The fission products resulting from reprocessing of commercial spent fuel are currently vitrified industrially by COGEMA at La Hague. The properties of 21 non-radioactive borosilicate glass samples containing between 4 and 6% of the platinum-group metals (PGM: Pd and Ru) compared with 1.6% in the industrial glass were investigated for chemical composition variations covering the full specification range. After a brief morphological description of the undissolved PGM in the glass, the viscosity variations at temperatures ranging from 1100 to 1200°C are discussed with emphasis on the effects of the particle inclusions on the rheological properties of the glass. Variations in the chemical durability of quenched glass specimens are then discussed. The initial leach rate V 0 at 100°C remained near the values obtained for the reference glass. The same tests were repeated on glass heat-treated to obtain maximum crystallization, and the results confirmed that the chemical durability of the glass is practically unaffected by the crystallization observed in this type of glass.

The high chemical durability of iodine-bearing apatite makes it strongly prospective for conditioning of radioactive iodine. The synthesis and consolidation of iodine-bearing compounds require low temperatures to avoid iodine volatilization. Spark plasma sintering therefore appears to be a suitable process because of its shorter treatment time and lower sintering temperature compared with other processes such as HUP or HIP. Two alternatives were examined: SPS sintering of iodine-bearing apatite powder and SPS reacting of a stoichiometric lead iodide and lead phosphovanadate powder mixture. The degree of densification and the microstructure of bulk materials in both cases are described and compared. Reactive sintering appears to involve a three-stage mechanism: (i) PbI2 coalescence, (ii) solid-state iodoapatite synthesis and consolidation and, (iii) iodoapatite consolidation in the presence of a liquid phase. The SPS reacted products reveal the finest and most homogeneous microstructure, and a density exceeding 96%.

In this study, the use of combustion synthesis for immobilization of 14C was considered. Ceramic matrices have been prepared by this method using two different devices: one non-conventional with preheating of the samples and the other conventional device where ignition was produced thanks to tungsten filament. These two devices gave rise to different mechanisms of reactions involving different amounts of unreacted carbon graphite inside the matrix. The SHS samples were characterized by using scanning electron microscopy (SEM) and X-ray diffraction (XRD).

Lead phosphovanadates can be used as reactants for the synthesis of iodoapatite. Because of its high chemical durability, iodoapatite has considerable potential interest for immobilizing radioactive iodine. Iodine-bearing compounds must be synthesized and consolidated at low temperatures to avoid iodine volatilization. Spark plasma sintering (SPS) thus appears to be a suitable sintering process because of its short processing time. This paper deals with spark plasma sintering of lead phosphovanadate powder prepared mechanically by attrition and planetary ball milling. The influence of sintering parameters such as the heating rate, temperature, and holding time on the degree of densification and the microstructure of bulk materials is discussed. The bulk characteristics were directly correlated with the shrinkage curves. The powder characteristics were determined (grain size and size distribution, specific area, crystallite size, etc.) to explain the singular sintering behavior of the attrited powder; we also investigated whether the latter exhibited the same singular behavior during conventional sintering and hot pressing.

The hollandite structure-type has received considerable attention as a nuclear waste form for the incorporation of radioactive Cs-135 and Cs-137, both of which are important fission product radionuclides in the high-level nuclear waste generated by the reprocessing of used nuclear fuel. A critical concern has been the effects of high doses of ionizing radiation from incorporated Cs on the long-term structural stability of the hollandite structure. Optimization of the synthesis conditions has resulted in the hollandite stoichiometry of Ba0.85Cs0.26Al1.35Fe0.77Ti5.90O16. To evaluate the effect of Cs-beta-decay on this stoichiometry, we have simulated the ionizing radiation using 200 kV electron beam using transmission electron microscopy (TEM) at 298 and 573 K. Complete amorphization was achieved at doses of 1.1 x 10(14) and 1.8 x 10(14) Gy at temperatures of 298 and 573 K, respectively. Electron energy-loss spectroscopy (EELS) of the Cs M-edge revealed the selective loss of Cs at the maximum doses. Hollandite irradiated using gamma rays, similar to 10(6) Gy, which has defects associated with the formation of Ti3+ and O-2(-) had a dissolution rate similar to that of the pristine hollandite, suggesting that the initial stage of defect formation does not influence chemical durability. Because the accumulated dose in the hollandite with 5 wt% of radioactive (Cs2O)-Cs-137 is estimated to be similar to 2.0 x 10(10) Gy after 500 years, the hollandite structure should be stable under the conditions anticipated for geologic disposal.

X-ray patterns of the Ba1Cs0.28Fe0.82Al1.46Ti5.72O16 compound evidence two wide peaks at low angle in addition to the well defined peaks of the I4/m hollandite structure type. Two hypotheses have been explored to account for these features: the coexistence of the hollandite phase with an amorphous phase and the appearance of a commensurate or incommensurate modulated structure associated with a cationic ordering, as proposed in the literature. Actually, even if the amorphous phase quantification by the Rietveld method reveals about 15wt% of non-crystalline phase in some of the powdered sample, the origin of the two wide peaks was found to stem from the incommensurate modulated character of the hollandite structure type (super space group I4/m(00γ)00) with a distribution of the modulation wavevectors presumably related to slight chemical composition changes.

