Traditionally the effective diffusivity in and the formation factor of intrusive igneous rock have been measured in the laboratory by through diffusion (TD) experiments, which are very time consuming in larger samples with low porosity. In previous work alternating current (AC) has been used to measure the formation factor directly in large samples. In this paper direct current is used to actually transport the tracers through the rock sample in so called through electromigration (TEM) experiments. In these experiments electoosmosis has to be corrected for. The experimental time is reduced substantially when adding an electromigratory flux to the diffusive flux. TD, TEM and AC experiments were performed on a 15 mm thick unweathered granite sample from Laxemar, Sweden. The tracers uranin and iodide were used. The formation factor measured with the three methods varied between 1.2·10-4 - 2.87·10-4.

To understand the influence of cations and silica in groundwater on spent fuel corrosion, the leaching behavior of newly reduced UO2.00 in different solutions was investigated. Uranium concentrations in the solutions were measured and the leached UO2 surface and precipitates on it were analyzed by XPS, laser Raman spectroscopy, SEM-EDS and TEM-EDS. It was found that, in air saturated 2 mM HCO3- solutions, the leaching rates of UO2(s) depend on the composition of the leaching solutions. Very similar leaching rates, 0.5 μg • cm-2 • day-1, were obtained in 0.1 M NaCl and 0.1 M KCl solutions, while that in 0.45mM Ca2+ -0.18 mM Mg2+ -0.2 mM SiO2 containing synthetic groundwater was 10 times smaller. Calcite and quartz like precipitates were detected on the corroded UO2(s) surface.

The chemical forms of carbon leaching from carbon-containing Zr and Fe-based metallic materials have been investigated to improve the estimation of the contribution of C-14 in the performance assessment of TRU waste disposal. Both organic and inorganic carbons were identified in the leached solution with carbon containing zirconium and steel, and the concentrations of total carbon (organic plus inorganic) in the leached solutions increased with time. The carbon concentrations in the leached solution for both metallic samples were higher at higher pH. With High Performance Liquid Chromatography (HPLC), organic carbons were identified to be low-molecular weight alcohols, carboxylic acids and aldehydes.

To explore the chemical state of carbon in the matrix materials, the leaching experiments were carried out also for ZrC, Fe3C, the powder mixtures of carbon and zirconium, and of carbon and iron. The low-molecular weight organic carbons were detected only in the case of carbides (ZrC and Fe3C). The chemical forms of carbon in the zirconium alloy were suggested to be carbide or carbon by H.D. Smith[1]. The present result suggests that the chemical forms of carbon in zirconium or iron are mainly in the form of carbide.

In the interest of performance assessment, the distribution coefficients of the organic carbon species identified in the leached solution for cement were obtained. As expected, some of them were shown to be larger than the values assumed in the performance assessment of Progress Report on Disposal Concept for TRU Waste in Japan[2].

In-situ coupled multielectrode array sensors were used to measure the non-uniform corrosion of carbon steel and stainless steel materials under KCl salt deposit in simulated dry repository environments. It was found that the initiation of non-uniform corrosion occurs at a relative humidity that is 14% lower than the deliquescence relative humidity of the chloride salt. It was found also that once significant corrosion had occurred, the non-uniform corrosion process for the carbon steel material under the salt deposit continues at relative humidities as low as 27%.

A probabilistic cell model to study the hydration, alteration, and elemental release into solution from nuclear waste glasses is presented. This computational model considers the glass alteration as the product of hydration and multiple phase transformations. The model reveals that the elemental redistribution occurring as a consequence of these transformations can produce transport barriers that would affect the glass alteration and release into a surrounding aqueous solution. Importantly, the model also exhibits an apparent saturation concentration in solution, which has a kinetic origin and is lower than the true thermodynamic value. We also highlight the interplay among the different mechanisms that give rise to the glass corrosion process.

Carbon steel is one of the candidate overpack materials for high-level waste disposal and is expected to assure complete containment of vitrified waste glass during an initial period of 1000 years in Japan. The lifetime of the carbon steel overpack will depend on its corrosion rate. The corrosion rate of carbon steel is reduced by the presence of buffer material such as bentonite. Buffer material will delay the supply of corrosive materials and discharge of corrosion products through it. Carbon steel overpack will be corroded by consuming oxygen introduced by repository construction after closure of repository and then will keep the reducing environment in the vicinity of repository. The reducing condition will be expected to retard the migration of redox-sensitive radionuclides by lowering their solubilities. Therefore, the diffusion of corrosion products of iron in buffer material is important to discuss the corrosion rate of overpack, properties of buffer material and migration of redox-sensitive radionuclides. Electromigration experiments have been carried out with source of iron ions supplied by anode corrosion of iron coupon. Iron ions migrated as fast as 2mm in 2 or 3 hours under electrical field of 100V/m to the direction of cathode. Because the iron ions displaced exchangeable sodium ions in bentonite at a ratio of 1 to 2 during the electromigration, migrating iron ion could be ferrous ion.

