Actinide behavior and radiation damage produced by α-decay in materials to solidify nuclear waste

Abstract

Solidification of High Level (radioactive) Waste, HLW, from reprocessing of nuclear reactor fuels in a suitable matrix and subsequent burying in a suitable repository are the envisaged final steps of the nuclear fuel cycle. Because of the long half-lives of some of the actinides (mainly Np, Pu, Am and Cm) and because of their decay by emission of an α-particle (∼5 MeV energy) and a heavy recoil atom ( e.g. U from Pu-decay, ∼100 MeV energy), the actinide behavior in the candidate materials for waste solidification deserves particular attention. The materials for solidification are glasses or certain minerals, or a mixture, so called glass-ceramics, i.e. ceramic phases in a glass matrix. Actinides often become enriched in these ceramics which are therefore also called host phases. As repositories for solidified HLW, salt dome, granite rock or clay layers are envisaged. p ]Main questions to be solved before realizing a repository concept and before selecting a HLW matrix are: corrosion of the solidified waste in water (so-called leaching) at realistic temperatures and pressures expected for repository conditions, mechanical and fracture properties and compatibility of the HLW matrix with the repository material. Once these properties have been determined for the as-produced HLW products, the possible changes brought about by the radiation damage accumulating during long time storage have to be investigated. Damage occurs by γ-rays, β-decay, (spontaneous) fission, and α-decay. The first 3 damaging sources contribute, however, much less to the total damage production than the α-decays. Within the α-decay, most of the (displacement) damage is due to the heavy recoil atoms. Two methods were therefore selected to simulate long time damage in shorther times. The most realistic simulation is the incorporation of a short-lived actinide that produces the same α-disintegrations per unit mass or unit volume in a reasonably short time (months to years) as would be produced under real storage conditions in some 10 3 years. Another versatile and fast simulation is external ion bombardment with ion beams of the energies (and of the approximate masses) of the recoil atoms of the α-decay. Often, Pb ions are chosen. In the present study, both methods were utilized and compared. In addition, damage was produced by irradiation with high energy α-particles. Different types of waste glasses, of glass ceramics and of host phases were studied. Before damage introduction, the leaching behavior of waste glasses was investigated, often using autoclaves and elevated temperatures and pressures. Rutherford back-scattering, RBS, was employed to determine thickness and composition of the corrosion layers. Electron microprobe analysis yielded results for thicker layers. Important enrichment factors were observed for actinides and for other elements of low solubility (fission products, but also glass components such as Ca and Ti) in the corrosion layers that formed after contact with water. The mechanical property of relevance to the storage problem, e.g. the fracture toughness, K Ic, was measured with the aid of the Hertzian identation technique. In this method, spherical indentors are used to produce cone-shaped ‘ring crack’. The method was developed to an extent that quantitative results were obtained without any empirical fitting parameters as are often used for similar determinations with Vickers indentations. Also, the compatibility of waste glass and repository salt under the expected storage conditions was studied. No measurable interaction was observed. Leaching layers and fracture toughness were also measured following Cm-doping or ion bombardment of waste glasses. No important effect of damage on layer thickness and composition was found for the leaching conditions used. Radiation damage usually increased the fracture toughness, a very beneficial effect. Fracturing of waste glass cylinders should be minimized since fractured cylinders, because of the larger available surfaces for leaching, could potentially deliver more radioactivity into the surroundings than unfractured cylinders. Causes for fracture could be mechanical or thermal stresses. Bombardment and damage production with Pb ions caused a decrease in fracture toughness. This decrease, however, was smaller than that caused by adding chemically the same amount of Pb to the glass. PbO bonds are known to be weaker than SiO bonds. For such experiments, ion bombardment with PB is thus not a very good damage simulation procedure. Many of the crystalline phases studied became amorphous during extended α-decay damage (metamictization). The amorphization is often connected to an essential expansion (swelling) of the product due to the lower density of the amorphous phases, and also due to accumulation of defects. Data on thermal recovery of damage and the recrystallization of the amorphous phases have also be obtained. For the materials and condition studied so far, the above radiation damage (accumulated at ambient temperature) did not cause any dramatic deleterious effects. The observed increased fracture toughness is a positive effect of radiation damage.