Thoria-urania-zirconia ceramics were studied in order to investigate the long-term behavior of potential thorium fuels in a repository environment. The ceramics were prepared by coprecipitation of the metal salts. Zirconia was added to determine if further stabilization against dissolution of the thoria-urania system could be achieved. In addition, 0.5 wt% MgO was added to some samples to increase stability and density. The inclusion of Zr in the ceramics did not dramatically decrease the leaching of thorium from the matrix.

Material properties of the ceramic were analyzed using electron microscopy techniques such as Energy Dispersive X-ray (EDX) analysis and Electron Energy Loss Spectroscopy (EELS). X-ray diffraction and synchrotron-based x-ray absorption studies including extended xray fine structure (EXAFS) and x-ray absorption near edge spectroscopy (XANES) were also used to reveal elements of the phase structure and chemistry of the ceramics. X-ray diffraction (XRD) and EDX show that these ceramics separate into a zirconium-based phase and an actinide-based phase with low mutual affinity of thorium and zirconium, as well as partial solubilization of uranium in zirconium. The comparison of EELS spectra collected for the ceramics with spectra collected for UO2 and U3O8 reference materials also allow the assessment of uranium oxidation state independently in the two separate phases. Assessment of the bulk oxidation state using XANES correlated well with the EELS analysis. Interatomic distances and the bulk crystal structure were determined using EXAFS.

Diffusion and sorption are potentially the most important factors governing the transport of radionuclides in clays and under stationary subsurface environmental conditions. Preliminary studies have been performed to measure the transport of 137Cs and 60Co with no advection, using reconstituted samples of three regolith materials collected from a region in South Australia. The samples were saturated with CaSO4 solution to imitate the pore water chemistry of the in situ environment. A double diffusion cell testing apparatus made of polycarbonate resin was used to measure the transport of the selected radionuclides through the samples. A curve fitting procedure employing one-dimensional contaminant equations with a “stop-start” technique was used to estimate the diffusion (D) and sorption coefficients (Kd) from the measured concentration-time profiles. Results from these experiments are compared with those obtained from batch sorption tests.

Titanium Grade 7 (Ti-7) was tested for both general and local corrosion in environments consisting of fluoride and chloride salts added to simulated groundwaters. Tests were conducted on both U-bends and disks in order to determine the extent of corrosion under stressed and unstressed conditions. The experiments were run over a broad range of temperatures and pH values for periods of up to 164 days. It was found that Ti-7 is susceptible to intergranular attack in some of the test solutions at moderately elevated temperatures, leading to conspicuous localized corrosion. Pitting was detected within 14 days, with pit depths on the order of 0.1 mm. Surface defects appeared to promote the initiation of pitting. The same environment caused stress corrosion cracking failure in a statically stressed U-bend.

Murataite A4B2C7O22-x, where A = Na+, Ca2+, REE3+, An3+/4+; B = Mn2+/3+, Zn2+; C = Ti4+, Fe3+, Al3+; 0≤x≤1, is an isometric, derivative of the fluorite-structure. Murataite is potentially suitable as a phase for the immobilization of rare earth (REE) and actinide elements (An). Murataite structures with three-(3C), five-(5C), and eight-fold (8C) multiples of the fluorite unit cell parameters have been identified. Radiation-induced amorphization of murataite has been investigated by 1 MeV Kr+ ion irradiation of three ceramic samples produced by melting in a resistive furnace and a cold crucible at 1400-1600 °C. The 1 MeV Kr+ ion irradiations were performed at room temperature using IVEM-Tandem Facility at Argonne National Laboratory. Radiation damage was observed by in-situ TEM. Initially, the irradiation caused disordering of the murataite structure. Murataite was rendered fully amorphous at a dose of (1.7∼1.9)Á1018 ion/m2. The pyrochlore structure phase (2C) is more radiation resistant to ion irradiation-induced amorphization than the murataite structure. Combining results on murataite with those pyrochlore and fluorite, a generally increasing trend in the susceptibility to ion beam damage is found in the fluorite-related structures as a function of the increasing multiples of the fluorite unit cells.

Previous studies show that some radionuclides present in reactor decontamination wastes form strong complexes with the organic complexing agents used to decontaminate reactor cores and piping. Further, the metal-ligand complexes exhibit reduced adsorption to soils. Flow through column tests were used to study the adsorption of metal-organic ligand complexes to two soils. The breakthrough curves for metals and organic ligands for four tests are shown. In all the column tests, the adsorption of the organic ligand-metal complex, or the free organic ligand and free metal (disassociated species) was reversible. That is, close to the total mass injected was recovered in the flushing stage with untraced background solution.

