Want CNET to notify you of price drops and the latest stories?

IBM calculates the force it takes to move atoms

IBM has figured out how much force it takes to move atoms. Next, it will try to build things with those atoms.

Michael Kanellos Staff Writer, CNET News.com
Michael Kanellos is editor at large at CNET News.com, where he covers hardware, research and development, start-ups and the tech industry overseas.
Michael Kanellos
3 min read
Seventeen piconewtons: that's the force required to move a cobalt atom over a copper surface.

It takes 210 piconewtons to move a cobalt atom over a smooth platinum surface, according to a new research paper from IBM's Almaden Research Center and the University of Regensberg.

A piconewton is a trillionth of a newton. A newton is the amount of force required to accelerate a kilogram one meter per second squared. Lifting a penny weighing 3 grams takes about 30 billion piconewtons. The atoms in IBM's experiments are moved with atomic force microscopes. (Andreas Heinrich, lead scientist in the scanning tunneling microscopy lab at IBM Almaden and the lead author of the paper, recently let us move some atoms with a scanning tunneling/atomic force microscope in his lab.)

The breakthrough marks the first time anyone has been able to measure the force required to move individual atoms around, according to IBM, and helps the company move toward its goal of molecular computing.

Photos: Aiming for breakthroughs at IBM Almaden

For more than 40 years, semiconductor companies have boosted the performance of chips, and hence computers, by steadily shrinking the size of transistors, tiny on-off switches embedded in chips. Transistors have been shrunk so much that some transistor substructures are only a few atoms thick.

IBM, along with Intel and several research universities, is dedicating a significant amount of time and energy to take the final leap to learn how to make transistors or even processors and memory devices that consist of strands of molecules or a few atoms. In turn, this could lead to databases capable of holding exabytes of data and computers that could sift through those mountains of data rapidly. (An exabyte is a quintillion bytes, or a billion gigabytes.)

"The problems we're looking at aren't computationally driven, per se, but more information management problems," said Mark Dean, an IBM fellow and director of IBM Almaden in a recent interview. "Computation is not the hard part anymore."

Greg Wallraff and Jennifer Cha at IBM Almaden, for instance, are experimenting with ways to use designer DNA to arrange carbon nanotubes into arrays. Conceivably, this could lead to far smaller, more powerful, and cheaper chips than can be made with semiconducting manufacturing equipment.

Stuart Parkin, meanwhile, is examining ways of storing data by manipulating and controlling the magnetic fields of specific atoms. Parkin's work on the giant magnetoresistive effect over the last few decades led to significant advances in hard drive density.

One of the chief considerations in moving atoms on a substrate is how the atoms interact with what they sit on, according to Heinrich. Ideally, the atoms should bond lightly to the surface. That way, the probe can move them without exerting undue force, and the atoms will stick once they're placed. (The probe is controlled by a scientist with a computer and a mouse. In a few clicks, you can place and shift titanium atoms.)

The probe in the atomic force microscope used in this experiment is mounted on a quartz tuning fork. Changes in the vibration of the tuning fork can then be extrapolated into how much force was exerted in moving an atom.