AI can reduce development time for new batteries – 01/25/2024 – Market

Battery industry executives believe that artificial intelligence (AI) can halve the time needed to develop new materials, helping to accelerate innovation that plays a crucial role in combating climate change.

Mathias Miedreich, chief executive of Belgium-based Umicore, which is working with Microsoft, said AI will have an “immense impact” on the discovery and commercialization of new battery materials.

“The ambition is, when you have a certain product that you want to develop, to reduce the time to market by half. We are now carrying out the first concrete projects to see if this is viable,” he said.

Over the past two years, Umicore has supplied materials for car batteries used by companies including Volkswagen, BMW and Stellantis, whose brands include Chrysler, Fiat and Peugeot.

He hopes AI will help shorten the research phase in developing new battery materials from five or six years to two years.

The partnership allowed the company to file AI-enabled patents, Miedreich said.

The technology appears to hold most promise for advances in solid-state materials and disordered rock salts, which would dramatically improve an electric car’s range and open up new applications for batteries, he added.

Battery makers are optimistic about the AI ​​revolution in discovering new materials, continuing a shift seen in the pharmaceutical industry where the technology is being used to accelerate the development of new drugs.

“It will be a substantial acceleration at different phases of the process,” said Siddharth Khullar, a former Apple AI researcher who leads machine learning at Swedish battery maker Northvolt.

“It’s like a recipe with multiple ingredients and we want to figure out if the absolute ingredient is the right one or in the right proportion. This is a huge space to explore. With AI, you can start to expand the comfort zone of where you’re looking” .

Earlier this month, Microsoft said its work with the Pacific Northwest National Laboratory in the US showed that its algorithm was able to narrow down 32 million materials to 18 promising candidates for use in batteries in just 80 hours.

The government-backed laboratory is testing one such material, a lithium-sodium hybrid, that could reduce lithium content by up to 70% — a material that concerns car manufacturers due to its possible scarcity.

Venkat Srinivasan, director of the Argonne Collaborative Center for Energy Storage Science, a US government-funded laboratory, said he was “pretty optimistic” about the impact of AI on battery materials.

This is because advances in AI have come together with the automation of robot experiments, allowing the collection of large volumes of experimental data.

Miedreich warned that there may not be enough computing power to make all the discoveries. And even if suitable materials are identified, researchers must test them under real-world conditions to find out if they work as expected.

The safety of batteries made with new substances is also critical, which means testing times are difficult to reduce, executives said.

However, with the US seen as a leader in AI over China, it could offer the West a chance to catch up to Asian rivals that dominate the global battery industry, one investor said.

“The U.S. could try to use AI to optimize the battery materials it has access to and doesn’t have to rely on China,” said James Frith, principal at Volta Energy Technologies, a battery supply chain investment vehicle.