Scientists build generative AI tool to fast-track quantum material discoveries

Scientists have created an artificial intelligence tool called SCIGEN that can potentially speed up the hunt for novel quantum materials. Such unusual substances, displaying odd electronic and magnetic characteristics, are promising candidates to be the next generation's building blocks of quantum computers, nanoscale electronics, and next-generation energy devices.

The study employs machine learning and rigorous geometric rules in combination to produce millions of candidate materials, some of which seem both stable and strange enough to be interesting.

Quantum materials are center-stage for modern physics and chemistry. Their strange behaviors, say superconducting or exotic magnetic properties, can power revolutionary technology. The problem is that they are very difficult to discover.

The number of atomic arrangements available is so enormous that it is virtually impossible to look through all of them. Despite a couple of decades of work, scientists have been successful in identifying only a few stable candidates for such phenomena as quantum spin liquids, which have the potential for quantum computing.

This bottleneck spurred researchers at MIT and collaborating institutions to try a different tack. Instead of allowing the AI to generate materials at random, they told it to replicate known patterns in order to induce quantum behavior.

SCIGEN, or Structural Constraint Integration in GENerative model, works by guiding a standard form of generative AI called diffusion models. They normally start with some random noise and progressively move that towards building a structure. But if left to their own devices, they like to stay near what they've been trained with and venture into very few unusual geometries.

What makes SCIGEN special is that it brings in rules into the game. At each step, the system guides the model toward specific geometries of the lattice, such as honeycomb, kagome, or Archimedean structures. Such structures are most interesting to physicists because these tend to host exotic states such as high-temperature superconductors or odd magnetic orders.

"We don't need 10 million new materials to save the world, we just need one really good material," says Mingda Li, MIT's Class of 1947 Career Development Professor and lead author of the study.

To test the method, the group used SCIGEN to generate about 10 million inorganic compounds that have Archimedean lattice tilings. These tilings, made of repetition shapes like triangles, squares, or hexagons, are aesthetically pleasing in mathematics and physically intriguing.

The researchers then screened them through a four-step process that cut out unstable or chemically unreasonable candidates. A million or so survived the first sieve. They selected 26,000 for more extensive simulations using density functional theory (DFT), a standard quantum mechanical workhorse.

The result was surprising. Fully more than 95 percent of the DFT calculations converged. Over half of those materials proved to be structurally stable, their atoms settling into low-energy structures. Better still, 41 percent showed magnetic ordering, a characteristic often linked with exotic physics.

Yes, it's easy to forecast materials on a computer; it's another thing to produce them in a lab. To push the idea further, the team tried to synthesize two of the forecasted compounds: TiPd₀.₂₂Bi₀.₈₈ and Ti₀.₅Pd₁.₅Sb. Both were subjected to tests as paramagnetic and diamagnetic.

While not the exotic magnets scientists want most, both findings were in line with the forecasts, proving that SCIGEN can in fact produce materials that can be synthesized and tested in reality.

"Quantum community members really care about these geometric borders," explains Li. "We created materials with kagome lattices because they can mimic the behavior of rare earths, which are of great technical importance."

SCIGEN has some advantages over conventional methods. It doesn't require retraining the entire AI model, so it is flexible and economical. It doesn't spend time going down dead ends by focusing on rules that are equivalent to scientific facts. And the rate of success of stable structures is substantially higher than would happen under blind guessing.

The fact that the researchers can go from AI predictions to synthesis in the lab enhances the argument. Princeton's Robert Cava and Michigan State's Weiwei Xie, who helped conduct experiments, state that SCIGEN could speed up the hunt for highly coveted compounds like quantum spin liquids and topological superconductors.

There is a tremendous search for the components of quantum computers and topological superconductors, and all of these are tied with the material's geometric patterns," Xie says. "Experimental progress, however, has been very, very slow."

The group considers SCIGEN an initial point, not a terminal one. Additional types of rules like bonding preference, electronic properties, or even defect structures can be added to the subsequent work. The approach is also amenable to other diffusion models, an advantage for other research groups.

There are problems. Not everything generated passes the filter, and most fail when synthesized in the lab experimentally. Actual materials also involve issues like impurities, defects, and synthesis difficulty. Despite this, the advancements are staggering. Instead of searching blindly in an astronomical search space, researchers now have a map that shows them to areas with higher likelihoods.

"People who want to change the world care more about material properties than stability," says Ryotaro Okabe, first author of the paper. "Our approach lowers the ratio of stable materials, but it enables us to synthesize a bunch of promising materials."

SCIGEN can potentially transform how scientists search for and test new substances. From billions of possible materials, the system filters out to those with the best chance of displaying beneficial behavior, eliminating years of trial and error. That means faster progress toward materials that have the potential to enable next-generation quantum computing, superconductors, and clean energy technologies.

Beyond the laboratory, the research has the potential to make an impact upon society over the long term. More efficient superconductors can minimize energy losses in power grids. New quantum magnets could provide the foundation for ultra-high speed computing.

By accelerating the efficiency of materials discovery, SCIGEN shortens the time between theory and practical innovations that benefit mankind worldwide.

-------------------------------------------------

The research team included contributors across multiple institutions: from MIT, co-authors ranged from students and postdocs to faculty, including PhD candidates Ryotaro Okabe, Mouyang Cheng, Abhijatmedhi Chotrattanapituk, and Denisse Cordova Carrizales; postdoctoral fellow Manasi Mandal; undergraduates Kiran Mak and Bowen Yu; visiting scholar Nguyen Tuan Hung; alumnus Xiang Fu ’22, PhD ’24; and Tommi Jaakkola, professor of electrical engineering and computer science and affiliate of CSAIL and the Institute for Data, Systems, and Society.

External collaborators included Yao Wang of Emory University, Weiwei Xie of Michigan State University, YQ Cheng of Oak Ridge National Laboratory, and Robert Cava of Princeton University.

-------------------------------------------------

Research findings are available online in the journal Nature Materials.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.