Artificial intelligence discovers a new material!

Scientists say the discovery could reduce lithium usage by 70 percent. The new material was tested in light bulbs and was successful.

The discovery was made by Microsoft and the Pacific Northwest National Laboratory (PNNL), which is under the US Department of Energy.

Researchers at Microsoft managed to identify 18 "promising candidates" out of 32 million potential inorganic materials in less than a week using artificial intelligence and supercomputers.

If traditional laboratory methods had been used, this screening could have taken more than 20 years to complete. From the start, the process of developing a working battery prototype took less than nine months.

Microsoft Corporate Vice President Jason Zander told the BBC that their goal is to "compress 250 years of scientific discovery into the next 25 years"; "We think technologies like this will help us in that goal. I think this is the way to do scientific developments like this in the future."

The newly discovered material, named N2116, is a solid-state electrolyte, and it was tested by scientists until it was transformed from a raw material into a working prototype.

This material also has the potential to be a sustainable energy storage solution because solid-state batteries are safer than traditional liquid or gel-like lithium batteries. It is also hoped that solid-state lithium batteries, which charge faster, will provide higher energy density in the near future thanks to their ability to be charged thousands of times.

What is the problem with lithium?

Lithium is often called "white gold" due to its market value and silvery color.

It is one of the most important components of rechargeable batteries (lithium-ion batteries). We use these batteries in almost every electronic device, from electric vehicles to our mobile phones.

The International Energy Agency warns that with the increasing demand for electric vehicles worldwide, we could start experiencing a lithium shortage within the next two years. The US Department of Energy also points out that demand for lithium-ion batteries could increase 10-fold by 2030. For this reason, manufacturers are constantly building new battery factories.

Lithium mining is also a controversial issue. It can take years to reach production, and it has serious environmental impacts. Extracting lithium requires huge amounts of water and energy. The process leaves large "scars" on the earth and produces toxic waste.

Dr. Nuria Tapia-Ruiz, who leads a team of battery researchers at Imperial College London, says that any material that contains little lithium and stores energy well would be "holy" for the lithium-ion battery industry; "Artificial intelligence and supercomputers will be vital for battery researchers in the coming years in identifying high-performance materials."

However, Dr. Edward Brightman from the University of Strathclyde in Scotland warns that the new technology should be "approached with caution"; "It may present artificial or seemingly good results at first glance. But these materials may be already known or materials that cannot be synthesized in the laboratory."

How is the AI used different from others?

The supercomputers used in the recent discovery process are computers that house a large number of computers and can perform complex scientific and mathematical operations.

The AI module developed by Microsoft is a technology trained on molecular data that can solve chemistry. Zander says, "This AI is based on scientific data and materials. The data is very safe to use in scientific discoveries."

After the artificial intelligence narrowed down the number of potential materials that could be used in batteries to 18, battery experts from PNNL examined these materials and determined the final material they would work on in the laboratory.

Karl Mueller from PNNL said that thanks to artificial intelligence, they "reached an area that could bear fruit much faster." According to Mueller, the latest discovery shows how much artificial intelligence can accelerate innovation processes.