US study of reactor and fuel types to enable naval reactors to shift from HEU fuel

Alan Kuperman, Frank von Hippel

In February 2020, the US Department of Energy's office of Defense Nuclear Nonproliferation released its report, Initial Evaluation of Fuel-Reactor Concepts for Advanced LEU Fuel Development, a screening study for potential fuel and reactor types that may be relevant to switching US naval nuclear propulsion away from reliance on highly enriched uranium (HEU) fuel. The DNN report was commissioned from three DOE national laboratories with reactor-design expertise: Idaho, Oak Ridge and Argonne. It ends up recommending two reactor types and seven fuels for further investigation in the next phase of its work. These options include pressurized water reactors (PWRs), and a number of possible high-density, low enriched uranium (LEU) fuels.

Currently, US and UK naval reactors are fueled by weapon-grade HEU (93.5% U-235). Russia and India also use HEU (≥ 20% U-235). The other two countries with nuclear submarines, China and France, use LEU.

The origin of the new report can be traced to a request from Congress that led in 2014 to the Office of Naval Reactors (NR) submitting to Congress a Report on Low Enriched Uranium for Naval Reactor Cores (see also an earlier IPFM post). In comparison to a report on the same topic written in 1995, the 2014 report was quite positive:

recent work has shown that the potential exists to develop an advanced fuel system that could increase uranium loading beyond what is practical today while meeting the rigorous performance requirements for naval reactors. Success is not assured, but an advanced fuel system might ... allow using LEU fuel with less impact on reactor lifetime, size, and ship costs.

The Congressional interest was in whether it might be possible to expand a longstanding US-led nonproliferation initiative, aimed at ending worldwide civilian use of HEU in research-reactor fuel and medical isotope production, to encompass all non-weapon uses of HEU, including those of the military. The benefits would include eliminating the possibility of theft of HEU from the naval fuel cycle, which has happened in the past in both the US and Russia. They also include making it more difficult for countries such as Iran to justify production of HEU for their planned naval reactors. Brazil, the only non-nuclear-weapons state currently with a nuclear submarine development program is working with LEU fuel.

Congress responded to the 2014 NR report by appropriating $5 million in FY16 to commence R&D of LEU fuel for Navy propulsion reactors. In 2016, DOE submitted a report, Conceptual Research and Development Plan for Low-Enriched Uranium Naval Fuel (see IPFM blog post), laying out a 15-year, $1 billion, program to develop Navy LEU fuel:

Development of an advanced naval fuel that uses LEU would demonstrate United States leadership toward reducing HEU and achieving nuclear non-proliferation goals...The advanced LEU fuel system concept has the potential to satisfy the energy requirements of an aircraft carrier without affecting the number of refuelings ... An LEU-fueled submarine with this fuel is expected to require at least one refueling, or the reactor (and hull) would need to be increased in size correspondingly ... For these reasons, an LEU-fueled submarine reactor is a larger challenge which would not be addressed until experience could be gained during the development of an LEU-fueled aircraft carrier reactor.

Congress provided another $5 million for FY17 but the members that were interested in LEU fuel indicated that they wanted NR to pursue LEU fuel options for submarines as well. It was understood that reactor cores sufficient to power submarines for their full design lives - a long-term NR goal which it believed that it had finally achieved in the Virginia-class attack submarine, which has begun deployment and is in ongoing production - might have to be larger if LEU fuel were used, which might require a reconfiguration of the reactor compartment in subsequent classes of submarines.

NR failed to spend the first two years' of funding for Navy LEU fuel R&D, so Congress appropriated another $5 million for FY18 but transferred the project to DOE's Office of Defense Nuclear Nonproliferation (DNN), and asked the Secretaries of the Navy and Energy to make a joint determination on the program (NDAA FY2018 Conference Report, § 7319, see also an IPFM post). In March 2018, the Secretaries of Energy and the Navy responded negatively, writing to the Committees on Armed Services that:

Funding developmental work of this magnitude is not possible without increasing risk to other existing naval nuclear propulsion efforts. A program to pursue R&D of an LEU advanced fuel system would compete for necessary resources against all other NNSA and Department of Defense priorities as part of a future budget request.

Nevertheless, Congressional advocates of Navy LEU fuel were undeterred and actually increased FY19 funding to $10 million in DNN's budget (p. 167). An additional $15 million was provided in FY20 (p. H11262), and DNN tasked DOE's three civilian laboratories with reactor-design expertise, led by Idaho National Laboratory (INL), to look into the matter.

Although the FY19 funding was explicitly for an "advanced naval nuclear fuel system," which was the language that NR had used in its 2016 report on LEU fuel for aircraft carriers and submarines, DNN was apparently concerned about being seen as intruding on NR turf. The report therefore describes the research as looking at concepts for "off-grid power sources for ship-borne systems and transportable systems to fixed locations, e.g., to a remote base or to a seawater desalination service location" - all possible applications other than the one that Congress had provided funding to research.

