Why Multivalent-Ion Batteries Could Power the Future

Unlike conventional lithium-ion batteries, multivalent-ion batteries use ions carrying two or three positive charges instead of just one. This enables them to store significantly more energy, making them a promising candidate for next-generation energy storage.

Figure 1. AI Unveils the Future of Batteries—Beyond Lithium.

The main challenge is that these larger, highly charged ions move less easily through standard battery materials. To address this, the NJIT team leveraged AI to identify materials specifically optimized for efficiently handling multivalent ions, overcoming a major hurdle in their development. Figure 1 shows AI Unveils the Future of Batteries—Beyond Lithium.

Turning to Generative AI for Battery Breakthroughs

“One of the biggest hurdles wasn’t the lack of promising battery chemistries—it was the impossibility of testing millions of material combinations,” said Datta. “We turned to generative AI as a fast, systematic way to explore this vast landscape and identify the few structures that could make multivalent batteries practical.”

By leveraging AI, the team could rapidly screen thousands of potential candidates, dramatically accelerating the search for efficient and sustainable alternatives to lithium-ion technology.

The Power of Dual AI: CDVAE and LLM

To tackle these challenges, NJIT researchers developed a dual-AI approach combining a Crystal Diffusion Variational Autoencoder (CDVAE) with a finely tuned Large Language Model (LLM). Together, these tools explored thousands of new crystal structures—an impossible feat with traditional lab experiments.

The CDVAE, trained on extensive datasets of known crystals, generated entirely novel materials with diverse structures. Meanwhile, the LLM focused on candidates closest to thermodynamic stability, essential for practical synthesis.

Discovery of Five Breakthrough Structures

“Our AI tools accelerated discovery, revealing five new porous transition metal oxide structures with exceptional potential,” Datta explained. “These materials feature large, open channels ideal for fast, safe transport of bulky multivalent ions—a critical breakthrough for next-generation batteries.”

The team validated these AI-generated structures through quantum mechanical simulations and stability tests, confirming their experimental viability.

Beyond Batteries: A Scalable Materials Revolution

Datta emphasized the broader impact: “This goes beyond battery materials—it’s a scalable, rapid method to explore advanced materials for electronics, clean energy, and more, without extensive trial and error.”

With these promising results, the team plans to collaborate with experimental labs to synthesize and test their AI-designed materials, moving closer to commercially viable multivalent-ion batteries.

Source: SciTECHDaily

Cite this article:

Priyadharshini S (2025), AI Discovers the Next-Generation Battery Beyond Lithium, AnaTechMaz, pp.850