[en] Models are developed to evaluate the performance of potential geological repositories for high level nuclear wastes systems. Source term models are used to study the stability of the spent nuclear fuel under repository conditions. These models increase the confidence in predicting the evolution and behaviour of the spent fuel matrix and of the radionuclides embedded in it. In this context, we will present a review of the assumptions and approaches considered in most of the studies to simulate the release of the radionuclides embedded in the spent fuel matrix. After this review exercise, we will show the radionuclide release model developed by our group, which is based on kinetic and thermodynamic approaches in order to study the evolution of the spent fuel-water interface as a function of time. The results of this modelling are used as input data in other models that simulate the transport of the released radionuclides through the near-and the far-fields up to the biosphere. Therefore, the model presented in this work should lead to a better definition of the source-term for performance assessment exercises. (Author)

[en] The aim of this report is to give the reader an overall insight of the different models, which are used to predict the long-term behaviour of the spent fuels and HLW disposed in a repository. The models must be established on basic data and robust kinetics describing the mechanisms controlling spent fuel alteration/dissolution in a repository. The UO2 matrix, or source term, contains embedded in it the , majority of radionuclides of the spent fuel (some are in the gap cladding). For this reason the SF radionuclides release models play a significant role in the performance assessment of radioactive waste disposal. The differences existing between models published in the literature are due to the conceptual understanding of the processes and the degree of the conservatism used with the parameter values, and the boundary conditions. They mainly differ in their level of simplification and their final objective. Sometimes are focused the show compliance with regulatory requirements, other to support decision making, to increase the level of confidence of public and scientific community, could be empirical, semi-empirical or analytical. The models take into account the experimental results from radionuclides releases and their extrapolation to the very long term. Its necessary a great statistics for have a representative dissolution rate, due at the number of experimental results is not very high and many of them show a great scatter, independently of theirs different compositions by axial and radial variations, due to linear power or local burnup. On the other hand, it is difficult to predict the spent fuel behaviour over the long term, based in short term experiments. In this report is given a little description of the radionuclides distribution in the spent fuel and also in the cladding/pellet gap, grain boundary, cracks and rim zones (the matrix rim zone can be considered with an especial characteristics very different to the rest of the spent fuel), and structural components such as Zircaloy, Inconel and stainless steel, to further on describing the radionuclides release from these zones. (Author)

[en] A general model of UO₂ corrosion is given UO₂ is an electrochemical process consisting of anodic dissolution UO₂ to UO''2₂ + 2e accompanied by a cathodic reduction of an oxidant such as oxygen: e. g. O₂+ 2H₂O+ 4e-4OH for neutral and basic solutions. The cathodic oxygen reduction is often the slower process. The rate-determining step being the oxygen diffusion or transport to the electrode surface. The equation for the partial current density of oxygen reduction at the surface is known and is directly proportional to the equilibrium exchange current density. Alternative cathodic processes ar4e hydrogen and water reduction reactions; their role depends upon the exact conditions of corrosion (temperature, atmosphere, pressure). At the potentials of interest for UO₂, it can be seen that oxygen reduction is the dominant and most important cathodic process. Taking into account the experimentally determined corrosion potentials and corrosion rates, the anodic dissolution of UO₂ can be described in a simpler form of the anodic current density i_uo2_const.exp (2.303/E/b where b=0.08 V and E is the polarisation potential in volts. Using these equations the corrosion potential of UO₂ can be calculated from the oxygen concentration via the oxygen reduction partial current densities. Natural UO₂ immersed in deaerated 0.1 M NaCl solution was monitored as hydrogen peroxide was dosed in the solution. A potential shift was observed and was also calculated from these equations, the form of the curves were very similar with an offset of approximately 50 mV. This offset was attributed to the influence of residual hydrogen peroxide and its decomposition products (other than oxygen) that are not taken into account in the equations. On the other hand, the shift of 50 mV in corrosion potential also represents an increase in corrosion rate of about 7 times. The interplay of the different partial current densities was then emphasised with a variety (logarithmic and linear) of graphical illustrations, and indicates how the environmental can affect the corrosion rates and reactions. (Author)