Reactor types. The labs chose to focus on a reactor sized to generate 100 Megawatts electric (MWe) for 30 years operating at a 20-percent average capacity factor. The power is about the geometric mean between public estimates of the power outputs of US submarine and aircraft carrier reactors. The capacity factor agrees with estimates for naval reactors if one takes into account the time that ships spend in port and the fact that they usually do not move at maximum speed. (Water resistance increases as the cube of speed. At 80 percent of maximum speed, the propulsive power required is therefore cut in half.) The 30-year life is between the 25-year core life for the two reactors on US Nimitz-class aircraft carriers and the 33-year design life of Virginia-class submarines, which have lifetime cores (p. 10). The reactor was also required to have "rapid and repeated load following capability from 0 to 100% power without rate or restart constraints due to Xe poisoning and other effects," which is appropriate for a military ship (p. 2).

The two reactor types recommended for further study are pressurized water reactors (PWRs), the current type used by all nuclear navies, and liquid-sodium-cooled fast-neutron reactors (SFRs), a reactor type that INL and Argonne National Laboratory have been promoting as an advanced power reactor.

The legendary leader of the US naval-reactor development, Admiral Hyman Rickover, actually tried an SFR in his second nuclear submarine, the Seawolf (SSN 575) but replaced the reactor with a PWR after about a year. According to the historians of the founding of the US nuclear navy, Rickover's verdict about sodium-cooled reactors was that they were

expensive to build, complex to operate, susceptible to prolonged shutdown as a result of even minor malfunctions, and difficult and time-consuming to repair. (Nuclear Navy: 1946-1962, p. 274.)

The Soviet Union also tried closely-related liquid-metal (lead-bismuth) cooled reactors in eight attack submarines: one November-class submarine, the K-27, which went into service in 1963 but was retired after five years as a result of a reactor accident and seven Alfa-class submarines, which went into service during 1971 and 1981 and were retired in 1996. One of the Alphas also had a serious reactor accident (The Use of Highly-Enriched Uranium as Fuel in Russia, p. 94).

Fuel type. For the enrichment of the LEU fuel, the lab study chose 19.75%, just below the 20% boundary line above which enriched uranium is considered weapon usable. It is believed US naval reactor fuel is ceramic UO2 particles embedded in zirconium metal ("cermet") (Effects of Variation of Uranium Enrichment on Nuclear Submarine Reactor Design, p. 65).

The labs narrowed down their candidate fuels to seven. The top four are UO2 (the "benchmark"), a uranium-molybdenum alloy (7-10 weight percent molybdenum), USi3 and uranium metal alloyed with 2% molybdenum and 1% silicon (Si). These compounds have uranium densities about 50% higher than UO2, which could theoretically result in the possibility of packing the same amount of uranium into a core with about two thirds of the volume of fuel. This would reduce the volume penalty from converting to a lower enrichment of U-235.

The volume penalty is further reduced because it is not necessary to have the same amount of U-235 to achieve the same core life, as during irradiation some of the U-238 in the LEU fuel is fissioned and some is converted into chain-reacting plutonium, some of which is later fissioned. In its 1995 report to Congress, NR reported that, because of these contributions, only two-thirds as much U-235 would be required at 20 percent enrichment as at 93% (Report on Use of Low Enriched Uranium in Naval Propulsion, p. 10). This effect, combined with a 50% increase of uranium density in the fuel would result in an LEU core having twice the volume of an HEU core for the same core life.

A life-of-the-ship LEU core might still require a larger reactor pressure vessel and a modest increase in submarine length to provide additional buoyancy to offset the increased weight of the reactor compartment. Alternatively, the Navy could revert to mid-life refueling, the traditional practice for pre-Virginia class submarines that still comprise the bulk of the fleet. France's navy, which has excellent nuclear submarines, has designed them to be refueled in weeks rather than years as the US does, which is an approach that could address the US Navy's main objection to refueling. Aircraft carrier reactor pressure vessels may be already large enough to accommodate a larger LEU core. DOE's 2016 report to Congress stated that, "Preliminary design work has shown that an initial application of LEU fuel in an aircraft carrier reactor might meet ship performance requirements in the available size envelope."

The recent February 2020 report is the first from DOE's advanced-LEU (aLEU) Fuel for Nonproliferation Applications Project that includes reactor modeling and experimental work on the candidate fuels. The project aims eventually "to offer a candidate LEU fuel system" for seaborne applications consistent with the reactor performance requirements. This project is housed in DNN's office of Research and Development and is the military counterpart to the longstanding civilian LEU fuel development program, the Reduced Enrichment for Research and Test Reactors (RERTR) program established in 1978, which is based in DNN's office of Material Management and Minimization.