In the autumn of 2012, the Japan Atomic Energy Agency (JAEA) launched a new research project named F-TRACE (Long-Term Assessment of Transport of Radioactive Contaminant in the Environment of Fukushima). The aims of this project are to develop a system for prediction of radiation exposure, taking into consideration the transport, deposition, and remobilization behavior of radiocaesium (RCs) from the highest contaminated mountain forests, down through the biosphere, before deposition in a number of different aquatic systems. Especially, it is important to understand balances of suspended and deposited particles and RCs inventory in inflow water, discharge water and bottom sediments of an upland reservoir. In this paper, we describe current research activities performed by JAEA at the Ogi Reservoir, Fukushima prefecture, Japan.

According to our analyses the specific sediment yield and the average rate of storage capacity loss at the Ogi Reservoir are 210 m3 km-2 year-1 and 0.15 % year-1, respectively. The vertical distribution of RCs exhibits clear peaks at several sites in the reservoir formed by deposition of eroded soil particles from the catchment that were contaminated by accident fallout. Above the depth of each of the RCs peaks, the distribution of RCs was found to be variable with depth, with concentrations ranging over five orders of magnitude for a single core. The peaks in the sedimentation profiles are probably formed from eroded soil particles entering the reservoir from the surrounding contaminated watershed (most probably during storm events). Results from grain size analyses suggest that contaminated fine sediment tends to be deposited thickly within deeper parts of the reservoir. In addition, above the depth of RCs peaks at these deeper sites, the concentration of RCs approximately increased or decreased as the proportion of fine sediments increased or decreased. However, some fine particles are possibly discharged downstream during operational releases from the dam.

We developed a simplified biofilm sorption and diffusion experiment method. The biofilms of the Bacillus cereus were incubated on cellulose acetate membrane filters (pore size 0.2 µm, diameter 47 mm) placed on thick NB broth agar medium (thickness was about 30 mm) to support sufficient biofilm growth of the Bacillus cereus. The thickness of the formed biofilms was about 1 mm. The formed biofilms were applied to through-diffusion method, which has been used to measure diffusion coefficient of crystalline and sedimentary rocks and clay minerals. The obtained copper sorption coefficient by batch experiments was about 100 ml/g (wet weight) at the case of the concentration of cupper ion was over 0.074mmol/L. And diffusion coefficients by through diffusion experiment was De=1.1 x 10-10 (m2/s). From these results, this simplified biofilm sorption and diffusion experiment may make possible to obtain these parameters with ease.

The potential effect of high pH plume caused by cementitious materials must be evaluated in the performance assessment for HLW geological disposal. Alkaline plume would lead to change sorption properties of host rock by primary mineral dissolution, secondary mineral precipitation and sequential change of pore water chemistry. In this study, the effect of alkaline alteration on sorption of Cs, Ni and Th was investigated using rock samples from the Horonobe Underground Research Laboratory. Crushed rock samples were reacted in high pH alkaline solution at 90 °C for 45 days, 95 days and 1,383 days, respectively. As a result of sample analysis, it was supposed that zeolitic mineral was precipitated as secondary mineral. The cation exchange capacity slightly increased in comparison with the unaltered sample. Distribution coefficients (Kd) of Cs, Ni and Th on unaltered and altered rock sample were measured by batch sorption experiment in synthetic groundwater. Kd of Cs increased with the alteration period. These results show that secondary minerals contribute to the increase in Cs sorption. By contrast, Kd of Ni and Th decreased with the alteration period. This change might be caused by dissolution of clay minerals and amorphous silicates controlling Ni and Th sorption by surface complexation. These results imply that effects of alkaline alteration on Kd of rocks depend on the dissolution/precipitation of minerals, their surface properties and sorption mechanisms.

