Whether commercial demand for high-assay low-enriched uranium (HALEU) fuel ultimately falls at the high or low end of divergent forecasts, one thing is certain: the United States is not ready to meet demand, because it currently has no domestic HALEU enrichment capacity. But conversations happening now could help build the commercial HALEU enrichment infrastructure needed to support advanced reactor deployments. At the 2022 American Nuclear Society Winter Meeting, representatives from three potential HALEU enrichers, the government, and industry met to discuss their timelines and challenges during “Got Fuel? Progress Toward Establishing a Domestic US HALEU Supply,” a November 15 executive session cosponsored by the Nuclear Nonproliferation Policy Division and the Fuel Cycle and Waste Management Division.
The session was chaired and organized by Christina Leggett, a nuclear consultant at Booz Allen Hamilton, member of the ANS Board of Directors, and vice chair of the Fuel Cycle and Waste Management Division; and by Morris Hassler (who also participated as a panelist), vice president and business unit general manager of joint services for IB3 Global Services. Six panelists talked through the issues of HALEU enrichment: Andy Griffith, the Department of Energy’s deputy assistant secretary for nuclear fuel cycle and supply chain in the Office of Nuclear Energy; Hassler; Ben Jordan, manufacturing manager at Centrus Energy; Stephen Long, chief executive officer of Global Laser Enrichment (GLE); Magnus Mori, vice president of sales and marketing at Urenco; and Everett Redmond, senior director of fuel affairs at Oklo.
Immediate needs: Leggett, who moderated the session, opened it by saying,” I think we all agree that everyone's really excited about nuclear right now. We have unprecedented private investment in nuclear. . . . But for those of us who are involved with the fuel cycle, there's one question that's foremost on our minds. It's ‘Got fuel?’”
Redmond, as the only representative of a reactor developer on the panel, explained that “advanced reactor companies are not in a position, at the moment, to sign 10-year contracts. . . . We're heading in that direction, but that's where . . . the Department of Energy [and] Congress [have] the foresight to help provide that market demand signal right now.” He continued, “When we talk about reactor startup, fuel fabrication has to occur typically at least a year ahead of that. So, we have to take delivery of enriched uranium a year to two years before we're going to start up. For Oklo, we have line of sight on our first core—we're going to use reprocessed EBR-II fuel. For subsequent cores, however, we're in the same boat as everybody else and we need HALEU up to 20 percent.”
Griffith leads the DOE's research and development on advanced nuclear fuel cycle technologies. “There's a range of enriched uranium programmatic needs, and the big difference between them is the timing,” he said. “At the Office of Nuclear Energy our commercial needs are immediate—as soon as possible we need to stand up a commercial supply.” As for other, noncommercial HALEU needs, including for research reactor fuel and defense programs, “anything we do for deployment of advanced reactors is only going to help and pave the way toward their long-term viability,” Griffith added.
Why not just downblend HEU? Hassler retired as senior director of global security and strategic partnerships at the Y-12 National Security Complex and Pantex sites after 30 years of supplying HALEU and other materials, and he talked about the government’s past management of supply and demand for HALEU and high-enriched uranium. He explained that while “we've been supplying HALEU since the Atoms for Peace initiative . . . we didn't enrich HALEU, we either enriched up to 5 percent or we enriched highly enriched material [and downblended it to HALEU], so there hasn't been that capability.”
Hassler said he’s heard people say, in answer to the urgent calls for HALEU supply, “Let's use all of the [HEU] stocks that are out there.” His response? “I will tell you . . . it is not there anymore. . . . We did a good job of getting rid of over 180 metric tons of the supply that was out there. We downblended that material to less than 5 percent and it powered a lot of homes in the United States as low-enriched fuel. . . . If we had it to do over again, it probably would have been nice to have downblended a lot of that material to HALEU and put it on the shelf. But that was not what the environmental impact statement and the record of decision at the time was, because no one was thinking of HALEU at that point, even though it was one of the alternatives in the analysis. . . . Believe me, we have cleaned out the cupboards.” Any material remaining is earmarked for the nuclear deterrent, naval production, or research reactor commitments.
Government boost: To address the challenge, the DOE has downblended some limited stocks of HEU held at national laboratories and has initiated the HALEU Availability Program, which was mandated by the Energy Act of 2020 and received initial funding of $700 million in the Inflation Reduction Act. The DOE recently awarded a contract to Centrus Energy to complete a HALEU enrichment demonstration project and ramp production up to the level of 900 kilograms a year level by 2024 (on December 1, Centrus announced the final contract had been signed). Through the HALEU Availability Program the DOE is looking for more—it hopes to buy 25 metric tons of HALEU of per year “as soon as possible” through competitively awarded offtake contracts. Both Urenco and Global Laser Enrichment are positioning themselves to meet some of that need.
