Radon emanation thermal analysis (ETA) during ”step by step”heating to selected temperatures up to 1200oC in argon and subsequent cooling yielded detailed information on the annealing of radiation damage and macroscopic defects in a natural brannerite. Thermogravimetry showed mass losses in the temperature ranges 230-315, 570-760 and 840-1040oC. Mass spectrometry of the evolvedgases indicated M/Z = 44, i.e. CO2. Crystallization of initially amorphous brannerite was indicated by ETA in the temperature range 900-1020oC, in agreement with the X-ray diffraction results. From the radon diffusion activation energy values determined in ETA, the degree of structural ordering was assessed at different temperatures.

Murataite –a complex actinide (An)- and rare earth element (REE)-bearing oxide with a cubic fluorite-related lattice – is one of the promising host-phases for immobilization of Pu-containing waste. The murataite phase composition corresponds to the empirical formula:A4B2C7O22-x, where the A-sites are occupied by Ca, Mn, REE, and An (U); B sites – by Mn, Ti,Zr, and An (U); and C sites – by Ti, Al, and Fe. The total amount of the actinides (U) and REE (Ce, Gd) in the murataite may exceed 20 wt%. In contrast to the other prospective hosts for actinide waste immobilization (cubic zirconia and pyrochlore), murataite accommodates higher amounts of corrosion products(Al, Fe) along with the actinides. The authors compared murataite-based ceramics having similar compositions and produced by melting in a high-temperature resistance furnace or via inductive melting in a cold crucible. Eight samples of the murataite-based ceramics were produced and investigated in detail. Murataite was found to be the major phase in four of the samples – with a basic composition, in wt%, of: 5.0 Al2O3, 10.0 CaO, 55.0 TiO2, 10.0 MnO, 5.0 Fe2O3, 5.0 ZrO2, and 10.0 UO2. These samples were produced by melting in a resistive furnace and in the cold crucible and included Gd-bearing samples and one Pu-bearing sample. The extra phases were other titanates: (from more to less typical) rutile, pyrochlore, zirconolite, crichtonite, pseudobrookite, and perovskite (in the Pu-doped samples only). Three varieties of the murataite, with 3-, 5-, and 8-fold fluorite-type lattices, were observed. Addition of uranium and rare earth oxides stabilizes pyrochlore as the major phase, whereas addition of zirconia yields zirconolite. Plutonium stabilizes the perovskite-type phase, probably due to the formation of Pu3+. The maximum waste oxide content in the murataite for the elements studied was found to be 10% ZrO2, 12% CeO2, 13% Gd2O3, and 14% UO2. Waste element partitioning among the murataite and all the other phases with similar fluorite-related structure (pyrochlore and zirconolite) was analyzed. The uranium leach rate for the sample with maximum murataite content was measured using a procedure similar to MCC-3. This leach rate was close to10-5 g/(m2*day) in a½-day test and decreased by more than one order of magnitude in 28-day tests. Investigation of the stability of the murataite structure after irradiation byheavy ions is in progress.

The effects of different methods of grinding or mechanical milling of precursor mixtures were determined for the structure and properties of perovskite ceramics with the formulation Ca1-xGdxTi1-xAlxO3 (0 ≤x ≤ 1). The ceramics were prepared by coldpressing finely ground precursor mixtures at 100-300 MPa to form pellets that were then sintered at 1300-1500oC. Ceramic samples prepared from precursor material batches treated in a ball or planetary mill or ground in a mortar contained unreacted oxides and had low mechanical integrity and chemical durability. The ceramics produced from precursor material treated by high performance mechanical milling using a high speed rotating vortex layer of ferromagnetic bodies were composed of a major perovskite-structured phase (90-95%) with a Ca-Ti-Gd-Al-O composition and a minor pyrochlore-structured phase with a composition close to that of gadolinium titanate; these ceramics had the highest bending strength (200-340 MPa) and density (~90% of theoretical) and the lowest leach rates for uranium, plutonium, and americium among all the samples studied.