Two tests that used the iron oxide coated sand show complicated behavior that was interpreted as being caused by ligand (both EDTA and picolinate) interacting with the ferric oxides. The Ni-organic ligand complex in both cases appears to exchange Ni for Fe to some extent such that free Ni+2 is produced and Fe (III)-organic complexes are formed that adsorb with different strengths. Further, we suspect that the organic ligand is dissolving some of the ferric oxide coatings and destroying sorption sites. The combination of all these reactions leads to rather complicated breakthrough curves. In tests with picolinate there is partial disassociation of the metal and ligand such that the breakthrough curves for the metal and ligand are different

A recommendation is made to not bury EDTA-laden decontamination wastes with cement. Another observation is that predictions that use simple constant Kd and constant source release constructs may not be exclusively conservative in predicting concentrations of contaminants in water down-gradient from disposal sites.

The hydrogen redistribution induced by the thermotransport in the modified Zircaloy-4 at temperatures likely to be encountered in nuclear power reactors (300-340°C) was investigated by means of steady state techniques. The modified Zircaloy-4 was prepared by changing the chemical compositions of Zircaloy-4, which is used widely as a nuclear fuel cladding material in pressurized water reactors. The change of Q for hydrogen, which describes the direction and magnitude of the thermotransport, with increasing hydrogen and oxygen concentrations was investigated in the modified Zircaloy-4. The value of Q for hydrogen in the modified Zircloy-4 alloys was found to be about 7 kcal/mol and it was not affected by hydrogen concentration in the hydrogen concentration range from 63.3 ppm to 91.7 ppm. While the value of Q for hydrogen decreased from 6.8 kcal/mol to 4.5 kcal/mol with increasing oxygen concentration from 0.2 wt% to 1.0 wt% and it was considered to be due to the trapping of hydrogen by oxygen. In addition, the hydrogen redistribution and Q in Zircaloy-4 was also investigated in order to compare the characteristics of thermotransport of hydrogen between Zircaloy-4 and modified Zircaloy-4. The hydrogen redistribution and Q in Zircaloy-4 showed the same results to those of the modified Zircaloy-4.

Baseline formulation titanate ceramics for surplus Pu disposition are based upon a target mineralogy of 95 wt.% pyrochlore (Ca0.89Gd0.22Hf0.23Pu0.22U0.44Ti2O7) plus 5 wt.% Hf-doped rutile (∼ Ti0.9Hf0.1O2), where Ce is used as an analogue for Pu and U, and Th for Pu. Typically, Pu/U, Th/U and Ce/U-baseline samples form major pyrochlore, plus minor brannerite (AnTi2O6) and rutile. Ce/Ce-baseline ceramics were similar but did not form brannerite. Sintering in air produced more U5+ in the ceramics than sintering in Ar. In the pyrochlore the charge compensation for U5+ principally occurred by an increase in Ca2+ and decrease in Ti4+ and Th4+. In the matrix these phase compositional changes result in an increase in brannerite content of the ceramic at the expense of pyrochlore in the air sintered ceramic relative to the Ar sintered sample. Sintering in reducing 3.7% H2 in Ar atmospheres eliminates the brannerite and rutile and results in 2M-zirconolite and perovskite in addition to the major pyrochlore phase. Varying the sintering temperature between 1250°C and 1400°C had little effect on the phase chemistry other than slightly incomplete reactivity of coarse actinide oxide at the lower sintering temperatures and a general increase in density (to ∼ 1350°C) and grain size with increased sintering temperatures. Th4+ appears to be a good analogue for Pu4+ from a crystal chemistry aspect.

This waste form is an alloy consisting of stainless steel with 15 wt% zirconium and acts as a host for the immobilization of radioelements that remain with the spent fuel cladding hulls following their treatment using an electrometallurgical treatment process. The results presented here are from 14, 34 and 90-day immersion tests conducted at 90 °C. These tests show that the release of uranium is considerably higher than that of all other major elements present (Fe, Cr, Ni, Zr), but that release of all constituents is comparable to or lower than that for borosilicate glass.

