Deploying microreactor technology for military applications could have huge impacts on logistics and reliability for the military of the future, and on the commercial use of similar technologies. That’s why the Department of Defense is developing Project Pele—a high-temperature, gas-cooled and TRISO-fueled microreactor, transportable within mobile shipping containers—for testing at Idaho National Laboratory in 2025.
Work is underway at BWXT Advanced Technologies in Lynchburg, Va., where a renovated facility will also host the assembly of another DOD reactor—a space-bound reactor for DRACO, the nuclear thermal rocket program being led by the Defense Advanced Research Projects Agency (DARPA) in collaboration with NASA.
Jeff Waksman, a program manager in the DOD’s Strategic Capabilities Office (SCO), and Joe Miller, president of BWXT Advanced Technologies, provided an update on Project Pele in an August 2 American Nuclear Society webinar (now available to view online). They were joined by moderator Alex Gilbert, director of space and planetary regulation at Zeno Power Systems, for a dynamic Q&A that explored the motivations and manufacturing behind Project Pele.
Military motivation: The DOD is interested in microreactors to provide power in remote or islanded locations, Waksman said, and also to offset fossil fuels—both “for climate purposes” and “from a strategic and operational perspective.”
“The exhaust on Pele is hot air, literally,” Waksman said. “So you could use that for heat. . . . I do think that using nuclear power for both heat and also process heat for chemical processes, desalination, synthetic aviation fuel—I think those are all applications that that the DOD is going to be looking at in the future.”
The idea for the SCO-led project originated at U.S. Indo-Pacific Command (INDOPACOM) in Hawaii, according to Waksman, and that is reflected in the project’s name. Pele is “a backronym for ‘Portable Energy for Lasting Effects,’ but no one uses the backronym. Pele is really just named after the Hawaiian goddess of fire and power,” Waksman said.
Scheduling: Waksman said the team is now in “the final bits of the design phase,” and hopes “to have the final design approved by the Department of Energy by the spring.”
“We are now ordering hardware, we are making fuel, we are forging the containment vessel, we're making moderator blocks—all that good stuff,” he added. “We have now reached the point in the famous Admiral Rickover essay where we are no longer a paper reactor or an academic reactor—we are now becoming very real.”
Waksman explained that reactor assembly will follow DOE design approval, and “if all goes according to schedule, by early 2025 we will have shipped the reactor to Idaho National Laboratory. At Idaho, it will then be fueled [and] shipped out to the desert to a location that we have selected to do the initial testing. The reactor will then go through a final operational readiness review, and if all—again—goes according to schedule, we will be able to turn the reactor on before the end of calendar year 2025.”
To meet the DOD’s goal of developing a transportable power source, the reactor will be delivered within a set of standard shipping containers. “There are several reasons why we decided to build this reactor in a series of boxes rather than in one big box,” Waksman said. “One of the advantages is that we have one reactor module that has the nuclear stuff in it. For lack of a more technical term, all the radioactive stuff is in one box. There are very few moving parts in that box. And the idea is that that box will be shut for the entire length of the operation, and then when you're done, you ship that whole box back and you get a whole new box. We do not want people opening up boxes and messing with used fuel at a remote military site.”
BWXT leads the build for Pele—and DRACO: In June 2022, BWX Technologies won the contract to build and deliver the Project Pele microreactor following two years under contract to deliver first an engineering design, and then a final design. Last month, BWXT was also selected to supply the nuclear reactor and fuel for the DRACO nuclear thermal rocket (formally known as the Demonstration Rocket for Agile Cislunar Operations). Miller said that BWXT is renovating a 170,000-square-foot building that will host both DRACO and Pele during assembly. In the case of Pele that means coordinating with partners, including Northrup Grumman and Rolls-Royce LibertyWorks.
“The fact that the government is putting out real money to build prototypes to demonstrate new nuclear is something we haven't seen in a very, very long time. [It’s] injecting a lot of vitality back into nuclear, and it's just great to be part of it,” Miller said.
Waksman said that the Project Pele and DRACO teams have supported each other during project planning and design stages. “Obviously there are several differences with the space reactor in terms of it's being an NTP [nuclear thermal propulsion] system, . . . and also the way you cool reactors is very different in space, but I'm a big believer that the hard part of nuclear engineering is really not those larger details. It's really the smaller details, how do you make real things real. And so in that way, there is just a ton of overlap between Pele and DRACO.”
Miller said that those similarities are beneficial when it comes to manufacturing. “The geometries are quite different, but the way in which you control the reactor is similar, regardless of the reactor, and then the fuel manufacturing capability can be pretty flexible, so being able to port one type of fuel manufacturing equipment and lay-down area and things like that offers a big benefit,” he said.
TRISO-fueled Pele will operate at a power level between 1 MWe and 5 MWe. The uranium-235 in the fuel “actually comes out of some old nuclear warheads that were decommissioned,” Waksman said. “In terms of actually taking that fuel and downblending it to HALEU and then turning it into TRISO, that's all being done by BWXT at the Lynchburg facility.”
From military to market: “The first commercial nuclear reactor ever built in the world in Shippingport, Pa., was built by the navy. And it was an aircraft carrier reactor design, and so that allowed the commercial companies to see a design—to see something that worked, to understand the manufacturing process and go and run with it from there,” Waksman said. “So our plan from the start for Project Pele has always been that we want to build reactors that can have commercial spin-offs so that we can get the number of reactors coming off the assembly line to be enough that it'll be cost competitive.”
Miller agreed, and said, “We've been able to parallel some of the activities within BWXT for our BANR reactor [BWXT Advanced Nuclear Reactor]. . . . Having the capability and the understanding and the supply chain knowledge and all the design-for-manufacturing effort that's gone into this program, we're able to port that for commercial purposes and look to how we can apply that commercial reactor to future markets.”
Operating plan: “We anticipate this reactor is largely autonomous. That said, there's always going to be some need for operator input,” Waksman said. “And even if you had a fully autonomous reactor, I view it the same way that you look at autonomous cars. Technically, we have cars that can drive themselves down the road. No one allows cars on the road without a driver at the steering wheel.”
Waksman anticipates similar policies for nuclear reactors and predicts that “it will be a very long time before someone allows an unaccompanied operating nuclear reactor. I don't think that's going to happen anytime soon, so we are assuming that at all times there will need to be two operators.”
Idaho National Laboratory has hired several future Project Pele operators, Waksman said, “mostly former naval reactor folks. . . . And these people will be trained to operate Pele. They will also be trained to operate MARVEL [a 100-kW thermal microreactor to be operated inside INL’s Transient Reactor Test facility], and they will then be available to operate other reactors that might come down the road at INL, so we're hoping to create an ecosystem of operators who are trained at handling different sorts of advanced reactors.”
Licensing: The licensing process has proven to be complicated, Waksman said. “You have to consider, ‘Well, what if this goes wrong, and there's an earthquake, and there's a fire, and all your computers go out? How do you make sure nothing bad happens?’ And that's why nuclear is so safe.”
The reactor will be licensed by the DOE, not the Nuclear Regulatory Commission, but it’s being designed to be “NRC certifiable,” according to Waksman, who added that “the NRC does sit in on all of our design reviews. They have the opportunity to express any input that they may have.”