Machines pave the way for the electric car: this is how they use artificial intelligence to understand fluorine batteries
Published by Serge Grenier in News the
15/06/2022 at 10:59
Fluoride-ion batteries have positioned themselves as one more alternative to lithium-ion batteries, potentially offering more energy density, much longer range and lower cost, although they suffer from the same problem as lithium-ion batteries . semiconductors: they still cannot be mass produced.
But thanks to artificial intelligence , a group of scientists has developed a method to investigate the innards of this technology and be able to exploit it as it deserves to compete with lithium.
Machine learning to find the right materials
As Chemistry World explains , a team of scientists from the University of North Carolina, USA, have developed a machine learning method using supercomputers that can quickly and accurately determine how easily fluoride ions travel in any known fluoride.
This is essential as few materials are known to conduct fluoride ions, a vital requirement as it is the movement of these ions that allows these batteries to charge and discharge.
Initially, the team used a common computational method to sift through a database of 140,000 known materials, narrowing it down to 10,000 fluoride-containing candidates.
"What's really cool is that many materials seem to be better conductors than those used in lithium-ion batteries," says Scott Warren, one of the researchers.
One such material is a fluoride-containing zinc-titanium compound , an "extremely inexpensive material with excellent fluoride conduction properties, and which should hold particular promise as an electrolyte for a fluoride-ion battery," says -he.
Experts point out that machine learning may also be the gateway to solid-state batteries , as it would allow the scientific community to focus efforts first on computer-selected compositions and thus accelerate the process of finding materials that are stable and that they can be synthesized.
Fluoride ions are light, small, and very stable in theory, and fluoride is cheaper than cobalt and lithium.
Brands like Toyota have developed a prototype battery that uses a negatively charged anode, or electrode, made up of fluorine, copper and cobalt, and a positively charged cathode made up of lanthanum with a higher theoretical energy density than ions. of lithium.