Mention "glass," and a window pane comes to mind. But under certain conditions, a metal can also form as a glass, possessing properties that make it perfect for electric transformers, golf clubs and other products.

Making useful metallic glasses is tricky, but Todd Hufnagel, an assistant professor of materials science and engineering, has taken on the challenge.

Todd Hufnagel has set up lab equipment in Maryland Hall to develop new metallic glass that will remain solid and not crystalize at higher temperatures.

Hufnagel is trying to produce new metallic glasses in bulk form with superior strength and elasticity and magnetic properties. In doing so, he hopes to learn more about the microscopic events that occur when molten metal cools into a solid. This is the critical period when a metallic glass is born.

To scientists, a glass is any material that can be cooled from a liquid to a solid without crystalizing. Most metals do crystalize as they cool, arranging their atoms into a highly regular spatial pattern called a lattice. But if crystalization does not occur, and the atoms settle into a nearly random arrangement, the final form is a metallic glass. Window glass possesses this same random atomic arrangement, although it is not metallic.

Unlike window panes, metallic glasses are not transparent, but their unusual atomic structure gives them distinctive mechanical and magnetic properties. And, unlike window glass, metallic glasses are not brittle. Many traditional metals are relatively easy to "deform," or bend permanently out of shape, because their crystal lattices are riddled with defects. A metallic glass, in contrast, will spring back to its original shape much more readily.

"If you rank materials for how springy they are, metallic glasses are off the chart," says Hufnagel. "They're far and away better than anything else out there."

Because they lack crystal defects, iron-based metallic glasses are very efficient magnetic materials. And like window glass and plastic, metallic glass softens as it is heated, making it easy to mold into a final shape.

In manufacturing, properties like these can have great appeal. But making a metallic glass in thick, bulk form is not easy because most metals rush to crystalize as they cool. To make a glass, the metal must harden before the crystal lattice has a chance to form. To create a glass from a pure metal, such as copper or nickel, one would have to cool it at about 1 trillion degrees Celsius per second, Hufnagel says. That's impractical.

In the 1950s, however, metallurgists learned how to slow the crystalization by mixing certain metals, such nickel and zir-conium. When thin layers of such alloys were cooled at 1 million degrees per second, they formed a metallic glass. Because of the rapid cooling requirement, this material could only be made as a thin ribbon, a wire or a powder.

Materials scientist Todd Hufnagel displays ribbons of metallic glass.

More recently, however, scientists have created about a dozen metallic glasses in bulk form_bars, for example_by mixing four or five elements that possess atoms of varying sizes. That makes it tougher for the mixture to form crystal lattices. One of these new metallic glass alloys is being used commercially to make powerful golf club heads.

Hufnagel has set up a lab at Hopkins to test new alloys. He is trying to create a new metallic glass that will remain solid and not crystalize at higher temperatures, making it useful for engine parts. The new metallic glass may also have military applications as armor-piercing projectiles. Unlike most crystalline metal projectiles, which flatten into a mushroom shape upon impact, Hufnagel believes the sides of a metallic glass head will sheer away on impact, essentially sharpening the point and providing more effective penetration.

His work follows in the footsteps of scientists throughout history who have stirred together ingredients, trying to make valuable new materials. "Metallurgy has a long tradition of being a 'black art,'" says Hufnagel. "For a long time, people did things because they knew they worked, without understanding why. The real contribution of metallurgy is in starting to figure out why things work and how we can make them better"

He adds, "Part of what we're doing is still sort of alchemy, mixing up new combinations to see how well they form a glass. But the other part of this is science. We want to know how the crystalization works, what's going on there. If you can understand how the crystalization happens, then, presumably, you can design your alloy to avoid it. There's a lot of basic research to be done on this stuff."

Hufnagel's research is funded by the National Science Foundation and the U.S. Army Research Office.

Hufnagel's home page is at www.jhu.edu/~matsci/people/faculty/hufnagel/hufnagel.html.