INTRODUCTIONThe disposal of radioactive "waste" generated by the nuclear fuel cycle is among the most pressing and potentially costly environmental problems of the 21st century, a heritage from the Atomic Age of the 20th century. Proposed disposal strategies are complicated, not only because of the large volumes and activities of waste, but also because of the political and public-policy issues associated with the long times required for containment and disposal (10 4 to 10 6 years). The development and use of highly durable waste forms, materials that have a high chemical durability and resistance to radiation damage effects, can simplify the disposal strategy (Ewing 2001).An interest in phosphate-based waste forms has developed because the high-level waste generated by reprocessing of spent nuclear fuel can contain substantial amounts of phosphates (up to 15 wt % P 2 O 5 ) that result from processing technologies that utilized either a bismuth phosphate or tributylphosphate process (Bunker et al. 1995). In addition to the high phosphate content, other metal oxides may achieve significant proportions (up to 15 wt % Fe 2 O 3 ; up to 30 wt % Bi 2 O 3 ; up to 30 wt % UO 2 ) (Lambert and Kim 1994). These complex compositions have presented special challenges in developing crystalline ceramics that can accommodate the full compositional range of the waste streams.The early work on phosphate glasses led to the idea that crystalline phosphates might make extremely durable waste forms, particularly for actinides. The earliest suggestion was for the use of monazite (Boatner 1978, Boatner et al. 1980, McCarthy et al. 1978, 1980. The attractive qualities of monazite as a nuclear waste form are: (1) a high solubility for actinides and rare earths (10 to 20 wt %); (2) evidence from natural occurrences of good chemical durability; (3) an apparent resistance to radiation damage, as natural monazites are seldom found in the metamict state, despite very high alphadecay event doses (Boatner and Sales 1988). There have been extensive studies of monazite and apatite as potential waste form phases, and a considerable amount of work on a number of synthetic phosphate phases has been completed.The crystal chemistry of monazite, apatite, and related phosphate minerals, has been discussed in detail (see this volume, Chapters 1, 2, and 4 by Hughes and Rakovan, Pan and Fleet, and Boatner, respectively) and will not be repeated here. Rather, we will summarize the work relevant to the consideration of these phosphate phases as nuclear waste forms. TYPES OF NUCLEAR WASTEThe design and evaluation of nuclear waste forms requires some understanding of the sources, volumes, compositions and activities of the various waste streams generated by the nuclear fuel cycle. There are three primary sources of radioactive waste in the United States (DOE 1997): the high-level waste (HLW) from the reprocessing of spent nuclear fuel, the spent nuclear fuel itself, and plutonium reclaimed by reprocessing or obtained by the dismantlement of nuclear weapons. 67...

O 4 ), and phenakite structures have been investigated using in situ and high-resolution transmission electron microscopy (HRTEM) over a wide temperature range (20-873 K). At low temperatures (<200 K), the olivine and phenakite are susceptible to radiation-induced amorphization with a critical amorphization dose of 0.2-0.5 displacement per atom (dpa). The critical amorphization dose increases with increasing irradiation temperature at varying rates for the various phases, resulting in a distinct critical amorphization temperature for each phase. For the Mg-Fe series of olivine, the susceptibility to amorphization at higher temperatures (room temperature or above) increases with increasing Fe content. Although the spinel phases are, in general, much more resistant to amorphization, a highpressure metastable spinel phase, ␥-SiFe 2 O 4 , is easily amorphized at doses <0.2 dpa at temperatures below 723 K. This phase decomposes after irradiation at 873 K. At 20 K, complete amorphization of the FeCr 2 O 4 spinel (chromite) is achieved at ∼4 dpa, but no evidence of amorphization is observed in MgAl 2 O 4 spinel after 5.4 dpa. The distinct differences in the relative susceptibility of these phases to amorphization are discussed in terms of the structural and chemical controls on the amorphization process.K. E. Sickafus-contributing editor Manuscript No. 189659.