Issues involving Storage and Transportation of High Burnup Nuclear Fuel
Marvin Resnikoff, Ph.D.
SCE Community Engagement Panel (CEP)
San Juan Capistrano Community Center
May 6, 2014
In the interests of full disclosure, I once worked for a public interest organization with the trademarked name, CEP, Council on Economic Priorities, and co-authored a book in 1983 on transportation issues, 3 years before Holtec, who supplies dry storage casks for the nuclear industry. The CEP book supported dry storage of nuclear fuel, but I never realized at the time the present situation, the amount of fuel and burnup that the industry would employ. In a way, part of the problem is my doing. As a member of the Sierra Club, we intervened against the only commercial reprocessing operation in the United States, Nuclear Fuel Services in West Valley, NY, and shut them down. The lack of reprocessing has led utilities to store more fuel in storage pools and in dry storage casks. The lack of a final repository is also partly my doing. I work for the State of Nevada as a consultant on nuclear transportation issues and have since 1986. My parents never gave me a middle name, but sometimes I think it’s “Trouble.”
So utilities are left with the problem of spent nuclear fuel and also faced with competition from natural gas. The economics has forced utilities to hold fuel in reactors longer, not 3 years, but 4 ½, which means less shutdown time. And the economics are also forcing the industry to put more fuel into each dry storage cask, moving from 24 PWR assemblies, to 32, which Transnuclear has requested for San Onofre, to 37 PWR assemblies, which Holtec has requested. I’m going to briefly discuss transportation and storage of nuclear fuel, and I’m going to focus on high burnup nuclear fuel (HBF). What and why is HBF? NRC has not fully investigated the technical issues and implications, which in my view, are major and should have required careful study and an EIS. This is work that should have been done before the NRC allowed utilities to go to high burnup, not after. By high burnup, I mean fuel greater than 45 GWD/MTU, but in clearer terms, allowing each assembly to remain in the reactor longer. The implications are the radioactive inventory in HBF is greater. NRC staff have focused on the heat in HBF, which is greater. But heat will decline over time. One implication is decommissioning will take longer. Fuel will sit in fuel pool for 20 years or more. San Onofre has high burnup fuel. The implication of a longer decay time is that the workers at the site will not be available for the decom process. Putting more fuel into the same space, moving from 24 fuel assemblies to 32, as Southern California Edison intends to do, will further the cooling off period. However, while heat is an important consideration, but perhaps of greater import is the impact on fuel cladding. It may surprise you to know that the NRC does not know how much HBF exists across the country. While the NRC has the power and the ability to identify how much HBF is at each reactor. The NRC has inspectors at each reactor. They simply have not made the effort. The Department of Energy (DOE) is conducting a survey which should be released in September. HBF has major implications for decommissioning, storage, transportation and disposal.
Let’s step back a second. Nuclear fuel assembly – collection of fuel rods. (fuel assembly) Each rod, about 12 feet long is composed of a tube, cladding, with nuclear fuel stacked like poker chips inside. But the cladding is quite thin, not much thicker than heavy duty aluminum foil. During operation and after, the cladding will develop defects. Studies by Argonne show that the zirconium cladding of HBF will become less ductile, or more brittle. How brittle? The NRC has contracted with Oak Ridge to examine cladding of HBF. The Oak Ridge study should have been completed in March, but has not been released. I call on the NRC to release the Oak Ridge study, before it is manicured by public relations specialists. This is a study that should have been done before HBF was licensed, not after the fact. In response the NRC would say, we do have technical support. The NRC will cite a study at Turkey Point reactor. But this demonstration project examined a cask loaded with lower burnup fuel (approximately 30 GWd/MTU average). Following 15 years of storage, the cask internals and fuel did not show any significant degradation (Einziger et al., 2003). According to that report, the data from this study can be extrapolated to maintain a licensing safety finding that low burnup SNF can be safely stored in a dry storage mode for at least 80 years with an appropriate aging management program that considers the effects of aging on systems, structures, and components (SSCs). The limits in ISG-11, Rev. 3, a peak cladding temperature of 400 oC, are all based on data available prior to 2002. None of this is directly relevant to HBF.
The NRC will also cite the 1988 report, PNL-6258, “Assessment of the Use of Extended Burnup Fuel in Light Water Power Reactors,” but this report did not address the cladding problems of HBF.
Cooling during storage may result in hydride-induced embrittlement. According to a more recent Argonne report, “pre-storage drying-transfer operations and early stage storage subject cladding to higher temperatures and much higher pressure-induced tensile stresses than experienced in-reactor or during pool storage.” The Argonne report discussed the problems of embrittlement of cladding of HBF. Due to thinning of cladding and lack of ductility, the cladding is weakened. As a result the cladding may not be an effective barrier to release of radioactivity to the cask canister. A report by the Nuclear Waste Technical Review Board goes into the matter in great detail. Thinning of cladding is correlated with the outer oxide layer on the cladding. As seen in the figure below, at a burnup of 60 GWD/MTU, the outer oxide layer is 115 microns. Considering the initial cladding thickness is on average 600 microns, NWTRB calculates a metal loss on the order of 70 microns or 12% at 60 GWD/MTU. Together with a hydride layer inside the cladding, this represents substantial weakening of the cladding.
