Diffusivities were measured for plutonium in brine-saturated compacted Wyoming bentonite. Complexities of the solution chemistry and retardation of transuranics necessitate diffusion studies under conditions that are specific for repository host rock types in this case salt. Diffusivity values in the range of 10−15 to 10−14 m2/s were obtained for bentonite at a packing density of 1800 kg/m3. That density was obtained by compaction at 15 i0Pa, a typical lithostatic pressure in a repository in salt at 650 m depth. Even a 0.05 m (2 inch) thick bentonite-containing engineered barrier could decrease radionuclide release rates by approximately 4 orders-of-magnitude if the diffusivity for that radionuclide were in the observed range of 10−15 to 10−14 m2/s. These results confirm the effectiveness of uncompacted bentonite-containing materials as engineered barriers for radioactive waste isolation.

Solubilities of Pu from leaching of a simulated HLW glass (C31–3-EC) spiked with 5 wt % Pu and from PuO2 are determined in salt solutions of different NaCl concentration and in groundwaters which are representative for the Gorleben area. Th.e leaching experiment is carried out either in an autoclave at 200 °C or under normal conditions (1 atm, 25°C). Solubilities of Pu in all investigated solutions are observed to be considerably higher than the values known from theoretical estimates. The predominant Pu species in solutions are found to be microcolloids, which are characterized by means of ultrafiltration and pulsed laser-induced photoacoustic spectroscopy.

Radionuclides breakthrough times as calculated through constant retardation factors obtained in dilute solutions are non-conservative. The constant retardation approach regards the solid as having infinite sorption capacity throughout the solid. However, as the solid becomes locally saturated, such as in the proximity of the waste form-packing materials interface, it will exhibit no retardation properties, and transport will take place as if the radionuclides were locally non-reactive. The magnitude of the effect of finite sorption capacity of the packing materials on radionuclide transport is discussed with reference to high-level waste package performance. An example based on literature sorption data indicates that the breakthrough time may be overpredicted by orders of magnitude using a constant retardation factor as compared to using the entire sorption isotherm to obtain a concentrationdependent retardation factor.

Temperature effects in the near-field radioactive waste disposal environment can result in changes in the adsorptive capacity and character of the substrate and the chemistry of the reacting fluids. This work examines the effect of temperature on 1) the kinetics of radionuclide sorption onto clays from 25°-75°C and 2) the degradation and metal-binding ability of two organic complexing agents found in chelated radioactive wastes and natural groundwaters. Clay-solution mixtures containing U, Co, Sr and Cs with kaolinite or montmorillonite were sampled over a 100-hour period for kinetic results at 25°, 50° and 75°C with and without EDTA. Three thermal degradation experiments with EDTA and oxalic acid were maintained at 250° for one week. Initial equilibrium concentrations were attained in the kinetics runs within 30 minutes in most cases. Possible sorption/retention mechanisms include ion exchange, physical adsorption and/or precipitation. EDTA decreased U and Co sorption, but slightly enhanced Sr and Cs sorption. Complexing of Al by EDTA may drive Sr and Cs to the clay surface to satisfy the resulting charge imbalance. Adsorption increased with increasing temperature for U, Co without EDTA and Sr, while Cs sorption decreased from 25°-75°C. Primary EDTA degration (at ∼200°C) products were mostly ethyl- and methyl-substituted amines with no metal-binding ability; therefore, to inhibit transport of low-level chelated radioactive wastes thermal or biological degradation is recommended. Interaction of high-level radioactive wastes with naturally-occurring organics in repository groundwaters may enhance migration of the actinides and 60Co.

In this study, the transport and capture rates of colloidal particles were calculated for a parallel-plate channel simulating a single fracture. The steady-state convective diffusion equation was solved with the particle velocity normal to the walls of the channel being the sum of the external forces acting on the particles. The forces considered were the gravitational, London-van der Waals and electric-double layer forces. The effects of parameters governing these forces and particle production mechanism on the rates of particle capture and transport are determined. The dynamic balance between particle production and capture has a significant effect on the concentration of particles leaving the fracture. The average particle velocity, though higher than the average fluid velocity, seems to be insensitive to phenomena governing particle capture.