C-14 contained in Hull waste is one of the most important radionuclides in the safety assessment of transuranic (TRU) waste disposal. For more realistic safety assessment, it is important to clarify the release mechanism and chemical species of C-14 from Hull waste. In this research, leaching tests were conducted using an irradiated Zry cladding tube from a boiling-water reactor (BWR) to obtain leaching data and to investigate the relationship between Zry metal corrosion and C-14 release behavior. Both organic and inorganic C-14 compounds existed in the the liquid phase, and some C-14 moved to the gaseous phase. The release rate of C-14 obtained from the BWR cladding tube after two-year leaching tests was lower than the release rate from a pressurize water reactor (PWR) cladding tube. It is considered that the BWR cladding tube used in this test did not easily corrode since it used a comparatively new material. The release rate of C-14 was slightly lower as compared with the corrosion rate of unirradiated Zry. This is thought to be the result of improved corrosion resistance conferred by neutron irradiation, which encouraged the dissolution of grain boundary precipitation elements, such as Fe, Cr, and Ni, into the crystal grains. The leaching tests will be continued for 10 years.

Corrosion behavior is a key issue in the assessment of disposal performance for activated waste such as spent fuel assemblies (i.e., hulls and end-pieces) because corrosion is expected to initiate radionuclide (e.g., C-14) leaching from such waste. Because the anticipated corrosion rate is extremely low, understanding and modeling Zircaloy (Zry) corrosion behavior under geological disposal conditions is important in predicting very long-term corrosion. Corrosion models applicable in the higher temperature ranges of nuclear reactors have been proposed based on considerable testing in the 523−633 K temperature range.

In this study, corrosion tests were carried out to confirm the applicability of such existing models to the low temperature range of geological disposal, and to examine the influence of material, environmental, and other factors on corrosion rates under geological disposal conditions. A characterization analysis of the generated oxide film was also performed.

To confirm applicability, the corrosion rate of Zry-4 in pure water with a temperature change from 303 K to 433 K was obtained using a hydrogen measuring technique, giving a corrosion rate for 180 days of 8 × 10-3 μm/y at 303 K.

To investigate the influence of various factors, corrosion tests were carried out. The corrosion rates for Zry-2 and Zry-4 were almost same, and increased with a temperature increase from 303 K to 353 K. The influence of pH (12.5) compared with pure water was about 1.4 at 180 days at 303 K.

The carbon-14 generated in Zircaloy (Zry) hull waste is considered an important radionuclide in the TRU waste geological disposal concept in Japan. Given that the metal Zry is highly corrosion-resistant in the anaerobic and low-temperature conditions of the repository, and that the C-14 release rate is assumed to be controlled by the corrosion rate, a variety of corrosion and leaching tests have been performed. However, since the Zry corrosion rate is extremely slow, it is not possible to predict long-term corrosion behavior through low-temperature corrosion tests conducted in a reasonable time period. A vast amount of testing has been conducted in the higher-temperature range of 523 to 633 K, and corrosion correlations have been obtained from these tests. Corrosion correlations have been used to predict the corrosion rate of Zry in a tuff repository. Long-term Zry autoclave corrosion data have been analyzed to develop new corrosion correlations. Extrapolating these correlations to a lower temperature range requires verification that the mechanisms do not change over the range of testing and extrapolation. Factors that influence corrosion rates under geological disposal conditions, such as material and environmental factors, should also be examined. Corrosion correlations, factors influencing corrosion rates, the results of corrosion and leaching tests, and a preliminary evaluation are discussed.

Modern concrete has a record of good performance of around 120 years although there are structures in perfect conservation made with roman concrete (mixture of lime and natural pozzolans). El Cabril repository has a design life of 300-500 years and therefore, it should keep its integrity much longer than the back experience we have on reinforced concrete structures, which makes necessary a closer monitoring with time on the aging of concrete in real conditions. With this purpose, Enresa has designed in collaboration with IETcc and Geocisa the installation of permanent sensors in a pilot nuclear waste container in buried conditions. The sensors were installed in 1995 for monitoring corrosion parameters and have been working until present. The non-destructive tests (NDT) applied are based in electrochemical measurements (corrosion rate, corrosion potential, electrical resistivity, concrete strains, oxygen availability). Relations between the climatic influence, the buried depth and the corrosion parameters are also presented. The results indicate that temperature is a very relevant variable influencing the measurements. All the other parameters evolve according to seasonal changes. Values of activation energies of the resistivity changes are given although it seems more adequate to model the evolution with time by simply plotting the values registered at 20 ± 2 °C.

