How To Reach C&ENACS Membership Number
Advanced Options


April 28, 2003
Volume 81, Number 17
CENEAR 81 17 pp. 27-29
ISSN 0009-2347

Modeling and theory are becoming vital to designing and improving nanodevices


CRIMPED A nanocantilever squishes a nanotube, acting as a nanovalve. COURTESY OF SANTIAGO SOLARES
These days, you can't swing a dead cat without hitting something with the word "nano" on it. Typically, the term refers to the extreme tininess of cleverly engineered devices and their promise to miniaturize, speed up, and revolutionize electronics, computers, and other technologies. Indeed, the proliferation of research into nanometer-scale objects promises to make them as important as their counterparts in the macroscopic world.

And just as with larger creations, where the calculations of, say, a mechanical engineer are vital to the stability of a skyscraper, the design of nanodevices benefits from the same predictive aspects performed by theoretical chemists and their computers.

A symposium on computational nanotechnology, held at the American Chemical Society national meeting in New Orleans last month, showcased the importance of computers in the design, improvement, and property prediction of nanodevices. Sponsored by the Division of Computers in Chemistry and organized by California Institute of Technology chemistry professor William A. Goddard III and Mario Blanco, director of Caltech's Materials & Process Simulation Center, the symposium featured dozens of talks, tackling topics from the prediction of nanotube properties to the behavior of nanocoatings and from the design of nanocantilevers to nanoswitches.

Nanoscale systems, though infinitesimal, are made up of thousands, even hundreds of thousands, of atoms. Thus, describing their electronic structures and dynamics requires significant theoretical skill and much computer power. Currently, chemists frequently employ combinations of theoretical strategies, marrying approximations with precise modeling techniques, even as computer power grows.

"Electronics is an area with likely near-term success for applied nanotechnology."
"We anticipate that computational nanotechnology's consumption of computer CPU cycles will soon rival quantum chemistry in the complexity of software and in the usage of massively parallel computer architectures," Goddard noted in his introduction to the symposium. 

NUMEROUS chemical nanoswitches, which flip off and on with voltage or change of solvent, have already been created by chemists such as James R. Heath at Caltech, Charles M. Lieber at Harvard University, and J. Fraser Stoddart at the University of California, Los Angeles.

"Electronics is an area with likely near-term success for applied nanotechnology," Goddard said.

So what do computations have to offer? Caltech graduate student Weiqiao Deng's research, presented by Goddard, has solved a puzzle in the rotaxane switch designed by Heath and Stoddart. The switch consists of a linear molecule that contains tetrathiafulvalene (TTF) and naphthyl groups, separated from each other along the strand. A ring of dibenzo-[24-crown-8] surrounds the molecule, like a wedding band on a finger.

When voltage is applied to the molecule, the ring moves between positions, settling over either the TTF or the naphthyl group. Experimentally, these are distinguishable because one has high resistance (low current), while the other has a resistance 20 times lower. But the experiment didn't indicate which position--ring on TTF or ring on naphthyl--was on or off.

Deng performed calculations of a model of the system, drawing on numerous techniques, including electron tunneling theories developed by Northwestern University chemistry professor Mark A. Ratner.

The answer came from the pictures of the system's highest occupied molecular orbitals (HOMO) and lowest unoccupied molecular orbitals (LUMO).

When the ring is over the naphthyl unit, Deng found that the energy states of the HOMOs and LUMOs are basically degenerate--that is, nearly the same--making it easy for an electron or hole to tunnel across the molecule. But when the ring is over the TTF, the HOMOs and LUMOs are localized and therefore separated, making it less likely that electrons will conduct, Goddard explained.

Therefore, when the ring is over the naphthyl group, the switch is on. When the ring is over the TTF group, the switch is off.

SWITCHES Dinitropyridines act as nanoswitches. At left, the neutral molecule's phenyl rings are at 90º to each other, and the switch is "off." The center and right molecules are anionic and their rings are coplanar, and thus the switch is "on." COURTESY OF PEDRO DEROSA
More than just satisfying a curiosity, the calculations also provide a design principle. Deng looked to improve the switch by modifying some of its components: He attached a cyano group to the naphthyl group, creating an even greater delocalization between the highest HOMO and lowest LUMO, leading to an increase in current. Goddard's group is now collaborating with Heath's and Stoddart's groups in an effort to push the nanoswitch technology envelope even further.

Another model nanosystem, this one involving 75,000 atoms, comes from Caltech graduate student Santiago Solares, who, with Blanco and Goddard, designed a tiny fluid control valve that won a design prize in the Institute for Molecular Manufacturing's 2002 prizes in computational nanotechnology. Solares coupled molecular mechanics with classical chemical engineering principles, because treating every atom quantum mechanically would use prohibitively expensive amounts of computer time and power.