The leaching behaviour of various ceramics was studied as part of a program to evaluate their potential use as container material for very long term containment of nuclear fuel waste under conditions of deep geologic disposal.

Samples of four grades of Al2O3, stabilized ZrO2, TiO2, a cermet (70% A12O3 - 30% TiC) and porcelain have been leached in deionized distilled water, Standard Canadian Shield Saline Solution (SCSSS), and SCSSS plus 20% Na-bentonite under static conditions at 100°C for periods up to 120 d.

The results of these ongoing leaching experiments suggest that the controlling features of the ceramic leaching process were leachant composition, and the presence and concentration of ions in the solution capable of preferentially precipitating on the ceramic surface, i.e., silicon, Mg2+, Ca2+, etc. Weight losses were determined after leaching. The initial leach rates showed a dependence upon ceramic purity and the processing used to prepare the ceramics, as well as the leachant composition. Also, it was apparent that the decrease of leach rates with time occurred as a result of the ceramic components in the leachant approaching their saturation limits.

Analysis of gas from a heated air-filled canister containing a spent fuel assembly before and after rupture of a fuel rod shows that about 1.5 mCi of 14C from the external surface of the assembly was rapidly oxidized and released as 14CO2 in excess oxygen at 275°C and 104rad/hr. After rupture, an additional 0.3 mCi was released, probably also from the external surface. The total 14C inventory in the entire 15×15 rod assembly including structural hardware is estimated to be 690 mCi. These measurements indicate that account will have to be taken of the time distribution of lifetimes of the canisters, and a broad definition of the “engineered system” may be necessary, in order to meet 1OCFR60 requirements with spent fuel in a repository in tuff.

Experimentally measured concentrations in spent UO2-fuel leachates are reported for uranium and plutonium for contact times ranging from 14 days to 2 years at different pH. The experimental results for uranium clearly indicated saturation. This interpretation was supported by comparison with theoretical solubilities, calculated using a modified SOLGASWATER program. The calculated solubilities were found to agree well with the experimentally determined concentrations.

Some ceramic materials may withstand granitic ground water for millions of years, or, ideally be thermodynamically stable under these conditions. Some conceptual design criteria and materials selection guidelines are discussed with reference to ceramic containers for spent fuel.

Rutile was chosen for practical studies. One reason for this choice was the possibility of joining lid and container below 950°C. TiO2 with 0.1–1.0 w/o B, added as B or B2O3, was densified at 900°C by hot pressing; 0.25 w/o B was used for bonding.

TiO2 from a sol-gel process was, without additions, used for bonding at 900°C and at 800°C. A new TiO2 raw material was, without sintering aids, pressed to high density at 930°C, and also used for joining experiments.

Some bonds are shown in micrographs. The major result is that nearly homogeneous bonding at 900°C seems feasible. The new raw material might possibly be processed to a monolithic ceramic body outside the fuel without any need for sealing.

Rutile was selected for some practical studies of processing and properties of ceramic containers. Hot isostatic pressing at 1280°C has resulted in reaction zones between the TiO2 powder and the steel canning. The phases ilmenite, pseudobrookite, rutile, and iron have been identified by X-ray diffraction and by microprobe analysis.

The microstructures have been interpreted by classical metallographic methods, and some microstructures obtained by hot pressing and rapid cooling have also been examined for purposes of comparison.

Some implications of the microstructures have been discussed in terms of microcracking and slow crack growth.

Part of a strategy for evaluating the compliance of geologic repositories with Federal regulations is a modeling approach that would provide realistic release estimates for a particular configuration of the engineered-barrier system. The objective is to avoid worst-case bounding assumptions that are physically impossible or excessively conservative and to obtain probabilitistic estimates of (1) the penetration time for metal barriers and (2) radionuclide-release rates for individually simulated waste packages after penetration has occurred. The conceptual model described in this paper will assume that release rates are explicitly related to such time-dependent processes as mass transfer, dissolution and precipitation, radionuclide decay, and variations in the geochemical environment. The conceptual model will take into account the reduction in the rates of waste-form dissolution and metal corrosion due to a buildup of chemical reaction products. The sorptive properties of the metal-barrier corrosion products in proximity to the waste form surface will also be included. Cumulative releases from the engineered-barrier system will be calculated by summing the releases from a probabilistically generated population of individual waste packages.

