[en] This presentation provides an overview of the workflow and methodology of the consequence analyses supporting the safety-based comparison of the siting regions. The consequence analyses are performed based on a reference scenario, which represents the expected evolution of the repository system. The radionuclide (RN) transport in both the aqueous and the gas phase is modelled based on two types of numerical models: a) a 1-D model for the RN transport in aqueous phase, and b) a so-called 3-D macro element modelling approach is used to simulate the gaseous C-14 migration by considering two phase flow processes. Both modelling approaches are processed in the same workflow. As the first step of the analyses, a conceptual model of each site is developed based on site specific geological information and data, such as the thicknesses and transport-relevant properties of the various geological layers. This provides a common foundation for the development of site-specific models of radionuclide retention and transport in the geosphere both in the aqueous phase and in the gas phase. The two models are then applied separately in carrying out the following steps: 1. Radionuclide inventory screening to identify safety-relevant radionuclides for each model. For the gaseous radionuclide migration, C-14 is the only radionuclide considered. Other gaseous radionuclides are either highly soluble (e.g., I-129) or are short-lived and/or present in only minor amounts (e.g., Rn-222). 2. Setup of the reference calculation using reference parameters (most likely or conservative assumptions). The objective of this calculation is to obtain the reference dose rate and to develop an understanding of the general behaviour of the investigated site. 3. Define probability density functions (PDFs) for uncertain model parameters using all available data. 4. Perform a Monte-Carlo simulation, sampling each uncertain parameter randomly from its respective PDF, and analyse the resulting uncertainty in the calculated maximum dose rate. Finally, the results from all the simulation realisations are aggregated and analysed. 5. Based on the results of the Monte-Carlo simulation, a global sensitivity analysis is performed to identify the sensitive parameters of the RN transport modelling. The maximum dose rate from the Monte-Carlo simulations within the period under consideration is taken as the quantity of interest for parameter sensitivity analysis, e.g., the maximum total dose rate from the geosphere within one million years for the HLW repository (SF and HLW) and within 100'000 years for the L/ILW repository. In order to ensure that all sensitive parameters are taken into account in the sensitivity analysis, several sensitivity measures are used in a systematic manner to determine the parameter importance. 6. Based on the parameter sensitivities, a set of deterministic calculation cases is defined and analysed. The values of the most sensitive parameters are assigned by taking a bounding value on the pessimistic side of their respective distributions individually in each deterministic case (according to the parameter sensitivities obtained in the sensitivity analysis). There is also an additional deterministic case for each site-specific model, combining the pessimistic bounding values of the sensitive parameters. 7. The results of the probabilistic uncertainty analysis and the deterministic calculation cases are combined and compared to assess the safety margins available for each siting region

[en] Management of spent fuel (SF), vitrified high-level waste (HLW) and long-lived intermediate-level waste (ILW) is based on the concept of deep geological disposal, namely long-term, effective isolation of the waste in suitable deep rock formations. The first project studies carried out by Nagra in this respect already lie more than 20 years in the past, when disposal in the crystalline basement and in clay was considered. The strategy developed by Nagra over the years agrees well with the concept of 'monitored long-term geological disposal' as formulated by (EKRA 2000) and contained in the new Nuclear Energy Act of 2003. This paper provides an overview of the concept for facilities and operation of a deep geological repository for SF/HLW/ILW, as prepared for the 'Entsorgungsnachweis' project, together with a geological synthesis report for the Zuercher Weinland and a report on long-term safety. The facilities and operation concept look at the feasibility of constructing a repository in the Opalinus Clay of the Zuercher Weinland. It also provides project-specific input for analysing and demonstrating the long-term safety of such a repository. The individual structural elements and facility components for which the feasibility study was conducted are brought together as a modular system to form a stand-alone reference project. They can be adapted later to meet local features and requirements. The main focus of the paper shall be on selected system elements concerning design, layout and operational aspects including operational safety. (author)