The valve, no more than 55 nm long, consists of a cantilever positioned over a single-walled carbon nanotube. The top portion of the cantilever is functionalized with a monolayer of acrylic acid, which causes the cantilever to bend with change in solvent pH. When the cantilever bends, it crimps the nanotube, restricting the flow of fluid. Such a device could have numerous applications, including drug delivery and ink-jet printing.

Another study shedding light on the property of a potential new nanoswitch design comes from the lab of University of South Carolina electrical engineering professor Jorge M. Seminario. Postdoc Pedro A. Derosa reported new insights into the switching properties of dinitropyridines, an achievement derived from work that the Army Research Office designated a "success story" in 2001.

One particular molecule, 3-nitro-2-(3'-nitro-2'-ethynylpyridine)-5-thiopyridine, has two phenyl rings--each with a nitro group--separated by an acetylene group. Derosa modeled the molecule, which is connected at each end to gold contacts. He found that when the molecule is neutral, its two rings are 90° out of plane with each other. When the molecule is anionic, the two rings are in the same plane.

The orbital-based explanation for the switching behavior derived from this picture is similar to Deng's. "Coplanarity implies a full delocalization of orbitals along the whole molecule," Derosa explained--which makes it easy for electrons to flow.

But when the rings are in the out-of-plane configuration, he said, "the molecular orbitals are localized in a smaller spatial region without making a path connecting both ends, thus yielding a much larger resistance to the electron transfer."

These computations are actually becoming a requirement in the design of molecular electronic devices, Derosa noted. "Not performing them would require trial-and-error experimentation."

Computations have also helped Alexander Hillisch, a chemist at EnTec GmbH in Jena, Germany, design nanoinstruments for measuring DNA deformation. DNA-protein complexes, ubiquitous in biology, form flexible structures that are difficult to study.

Hillisch's method involves using a technique known as fluorescence resonance energy transfer (FRET) to determine distances between donor and acceptor dyes attached to molecules.

To test their method, Hillisch's group synthesized DNA strands with bulges introduced at points along the helices and dye molecules at the ends. They obtained nuclear magnetic resonance structures of the bulge and then created computational models of the positions of the dyes at the ends of the DNA. They were then able to use the FRET data, combined with the other information, to reveal "even subtle changes" in these large molecules, Hillisch noted.

NO FRETTING Detailed picture of DNA molecule with 2 dA5 bulges made possible by local NMR structure of the bulge (inset), combined with global FRET measurements and modeling. COURTESY OF ALEXANDER HILLISCH

NUMEROUS TALKS focused on carbon nanotubes. With their unusual electronic and structural properties, they hold promise for myriad nanodevices. Carter T. White, a chemist with the Naval Research Laboratory in Washington, D.C., discussed the importance of communication between theory and experiment, including the observation of the importance of helical symmetry in computational studies of nanowires. Ronald C. Brown, chemistry professor at Mercyhurst College, Erie, Pa., described energetics of hydrogen as it chemisorbs to carbon nanotubes, while chemist Petros Koumoutsakos, at ICOS, ETH Zentrum, Zurich, described molecular dynamics simulations of carbon nanotube arrays in water.

"We think theory's at the point where we can use to it look at systems, to help design systems. ... Now we hope to leverage both theory and experiment to do them together."
Other highlights included theoretical characterization of carbon nanotubes and fullerenes from Deepak Srivastava, a computational chemist at the National Aeronautics & Space Administration's Ames Research Center in Moffett Field, Calif.

Gustavo E. Scuseria, chemistry professor at Rice University, discussed new efforts, using techniques such as density functional theory and programs such as Gaussian, to model single-walled carbon nanotubes.

Theory hasn't yet explained some of the quantum effects, such as confinement and surface effects of nanoparticles, of group 4 elements, noted Giulia Galli, a theoretical chemist at Lawrence Livermore National Laboratory. She reported results from ab initio--or first principles--simulations that can help expand chemists' understanding of the physical and chemical properties of carbon, silicon, and germanium nanoparticles only 2 to 3 nm in diameter.

"We think theory's at the point where we can use to it look at systems, to help design systems," Goddard said. "In many respects, it's as fast to do theory as it is the experiments. Now we hope to leverage both theory and experiment to do them together."


Chemical & Engineering News
Copyright © 2003 American Chemical Society


Visit SGI

Related Stories
[C&EN, September 30, 2002]

[C&EN, March 17, 2003]

Nanotechnology Archive

E-mail this article to a friend
Print this article
E-mail the editor

Home | Table of Contents | Today's Headlines | Business | Government & Policy | Science & Technology |
About C&EN | How To Reach Us | How to Advertise | Editorial Calendar | Email Webmaster