In nuclear reactors, radiation causes defects to form inside materials, and this process can change those materials’ overall properties—usually for the worse.
One approach for mitigating this radiation damage is heating those damaged materials. This process rearranges the materials’ atoms and recovers the materials’ original properties.
Now, a team of University of Wisconsin-Madison engineers better understands how reactor materials recover from radiation damage when they’re heated.
Charlie Hirst, a UW-Madison assistant professor of nuclear engineering and engineering physics, and his collaborators visualized how radiation damage evolves in real time. What they learned could not only inform strategies for heating irradiated reactor materials to recover their properties—but also, importantly, extend the lifetime of current nuclear reactors.
The researchers detailed their advance in a paper published online in September 2025 in the journal Scripta Materialia.
To study radiation damage recovery, scientists have typically used transmission electron microscopes to take images of the microstructures of irradiated materials before and after they’ve been heated.
“However, this approach only gives you snapshots of the material before and after,” Hirst says. “It’s like watching only the first few minutes and the last few minutes of a movie and then trying to figure out what happened during the movie. There’s a lot of important information about how the defects are evolving that isn’t captured with this approach.”
To address this challenge, Hirst and his team used a new approach. In collaboration with Idaho National Lab, the researchers placed a sample of neutron-irradiated titanium on a special chip that can be heated up in a transmission electron microscope. This allowed them to visualize how the titanium’s microstructure changed while it heated up.
The researchers discovered that defect clusters moved around during heating—actually leading to recovery of the microstructure. “Crucially, this motion is not described by the existing theory for radiation damage recovery, which assumes defect clusters will shrink or grow—not physically move around,” Hirst says. “Our approach gave us really useful data and insight about what’s happening in the material.”
In previous work published in Science Advances, Hirst and his collaborators conducted atomistic-scale simulations of radiation damage recovery that predicted the defects’ motion. “It’s exciting that our experiments directly visualized and validated our previous hypothesis about the defects moving around,” Hirst says. “It’s sometimes quite rare for simulations and experiments to actually match up as they did here.”
Ian Steigerwald, a co-author on the paper and graduate student at the University of Michigan, manually identified the defects in the microscope images, which was a laborious process. In future work, Hirst plans to explore how machine learning could be harnessed to detect defects much faster from large datasets generated from similar types of in situ experiments.
View video clips showing the microstructural radiation damage recovery here.
Charlie Hirst is the Steven J. and Teresa M. Zinkle Nuclear Materials Assistant Professor.
Additional co-authors on the paper include Boopathy Kombaiah of Idaho National Laboratory, Kevin Field of the University of Michigan, and Michael Short of the Massachusetts Institute of Technology.
This work was supported by the U.S. Department of Energy Office of Nuclear Energy under DOE Idaho Operations Office Contract DE-AC07-051D14517 as part of a Nuclear Science User Facilities award number 4238, as well as the DOE Office of Basic Energy Sciences under award number DE-SC0021529, and the National Science Foundation Faculty Early Career Development Program grant DMR-1654548.
Featured image caption: Assistant Professor Charlie Hirst works on equipment in the UW-Madison Ion Beam Laboratory. Credit: Joel Hallberg.