A disposal system for high-level and long-lived radioactive waste should provide for adequate isolation of radionuclides from the bio-sphere.
During the work of finding such a system, and in demonstrating its long-term isolation, certain unprecedented requirements become apparent: the disposal system must function over very long times without any need for maintenance or repair. Furthermore, no experience from operation can be made available for the analysis, and there will be no results from tests with durations comparable in magnitude to the life-time of the facility.
In the systems studied in Sweden and Finland, copper containers are deposited in crystalline bedrock with bentonite as a backfill. The groundwater is reducing and the flow is very low due to the low permeability of the bentonite clay and of the rock near the deposition hole.
The chemical durability of a container in such a disposal system will depend primarily on three factors: the mechanical tensile stresses in the container, the chemical environment, and the copper material itself. The tensile stresses depend on manufacturing factors as well as external forces. The chemical environment will depend on the composition of the groundwater, theibentonite clay and possibly also phenomena that might take place in the vicinity of the container. The copper material can corrode primarily by oxidation or sulphidation. If the salt concentration is high (and a reducible specie is available), copper may also react by forming chloride complexes.
If the copper container has thick walls, penetration by general corrosion will take considerable time. It is nevertheless of interest to study general corrosion since - at least under certain conditions - it will give rise to a protective and passivating surface layer. Such a layer may be of significance when the possibilities of localized corrosion and stress corrosion cracking are to be evaluated. It is pointed out that the development of localized corrosion is strongly dependent on the specific chemistry. Concern is expressed regarding certain treacherous aspects of stress corrosion cracking. Areas of research that may be fertile to pursue have also been identified.
A point is made that the safety analyses and scenarios should include both probable and less probable developments. Examples are given of the latter type.
A suggestion is made that samples of the container and of container material be left in the repository to enable evaluation and thus to facilitate improvements that future generations might wish to make.