The artificial intelligence mainstream generates controversy, but it can be a great help in areas such as medicine or industrial developments. It is a great ally when it comes to finding patterns in data sets, such as alloys of new materials.
MIT has developed a new alloy five times harder and just as light as traditional aluminumusing a 3D printer, thanks to artificial intelligence.
An alloy is a mixture of two chemical elements, one of them a metal.. Alloys change the composition of the resulting material. The problem is that there are millions of combinations of chemical elements, and that is without counting the proportion of each product. Logically, each combination cannot be tested one by one. This is where AI lends a hand.
Deep learning to create new materials
In this research, MIT has provided the AI with a database of existing aluminum alloys, and millions of possible combinations with new chemical elements.
Using deep learningthe AI identifies patterns and correlations between the properties of elements, selecting the combinations that are most likely to be successful.
According to the media Techno-Science, This new 3D printed aluminum alloy is five times stronger than traditional aluminumthanks to a fusion of the metal with nanometric precipitates. These precipitates are organized into regular patterns with exceptional strength.
After analyzing millions of combinations, artificial intelligence selected 40 candidate alloys. That is a number that can be tested one by one.
The researchers found an alloy that benefited from 3D printing, what is professionally called laser powder bed fusion technique. This layered melting method allows for rapid cooling, which promotes the formation of fine, dense microstructures.
The alloy proposed by the AI has a placement pattern of the nanometric precipitates that prevents their growth during solidification. This is what allows the new alloy to be five times stronger than conventional aluminum, with the same lightness.
Another good news is that It also retains its stability at elevated temperatures, up to 400 °Cwhich is notable for this type of material.
Using AI deep learning to analyze millions of combinations, and narrow them down to those most likely to be successful, is a technique that is also used in medicine to find new drugs.