The occurrence of Atmospheric chloride-Induced Stress Corrosion Cracking (AISCC) under wetted deposits of MgCl2 or sea-salt at 70°C has been investigated at various Relative Humidities (RH). The appearance of AISCC is a function of the environmental RH. At 33% RH (the deliquescence point of MgCl2), AISCC generated under MgCl2 or sea-salt deposits is of a similar appearance with regards to the number of cracks produced and average crack length. At 50% RH sea-salt seems to be more aggressive at least in terms of crack frequency. This observation may highlight the significance of carnallite (KMgCl3.6H2O) in promoting AISCC in types 304L and 316L stainless steels. The use of accelerated testing methods to validate apparent thresholds in chloride deposition density and other critical factors that influence the initiation and propagation of AISCC is briefly discussed.

The release of radionuclides measured during washing cycles of spent nuclear fuel samples in a series of experiments using different solutions are analyzed with respect to the fission products Cs, Sr, and Tc and the actinides U, Pu, and Am. Based on the concentrations of the dissolved radionuclides, their release rates are evaluated in terms of fraction of inventory in the aquatic phase per day. The application of this information on the fast/instant radionuclide release fraction (IRF) is discussed and following issues are addressed: Duration of the wash steps, solution chemistry, and radionuclide sorption onto surface of the experimental vessels. Data for the IRF are given and the correlation between the mobilization of the various elements is analyzed.

Preparation and characterization of a Simulated Spent Nuclear Fuel (SIMFuel), which replicates the chemical state and microstructure of Spent Nuclear Fuel (SNF) discharged from UK Advanced Gas-cooled Reactor (AGR) after a cooling time of 100 years is described. Thirteen stable elements were added to depleted UO2 and sintered to simulate the composition of fuel pellets after burn-ups of 25 and 43 GWd/tU and, as a reference, pure UO2 pellets were also investigated. The fission product distribution was calculated using the Fispin code provided by NNL. SIMFuel pellets exhibit a microstructure up to 92% TD. During the sintering process in H2 atmosphere Mo-Ru-Rh-Pd metallic precipitates and grey-phase ((Ba, Sr)(Zr, RE)O3 oxide precipitates) formed within the UO2 matrix. These secondary phases are present in real PWR and AGR SNF, although they are smaller in size than those examined in this study. The grain size of the produced SIMFuel is in good agreement with literature references.

The solubility of molybdenum in borosilicate glasses is low. The UK National Nuclear Laboratory has developed a new glass formulation containing calcium and zinc for the vitrification of high molybdenum containing waste arising from the Post Operational Clean Out of the Highly Active Storage Tanks at Sellafield that will decrease the number of product containers required, reducing both production and disposal costs. The new formulation increases the quantity of molybdenum that can be vitrified through the formation of a durable CaMoO4 phase once the solubility limit of molybdenum in the glass has been exceeded. Extensive laboratory trials confirmed the potential to increase the Mo loading significantly. Recently full scale testing has been performed on the Vitrification Test Rig using highly active liquor simulants to determine the maximum MoO3 loading that can be achieved. This paper explores the full scale testing and product quality of the glass manufactured during this study.

The rapid release of fission products segregated either to the gap between the fuel and the cladding or to the UO2 grain boundaries from spent nuclear fuel in contact with water (often referred to as the instant release fraction - IRF) is of interest for the safety assessment of geological repositories for spent fuel due to the potential dose contribution. In September 2012 a study was initiated with the aim of comparing the instant release behavior of fuels with and without additives/dopants. Preliminary results from this (ongoing) study indicate that the release of uranium during the first contact periods was higher than during the tests with fuel segments, even though the fuel was cut open recently [1]. This could be due to the sample preparation method which included axial cutting of the cladding in order to remove the fuel fragments used in the study. In the present work, leaching data from both studies are presented and the releases are discussed comparing the two sample preparation methods and considering the effect of matrix composition. The leaching studies have been performed in air using 10 mM NaCl + 2 mM NaHCO3 as leaching solution.

