Initial Evaluation of Processing Methods for an Epsilon Metal Waste Form
During irradiation of nuclear fuel in a reactor, the five metals, Mo, Pd, Rh, Ru, and Tc, migrate to the fuel grain boundaries and form small metal particles of an alloy known as epsilon metal (ε-metal). When the fuel is dissolved in a reprocessing plant, these metal particles remain behind with a residue - the undissolved solids (UDS). Some of these same metals that comprise this alloy that have not formed the alloy are dissolved into the aqueous stream. These metals limit the waste loading for a borosilicate glass that is being developed for the reprocessing wastes. Epsilon metal is being developed as a waste form for the noble metals from a number of waste streams in the aqueous reprocessing of used nuclear fuel (UNF) - (1) the ε-metal from the UDS, (2) soluble Tc (ion-exchanged), and (3) soluble noble metals (TRUEX raffinate). Separate immobilization of these metals has benefits other than allowing an increase in the glass waste loading. These materials are quite resistant to dissolution (corrosion) as evidenced by the fact that they survive the chemically aggressive conditions in the fuel dissolver. Remnants of ε-metal particles have survived in the geologically natural reactors found in Gabon, Africa, indicating that they have sufficient durability to survive for ≈ 2.5 billion years in a reducing geologic environment. Additionally, the ε-metal can be made without additives and incorporate sufficient foreign material (oxides) that are also present in the UDS. Although ε-metal is found in fuel and Gabon as small particles (≈10 µm in diameter) and has survived intact, an ideal waste form is one in which the surface area is minimized. Therefore, the main effort in developing ε-metal as a waste form is to develop a process to consolidate the particles into a monolith. Individually, these metals have high melting points (2617°C for Mo to 1552°C for Pd) and the alloy is expected to have a high melting point as well, perhaps exceeding 1500°C. The purpose of the work reported here is to find a potential commercial process with which ε-metal plus other components of UDS can be consolidated into a solid with minimum surface area and high strength Here, we report the results from the preliminary evaluation of spark-plasma sintering (SPS), hot-isostatic pressing (HIP), and microwave sintering (MS). Since bulk ε-metal is not available and companies could not handle radioactive materials, we prepared mixtures of the five individual metal powders (Mo, Ru, Rh, Pd, and Re) and baddeleyite (ZrO 2 ) to send the vendors of SPS, HIP, and MS. The processed samples were then evaluated at the Pacific Northwest National Laboratory (PNNL) for bulk density and phase assemblage with X-ray diffraction (XRD) and phase composition with scanning electron microscopy (SEM). Physical strength was evaluated qualitatively. Results of these scoping tests showed that fully dense cermet (ceramic-metal composite) materials with up to 35 mass% of ZrO 2 were produced with SPS and HIP. Bulk density of the SPS samples ranged from 87 to 98% of theoretical density, while HIP samples ranged from 96 to 100% of theoretical density. Microwave sintered samples containing ZrO 2 had low densities of 55 to 60% of theoretical density. Structurally, the cermet samples showed that the individual metals alloyed in to ε-phase - hexagonal-close-packed (HCP) alloy (4-95 mass %), the α-phase - face-centered-cubic (FCC) alloy structure (3-86 mass %), while ZrO 2 remained in the monoclinic structure of baddeleyite. Elementally, the samples appeared to have nearly uniform composition, but with some areas rich in Mo and Re, the two components with the highest melting points. The homogeneity in distribution of the elements in the alloy is significantly improved in the presence of ZrO 2 . However, ZrO 2 does not appear to react with the alloy, nor was Zr found in the alloy.