Researchers from the Max Plank institute have identified 17 possible new metals with useful properties such as resistance to rust, and extreme temperatures using the power of Machine Learning. Scientists typically run experiments to find such metals to find ways to combine metals to create new ones. They start off with one well-known element, like iron, which is cheap and malleable, and add one or two others to see the effect on the original material. It is a laborious process of trial and error that inevitably yields more failures than useful results.
But, using AI, researchers can far more precisely predict which combinations of metals will show promise. The team from the Max Plank institute was hunting for metals with a low level of “invar,” which refers to how much materials expand or contract when exposed to high or low temperatures. Metals with low invar do not change size under extreme temperatures.
This could be useful in a range of sectors, for example, metals that perform well at lower temperatures could improve spacecraft, or boats to ensure they remain resistant to corrosion and rust.
Hundreds of data points representing the characteristics of current metal alloys were used to train the models. This was used by the AI to forecast the appearance of new metals with low invar. The findings of the measurements were then given back into the machine-learning model when those metals were produced in a lab. The researchers tested the suggested metal combinations, fed the results back into the model, and so on until the 17 potential new metals emerged.