To assess the stability of spent fuel in the highly alkaline chemical environment of the Belgian Supercontainer design, static leach experiments were performed with depleted UO2 and 238Pu-doped UO2 at different SA/V ratios for 1.5 years in cement waters (11.7< pH < 13.5) at ambient temperature and under argon atmosphere. The influence of the calcium concentration on the uranium release was also investigated. While the ultrafiltered U(IV) concentration was 10-9-10-8 mol.L-1 and independent of the pH, the U(VI) release from the UO2 surface was enhanced by the OH- concentration, leading to soluble U concentrations up to 10-5 mol.L-1 at high SA/V and under the influence of radiolysis. Together with the high Ca concentration, this can lead to the formation of Ca-U(VI) colloids as precursor of Ca-U(VI) secondary phases, decreasing the soluble U concentration. The precipitation of Ca-U secondary phases was however not clearly evidenced by surface analyses.

The Knudsen Effusion Mass Spectrometer (KEMS) and the mechanistic MFPR (Module for Fission Product Release) code are tools which seem particularly interesting to support studies of the Instant Release Fraction (IRF) of Cs from spent nuclear fuel in a final repository. With KEMS, the thermal release of 137Cs and 136Xe were analysed by annealing up to total vaporization (2500K) of high burn-up (60 GWd/tU) Spent Nuclear Fuel (SNF) samples. Powder samples from the centre of the fuel, without high burn-up structure, were used. To determine the IRF, samples were analysed before and after being submitted to corrosion experiments in bicarbonated aqueous media.

MFPR was applied to determine the localization of Cs and fission gases in the SNF at the end of irradiation; the results are compared and supported by dedicated thermodynamics calculations performed for equilibrium conditions at various temperatures and fuel oxygen potentials by the non-ideal thermodynamic MEPHISTA (Multiphase Equilibria in Fuels via Standard Thermodynamic Analysis) database. A possible mechanism for Cs release during thermal annealing is proposed, taking into account inter-granular release and Cs oxide vaporization, atomic diffusion, ternary oxide phase formation and bubble release.

Differences in KEMS release profiles before and after submitting the samples to aqueous corrosion are attributed to the IRF and to changes in the vaporisation mechanism because of differences in the oxygen potential (pO2). The IRF of Cs estimated from the KEMS spectra, consisting on the part located at the grain boundaries and in inter-granular bubbles, is not significantly different from that corresponding to the experimental results found using classical static leaching experiments.

New experimental campaigns are being designed to confirm our interpretation proposed after this first run.

Two adjacent fuel rod segments were irradiated in a pressurized water reactor achieving an average burn-up of 50.4 GWd/tHM. A physico-chemical characterisation of the high burn-up fuel rod segments was performed, to determine properties relevant to the stability of the spent nuclear fuel under final disposal conditions. No damage of the cladding was observed by means of visual examination and γ-scanning. The maximal oxide layer thickness was 45 µm. The relative fission gas release was determined to be (8.35 ± 0.66) %. Finally, a rim thickness of 83.7 µm and a rim porosity of about 20% were derived from characterisation of the cladded pellets.

A simulated Magnox glass which is Mg- and Al- rich was subjected to aqueous corrosion in static mode with deionised water at 90 °C for 7-28 days and assessed using X-Ray Diffraction (XRD), Scanning Electron Microscopy (SEM) with Energy X-Ray Dispersive Spectroscopy (EDS) and Inductively Coupled Plasma – Optical Emission Spectroscopy (ICP-OES). XRD revealed both amorphous phase and crystals in the glass structure. The crystals were Ni and Cr rich spinels and ruthenium oxide. After two weeks of incubation in deionised water, the glass surface was covered by a ∼11 μm thick Si-rich layer whilst mobile elements and transition metals like Na, B, and Fe were strongly depleted. The likely corrosion mechanism and in particular the role of Mg and Al in the glass structure are discussed.

