A theoretical approach has been developed that allows one to obtain thermodynamic properties of plutonium-based alloys and compounds from first-principles electronic structure calculations based on density functional theory. The approach is applied to study the defect structure in non-stoichiometric PuO2±δ.

We report on the spectroscopic analysis of several samples relevant to the processes governing the behavior of oxidized uranium species in groundwater solutions under anoxic conditions. Both Fe samples with different times of exposure to the U(IV) solution and Fe metal-solution inetrfaces in the liquid cell ex-situ and in-situ, respectively. Resonant inelastic soft x-ray scattering is shown to be sensitive to the chemical state of uranium. The measurements were performed at a number of energies of the primary photon beam across the U 5d absorption edge. The results unambiguously indicate the reduction of U(VI) to U(IV) on the Fe surface.

An alkali-containing (mainly sodium) aluminophosphate glass waste form was studied by analytical electron microscopy. Bright-field imaging showed that small bubbles formed under the electron irradiation even at a low electron dose (8×1022e/m2). These bubbles grew with the increase of the electron dose and were finally released at the surface of the sample. At the same time alkali elements were also lost under irradiation. At an electron dose of about 2.2×1026e/m2 (6.6×1011Gy), all the bubbles were released and no bubbles were formed with further irradiation. The glass was finally transformed to a pure aluminophosphate glass. Further irradiation resulted in the phase separation between the Al-rich and the P-rich phases. The electron irradiation damage effects on the aluminophosphate glass are compared to those observed in iron phosphate glasses.

Simplified glass compositions were chosen to improve our knowledge of the alteration kinetics of complex glasses dedicated to the confinement of high-level waste. Since 1998, the sodium borosilicate glass system is at the center of a passionate debate between an affinity-based kinetic rate law and a protective surface layer theory. All the authors who have investigated ternary 68/14/18 SiO2–B2O3–Na2O glass agree on the fact that the affinity law cannot satisfactorily account for its alteration kinetics. Some authors explained that these discrepancies between classical kinetic rate law and experimental findings could be due to macromolecular amorphous separation in the 68/14/18 sodium borosilicate system and that this simplified glass could be divided into 90% reedmergnerite (NaBSi3O8) and 10% diborate (Na2O–2B2O3). This article provides evidence of the homogeneity of ternary 68/18/14 SiO2–B2O3–Na2O glass at nanometric scale and shows that even phase separation at less than nanometric scale could not explain the inability of hydrated glass-solution affinity laws to describe its alteration. The relative simplicity of the SiO2–B2O3–Na2O chemical system allows a critical examination of the macroscopic alteration laws developed over the last twenty years based only on the hydrated glass-solution chemical affinity without taking into account the formation and reactivity of the gel or its passivating properties.

A new approach is proposed for discussing the reliability and predictability of the models intended to evaluate the performance of glass packages under geological disposal conditions. The r(t) model developed by the CEA is used in this study to simulate original laboratory test results, and the validity of the predictions is then verified experimentally. This approach allows us to check that the key mechanisms are correctly simulated, even if they are simplified. A review of the experimental results (glass alteration kinetics and morphology of the alteration film) suggests that physically separating the glass and clay under geological disposal conditions could considerably diminish the silica pump effect attributed to reactive clays such as Boom clay or FoCa7 clay.

This work examines secondary phases formed on the surface of radioactive borosilicate glass K-26 during 12 years of testing in the loamy soil. This glass has been produced to immobilise intermediate level operational nuclear power plant (NPP) radioactive waste. An altered layer has been formed at the glass surface that had been contacting with the repository environment. The layer is morphologically following the bulk waste glass. The layer is chemically inhomogeneous and non-uniform in thickness and structure. It contains in addition to the major vitreous phase, alpha quartz, calcite, aluminium hydrochlorides, and molybdenum compounds.

