MIT Harnesses Generative AI To Unlock Quantum Materials For Next-Gen Computing

Now, MIT researchers have developed a technique that lets popular generative materials models create promising quantum materials by following specific design rules. The rules, or constraints, guide models in developing materials with unique structures that give rise to quantum properties.

"The models from these large companies generate materials optimized for stability," says Mingda Li, MIT's Class of 1947 Career Development Professor. "Our perspective is that's not usually how materials science advances. We don't need 10 million new materials to change the world, we just need one really good material."

Publication and Early Results

The approach is described in a paper that will be published by Nature Materials. The researchers applied their technique to generate millions of candidate materials comprising geometric lattice structures with associated quantum properties. From that pool, they synthesized two actual materials with exotic magnetic traits.

"People in the quantum community really care about these geometric constraints, like the Kagome lattices that are two overlapping, upside-down triangles. We created materials with Kagome lattices because those materials can mimic the behavior of rare earth elements, so they are of high technical importance," Li says.

Li is the senior author of the paper. His MIT co-authors include PhD students Ryotaro Okabe, Mouyang Cheng, Abhijatmedhi Chotrattanapituk, and Denisse Cordova Carrizales; postdoctoral scholar Manasi Mandal; undergraduate researchers Kiran Mak and Bowen Yu; visiting scholar Nguyen Tuan Hung; Xiang Fu '22, PhD '24; and Professor of Electrical Engineering and Computer Science Tommi Jaakkola. Additional co-authors include Yao Wang of Emory University, Weiwei Xie of Michigan State University, YQ Cheng of Oak Ridge National Laboratory, and Robert Cava of Princeton University.

Steering Models Toward Impact

Its structure determines a material's properties, and quantum materials are no different. Certain atomic structures are more likely to give rise to exotic quantum properties than others. For instance, square lattices can serve as a platform for high-temperature superconductors, while Kagome and Lieb lattices can support the creation of materials useful for quantum computing.

To help diffusion-based generative models produce materials with specific geometric patterns, the researchers created SCIGEN. SCIGEN is a computer code that ensures diffusion models adhere to user-defined constraints at each generation step. With SCIGEN, users can give any generative AI diffusion model geometric structural rules to follow as it generates materials.

AI diffusion models work by sampling from their training dataset to generate structures that reflect the distribution of the dataset. SCIGEN blocks generations that don't align with the structural rules.

To test SCIGEN, the researchers applied it to a popular AI materials generation model known as DiffCSP. They had the SCIGEN-equipped model generate materials with Archimedean lattices, which are collections of 2D lattice tilings of different polygons. Archimedean lattices can give rise to a range of quantum phenomena and have been the subject of extensive research.

"Archimedean lattices give rise to quantum spin liquids and so-called flat bands, which can mimic the properties of rare earths without rare earth elements, so they are extremely important," says Cheng, a co-corresponding author. "Other Archimedean lattice materials have large pores that could be used for carbon capture and other applications. In some cases, there are no known materials with that lattice, so I think it will be really interesting to find the first material that fits in that lattice."

Testing and Synthesis

The model generated over 10 million material candidates with Archimedean lattices. One million of those materials survived a stability screening. Using supercomputers at Oak Ridge National Laboratory, the researchers simulated 26,000 materials to analyze the atomic behavior of these materials. Magnetism was found in 41 percent of those structures.

From that subset, the researchers synthesized two previously undiscovered compounds, TiPdBi and TiPbSb, in the labs of Xie and Cava. Experiments confirmed that the AI model's predictions aligned primarily with the actual properties of the materials.

"We wanted to discover new materials that could have a huge potential impact by incorporating these structures that have been known to give rise to quantum properties," says Okabe, the paper's first author.

Accelerating Material Breakthroughs

Quantum spin liquids could unlock the potential of quantum computing by enabling stable, error-resistant qubits. But no quantum spin liquid materials have been confirmed. Xie and Cava believe SCIGEN could accelerate the search.

"There's a big search for quantum computer materials and topological superconductors, and these are all related to the geometric patterns of materials," Xie says. "But experimental progress has been very, very slow," Cava adds. "Many of these materials must be in a triangular or Kagome lattice. By generating many materials that satisfy these constraints, experimentalists gain hundreds or thousands more candidates to accelerate research."

The researchers emphasize that experimentation remains crucial for determining whether AI-generated materials can be synthesized and for comparing their actual properties with model predictions. Future work on SCIGEN could incorporate additional design rules, including chemical and functional constraints, to further enhance its capabilities.

"People who want to change the world care about material properties more than stability," Okabe says. "With our approach, the ratio of stable materials goes down, but it opens the door to generate a whole bunch of promising materials."

Funding

The U.S. Department of Energy, the National Energy Research Scientific Computing Center, the National Science Foundation, and Oak Ridge National Laboratory supported the work.