Martin Gruebele got an idea while reading the 2008 New York Times article “The Nature of Glass Remains Anything but Clear.” Gruebele thought he could make the picture a lot clearer by applying technology first developed at Illinois by Beckman Institute colleague Joe Lyding to questions about the fundamental physics of glass. Within a year Gruebele and Lyding were able to do just that and visualize the molecular movements of glass in a way that has never been done before.

Gruebele, a member with Lyding of Beckman’s Nanoelectronics group, has a research focus on the dynamics of complex chemical and biological systems. One of his more notable recent discoveries was the development of a method for inducing protein folding that is much faster and milder than previous methods, allowing for improved computer simulations of this important biological function.

A little over a year ago Gruebele read the Times story about a scientific controversy surrounding the nature of glass or glasses, which are more broadly defined than the common term infers and includes metallic glasses, plastics, and ceramics. It quoted Nobel Laureate and physicist Philip W. Anderson as writing that the “deepest and most interesting unsolved problem in solid state theory is probably the nature of glass and glass transition.” It also quoted Australian chemistry professor Peter Harrowell as saying that glasses “sit right at this really profound sort of puzzle.”

Gruebele had an interest in the field of glasses but had not done research in the area. He was, however, extremely familiar with the physics of dynamic molecular reactions and questions such as the chemistry and energy flow of molecules in complex chemical systems. He is also very familiar with tools like lasers and computational methods for studying molecules and the processes that govern them. When he read the article, Gruebele saw an opening.

“It turns out that there is no real, complete theory of how glasses work, why they melt when they do, how they turn back into liquids again,” Gruebele said. “Part of the reason is that people who have done experiments on glass in the past, the only thing they can ever observe is the time it takes for the glass to change, but not what the microscopic pieces of glass that move around look like.”

The main problem those in the field face in describing glass dynamics is one of taking measurements over time, or as the Times article stated, “a rate of flow so slow that it is too boring or too time-consuming to watch.” Gruebele had just the answer to that problem: scanning tunneling microscopy and a very long form of moviemaking.

“The bottom line is if you want to do engineering on any material, first you have to understand the basic physics,” Gruebele said. “I was thinking about it beforehand but when I read that article – sometimes scientists learn something out of the popular press – I said that’s the thing that’s missing, actually having images of what the smallest pieces of glass are that move around, how big are they, what shape they are. Things that people had not been able to tell from bulk experiments.”

Gruebele believed understanding some fundamental facts about glasses could be found through a scanning tunneling microscope (STM), a type of electron microscope that can provide atomic scale three-dimensional images. Lyding, who built the first STM at Illinois and invented the ultrastable STM, provided assistance as the Gruebele group built an STM in their lab. They then recorded a “movie” of the scanned images of extremely slow, nanometer scale flow rate of a glass below its “melting” temperature.

“When I say movie we’re not talking about 30 frames per second, we’re talking about a frame every few minutes, and the movie lasts for perhaps a hundred hours,” Gruebele said. “We scan an STM frame, and it takes several minutes, and then we go right back and scan the next one. So what happens is these microscopic domains of glass are sitting on the surface and sometimes in two minutes or sometimes in two hours, one goes and moves over.”

What the Gruebele group accomplished was the unique feat of visualizing the movements of glass. They observed a mysterious process in the field of glass dynamics called beta-relaxation and were able to completely separates homogeneous (same at all sites) and heterogeneous (changing even at one site) movements of glass. What they discovered in these movements has important ramifications for glass dynamics and applications.

“The reason it is so important is because that difference is going to determine how you can control the properties of the material,” Gruebele said. “If it is heterogeneous that means you have well-defined properties but in different parts of the surface. If it’s homogeneous that means your properties are always in flux everywhere on the surface. Depending on which of these is the case you’ll have to do completely different things with the glass if you want to change its behavior.”

Current applications besides common usages like window panes include products such as metallic glasses that are often used in everyday electronic devices like cell phones; if scientists understood more about glasses future applications could include more stable glasses for improved drug manufacturing.

Gruebele, a professor in the University of Illinois departments of Chemistry, Physics, and a Center for Biophysics winner of the Raymond and Beverly Sackler International Prize in Biophysics for 2008, said that scientists could have performed these experiments a decade or more ago but didn’t because they were focused on using other experimental tools such as lasers or nuclear magnetic resonance. Scientists who use imaging may have had more aesthetic reasons for not taking an interest.

“Imaging people are more interested in an image that looks regular,” Gruebele said. “If you look at an image of glass it just looks like an Impressionistic painting; in some ways that is not as exciting as if you have a really regular pattern and say ‘ahh that did something.’

“When you look at the pictures of these surfaces, they look like lumpy, bumpy random things. Maybe that is part of the problem; they are not as visually exciting perhaps as something that looks more regular. In this case it looks lumpy and randomly bumpy but there are still patterns emerging out of that.”

And those patterns are being detected in the Gruebele Lab, with a paper expected soon describing their findings.