Optical emission spectra in the 300-700 nm range were collected from zirconolite and rutile specimens irradiated with a 3 μs pulsed electron beam using a Febetron 706 variable energy pulsed electronbeam generator. The long-lived emissions (up to microseconds after the electron pulse) consist of broad (halfwidths ~ 100 nm) bands centred around ~400 nm. Over the range 0.2 MeV to 0.6 MeV, the emission intensity per unit dose versus electron beam energy data from the rutile sample showed a single stage dependence on electron beam energy, whereas the zirconolite data suggested a two stage dependence. Rutile has a threshold of 0.23 ½ 0.02 MeV, which gives an Ed value of 39 ½ 4 eV for oxygen. Zirconolite has a threshold of 0.26 ½ 0.02 MeV, which gives an Ed value of 45 ½4 eV for oxygen. These data are discussed in the context of previously measured and calculated Ed values for other oxides.

Rb-doped Ba-hollandite and synroc-C samples were prepared for aqueous dissolution studies at 90°C in deionized water using MCC-1 protocols. Small amounts of possibly Rb-bearing BaTiO3, hibonite and alumina accompanied the major Rb-doped (0.1 formula units) hollandite phase. The Rb loadings in the synroc samples ranged from 0.89 to 2.67 wt%. Rb leached at <0.2 g/m2/d from hollandite samples and <0.1g/m2d from synroc, with the values decreasing with increasing dissolution time. The variance between the dissolution rates for the different Rb-doped hollandite and synroc samples is discussed, in terms of Rb-bearing minor phases.

UO2 containing short-lived α-emitters, the so-called α-doped UO2, can simulate type (i.e. α- decay) and level of activity of spent fuel at the time when it might become exposed to groundwater in a geologic repository during storage. This allows studying α-radiolysis effects on the dissolution of the fuel matrix. Additionally, UO2 with high concentrations of α-emitters accumulate, during experimentally acceptable short times, the amount of decays, hence of property modifications, corresponding to long storage times for spent fuel. UO2 containing ∼10 and ∼0.1 wt% 238Pu was fabricated and tested. Leaching experiments in deionized water under unaerated conditions, with continuous monitoring of the evolution of the redox potential and pH in the leaching solutions, were performed. The Eh measurements showed a fast increase of the redox potential in the case of the material with the highest α-activity, while the UO2 containing ∼0.1 wt% 238Pu increased its potential more slowly. The redox potential for undoped UO2 decreased steadily during the experiment. As previously observed, higher fractions of U were released in the case of α-doped UO2 compared to undoped UO2. The fractions of U and Pu released during leaching from the α-doped materials were very similar, suggesting that congruent dissolution occurred. After leaching times longer than 10 h, only dissolved species were present in the solutions. Under these experimental conditions, characterized by relatively low values of the ratio sample surface/leachant volume, a dependence of the released amounts on the α-activity of the samples was observed. Periodical measurements of parameters like hardness, showed a rapid buildup of radiation damage in the material with the high α-activity. After more than two years, noticeable changes, namely an increase of the hardness, have begun to be observed also for the material with the low concentration of 238Pu.

For a potential performance assessment of direct disposal of spent fuel in a nuclear waste repository, the chemical reactions between the fuel and possible intruding water must be understood and the resulting radionuclide release must be quantified.

Leaching experiments were performed with five spent fuel samples from French power reactors (four UO2 fuel samples with burnup ratings of 22, 37, 47 and 60 GWd·tHM−1 and a MOX fuel sample irradiated to 47 GWd·tHM−1) to determine the release kinetics of the matrix containing most (over 95%) of the radionuclides. The experiments were carried out with granitic groundwater on previously leached sections of clad fuel rods in static mode, in an aerated medium at room temperature (25°C) in a hot cell.

After 1000 or 2000 days of leaching, the Sr/U congruence ratios for all the UO2 fuel samples ranged from 1 to 2; allowing for the experimental uncertainty, strontium can thus be considered as a satisfactory matrix alteration tracer. No significant burnup effect was observed on the alteration of the UO2 fuel matrix. The daily strontium release factor was approximately 1 à 10−7 d−1 for UO2 fuel, and five to six times higher for MOX fuel. Several alteration mechanisms (radiolysis, solubility, precipitation/clogging) are examined to account for the experimental findings.

The dissolution rate of unirradiated uranium dioxide was studied in batch experiments as a function of hydrogen peroxide concentration (from 10−5 to 10−3 mol dm−3). Unirradiated UO2(s) was used in order to differentiate surface chemical processes from radiolytic effects. Dissolution rates were determined from both uranium release and hydrogen peroxide consumption. Results showed that H2O2consumption rate was higher than UO2 dissolution rate. This observation may indicate that the overall UO2 oxidative dissolution process would be controlled by the dissolution of the oxidized solid surface. The calculated hydrogen peroxide reaction order was 1 in the H2O2 concentration range from 10−5 to 10−4 mol dm−3, while at higher concentrations no clear dependence was observed.