The hollandite Ba1Cs0.28Fe0.82Al1.46Ti5.72O16, which has been proposed for the cesium-specific conditioning, can be synthesized either by an alcoxyde or a dry route. In both cases, a two-step protocol is applied, i.e., a calcination at 1000°C followed by a sintering at 1200°C. After sintering, both synthetic processes lead to a tetragonal form. According to the X-ray diffraction (XRD) patterns collected at the barium and the cesium K absorption edges, the different positions of these two elements have been evidenced with a more centered position in the oxygen cubic site of the tunnel for Ba than for Cs. On the contrary, after calcination, the two synthetic routes yield different products. The alcoxyde route gives rise to a mixture of the aforementioned Cs- and Ba-containing tetragonal I4/m hollandite, a Cs-only-containing monoclinic I2/m hollandite and an unidentified phase with a weak coherence length containing only Ba. The dry route yields a single tetragonal hollandite material containing Ba and Cs slightly different in composition from the targeted compound.

In France, there is strong interest in developing new ceramic wasteforms for containment of separated long-lived radionuclides. Barium hollandite (BaAl2Ti6O16) ceramics are suitable for cesium immobilization by virtue of their loading capacity and high chemical durability. A 5 wt% CS2O-doped Fe-substituted hollandite was synthesized at laboratory scale (100 g), using an alkoxide process. Hollandite crystallization was achieved after calcination for 5 hours at 1000degreesC in air. After milling and cold pressing, ceramic pellets were sintered in air for 15 hours at 1250degreesC. The resulting materials were dense single-phase pellets. Leaching experiments were carried out to assess the chemical durability of this specific hollandite. Initial leaching rates were measured at 1000degreesC using a Soxhlet apparatus, and long-term alteration rates at 90degreesC were determined by static leaching of powder samples. The chemical durability of iron hollandites, loaded with 5 wt% cesium oxide was very good, and comparable to that of Synroc hollandite.

Studies have been conducted in France to minimize the potential long-term impact of nuclear waste by enhanced chemical separation of the minor actinides (Np, Am, Cm) and some long-lived fission products (I, Cs, Tc). Two options may be considered following this work: (i) the initial reference option is transmutation by neutron bombardment in nuclear facilities, (ii) the second option would be to incorporate the separated elements into an inorganic matrix ensuring long-term stability. In the case of specific conditioning, zirconolite and hollandite are the potential host phases for the minor actinides and caesium, respectively. Both of these matrices have shown strong potential: (i) for incorporating the respective radioelement in the crystalline structure, (ii) for fabricating the ceramic by natural sintering in air, (iii) for chemical durability with a very low initial alteration rate (about 10(-2) g m(-2) d(-1) at 100 degreesC), then very rapidly reach alteration rates more than four orders of magnitude lower. In the case of zirconolite ceramics, the high chemical durability is conserved even after amorphization of the crystalline structure by external irradiation with heavy ions or by self-irradiation in natural zirconolites 550 million years old. (C) 2004 Academie des sciences. Published by Elsevier SAS. All rights reserved.

A new glass formulation, a calcium and zirconium-enriched alumino-boro-silicate matrix, able to incorporate 12 wt% of molybdenum oxide has been developed in order to trap molybdenum as CaMoO4, a stable phase in a vitreous medium. Laboratory studies on this matrix have shown the impact of the glass thermal treatment on its microstructure. Thus, four cooling rates and heat treatments presented in this paper were investigated to determine if Ca and Mo totally reacted to give CaMoO4 and to localize the powellite phase in the different glasses. Using X-ray diffraction and Rietveld refinements, we were able to quantify both crystallized and amorphous CaMoO4 in all the samples. EDS analyses allowed us to localize it in the glass microstructure. Then coupling together the micrograph analysis and the X-ray diffraction results proved the presence of amorphous powellite and showed its place in the glasses.

A detailed study of the XPS binding energy shifts of Si 2p, O 1s and Zr 3d in a series of aluminosilicate glasses (a three oxide glass: SiO2–Al2O3–CaO, three four-oxide glasses: SiO2–Al2O3–CaO–TiO2, ZrO2 or CeO2, along with a six-oxide glass SiO2–Al2O3–CaO–TiO2–ZrO2–CeO2) is presented. Their composition is such that these glasses have the same mean electronegativity, so that no changes in the atomic charges is expected. The binding energy shifts are interpreted in terms of initial and final state effects, and the balance of charge transfer contribution and electrostatic effects is discussed. Referred to the ternary glass, the binding energy shifts of the Si 2p, O 1s and Zr 3d lines in the complex glasses are due to an initial state effect, as the extraatomic relaxation is similar along the glass series. These shifts originate from electrostatic Madelung effects, likely coming from a structural change induced by the presence of the other oxides TiO2, ZrO2 and CeO2. These structural effects make it impossible to deduce information on the atomic charges directly from the binding energy shifts, even in a qualitative way.