Leveraging AI to minimize rare earth consumption: Tech giant Microsoft is partnering with a team of scientists from US government-backed Pacific Northwest National Laboratory on a study aiming to leverage machine learning to identify and develop alternative materials to lithium metals used in EV battery production, according to a statement.

How will they do it? Tapping into high performance computing — a type of cloud-based computing that combines large numbers of computers to solve complex scientific and mathematical tasks — Microsoft sifted through 32 mn potential replacements to lithium, watering them to 18 promising candidates that could be used in battery development in just 80 hours, the company notes.

What they know so far: Although more assessments are needed, a newly discovered material by the AI system combining sodium-lithium is one of the most viable elements for EV mass production beyond lithium, with potential to push down lithium usage in the EV batteries by as much as 70% from current rates.

Not the first sodium EV battery on the block: Interest in sodium battery cells is gaining traction given the scarcity of lithium in parallel to its projected increase in its demand by up to 10x by 2030 from the current 720k ton annual rate. Swedish electric battery maker Northvolt said back in November it developed a new sodium-ion battery which should contribute to the expansion of cost-efficient and sustainable energy storage systems globally by offering an energy density of over 60 watt-hours per kg.

Tags: