Researchers at the University of Toronto have harnessed AI and robotics to develop six innovative alloys that withstand extreme heat, promising advancements for aerospace, power, and manufacturing sectors.
Researchers at the University of Toronto Engineering have used artificial intelligence and robotics to identify six new nickel-cobalt-chromium alloys that keep their strength at temperatures that would defeat many conventional metals, a development with clear implications for sectors such as aerospace, power generation and advanced manufacturing.
The work, published in npj Advanced Manufacturing, centres on a self-driving laboratory that combines machine learning, computer simulation and robotic experimentation to narrow the search for promising compositions. Rather than relying on the slow process of trial and error, the system predicts which alloy combinations are most likely to perform well, then has robots fabricate and test small samples before feeding the results back into the model. The result is a rapid cycle of discovery that can adapt as new data emerge.
That matters because the market for high-temperature materials is constrained by a familiar problem: many of the components needed in turbines, jet engines and nuclear facilities must operate in punishing heat and under heavy mechanical stress, yet the metals traditionally used for such tasks can lose performance or oxidise over time. Industrial users are also increasingly looking for materials that can be made by metal additive manufacturing, which allows engineers to produce complex geometries and tailor different properties within a single part.
Professor Yu Zou, who led the study, and his team focused on a three-element system built around nickel, cobalt and chromium. Even within that relatively simple chemistry, the AI-guided platform identified six viable new alloys within weeks. One formulation, containing 12% nickel, 62% cobalt and 26% chromium, proved especially strong at around 600C and was harder than Inconel 625 by 4.5% in laboratory testing. Another, with 36% nickel, 14% cobalt and 50% chromium, was designed for far harsher conditions and showed an 85% improvement in oxidation resistance at about 1,000C.
The broader significance goes beyond the individual alloys. The University of Toronto says its Acceleration Consortium is built around this kind of self-driving laboratory, with the aim of compressing materials discovery from a process that can traditionally take years and require substantial funding into something far faster and more economical. For industries trying to decarbonise, that could be important: lighter, more durable components can improve efficiency, while more heat-resistant materials can support equipment operating closer to thermal limits in turbines, reactors and other high-duty systems.
The new alloys are still at an early stage, but the researchers say the approach is scalable. Future work will examine more complex blends involving up to 10 or 12 elements, raising the possibility of materials with even better combinations of strength, durability and heat tolerance. In a field where incremental gains can translate into major performance and efficiency improvements, the combination of AI, robotics and additive manufacturing is beginning to look less like a laboratory curiosity and more like a practical industrial tool.

