Stanford software brings precision and practicality to nanotechnology

May 24, 2005

Although mechanical engineering Assistant Professor Kyeongjae Cho works with theory and software simulations, it is in his Multiscale Simulation Laboratory where nanoscience and nanotechnology get real. That's because Cho's emphasis is not on speculation about fanciful applications in the distant future but on building a framework for productive research today.

"There's so much hype about nanotechnology, but certainly we don't have enough control yet to make many practical applications," Cho says. Nanoscience and nanotechnology refer to manipulating matter on the nanometer, or billionth of a meter, scale of atoms and molecules. At that size, materials have different properties that researchers and entrepreneurs are still only beginning to understand, much less exploit.

Before nanotechnology can become truly practical, scientists must close a big gap between what they have accomplished—mostly demonstrations of moving around atoms or making proof-of-concept carbon nanotubes, silicon nanowires and nanoparticles—and what they want to do: make exactly the component with the properties an industrial application requires. It is one thing to make a silicon nanowire that can emit light. It is another to make a silicon nanowire that has exactly the right structure for emitting a particular color of light for, say, a photonics application. This is the difference between crafting "things" and machining "parts."

The groundwork is the ground rules

Engineers cannot efficiently achieve this kind of exactness through experimental trial and error. They need to predict whether something they might make will have the properties they are looking for. This is what Cho's software is designed to do. Tell it structural attributes of a particle, wire or tube, and the program will predict with a fine degree of accuracy important properties, such as its electrical attributes.

"The kind of thing we are doing doesn't exist [elsewhere] at all," says Cho, whose research has been funded mostly by the National Science Foundation and the U.S. Department of Energy.

To enable simulations, Cho and his students observe how the "first principles" of quantum physics and physical chemistry that govern atoms and molecules scale up to nanocomponents, such as particles, wires and tubes. From these principles and experimental validation of them, Cho's research group is discerning ground rules for nanotechnology that govern the relationship between a nanocomponent's structure and its properties. By encapsulating these rules in software algorithms, Cho and his students give engineers a tool for "rational design" of nanocomponents—an alternative to intuition, guesswork and tinkering.

As an example of the industrial value of simulations, Cho points to an early version of the software developed under his guidance by doctoral student Byeongchan Lee. The work led to the founding in 2003 of Nanostellar, a company striving to develop more cost-effective catalytic converters for cars. Traditional converters use a lot of expensive platinum to turn carbon monoxide into carbon dioxide. The race is on to develop devices that use nanoparticles to catalyze the reaction with cheaper materials. Competing companies have been relying on arduous trial-and-error methods to do this, but Cho says Nanostellar has used the simulation software to find candidate materials much faster than some competitors.

The National Nanotechnology Advisory Panel, in a major report on American nanotechnology policy earlier this month, singled out Nanostellar and Stanford's progress: "Nanostellar has dramatically reduced the amount of platinum required for automotive emission control," the report says.

Applications move within reach

Cho's research emphasis on codifying fundamental nanocomponent structure-property relationships in software does not preclude him from working on specific applications. A few, such as controlling auto emissions, are coming within reach.

In February, for example, Applied Physics Letters published research of immediate interest to the semiconductor industry. Cho co-authored the research with Yoshio Nishi, an electrical engineering professor and director of the Stanford Nanofabrication Facility; Seongjun Park, a research associate; and Luigi Colombo, a scientist at Texas Instruments. The paper describes the proper structure for a new kind of metal electrode to accompany novel insulating materials in transistors on computer chips. Such an advance is necessary for transistors to continue to shrink. Other Stanford researchers also are focused on using nanotechnology to extend the life of silicon microchip technology. In January 2004, a group of Stanford and Berkeley researchers, including chemistry Associate Professor Hongjie Dai, announced the first working silicon circuit to incorporate nanotubes.

Meanwhile, last spring Cho signed on with Dai; Bruce Clemens, professor of materials science and engineering; and Anders Nilsson, associate professor at the Stanford Synchrotron Radiation Laboratory, to study whether nanotubes or other nanostructures with a lot of surface area are useful for storing hydrogen. Safe and efficient hydrogen storage is necessary to realize popular predictions of a "hydrogen economy," in which electronics, vehicles and buildings are powered by hydrogen fuel cells.

Cho also has published a few other papers this year offering insights into the distinct properties of uniquely structured nanotubes. With each effort he adds to a solid platform of fundamentals that will make nanoscale engineering easier to achieve.

"Once you discover the fundamental scientific principles, then you can figure out how to bring them into the practical engineering domain and industrial applications," he says.

This summer he plans to make nanotechnology simulation software developed in his lab freely available for academic use through http://www.nanohub.org , a site funded by the National Science Foundation.

Source: Stanford University