Table of Contents

New Silicotitanate Waste Forms: Development and Characterization M. L. Balmer, Y. Su, E. Bitten,(a) A. Navrotsky,(b,c) H. Xu,(b,c) T. Nenoff,(b,d) M. Nyman,(d) and R. Roth(e) Supported by the Environmental Management Science Program. (a) Student. (b) EMSL User. (c) University of California, Davis. (d) Sandia National Laboratories. (e) National Institute of Standards and Technology. This program has identified new waste forms and disposal strategies specific to crystalline silicotitanate (CST) secondary waste that is generated from cesium and strontium ion exchange processes. In particular, in-situ heat treatment of CSTs to produce an alternative waste form is being examined. Waste forms that are developed in this work will offer an alternative to current disposal plans, which call for recombining the separated cesium, strontium-loaded CST into the high-activity waste (HAW) streams, then dissolving it in borosilicate glass. The goals of the program are to reduce the costs associated with CST waste disposal, minimize the risk of contamination to the environment during CST processing, and provide DOE with technical alternatives for CST disposal. Because there is uncertainty in repository availability and in waste acceptance criteria, it is likely that cesium and strontium-loaded ion exchangers will require short-term storage at Hanford or that new scenarios for long-term storage or disposal of nuclides with relatively short half-lives (such as 137Cs and 90Sr) will arise. Research activities in this program have generated information on the durability and stability of thermally consolidated CSTs so that the potential of these options as viable storage or disposal scenarios can be evaluated. The technical objectives of the proposed work are to fully characterize the phase relationships, structures, and thermodynamic and kinetic stability of crystalline silicotitanate waste forms and to establish a sound technical basis for understanding key waste form properties, such as melting temperatures and aqueous durability, based on an in-depth understanding of waste form structures and thermochemistry. Plans are in place to retrieve, separate, and immobilize radioactive waste contained in 177 underground storage tanks at the Hanford Site. Likewise, nuclear wastes at other DOE sites across the country must be immobilized in a stable waste form for storage in underground repositories. A viable waste form must be chemically durable under environmental storage conditions (aqueous environments are of primary concern) and thermally stable under repository conditions over a geologic time scale. In addition, an adequate waste form should be capable of incorporating specific waste feeds to form a stable glass or ceramic material with a minimum of waste dilution (to minimize waste volumes) and be easy to process under remote handling conditions. Relatively low processing temperatures are desirable, as are simple heat-treatment cycles. Borosilicate glass has been chosen as the baseline host for immobilization of HAW present at the Hanford Site. However, CST, the most promising candidate for removal of cesium and strontium from tank wastes, has been identified as a risk to the borosilicate vitrification process. TiO2 in the CST promotes crystallization and immiscible phase separation, and affects the redox state and solubility of uranium in glass (Ewest et al. 1987; Galakhov et al. 1988; Bickford et al. 1990; Plodinec 1980). Because of this, a TiO2 limit of 1 wt% is set for borosilicate waste glass at the Savannah River Defense Waste Processing Facility (DWPF) (Plodinec 1980). If these high levels of waste dilution are required to stabilize CST waste, the volume of expensive high-activity borosilicate waste glass produced for subsequent storage will be substantially increased. Dissolution of the CST in borosilicate glass (as opposed to direct thermal conversion) necessitates removal and transfer of the CST from the column, mixing with glass frit, and melting. Each of these steps significantly increases the risk of contamination to workers and/or the environment. The volume of HAW can be minimized and the process can be simplified by converting the separated, compositionally homogenous loaded exchanger into an alternate waste form rather than recombining it with the HAW for dissolution in glass. Cesium-loaded silicotitanate ion exchangers contain the basic ingredients that can form a ceramic or glass at high temperature. The premise of this work is that for CST ion exchange waste, waste forms can be tailored to specific waste feeds rather than attempting to tailor waste feeds for accommodation by a single waste form. Direct in situ, thermal conversion of the CST will