1. SCOPE
Comprehensive research has been conducted since the eighties in countries like Sweden, Finland, France and the UK for developing methods for safe disposal of highly radioactive waste [1]. Most of the work has concerned crystalline rock, which provides excellent stability of repositories located at a depth of a few hundred meters but which makes prediction of groundwater flow and contamination difficult because of the high content of fracture zones. This problem has led to reduced faith in the waste-isolating capacity of rock and a tendency of relying on engineered barriers, as exemplified by the present paper. Currently proposed concepts for safe isolation of highly radioactive waste imply construction of deep repositories in crystalline, sedimentary or salt rock. Waste in the form of spent fuel is planned to be encapsulated in metal containers, termed canisters, surrounded by very dense smectite-rich clay, and placed in bored vertical holes or horizontal tunnels. The concepts favoured in Sweden and Finland are of multibarrier type with redundance provided by the host rock, canisters and the clay termed “buffer clay”. Despite the tremendous construction cost the planned isolation has several weaknesses and may not serve acceptably for the required 100,000 years. This matter is dealt with in the paper that describes an alternative concept for long-term safe isolation, implying only one type of effective engineered barrier and a potential to be implemented in abandoned mines or even—at a safe distance—in mines where mining is still going on.
2. PRINCIPLES FOR DISPOSAL OF HIGHLY RADIOACTIVE WASTE (HLW)
2.1. Rock
Because of the risk of erosion and other exogenic processes and the fact that the average hydraulic conductivity of rock decreases with depth there is general consensus that repositories should be located at a depth of at least some hundred meters [1]. Larger depth means that the rock stresses are higher and can cause collapse of deposition holes and tunnels. A risk that is particularly obvious if sedimentary rock is considered for hosting the waste. We will confine ourselves to consider only crystalline rock in this paper.
Structural modeling using the simple concept of plane, parallel fracture zones of different extension and width and assumed bulk hydraulic properties has been performed for calculating the contamination of groundwater that can occur if radionuclides are released from the waste containers [1,2]. The common spacing of major fracture zones, termed 2nd order discontinuities, is 100 - 1000 m, which makes it possible to fit in deposition tunnels and holes so that they will not interact with these major weaknesses. Smaller zones of 3rd order with lengths of 100 to 1000 m and spacings of 10 to 100 m, which are not allowed to intersect canister positions, will not be known until repository construction has begun [3]. Generalized structural models will look like the one in Figure 1.
2.2. Engineered Barriers
2.2.1. Presently Considered Canisters for Spent Fuel
Figure 2 shows the engineered barriers according to most of the proposed concepts for isolating HLW from the groundwater. There is only one criterion for waste containers of HLW, i.e. the canisters, except that they must be constructible and reasonably costly, and that is that they must be tight for a very long period, usually taken as 100,000 years. For this purpose they must not be through-corroded or broken in this period, which puts demands on the chemical integrity and mechanical strength.
Internationally, steel canisters are a primary choice due to their mechanical strength and because the oxygenfree conditions at deep-geological disposal suggest that corrosion will not be significant [1]. In Sweden and Finland canisters consisting of a core of cast iron with room for spent fuel and surrounded by a 50 mm copper lining is favoured [2]. The iron core, separated from the lining by a 2 mm space, provides mechanical strength and the copper liner corrosion protection.
One problem with the iron/copper canisters is their sensitivity to rock strain. They must sustain seismically or tectonically transversal shearing and resist axial tension caused by movements in that direction of the surrounding buffer clay. Transversal shearing indicated in
Figure 1. Simplified and generalized model of candidate site. The green area represents the ground surface and the blue plates 2nd order discontinuities with 100 m width. The red plates are 2nd and 3rd order discontinuities with 50 m width. No repository tunnels and rooms have been marked [3].
Figure 2. Isolation of HLW. The radioactive waste is contained in canisters (inner circle) embedded in “buffer” clay that fills up the deposition hole bored in rock [4].
Figure 3, can take place along a fracture or fracture zone. The Figure refers to Swedish and Finnish canisters with an inner core of cast iron with channels for the spent fuel, surrounded by a copper liner separated from the core. The ductility of the clay “buffer” surrounding the containers reduces the stresses generated by the shearing but calculations have shown that there is a critical condition with respect to shear strain as indicated in the figure.