[en] Document available in extended abstract form only. An in-situ test in the Opalinus Clay formation, termed pore water Chemistry (PC) experiment, was run for a period of five years. It was based on the concept of diffusive equilibration whereby traced water with a composition close to that expected in the formation was continuously circulated and monitored in a packed off borehole. The main original focus was to obtain reliable data on the pH/pCO₂ of the pore water, but because of unexpected microbially- induced redox reactions, the objective was then changed to elucidate the biogeochemical processes happening in the borehole and to understand their impact on pH/pCO₂ and pH in the low permeability clay formation. The biologically perturbed chemical evolution of the PC experiment was simulated with reactive transport models. The aim of this modelling exercise was to develop a 'minimal-' model able to reproduce the chemical evolution of the PC experiment, i.e. the chemical evolution of solute inorganic and organic compounds (organic carbon, dissolved inorganic carbon etc...) that are coupled with each other through the simultaneous occurrence of biological transformation of solute or solid compounds, in-diffusion and out-diffusion of solute species and precipitation/dissolution of minerals (in the borehole and in the formation). An accurate description of the initial chemical conditions in the surrounding formation together with simplified kinetics rule mimicking the different phases of bacterial activities allowed reproducing the evolution of all main measured parameters (e.g. pH, TOC). Analyses from the overcoring and these simulations evidence the high buffer capacity of Opalinus clay regarding chemical perturbations due to bacterial activity. This pH buffering capacity is mainly attributed to the carbonate system as well as to the clay surfaces reactivity. Glycerol leaching from the pH-electrode might be the primary organic source responsible for bacterial growth and the subsequent perturbation in the borehole. Thanks to an accurate description of the pristine boundary conditions this modelling exercise allows now for the first time a better constrain of the chemical controls of the major cations and anions and though enables a realistic simulation of a biological perturbation. (authors)

[en] Document available in extended abstract form only. Indurated argillaceous formations are investigated in a number of countries (e.g. France, Belgium, Switzerland) as host rocks for radioactive waste disposal. Such formations are characterized by very low permeability and thus may retard the migration of radionuclides for very long time periods. The Mont Terri underground rock laboratory in the northwest of Switzerland offers the unique opportunity to study radionuclide migration in the Opalinus Clay. In the DI-A2 experiment, several non-reactive and reactive tracers were injected as a pulse in a packed-off borehole. Unlike the previous DI-A1 test, the design of the Teflon filter in the injection borehole forced the water to flow through the filter and the open space between the filter and the borehole wall (the filter itself did not act as a diffusion barrier between the circulating solution and the rock). The decrease in tracer concentration in the liquid phase was monitored during a period of one year. Afterwards, the borehole section was over-cored and the tracer profiles in the rock were analyzed. A main interest of this experiment was to understand the chemical behavior of sorbing tracers: Cs⁺ (stable), ⁸⁵Sr²⁺, ⁶⁰Co²⁺ and Eu³⁺ (stable). The complete dataset (except for Eu³⁺ because of strong sorption to experimental equipment) was analyzed with a 2D diffusion-reaction model and the derived diffusion and sorption parameters were compared with laboratory data. As in DI-A1, a difference by a factor of about 2 was obtained for the sorption capacity of Cs⁺ between in-situ and laboratory batch sorption experiments. Recent experimental and modeling studies have shown equivalent Cs⁺ sorption on intact and dis-aggregated Opalinus Clay samples. Cation exchange instead of a simpler Freundlich isotherm was used in the modeling. Additionally, Jakob et al. (2009) obtained a larger (x 4) effective diffusion coefficient for Cs⁺ normal to bedding (Dell = 1.8 e-10 m²/s), compared to previous studies. In view of these developments, new modeling of Cs⁺ diffusion and retention in the DI-A2 experiment has been performed, including a cation exchange model to account for the retention of Cs⁺. A difference in sorption capacity by a factor of about 2 still remains depending on the use of the original model by Bradbury and Baeyens (2000) or the modified model by Van Loon et al. (2009). Van Loon et al. (2009) decreased Cs⁺ sorption on the type-II sites to refine the fit of the exchange model to the experimental data in both compacted and dis-aggregated clay samples. A value of Dell (parallel to bedding) equal to 2 e-10 m²/s was obtained when using the modified model by Van Loon et al. (2009); Dell 4 e-10 m²/s was obtained when using the original model by Bradbury and Baeyens (2000), which was also used by Jakob et al. (2009). Clearly, the values of De obtained are correlated with the strength of sorption in the model, with higher sorption leading to larger De values. This correlation may be the reason, at least in part, behind the larger value for Dell obtained by Jakob et al. (2009). Discrimination between the two versions of the exchange model is not possible when using only the results of the in-situ test. Additionally, at early times (t < 10 d) the drop in Cs⁺ concentration in the circulation system is slower than expected. Due to the experimental setup, this slow decrease in concentration cannot be caused by the filter in the contact between borehole and rock. Poor mixing in the circulation system could explain this effect. (authors)