consolidate and immobilize the loaded ion exchange particulate, minimize handling risks, and remove water and hydroxyl groups, thus eliminating radiolytic hydrogen generation during storage. The research strategy for developing an alternate waste form for CSTs is based on an understanding of ceramic and glass structures and phase stability. The key components of the research include determining stable and metastable phases that form from heat treatment of the CST and development of phase diagrams for the component systems characterizing the structures of ceramic and glass waste form candidates developing structure-property relationships that will provide predictive capabilities regarding key performance parameters, including aqueous leach resistance, melt temperatures, melt viscosities, and ease of crystallization for select waste forms determining the enthalpies and free energies of formation of key compounds by high-temperature oxide melt solution calorimetry. Work on this program has shown that thermally converted CSTs have very high aqueous durability (Su et al. 1998). These thermally converted waste forms are several orders of magnitude more durable than borosilicate glass. For example, a standard engineering assessment (EA) of borosilicate glass shows a seven-day leached concentration of alkali of 13.3 g/L (there is no cesium in this standard, only sodium) (Ferrara et al. 1997). In comparison, IE-911 heat-treated to 900°C has a seven-day cesium concentration of 0.008 g/L, or in the worst case for the 700°C heat-treatment, the seven-day cesium concentration is 0.175 g/L. Heat treatments of 900° and 500°C yielded the most durable ceramics; however, all residual water and hydroxy groups are not removed until 800° C. Therefore, 900°C is the optimum heat treatment because the possibility of hydrogen formation from radiolytic decay of physisorbed or chemisorbed water is eliminated. Phase stability and crystal chemistry studies for compositions related to the exchanger are vital to predicting long- and short-term performance of waste forms. Transmission electron microscopy (TEM), x-ray diffraction (XRD), and nuclear magnetic resonance studies have been performed to elucidate the crystal structure of each phase in the heat-treated IE-911. Figure 9.1 shows a micrograph of a thin section of 12 wt% cesium-exchanged IE-911 heat-treated to 900°C. Energy dispersive spectroscopy on individual grains revealed that several oxide phases with unique compositions exist in the sample. Figure 9.1. TEM micrograph of a thin section of IE-911 that was treated in NaOH, exchanged with 12 wt% cesium, then heat-treated to 900°C. Energy dispersive spectroscopy of individual grains revealed the approximate compositions of the phases in this proposed waste form.. Among these were Cs, X1, Si oxide, Na, Ti oxide, Na, Ti, X2 oxide, and very minor Na, Cs, Ti, Si, X1 oxide, where X1 and X2 are proprietary components of the ion exchanger. Of these four phases, positive identification by XRD had already been achieved for the sodium titanate phase, Na2Ti6O13. This structure is a layered structure. While there are known isomorphs for potassium and rubidium substitutions on this structure, cesium substitution had not been documented. The structure of the pure cesium analog, Cs2 Ti6O13, has a different structure than the sodium, potassium, or rubidium forms. To determine whether small amounts of cesium could reside in the Na2 Ti6O13 structure, a solid substitution series of (Na, Cs)2 Ti6O13 was synthesized and the lattice parameters were measured by XRD. This study surprisingly revealed that there is no measurable substitution of cesium for sodium on the Na2 Ti6O13 lattice, eliminating the possibility that small amounts of cesium could reside in this phase in heat-treated IE-911. TEM revealed that the majority of the cesium is contained in a Cs, X1, Si oxide. This phase could not be synthesized by solid-state reaction; however, it was successfully synthesized using sol-gel and hydrothermal reactions. The stoichiometry of the phase is Cs2X1Si3O9. While a similar phase is reported in the literature, the crystal structure of this compound is unknown and the powder XRD data are incomplete. A search of the powder diffraction database revealed a compound that is a structural isomorph to Cs2X1Si3O9. From this it was determined that the new cesium compound is hexagonal with a lattice parameter of a = 7.23 Å and c = 10.271 Å (space group P63/m). Efforts are under way to refine the structure to obtain the exact atomic positions. The simulated crystal structure based on the atomic positions of the isomorph and the lattice parameters of Cs2X1Si3O9 is shown in Figure 9.2. It consists of silica tetrahedra and