The 103 nuclear power plants operating in the USA contain massive amounts of radioactive material in their reactor cores. In addition, nuclear plants have discharged more than 43,000 tonnes of spent fuel, representing an amount of long-lived radioactive material that substantially exceeds the amount in the reactor cores. Cumulative national production of spent fuel is likely to exceed 80,000 tonnes over the currently-licensed lifetimes of existing nuclear plants. Most of the spent fuel is now stored in high-density spent-fuel pools adjacent to reactors, and the nuclear industry intends to continue using these pools. As pools become full, independent spent fuel storage installations (ISFSIs) are being built to accommodate the growing inventory of spent fuel. Present and proposed ISFSIs are generally at reactor sites, but away-from-reactor ISFSIs may be established at Skull Valley, Utah, and elsewhere. In the USA, ISFSIs store spent fuel under dry conditions inside storage modules that are arrayed on concrete pads in the open air.

This situation poses a very high risk, because the loss of water from a high-density pool will cause spent fuel in the pool to heat up, self-ignite, burn and release a huge amount of radioactive material to the atmosphere. Water could be lost from a pool by evaporation, displacement, siphoning, pumping or through a breach. These mechanisms could be exploited in various ways by knowledgeable and determined attackers, who could thereby contaminate large areas of US territory with radioactive material. Nuclear reactors are also vulnerable to attack. ISFSI modules have safety advantages in comparison to pools and reactors, but are not designed to resist a determined attack.

Thus, nuclear power plants and their spent fuel can be regarded as pre-deployed radiological weapons that await activation by an enemy. The US government seems to be unaware of this threat. Responsibility for overseeing the security of civilian nuclear facilities has been delegated to the US Nuclear Regulatory Commission (NRC). This agency has a longstanding policy of not requiring its licensees to protect their facilities against enemy attack, and has continued this policy with little change since the terrorist attacks of September 2001. As a result, US nuclear facilities are lightly defended. This situation is symptomatic of an unbalanced US strategy for national security, in which offensive capabilities are assigned a higher priority than homeland defense. The lack of balance is a potentially destabilizing factor in the current international environment, and also exposes US citizens to the risk that an enemy will create widespread radioactive contamination.

This report offers a way forward in an important area of national defense. The report articulates a strategy for providing robust storage of US spent fuel, where the word "robust" means that a facility for storing spent fuel is designed so as to be resistant to attack. Implementation of robust storage will be needed whether or not a repository is opened at Yucca Mountain, Nevada. The proposed robust-storage strategy can and should be implemented as a major element of a defense-in-depth strategy for US civilian nuclear facilities. In turn, that defense-in-depth strategy should be a component of a homeland-security strategy that provides solid protection of our critical infrastructure.

The highest priority in a robust-storage strategy for spent fuel should be to re-equip spent fuel pools with low-density, open-frame racks, as was the case when the present generation of nuclear plants began operating. This step would prevent fuel from igniting and burning if water were lost from a pool. Fuel that can no longer be accommodated in the pools would be stored in ISFSIs. As a further measure of risk reduction, ISFSIs should be re-designed to incorporate hardening and dispersal. "Hardening" means that each fuel-storage module would be shielded from instruments of attack by layers of concrete, steel, gravel or other materials. "Dispersal" means that fuel-storage modules would not be concentrated at one location, but would be spread more uniformly across a site.

Hardening and dispersal of ISFSIs should not be conducted in a manner that encourages society to extend the life of an ISFSI until it becomes, by default, a repository. Therefore, a hardened ISFSI should not, unless absolutely necessary, be built underground. Also, the cost of implementing hardening and dispersal should be minimized, consistent with meeting performance objectives, and the timeframe for implementation should be similarly minimized. These considerations argue for the use, if possible, of dry-storage modules that are already approved by the NRC and are in common use.

The design of a hardened, dispersed ISFSI would be governed by a design-basis threat (DBT). This report articulates a two-tiered DBT. The first tier requires high confidence that no more than a small release of radioactive material would occur in the event of a direct attack on the ISFSI by a TOW missile, a manually-placed charge, a vehicle bomb, an explosive-laden general-aviation aircraft or a fuel-laden commercial aircraft. The second tier requires reasonable confidence that no more than a specified release of radioactive material would occur in the event of a ground burst of a 10-kilotonne nuclear weapon at the ISFSI.

An ISFSI design approach that offers a prospect of meeting this DBT involves an array of vertical-axis dry-storage modules at a center-to-center spacing of perhaps 25 meters. Each module would be on a concrete pad slightly above ground level, and would be surrounded by a concentric tube surmounted by a cap, both being made of steel and concrete. This tube would be backed up by a conical mound made of earth, gravel and rocks. Channels for air cooling would be inclined, to prevent pooling of jet fuel, and would be configured to preclude line-of-sight access to the dry-storage module.

An alternative design approach could be used at a few sites where space is insufficient to allow wide dispersal. In this approach, a number of dry-storage modules would be co-located in an underground, reinforced-concrete bunker. Similar bunkers would be dispersed across the site to the extent allowed by the site's geography. At especially-constricted sites, it might be necessary to ship some spent fuel from the site to an ISFSI elsewhere.

Three major requirements must be met if a robust-storage strategy for spent fuel is to be implemented nationwide. First, full-scale experiments are needed to determine the ability of various dry-storage design approaches to accommodate various DBTs. Second, performance-based specifications for dry storage must be developed with stakeholder input. Third, robust storage for spent fuel must be seen as an important component of homeland security, to ensure that sufficient funding is available and robust storage is implemented quickly.

Dr. Thompson is the executive director of the Institute for Resource and Security Studies, Cambridge, MA and a Research Professor at Clark University, Worcester, MA. He received his Doctorate in Applied Mathematics from Oxford University, UK, in 1973. Dr. Thompson has extensive experience in assessing the safety and security hazards associated with nuclear facilities, and in identifying alternative designs and modes of operation that can reduce a facility's hazard potential.