Oak Ridge and Idaho National Laboratories partner to 3D print and AI-qualify nuclear pressure vessels for U.S. energy supply chain

Two of the country’s leading national laboratories are teaming up to rethink how nuclear pressure vessels get built. On August 19, 2026, Oak Ridge National Laboratory and Idaho National Laboratory announced a formal research collaboration to expand domestic manufacturing of these critical components—using wire arc 3D printing and AI-driven qualification tools—at the Materials and Manufacturing Innovation Days event hosted by ORNL.

The effort targets a real gap in U.S. industrial capacity, and it comes at a moment when demand for nuclear energy infrastructure is picking up fast.

Two national labs announce joint effort to manufacture nuclear pressure vessels domestically

The collaboration between ORNL and Idaho National Laboratory (INL) is a direct response to a concrete industrial problem: the U.S. doesn’t have enough domestic forging capacity to keep up with its nuclear ambitions. Pressure vessels—the large, heavy-walled containers that house reactor cores and other high-pressure systems—have traditionally been made through forging, a process that demands specialized equipment and expertise that’s in short supply domestically.

They have to handle intense heat, pressure, and radiation over decades of service, which is why manufacturing standards for them are so demanding.

The two labs announced the partnership on August 19 at M2IND, ORNL’s Materials and Manufacturing Innovation Days event, which brought together industry, government, and research leaders to discuss AI-enabled manufacturing, energy security, and domestic supply chains. Wire arc additive manufacturing sits at the center of the initiative—and it could seriously change how these components get made.

Limited domestic forging capacity drives the search for alternative manufacturing methods

Pressure vessels operate under extreme conditions. They have to handle intense heat, pressure, and radiation over decades of service, which is why manufacturing standards for them are so demanding. Forging has long been the go-to method because it produces dense, structurally sound metal parts—but U.S. forging infrastructure for large-scale components simply hasn’t kept pace with the growing nuclear energy pipeline.

Wire arc additive manufacturing offers a different path. Instead of shaping metal through force, it builds components layer by layer, melting metal wire with electric arcs and depositing it precisely to form complex shapes. Large parts can be produced without the constraints of traditional forge capacity. ORNL’s primary tool for this work is MedUSA, a platform using three coordinated robotic arms to print large, complex metal components—the infrastructure behind the early prototypes already coming out of the lab.

AI tools aim to qualify components in real time during printing, replacing post-production testing

One of the most ambitious goals of the collaboration is producing what researchers call “born-qualified” components. The idea: a part’s fitness for service gets verified continuously during the printing process itself, using data gathered in real time, rather than through lengthy post-production inspection.

INL brings expertise in AI, digital engineering, and data science to the partnership. One of its tools is Prometheus, an agentic AI initiative designed to verify shape and material properties as a component is being built. ORNL contributes its additive manufacturing leadership and the capabilities of the Manufacturing Demonstration Facility.

“We would like to achieve born-qualified pressure vessel components using data gathered during printing to confidently assess their worthiness for extreme environments,” said ORNL lead researcher Patxi Fernandez-Zelaia. INL’s Jorgen Rufner, group lead for advanced manufacturing, framed it the same way—evaluate a part’s performance while it’s being printed, not after. Cutting out post-production testing delays could meaningfully accelerate how fast nuclear components move from manufacturing to deployment.

Early milestones include a printed pressure vessel prototype and sensor brackets for an operational microreactor

The collaboration already has something concrete to show. In July, MDF scientists used the MedUSA platform to print a small nuclear pressure vessel—roughly 3 feet by 5 feet—from a steel alloy relevant to nuclear applications. Displayed at M2IND, it represents an early but real milestone: proof that the technology can produce a large, closed vessel at meaningful scale.

Around the same time, Oak Ridge National Laboratory used the same printer and material to produce neutron sensor brackets for Antares Nuclear Inc.’s R1 Mark-0 microreactor, a novel design that reached criticality at INL in June. That’s a functional reactor component, not a demonstration piece. Together, these two outputs show the technology can handle both large-format structural printing and precision parts for operating reactors. Researchers are now shifting focus toward integrating AI-assisted in-process verification—moving from proving the printing works to proving the qualification process can work alongside it.

Applications extend beyond nuclear energy to defense, aerospace, and chemical industries

The methods being developed here aren’t limited to nuclear pressure vessels. As the technology matures, researchers see potential to apply the same printing and AI-verification approach to other reactor components. Beyond nuclear energy entirely, industries that depend on large structural metal parts—chemical refining, oil and gas, defense, aerospace—could eventually benefit from the same advances.

Both laboratories operate under DOE management. INL is managed by Battelle Energy Alliance for DOE’s Office of Nuclear Energy and serves as the nation’s center for nuclear energy research and development. ORNL is managed by UT-Battelle for DOE’s Office of Science. The Manufacturing Demonstration Facility at ORNL, where the pressure vessel prototype was printed, is the only large-scale, open-access research institution in the country focused on early-stage advanced manufacturing technologies ready for industry adoption.

The core takeaway: two national labs are combining wire arc 3D printing with real-time AI qualification to address a genuine gap in U.S. nuclear manufacturing capacity—and they’ve already got working hardware to back it up.

Kelly Lippke

Kelly is an experienced writer with 15 years of experience exploring the big stories that shape our world, from tech breakthroughs and space exploration to climate, energy, and the fascinating quirks of science. She has a talent for turning complex ideas into sharp, memorable insights that stay with readers long after they’ve finished reading.