A mixed-alkali modified borosilicate base glass used in the vitrification of high-level nuclear waste (HLW) has been doped with a number of waste simulants to between 2 and 12 mol%. The simulants have been chosen to give two distinct series of glasses: one consisting of trivalent ions having the form M2O3

(where M is La, Bi, Al or Fe) and the other consisting of divalent simulants of the form MO (where M is Pb, Zn or Ba). An international standard Soxhlet leach test procedure was performed on each glass to study the effect of prolonged, moderate-temperature, dynamic water corrosion. Results of these studies show that, except for BaO, as the amount of simulant is increased, the amount of Na and Li leached decreases showing them to become more chemically resistant. These corrosion tests have been correlated to ionic (D.C.) conductivity measurements, which show a decrease in the conductivity of the glass as the amount of waste simulant is increased, and 11B magic-angle spinning nuclear magnetic resonance (MAS-NMR) studies, which have shown that, as more waste-simulant is loaded into the glasses the rate of conversion of [BO4] to [BO3] units increases. All of the data from these studies reflect the different network forming abilities of the divalent and trivalent cations.

This study looks at the interactions between simulated calcined high level waste from fuel reprocessing and mixed alkali borosilicate glass frit in the early stages of melting, and the possibility of the formation of yellow phase during these stages. Simulant “calcine” from a full scale inactive trial (Magnox: oxide “blend” 25:75) was pre-mixed with alkali borosilicate glass, to achieve a 25wt% waste loading, and melted at 1050°C at various times. It is shown that dissolution occurs in two separate stages; the first involves formation of a low density CsLiMoO4 fluid, which separates and forms a yellow/green layer on the surface of the melt, accompanied by some dissolution of rare- earth elements (Nd, Ce, Gd) and Zr from the waste into the glass matrix. The second stage entails more extensive migration of these rare-earth elements into the glass, and the disappearance of the surface layer on the melt. The glass appears more homogenized at the later stages of melting, but still contains undissolved particles of calcine after 16 minutes.

New glassy matrices, able to incorporate new highly concentrated radioactive liquid wastes (HLW), are being studied. Investigations were performed on rare earth-rich glasses, known as very durable matrices. The selected basic glass composition was (wt. %): 51.0 SiO2 – 8.5 B2O3–12.2 Na2O – 4.3 Al2O3 – 4.8 CaO – 3.2 ZrO2 – 16.0 Nd2O3. To determine both the environment around the rare earth in this glass and its evolution according to its concentration (1.3 – 30 wt. % Nd2O3), EXAFS (Extended X-Ray Absorption Fine Structure) spectroscopy at the LIII-edge of neodymium and optical absorption spectroscopy were used. By coupling these two characterisation methods, several hypotheses are proposed about the nature of the rare earth neighbouring in the glass.

Laboratory scale unit for registration of acoustic emission (AE) signals was used for integrity control of glass granules and ceramic bodies. AE waveform recorded in course of the temperature changes has been shown to allow for an unambiguous interpretation and may be used for nondestructive quality control of glass and ceramics bodies.

The redox state of Mo in glasses containing simplified simulated high level nuclear waste (HLW) streams has been investigated using Electron Spin Resonance (ESR) and X-ray Photoelectron Spectroscopy (XPS). Melts performed in highly oxidising or neutral (air) atmospheres contained Mo6+, but our study indicates that the Mo can be at least partially reduced to Mo4+ or Mo3+ by melting in a reducing atmosphere. The implications for glasses containing a full simulated HLW stream are that the formation of detrimental crystalline phases containing Mo6+ upon heat treatment may be avoided by reducing the oxidation state of Mo during melting.

Corrosion experiments with the French borosilicate glass SON 68 were conducted under gamma (60Co source) and alpha (cyclotron) irradiation conditions. Static tests with glass powder were conducted at 90°C under saturation conditions with synthetic solutions rich in Si, B and Na. The initial pH was 9.8 and the SA/V was 3970 m−1. For gamma irradiation tests with the highest dose (∼ 58000 Gy) the pH decreased by almost a unit, which lasted for two weeks. The ion-exchange between glass and solution was enhanced as evidenced by the increase of the Li-normalized mass loss within 93 days. The measured H2O2 concentration in the experiment with the glass was as high as 1.51 10−5. The alpha irradiation tests with a total dose of 1800 Gy did not affect the solution pH and therefore the leaching rate of the glass remained similar to that in the blank experiment after 59 days. However, the measured H2O2 concentration was as high as 2.32 10−5. This work indicates that high irradiation doses may enhance the ion-exchange process due to the pH decrease.