Thermodynamic calculations were conducted to determine the deliquescence behavior of salt mixtures and to simulate the evaporation of Yucca Mountain groundwaters. The results are consistent with published experimental data that show the deliquescence points of salt mixtures are lower than that of individual salts. For mixtures of NaCl and KCl salts, the deliquescence point of pure NaNO3 salt is an appropriate lower bound. However, mixtures containing magnesium and calcium salts have much lower deliquescence points than pure NaNO3. If magnesium and calcium salts are deposited on waste package and drip shield surfaces, it could lead to earlier initiation of aqueous corrosion than assumed by the DOE in its performance assessment abstractions. Such salt mixtures can be formed by evaporation of waters with compositions similar to some Yucca Mountain porewaters and would be characterized by low deliquescence relative humidity, high chloride concentration, and low concentrations of anions such as free (uncomplexed) nitrate and sulfate that could mitigate against the chloride-enhanced corrosion of the waste package. Evaporation of Yucca Mountain groundwaters also could lead to fluoride concentrations that are above the threshold for accelerated corrosion of the titanium drip shield.

Iodine release from silver iodide (AgI) to water was evaluated under reducing conditions in the presence of iron-bearing minerals, goethite (FeOOH), magnetite (Fe3O4) and Wüstite (FeO). The release tests were performed in a glove box purged with gas mixture (Ar+5%H2), and the concentrations of dissolved iodine, silver and iron were measured. The apparent equilibrium concentration of dissolved iodine was 4.0 ×10−8, 4.3×10−8 mol/l for the tests with FeOOH and Fe3O4, respectively. The values were almost the same as in the test without minerals. For the test with FeO, the concentration of dissolved iodine was 5.4×10−3 mol/l, which is five orders of magnitude higher than for the other tests. Solid phase analyses by using XRD and SEM/EDS indicated that metallic silver precipitated on the surface of the original AgI for the test with FeO and on the surface of Fe3O4 for the test with Fe3O4, but not for that with FeOOH. These results showed that only FeO could reduce AgI effectively to increase iodine release. The amount of dissolved Fe(II) was a key factor affecting AgI reduction instead of redox potential of solution. Kinetic processes may also play an important role in AgI reduction and associated iodine release under reducing condition.

An unconformity underlying the Tono uranium deposit in central Japan represents the approximate location of a redox front separating relatively oxidizing groundwaters (Eh ≈0 mV) in the weathered, fractured Toki granite (TG) from strongly reducing pore fluids (Eh ≈-360 mV) in sedimentary rocks of the overlying Lower Toki Lignite-bearing Formation (TL). Uranium has been effectively immobilized in the TL during the past 10 million years. Stable and reversible redox potentials measured in-situ in boreholes penetrating the sedimentary rocks and granite appear to be controlled by the Fe(III)-oxyhydroxide – Fe2+ redox couple. A simplified analytical model of front migration suggests that chemical buffering by pyrite alone would limit the propagation velocity of the front into the TL to less than 8x10−6 m yr−1. The model is constrained by Darcy fluxes derived from groundwater flow models and relative 14C groundwater ages, average modal abundances of pyrite in the TL, and the analytical detection limit for dissolved oxygen in TG groundwaters (2 ppm). Model results also suggest that the redox buffering capacity of the TL would be exhausted within 10 million years if an upper bound O2(aq) concentration in TG groundwaters fixed by equilibrium with atmospheric O2(g) (8.5 ppm) is assumed. Immobilization of uranium in the TL is thus attributable to oxidation-reduction reactions that minimize O2(aq) concentrations primarily in the TG, and secondarily in the TL.

Waste management and disposal decisions at the Hanford Site, Washington, depend in part on an understanding of the risks and impacts associated with alternate disposal and remedial actions. A proof-of-principle site-wide assessment of the risks and impacts associated with all wastes that will remain at the Hanford Site following cleanup has been performed for the first time. It simulates contaminant release, migration, and fate from the initiation of site operations in 1944 forward, and, thus, illustrates historical and near-term influences on long-term risk and impact.

A stochastic simulation tool capable of addressing 1000 waste discharge and disposal sites and 10 contaminants for a period of 1000 years has been created and applied. Human health and ecological risks as well as impacts to the regional economy and local cultures are estimated. The methodology developed is known as the System Assessment Capability (SAC). It is currently in a Revision 0 state corresponding to a proof-of-principle demonstration.

An initial assessment based on the planning baseline of the U.S. Department of Energy (Richland Operations office and office of River Protection) has been undertaken. Preliminary results of the assessment indicate variability in predictions is most influenced by uncertainty in:

- geochemical adsorption (i.e., distribution coefficients, Kd) for contaminant release, vadose zone, and groundwater models; especially for uranium and iodine which are somewhat but not greatly adsorbed,

- solid waste burial ground inventories of iodine-129, and

- liquid discharge site inventories of technetium-99.

It is apparent that variables governing change in performance are a function of space and time. Initially, these results point to the need for related models and data to be examined, and, if necessary, augmented through future laboratory and field studies to better quantify or reduce uncertainty.

