AI lab teaches itself to create superalloys – Metal Tech News
AI lab teaches itself to create superalloys
Metal Tech News – July 17, 2026
Share
Vladislav Ociacia at adobe.stock.com
University of Toronto researchers have developed an AI-powered active learning lab to create alloys that can be 3D printed into complex designs for jet engines and other extreme temperature applications
Tyler Irving / University of Toronto
University of Toronto PhD student Ajay Talbot holds up samples of metal alloys made of nickel, cobalt, and chromium created in the self-driving lab
Tyler Irving / University of Toronto
Computer modeling, machine learning, and robot-assisted manufacturing are being used to design, fabricate, and test samples of nickel-cobalt-chromium alloys like these. Information gained from those tests is fed back into the model to inform the next iteration of the process
University of Toronto scientists alloy computer modeling, machine learning, and robot-assisted manufacturing into a self-driving lab
University of Toronto Engineering researchers have used artificial intelligence to discover six new 3D-printable superalloys in a matter of weeks, including nickel-cobalt-chromium formulations that outperformed an industry-standard aerospace alloy under extreme temperatures
The researchers developed the alloys with a largely autonomous laboratory that uses computer modeling and machine learning to select metal compositions, robotic equipment to manufacture and test the samples, and the resulting data to determine which formulations to explore next
This closed-loop approach could dramatically accelerate the discovery of alloys designed to withstand the extreme heat, pressure, and corrosive conditions found inside jet engines, nuclear power plants, and other demanding environments
“There’s enormous demand for materials that can stand up to huge swings of temperature and pressure,” said University of Toronto Professor Yu Zou, Canada Research Chair in Materials and Manufacturing for Extreme Environments and leader of the project
Additive manufacturing, commonly known as 3D printing, adds another layer of complexity to alloy design
In addition to surviving extreme operating conditions, the metals must respond predictably to the rapid heating and cooling involved in laser-based printing. Alloys developed for traditional casting or forging can crack, deform or otherwise perform poorly when printed layer by layer
Printable alloys could allow manufacturers to create intricate parts that cannot be produced through conventional methods and vary material properties across a single component. A part could, for example, transition from a hard, heat-resistant exterior to a lighter interior
Many high-performance alloys used today consist primarily of nickel or cobalt mixed with smaller amounts of up to 10 other elements. Even slight changes in the elements or their proportions can significantly alter an alloy’s strength, hardness, oxidation resistance and printability
This creates tens of thousands of potential formulations, making the discovery of new alloys through traditional trial-and-error experimentation a costly and time-consuming undertaking
To explore this enormous design space more efficiently, Zou’s team collaborated with University of Toronto materials science Professor Jason Hattrick-Simpers to develop an AI-guided alloy discovery process
Tyler Irving / University of Toronto
Computer modeling, machine learning, and robot-assisted manufacturing are being used to design, fabricate, and test samples of nickel-cobalt-chromium alloys like these. Information gained from those tests is fed back into the model to inform the next iteration of the process
Self-driving alloys lab
Partially supported by the University of Toronto’s Acceleration Consortium, the project combines computer modeling, machine learning and robot-assisted manufacturing into what the researchers describe as a self-driving lab
The Acceleration Consortium is a global partnership of governments, academic institutions, and industries applying AI and automation to materials discovery
Rather than requiring a large database of information about existing alloys, the University of Toronto platform uses data-lean active-learning models that begin with a relatively small number of experiments
The system selects several compositions to manufacture and test. Data from those experiments is then fed back into the machine-learning model, which uses the results to identify the most promising formulations for the next round
This process allows the system to progressively map largely unexplored areas of alloy design with minimal human intervention
Ajay Talbot, a doctoral student in Zou’s laboratory and lead author of a paper describing the research in npj Advanced Manufacturing, said the self-driving lab is helping to overcome one of the largest obstacles to applying AI to materials science
“One problem you often run into when trying to use AI to design materials is that most machine learning models require lots of data about material properties to learn from,” he said. “But if you’re working in part of the design space that hasn’t been explored yet, that data doesn’t exist, so you’re kind of flying blind.”
The active-learning model navigates this uncharted territory by strategically choosing each new group of experiments based on what it learned from the previous results
To demonstrate the platform, the researchers focused on compositionally complex alloys – the combination of three or more primary metals – made from nickel, cobalt and chromium
Within weeks, the self-learning lab was combining these three metals into alloys that rival aerospace superalloys used by NASA
Tyler Irving / University of Toronto
University of Toronto PhD student Ajay Talbot holds up samples of metal alloys made of nickel, cobalt, and chromium created in the self-driving lab
Discovering aerospace superalloys
Nickel, cobalt, and chromium are important ingredients in aerospace superalloys because of the strength, heat resistance, and oxidation protection they can provide
The University of Toronto researchers first tasked the self-driving lab with identifying alloys capable of retaining hardness at temperatures of up to 1,112 degrees Fahrenheit (600 Celsius), comparable to conditions found toward the front of a jet engine
“The industry standard in this space is nickel-based alloys such as Inconel 625. But we found one made of 12% nickel, 62% cobalt and 26% chrome that was great for retaining hardness at extremely high temperatures,” said Talbot. “Even with just three components, our alloy outperformed Inconel 625 – an alloy of more than 10 different elements – by 4.5% in our lab tests.”
A more nickel- and chromium-forward alloy was designed to resist high-temperature oxidation at the back end of jet engines, where temperatures can reach 1,830F (1,000C)
At such temperatures, oxide scale can form on exposed metal, gradually consuming and degrading the material
The self-driving lab identified a formulation consisting of 36% nickel, 14% cobalt, and 50% chromium that performs extremely well at resisting oxidation at high temperatures, outperforming Inconel 625 by 85%
These laboratory results place the new formulations firmly in the realm of high-temperature superalloys, though the research remains at the materials-discovery stage rather than representing flight-qualified jet-engine components
The team is now working toward alloys capable of withstanding temperatures of up to 2,192 F (1,200 C)
The research team also plans to increase the complexity of the system by expanding beyond three principal metals
The next generations could incorporate 10 to 12 elements, opening the door to additional strengthening mechanisms and combinations of useful properties
“This nickel-cobalt-chrome system has just three elements in it. In the grand scheme of things, it’s a relatively simple system,” said Talbot
The three-metal experiments, however, demonstrate that the closed-loop discovery platform can rapidly identify high-performing materials far removed from the conventional equal-parts formulations that researchers might otherwise prioritize
As the self-driving lab generates more experimental data, it will be able to navigate increasingly complex alloy systems and potentially tailor new superalloys for specific aerospace, power-generation, and advanced-manufacturing applications
“There’s a lot more out there just waiting to be discovered,” Talbot said
Shane Lasley, Metal Tech News
With more than 18 years of covering mining, Shane is renowned for his insights and in-depth analysis of mining, mineral exploration, and technology metals
- Email: [email protected]
- Phone: 907-726-1095
- https://www.linkedin.com/in/shane-lasley-ab073b12/
You might be interested in:
Â
Most Popular
Trail anchors Canada’s critical minerals push
Shane Lasley | Jul 8, 2026
USA Rare Earth produces magnet REE oxides
Shane Lasley | Jul 14, 2026
War Department commits $25M to ReElement
A.J. Roan | Jul 13, 2026
Missouri lands critical minerals funding
Shane Lasley | Jul 14, 2026
Â
Rendered 07/17/2026 21:33
