A new material with the potential to significantly reduce lithium usage in batteries has been discovered through the use of artificial intelligence (AI). In a joint venture between Microsoft and the Pacific Northwest National Laboratory (PNNL), advanced AI and high-performance computing analyzed over 32 million potential inorganic materials in just 80 hours, identifying 18 promising candidates for battery development.
One solid-state electrolyte material, initially considered improbable due to the combination of lithium and sodium ions, was found to be viable by the AI system. The resulting electrolyte reduces required lithium by up to 70%. Microsoft and PNNL have synthesized and prototyped batteries with the material, currently undergoing real-world testing. The discovery showcases AI’s capability to expedite material exploration for various applications, including sustainable energy solutions.