The five-metal epsilon particle represents an important component with respect to the corrosion of spent nuclear fuel as it is the principal host for 99Tc. This radionuclide has a high solubility in oxidizing environments (as TcO4 -), a half-life of 213,000 years, and has been proposed as a monitor for the corrosion rate of spent fuel. As such, an understanding of the corrosion processes affecting epsilon particles may have important implications on our ability to accurately assess radionuclide release rates from spent nuclear fuel.

Non-radioactive metal powders were mixed in an atomic percent ratio of 40% Mo, 30% Ru, 15% Pd, 10% Re, and 5% Rh (Re used as a surrogate for Tc), a composition that simulates epsilon particles in spent reactor fuel [1,2]. The powders were electric arc-melted in the presence of an Ar-purged atmosphere. Scanning electron microscopy images of the as-cast samples denote a primary sample heterogeneity, with two distinct zones, each composed of a mixture of very fine-grained Mo-, Mo-Ru-Re, and Mo-Ru-Pd rich crystals. Results from vapor hydration tests (200°C) for time periods up to 35 days indicate the formation of Mo- and Reenriched alteration phases. Phase growth was enhanced when air was periodically replenished in the test vessels suggesting that oxide and/or hydroxide alteration phases may play a role in the corrosion process. MCC-1 tests (90°C) were conducted for time periods up to 182 days in leachant solutions that were prepared with either a) deionized water or b) deionized water acidified with nitric acid to a pH of 3.0. Solutions from the deionized water tests progressively decreased in pH with increasing reaction time (5.5 to 4.0), while the pH values remained at a near-constant value of 3.0 in the acidified solutions. Solution results indicate a preferential release of Mo followed by Re. Normalized Re/Mo release ratios varied from 0.20 to 0.86, and increased as a function of reaction time and acidity. Release of the platinum group metals (Pd, Rh, and Ru) was relatively insignificant in comparison to Mo and Re.

Two potential applications of radioisotope fractionation and decay-series secular disequilibrium in performance assessment for geologic repositories for nuclear waste are preferential radionuclide release in source term analysis and characterization of system closure as a measure of the capacity of the geologic system to isolate waste. A primary mechanism of radioisotope fractionation is selective release and mobility of alpha decay products because of nuclear recoil effects, which is evident in natural system data. Preferential release of radioisotopes from nuclear waste forms or solubility limiting solid phases could affect repository performance; however, consequences of differential radioisotope releases are neglected in performance assessments. Decay-series disequilibria are useful also to characterize open/closedsystem behavior in natural systems. Systems that are closed on time scales that are long relative to the half-lives of decay chain nuclides achieve secular equilibrium characterized by unit activity ratios among series nuclides. For geologic disposal of nuclear waste, measures focused on chemical system closure could capture the essential characteristics of the natural system with respect to radionuclide isolation and could be based quantitatively on uranium and thorium decay series secular equilibria/disequilibria.

Beginning in 1952, waste materials, including volatile organic compounds (VOCs)contaminated with transuranic radionuclides, were generated during the fabrication, assembly, and processing of nuclear weapons components in the US Department of Energy (DOE) weapons productions complex at the Rocky Flats Plant (RFP). Following processing and containerization, drums were shipped to the Subsurface Disposal Area (SDA) at the Idaho National Engineering and Environmental Laboratory (INEEL). During 1968 approximately 9,691 drums were buried there.

In subsequent years, observations made during drum retrieval studies indicated that many of the drums were compromised on impact or suffered physical damage by compaction equipment shortly after burial. Corrosion also appears significant on drums buried for a few years. A large vadose-zone contaminant plume composed of solvents buried in the drums has been found beneath the burial area.

Phase partitioning calculations show it unlikely that separate-phase solvent has leaked from the compromised drums deep into the soil profile or that solvents have dissolved into infiltrating water. Rather, it appears that the solvents are evaporating out of the barrels into the air phase and further partitioning from there throughout the subsurface.

This paper describes the history of mixed wastes buried at the SDA, phase partitioning, and preliminary computer simulation results on gas contaminant mobility in the vadose zone.

The phase stability of C-22 alloy (UNS N06022) gas tungsten arc welds was studied by aging samples at 593, 649, 704, and 760°C for times up to 6,000 hours. The tensile properties and the Charpy impact toughness of these samples were measured in the as-welded condition as well as after aging. The corrosion resistance was measured using standard immersion tests in acidic ferric sulfate (ASTM G 28 A) and 2.5% hydrochloric acid solutions at the boiling point. The microstructures of weld samples were examined using scanning electron microscopy (SEM). Precipitate volume fraction measurements were made using optical microscopy.

