Full excavation of a water-conducting shear zone used for radionuclide migration experiments is part of a comprehensive study to physically determine the sites of in-situ radionuclide retardation. Prior to the excavation, multi-tracer solutions containing strongly sorbing y- and αemitting nuclides were injected into the shear zone and the water extracted was analyzed on-line using in situ, high-resolution y-spectrometry. In addition, water samples were collected to compare quantitatively the on-line data with subsequent laboratory measurements α- and γ- spectrometry) and also to analyze samples with activities below the detection limits of the in-situ analysis. The breakthrough curves indicate that Se had passed through the fracture with a velocity comparable to a conservatively behaving dye (uranine), but very strong retardation of Co. Eu and Sn was observed. The recoveries of these nuclides were found to be in the order of only a few percent when the dipole flow-field was interrupted about one month after radionuclide injection and the fracture was sealed off for excavation. Significant retardation was also obtained for the oxemitters U, Np and Pa. Almost identical breakthrough curves for these actinides were measured, suggesting a similar spatial distribution in the fracture at any given time.

For several years, tracer migration experiments at Nagra's Grimsel Test Site have been performed with an exceptionally high degree of precision. A summary of the main findings obtained by modelling these experiments is presented, addressing, in particular the influence of sorption. A consistent description of tracer behaviour is achieved, giving confidence when applying broadly similar methodologies and models in geosphere performance assessment.

Quantitative autoradiography for the investigation of rock matrices has been further developed by use of tritium-labeled polymethylmethacrylate (3H-PMMA), the use of thin sections and image analysis functions allowing correlations of mineralogic-petrographical and analytical or porosity information. The autoradiographic system was studied experimentally and theoretically. The autoradiography could achieve a resolution down to 10 and 20 μm for 3H and 14C, respectively. Possible candidates for intact, low-porous rock standard materials were studied and the measurement of their transport properties discussed. Porosities and diffusivities measured in rock types dominated by very small pores were depending on the probe molecule used. The effect was most pronounced in flint which gave low porosities for MMA, Hg and He and no diffusivities at all for He.

The channels of polonium nuclide formation in lead-bismuth target irradiated by high energy protons and secondary neutrons are considered. Under off blanket condition the activity of 200po -210Po, as well as the fission and spallation product activity are calculated for 15 MW target after 1 year operation. Polonium separation methods, including removal using alkaline extraction, vacuum sublimation, and so-called carrier method are analyzed. The engineering design for experimental apparatus to test polonium removal and containment using alkaline extraction using tellurium as a chemical analog of polonium is developed. The necessary degree of polonium removal to keep concentration below levels appropriate to meet the US and Russia safety standards is estimated. Comparative analysis of alkaline extraction and vacuum sublimation techniques from the viewpoint to achieve the necessary removal depth is carried out.

Lead, the final decay product of uranium, is found in natural uranium ore deposits. The isotope composition of lead in uranium-bearing minerals reflects their age and the migration behavior of lead. Secondary Ion Mass Spectroscopy (SIMS) can be used to analyze the isotopic composition of minerals. SIMS analysis of lead contained in the Koongarra uranium deposit, Australia, revealed that uraninite and uranyl minerals with different chemical composition in the primary ore region have homogeneous lead isotope, from the highest grade area to a uranyl silicate zone at a distance of 6.1m. Uranyl minerals, which have good crystallinity and retain stoichiometric composition, have “old” lead isotope composition identical to that of uraninite. Uranyl minerals keep exchanging lead with ground water in the primary ore region through metamictization and recrystallization by alpha-decay damage. Sulfide minerals just outside the primary ore body contain only radiogenic lead with an isotopic composition different from that of uranium minerals. Lead that migrated from the primary ore body formed sulfide minerals in a specific geological event. Since then, lead produced from uranium decay has been retained in the primary ore body, in spite of recrystallization of uranyl minerals and exchange and homogenization of lead isotopes. Uranium minerals may retain minor elements despite intense alpha-decay dose.

The possibility of furnace free vitrification of ash residue was considered by using heat generation by means of exothermal chemical reactions which occur in the mixture of heat generating materials (heating batch), glass forming batch and ash residue.