Mori, representing Urenco, which is the only company enriching uranium commercially in the United States today, suggested that “the current approach that the DOE is taking to have an offtake for HALEU would then justify automatically the expansion of the LEU feed that is needed to supply that facility.”
It matters where that natural uranium and LEU feedstock comes from. Long noted that of the four or five key criteria that GLE will factor into a decision to accelerate its commercialization plan, “Number one, we think that there needs to be long-term clarity regarding restrictions on Russian fuel supplies regardless of what incentives the government puts out or how we work with industry. We can't ignore the fact that Russian imports of material into this country perturbate the naturally competitive market forces, and they've discouraged investment in the nuclear fuel cycle domestically for some time.”
Building out capacity: Centrus is completing a 16-centrifuge HALEU demonstration cascade in Piketon, Ohio, the only facility currently licensed to enrich uranium up to 19.75 percent U-235. The “massive facility” has room for 11,520 centrifuges, according to Jordan, which makes the demo program of “16 and two spares a very, very small starting point, but an encouraging and very important first step. . . . We're going to start production in 2023 and we will create the capability to make 900 kilograms of HALEU per year."
Urenco is planning to increase enrichment capacity and levels at its Eunice, N.M., gaseous centrifuge facility, which currently enriches uranium up to 5.5 percent. “Phase One is what we call LEU+ for enrichments up to 9.9 percent,” said Mori. “That's fully funded, engineering is being completed, and we anticipate production at the facility in New Mexico starting in [the first quarter of] 2025. Phase two, which is also ongoing, is HALEU between 10 and 20 percent. That . . . requires a new facility . . . different regulatory requirements, different safeguards requirements. For that we anticipate an expenditure of about $250 to $400 million, and we anticipate it will be operational in 2028.”
Unlike Centrus and Urenco, GLE’s enrichment plans do not rely on gaseous centrifuge technology. Long reminded attendees that in 2012 GLE received “the first-ever [Nuclear Regulatory Commission] license to construct and operate a commercial-scale laser enrichment facility.” GLE’s Wilmington, N.C., site was licensed for 6 million separative work units (SWU) at an enrichment up to 8 percent. Long explained that technology development slowed after then–part owner GE scaled back its operations in 2013, but in 2016 GLE secured the rights to re-enrich 200,000 metric tons of depleted uranium tails inventories at the DOE’s shuttered Paducah gaseous diffusion facility.
GLE is “currently installing the first of two pilot scale laser modules that we plan to use in our pilot-scale technology demonstration in the next two years,” Long said, while eyeing operations at a commercial-scale Paducah Laser Enrichment Facility by 2030, or—if the accelerated path is chosen—by 2027.
The deconversion debate: LEU for the operating fleet is transported to fuel fabrication facilities as UF6, where it is deconverted to uranium oxide prior to fabrication. For HALEU-fueled advanced reactors, “many of the reactor designers also want to fabricate their fuel,” Hassler explained, adding that “from an economic standpoint it's very hard to think about shipping HALEU UF6 on a good commercial scale,” because criticality safety considerations limit the volume that can be sent in one package.
Enrichers are interested in hosting deconversion services. Mori said that Urenco is “in active talks with a partner to collocate the conversion facility in New Mexico. I can't say more at this point, but we are progressing the project.”
Jordan said that at Centrus, “We believe deconversion needs to have the option to be at the point of enrichment, and that is our intention. We intend to have deconversion capability to at least metal, if not metal and oxide, at our production facility. But of course there's also going to be some that want us to be able to send UF6 elsewhere and to their production facilities. That certainly is an intention as well.”
Long had a slightly different perspective. Before joining GLE, he worked for 13 years with GE Hitachi and Global Nuclear Fuel. “Coming from a previous life as a fuel fabricator, my initial inclination was to lean towards deconversion at the fabricator, especially with the potential different forms being explored there,” he said. “But . . . it's really going to come down to the detailed economic analysis to support which collocation, if any, is the most appropriate.”
Price point: As the likely first buyer of HALEU, the DOE is carefully considering the price it will pay for bankable HALEU and the price it will offer to reactor developers in turn. “I think the RFP that we're planning to socialize in draft form by the end of the calendar year and . . . actual proposals for what it's going to cost to stand up that commercial supply—I think that's going to ‘set the ceiling,’ if you will,” Griffith said. “Initially our plans are to adjust the price that we distribute and provide that to the two demos and those advanced reactors beyond that. . . . We're going to have to put some effort into calculating [that distribution price] so that it's more reflective of a long-term stable market price. . . . We want to do that in a thoughtful way that encourages the market and doesn't dampen it.”
