Skip to Content

Nanotech’s Super Models

Simulation teams unravel the mysteries of the mesoscale.
November 8, 2001

Simulating the growth of a film of silver crystals, atom by atom, on a powerful supercomputer, James Evans got a big surprise.

Conventional wisdom held that as the growing temperature went down, the film’s surface would become rougher, with crystals jutting off in uneven bumps. Although calculations by the Iowa State University professor and his colleagues confirmed that theory down to 200 °K, below that point the simulation began to get…smoother.

The researchers trusted their model and in 1999 published their findings in Surface Science. A year later, experiments using real silver proved them right: atomic-scale images acquired with a scanning tunneling microscope clearly showed smoother surfaces below 200 °K. “That was a nice example of a success story where modeling preceded the experiments,” Evans says.

Simulation success stories like Evans’s are becoming more common. Last year, the National Science Foundation selected his project and six others for three-year investigations of nanoscale modeling and simulation. Their goal: to construct a more detailed and accurate picture of how atomic-scale interactions shape big-world phenomena. Along the way, they just might revolutionize the design of new materials.

“We want a stronghold in this field,” says Mihail Roco, NSF senior advisor for nanotechnology, “because modeling will be key in understanding how to scale up and develop manufacturing methods.” In addition to Evans’s work with thin films-an area important to semiconductor makers, among others-researchers are modeling microfluidics, polymers, silicon and molecular electronics, membrane transport and quantum computing.

Composite Picture

At the University of Pittsburgh, chemical engineering professor Anna Balazs concentrates on the so-called mesoscale, where atomic phenomena begin to influence bulk properties.

Balazs uses complex calculations to predict the properties of nanoparticle composites-combining, for example, a material with certain optical properties with one chosen for its strength.

Her research promises to pay off for industry with a new breed of strong materials. In 2000, she published a paper in Science predicting a novel way to strengthen nanocomposites using clay.

Clay, she explains, arranges itself in closely packed sheets. “The clay particle looks like a book,” she says. “To strengthen the [composite], you need to explode the binding of the book, so the ‘pages’ are ripped out, no longer a stack of sheets, but dispersed throughout the material.”

Last summer, scientists at Dow Chemical published a second Science paper that supported Balazs’s predictions with experimental results. Balazs and the company have applied jointly to patent their discovery, she says.

Balazs’s collaborators are trying to fill in other parts of the picture. When Balazs generates a mesoscale model, she gives her results to Craig Carter, a materials engineering professor at the Massachusetts Institute of Technology, who runs his own calculations to determine the material’s mechanical properties. Ultimately, Balazs says, they hope to include data at the level of individual atoms, to produce a model that scales seamlessly from the atomic to the macroscale.

Quantum Simulator

Another group of MIT researchers is tackling the sticky problem of quantum simulation. Their goal-one of the most visionary funded by the NSF simulation program-is to throw out the supercomputers on which most simulations rely and replace them with the world’s first quantum computers.

At the atomic level, explains MIT professor Seth Lloyd, Newtonian physics breaks down and the rules of quantum mechanics kick in. The result is such seeming impossibilities as an electron that spins in opposite directions, simultaneously.

That makes modeling the behavior of groups of atoms extremely difficult, since a model must account for every possible combination of states. A model of even 100 atoms is beyond the combined power of every computer on earth, because a single picture of the atoms’ quantum state would require 2100 bits. A picture of 300 atoms would require 2300 bits-as many bits as there are particles in the universe.

“That’s what’s known in the technical jargon as a problem,” says Lloyd.

To get around this problem, Lloyd and MIT nuclear engineering professor David Cory decided to fight fire with fire-or in their case, to count quanta with quanta. In 1996, Lloyd, the theorist, proposed an “analog quantum computer.” Cory, the experimentalist, went on to build one.

Using nuclear magnetic resonance-the same technology hospitals use to scan patients-the duo found they could not only observe, but manipulate the quantum state of atoms within a calcium fluoride crystal. By controlling the crystal’s quantum state with magnetic pulses, they learned to mimic certain other types of quantum behavior. Their analog obeyed the rules of quantum physics that so overwhelm traditional computers.

Lloyd and Cory now are investigating “quantum chaos”-the sensitivity of quantum phenomena to tiny changes in their system, a problem first proposed by Einstein. “We want to use these quantum computers to create quantum chaos in the laboratory and see how it propagates up to larger scales,” he says. Like the projects of Evans and Balazs, this research could eventually improve the design of new materials.

Sim City

Researchers at the other NSF-funded projects are asking similar questions about different phenomena. At Northwestern and Purdue, researchers are simulating molecular circuitry, working toward smaller and faster chips. A Stanford team investigates patterned magnetic nanostructures, helping the quest for denser data storage.

Other projects aim to further biotech research. At the University of Wisconsin, Madison, researchers are using nanoscale simulation to predict the flow of molecules through microfluidic devices. A joint project at the University of Delaware and Pennsylvania State is modeling artificial membranes for use in gas and biomolecule separation, fuel cells and other applications.

“We are seeing significant progress moving from very idealistic solutions to more realistic ones,” says the NSF’s Roco. “Before the national nanotechnology initiative started, the question was, is this real or is this science fiction? Now the industry question is, who will be the leader?

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.