Two groups of Cornell University researchers have been awarded U.S. defense agency contracts aimed at exploring a new generation of electronics technology at the molecular and nanoscale levels. The goals of the two programs are to investigate the possibility of developing new devices that ultimately could lead to huge increases in data storage and processing speed.

At the molecular level, George Malliaras, assistant professor of materials science and engineering, has been awarded a four-year, $400,000 contract by the Defense Advanced Research Projects Agency (DARPA), the central research and development organization of the U.S. Department of Defense. His research into understanding how to make molecules work like switches is part of a larger multi-disciplinary DARPA program, with the goal of demonstrating the feasibility of building functional molecular electronic devices. Also involved in this larger effort are the Naval Research Laboratory, the Air Force Research Laboratory, the University of North Texas and Scripps Research Institute.

DARPA's hope, according to Malliaras, is to develop molecular electronics devices that would leapfrog current silicon chip technology by increasing the number of transistors on a chip to the hundreds of millions. A Pentium II processor, one of the densest chips in use today, contains 7.5 million transistors.

Malliaras's project, which is Cornell's Þrst entry into molecular electronics research, is to investigate the electrical properties of individual molecules. To do this, he will build test structures that will contain a very small number of molecules between two metal electrodes. The measurements will take place at a probe station integrated with a cryostat and electrical characterization equipment that can measure down to 400 ato-Amperes (aA), an incredibly small measure of current. One aA is about 5 electrons a second flowing through the molecules. The cryostat will keep the molecules at temperatures as low as 4 degrees Kelvin (minus 452 degrees Fahrenheit).From these measurements, says Malliaras, "we can expect to extrapolate to a single molecule." The experiments will be complemented with a theoretical effort, to be undertaken in collaboration with David Dunlap of the University of New Mexico.

In theory, molecular electronics research could lead to circuits in which each element of a system, such as a transistor, diode or conductor, would be replaced by an individual molecule. Such molecular microchips, Malliaras says, could provide greatly improved computing speed and immense storage with a minimum of power demands, leading to such applications as a camera that could store millions of pictures or a watch with the computing power of a desktop PC. However, Malliaras believes that molecular electronics "is a fairly long shot" that could take a decade or more to develop.

At the nano scale (a nanometer is one billionth of a meter, or equal to the diameter of three silicon atoms), Robert Buhrman, professor of applied and engineering physics at Cornell, together with Dan Ralph, associate professor of physics, and colleagues at four other universities have been awarded a contract by the U.S. Army Multidisciplinary University Research Initiative (MURI) to study and exploit the spin property of electrons, which can be investigated through measurements of the magnetic-field dependent transport properties of electronic nanostructures.

The contract, which is for nearly $3 million over three years, will fund the work of Buhrman and Ralph at Cornell, David Awschalom at the University of California-Santa Barbara, Michael Flatte at the University of Iowa, Michael Roukes at the California Institute of Technology, and Ali Yazdani at the University of Illinois at Urbana-Champaign.

At the heart of the study of spin manipulation and spin interactions, says Buhrman, is the future hope of using wave function and spin instead of, or in addition to, electric charge to maintain and process stored information, a technology called quantum manipulation. (Spin, which can be in one of two states, is an intrinsic property of electrons and is part of the angular momentum of the particle.)

The MURI contract grew, in part, out of the demonstration by Buhrman and Ralph of a new way to write information on magnetic material on devices measuring between 10 and 100 nanometers in width. The devices could lead to new computer memory chips with very high data storage capacity. "The basic idea of this research is to develop techniques for the measurement of phenomena in electronic solid state systems that depend on spin, to learn how to control these phenomena and then to figure out ways to manipulate, store and process information using spin-based effects," Buhrman says.

"Magnetics and spin have the possibility of replacing silicon memory with magnetic memory integrated onto a silicon chip," he says. "Beyond that, magnetics and spin phenomena have the possibility of implementing quantum computing, an extremely long range and challenging program."