University of Vermont

Office of Undergraduate Research

Strong Gold and Weird Kinks at the Nanoscale

Frederic Sansoz
Like an alchemist's dream come true, engineer Frederic Sansoz has transformed gold, changing it from nature's softest metal into wire far stronger than steel. Sansoz works at the intersection of nanotechnology and materials science -- putting his ideas to the test in real-world nanowires and in simulations at UVM's Vermont Advanced Computing Center. (Photo: Joshua Brown)

One nanometer is about five atoms wide. That’s roughly 100,000 times smaller than the diameter of a human hair. Weird stuff happens at this scale. Or, as the great physicist Richard Feynman presciently noted in 1959, “atoms on a small scale behave like nothing on a large scale.”

Take gold. It’s the softest of all metals, prized since ancient times for both its beauty and malleability. It can be easily bent into rings or pounded into foil. But that can be changed at the nanoscale.

A recent experiment — reported in Nature Communications and led by University of Vermont engineer Frederic Sansoz and Scott Mao at the University of Pittsburgh — demonstrated that gold nanowires could be made to have Herculean strength: ten times stronger than steel.

Nearly perfect

To do this, the scientists created tiny layers called “twin boundaries” in the crystalline structure of the gold. Each of these layers was especially tiny even by the yardstick of nanotechnology: about 0.7 nanometers thick, only a few atoms across.

The microscopic wires were all pure gold, but “just by changing the structure of the material at the atomic scale we increased the strength fifty times,” Sansoz says. “It’s higher than titanium alloys and comparable to Kevlar fibers. It becomes a super-strength material.”

In fact, the gold wires that they tested in both the lab and in computer simulations on the Bluemoon supercomputer at UVM’s Vermont Advanced Computing Center come very close to a state of matter that physicists call “theoretical strength.” This is the strongest possible arrangement of atoms of that element.

“We demonstrated that we attained nearly the maximum strength that Mother Nature could achieve,” Sansoz says.

To strengthen the gold, they relied on one of the basic principles of nanotechnology: that when you make things extremely small they are going to become more perfect.

"Perfect in the sense that their arrangement of atoms in the real world will become more like an idealized model," says UVM’s Sansoz. "With smaller crystals — in for example, gold or copper — it's easier to have fewer defects in them.”

And eliminating the defects at the interface separating two crystals, or grains, has been shown by nanotechnology experts to be a powerful strategy for making materials stronger, more easily molded, and less electrically resistant — or a host of other qualities sought by designers and manufacturers.

Not so perfect

But while Sansoz’s work with gold pushes it toward a kind of perfection of strength, another experiment he and colleagues recently completed demonstrates that one perception of perfection at the nanoscale may not actually be so perfect after all.

Since 2004, when a seminal paper came out in Science, materials scientists have been excited about "coherent twin boundaries" or CTBs. Based on theory and experiment, these CTB's are often described as "perfect," appearing like a dead-flat, one-atom-thick plane in computer models and electron microscope images.

Over the last decade, a body of literature has shown these coherent twin boundaries — found at the nanoscale within the crystalline structure of common metals like gold, silver and copper — are highly effective at making materials much stronger while maintaining their ability to undergo permanent change in shape without breaking and still allowing easy transmission of electrons — an important fact for computer manufacturing and other electronics applications.

But new research by Sansoz and colleagues from the Lawrence Livermore National Laboratory and elsewhere shows that coherent twin boundaries found in copper "are inherently defective."

With a high-resolution electron microscope, using a more powerful technique than has ever been used to examine these boundaries, they found tiny kink-like steps and curvatures in what had previously been observed as perfect.

Even more surprising, these kinks and other defects appear to be the cause of the coherent twin boundary's strength and other desirable qualities.

Kinks and curves

"Everything we have learned on these materials in the past 10 years will have to be revisited with this new information," Sansoz says.

The experiment, reported in the May 19 edition of Nature Materials and led by Morris Wang at the Lawrence Livermore Lab, applied a newly developed mapping technique to study the crystal orientation of CTBs in so-called nanotwinned copper and "boom — it revealed these defects," says Sansoz.

This real-world discovery conformed to earlier intriguing theoretical findings that Sansoz had been making with "atomistic simulations" on a computer. The lab results sent Sansoz back to his computer models where he introduced the newly discovered "kink" defects into his calculations. At UVM’s Vermont Advanced Computing Center, he theoretically confirmed that the kink defects observed by the Livermore team lead to "rather rich deformation processes at the atomic scale," he says, "that did not exist with preconceived-to-be perfect twin boundaries."

With the computer model, "we found a series of completely new mechanisms," he says, for explaining why coherent twin boundaries simultaneously add strength and yet also allow stretching (what scientists call "tensile ductility") — properties that are usually mutually exclusive in conventional materials.

"We had no idea such defects existed," says Sansoz. "So much for the perfect twin boundary. We now call them defective twin boundaries."

Better understanding

For several decades, scientists have looked for ways to shrink the size of individual crystalline grains within metals and other materials. Like a series of dykes or walls within the larger structure, the boundaries between grains can slow internal slip and help resist failure. Generally, the more of these boundaries — the stronger the material.

Originally, scientists believed that coherent twin boundaries in materials were much more reliable and stable than conventional grain boundaries, which are incoherently full of defects. But the new research shows they could both contain similar types of defects despite very different boundary energies.

"Understanding these defective structures is the first step to take full use of these CTBs for strengthening and maintaining the ductility and electrical conductivity of many materials," Morris Wang says. "To understand the behavior and mechanisms of these defects will help our engineering design of these materials for high-strength applications."

For Sansoz, this discovery underlines a deep principle, "There are all manner of defects in nature," he says. "With nanotech, you are trying to control the way they are formed and dispersed in matter, and to understand their impact on properties. The point of this new Nature Materials paper is that some defects make a material stronger."