A wetland located on the Savannah River Site in Aiken, South Carolina has become contaminated by a number of radionuclides as a result of operating a pilot-scale nuclear test facility. Sediment Th-232 concentration at the site exceed 700 mg/kg, whereas background concentrations are ∼7 mg/kg. The Th is relatively mobile, having moved >300-m since 1958 and greatly exceeding contaminant transport model predictions. Studies were initiated to determine the chemical cause for the enhanced transport of Th. Particle-size fractionation studies showed that colloids alone were not likely responsible for all of the enhanced transport of Th. Sediment sequential extraction data showed that >80% of the Th was associated with the sediment organic fraction. In an experiment in which varying soil amendments resulted in the release of dissolved organic carbon (DOC) to the aqueous phase, a significant correlation (r = 0.91; p ≤ 0.01, n = 8) was calculated between aqueous Th and DOC. Pore water Th partitioning to hydrophobic resins suggested that 33% of the aqueous Th was associated with organic matter. Together, these results provide indirect evidence support the notion that enhanced mobility of Th at the site may in part be due to the presence of Th-DOC complexes.

The chemical properties of groundwater entering a repository for nuclear wastes, have a major impact on the performance of natural and engineered barriers. Thus, accurate models for the chemical evolution of the groundwater are important in connection with performance assessments. Simple mixing of waters is inadequate for describing most chemical properties, like concentrations of iron, pH or redox potential.

In the present work, ground waters from Äspö at the Swedish east coast have been modeled. In the model it is assumed that ancient waters (end member waters) have been able to reach a steady state with respect to the minerals in the fracture walls. Due to groundwater flow, end member waters have been mixed to form intermediate waters. In the sampling process (by pumping), waters from a large number of small units are mixed too rapidly for mineral reactions to be of importance. After the sampling mixture has been formed, the water reaches equilibrium with respect to homogeneous reactions (in the aqueous phase) but not with respect to heterogeneous reactions.

The CRACKER program has been used to simulate groundwater formation and sampling process according to the model. It has been found that the end member waters are fairly stable with respect to observed mineral sets.

Conservative elements like Cl have been used to find the initial mixing ratios for intermediate waters. Evolution of the mixtures was simulated and finally mixing and speciation of waters from some 900 locations have been simulated to describe the final solution. It is found that the model is able to give a good agreement between observed and modeled concentrations for the ten most important elements as well as pH and redox potential.

Both concentrated and dilute simulated solutions of saturated J13 and unsaturated UZ pore water were concentrated through distillation of the solutions under atmospheric pressure. It was observed that condensed vapors from the pH of J13 waters steadily rose during the distillations to a value of 10, while the pH of UZ waters remained steady until 90% of the volume of the solution had been distilled, after which the pH of the condensed vapors dropped precipitously, often below 1. Residual solutions analyzed when most of the solution had been distilled away were also found to be extremely acidic. The temperature of these residual solutions was around 144°C due to their high solute content causing boiling point elevation. All experiments were performed with the condenser open to ambient air at atmospheric pressure. The pH drop during the distillation of UZ water is attributed largely to the presence of large amounts of magnesium. Specimens of Alloy 22 tested in the residual solutions of at their boiling temperature (around 144°C) showed significant rates of general corrosion over a broad range, often approaching 1 mm/year. Similarly high corrosion rates were also observed in tests on Alloy 22 specimens in condensates obtained during the late stages of the distillation. These tests were performed either in situ at 75–80°C using a Soxhlet extractor, or in separate pressure vessels at temperatures between 90 and 130°C.

HLW after the separation of U, Pu and Np contains two major radionuclide groups: fission products and TRU elements. The later ones are the most dangerous and should be separated in the special actinide fraction. This process has been realized at PA “Mayak”.

The composition of the actinide fraction depends on a separation method:

A. Radionuclides of Am and Cm with minor amounts of U, Np and Pu.

B. Radionuclides of Am and Cm with minor amounts of U, Np and Pu and also REEs.

C. Radionuclides of Am and Cm with minor amounts of U, Np and Pu and also REEs and perhaps some amount of Zr.

The fraction with composition A can be burned off in fast reactors whereas the fractions with compositions B and C should be immobilized in stable matrices. Alpha-Activity of actinide fraction is subsequently determined by 244Cm, 241Am, 244Am, 239Pu and 242Pu during a long-term storage. The storage time depends on conditions and can reach several hundreds of thousands years. The main requirements to matrices are chemical and radiation resistances. The fraction with composition B can be incorporated in perovskite ceramics whereas in case of composition C the additional zirconolite and/or pyrochlore phases are necessary.