With respect to the assessment of the long-term behavior of the waste form spent fuel it is of high importance to study in particular the UO2 matrix dissolution behavior and the associated release/retention of radionuclides in contact with groundwater. During long term fuel storage, fuel oxidation may not be avoided. Main issue of this work is to identify the impact on the corrosion of partly oxidized fuel of environmental conditions such as (1) the nature of solution contacting the matrix, the (2) presence/absence of CO2, (3) fixed pH values within a range between pH 7- pH 11, and (4) the presence/absence of corroding container material (Fe-powder). Dissolution tests with powdered oxidized spent fuel in various granite waters, and NaCl-brine resulted in matrix dissolution rates in the same order of magnitude for all investigated media (ca.5×10−4/d). The presence of CO2 and fixed pH values (pH 5 – 11) was without a distinct effect. The independence of the dissolution rate of the oxidized fuel matrix upon the nature of solution, pCO2, fixed pH values (5-11) can probably be explained by a masking effect of radiolysis. In presence of Fe powder the matrix dissolution rate was found to be slowed down by a factor of ca. 20, associated with strong retention effects of radionuclides.

The results of leaching tests on the synthesised 46%Mo-29%Ru-21%Pd-4%Rh alloy show selective release Mo at both reducing and air saturated conditions. The leach rate increases with rising redox potential (Eh) of leaching solution and temperature, and slightly decreases with leaching time. The electrochemical study indicates that the alloy is more stable than carbon steel against corrosion but less stable than copper and stainless steel. Based on the leach rates of Mo, Ru, Rh, Pd and oxidation potentials of these four metals and Tc0, the leach rates of Tc from alloy aggregates, 40%Mo, 30%Ru, 10%Tc, 15% Pd, 5%Rh in spent fuel at reducing and air saturated conditions are estimated, to be 4.10−9 g.cm−2.day−1 and 4. 10−7 g.cm−2.day−1, respectively.

The chemical stability of spent fuel will be greatly influenced by the redox potential of the near field. Presence of reductants such as iron is likely to be an important factor to maintain the original integrity of spent fuel. In this work experimental data about the influence of metallic iron (container base material) on SIMFUEL leaching behavior under simulated granite and saline repository conditions is presented. In the presence of iron uranium concentration undergoes a sharp decrease. This is much more noticeable in the experiments performed under initial oxic conditions. The effect of iron on simulated fission products of SIMFUEL is very important for the elements with high redox sensitivity such us molybdenum. On the contrary, strontium remains stable during the entire tests and it seems not be affected by changes in redox potential.

The long-term dissolution behavior of spent PWR fuel in synthetic granitic groundwater has been investigated since June of 1998 in order to identify the release mechanism of spent fuel in contact with domestic Ca-bentonite, which was compacted as a density of 1.4g/cm3. Several spent fuel specimens were cut to around 3-mm thick, and these specimens were then loaded into each leaching cell. Every half year, one specimen was collected and the bentonite block was sliced into several pieces. The element distribution on the specimen surface was measured by EPMA and an optical microscope, and the gamma activities in the sliced bentonite pieces were measured. All the leachates were sampled and then the gamma and gross alpha activities were measured. By the results, coming from leaching up to about 1.5 years, the fraction of radiocesium released through the compacted bentonite layer from the spent fuel is approximately a hundredth lower than that without the bentonite layer. By the depth profile of the gamma activity on the bentonite, the activity gradually decreased in accordance with more distance from the specimen and the fractional release rates of cesium were nearly the same. On the other hand, the EPMA results indicate that more time and more information is required in order to identify the surface alteration of spent fuel by leaching. This experiment will therefore be continued at least into the year 2006.

In a spent fuel repository the processes that govern the release of radionuclides are dissolution and transport in a possible groundwater flow. The cladding will be the last barrier before the water comes into contact with the fuel, namely with the outer rim of the pellet. Here the heterogeneity of the material due to the irradiation process is responsible for a complex release process. Fission products and minor actinides inventories are considerably higher at the pellet periphery as a result of increased epithermal neutron capture and of migration in the case of the volatile fission products.

