Artificial intelligence (AI) and large-scale cloud computing are accelerating the search for new materials for batteries. The AI-powered collaboration between Microsoft and the Pacific Northwest National Laboratory (PNNL) has already produced a promising new material, which the two are sharing publicly today.
They discovered a new type of solid-state electrolyte, the type of material that could lead to a battery that is less prone to catching fire than current lithium-ion batteries. It also uses less lithium, which is becoming more difficult to obtain as demand for rechargeable electric vehicle batteries rises.
There is still a long way to go to see the viability of this material as an alternative to traditional lithium-ion batteries. What scientists are most excited about is the potential of generative AI to speed up their work. This discovery is the first of many materials they will test in the search for a better battery.
“If we can see this kind of acceleration, my bet would be that this is the way of the future to find these types of materials.”
“The important point to make is how quickly we get to a new idea, to a new material,” says Karl Müller, a physical chemist and director of the Program Development Office at PNNL. “If we can see that kind of acceleration, my bet would be that this is the way to go.” “For the future to find these types of materials.”
Microsoft approached PNNL researchers last year to offer Azure Quantum Elements (AQE), a platform that combines high-performance computing and artificial intelligence — and, ultimately, quantum computing, according to Microsoft. The company launched it last year as a tool designed for discoveries in chemistry and materials science.
The researchers inquired about the AQE for battery materials that use less lithium, and quickly suggested 32 million different candidates. From there, the AI system had to determine which of these materials would be stable enough to use, which ended up being around 500,000. They used more filters to infer how well each material conducts energy, simulate how atoms and molecules move within each material, and discover how practical each filter is when it comes to cost and availability.
Ultimately, only 23 candidates remained, five of which were already known subjects. The entire downsizing process took just 80 hours – a feat so fast that it would have been nearly impossible without AI and AQE.
“Thirty-two million is something we will never be able to do… Imagine a human being sitting down and going through 32 million materials and picking out one or two of them. That’s not going to happen,” says Vijay Murugesan, a staff scientist and head of the Materials Science Group at PNNL.
PNNL has assembled a promising candidate from this research to test. They were able to produce a working battery from it and use it to power a light bulb and a clock. Hundreds of prototype batteries will need to be tested and modified for this new material to prove effective. So, don’t expect it to hit store shelves anytime soon — there’s a lot of research on promising new materials that will never reach the market.
What’s interesting about this particular filter is that it uses a combination of lithium and sodium, which is an abundant element and the main component of salt. Microsoft says the new material can reduce the amount of sodium used in the battery by up to 70 percent.
Furthermore, it could be used to create a solid-state battery that is safer than current lithium-ion batteries made with liquid electrolytes that are more susceptible to overheating. The tricky part is that solid electrolytes generally haven’t been as good at delivering energy as their liquid counterparts. Researchers are still trying to overcome this challenge with this new material, as in laboratory tests it has shown a lower conductivity than initially expected.
Fortunately, there are still other promising candidates for researchers to manufacture and test as they try to create the next generation of batteries needed to power the world with renewable energy. Keep in mind that generative AI has an increasing environmental impact in its own right, especially the greenhouse gas emissions associated with all the energy burned using computing. This makes it important to increase the energy efficiency of computing and power data centers with clean energy at the same time, which requires better batteries.
“We really need to compress the next 250 years of chemicals science into the next two decades, right? That’s because we want to save our planet,” says Christa Spohr, who leads the Microsoft Quantum – Redmond (QuArC) group at Microsoft Research. “From these findings, AI and high-performance computing together have the potential to accelerate this scientific discovery.”