We synthesized powder samples of NpxU3-xO8 with x between 0 and 0.33. Analyses of solids by x-ray absorption spectroscopy indicate that Np occupies U sites and is predominantly tetravalent. Lattice parameters vary smoothly as a function of Np concentration. The a and b parameters (parallel to structural sheets in U3O8) depend strongly on Np concentration below x = 0.3; however, extrapolation of a least-squares fit up to x = 1 (the maximum value of x if Np is tetravalent) suggests little expected change in a or b for values of x greater than approximately 0.5. The c cell parameter (perpendicular to the plane of the sheets) shows relatively small dependence on Np concentration for all values of x. The a and b cell parameters vary in a complementary fashion, such that the a-b plane and unit-cell volumes remain nearly constant for small x (below ∼0.2). The projected volume increase at higher x may be accommodated by expansion along c; the maximum volume expansion for NpxU3-xO8 is projected to be less than 0.2% up to x = 1. Our results demonstrate that NpxU3-xO8 forms a homogeneous solid solution up to (at least) x = 0.33, and that a complete solid solution may exist between α-U3O8 and NpU2O8; however, the structure of NpU2O8 may resemble that of β-U3O8.

The uptake of dissolved oxygen (O2) has been studied in bentonite suspensions in 0.1 M NaCl media at (25±2)°C. MX-80 and Montigel bentonites were used in concentrations varying from 18 to 73 g/L. The experiments were performed in a magnetically stirred closed glass vessel, in an N2-glove box. Redox potentials where measured with Pt-wires, and dissolved O2 was measured both with a membrane electrode and with an optode. The experiments with MX-80 show that dissolved O2 disappears in ∼5 days under these conditions. Redox potentials decreased from ∼ +500 to ∼ +125 mVSHE (versus Standard Hydrogen Electrode). The data for the Montigel bentonite show similar time scales for O2 uptake but lower redox potentials at the end of the experiments ∼ −175 mVSHE. Pyrite oxidation is perhaps not the main process for O2 uptake, as MX-80 contains 0.3% FeS2 while Montigel bentonite only has a negligible amount.

Reaction sintering under pressure can be used to vitrify radioactive waste. We used defense waste simulate with high concentrations of refractory oxide ZrO2 (∼36 wt%). Silicate glasses with 30 to 50 wt% of waste loading were prepared by hot isostatic pressing (800 °C, 28 MPa). Amorphous silica (50 to 70 wt%) was added to calcined waste simulants. The reaction sintered final products were characterized by scanning and analytical electron microscopy. Our results showed that a continuous glass phase formed. Waste components such as Na2O, CaO, MnO2, Fe2O3, and Al2O3 dissolved in the glass phase. ZrO2 dissolved as well but reached its solubility limit. Nanometer size crystals of baddeleyite (ZrO2) suggest that their presence is a result of super-saturation of the glass phase rather than incomplete dissolution. The sodium concentration increases in the early stages of reaction sintering and decreases towards the end and so does solubility of ZrO2 and other oxides. The microstructure of reaction sintered glasses is determined by a) the concentrations of refractory oxides, b) the concentration of reaction sintering agents, c) the particle size of glass formers, and d) sintering conditions, e.g., temperature and time. Evidence is provided by respective analytical data and micrographs.

Ceramics within the compositional series Ca4-xGdxMn2Ti7O20+x/2 (x = 0, 1, 2, 3, 4) and a sample with Ca2U2Ti7O20 formulation were studied as promising matrices for immobilization of rare earth (RE) and actinide (An) constituents of high level waste (HLW). The samples were prepared by cold pressing of oxide mixtures in pellets at 200 MPa followed by their sintering at 1400 °C or melting at 1500 °C and examined with X-ray diffraction, scanning and transmission electron microscopy. At x=0 and x=1 a perovskite – pyrophanite assemblage occurred. The sample with x=2 consisted of murataite and perovskite. Murataite was a major phase in the sample with x=3 (pyrochlore and perovskite were minor phases) and the only phase in the sample with x=4 prepared under oxidizing conditions (in air). The latter was composed of two murataite varieties with seven- and five-fold fluorite unit cells. The sample with the same formulation but synthesized under reducing conditions contained pyrochlore as an extra phase. Coupled substitution 2 Gd3+ = Ca2+ + U4+ resulted in formation of pyrochlore as the major phase. Mu-rataite and perovskite are considered as the host phases for rare earths and actinides mainly trivalent, including Pu(III), Am(III), and Cm(III), and corrosion products (Mn, Fe, Al) whereas pyrochlore is the host phase for rare earths and tetravalent actinides (U(IV), Np(IV), Pu(IV)). This makes the system of calcium, gadolinium, manganese, and titanium oxides prospective for immobilization of RE – An fraction of HLW containing minor corrosion products (iron group elements).