The present paper gives a review of experimental activities at the Institute for Transuranium Elements (ITU). Both single effects studies and integral tests are carried out to study the behavior of spent fuel under storage conditions.

Leaching of irradiated UO2 (up to 50 GWd/tU) and MOX (up to 25 GWd/tU) fuel rods with preset cladding defects at 100°C under anoxic or reducing conditions should simulate the realistic case of groundwater coming into contact with a spent nuclear fuel repository. For all main radionuclides the release process can be described considering a two-step dissolution mechanism that includes the initial dissolution of an oxidized layer present on the fuel surface followed by a long-term oxidative matrix dissolution. By means of α-doped (238Pu) UO2 it could be demonstrated, that radiolysis has a significant influence on this dissolution. Especially high initial release rates were found for the volatile cesium and iodine for the reasons mentioned above.

Besides the conventional leaching experiments electrochemical techniques are used to investigate for instance the complex corrosion behavior of the heterogeneous MOX fuel materials or the influence of α-radiolysis on spent fuel dissolution.

In the integral tests mentioned above with large S/V values, reprecipitation of U is likely to happen. Therefore special dynamic test are conducted where this reprecipitation is prohibited and true U solubility can be determined.

Thin layer of UO2 and (U,Pu)O2 doped with various fission products and minor actinides are prepared to study the influence of these elements on the matrix dissolution. When Cs is for instance co-deposited, the U oxidation state changes from U4+ to U6+ for the same O2 pressure possibly indicating a stable Cs uranate. This could be an indirect proof of the existence of such a species in irradiated fuel (e.g. at the grain boundaries).

In the framework of the Enresa 2000 PA exercise and as a continuation of the developments made during SR 97, we have developed a conceptual and numerical model to calculate the release of radionuclides from spent fuel under repository conditions. The model includes both thermodynamic and kinetic considerations. Hence, although certain radionuclides are solubility controlled, for other radionuclides their release is governed by kinetic processes such as radiolytically promoted oxidative dissolution of the matrix and the associated water turnover inthe gap. The fluxes of selected radionuclides are calculated as an indication of the relative importance of the various processes considered to define source term concentrations in the performance assessment of the spent fuel repository.

Under the geological disposal conditions, spent fuel (SF) is expected to evolve during the 10,000 years while being maintained isolated from the biosphere before coming in contact with water. Under these circumstances, several driving forces would lead to the progressive intrinsic transformations within the rod which would modify the subsequent release of radionuclides. The major mechanisms are the production of a significant volume of He within the UO2 lattice, the accumulation of irradiation defects due to the low temperature which avoids any annealing, the slow migration of radionuclides (RN) within the system by (i) the α self-irradiation-induced athermal diffusion and (ii) locally the building-up of internal mechanical stresses which could turn the pellets into powder. However, the current RN source terms for SF have never accounted for this evolution as they are based on existing knowledge of the fresh SF. Two major mechanisms were considered, the leaching of the readily available fraction (one which was supposed to be instantly accessible to water), and the release of RN through alteration of the UO2 grains. We are now proposing a new RN source term model based on a microscopic description of the system in order to also take account of the early evolution of the closed system, the amplitude of which increases with the burnup and is greater for MOX fuels.

In Sweden, it is proposed that spent nuclear fuel should be encapsulated in sealed cylindrical canisters for disposal in a geologic repository. The canisters would consist of a thick ferrous inner container and a copper overpack. If mechanical failure of the copper overpack occurred, allowing water to enter, there would be a build up of ferrous corrosion product, which could induce stresses in the outer copper canister. This paper describes an apparatus, the ‘stress cell’, which was designed to measure the expansion caused by the anaerobic corrosion of steel under compressive loads. The apparatus consisted of a stack of steel and copper discs, which were immersed in simulated anoxic groundwater. A system of levers amplified the change in height of the stack, and the displacement was measured using sensitive transducers. Three cells were set up; two contained alternate mild steel and copper discs, and the third, a control cell, consisted of alternate stainless steel and copper discs. A slight contraction of the control cell was observed but no expansion was measured in the mild steel - copper cells.