[en] The generation of radiolytic products as a result of alpha radiation in the surface of the spent fuel is a key process in order to understand how it becomes degraded in repository conditions. The present work has established a radiolytic model based on a set of reactions involving fuel oxidation-dissolution and radiolytic products recombination. It also includes the decrease of the dose rate as the main alpha emitters decay away. Four cases, with varying parameters of the system, have been assessed. The results show a decrease in both the concentration of the radiolytic products in the gap water and the degradation of the fuel matrix. It has been estimated that in the period of the evaluation (10''6 years) up to 52% of the pellet is altered in the conservative cases, whereas only 11% is altered in the realistic cases. No significant differences were observed when the carbonates reactions were included in the system. (Author)

[en] Radiolytic models are usually to study the effect of a radiation field on the chemistry of a given system. The spent fuel contain a large amount of radionuclides that leads to a complex radiation field in the vicinity of the spent fuel surface and beyond. The effect of this radiation field on the surrounding environment of the nuclear waste under repository conditions has been the focus of many modelling efforts. in this work, we are interested in the effects of alpha radiolysis on the stability of the spent fuel matrix after failure of the containing system and the subsequent intrusion of groundwater into the system. A review of the different approaches to model the alteration of the spent fuel by alpha radiolysis if given. Several authors have developed radiolytic models based on different assumptions, leading to differences in the predicted behaviour of the system. We have developed a radiolytic model for the long-term alteration of the spent fuel matrix under repository conditions. The model takes into account several processes: generation and recombination of radicals, oxidative dissolution of the spent fuel surface, evolution with time of the alpha dose rate and water turnover in the gap between the fuel pellet and the cladding. The results are encouraging and gives us confidence in further developing the model to include more complex issues, such as the geochemistry of the contacting waters. (Author)

[en] A kinetic-based model to predict the dissolution of UO₂ α-doped pellets under initial anoxic conditions is presented and compared with experimental results previously obtained. The uranium and plutonium concentrations in solution are predicted by considering the presence of a α-radiation field and its influence due to radiolysis of water on the pellet surface oxidation and subsequent dissolution. The initial parameters required by the model in order to reproduce the pellet alteration process are: system geometry, chemical composition of the leachant, physicochemical characteristics of the leachate and the oxidation conditions of the pellet surface (expressed in terms of U(VI)/U(IV) ratio). The last one is the key parameter in the model for simulating the initial quick dissolution process. The results obtained are compared with experimental data. The agreement between the predictions obtained and the published data is good. The influence on the matrix oxidation-dissolution process due to the α-radiation field intensity and the release of Pu are reproduced by the model. The Pu concentration trends as a function of time are explained in connection with the matrix dissolution process and are controlled by the formation of the secondary phase Pu(OH)₄(S)''. (Author)

[en] In this work we made a review on the different models and mechanisms that have been developed by different authors to explain the dissolution of spent nuclear fuel under oxic conditions. In most cases the oxidizing reagent used has been the molecular oxygen, but also some works with hydrogen peroxide or even with hypochloric acid can be found. Leaching experiments have been carried out with different types of spent nuclear fuel as well as with either chemical or natural analogues such as non irradiated uranium dioxide or natural uraninites, respectively. In oxygen and in the absence of bicarbonate ion, the data found in literature can be fitted considering the two-step oxidative dissolution mechanism developed by Torrero et al. (1998). This mechanism is able to explain the different reaction orders for pH oxygen concentration obtained depending on the experimental conditions. In the presence of bicarbonate, the data can be fitted considering the mechanism described de Pablo et al. (1999), which consists on two different steps: (1) oxidation of the surface of the solid and (2) surface co-ordination of the bicarbonate ion and dissolution of the complex formed. This model allows to explain different reaction orders for bicarbonate and oxygen concentration obtained by different authors. The development of a mechanism of UO₂ oxidation and dissolution in the presence of hydrogen peroxides is much more complied than in the case of oxygen because of the decomposition of the hydrogen peroxide, which is probably catalysed by the UO₂(s). At present, more work is being directed to the elucidation of this mechanism, including the study of the influence of some radicals such as OH on the UO₂ dissolution. (Author)

