Radionuclide migration in crystalline rock fractures

Crystalline rock has been considered as a host medium for the repository of high radioactive spent nuclear fuel in Finland. The geosphere will act as an ultimate barrier retarding the migration of radionuclides to the biosphere if they are released through the technical barriers. Radionuclide transport is assumed to take place along watercarrying fractures, and retardation will occur both in the fracture and within the rock matrix. To be able to predict the transport and retardation of radionuclides in rock fractures and rock matrices, it is essential to understand the different phenomena involved. Matrix diffusion has been indicated to be an important mechanism, which will retard the transport of radionuclides in rock fractures. Both dispersion and matrix diffusion are processes, which can have similar influences on solute breakthrough curves in fractured crystalline rock. In this work, the migration of radionuclides in crystalline rock fractures was studied by means of laboratory scale column methods. The purpose of the research was to gain a better understanding of various phenomena - particularly matrix diffusion - affecting the transport and retardation behaviour of radionuclides in fracture flow. Interaction between radionuclides and the rock matrix was measured in order to test the compatibility of experimental retardation parameters and transport models used in assessing the safety of underground repositories for spent nuclear fuel. Rock samples of mica gneiss and of unaltered, moderately altered and strongly altered tonalite represented different rock features and porosities offering the possibility to determine experimental boundary limit values for parameters describing both the transport and retardation of radionuclides and rock matrix properties. The dominant matrix diffusion behaviour was demonstrated in porous ceramic column and gas diffusion experiments. Demonstration of the effects of matrix diffusion in crystalline rock fracture succeeded for the first time in a series of experiments where the experimental arrangements enabled very low water flow rates. FTRANS, a modified numerical code that can simulate both groundwater flow and the transport of radionuclides in porous or fractured medium, was tested in order to interpret laboratory scale migration experiments. The code was able to interpret in-diffusion of calcium into altered tonalites. The elution curves of calcium for the altered tonalite fracture columns were explained adequately by the code when using parameters obtained from in-diffusion calculations. The Kd-values for intact rock obtained on the basis of the fracture column experiments were one order of magnitude lower than the Kd-values for crushed rock, indicating that batch experiments overestimate the retardation of sorbing radionuclides onto the rock matrices owing to the larger surface areas that are available. The greater sorption on altered tonalites was explained by the composition of the sorptive alteration minerals and the large specific surface areas. In this research, sodium, calcium and strontium were used as tracers in order to compare various experimental techniques. Experiments with relevant nuclear waste nuclides are needed to enable more detailed discussion of the differences in Kd-values and the implication for performance assessment calculations. Data obtained from transport experiments conducted in well-defined laboratory conditions is the basis for the block scale and field experiments that are necessary to validate the radionuclide transport concept and to test the transferability of laboratory data to field conditions.(orig.)