Underground autocatalytic-criticality potential and its implications to weapons fissile- material disposition

Several options for weapons fissile-material disposition, such as once-through mixed- oxide (MOX) fuel in reactors or immobilisation in waste glass, would result in end products requiring geologic disposal. The criticality potential of the fissile end products containing U-235 and Pu-239 and the associated consequences in a geologic setting are important considerations for the final disposal of these materials. The possibility of underground criticality, and especially autocatalytic criticality, is affected by (1) groundwater leaking into a failed waste container, (2) preferential leaching of neutron absorbers or of fissile material from a failed container, and (3) preferential deposition of fissile material in the surrounding rock. Bowman and Venneri have pointed out that fissile material mixed with varying compositions of water and silica can undergo a nuclear chain reaction. Some configurations can become autocatalytically supercritical resulting in considerable energy release, terminated finally by disassembly. Some reviews rejected the Bowman and Venneri warning as implausible because of low probabilities of scenarios that could lead to such configurations. Sanchez et al. reported possible supercritical conditions in systems of Pu-SiO2-H2O and Pu-tuff-H2O but concluded that the probability of forming such combinations is extremely low. Kastenberg et al. studied the potential for autocatalytic criticality of plutonium or highly enriched uranium in the proposed Yucca Mountain geologic repository. They concluded that plutonium or uranium could, theoretically, become supercritical, but that such criticality is unlikely given the hydrology, geology and geochemistry of the Yucca Mountain site. These studies are not definitive. The possibility of criticality exists. Detailed mechanisms have not been sufficiently studied for clear conclusions on the probabilities of occurrence. More technical analysis is needed to understand the potential for underground autocatalytic criticality and the associated consequences. Here we present our calculations of possible conditions and a mechanism for criticality in a container containing fissile material in various different chemical forms, including conditions for autocatalytic criticality. We discuss criteria for designing waste solids for the disposition of weapons fissile material so that long-term dissolution of a neutron-absorber or of fissile material would not cause criticality. These results for simplified geometries outline the conditions for a more detailed analysis of practical systems. The potential for underground criticality is also studied for other fissile waste forms, including spent fuel from commercial reactors, research reactors, and naval reactors, all of which may be destined for geologic disposal. If not carefully resolved on scientific grounds, underground criticality could potentially become the 'Achilles heel' of programmes to dispose of weapons materials and nuclear wastes. 6 refs