In parallel, coupons of mild steel and cast iron were corroded in anoxic, artificial groundwater at 50°C and 80°C for several months. The coupons were examined by atomic force microscopy (AFM) to determine the mechanical properties and the structure of the corrosion product films, and X-ray photoelectron spectroscopy (XPS) to identify the chemical composition of the film.

In Sweden, it is proposed that spent fuel should be encapsulated in sealed cylindrical canisters for disposal in a geologic repository. The canisters would consist of a thick ferrous inner container and a copper overpack. If mechanical failure of the copper overpack occurred, allowing water to enter, the ferrous inner container would corrode anaerobically and liberate hydrogen. The rate of hydrogen generation due to the anaerobic corrosion of steel in anoxic groundwater has been measured using barometric cells. The aim of the work presented in this paper was to measure the redox potential, Eh and pH in the presence of anaerobically corroding steel, in a barometric cell. Two specially designed barometric cells were constructed. They were equipped with a silver chloride or calomel reference electrode, a gold Eh electrode, a glass electrode, and a steel electrode. The electrodes were allowed to stabilize in anoxic artificial groundwater and then a mass of pickled steel wire was introduced into the test cell. As the wires were added, the redox potential rapidly became more negative due to the rapid consumption of the residual oxygen. The corrosion potential of the steel was stable and a slow drift in pH was observed. The results are compared with the results of geochemical modelling. Extension of the work to investigate the electrochemical parameters in the presence of dissolved uranium species is discussed.

A repository for low and intermediate level radioactive wastes will contain a great deal of carbon steel as reinforcement materials and waste containers. These steel components are expected to generate hydrogen gas due to reductive corrosion of carbon steel under repository condition. In this study, we have investigated the influence of the environmental factors such as oxygen concentration and pH in the solution on the rate of the gas generation. This rate was measured with a gas flow type test system under the controlled oxygen concentration, which simulates the redox conditions from initial to final environment in a repository. In addition, the corrosion mechanism was examined through the electrochemical tests.

As a result, it has been found that pH in the solution and the oxygen concentration considerably affected the generation rate of hydrogen gas in various way. As to the influence of pH of the solution, it has been found that the gas generation rate increased with the increase of pH between 11.8 and 14. Further, as to the influence of the oxygen concentration, when the oxygen concentration in the blowing nitrogen gas was 2 ppm or more, the gas generation rate increased with the oxygen concentration. This trend was possibly caused by local decrease of pH due to localized corrosion. The gas generation rate was very low when the oxygen concentration was in the passive region. The rate was equal to 0.01 μm/y or less if the corrosion occurs with an equation: 3Fe+4H2O → Fe3O4 + 4H2 ↑ to 0.01 *m imcro;/y (hereafter we call this the equivalent corrosion rate). However the gas generation rate increased when the oxygen concentration was controlled to the extent that sufficient passive film cannot be produced and thus it shifts from passive state to active state.

Alloy 22 is the material preferred by the U.S. Department of Energy for the waste package outer container for geological disposal of high-level radioactive waste at the proposed site in Yucca Mountain, Nevada. Alloy 22 is considered to be extremely resistant to various modes of aqueous corrosion over broad ranges of temperature, pH, and concentration of anionic and oxidizing species. Uniform corrosion under passive dissolution conditions, localized corrosion in the form of crevice corrosion, and stress corrosion cracking are discussed on the basis of experimental results obtained with mill annealed, thermally treated, and welded specimens using electrochemical techniques. The approach developed for long-term performance prediction, including the use of empirically derived parameters for assessing localized corrosion and the modeling of the passive dissolution behavior, is described.

Ti grade 7 has been selected by the U.S. DOE as the current material of choice for the drip shield in the proposed high level waste (HLW) repository design. Due to the addition of Pd, Ti grade 7 exhibits enhanced resistance to hydrogen embrittlement (HE), yet there is relatively little data on HE of this material. Calculations of hydrogen absorption/recombination, solubility, and free energy of hydride formation in Ti and Pd are presented to qualitatively evaluate Keff, the stress intensity factor for crack propagation induced by hydride formation, of Ti grade 7 in relation to other Ti alloys without Pd. Calculations were performed that show concentration of hydrogen in Ti grade 7 may exceed the critical hydrogen concentration, Hc, where the material becomes embrittled, when accelerated passive dissolution of Ti grade 7 in concentrated Cl- and Cl-+F- solutions as the source of hydrogen is considered.