[en] In France, different types of spent fuel (UOX2, UOX3, URE and MOX) must be studied for their possibility of deposited in deep repository. The spent fuel assembly is quite a complex and heterogeneous system; it is composed of different materials and its physico-chemical and radiological evolution depends on few parameters as composition of fuel, burnup but also external parameters linked to the environment. In the frame of the safety assessment of the repository, Andra has to define the radionuclides release by the spent fuel in the environmental conditions of the deep repository but she must demonstrate the robustness of the evaluation, the source term, which is based on the choice of prediction models to evaluate the long term behavior of spent fuel and the radionuclides release. To built the source term PA model, it was necessary to know three main data: - the physico-chemical characteristics of the spent fuel matrix and their evolution with time: - the location and the inventory of the radionuclides in different components of the assembly and their evolution with time. - the matrix alteration and the radionuclide release because the matrix is where the majority of radionuclides is located. This data is based on the state of the art on the knowledge on the mechanisms involved in controlling the radionuclides release and the main parameters associated. Because the spent fuel is very complex and the knowledge is not sufficient today, a strategic is adopted to deal with uncertainties. Therefore some simplifications must be proceeded and justified in regards of: - french industrial production of spent fuel, - the concept of geological disposal in France, - the state of the art and the level of validation for different models available on the duel corrosion: - and the knowledge accessible until 2006 which is the deadline in France to return the deep repository safety report. (Author)

[en] This presentations is mainly based on the electrochemical studies carried out by the Canadian team and the research group of the Berlin University. Electrochemical studies allow to study separately both the anodic reaction which corresponds-sources on UO₂-electrodes response is one of to the UO₂ dissolution and the cathodic reaction that is the reduction of the oxidants. By using intensity current-potential plots a mechanisms of UO₂ corrosion has been established. At-300 mV (vs SCE), irreversible oxidation of UO₂ takes place and dissolution begins. In the absence of complexing agents like carbonate, an oxidised layer is formed at 100 mV a stoichiometry close to UO₂. In carbonate medium, the oxidized layer is not formed because the U(VI) formed is rapidly dissolved. Results in terms of dissolution rates obtained by electrochemical measurements are similar to the ones obtained in dissolution experiments by using flow through reactors and similar kinetic laws are obtained. The effect of external α and γ-sources on UO₂-electrodes response is one of the few available data on the effects of radiolysis on the UO₂ dissolution rate and can offer a complementary knowledge to the spent fuel and α-doped pellets dissolution experiments. (Author)

[en] Between the years 1996 and 2000, the BFS (federal Office for Radiation Protection, Germany) supported the development of an integrated source term for disposal of different wastes forms, including spent fuel elements in a salt formation. The aims of this project covered reviews on the present state of knowledge on waste form reactions in salt solutions, experimental and theoretical short term radionuclide release, and modelling of the long-term evolution of the geochemical environment and the maximal radionuclide concentrations. This presentation covers specific findings by a combination of experimental and theoretical results relevant for the source term (radionuclide concentrations) in the near-field of the spent fuel elements. Geochemical modeling is based on the concept which assumes a quasi closed system in the rock salt disposal including the spent fuel and all other components present in the near-field. Parts of these components are the solutions, primary and secondary solid phases, and gaseous components. Due to an effective functioning of the near-field barriers,one can assume that exchange processes between this quasi closed system and the open natural system are limited and controlled by slow diffusive processes. The benefit from this approach lies in the fact that thermodynamic reaction path models effectively describe the systems under consideration. Kinetically controlled radionuclide release processes account only for the easily soluble radionuclides (e.g. Cs), but not for the actinides. Thermodynamic/geochemical modeling show that under the conditions of a deep repository in rock salt, the dominating corrosion of steel canisters are Fe(II)-Fe(III) compounds. Fe(II) may be oxidized to Fe(III) and which compete for oxidizing species produced e.g. by radiolysis. Results of experiments in the presence and absence of corroded iron significantly: In the presence of iron, release rates were found to be lower and the measured actinide concentrations were lower,too. In the experiments, an increase of the pH was measured. This effect can be modeled by assuming an inhomogeneous Cs distribution on the fuel grains. (Author)