Boston University Researcher Finds that Big and Little May Not Be the Same

Contact: Bob Zalisk, 617/353-7628 |

Boston, MA — Flip a switch and the lights go on. Same with a computer: turn it “on,” and electricity flows through its circuits, enabling it to work. At bottom, the work the computer does is simply the end product of millions of ingenious manipulations–manipulations of tiny little on-off switches. The great achievement of the computer industry over the past decades has been its ability to make those switches smaller and smaller and still operate.

The computer wizards did have an advantage. Whatever happens big also happens small. The science is all the same. And also, electricity passing through wires is a pretty basic technology that’s been in use for well over a hundred years.

But what if the science–what is known about what happens when electricity flows through a wire–became different?

According to the research findings of Boston University physicist Pritiraj Mohanty and his co-researcher at the University of Maryland, Richard A. Webb, just that might be the case.

Experimenting with “quasi-one-dimensional gold wires”–wires only 30 or 40 atoms thick–Mohanty found that when an electric current was passed through the exceedingly small wires they did not behave exactly as they should have according to the theory that’s been used to design computer and other circuits for the past quarter-century.

Current computers–which is to say, the chips that do all the actual “work”–operate at a level where objects are measured in microns, millionths of a meter. The circuitry of these chips is designed with the expectation that a single parameter determines how a current will pass through a wire. Whether a material–Mohanty’s gold wires, for example–resists the movement of electrons through it, or, alternatively, facilitates or conducts their movement, depends on only one thing: the mean conductance of that material itself.

This relationship was first articulated in a paper published in 1979 by the Nobel Prize physicist Philip W. Anderson and three colleagues. This work and its further elaboration is known as the scaling theory of localization, and for the past 20 years it has been the basic understanding of how conductors behave. It describes when a material is a conductor or an insulator and how it will behave at different temperatures and lengths or thicknesses.

Mohanty’s research was conducted at a scale where things are measured not in microns, but in nanometers, yet a thousand times smaller. The wires Mohanty and co-researcher Webb monitored were only 20 nanometers thick. The conductance the wires registered showed a small but significant difference from what scaling theory says it should be. The difference was observed only when the wires were tested at a temperature very close to absolute zero: 30 milikelvins.

This result indicates that at this extremely small scale and low temperature something in addition to mean conductance accounts for how a material behaves. That “something,” the researchers believe, is that the material is affected by quantum interactions among the atoms and electrons of which it is made and the electrons that pass through it.

Current theory, which has been so powerful in its utility for so long, does not account for such interactions and their consequent effects. Mohanty and Webb conclude their report with the comment: “The necessity of additional parameters? to describe the conductance possibly signals the breakdown of the one-parameter scaling hypothesis.”

Chip makers continue to race to make their devices ever smaller. They may find, this new research suggests, that when they get down to the long-sought nanoscale, they will need a new yardstick with which to survey that promised land.

Mohanty and Webb’s report, “Anomalous Conductance Distribution in Quasi-One-Dimensional Gold Wires: Possible Violation of the One-Parameter Scaling Hypothesis,” appears in the “Physical Review Letters” of 8 April, 2002.