In the KBS-3 repository concept and safety analysis, the copper container with a cast iron insert plays a central role in assuring isolation of the waste from the surrounding during long periods of time. All processes that affect its stability are thoroughly analysed, including potential detrimental processes inside the canister. For this reason, an estimation of the helium produced during the long term decay of alpha emitters in the spent fuel is necessary to evaluate if the pressures generated inside can have consequences for the canister.

The spent nuclear fuel to be disposed of in Sweden is mainly LWR fuel. The maximum burn-up expected is 60 MWd/kg U for BWR and PWR. A small quantity of BWR MOX is expected to be stored with a maximum burn-up of 50 MWd/kg U.

This work has focused on carrying out calculations of the amounts of He generated during more than 1 million years in Swedish spent nuclear fuels with a benchmarking exercise by using both codes AMBER and Origen-ARP. The performance and agreement of the codes in the He generation from α-decay have been checked and validated against data reported in literature [1].

In the calculation of the maximal pressure inside the canister, the quantity of helium used to pre-pressurise the fuel rods has been accounted for. The pressure inside the canister due to He generation is at all times much lower than the hydrostatic pressure and/or the bentonite swelling pressure outside the canister.

A flow-through experimental reactor has been designed in order to perform studies at both high pressure and high temperature conditions. A chromatographic pump is used to impulse the leachant throughout the reactor in order to work at very low flows but high pressures. Therefore, high surface solid to volume leachant ratios, similar to the ones predicted in the final repository, can be obtained. The reactor allows working at different atmospheres at pressures up to 50 bars. The temperature inside the reactor can be set using a jacket.

Using this new reactor the evolution of uranium concentrations released from an UO2 sample was studied at different conditions.

The results show that at hydrogen pressures between 5 and 7 bars, hydrogen peroxide does not seem to significantly oxidize the uranium (IV) oxide. Uranium concentrations in those experiments remain between 10-8 mol·l-1 and 10-9 mol·l-1.

Natural metamict mineral found as large (1-3 cm in size) homogeneous grains (as assumed, former single crystals), was investigated by X-ray powder diffraction (pXRD), high-temperature pXRD, scanning electron microscopy (SEM) and electron microprobe analysis (EMPA). The average chemical composition obtained by EMPA is (wt. %): Nb2O5 – 42.6; Ta2O5 – 4.4; TiO2 – 9.2; UO3 – 4.4; ThO2 – 1.0; MnO – 1.3; FeO – 19.4; Y2O3 – 16.6.

The untreated (original) sample is X-ray amorphous. The sample remained amorphous after annealing at 400 °C for 1 hour. The sample became almost fully crystalline after annealing at 700 °C for 1 hour with an X-ray diffraction pattern similar to that of Fe-columbite (ICCD: 01-074-7356). Further annealing at 1000 °C and higher temperatures caused changes in the phase composition of the sample. It was proposed that under self-irradiation a single-phase U-Th-bearing solid solution, based on monocrystalline Y-niobate, became metamict but remained homogeneous without evidence of solid solution destruction. However, this metamict solid solution is unstable under thermal treatment and recrystallization.

Apatites are often seen as good potential candidates for the immobilization of halide-rich wastes and, in particular, chlorapatite (Ca5(PO4)3Cl) has received much attention in recent years. However, synthesis of chlorapatite waste-forms can produce a complicated multi-phase system, with a number of secondary phases forming, including β-TCP (Ca3(PO4)2), spodiosite (Ca2(PO4)Cl) and pyrophosphate (Ca2P2O7), many of which require elevated temperatures and extended calcinations times to reduce. Calcium vanadinite (Ca5(VO4)3Cl) demonstrates a much simpler phase system, with calcination at 750 °C yielding Ca5(VO4)3Cl together a small quantity of a Ca2V2O7 secondary phase, the formation of which can be retarded by the addition of excess CaCl2. Characterization of compositions doped with SmCl3 as an inactive analogue for AnCl3 show the Cl to be immobilized in the vanadinite whilst the Sm forms a wakefieldite (SmVO4) phase.