The moment a nuclear reactor begins operation, a complex chain of events is initiated within the fuel: Heavy atoms split into fission products, knocking other atoms out of place and creating defects that can change how the fuel swells, transfers heat, and reacts chemically over time.
Understanding those processes is key to understanding how safe and efficient a nuclear reactor will be. But even for some of the most-studied fuel types, the mechanisms controlling those processes are unclear.
Such is the case with a particular kind of metallic fuel, uranium alloyed with 10 percent zirconium by weight, also known as U-10Zr. This fuel was extensively tested in historic sodium-cooled fast reactors such as the Experimental Breeder Reactor-II (EBR-II) in Idaho and the Fast Flux Testing Facility (FFTF) in Washington state, helping establish the foundation for metallic fuel development in the U.S. Today, U-10Zr is again attracting attention for use in next-generation advanced reactors.
But most studies of U-10Zr took place decades ago, leaving unanswered questions about exactly how the fuel changes when it undergoes nuclear fission in a reactor and how it interacts with the protective fuel cladding surrounding it.
Now, together with Idaho National Laboratory (INL), MIT researchers have led one of the most detailed three-dimensional studies of irradiated U-10Zr to date. The researchers used a technique known as high energy synchrotron X-ray computed tomography at Brookhaven National Laboratory (BNL) in New York to analyze the pore networks and chemical changes that formed under irradiation during use inside the FFTF reactor, providing new insights into how the material swells, transfers heat, and interacts with the fuel cladding.
The findings could help keep some nuclear reactors running for longer, while also informing the next generation of nuclear reactor fuel systems.
"This study helps us model the pore distribution in the fuel more accurately," says senior author Ericmoore Jossou, MIT's John Clark Hardwick (1986) Professor of Nuclear Science and Engineering. "It also helps us design for the safe operation of metallic fuels in reactors by giving us a better understanding of the role of pores and their importance."
Joining Jossou on the paper are first author and MIT postdoc Anthony Harrup; Riley Moeykens '25, SM '25; BNL researchers Michael Drakopoulos and Nghia Vo; and INL researchers Jana Howard, Colby Jensen, and Tiankai Yao.
Understanding nuclear fuel
A class of nuclear reactors known as sodium-cooled fast reactors generate energy from rods of metallic fuels that are sealed inside metal tubes called cladding. In each rod, heat generally moves outward from the center to the edge and then to the cladding, where liquid sodium carries heat away to be harvested into power.
"As you operate the reactor, the contact between the fuel and the cladding material creates chemical interactions that can be problematic," explains Jossou. "There is a migration of materials from the fuel to the cladding, like fission gases and rare earth elements called lanthanides, which can react with the cladding, cause embrittlement, and damage the fuel system."
Studies of previously irradiated fuel and its cladding have captured mostly two-dimensional snapshots, preventing scientists from seeing the full scale of the pore networks that influence heat transfer and transport materials like lanthanides. Previous studies also mainly focused on specific sections of the fuel system, such as the fuel center or the fuel cladding interface.
For their study, the MIT researchers used fuel samples from the Fast Flux Testing Facility reactor, a sodium-cooled fast neutron reactor located in Washington state that operated from 1982 to 1992.
The Idaho National Lab managed the samples and prepared the samples. The team studied the prepared samples using high-energy synchrotron X-ray tomography at the Brookhaven National Laboratory. The synchrotron generated high-energy X-rays that allowed the researchers to reconstruct the fuel's internal pore networks in three dimensions, revealing how porosity, chemistry, and fuel-cladding interactions evolve across the fuel radius.
The researchers found porosity increased modestly from the center of the fuel toward the fuel edge, but pore density jumped by over two orders of magnitude at the fuel's edge by the cladding. The researchers also characterized the size and shape of pores, finding small pores at the center that turn into larger pore networks pointing outward toward the edge.
"The pores are currently modeled as spheres; however, in reality they are more complex, especially when many pores merged together," Harrup says. "That's true from the center all the way to the cladding. It explains why the cladding reacts the way it does, and why we see cladding chemicals in the fuel."
The pore networks toward the edge allow fission products and lanthanides to move but slow down heat transport, impacting the fuel's performance and lifetime. The researchers also mapped their microstructural findings with changes in the chemistry of the fuel in different areas.
"With this study, we've conducted an in-depth analysis enabled by advanced computational imaging methods that has never been done before, with correlations between local chemical environments and the formation of pores," Harrup says. "It turns out that whether the environment is uranium rich or zirconium rich impacts the morphology and the channels of the pores. That has never been reported before."
"The ability to directly visualize pore connectivity and fuel cladding interaction in three dimensions gives us important insight for improving fuel performance for advanced metallic fuel for sodium fast reactors," says Tiankai Yao of INL.
Informing reactor designs
The experimental findings differed from some models of how pores form and how the fuel system swells, which could improve simulations to help keep reactors running for longer. They also give a more nuanced picture of how pores influence reactor performance and safety.
"This helps optimize the current metallic fuel proposed for sodium fast reactors," Jossou says. "Now, together with INL, we better understand how pores are influencing the thermal performance of metallic fuel in reactors. At high temperature, the pores are not all bad, because we found they act as pathways for liquid sodium metal to flow through the fuel and sustain thermal conductivity. Connected pores could also serve as releasing channels for fission gases which reduce the internal fuel matrix stress."
The findings could also be used to design better fuel systems for next generation of sodium fast reactors.
"This excellent piece of work generated by Professor Jossou's group in collaboration with INL and BNL has elegantly combined the strength of attenuation-based X-ray tomography and focused ion beam lift-outs and produced valuable insights to the location-specific 3D porosity distribution in neutron-irradiated U-10Zr fuel," says Dong Liu, a professor at Oxford University who was not associated with this work. "What is also impressive is that they correlated 3D porosity to the thermal properties of the fuels: The total volume fraction is not the only parameter that is important, the 3D topology also matters. This is extremely informative for the study of other types of porous nuclear materials."
The work was supported by the U.S. Department of Energy Office of Nuclear Energy and utilized resources at BNL and INL. The sample preparation was carried out at INL, which is part of the Nuclear Science User Facilities, through a Rapid Turnaround Award.