The Basalt Waste Isolation Project (BWIP) is conducted for the U.S. Department of Energy (DOE) by Rockwell Hanford Operations (Rockwell). The BWIP has undertaken an experimental program to test the behavior of defense waste glass under conditions relevant to a repository in basalt. Three autoclave experiments utilizing a nonradioactive simulated waste glass (Savannah River Laboratory (SRL) Type 131/TDS-3A) have been completed. Monolithic discs and crushed glass particles were reacted with crushed basalt and synthetic basalt groundwater for ∼1 yr at temperatures of 90 °C to 150 °C. The crushed qlass proved considerably more reactive than the monoliths at similar temperatures. Raising the temperature from 90 °C to 150 °C produced a striking increase in the extent of glass alteration. At 150 °C, essentially all of the boron and most of the sodium in the glass particles were released. The major reaction products in all three experiments were smectite clay and zeolites. The two tests utilizing crushed glass also yielded analcime. Nickel, cesium, and strontium, important elements in defense waste, were released from the glass in small quantities. The nickel was incorporated into the clay, while the cesium was incorporated into the analcime.

Hydrothermal waste package interaction tests were conducted with a mixture of crushed glass, basalt, and steel in methane-containing synthetic basalt groundwater. In the absence of gamma radiolysis, methane was found to have little influence on the corrosion behavior of the waste package constituents. Under gamma radiolysis, methane was found to significantly lower the solution oxidation potential when compared to identical tests without methane. In addition, colloidal hydrocarbon polymers that have been produced under the irradiation conditions of these experiments were not formed. The presence of the waste package constituents apparently inhibited the formation of the polymers. However, the mechanism which prevented their formation was not determined.

The change in permeability with time of granite, quartzite, anorthosite and gabbro was measured while these rocks were subjected to a temperature gradient. The highest temperature (at the heat source) was fixed at 250°C, while low temperatures ranged from 60 to 111°C, depending on rock type. Permeability reductions of up to two orders of magnitude were observed, with the greatest reactions occurring in the quartzite. These changes are thought to be caused by dissolution of minerals at high temperatures, and redeposition of the dissolved material at lower temperatures. Quartz appears to be an important mineral in this self-sealing process. If very low permeability is desired around a nuclear waste repository in crystalline rocks, then a quartz-rich rock may be the most appropriate host.

Previously reported results for Palo Duro deep brines show that Ra is highly soluble and not retarded. Relative to Ra, U and Th are highly sorbed. Uranium, like thorium, is in the +4 valence state, indicating a reducing environment. Additional data reported here support these results. However, one Wolfcamp brine sample gives somewhat different results. Radium appears to be somewhat sorbed. Uranium is largely in the +6 valence state, indicating a less reducing condition. In all brines, kinetics for sorption (228Th) and desorption (224Ra) are rapid. This Wolfcamp brine was tested for the effects of colloids for Ra, U, and Th concentrations. No effects were found.

The work reported here involved application of the ion-interaction approach of Pitzer [1] [2] to model the activities of some major and minor cations and major anions in deep brine systems at elevated temperatures and pressures.

The solubilities of anhydrite (CaS04), gypsum (CaSO4·2H2O), celestite (SrS04), barite (BaS04), and radium sulfate (RaSO4) in brines from the Wolfcamp Formation and granite wash facies, Palo Duro Basin, Texas, were modeled using ion-interaction equations. Waters with ionic strengths ranging from 2.89 to 4.76 m, and temperatures and pressures up to 40°C and 130 bars were sampled from five horizons in three wells, at depths between 970 and 1,670 meters. Theoretical solubility products as a function of temperature and pressure were obtained for comparison with ion activity products computed using ion-interaction theory. The effect of temperature on model calculations was found to reside almost entirely in the Debye-Hückel AΦ parameter, which accordingly was corrected for brine temperatures [3].

Modeling results indicated that all five brines are saturated with respect to anhydrite and celestite, and three of the five with respect to barite. Saturation may result from hydrologic connection with adjacent overlying evaporite units, or increased Ca, Sr and Ba concentrations caused by their desorption from clays. Radium concentrations, which range from 10-11.3 to 10-12.7 m, are not controlled by RaSO4 solubility or adsorption, but probably by solid solution in other sulfate minerals in these low pH (4.4 – 6.3) high calcium (0.18 – 0.52 m) brines [4] [5].