While an established LEU market serves the current U.S. fleet with about 2,000 metric tons per year, even the highest demand estimates for HALEU (in December 2021 the Nuclear Energy Institute surveyed 10 reactor developers and estimated that they could need a total of as much as 215 t per year by 2030) are a small fraction of that volume. A smaller market is less able to absorb the cost of building new infrastructure, which is one reason enrichers have been waiting for clear demand signals.
Jordan was asked by the panel organizers to discuss HALEU economics. “The key to remember is that HALEU enrichment is something that is done to fuel that's already been enriched to an LEU level,” he said. “You've got 4.5 kilograms of LEU being used to make 1 kilogram of 19.75 [HALEU]. And we checked [the LEU price] just within the last couple of weeks—our LEU sales folks were telling us that you'll need $11,200 of LEU for each kilogram of HALEU that you produce. So that's key to keep in mind. It's not going to be cheaper than that in today's market.”
Jordan explained some of the costs that will go into HALEU production at any facility. “You can't just repurpose an existing cascade and make it produce 20 percent material,” he said. “In general, you're going to need to build a new facility that meets NRC Category II requirements and the nuclear criticality safety requirements . . . [and] that's going to entail higher fixed costs. And unfortunately, right now, as you all have seen, you're not going to be making anywhere near as much HALEU, no matter whose numbers you believe, as the folks producing LEU. So you've got a smaller bucket to spread that cost across. . . . As you keep adding capacity, it is going to come down in price.”
Mori pointed out that inflation is adding to Urenco’s unknowns. “When I asked my design team to give me projections [for the cost of the HALEU facility], they told me, ‘Well, look, for some of the materials we need for the centrifuges, prices have tripled in the last 10 months . . . this is our uncertainty right now,’” Mori said. “The infrastructure itself will not change whether we have one, two, or possibly even three cascades. So, if we can sell the output of three cascades, that capex will get depreciated very quickly and price will come down very quickly as well. But we’ll have to see what that demand is.”
Flexible output: An audience member asked whether enrichers had the flexibility to supply HALEU in different forms—as uranium oxide, uranium metal, or enriched UF6—and at different levels of enrichment between 10 and 20 percent to meet reactor design needs.
The answer from enrichers? It depends on demand.
Jordan said that if a hypothetical company wanted to buy a small amount of HALEU at 12 percent enrichment from Centrus they would probably receive material that had been downblended from 19.75 percent (a potentially less economical option). “But if they gain traction and begin to deploy the number of reactors and fuel consumption that could take an entire cascade of our centrifuges, then if they’re confident and we’re confident it would be appropriate to build them a cascade dedicated to producing the assay and material that they are looking to obtain. So not only does it depend, but it could change over time as they get further down their business plan,” Jordan said, adding that being prepared to offer HALEU in both metal and oxide forms “is extremely important, and that’s the direction we’re headed.”
At Urenco, Mori said, “We’ll design the plant to be safe and to go up to 20 percent. And then . . . it will depend on the trade-off between volume and the enrichment and output required. We can do anything really—if there's demand.”
Long pointed out that GLE’s laser enrichment technology would be “modular and flexible . . . we don't have the issue of not being able to turn off our technology.” He added, that “on the reactor side, as I understand it, many of the technologies they have some flexibility to vary their core design around the enrichment.” Long suggested a “meet-in-the-middle approach until we see some consolidation in the market as to the widely adopted designs, and then we can start making decisions around that.”
What happens next? “From our perspective at DOE, we're looking at a temporary presence to incentivize the establishment of that supply chain,” Griffith said. “Now, what happens after those two demonstrations is going to be absolutely critically important to what’s next. . . . I think what happens with the TerraPower Natrium evaluation of five additional sites . . . how the X-energy cooperation with Dow unfolds for their next four-pack—those are going to be very telling of what the true demand signal will be.”
Griffith is not banking on the demand estimates released by the NEI, which he said were “very aspirational and assumed that all the regulatory process went like clockwork, all the financing went like clockwork, all the design and supply chain went like clockwork.” He does, however, want to see “the developers, the utilities, the industrial users, or the advanced reactors [setting] independent contracts outside of our commitments to add to the commitment that we make to get those commercial juices flowing and build that market.”
As the panel session was nearing an end, one audience member asked, given the growing demand for clean energy and advanced reactors, “Won’t this economic question take care of itself?”
In response, Jordan provided some insights into the decision-making process for enrichment companies and investors when he said that “the power producers themselves are making decisions that really put a question mark beside HALEU.” He explained that new-build contracts for light water–cooled small modular reactors fueled with conventional LEU “are going to keep companies like mine from writing huge checks and from getting investment institutions to write huge checks that allow us to build the type of HALEU capacity that would be needed to really support all the HALEU contenders out there that want a fair chance to make their business plan come true. I think that's why there's a government role in answering this question, and it's not as open-and-shut a case as we might wish that it was.”