A new approach predicts the ever-changing behavior of glass at different temperatures

Everything about the glass is not clear. The way in which its atoms are arranged and behave, in particular, is surprisingly opaque.

The problem is that glass is an amorphous solid, a class of materials that lies in the mysterious realm between solid and liquid. Glassy materials also include commonly used polymers or plastics. Even though it may seem stable and static, glass atoms are constantly moving in a desperately futile quest for balance. This sneaky behavior made the physics of glass almost impossible to pin down.

At present, a multi-institutional team including Northwestern University, North Dakota State University and NIST (National Institute of Standards and Technology) has developed an algorithm to further clarify polymer lenses. The algorithm allows researchers to create coarse-grained models to design materials with dynamic properties and predict their evolving behaviors. Called "energy renormalization algorithm", it is the first to accurately predict the mechanical behavior of glass at different temperatures and could allow the rapid discovery of new materials, designed with optimal properties.

"The current material discovery process can take decades," said Sinan Keten, from the Northwest, who co-led the research. "Our approach is multiplying by a thousand molecular simulations, which allows us to design materials faster and examine their behavior."

"Although vitreous materials are all around us, scientists still struggle to understand their properties, such as their fluidity and diffusion as a function of temperature or composition," said researcher Jack F. Douglas. NIST, who led the work with Keten. "This lack of understanding is a serious limitation in the rational design of new materials."

The study recently published in the journal Progress of science. Wenjie Xia, an assistant professor of civil and environmental engineering at North Dakota State University, was the first author.

The strange behavior of Glass comes from the way it is made. This starts with a hot bath of molten material which is then rapidly cooled. Although the final material wants to reach equilibrium in the cooled state, it is very sensitive to temperature changes. If the material is heated, its mechanical properties can change considerably. It is therefore difficult for researchers to effectively predict mechanical properties using existing molecular simulation techniques.

"As simple as the appearance of glass, it's a very strange material," said Keten, associate professor of mechanical engineering and civil and environmental engineering at the McCormick School of Engineering's Northwestern. "It is amorphous and has no equilibrium structure, so it is constantly evolving because of the slow movements of its molecules, and then its evolution varies a lot depending on the temperature and the molecular characteristics of each molecule. vitreous material, a very long calculation time in molecular simulations Accelerating computations is only possible if we can map the positions of molecules into simpler structural models. "

The structure of the glass strongly contrasts with a crystalline solid in which the atoms are arranged in an orderly, predictable and symmetrical manner. "It's easy to map atoms in crystalline materials because they have a repetitive structure," Keten explained. "While in amorphous material, it is difficult to map the structure because of lack of long-term order."

"Due to the amorphous and disordered nature of the glass, its properties could vary considerably with temperature, which would greatly complicate the prediction of its physical behavior," added Xia. "Now we have found a new way to solve this problem."

To meet this challenge, Keten, Douglas, Xia and their collaborators designed their algorithm to take into account the many ways that glass molecules might or might not move, depending on the variation in temperature over time. Calculating the position of each atom in the glass would be laboriously slow and tedious, even for a very powerful algorithm, to calculate. Keten and his collaborators have therefore used a "coarse grained modeling", a simplified approach that considers clusters of atoms rather than simple atoms. Their new methodology effectively creates parameters for the interactions between these coarser particles so that the model can capture the dramatic slowing of molecular motion as the vitreous material cools.

"We can not do atom-by-atom simulation, even for nano-sized glass films, because even that would be too big," said Keten. "That's still millions of molecules, and coarse-grained models allow us to look at larger systems comparable to laboratory experiments."

Until now, Keten and his team have checked their algorithm against three types of polymeric glass-forming liquids, already well characterized and very different. In each case, the algorithm accurately predicts the known dynamic properties over a wide temperature range.

"Explaining the physics of glasses has been one of the biggest problems that scientists have not been able to solve," Keten said. "We are getting closer to understanding their behavior and solving the mystery."