AI-driven system discovers six new metal alloys that could reshape jet engines and nuclear power

By NewSunSEO | July 26, 2026

Researchers at the University of Toronto have used an artificial-intelligence-driven active learning platform to identify six new metal alloys that retain strength under extreme heat and pressure, with potential applications in jet engines and nuclear power plants. The system functions as a self-driving laboratory, using AI to select promising metal combinations, directing robots to manufacture and test them and feeding results back into the model to compress years of materials discovery into weeks.

How does the self-driving laboratory work?

The team, led by Canada Research Chair Yu Zou in the university’s Department of Materials Science and Engineering, developed the alloys within a few weeks using a method called active learning, which combines computer modeling, machine learning and robot-assisted manufacturing. Rather than manually testing thousands of metal combinations, the AI picks the most promising options, directs robots to manufacture them, evaluates their performance and uses those results to guide the next round of experiments. The new materials are also compatible with 3D metal printing, enabling the production of complex components that cannot be made using traditional methods.

The findings were published in the journal npj Advanced Manufacturing in June 2026. The paper is titled "Active learning for the accelerated discovery of complex concentrated NiCoCr alloys in additive manufacturing," and lists Ajay Talbot as first author and Yu Zou as corresponding author.

What problem does data-lean active learning solve?

Most machine-learning models require large datasets to make accurate predictions, but such data often does not exist for unexplored material combinations. According to the study, the team worked around that limitation with data-lean models that strategically select only a few samples to manufacture and test, then feed those measurements back into the model to inform the next round. As Talbot explained, "Our active learning model strategically selects a few samples to manufacture and test, and the data from those experiments is fed back into the model to inform where we go next. It really speeds things up."

What did the new alloys actually achieve?

The study focused on compositionally complex alloys built from nickel, cobalt and chromium. Among the six new printable alloys the platform identified, two were singled out for standout performance against the popular equiatomic NiCoCr reference and against industry-standard superalloys.

• Hot-section hardness: An alloy of 12% nickel, 62% cobalt and 26% chromium retained about 50% higher hardness than equiatomic NiCoCr at 600 °C and demonstrated exceptional hardness at temperatures up to 1,112 °F. In lab tests it outperformed Inconel 625, a nickel-based industry-standard alloy made from more than 10 elements, by 4.5%.
• High-temperature oxidation resistance: An alloy of 36% nickel, 14% cobalt and 50% chromium reduced oxidation mass gain by 85% at 1,000 °C compared with conventional superalloys, including Inconel 625, targeting even hotter sections of jet engines reaching 1,832 °F.

The study reports that, taken together, the six new alloys are up to about 40% harder than equiatomic NiCoCr at room temperature. According to the abstract, the results show that optimum performance can reside far from equiatomic proportions and can be uncovered with substantially fewer experiments when exploration is guided by application-specific objectives.

What applications is the team targeting?

Zou framed the industrial need plainly: "There’s enormous demand for materials that can stand up to huge swings of temperature and pressure, such as what you would find inside a jet engine or in the steam generators inside nuclear power plants, anywhere conventional steel just can’t survive." Talbot added that the team is aiming higher, saying, "We’re eventually aiming to ramp up to even higher temperatures, up to 2,192 °F."

What comes next for the platform?

The researchers describe the current results as an early demonstration of what the closed-loop platform can achieve. The nickel-cobalt-chrome system tested here uses only three elements, and the team plans to push the approach further by developing alloys with up to 10 or 12 different elements. As Talbot put it, "It’s great for showing that this whole closed-loop discovery platform really works. What we want to do next is ramp up the complexity a bit more to make even crazier stuff, with maybe up to 10 or 12 different elements."

The project was partially supported by the University of Toronto’s Acceleration Consortium, a group that uses AI and automation to accelerate materials discovery. Funding for the study came from the Natural Sciences and Engineering Research Council of Canada, the New Frontiers in Research Fund, the Digital Research Alliance of Canada, the Canadian Foundation for Innovation, and the Canada First Research Excellence Fund through the Acceleration Consortium. Co-authors Cristian Cojocaru is affiliated with the National Research Council Canada, and Jason Hattrick-Simpers is affiliated with the Acceleration Consortium, the Vector Institute for Artificial Intelligence and the Schwartz Reisman Institute for Technology and Society in addition to the Department of Materials Science and Engineering.

FAQ

Who led the AI alloy discovery research?

The work was led by Yu Zou, Canada Research Chair in Materials and Manufacturing for Extreme Environments in the University of Toronto’s Department of Materials Science and Engineering, with Ajay Talbot as first author of the paper published in npj Advanced Manufacturing in June 2026.

How does the AI platform find new alloys so quickly?

It uses active learning in a self-driving laboratory: the AI selects a small number of promising metal combinations, robots manufacture and test them, and the resulting data are fed back into the model to guide the next experiments, cutting what used to take years down to a few weeks.

How do the new alloys compare with Inconel 625?

The 12% Ni, 62% Co, 26% Cr alloy was 4.5% harder than Inconel 625 at temperatures up to 1,112 °F, and the 36% Ni, 14% Co, 50% Cr alloy showed 85% better oxidation resistance than Inconel 625 at 1,832 °F, conditions found inside jet engines.