The Gas Migration Test in the engineered barrier system (GMT) investigates the migration of waste-generated gas from low and intermediate level waste in a silo-type disposal concept. The EBS has now been emplaced and saturation was initiated in August 2001. The saturation patterns show heterogeneity within and between different layers of the EBS. Plans for the remaining test sequence are also presented.

Titanate ceramics have been proposed as candidate materials for immobilizing excess weapons plutonium. This study focuses on the characterization of a titanate-based ceramic through X-ray diffraction (XRD), electron probe microanalysis, and electron energy-loss spectroscopy (EELS). Three distinct phases have been identified, and their volume fraction was determined from element distribution maps using Scionimage-NIH Analysis software. This analysis revealed that the pyrochlore-group phase betafite (A2Ti2O7) forms the matrix of the ceramic and occupies 90.4% of the volume. Uniformly distributed in this matrix are perovskite (A2Ti2O6) and Hf-enriched rutile (TiO2), which account for 6.4 vol% and 3.1 vol%, respectively. The studied ceramic exhibits a very low porosity (0.3 vol%), which is characterized by small (<6 μm), rounded and isolated voids. In the studied ceramic, A-site cations are represented by Ca, rare earth elements, and Hf. The powder XRD pattern of the ceramic allowed refining the unit cell parameters for the cubic betafite, which is characterized by a cell edge of 10.132±0.003Å. The EELS data indicate that Ce is present as both Ce3+ and Ce4+ in betafite, whereas in perovskite, all Ce is trivalent.

Samples of Alloy 22 were tested in solutions containing various anions in order to determine their effect on the corrosion of the alloy. It was found that Alloy 22 is relatively corrosion resistant in HCl and HNO3 at pH 1 and 160°C (general corrosion rates on the order of 10 μm/year), but more susceptible to phosphoric acid, especially under reducing conditions. The presence of fluoride raised the corrosion rate of Alloy 22 to the order 1 mm/year at pH 1, and fluoride is still active towards Alloy 22 at pH levels as high as 3.5. Samples tested in solutions of 1000xJ13 in which the pH was altered during testing showed an increase in corrosion rate over solutions of constant pH. Preliminary electrochemical tests suggest that nitrate may be an effective corrosion inhibitor in fluoride containing solutions, while sulfate is not.

With respect to the assessment of the long-term behavior of the waste form spent fuel it is of high importance to study the fuel alteration in contact with groundwater and near field materials. The aim of this work is to evaluate the impact of candidate backfill materials hydroxylapatite and magnetite on the overall corrosion behavior of this waste form in salt brine; both materials are used in corrosion tests together with spent fuel. The instant releases and the matrix dissolution rates appear to be similar in presence and in absence of any backfill material under study. However, Am,Np,Pu,U and Sr are retained at different ratios on the hydroxylapatite, on the magnetite and on the fuel sample, indicating possibly the formation of different radionuclide containing new solid phases.

The stability constant for isosaccharinic acid dissociation is of importance in complexation studies involving species present in radioactive waste. As there are very few stability constants reported in the literature, this study has determined stability constants at 22.8, 34.5 and 44.2° C, using potentiometric titrations in a 1 M NaClO4 solution. The stability constants were linear functions of the inverted temperature, thus allowing for a simple determination of the entropy and enthalpy of the reaction.

Results of migration experiments with fractured drill cores performed in laboratory and in the CHEMLAB 2 probe under in-situ conditions at Äspö HRL are presented. Drill cores for both experiments are prepared by the same method and provide similar hydraulic properties. As tracers, actinides Am(III), Pu(IV) and Np(V) are applied. Breakthrough of Np is found to be unre-tarded in comparison to inert HTO tracer. Recovery of Np amounts to less than 40%. Am and Pu are not eluted from the cores. Lower limits of the retardation factor of 135 are calculated for both Am and Pu. Post mortem investigations of the fractured cores are performed by cutting perpendicular to the cylinder axis and subsequent chemical and radiochemical analysis of abraded material and slices. Imaging of the slices reveals the geometry of the flow path. α-radiography of the slices shows similar distribution patterns of Np and Am. It is shown by TTA extraction that Np bound onto the slices is in the tetravalent state. Hence, Np(V) undergoes reduction during migration. Retention of Am and Np is attributed to different processes.