Degradation of the mechanical and corrosion properties of C-22 welds due to aging at all temperatures investigated was seen to occur sooner than was seen in C-22 base metal. An evaluation of the kinetics of nucleation and growth of the precipitates forming at these temperatures, however, indicated that no significant changes in TCP phase morphology would occur at temperatures below approximately 300°C.a

The effect of Pb, Hg, thiosulfate and sulfide on the corrosion of Alloy 22 was studied. In solutions of 1000xJ13 groundwater doped with these contaminants, disks and U-bend specimens exhibited general corrosion and pitting at pH 1, and some dissolution at pH 13. One specimen exposed to the presence of Pb cracked catastrophically. In basic media thiosulfate and sulfide caused both oxide spalling and enhanced general corrosion.

Contaminated groundwater from a former liquid waste discharge area at the Chalk River Laboratories was treated with ion exchange resins to isolate aqueous extracts of selected radionuclides (60Co, 106Ru, 137Cs, 238Pu, 239/240Pu, 241Am and 244Cm). The extracts were then mixed with native uncontaminated soil material to determine their liquid-solid partitioning coefficients (Kd). Separate Kd values of the same elements were also obtained with the usual short-term “batch” technique for comparison, but using different radioisotopes as tracers (57Co and 134Cs), spiked to the extracts. The comparison included separate Kd determinations with 239/240Pu and 241Am as tracers, added to uncontaminated groundwater.

In all cases, the radionuclides originally present in the contaminated groundwater extracts exhibited low Kd values, compared to the values obtained in the “batch” method with tracers (57Co, 134Cs, 239/240Pu and 241Am). The difference was up to two orders of magnitude. This was attributed to differences in aqueous speciation of the nuclides in the contaminated groundwater, allowing limited interactions of complexed radionuclide species with soil material. Our results indicated that all the radionuclides were predominantly in anionic form in the groundwater, whereas in the tracer “batch” experiment, 57Co and 134Cs were predominantly cationic, Pu and Am were predominantly anionic. Hydrolysis partially explains the dominant anionic character of Am and Pu in the tracer experiments, however dissolved organics are suspected to dominate the speciation of the radionuclides in contaminated groundwater. Our experiment implies that in some cases, if Kd values obtained using the “batch” method with tracers are used in transport models, radionuclide transport could be underestimated.

A ten-year-long laboratory experiment was recently started in order to study interactions between the groundwater, the bentonite, and the canister in the near-field of the planned Finnish repository for spent nuclear waste. The experimental set-up consists of a number of samples containing sodium MX-80 bentonite and cast iron cylinders placed in copper vessels. The bentonite stays in contact with an external solution, either distilled water or a 0.5 M NaCl solution, via metal sinters. The experiments are performed under anoxic conditions (N2 atmosphere) at ambient room temperature (∼30 °C). The experimental measurements focus mainly on the development of (1) the water chemistry, (2) the cast iron corrosion, and (3) the diffusion of corrosion products inside the bentonite, since these processes are the most likely ones to produce detectable changes during the ten-year-long programme.

The first analyses, after 9 months, showed that the iron corrosion had started and that the corrosion products had migrated into the bentonite. The corrosion products could not be identified by XRD, which either meant that the material was amorphous or present in concentrations too low to be detectable. The analyses of the gas phase indicated the presence of hydrogen, which most probably was the result of the iron corrosion. The compositions of the external water and the porewater were determined with regard to the major species of interest. A closer evaluation of the results will not be given until more data are at hand.

Incorporation of paramagnetic species in NaZr2(PO4)3 (NZP) single crystal made possible to investigate local distortion of the NZP structure from EPR data. Cr3+ and Fe3+ ions were used as paramagnetic probes. We mapped invariant sum of cubic component of the fourth rank tensor of crystal field, the indicative surface of the B04 element of this tensor as well as 2nd, 3rd, and 4th orders axles of the cubic component of the same tensor. From these data the coordination polyhedron of the ion-probe has been modeled and direction to charge compensator has been determined. Based on the coordination polyhedron model obtained it has been shown how would the local symmetry be lowered from C3 (initial symmetry of the Zr-polyhedron) to C1 (Zr substitute ion-probe). A comparison of dynamics of the structure occurring at lowering the symmetry with allowed dynamics of the NZP structure (Roy -Alamo-Agrawal model) makes easily explainable why the dynamics observed results in an instability of the NZP structure and breaking the symmetry from R 3c to C2/c.