Vapour phase hydration studies of a blended Oxide / Magnox simulant high level waste glass were undertaken at 200°C, over a period of 5 – 25 days. The alteration of this simulant waste glass is characterised by a short incubation time of less than 5 days, leading to the formation of an alteration layer several microns thick. Following the incubation period, the alteration proceeds at a constant rate of 0.15(1)μmd−1. The distribution of key glass matrix (Si, Na) and waste (Cs, Zr, Nd, Mo) elements was found to vary significantly across the alteration layer. Vapour phase hydration leads to formation of surface alteration products, identified as smectite, zirconium silicate and alkaline-earth molybdate phases.

A simulated sodium bearing waste (SBW), which represented a type of high sodium and sulfate waste, was successfully vitrified in iron phosphate glasses (IPG), at a maximum waste loading of 40 wt%, using a cold crucible induction melter (CCIM). Scanning electron microscopy (SEM) and X-ray diffraction (XRD) showed that all of the IPG waste forms did not contain sulfate salt segregation or crystalline phases. The calculated composition and the average analytical composition obtained by Electron Probe Microanalysis (EPMA) were in good agreement. The major elements were uniformly distributed throughout the samples. The chemical durability of the IPG waste forms containing 40wt% SBW was evaluated by the product consistency test (PCT) and met current DOE requirements. IPG waste forms were melted at a relatively low temperature and for short times compared to borosilicate glasses. These advantages, combined with those of a significantly higher waste loading and the feasibility for CCIM melting, offer a considerable savings in time, energy, and cost for vitrifying this high sodium and sulfate waste.

The alteration kinetics of the French borosilicate glass SON 68 have been investigated in a dynamic system at 50°C and 90°C under solution saturation conditions. The pH was adjusted to 4.8, 7.2 and 9.8 with addition of chemical buffers or/and by bubbling CO2 in solution. In all experiments, Li and Cs leaching seems to be controlled by a diffusion process. The Li- and Cs-concentrations were used to calculate the sum of ionic exchange and matrix dissolution rates of the glass while Mo-concentrations indicate matrix dissolution. The final leaching rates in saturation condition of Mo were in the order of 10–4–10- g.m-2.d-1 in good agreement with those given in literature for static tests. The glass surface was studied by scanning and transmission electron microscopy (SEM, STEM) for analysis of the corrosion products and by infrared spectroscopy (FTIR) for water speciation and concentration. A good inverse correlation between the water content and the alkali concentrations released from the glass has been obtained. About three hydrogen ions replaced one alkali ion. Modeling of the experimental data using GM 2001 model gives water diffusion coefficients between 10–20 and 10–22 m2.s-1.

The behaviour of a simulant Magnox waste glass was investigated under the likely conditions of high temperature, very deep, geological disposal, viz. 760°C and 0.15GPa. Under these conditions partial crystallisation of the glass is observed with the formation of LnBSiO5 (Ln = Y, La, Pr, Ce, Nd, Sm and Gd), LiNaZrSi6O15 and a palladium (ruthenium) telluride phase. The significance of these findings is discussed in the context of the high temperature, very deep, geological disposal scheme.

Since 1990, the Laboratory of Applied Mineralogy and Radiogeochemistry of the V.G. Khlopin Radium Institute (KRI) has been developing several different types of crystalline host-phases acceptable for the economically feasible and environmentally safe immobilization of actinide wastes. We proposed that ceramics that are based on host phases similar to naturally occurring accessory minerals including zircon, (Zr,Hf,…)SiO4; hafnon, (Hf,Zr,…)SiO4; baddeleyite (monoclinic zirconia), (Zr,Hf,…)O2; tazheranite (cubic zirconia), (Zr,Hf,Ca,Ti,…)O2; garnet, (Ca,Fe,Gd,…)3(Al,Fe,Si,…)5O12; perovskite, (Ca,Gd,…)(Al,Fe,Ti,…)O3, and monazite, (La,Ce,…)PO4, are the most efficient materials for actinide immobilization in deep geological formations. Solid solution of Pu in zirconia, (Zr,Pu)O2, could be used as a ceramic nuclear fuel that is competitive with mixed oxide fuel (MOX). To date, the following crystalline materials doped with 239Pu, 238Pu and 243Am have been successfully synthesized and studied at KRI: zircon; hafnon; cubic and tetragonal zirconia; monazite; aluminate garnet and perovskite. The maximum actinide loading was (in wt.% el.): 239Pu -37; 238Pu-10; 243Am-23. All Pu-Am-doped samples were made in air atmosphere under glove boxes conditions. Polycrystalline (ceramic) materials were made by sintering or melting of sol-gel, co-precipitated hydroxides, oxalates and phosphates or ground oxide precursors; single crystals were grown by a flux method. It was demonstrated that all ceramic samples obtained are characterized by high chemical durability and typical normalized actinide losses in deionized water at 90°C do not exceed 10−2–10−3 g/m2 (without correction for ceramic porosity). However, investigation of long-term behavior of ceramic waste forms requires taking into account the results of accelerated radiation damage study and modeling of ceramic alteration by underground solutions. The principal features of Pu-Am-doped samples obtained so far at KRI and their synthesis conditions are discussed.