Moving closer to home, for this reason, we are of the opinion, Edison should consider the HBF fuel assemblies to be damaged fuel that should be individually canned; the canned assemblies would then be stored in a HUHOMS concrete containment (NUHOMS being inserted) or a Holtec vertical silo (Holtec silo) for an indefinite period.
Passive cooling works like a chimney. Once fuel is removed and put into storage, after 18 to 20 years, the NRC license can be converted to storage. Here is what remains of CT Yankee reactor (photo). Nuclear fuel in 40 Holtec casks, and reactor internals in 3 casks. San Onofre will have many more casks. But one additional feature distinguishes the San Onofre situation, the salt environment. Documents show that the stainless steel canister has pitting corrosion, after less than 20 years. This is a major concern if casks are going to remain on-site for an extended period, say 40 to 100 years. NRC’s NUREG/CR-7030 states that atmospheric corrosion of sea salt can lead to stress corrosion cracking within 32 and 128 weeks in austenitic [corrosion resistant] stainless steel canisters. How will this corrosion be prevented? Can the canisters be coated to prevent corrosion We do not believe the industry has the experience in transferring failed (damaged) fuel from one cask to another and no procedures for doing this. In fact, no spent fuel bundle, damaged or not, has ever been transferred from one dry cask to another. Since high burnup fuel is more likely to fail sooner in storage, this becomes an even bigger and more urgent problem.
This is not a theoretical problem. Three examples of stress corrosion cracking at San Onofre have already been seen. In the fall of 2009, three examples of chloride-induced SCC which extended through-wall were discovered at the San Onofre Nuclear Generating Station (SONGS) in the weld heat-affected zone (HAZ) of Type 304 stainless steel piping. The piping included 24-inch, Schedule 10 emergency core cooling system (ECCS) suction piping; 6-inch, Schedule 10 alternate boration gravity feed to charging line piping; and an ECCS mini flow return to refueling water storage tank. While the through-wall failures were attributed to chloride-induced SCC, surface pitting was also observed on the surface of the pipes, with a greater concentration in the weld HAZ. All three pipes were exposed to the outside ambient marine atmosphere. Through-wall cracks developed after an estimated 25 years of service….
These are my takeaways on the HBF and storage issue:
• Little technical support for NRC approval of high burnup fuel (HBF). Experiment taking place in the field.
• Total amount of HBF unknown. At a minimum, the NRC should survey utilities.
• HBF will postpone storage up to 20 years; 32 PWR canister extends cooldown period.
• Cladding defects are a major problem for HBF; HBF may not be retrievable. HBF should be canned.
• Because of corrosion, long-term storage may not be possible in a salt environment.
Brittleness is important when considering transportation and disposal. One utility, Maine Yankee, has taken the important step of canning the HBF, that is, individually enclosing each fuel assembly in a stainless steel container. Concern is vibrations when transported, and potential shattering of cladding in a transportation accident. Transportation casks must satisfy regulatory accidents. Casks must withstand 30 foot drop onto an unyielding surface. In a hypothetical transportation accident, cask must withstand an end drop (drop from Holtec rpt) where 140 ton casks are cushioned by impact limiters. But a more serious accident involves a side impact where impact limiters are not present. One example is a RR crossing where a cask could be struck by the sill of a locomotive. (picture from NV rpt). NRC has not carefully evaluated such an accident, including the impact limiters. NRC hypothetical accident requires the cask to withstand a 30 inch drop onto a punch.
Another type of accident involves fire. Several major train fires have occurred recently. 140 ton casks would be shipped by train, on the same routes used by oil tankers. Right now, nuclear fuel has nowhere to go, no final repository. But NRC has not done the statistical analysis to determine the statistical likelihood of a nuclear shipment caught in an oil tanker fire. A study of the likelihood of an accident involving an oil tanker fire and a nuclear shipment requires a sophisticated Monte Carlo analysis. In addition to the likelihood of a long duration fire involving a nuclear cask, the NRC must also analyze the consequences of a radioactive release In my opinion, the NRC has not properly taken into account a long duration fire, by not properly taking into account the conduction of fire heat into the cask interior. As seen, fuel sits within a sealed canister, welded shut. The transportation overpack is metal, but this is surrounded by a neutron absorber, generally boronated, hydrogenated plastic, with an outer metal envelope. (picture of cask crossection). Plastic does not effectively conduct heat, so additional metal pieces serve to transfer heat out of the cask, but also conduct heat into the cask in a fire. Oil fire may burn at 1850 oF or higher depending on the air supply. The hypothetical accident fire consists of an all engulfing fire at 1475 oF for 30 minutes, while an oil fire can burn for many hours. The most recent NRC report NUREG-2125 does not correctly take into account a long duration high temperature fire and should be redone.
Here are my takeways on the transportation issue:
• Realistic low probability, high consequence accidents should be examined.
• Side impact rail accidents may shatter HBF cladding.
• Long duration, high temperature fires may involve oil tankers that travel the same tracks. NRC has not properly quantified the statistical likelihood.