[en] The suitability of different porous materials (stainless steel, VYCOR ^registered glass, Al₂O₃ and PEEK) for use as confining filters in diffusion experiments was evaluated by measuring the effective diffusion coefficients (D_e) of neutral (HTO) and ionic solutes (Na⁺, Cs⁺, Sr²⁺, Cl^-, SeO₄^2-) in the materials in through-diffusion experiments. For stainless steel filters, the D_e values of the target solutes correlated satisfactorily with their bulk diffusion coefficient in water (D_w); thus, the diffusion process in the stainless steel filters was primarily controlled by the diffusivity of the solvated ions. For the remaining materials, the D_e and D_w values were also correlated for the target solutes, and the geometric factors were in the sequence: VYCOR ^registered glass < Al₂O₃ < PEEK. Stainless steel and VYCOR ^registered glass were the most appropriate materials because of their high D_e values, but a specific interaction of caesium with VYCOR ^registered glass was hypothesised because the D_e values obtained for this solute were slightly higher than expected.

[en] Highlights: ► An in situ diffusion experiment at Mont Terri has been modeled. ► Cs⁺ migration can be explained using sorption parameters measured in the laboratory. ► Cs⁺ batch-sorption data can be applied to in situ experiments (no upscaling). ► Differences in the sorption model translate into different diffusion coefficients. ► Effect of poor mixing in the circulation system has been included. - Abstract: In the DI-A2 experiment several non-reactive and reactive tracers were injected as a pulse in a packed-off borehole in the Opalinus Clay. Unlike the previous DI-A1 test, the design of the Teflon filter in the injection borehole forced the water to flow through the filter and the open space between the filter and the borehole wall (the filter itself did not act as a diffusion barrier between the circulating solution and the rock). The decrease in tracer concentration in the liquid phase was monitored during a period of a year. Afterwards, the borehole section was overcored and the tracer profiles in the rock were analyzed. A main interest of this experiment was to understand the chemical behavior of sorbing tracers: Cs⁺ (stable), ⁸⁵Sr²⁺, ⁶⁰Co²⁺ and Eu³⁺ (stable). The complete dataset (except for Eu³⁺ because of strong sorption to experimental equipment) was analyzed in a previous study with a 2D diffusion–reaction model and the derived diffusion and sorption parameters were compared with laboratory data. As in DI-A1, a difference by a factor of about 2 for sorption (magnitude of the Freundlich isotherm) was obtained between in situ and laboratory batch sorption experiments. Recent experimental and modeling studies have shown equivalent Cs⁺ sorption on intact and disaggregated Opalinus Clay samples. In view of these developments, new modeling of Cs⁺ diffusion and retention in the DI-A2 experiment has been performed using CrunchFlow. The calculations include transport by diffusion and a multisite cation exchange model to account for the retention of Cs⁺. The new results show that upscaling of Cs⁺ sorption from laboratory to field is no longer required. However, a difference in sorption by a factor of about 2 is still explained by the use of different versions of the same cation exchange model (a small difference in the selectivity coefficient for one type of site). This uncertainty in sorption leads to an uncertainty in the effective diffusion coefficient (D_e) for Cs⁺, also by a factor of 2 (2–4 × 10⁻¹⁰ m²/s). Clearly, the values of D_e obtained are correlated with the strength of sorption in the model, with stronger sorption leading to larger D_e values. Discrimination between the two versions of the exchange model is not possible when using only the results of the in situ test. Additionally, during early times (t < 10 days) the drop in Cs⁺ concentration in the circulation system is slower than expected. Due to the experimental setup, this slow decrease in concentration cannot be caused by the filter in the contact between borehole and rock. Poor mixing in the circulation system could explain this effect

[en] Solute diffusion in compacted clays depends on ionic strength through its control on the thickness of the electrical double layer (EDL) on the charged clay surfaces. In the DR-A field experiment (Mont Terri, Switzerland), synthetic porewater was circulated through a borehole for 189 days, leading to the out-diffusion of a variety of tracers into the Opalinus Clay. The borehole solution was then replaced with a higher-salinity solution for an additional 540 days, leading to the diffusion of Cs⁺, Ca²⁺, Mg²⁺, and Sr²⁺ back into the borehole and to an increase in the out-diffusion of anions (I^-, Br^-) and ³H. The experimental results were interpreted using the CrunchClay code, which includes a mean electrostatic potential model for the EDL. The EDL corresponds to a second continuum in addition to bulk electrically neutral porewater. Species-specific diffusion (Nernst-Planck equation) occurs through both domains. A 1D radial model considered a single pore diffusion coefficient (Dp = 10-9 m2/s) for cations and 3H in the bulk porosity, and a smaller D_p (3 × 10^-10 m²/s) for anions. Dp values in the EDL were smaller (10^-11 m²/s), except for Cs⁺ and K⁺ (5 × 10^-10 and 2 × 10^-10 m²/s, respectively). The model reproduced well the experimental results and showed the capability to consider temporal changes in geochemical conditions affecting the transport and retention of potentially important radionuclide contaminants (e.g., ¹³⁷Cs⁺, ⁹⁰Sr²⁺, ¹²⁹I^-) in underground geological nuclear waste repositories. Finally, coupled multicomponent diffusion together with the electrostatic properties of the charged surfaces are essential in the development of predictive models for ion transport in clays.