A computer programme (FANFARE) for facilitating scenario analysis and conceptual model development is discussed. FANFARE implements a systematic methodology, based on the use of Directed Diagrams, that makes the assessment of comprehensiveness of FEPs and scenarios a tractable problem. Coupled with its ability to store a traceable audit trail of judgments and decisions, this makes FANFARE a useful tool for carrying out FEP analysis, scenario analysis and conceptual model development for performance assessments of radiological disposal facilities. It can also be applied to other systems that require consideration of the underlying phenomena to determine system performance.

To determine the potential effects of alkaline plume migration on the near-field environment of the proposed high-level radioactive waste geologic repository at Yucca Mountain, Nevada, calculations are conducted simulating interactions between cement and tuff with pure diffusive transport of solute species. The calculations used the reactive transport submodule GEM of the computer code MULTIFLO [6]. The results suggest that strong alteration of the tuff host rock and of cement in contact with the tuff could result from these interactions. Porosity reduction within the tuff could isolate the matrix from fracture pore water. The model calculations predict calcification of the cement as would be expected. In simulations involving counter-diffusive transport across the cement-tuff contact, calcification is more pronounced in a partially-saturated environment compared to a fully-saturated one.

Calculations of risk or dose from the groundwater flow pathway are a central part of the assessment of post closure performance for a deep repository system for radioactive waste. These calculations need to be supported by a knowledge of the various components of the disposal system. Site characterisation provides a site specific geological database which can be used to build, test and refine numerical models of the geosphere and, to some extent, the biosphere components of the disposal system in order to evaluate its future evolution.

Site characterisation programmes, and groundwater flow modelling within the performance assessment need to progress in an interactive manner to ensure that both remain focused on the resolution of key uncertainties which impact the overall performance assessment. The data derived from site investigations can be classified in terms of their application within the performance assessment: establishing the framework for numerical models; enabling the parameterisation of numerical models; testing model output; refining models; and demonstrating their credibility. Site characterisation programmes need to be established to provide the appropriate level of information for each of these stages. It is therefore not particularly useful to classify information into the traditional disciplines of geology, hydrogeology and geochemistry.

A key issue for developing a geological and hydrogeological database for performance assessment is determining when sufficient information has been obtained. Primarily, this is dictated by the needs and sensitivities of the performance assessment and any regulatory requirements. However, radioactive waste disposal agencies work in a political environment that is strongly influenced by public perception. In order to progress, any programme must attain a broad measure of credibility and acceptability to other key “stakeholders”.

The paper draws on examples and experiences from the extensive programme of work which has been performed by Nirex to understand the characteristics of a site near Sellafield in Cumbria, and to develop groundwater flow models to support the assessment of the post closure performance of a potential repository located at that site. Uncertainties important in influencing calculated risk within an interim assessment, published as the “Nirex 95 assessment” were used to develop further site-specific investigations. The resulting site characterisation activities generated an enhanced and refined site-specific geological and hydrogeological database for a subsequent assessment - the “Nirex 97 assessment”.

Constraints on oxidation phenomena and processes in Opalinus Clay are derived from studies at three localities: the 140-year old Hauenstein tunnel, the six-year old Mt. Terri tunnel, and an open clay pit at Siblingen that documents processes active over thousands of years. The excavationdisturbed zone (EDZ) in both tunnels is characterized by an extensive fracture network that extends to at least 1.6 m. The network consists of saturated fractures as well as some air-filled fractures with apertures up to several mm that contain effervescent gypsum. At Hauenstein, the fracture network displays regularly developed brownish oxidation zones extending 3-15 mm into matrix rock. Such oxidation zones are absent at Mt. Terri. Siblingen displays a remarkable decrease of both fracturing and oxidation phenomena with depth below surface.

Microscopic observations, mass balance constraints, and estimated diffusion coefficients of oxygen indicate a minor extent of pyrite oxidation at Hauenstein possibly controlled by diffusionlimited oxygen transport into water saturated rock. Oxidation of siderite is quantitatively dominating over pyrite oxidation at Siblingen. There is evidence at Hauenstein and Siblingen that rock alteration penetrated deeper than the visible alteration zone.

