Michigan University scientists have developed a new machine learning model that can predict the density and stiffness of glasses.
Such information could be used to design better reinforcing fibers for strong and lightweight composite materials in automobiles and wind turbines.
According to team led by Michigan professor Liang Qi, all solid materials, including glass, have a property called elastic stiffness – also known as elastic modulus.
It’s a measure of how much force per unit area is needed to make the material bend or stretch.
If that change is elastic, the material can totally recover its original shape and size once you stop the force.
Elastic stiffness is critical for any materials in structural applications, contends Qi.
Higher stiffness means that you can sustain the same force loading with a thinner material.
For example, the structural glass in car windshields, and in touch screens on smartphones and other screens, can be made thinner and lighter if the glasses are stiffer.
Glass fiber composites are widely used lightweight materials for cars, trucks and wind turbines, and the scientists believe their new method can make these parts even lighter.
The scientists had to first overcome several barriers while designing their light and resilient glasses.
Because glasses are amorphous – or disordered – materials, it’s hard to predict their atomistic structures and the corresponding physical/chemical properties.
Secondly, there isn’t enough data about glass compositions for machine learning to be effective at predicting glass properties for new glass compositions.
In order to remedy this, Qi’s team used existing high-throughput computer simulations to generate data on the densities and elastic stiffnesses of various glasses.
They then developed the machine learning model that is more suitable for a small amount of data.
While training their machine learning model with glasses made of silicon dioxide and one or two other additives, the scientists found that it could accurately predict the lightness and elastic stiffness of more complex glasses, with more than 10 different components.
This implies it can screen as many as 100,000 different compositions at once.
The scientists are now planning to open and share their data so that it becomes easier to understand the strength, toughness, and melting temperatures of different glasses.
Image and content: Qi Group/University of Michigan