[en] Hydrogen gas (H_2) may be produced by the anoxic corrosion of steel components in underground structures, such as geological repositories for radioactive waste. In such environments, hydrogen was shown to serve as an electron donor for autotrophic bacteria. High gas overpressures are to be avoided in radioactive waste repositories and, thus, microbial consumption of H_2 is generally viewed as beneficial. However, to fully consider this biological process in models of repository evolution over time, it is crucial to determine the in situ rates of microbial hydrogen oxidation and sulfate reduction. These rates were estimated through two distinct in situ experiments, using several measurement and calculation methods. Volumetric consumption rates were calculated to be between 1.13 and 1.93 μmol cm"−"3 day"−"1 for H_2, and 0.14 and 0.20 μmol cm"−"3 day"−"1 for sulfate. Based on the stoichiometry of the reaction, there is an excess of H_2 consumed, suggesting that it serves as an electron donor to reduce electron acceptors other than sulfate, and/or that some H_2 is lost via diffusion. These rate estimates are critical to evaluate whether biological H_2 consumption can negate H_2 production in repositories, and to determine whether sulfate reduction can consume sulfate faster than it is replenished by diffusion, which could lead to methanogenic conditions. - Highlights: • Hydrogen and sulfate are consumed by bacteria in Opalinus Clay borehole water. • The consumption rates are respectively 1.53 and 0.17 μmol cm"−"3 day"−"1. • More H_2 is oxidized than can be accounted for by sulfate reduction. • These are crucial parameters for assessing the safety of deep geological repositories.

[en] Highlights: → Reactive transport modelling was used to simulate simultaneously solute transport, thermodynamic reactions, ion exchange and biodegradation during an in-situ experiment in a clay-rock formation. → Opalinus clay formation has a high buffering capacity in terms of chemical perturbations caused by bacterial activity. → Buffering capacity is mainly attributed to the carbonate system and to the reactivity of clay surfaces (cation exchange, pH buffering). - Abstract: Reactive transport modelling was used to simulate solute transport, thermodynamic reactions, ion exchange and biodegradation in the Porewater Chemistry (PC) experiment at the Mont Terri Rock Laboratory. Simulations show that the most important chemical processes controlling the fluid composition within the borehole and the surrounding formation during the experiment are ion exchange, biodegradation and dissolution/precipitation reactions involving pyrite and carbonate minerals. In contrast, thermodynamic mineral dissolution/precipitation reactions involving alumo-silicate minerals have little impact on the fluid composition on the time-scale of the experiment. With the accurate description of the initial chemical condition in the formation in combination with kinetic formulations describing the different stages of bacterial activities, it has been possible to reproduce the evolution of important system parameters, such as the pH, redox potential, total organic C, dissolved inorganic C and SO₄ concentration. Leaching of glycerol from the pH-electrode may be the primary source of organic material that initiated bacterial growth, which caused the chemical perturbation in the borehole. Results from these simulations are consistent with data from the over-coring and demonstrate that the Opalinus Clay has a high buffering capacity in terms of chemical perturbations caused by bacterial activity. This buffering capacity can be attributed to the carbonate system as well as to the reactivity of clay surfaces.

[en] Highlights: • Initial anaerobic corrosion rates of steel in bentonite outpace the diffusion of water. • A hydrogen gas phase is temporarily formed close to the steel surface. • Gas and the lack of water desiccate the bentonite resulting in shrinkage microfractures. • Diffusing Fe²⁺ is oxidized by sorbed oxygen and precipitates in the porespace. • The safety relevant properties of bentonite are not jeopardized by steel corrosion. Experimental evidence related to the interactions between steel corrosion and bentonite in deep geological repositories is reconsidered. The released Fe²⁺ interacts with the smectite and precipitates as Fe³⁺ when oxidised by residual immobile oxygen. This leads to the development of coloured fronts in the bentonite. The evolution of these fronts can be reproduced by a reactive transport model that takes account of the competition between diffusion and precipitation. Furthermore, two-phase flow modelling indicates that the consumption of water by the corrosion outpaces the diffusion of water through the bentonite, leading to shrinkage microfractures in bentonite that act as preferential pathways for corrosion products.