peeled the tips off
carbon nanotubes to make seemingly frictionless bearings so small that some 10,000
would stretch across the diameter of a human hair.
The minuscule bearings are actually telescoping nanotubes with
the inner tube spinning about its long axis. When sliding in and
out, however, they act as nanosprings.
Both the springs and bearings, which appear to move with no
wear and tear, could be important components of the microscopic
and eventually nanoscale machines under development around
the world.' Based on the principles of Vander Waal's forces this breakthrough, they state that, 'Our results demonstrate that multiwall carbon nanotubes hold great promise for
nanomechanical or nanoelectromechanical systems (NEMS) applications,' they conclude in
their paper. 'Low-friction, low-wear nanobearings and nanosprings are essential ingredients
in general NEMS technologies.' Read the news release and visit the lab Web page."
This time they got the components to move by attaching the tip of their scanning microscope to the inner bearning and simply pulling it out and pushing it in. I guess if you can logde a few condictive metal atomss inside the inner bearing you can do the same with electromagnetical forces. And then you can have computer controlled frictionless nano motion after working out quite a few complicated details.
Of those to whom much is given, much is required.
It is true that researchers--both pure and applied, "scientists" and "engineers"--sometimes present their discoveries so dryly and with so many qualifications that you can't be sure exactly what the heck they've actually done or what it might conceivably be useful for. In this case what they've done is basic and essential to nanotechnology, and it's important to get past the "so what?" reflex. (For a quick intro on the basics of molecular nanotechnology, take a look at http://www.zyvex.com/nano/)
Consider what they've demonstrated here: nano-scale springs and linear bearings. Presumably these bearings will also work rotationally, but that hasn't been tested yet, so I'll concentrate on the former. If you want to move molecules around one at a time with enough positional accuracy that you can place individual molecules where you want them to chemical react with other specific molecules--and no others--then you must position your molecule to within about 1/10th nanometer of the desired position.
In the face of thermal vibrations--and thermal vibrations dominate at this scale, not quantum mechanical effects--you must have a very stiff positioner to do this. Buckytubes are among the stiffest known structures, so it's very useful that we can make linear bearings and springs from them. If a way can be found to attach the ends of these telescoping buckytubes to spherical joints and, further, to provide a means to actuate them, then it would be possible to combine a minimum of six such telescoping tubes into a nano-scale Stewart platform.
A Stewart platform is a parallel linkage resembling a space frame of connected variable-length struts. The linkages form an octahedron in which one triangular face is the "platform" and the opposing triangular face is the "base." By varying the lengths of each strut, the position and orientation of the platform with respect to the base will vary in six degrees of freedom (translation in X,Y, and Z, and rotation in pitch, roll, and yaw). Stewart platforms are much stiffer than ordinary serial linkages (what we normally think of when we talk about robot arms), and thus they are commonly used for flight motion simulators. (Check here http://www.zyvex.com/nanotech/6dof.html for more details.)
OK, now you may rightly say, "So what?" again. Being able to build Stewart platforms from buckytubes means that we can then make robotic nanosystems conceptually capable of assembling anything else given the proper raw materials, energy, and software. And that means they can assemble copies of themselves, replicating exponentially. Even if it costs a billion dollars to build the first one,it takes just a few generations before the cost of each assembler/replicator drops to the cost of the raw materials, and since carbon atoms are plentiful and cheap, that cost can be very, very low.
To get all the details, consult Drexler's Nanosystems: molecular machinery, manufacturing, and computation, http://www.zyvex.com/nanotech/nanosystems.html