autoevolution

The Missing Link in Efficient Fluoride-Ion Batteries Could Be Found Using Machine Learning

Li-Ion batteries may be all the rage today, but scientists believe fluoride-ion batteries are better. The only problem is they haven’t found the perfect material to conduct the fluoride ions. A new approach uses machine learning to narrow down the choice of materials that can be used, greatly accelerating the research.
The missing link in efficient fluoride-ion batteries could be found using machine learning 7 photos
Photo: Mullen
The missing link in efficient fluoride-ion batteries could be found using machine learningSolid-state batteries are though to be safer, but a new study contradicts thatSolid-state batteries are though to be safer, but a new study contradicts thatSolid-state batteries are though to be safer, but a new study contradicts thatSolid-state batteries are though to be safer, but a new study contradicts thatSolid-state batteries are though to be safer, but a new study contradicts that
The theory says that fluoride-ion systems are ideal for batteries that could power everything from consumer electronics to electric vehicles. Fluoride offers tangible advantages over lithium, being lightweight and highly stable. It’s also cheaper than lithium and cobalt used in today’s Li-Ion batteries. To top it off, scientists calculated that fluoride-ion batteries could also offer higher energy density than their lithium-ion equivalents.

Considering all these theoretical advantages, you might ask why we don’t have yet any such batteries. Well, it’s because, between theory and practice, there’s often a huge rift. Fluoride-ion battery (FIB) research is still in its infancy, and the first rechargeable fluoride battery samples appeared only in 2011. The problem is that the scientists haven’t yet found the perfect materials to conduct the fluoride ions. More specifically, those with high ionic conductivity lack stability and vice versa.

Jack Sundberg and his colleagues in Scott Warren’s lab at the University of North Carolina have devised a new method to speed up the search for the best fluoride-conductive materials. The team uses machine learning algorithms and supercomputers that can rapidly and accurately predict how easily fluoride-ions move in any known fluoride-containing crystal.

The University of North Carolina team has got to a database of 10,000 fluoride-containing candidates out of 140,000 known materials. They randomly picked 300 and ran accurate calculations for each material’s fluoride-transporting ability. This phase took about one week per material. Still, the results were used to train the machine learning algorithm, which sped up the calculations to only one hour per material.

Their research already yielded material candidates described as “better conductors than those used in lithium-ion batteries.” One prime example is a fluoride-containing zinc-titanium compound, ZnTiF6. This is cheap, highly fluoride-conductive, and shows promising characteristics in other areas. It is not the only one, though, and the team has already patented the most promising compositions.

Having a solid starting point should accelerate the development of fluoride-ion batteries. Although it’s too early to tell, this type of chemistry might prove better in future solid-state batteries. After all, the materials used to make today’s Li-ion batteries are expensive and sometimes difficult to source. Scaling to the levels required by transitioning the whole car industry to all-electric automobiles requires a new approach if we want to avoid shortages and price hikes in the future.
If you liked the article, please follow us:  Google News icon Google News Youtube Instagram
About the author: Cristian Agatie
Cristian Agatie profile photo

After his childhood dream of becoming a "tractor operator" didn't pan out, Cristian turned to journalism, first in print and later moving to online media. His top interests are electric vehicles and new energy solutions.
Full profile

 

Would you like AUTOEVOLUTION to send you notifications?

You will only receive our top stories