Gradients of temperature, pressure, and fluid composition in saturated clays give rise to coupled transport processes (thermal and chemical osmosis, thermal diffusion, ultrafiltration) in addition to the direct processes (advection and diffusion). One-dimensional transport of water and a solute in a saturated clay subjected to mild gradients of temperature and pressure was simulated numerically. When full coupling was accounted for, volume flux (specific discharge) was controlled by thermal osmosis and chemical osmosis. The two coupled fluxes were oppositely directed, producing a point of stagnation within the clay column. Solute flows were dominated by diffusion, chemical osmosis, and thermal osmosis. Chemical osmosis produced a significant flux of solute directed against the gradient of solute concentration; this effect reduced solute concentrations relative to the case without coupling. Predictions of mass transport in clays at nuclear waste repositories could be significantly in error if coupled transport processes are not accounted for.

The Office of Nuclear Waste Isolation (ONWI) has planned performance assessments that will quantify various performance measures for the eventual licensing of a repository in salt. In addition to studying “expected” conditions, selected discrete event scenarios must be analyzed. This paper presents a probabilistic analysis of a human intrusion scenario in which boreholes may be drilled into the repository, contact nuclear waste, and release the radionuclides to the environment. A stochastic analysis using draft EPA limiting factors regarding probability of drilling and quantities of water flow is compared to the proposed EPA 10,000 year limits. The years in which boreholes penetrate the repository are probabilistically determined in each iteration of the analysis. For each borehole, the probability of contacting radioactivity is an increasing function of time. When a borehole does make contact with radioactivity, the curie release of a given nuclide is a function of solubility limits and current inventory. Prediction intervals and complementary cumulative probability distribution functions are developed for the curie release as a fraction of the EPA limits. Radionuclide releases are small fractions of the EPA limits. This study shows that there is a high expectation that the EPA requirements would be met with a repository in salt.

Recent modifications to the EQ3/6 geochemical modeling software package [1–3] provide for the use of Pitzer's [4] equations to calculate the activity coefficients of aqueous species and the activity of water. These changes extend the range of solute concentrations over which the codes can be used to dependably calculate equilibria in geochemical systems, and permit the inclusion of ion pairs, complexes, and undissociated acids and bases as explicit component species in the Pitzer model. Comparisons of calculations made by the EQ3NR and EQ6 computer codes with experimental data confirm that the modifications not only allow the codes to accurately evaluate activity coefficients in concentrated solutions, but also permit prediction of solubility limits of evaporite minerals in brines at 25°C and elevated temperatures. Calculations for a few salts can be made at temperatures up to ∼300°C, but the temperature range for most electrolytes is constrained by the availability of requisite data to values ≤100°C. The implementation of Pitzer's equations in EQ3/6 allows application of these codes to problems involving calculation of geochemical equilibria in brines; such as evaluation of the chemical environment which might be anticipated for nuclear waste canisters located in a salt repository.

Packed columns of canister packing material containing 25% bentonite and 75% quartz or basalt sand, were exposed to water vapor at temperatures of up to 260°C. The permeabilities of the columns were subsequently measured after complete saturation with liquid water in a pressurized system. Exposure to water vapor caused irreversible increases in permeability by factors of up to 105. After saturation with liquid water, the permeability was nearly independent of temperature. The increases in permeability were due to a large decrease in the ability of the bentonite to swell in water. Calculations suggest that swelling of bentonite exposed to water vapor at 250°C was not sufficient to fill the pore spaces. If the pore spaces are filled, the mixture will form an effective barrier against flow, diffusion, and transport of colloids.

The results suggest that if bentonite-based canister packing material is exposed even briefly to water vapor at high temperatures in a high-level nuclear waste repository, its performance will be seriously impaired. The problem will be less severe if the proportion of bentonite is high and the material is highly compacted. Previous results show significant degradation of bentonite by water vapor at temperatures as low as 150°C. This suggests that in some repositories, backfill in tunnels and drifts may also be affected.