Skip to content ↓

Bubbles used to simulate atoms' behavior

Professor Subra Suresh holds images of the bubble experiment that has informed his study of the movement of atoms.
Caption:
Professor Subra Suresh holds images of the bubble experiment that has informed his study of the movement of atoms.
Credits:
Photo / Donna Coveney

The performance of microscopic and nanoscale devices can now be better predicted and improved, thanks to MIT work with a substance familiar to any kid: bubbles.

In a communication published June 7 in Nature, the researchers explain how they used a raft of soap bubbles to simulate the behavior of atoms on and near the surface of a material when that surface comes in contact with another surface or object.

The work could be used to gain valuable insights into the mechanisms of atomic scale contact and wear (nanotribology) and defect nucleation at surfaces, which have numerous applications in nanotechnology, including nanoelectromechanical systems, nanoindentation and atomic force microscopy (AFM). Nanoscopic deformation and strength of surfaces are also of paramount importance to the creation of new materials in nanotechnology.

"As devices get smaller and smaller, understanding the phenomena of contact and defect nucleation at surfaces becomes more and more important," said Subra Suresh, the R.P. Simmons Professor and head of the Department of Materials Science and Engineering. His co-authors are Andrew Gouldstone (PhD 2001) and graduate student Krystyn J. Van Vliet.

Scientists can measure various properties associated with how surfaces respond mechanically to being probed with nanoindenters -- objects with tip sizes smaller than one-thousandth of the diameter of the human hair. However, "we cannot see how the atoms move and how defects form. Atoms are simply too small," said Professor Suresh, who also holds an appointment in the Department of Mechanical Engineering.

About a year ago, the three researchers were frustrated by their inability to conclusively explain certain results associated with indenting a few billionths of a meter into the surface of various metals. "Then we thought, 'why don't we create a layer of soap bubbles to simulate the atoms?'" Professor Suresh recalled.

So the team created a raft of bubbles a single layer thick to represent an atomic layer of a material's surface. And since each bubble was nearly a million times larger than an atom, by using a high-speed digital camera the researchers could do real-time monitoring of what happened when they indented the surface from the side.

Although soap bubbles had previously been used to study deformation of bulk metals, this work constitutes the first attempt at the analysis of nanoscale deformation at surfaces.

The team then compared data from the bubble simulation with data from nanoindentation of a real metal. "We found that the two sets of data matched, both qualitatively and quantitatively," Professor Suresh said. Suddenly they had a macroscopic system that could represent behavior on the nanoscale.

With insights gained from the bubble model, they formulated a mechanistic theory for defect nucleation at surfaces during nanoindentation. They were also able to explain with remarkable accuracy the unusually high strength of crystalline surfaces subjected to nanoindentation.

The team has since used the bubble system to explore how atoms move and defects form under a variety of surface conditions. They have experimentally simulated, for the first time, the effects of atomic level surface roughness on defect nucleation at surfaces.

"For the first time, these bubble experiments give insight into the real-time behavior of atoms near the surface of a material during mechanical contact deformation," Professor Suresh said. "Our ultimate goal is to use them to predict how defects will form on the nanolevel, because such defects contribute to the performance of these surfaces and nanoscale devices."

He and his research group have already made considerable progress in developing fully three-dimensional, molecular dynamics computer simulations of nanoindentation in engineering materials, in collaboration with Professor of Nuclear Engineering Sidney Yip and Dr. Ju Li (PhD 2001).

The work was supported by Defense University Research Initiative on NanoTechnology in the area of nanostructured materials, which is funded at MIT by the Office of Naval Research, and by a Department of Defense graduate fellowship to Ms. Van Vliet.

A version of this article appeared in MIT Tech Talk on June 13, 2001.

Related Topics

More MIT News