Researchers at the New Jersey Institute of Technology (NJIT) have used artificial intelligence to address energy storage concerns around lithium-ion batteries, with the task of developing a more sustainable alternative.
The findings could hold enormous potential for battery operated AV devices, as energy efficiency and battery life remain a concern.
The NJIT team research, published in Cell Reports Physical Science, saw generative AI techniques applied to quickly discover new porous materials capable of developing more sustainable, multivalent-ion batteries, using abundant elements such as magnesium, calcium, aluminium and zinc, with the aim of addressing global supply challenges and sustainability issues.
Multivalent-ion batteries use elements whose ions carry two or three positive charges, enabling the potential to store more energy, creating a more viable battery candidate for future energy storage solutions.
Multivalent ions feature a larger size and greater electrical charge; however, challenges have been found that make them difficult to accommodate efficiently in battery materials. The researchers relied on generative AI as a systematic way to sift through many material combinations, identifying the few structures that can make multivalent batteries practical and efficient.
The NJIT team developed a novel dual-AI approach, using a finely tuned large language model (LLM) and a crystal diffusion variational autoencoder to explore thousands of new crystal structures, something typically considered impossible using traditional laboratory experiments.
The AI-generated structures were validated using quantum mechanical simulations and stability tests, confirming that the materials could be synthesised experimentally and hold real-world application potentials.
The CDVAE model was trained on vast datasets of known crystal structures, enabling it to propose completely novel materials with diverse structural possibilities. Meanwhile, the LLM was tuned to zero in on materials closest to thermodynamic stability, crucial for practical synthesis.
Dibakar Datta, professor, NJIT, commented: “Our AI tools dramatically accelerated the discovery process, which uncovered five entirely new porous transition metal oxide structures that show remarkable promise,” said Datta. “These materials have large, open channels ideal for moving these bulky multivalent ions quickly and safely, a critical breakthrough for next-generation batteries.
This is more than just discovering new battery materials — it’s about establishing a rapid, scalable method to explore any advanced materials, from electronics to clean energy solutions, without extensive trial and error.”