Powerful permanent magnets are at the heart of many modern technologies, from the phones in our pockets to personal computers, medical devices, electric vehicles, power generators and even guided missiles. The magnets used in these technologies are mostly derived from rare earth elements – a set of 17 heavy metals that can be difficult to obtain and expensive to import. China’s grip on rare earth elements has become a political hot potato and sits at the heart of US President Donald Trump’s trade war with the country. The US is reportedly ploughing almost half a billion dollars into a Brazilian rare earths mine, and Greenland is increasingly being viewed as a possible new mining frontier.
Now, researchers at the University of New Hampshire have unveiled a new database of magnetic materials that could potentially reduce reliance on rare earths to begin with. No new permanent magnets have been discovered from the many magnetic compounds known to exist, but that could change thanks to the harnessing of artificial intelligence (AI). AI was used to produce the North East Materials Database (NEMAD), which currently has entries for more than 67,500 magnetic materials. Crucially, these include 25 previously unknown compounds that retain their magnetism even at high temperatures.
AI trawls through existing scientific literature for key data
The AI systems used are able to autonomously sift through and analyse scientific literature to find relevant data, including details such as elemental composition, magnetic ordering and Curie temperatures. In turn, this allows for disparate datasets to be combined into one easily searchable database. The process also goes beyond compiling available data. The AI is able to use predictive modelling techniques that can assess the magnetic properties of a given material, as well as its thermal stability. This is the temperature beyond which the magnetism fails and is a key property for magnets used in different kinds of technology.
Testing what could potentially be millions of combinations of elements individually in the laboratory is not feasible due to costs and time constraints, but the AI process is much faster. “Collecting this kind of information by hand would take an enormous amount of effort,” said lead study author Suman Itani. “Our AI system can do it quickly and automatically organize everything into a single, searchable database.”
Today’s news was brought to you by TD SYNNEX – the UK’s number one solutions distributor.
Destination AI
Destination AI from TD SYNNEX is our comprehensive end-to-end programme designed to help you and your customers get ahead of the AI curve.
