The nature of diffusivity and porosity in rock was studied as a function of various parameters. The phenomena of main interest were dead-end porosity, ion-exclusion and sorption. The rock types studied were rapakivi granite, granite and gneiss, and tracer techniques with 36Cl, 22Na+ and 3H (HTO) were used as a research method. A mathematical solution for outdiffusion from a porous cylinder was developed by applying a corrected form of Fick's second law for a case where part of the pores are so-called dead-end pores. With this model the theoretical curve could be closely fitted to the measured values. It was found that the rock-capacity factor is an increasing function of the ionic concentration of the solution in the case of Cl indicating ion-exclusion, while the opposite is true in the case of Na+ indicating ion-exchange type sorption. The effective diffusion coefficient was also found to vary as a function of the salinity in the case of 36Cl. In the case of 22Na, the effect was opposite and weaker. The diffusion of tritium through the rock samples was clearly higher than the diffusion of 36Cl. Part of the difference is explained by the smaller effective porosity for 36Cl. The rest can probably be explained by the steric effects on the chloride ion caused by the negatively charged pore surfaces in the narrow pores.

A numerical simulator of reactive chemical transport with coupling from precipitation-dissolution reactions to fluid flow, via changes of porosity and permeability, is applied to precipitation-dissolution of quartz and calcite in spatially and temporally variable fields of temperature. Significant effects on fluid flow are found in the quartz-silicic acid system in the presence of a persistent, strong gradient of temperature. Transient heat flow in the quartz-silicic acid system and in a calcite-calcium ion-carbonato species system produces vanishingly small effects on fluid flow.

A model is proposed to describe flow and transport in fractured rocks. It is based on the concept of a network of channels. This approach is backed by observations in drifts and tunnels that flow in fractured rocks takes place in sparse narrow channels with widths typically less than 10 cm and a channel frequency of one channel per a few square meters to one channel per more than a hundred square meters. Observations in boreholes also indicate that there are large distances, tens to hundreds of meters, between the most conductive sections in boreholes.

For visualization purposes our model is displayed on a rectangular grid. The individual channels are given stochastically selected conductances and volumes. Flowrate calculations have been performed in grids of sizes 20*20*20 channels in most cases but larger grids have also been used. For large standard.deviations in conductances, greater than 1.6 in the log normal distribution (base 10), channeling becomes pronounced with most of the water flowing in a few paths. The effluent patterns and flowrate distributions obtained in the simulations have been compared to three different field measurements in drifts and tunnels. Standard deviations of channels conductances were found to be between 1.6 and 2.4 or more in some cases. Channel lengths were found to vary between 1.2 m and 10.2 m in the different sites. In one site where detailed borehole measurements were available the channel length could be assessed independently and was found to be 1.2 m as compared to the 1.7 m obtained from the drift inflow measurements.

A particle tracking technique was used to simulate solute transport in the network. Nonsorbing as well as sorbing tracer transport can be simulated and by a special technique also tracers which diffuse into the rock matrix can be simulated.

Tracer measurements in one site, Stripa, were used to compare dispersivities. These were found to be large, having Peclet numbers less than 5 both in simulations and the field results. From the Stripa tracer data it was also found that the tracers were taken up into the rock matrix by molecular diffusion. The surface area needed for this uptake was estimated to be between 0.2-20 m2/m3 for different tracers. The wetted surface for the model estimated from flowrate distribution data indicate a wetted surface of 0.2- 0.4 m2/m3.

Radionuclides from a damaged canister for spent fuel will leak out through a damage in the canister wall and spread through the surrounding backfill. They will further migrate into water bearing fractures in the rock, up through the backfill into the damaged zone around the drift and into the drift itself. Some substance may also diffuse through the rock to adjacent fracture zones. Underway the nuclides will sorb on the materials along the transport paths. This very complex and variable transport geometry has been modelled using a compartment model which is based on simplifying a full 3 dimensional integrated finite difference model. The simplifications are supplemented by introducing analytical and semianalytical solutions at sensitive locations such as entrances and exits from damages and fractures and in the flowing water. The model has been tested against full 3D solutions with good results. Sample calculations are presented and discussed for a nuclide with the chemical properties of Pu-239.

The formation and impact of precipitation fronts on the diffusional mass transport of radionuclides from a high-level nuclear waste canister through a bentonite buffer has been investigated in a series of numerical simulations. The precipitation fronts arise from chain decay and ingrowth, coupled with differences in elemental solubility limits and sorption properties. The fronts influence particularly the behavior of uranium, plutonium and neptunium isotopes. The isotopic concentration profiles across the buffer differ considerably from results obtained with models that employ elemental solubility limits simply as a boundary condition at the waste-bentonite interface.

A large scale tracer test has been performed as a part of the Site Characterization and Validation (SCV) Project in the Stripa experimental mine. Tracers were injected in a fracture zone intersecting a newly excavated drift and collected in plastic sheets and sump holes. The experiment gave information on the variability of flow and transport properties in a fracture zone.

This paper reviews the current status of the Grimsel radionuclide migration experiment, with especial reference to the advances of the last four years. Details are provided on the unusually extensive support programme and the general philosophy of undertaking such a large and costly field experiment.

A series of batch laboratory experiments (‘black boxes’) were set up to study the gross effects of microbial activity on repository geochemistry, radionuclide sorption and the integrity of repository and host rock materials in a Swiss type B repository. The observed principal chemical exchanges and precipitations were confirmed by modelling and could be interpreted by excluding microbiological effects. However, mineralogical studies showed steel corrosion to be localised in deep pits with microbiology playing a possible role. Talc was precipitated in all of the cells which lowered ambient pH through removal of OH- causing dissolution of CSH compounds. This has implications for the long term stability of cements. Microbiology influenced far-field radiochemistry experiments in which added microbes increased Cs sorption particularly under anaerobic conditions.

Recently we proposed [1] that high-level nuclear waste be deposited in very deep boreholes, filled with a substance of low melting temperature, of high density, not generating gases and water-insoluble.

The first paper outlined nuclear waste disposal in deep boreholes, discussions of physical and chemical processes in a filled borehole and preliminarily comments on the ecological safety of the disposal. The given paper covers some economical aspects of the new proposal as well as the principal steps of implementation of the R&D program.