During the operational phase (20-100 years) of a deep repository at least the major EDZ fractures are expected to become unsaturated, and gypsum to be precipitated. Fracture apertures reach locally several millimeters within a few years, which enhances hydraulic and gas conductivities. Oxidation rims bordering fractures in the EDZ develop over timescales of 10-100 years, but mass transfers are quantitatively minor.

Gas threshold pressure tests and gas tracer tests have been performed at the Grimsel Test Site to study two-phase flow processes in a shear zone. In addition, capillary pressure and gas permeability measurements were carried out in the laboratory on drillcore samples. The laboratory investigations were complemented by assessing the pore structure of the shear zone material. The interpretation of the field tests with numerical models indicated that the structural and two-phase flow parameters to be determined are highly correlated with one another and, consequently, the parameter estimates can be rather uncertain. The joint interpretation of field and laboratory results, however, led to a more stringent description of the two-phase flow processes, expressed by a better overall fit of the test data and smaller uncertainty ranges of the estimated parameters. The results showed that the gas mobility in the shear zone was very high even at high water saturation and gas flow was limited to the narrow zones of brittle deformation along the shear zone.

The Wellenberg K-model describes the distribution of the hydraulic conductivity of the host rock at Nagra's proposed site for a LLW/ILW repository. The description takes the form of the expected value of hydraulic conductivity at a location and the uncertainty around this expected value. The development of the K-model is based on geostatistical concepts and, in particular, on kriging and conditional simulation methods. The model is built of cubes with sides of 100 m. Upscaling of the transmissivity profiles at the borehole locations to effective K-values on the 100 m scale is realised by fracture network modelling. The K-model accounts explicitly for the hydraulic conductivities estimated at each borehole. It is consistent with the hydrogeological conceptualization of the host rock and generates realistic hydraulic conductivity distributions with a degree of variability similar to that observed in the data.

Batch sorption experiments have been used to assess the sorption behaviour of four radionuclides, important in the context of low-level waste disposal, on a glacial substrate. Data for sorption of 238U, 237Np, 134Cs and 85Sr are compared and agree well with independent studies. A series of well-established kinetic models have been used to describe the individual uptake mechanisms and rate parameters reported. Sorption occurs via both equilibrium and kinetically controlled pathways, with neptunium sorption being under kinetic control to the greatest extent.

The application of thermodynamic models to calculate Kd values for Sr/Ra and Pb on bentonite is demonstrated. The model discussed here can be used to identify relevant sorption processes, as well as other factors that determine Kd values, such as impurities contained in bentonite or artifacts (precipitation); i.e. it can be used to conduct sensitivity analyses of Kd values. Kd values can be modeled over a very wide range, indicating that such models may also be used for an evaluation of experimental data, and for predictive purposes where no such data are available.

The solubility of Ni, Sn, Pb, Eu, and Sr was investigated at 22°C and 50°C in simulated cement porewaters corresponding to fresh cement (‘FPW’, pH 13.2) and cement leached of soluble alkalis, whose porewater chemistry is dominated by equilibrium with portlandite (‘APW’, pH 12.5). Solubility limits were approached from undersaturation, using Ni(OH)2(c), SnO2(c) (cassiterite), PbO(red), and Eu(OH)3(c) as solids. For Sr, no solubility experiments were carried out since the cement contains enough Sr to reach equilibrium with strontianite and celestite. Solubilities generally increase with increasing pH and temperature. The measured solubility of Pb ranges from about 4× 10−3 to 1×1O−2M and is within thermodynamically predicted values. The measured solubility of Ni is <O.5-1.3×10−7M in APW and about 3-4×10−7M in FPW, which is lower than predicted by most thermodynamic data. The measured solubility of Eu ranges from <6.6×10−10M in APW to 1.8-2.7×10−9M in FPW, which agrees well with thermodynamic calculations. The solubility of cassiterite is low, with dissolved Sn concentrations ranging from 3.5× 10−8M in APW to 3.3× 10−6M in FPW. Thermodynamic calculations predict much higher solubilities, presumably for less crystalline phases. However, oversaturation measurements in APW reproduced the solubility limits obtained by undersaturation within a factor of two. To our knowledge, no relevant thermodynamic data are available.

