A team of researchers from the New Jersey Institute of Technology has used powerful artificial intelligence tools to discover five brand-new porous materials, potentially revolutionizing how we store energy. Unlike today’s common lithium-ion batteries, these newly identified materials could be the key to building batteries based on more sustainable and widely available metals, such as magnesium, calcium, aluminum, and zinc.
How Did AI Make This Possible?
Traditional material discovery is a slow and painstaking process. Millions of possible material combinations exist, and testing them all by hand would take decades. The NJIT team, led by Professor Dibakar Datta, overcame this by combining two AI tools: a generative model called Crystal Diffusion Variational Autoencoder (CDVAE) and a large language model (LLM) customized for scientific research.
While the CDVAE model explored a huge range of possible crystal structures, the LLM quickly identified structures that were both stable and promising for battery use. “It was like searching for a needle in a haystack, but with AI, we could scan the entire haystack in hours instead of years,” Professor Datta explained.
The Breakthrough: New Porous Materials
The AI system pinpointed five completely new types of porous transition metal oxides. What makes these materials special? They have large, open channels that allow heavy, multi-charged ions—like magnesium or calcium—to move through them efficiently. That’s a major advance, since multivalent ions carry more charge than lithium ones, potentially leading to batteries that are longer-lasting and can store more energy.
The researchers didn’t just rely on computer predictions. They ran detailed quantum simulations and stability tests, showing these materials should work in the real world and could be synthesized in a lab.
Why Does This Matter?
This discovery is about more than better batteries. It opens the door to faster, smarter exploration of new materials for all sorts of technologies, from clean energy to advanced electronics. By harnessing AI, the timeline for finding new materials can drop from years to just a few hours.
The team is now collaborating with experimental scientists to create and test these AI-designed materials, with the ultimate goal of building practical, affordable alternatives to lithium batteries. This breakthrough comes at a critical time, as the world seeks safer, more sustainable solutions for energy storage.
Leave a Reply