To investigate the resistance of cubic zirconia to accelerated radiation damage, which simulates effects of long term storage, 238Pu-doped polycrystalline samples of cubic zirconia, (Zr,Gd,Pu)O2, were obtained and studied using X-ray diffraction analysis (XRD), electron probe microanalysis (EPMA), optical and scanning electron microscopy (SEM), and modified MCC-1 static leach test. The ceramic material was characterized by the following chemical composition (from EPMA in wt.% element): Zr = 50.2, Gd = 15.4, Pu = 12.2. This corresponds to the estimated formula, Zr0.79Gd0.14Pu0.07O1.99. The content of 238Pu estimated was approximately 9.9 wt.%. The XRD measurements were carried out after the following cumulative doses (in alpha decay/m3 × 1023): 3, 27, 62, 110, 134, 188, 234, and 277. Even after extremely high self-irradiation, cubic zirconia retained its crystalline structure. All XRD analyses showed no phases other than a cubic fluorite-type structure. The following results of normalized Pu mass loss (NL, in g/m2, without correction for ceramic porosity) were obtained from static leach tests (in deionized water at 90°C for 28 days) for 4 cumulative doses (in alpha decay/m3 × 1023):

The results obtained confirm the high resistance of cubic zirconia to self-irradiation. This allows us to consider zirconia-based ceramic as the universal material that is suitable for actinide transmutation and geological disposal.

Nd-bearing zirconolite was leached at 90°C for 6 months in 0.001M citric acid, and also in deionised water, to determine the effect of organic-bearing solutions on durability. The pH of the citric acid solution was adjusted to 5 using KOH, approximating that of the water in the parallel tests, to avoid the influence of pH on chemical durability of the zirconolite.

Releases were incongruent in the tests carried out in water. Release rates of Ti, Zr and Nd were comparatively very low (commonly too low to be measured) over the first 80 days of leaching. Rates for Ca and Al were 2 to 4 orders of magnitude higher than Ti, Zr and Nd over this same period. At about 80 days, there was an anomalous decrease in pH from 6 to 4 which enhanced release rates of Ti and Nd in particular. There was development of titania crystals, and the suggestion of hydrolysed titania, on the surface after 6 months. Thermodynamic equilibrium between the leachates and hydrolysed species on the surface of the zirconolite may be the key to apparent cessation of alteration, at least during thefirst 80 days of leaching.

By contrast, zirconolite leached in 0.001M citric acid maintained release rates of Ti, Zr and Nd 2 to 4 orders of magnitude greater than those in water for the first 80 days, values sustained, within an order of magnitude, for the remainder of the leach tests. Releases were congruent. The surface of the zirconolite showed no signs of secondary phase development. This suggests complexation by citrate ions prevented control by hydrolysed species on zirconolite solubility.

Synthetic murataite, an isometric, derivative of the fluorite-structure, has been proposed as a potential host phase for the immobilization of rare earth elements (REE) and actinides. A 1 MeV Kr+ ion irradiation has been performed on synthetic murataite ceramics in the system Ca-Ti-U-Mn-Al-Zr-Ce-O for different structural multiples of the fluorite unit cell. The temperature dependence of the amorphization dose has been determined. A higher critical temperature was obtained for the disordered murataite as compared to that of murataite superstructures, suggesting that murataite becomes more “resistant” to ion beam-induced damage with increasing degrees of structural disorder.