ADVANCED MATERIALS & PROCESSES | OCTOBER 2025 10 MACHINE LEARNING | AI AI FINDS ENERGY STORAGE OPTIONS Researchers at New Jersey Institute of Technology (NJIT), Newark, are using artificial intelligence to find affordable and sustainable alternatives to lithium- ion batteries. The NJIT team led by Professor Dibakar Datta applied generative AI techniques to discover porous materials capable of revolutionizing multivalent-ion batteries. These batteries, by using abundant elements like magnesium, calcium, aluminum, and zinc, offer a cost-effective alternative to lithium-ion batteries, which face global supply challenges. Multivalent-ion batteries use elements whose ions carry two or even three positive charges, which means they could store significantly more energy. However, the larger size and greater electrical charge of multivalent ions make them challenging to work with in battery materials—an obstacle the team’s new AI-driven research directly addresses. “One of the biggest hurdles wasn’t a lack of promising battery chemistries—it was the sheer impossibility of testing millions of material combinations,” says Datta. “We turned to generative AI as a fast, systematic way to sift through that vast landscape and spot the few structures that could truly make multi- valent batteries practical.” The researchers developed a dual-AI approach—a crystal diffusion variational auto-encoder (CDVAE) and a finely tuned large language model (LLM). Together, these tools quickly explored thousands of new crystal structures. The CDVAE model was trained on vast datasets of known crystal structures, enabling it to propose new materials with diverse structural possibilities. Meanwhile, the LLM was tuned to focus on materials closest to thermodynamic stability. “Our AI tools dramatically accelerated the discovery process, which uncovered five entirely new porous transition metal oxide structures that show remarkable promise,” says Datta. “These materials have large, open channels ideal for moving these bulky multivalent ions quickly and safely, a critical breakthrough for next-generation batteries.” The scientists validated their AI-generated structures using quantum mechanical simulations and stability tests, confirming that the materials could be synthesized experi- mentally and hold great potential for real-world applications. njit.edu. ADVANCED CORROSION ASSESSMENT WITH AI Researchers at the Indian Institute of Science (IISc) and the Qatar Science and Technology Research Center (QSRTC) developed an Microscopy images marked with deposit thickness measurements. Courtesy of Ashwin RajKumar. automated method to assess corrosion in industrial equipment using advanced machine learning and image analysis. The work involves a new algorithm that can analyze microscope images of corroded metal surfaces to estimate corrosion severity without human input. The AI-based technique focuses on two key indicators of corrosion: the thickness of corrosive deposits on the surface of metals and the porosity within those deposits. When microscopy images of metal surfaces are fed to the algorithm, it can quantify these two characteristics and infer certain features that indicate how much corrosion has occurred. These features include the concentration of corrosive chemicals and the acidity of the environment beneath the deposits. The team tested their method by checking the corrosion of steam generator tubes. They found that the algorithm achieved 73% accuracy and is more consistent than having people manually examine optical micro- scopy images. The next step is to validate the algorithm on larger and more diverse datasets. www.iisc.ac.in. The spongelike network inside a porous transitionmetal oxide lets larger ions travel during a battery’s charge/discharge cycles. Courtesy of NJIT.
RkJQdWJsaXNoZXIy MTYyMzk3NQ==