Although we hope that Artificial Intelligence never reaches the point of The Matrix, a movie where machines use humans as batteries, the first steps are already being taken.
An article has just been published detailing how Artificial Intelligence is helping to create better battery electrolytes, the component that transfers ions back and forth between the two battery electrodes.
One of the authors of the article is Shirley Meng, chief scientist at the Argonne Collaborative Center for Energy Storage Science (ACCESS), who commented that the development of electrolytes is a key to cheaper, longer lasting and more powerful batteries. For this it is necessary to try different substances. For example, switching from a nickel-containing oxide to a sulfur-based material as the primary component of the positive electrode of a lithium-ion battery could yield significant performance benefits and lower costs, but to do so requires figuring out how to retun the electrolyte.
They need to design electrolytes with the right chemical and electrochemical properties to enable optimal interface formation at the battery’s positive and negative electrodes, and to do this they get help from artificial intelligence (AI) that digitally searches through many more possible candidates, speeding up what had been a slow and laborious process of laboratory synthesis.
Thanks to AI, they are able to identify the best descriptors and characteristics that will allow the personalized design of various electrolytes for specific uses. Before they only managed to test several dozen a year, now they manage several thousand, since electrolytes have billions of possible combinations of components, and AI can act as an automated laboratory.
[…] machines can single-handedly perform increasingly refined and calibrated experiments to ultimately determine what combination of components will form the perfect electrolyte […] the machines can work around the clock and reduce the potential for human error.
You can read more in the article “Designing better electrolytes”, which appeared in Science on December 8.