Battery health is a growing issue as battery raw materials get more expensive and trade restrictions obstruct the supply chains. Long battery life will reduce both costs and the environmental impact. Chemists have now invented ‘battery medicines’ to extend the healthy performance curve of batteries. US Argonne National Laboratory is using machine learning to increase the lifespan of batteries.
Argonne researchers have studied how known electrolyte additives will affect battery health and performance. Through machine learning they have been able to predict combinations that prolong the life of the battery and increase the capacity. This could be compared to medicine to cure the diseases of batteries, caused by malfunctions in the chemistry applied.
The researchers trained models to forecast key battery health metrics, like resistance and energy capacity, and applied these models to suggest new additive combinations for testing.
One example is LiNi0.5Mn1.5O4 batteries, known as LNMO, which operate at a high voltage and offer significant advantages to traditional batteries. They have a higher energy capacity and eliminate the need for cobalt, a critical material associated with supply chain concerns, but this battery also presents significant challenges. Cell phone batteries and individual electric vehicle cells typically operate at low voltage, around four volts. But an LNMO battery operating at five volts far exceeds the stability limit of any known electrolyte.
“High voltage usually indicates high energy density,” explained Chen Liao, an Argonne chemist and senior scientist at the University of Chicago. Introducing an electrolyte additive to the LNMO battery could help limit decomposition and improve battery performance. An ideal additive decomposes during the first few battery cycles, forming a stable interface on both electrode interfaces.
“Think of an additive like medicine,” Liao said. “It makes the battery better.”
