Abstract This review describes nuclear waste forms for high-level waste (HLW), that is, glasses, ceramics, and glass-ceramics, as well as for low- and intermediate-level waste (LILW), that is, cement, bitumen, glass, glassy slags, and ceramics. Ceramic waste forms have the highest chemical durability and radiation resistance, and are recommended for HLW and actinide (ACT) immobilization. Most radiation-resistant materials are based on phases with a fluorite-related structure (cubic zirconia-based solid solutions, pyrochlore, zirconolite, murataite). Glass is also a suitable matrix for HLW containing fission and corrosion products, and process contaminants such as Na salts. Within the framework of the HLW partitioning concept providing separation of short-lived (Cs, Sr) and long-lived (rare earth element-ACT) fractions, glass may be used for immobilization of the Cs-Sr-bearing fraction, whereas the rare earth-ACT fraction may be incorporated in ceramics. Glass-based materials or clay-based ceramics are the most promising LILW forms, but cement and bitumen may also be applied as matrices for low-level wastes (LLW).

Abstract A number of polyphase or single-phase ceramic waste forms have been considered as options for the disposal of nuclear waste in geological repositories. Of critical concern in the scientific evaluation of these materials is their performance in natural systems over long periods of time (e.g., 10 3 to 10 6 years). This paper gives an overview of the aqueous durability of the major titanate host phases for actinides (e.g., Th, U, Np, Pu, Cm) and important fission products (e.g., Sr and Cs) in alternative crystalline ceramic waste forms. These host phases are compared with reference to some basic acceptance criteria, including the long-term behaviour determined from studies of natural samples. The available data indicate that zirconolite and pyrochlore are excellent candidate host phases for actinides. These structures exhibit excellent aqueous durability, crystal chemical flexibility, high waste loadings, and well-known processing conditions. Although both pyrochlore and zirconolite become amorphous due to alpha-decay processes, the total volume swelling is only 5–6% and there is no significant effect of radiation damage on aqueous durability. Hollandite also appears to be an excellent candidate host phase for radioactive Cs isotopes. Brannerite and perovskite, on the other hand, are more prone to alteration in aqueous fluids and have a lower degree of chemical flexibility. With the exception of hollandite, many of the properties of these potential host phases have been confirmed through studies of natural samples.