Engineering news
Researchers at the US Department of Energy have used machine learning tools to accurately predict the physical, chemical and mechanical properties of nanomaterials.
The team created an atomic-level model that accurately predicts the thermal properties of stanene, a 2D material made up of a one-atom-thick sheet of tin.
Mathew Cherukara, one of the lead authors of the study, said: “Predictive modelling is particularly important for newly discovered materials, to learn what they’re good for, how they respond to different stimuli and how to effectively grow the material for commercial applications, all before you invest in costly manufacturing.”
Traditionally, atomic-scale materials models have taken years to develop, and researchers have had to rely largely on their own intuition to identify the parameters on which a model would be built. By using a machine learning approach, the researchers were able to reduce the need for human input while shortening the time to craft an accurate model down to a few months.
Badri Narayanan, another lead author of the study, said: “We input data obtained from experimental or theory-based calculations, and then ask the machine if it can give a model that describes those properties. We can also ask if we can optimise the structure, induce defects or tailor the material to get specific desired properties.”
The machine learning model can capture bond formation and breaking events accurately, yielding more reliable predictions of material properties and enabling researchers to capture chemical reactions accurately and better understand how specific materials can be synthesised, according to the researchers.
Building models using machine learning is not material-dependent, meaning researchers can look at many different classes of materials and apply machine learning to various other elements and their combinations.
The study was published in The Journal of Physical Chemistry Letters.