Using AI-driven materials design, a team of researchers at University of Toronto Engineering has discovered a new set of metal alloys that retain their strength under extreme conditions.
The materials are well suited to additive manufacturing, also known as 3D metal printing, and could lead to enhanced, custom-made parts for aerospace, power generation and more.
"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," says Professor Yu Zou, the Canada Research Chair in Materials and Manufacturing for Extreme Environments, who led the project.
"We also need materials that can be printed layer by layer, enabling us to make components that can't be created by traditional manufacturing processes. For example, to make a material that is both lightweight and strong, you can vary the composition: a hard, tough alloy on the outside to something softer and lighter on the inside."
Zou says that many of the high-performance metal alloys used today are made primarily of a single component — often nickel or cobalt — with small amounts of up to ten other elements mixed in.
But there are many other formulations that haven't yet been explored, largely because of physical limits: identifying new 3D-printable alloys out of tens of thousands of possible combinations is a daunting task, particularly for complex alloys of three or more principal elements.
To overcome this problem and explore new regions of the potential design space, Zou's team, in collaboration with Professor Jason Hattrick-Simpers, is leveraging the power of AI.
Their technique, which they call active learning, combines computer modeling, machine learning and robot-assisted manufacturing to create a self-driving lab, capable of following up on promising leads with minimal human intervention.
The project is partially supported by the University of Toronto's Acceleration Consortium , a global community of government, academia and industry that uses AI and automation to accelerate the discovery of materials.
"One problem you often run into when trying to use AI to design to materials is that most machine learning models require lots of data about material properties to learn from," says Ajay Talbot, a PhD student in Zou's lab and lead author on a paper in npj Advanced Manufacturing that describes the work .
"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 way we get around that challenge is to use data-lean models that essentially feel their own way along. Our active learning model strategically selects a few samples to manufacture and test, and the data from those experiments is ingested back into the model to inform where we're going to go next. It really speeds things up."
To demonstrate the value of this approach, Talbot and his collaborators targeted what they call compositionally complex alloys that contain relatively large amounts of just three different elements: nickel, cobalt and chromium.
In just a few weeks of work, their self-driving lab had zeroed in on six new alloys with promising new properties.
"One of the properties we were targeting was puncture resistance at temperatures of up to 600 Celsius, which is what you'd find in the front section of a jet engine," says Talbot.
"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. 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."
Another alloy is designed for the back of jet engines, where temperatures can get even hotter, up to 1000 Celsius.
"One of the things that happens in an environment like that is the formation of oxide scale, which essentially means that your material is just getting burnt away," says Talbot.
"We found a material made of 36% nickel, 14% cobalt and 50% chrome that was extremely good for oxidation resistance at these high temperatures: it even outperforms Inconel 625 by 85%. We're eventually aiming to ramp up to even higher temperatures, up to 1,200 Celsius."
Talbot says that the new alloys reported in the paper are just the beginning.
"This nickel-cobalt-chrome system has just three elements in it. In the grand scheme of things, it's a relatively simple system," he says.
"But 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 ten or twelve different elements."
"As you add more components, you can get different strengthening mechanisms, different kinds of useful properties. There's a lot more out there just waiting to be discovered."