X1 octahedra that form three-membered and six-membered rings. The largest free aperture of the rings is approximately 2.2 x 2.6 Å, which is smaller than a cesium atom (~3.5 Å). Therefore, the cesium in this phase will be immobile. Removal of cesium from the structure will require the cleavage of the strong, covalent Si-O and X1-O bonds. This structural feature in part explains the high resistance to leaching of cesium in thermally converted IE-911 that is exposed to aqueous solutions. Figure 9.2. View of Cs2X1Si3O9 structure down the [001] directions. This is the major cesium-containing phase in thermally converted (900°C) IE-911. A second new phase discovered in TEM (Na, Ti, X2, oxide) has a structure similar to NaX2O3. To determine the extent of titanium substitution in the mixed phase, a series of compounds with up to 20% titanium substituted for X2 were synthesized using a sol-gel technique. A systematic shift of the lattice parameter as a function of titanium substitution could clearly be observed. Comparison of the heat-treated IE-911 with the synthesized compounds revealed that the phase in IE-911 has 15% titanium substitution on the lattice. The structure of the new compound is related to a perovskite; however, the distribution of the cations in the structure is unknown. With the discovery of the two new oxide phases in heat-treated CST, the phase identification is nearly complete. Figure 9.3 shows the XRD pattern of cesium-exchanged heat-treated IE-911 compared with a simulated x-ray pattern, which is a combination of the patterns of the three phases identified in this program, Na2Ti6O13, Cs2X1Si3O9, and Na(Ti, X2)O3. It can be seen that all of the major peaks have been identified. Several very small peaks that appear in the thermally converted IE-911 pattern are not present in the simulated pattern. The intensity of these peaks decreases significantly with longer heat treatment at 900°C or with heat treatment to 1000° C, indicating that this phase is metastable. Some of these minor peaks may also correspond to the minor Cs, Na, Si, Ti, X1 oxide identified with TEM. Work to determine the structure of this minor phase is continuing. Figure 9.3. Comparison of the experimental XRD pattern of mixed-phase cesium-exchanged IE-911 heat-treated to 900°C (upper pattern) with a simulated pattern that is a compilation of the x-ray patterns from synthesized compounds Na2Ti6O13, Cs2X1Si3O9, and Na(Ti, X2)O3 (lower pattern). The excellent match between the two patterns suggests that these compounds represent the major phases in the waste form. In summary, efforts on this program have focused on determining the stable and metastable phase development in a heat-treated, cesium-exchanged crystalline silicotitanate ion exchanger (IE-911). Transmission electron microscopy, XRD, and synthesis studies have revealed that the major phases in the ion exchanger are Na2Ti6O13, Cs2X1Si3O9, and Na(Ti, X2)O3. The network structure of the cesium-containing phases precludes facile migration of the cesium ion, resulting in extremely high aqueous durability. Information on phase selection as a function of composition, chemical durability, and thermodynamic stability can be used to determine processing windows and to predict long- and short-term stability of thermally-converted CST ion exchangers. References Bickford, D. F., A. Applewhite-Ramsey, C. M. Jantzen, and K. G. Brown, "Control of Radioactive Waste Glass Melters: I. Preliminary General Limits at Savannah River." J. Am. Cer. Soc. 73 [10], 2896-2902 (1990). Ewest, E., and H. Wiese, "High Level Liquid Waste Vitrification with the Pamela Plant in Belgium." IAEA-CN-48/177, pp. 269-280, Vienna, Austria (1987). Ferrara, D. M., M. K. Andrews, J. R. Harbour, T. L. Fellinger, D. T. Herman, K. M. Marshall, and P. J. Workman, Vitrification of Ion Exchange Material. WSRC-TR-97-00320, Westinghouse Savannah River Corporation, Aikin, South Carolina (September 30, 1997). Galakhov, F. Y., V. T. Vavilonova, V. I. Aver’yanov, and T. V. Slyshkina, "An Experimental Determination of the Region of Liquid-Liquid Phase Separation in the Li2O-Al2O3-TiO2-SiO2 System." Translated from Fiz. khim. Stekla 14(1), 38-46 (1988). Plodinec, M. J., "Development of Glass Compositions for Immobilization of Savannah River Plant Waste." In Scientific Bases for Nuclear Waste Management, CJM Northrup Jr., ed. Vol. 2, pp. 223-229 (1980). Su, Y., M. L. Balmer, L. Wang, B. Bunker, M. Nyman, T. M. Nenoff, and A. Navrotsky, "Evaluation of Thermally Converted Silicotitanate Waste Forms II." Proc. Mat. Res. Soc. Fall Meeting (1998). William R. Wiley Environmental Molecular Sciences Laboratory Feedback: webmaster@emsl.pnl.gov Revised: April 17, 2000 Security & Privacy PNNL-13147