Spent fuel, on-site storage and Idaho's limits

Every reactor in this course will make spent fuel, and the United States still has nowhere permanent to put it. The spent fuel and nuclear waste lesson covers the national picture: how pools and dry casks work, the Yucca Mountain stalemate, interim storage and reprocessing. This lesson asks what is different about spent fuel from small and advanced reactors, and what Idaho's agreements mean for the reactors being built at INL.

Different fuels, different leftovers

Fuel	Reactors	What comes out	Storage notes
LEU oxide pellets in rods	NuScale, BWRX-300, SMR-300	Assemblies like today's fleet produces	Pool cooling, then dry casks; the industry knows this path well
TRISO pebbles or compacts	Xe-100, Kairos, Kaleidos, Pele	Spent pebbles: a small amount of fuel inside a lot of graphite	Very durable particles, but large volume per unit of energy
Metallic fuel	Natrium, Aurora	Metal alloy pins	If sodium-bonded, as EBR-II's was, needs treatment before disposal
Fuel salt	Terrestrial IMSR, Natura MSR-1	Used salt containing fission products	Needs processing into a solid waste form; little commercial experience
Microreactor cores	Kaleidos, eVinci and others	Whole fueled cores, in some business models the whole reactor	Vendors plan to ship units back; where they go next is not settled

Idaho has direct experience with the metallic case. EBR-II's driver fuel had liquid sodium inside each pin to carry heat, and sodium reacts with water and air. DOE has been treating that fuel at INL for years, and under the 2019 supplemental agreement with Idaho it must turn EBR-II driver fuel into HALEU at a minimum rate and finish by December 31, 2028. That is the same material feeding first cores like Aurora's (lesson 5).

The Stanford-led waste study

In May 2022, Lindsay Krall, Allison Macfarlane and Rodney Ewing published a study in the Proceedings of the National Academy of Sciences looking at waste from three SMR designs: from Toshiba, NuScale and Terrestrial Energy. Their findings, as summarized by Stanford:

• Most SMR designs would increase the volume of waste needing management and disposal by factors of 2 to 30 for the reactors studied.
• SMRs would produce at least nine times more neutron-activated steel than conventional plants.
• Spent fuel would be greater in volume per unit of energy and more chemically complex.
• After 10,000 years, the plutonium in SMR spent fuel would be at least 50 percent more radiotoxic per unit of energy than in conventional spent fuel.

The main physical reason: small cores leak more neutrons out of the core, so they use fuel less efficiently, and those escaping neutrons activate surrounding steel. The study has limits the authors acknowledged. None of the reactors was operating, and proprietary design details made the analysis harder. Industry analysts have long argued that advanced designs will produce fewer radioactive byproducts, especially fast reactors that burn more of their fuel and designs paired with recycling. Where the evidence points: the study is the most detailed independent analysis available, and its core physics argument about neutron leakage is sound. Exact waste figures for any design will only be known when it runs.

On-site storage

With no repository, spent fuel from any new SMR would stay at the plant. For light-water designs that means a pool for several years, then welded steel canisters in concrete or steel overpacks on a pad, cooled by air flowing around them with no pumps or power. The NRC says dry casks have been used at U.S. plants for more than 30 years with an excellent safety record. A new site adds one more pad to the more than 70 sites in 35 states where commercial spent fuel already sits, per DOE.

Idaho's 1995 settlement agreement

The waste lesson explains how Idaho got its agreement. The terms that matter for new reactors, from INL's and DOE's summaries:

• A cap. DOE may send 55 metric tons (about 121,000 pounds) of spent fuel into the Idaho site between 1995 and 2035. Research quantities of commercial fuel count against that total under a 2011 agreement.
• Commercial power plant fuel is excluded, except for waivers the state has granted for small research quantities: 25 rods received in 2023, and a 2025 waiver for a high-burnup fuel cask and limited university research reactor fuel.
• A deadline. All spent fuel must be removed from Idaho by 2035.
• Teeth. If DOE misses a deadline, Idaho can suspend shipments into the state. In 2014, after DOE missed a 2012 waste treatment deadline, the state refused research fuel shipments, and they did not resume until after the 2019 supplemental agreement.

The Advanced Test Reactor showed how the 2035 deadline collides with reactors that keep running. In February 2020, Governor Brad Little and Attorney General Lawrence Wasden announced an agreement letting ATR keep operating: after January 1, 2035, its spent fuel (apart from test specimens) must leave Idaho within a year of going into dry storage.

The open question for new INL reactors

Aurora, Aalo-X, MARVEL, Pele and the DOME tests will all make spent fuel in Idaho, and some may run past 2035. The public INL and DOE settlement summaries reviewed for this lesson do not say how that fuel is treated: whether it counts against the 55-ton cap, falls under the 2035 removal requirement, or needs its own agreement like ATR's. Some of it is generated from material already in Idaho (Aurora's EBR-II-derived HALEU), which complicates the cap question, since the cap governs shipments into the state. The ATR precedent suggests the state and DOE will need written terms. That makes it one of the best questions an Idaho reporter or official can ask right now.

Supporters of the INL projects say spent fuel volumes from test reactors are small, the fuel is well characterized and DOE has managed fuel on the site for more than 70 years. Idaho's position, held by governors of both parties since Cecil Andrus and Phil Batt, is that research is welcome and a de facto repository is not. Both can be true, but only if the paperwork says so.