The conditions that are needed to achieve criticality in a high level waste repository for spent nuclear reactor fuel are reviewed. The effect of initial enrichment of the fuel, burnup, and of mixed oxide fuels on the conditions for criticality are discussed. The situations that produced criticality at Oklo, Gabon, 2000 million years ago are summarized. A model based on the Oklo conditions is presented for estimating the amount of fissile material that must be assembled to create a critical mass in typical granitic rocks. Mechanisms for movement of uranium and plutonium to achieve a critical configuration are discussed and compared to the conditions that are likely to occur in a repository in granite. The sequences of events needed to produce a critical assemblage are shown to be in conflict with the conditions expected in the repository and, in some cases, to require internally inconsistent assumptions to produce the postulated sequence of events. No credible scenario for achieving criticality in a high level waste repository has been found.

The chemical evolution of groundwater in the Palfris marl at Wellenberg has been simulated using a reactive transport model. The results were tested against the chemical and stable carbon isotopic composition of water samples from exploratory boreholes. Water chemistry is constrained by mineral and cation exchange equilibria. To reproduce measured CO2(tot) and δ3C values requires H+ ion exchange. Matching measured ratios between Cl– and other dissolved constituents constrains the relative amounts of reacting water and rock to porosities between I and 10%. NaHCO3 waters sampled from the Palfris are formed by replacement of the initial Na-Cl water by one to five pore volumes of infiltrating Ca-HCO3 recharge water. To entirely exhaust the exchange capacity of the formation so that Ca-HCO3 water persists requires several hundred to several thousand pore volumes of flow. The agreement between model results and measured water chemistry demonstrates a quantitative understanding of the geochemical processes controlling the chemistry of water naturally present in the Palfris marl. These processes will also determine the behaviour of material that might emanate from a repository. In addition, the modelling provides water flow information of use in testing groundwater flow models.

The input data required for transport models for use in long-term risk assessments of repositories for radioactive waste, in geological media, are intrinsic to the performance of the models. The flow parameters utilized in these models typically come from 2 or 3D hydro-geological calculations done in a prior phase of an assessment. This paper examines some of the standard simplifications introduced when hydrogeological data are reduced to ID as is often required for radionuclide transport models. Two key aspects of data reduction are the determination of average properties of fractured media between and along transport pathways. To quantify possible errors associated with these reduction procedures, two computer experiments have been done. We show that the use of effective flow parameters, representing the average properties of a set of independent transport pathways, in a 1 D radionuclide transport model can result in an underestimation of peak releases by one order of magnitude or more. This result is valid for short-lived nuclides whenever retardation is an important factor. On the other hand, averaging of spatially varying transport properties along a transport pathway may lead to unjustified conservatism. A simple calculation example using Monte Carlo technique, shows that a model that does not take into account spatially varying retardation properties along the transport pathways may overestimate peak release rates by several orders of magnitude. We conclude that more sophisticated transport models taking into account available hydrogeological information on spatial variability are needed to fully understand the potential errors associated with consequence calculations in the performance assessment

An experiment degrading a porous cement disc by pure water was modelled by a reactive transport model including chemical reactions, mineral precipitation/dissolution reactions and changing hydraulic properties of the cement. For the early cement degradation or high pH phase, an equilibrium sorption approach is used for the release of alkali metals. Porosity changes of the porous cement disc during degradation induce conductivity changes that are taken into account in the modelling by coupling back transport to geochemical parameters.

The excavation disturbed zone (EDZ) is defined as the rock zone where rock properties and conditions have been changed due to the processes induced by excavation, such as excavation damage, stress redistribution and desaturation. In-situ excavation disturbance experiment has been performed to determine the rock properties and width of the EDZ at the Kamaishi and Tono mines in Japan. Rock mass fails when blasting vibration velocity exceeds limit velocity so that we can estimate excavation damaged zone from vibration measurement. The width of the zone where blasting vibration velocity exceeded the limit velocity is roughly consistent with the width of low seismic velocity layer detected from the seismic refraction survey at both mines.