New model facilitates the prediction of crystallization in glass particles

A study by researchers affiliated with a research center supported by FAPESP can help scientists develop new glass and glass-ceramic materials and predict glass stability during use.

By Elton Alisson in São Carlos (Brazil)  |  Agência FAPESP – A new theoretical model that affords a better description of glass particle crystallization and can be used to predict its speed has been developed by scientists affiliated with the Center for Research, Technology and Education in Vitreous Materials (CeRTEV), hosted by the Federal University of São Carlos (UFSCar) in São Paulo State, Brazil.

Published in Acta Materialia, the study can be used in the development of novel glass-ceramic materials and the determination of the stability of glass during use. This would help, for example, to avoid the uncontrolled crystallization and shattering of a kitchen oven’s glass door while heating up.

Edgar Dutra Zanotto, a professor at UFSCar and Director of CeRTEV, explained that crystallization is a critical factor in the production of glass and glass-ceramics.

“It’s important to understand crystallization in glass production so that it can be avoided and the material can become rigid without crystallizing,” he said. “On the other hand, crystallization can be induced and controlled in special glasses by means of heat treatment. This is the case when producing glass-ceramic materials, which are partially crystallized.”

Controlled crystallization of glass-ceramics is used to produce microstructures with certain properties, such as high resistance, for use in various applications, including substrates for PC hard disks, cooktops, telescope mirrors and artificial teeth.

The study has won a Spriggs Phase Equilibria Award from the American Ceramic Society (ACerS) for the most valuable contribution to phase stability relationships in ceramic-based systems literature during the year. Phase stability is a key phenomenon in materials science and engineering. Zanotto will receive the award in September at the 121st Annual Meeting of AcerS in Portland, Oregon (USA).

Raphael Reis, who has a PhD in materials engineering from UFSCar and is currently a professor at Universidade Federal Fluminense (UFF), also participated in the development of the model.

“Our academic research group at CeRTEV is one of the five largest in the world in glass. Nucleation and crystallization are among the topics we’ve studied,” Zanotto said in a presentation to the first symposium on research and innovation in functional materials held by the Center for Development of Functional Materials (CDMF) on May 23-24, 2019, at UFSCar.

CDMF and CeRTEV are Research, Innovation and Dissemination Centers (RIDCs) funded by FAPESP.

According to Zanotto, crystallization begins with the nucleation of crystals on the surface or in the interior of glass particles when the material is heated to a high temperature for a certain time.

The first theoretical models of this phenomenon were developed 60 years ago. One of these models proposed that cube-shaped crystals grow at regular spatial intervals on the surface of glass particles and that crystals can grow only toward the particle center once a totally crystalline surface is formed.

While these models remain important, they are mathematically complex and require solutions based on computational methods that make them hard to use, Zanotto explained.

To overcome these barriers, the researchers developed a new model that describes crystallization in nonisothermal conditions (i.e., with temperature fluctuations) from nucleation sites on the surface or in the interior of glass particles.

“Our model is based on a very simple equation, greatly facilitating its use and the interpretation of the results. It very accurately describes crystallization in nonisothermal conditions,” Zanotto said.

The researchers tested the new crystallization model to evaluate crystallization from a finite number of nuclei on the surface of diopside (calcium magnesium silicate) glass particles and simultaneously on the surface of and inside lithium disilicate glass particles.

They used a technique called differential scanning calorimetry, which identifies glass transition and crystallization temperatures. They compared the results of the new model with those of the older more complex models based on three parameters: the number of surface nuclei, the number of internal nuclei, and nondimensional time.

The comparisons showed that although its analytical equation is much simpler, the new model produced similar results to those of the complex models.

Moreover, according to the researchers, it produces a more realistic approximation for particles that sinter (compact into a solid mass by heat without melting) during phase transformation, such as glass particles that crystallize simultaneously during sintering.

In addition, the new model can be used to predict transformations starting from grain boundaries in polycrystalline materials such as glass-ceramics.

“Theoretical models like the one we’ve developed are important because they help us understand phenomena and make predictions without the need to perform experiments,” Zanotto said.

The article in Acta Materialia entitled “Simple model for particle phase transformation kinetics” (DOI: 10.1016/j.actamat.2018.05.039) by Raphael M. C. V. Reis and Edgar D. Zanotto can be retrieved from: www.sciencedirect.com/science/article/abs/pii/S1359645418303999.