A series of experiments was conducted to evaluate the significance of redox potential, Eh, as an independent opensystem variable in the dissolution of nuclear waste glasses and ceramics. The standard platinum electrode can be used in leach solutions. Measurements on couples with rapid kinetics such as Fe+++/Fe++ give results in agreement with thermodynamic calculations. Highly reducing leaching solutions can be obtained by passing hydrogen through the solvent in the presence of a palladium catalyst.

Very rough estimates have been made of the total amount, the formation rate and spatial distribution of the Na metal colloid particles induced in rock salt adjacent to four types of radioactive waste canisters. A number of extrapolations were required. Salt immediately adjacent to a lightly shielded, 2.16 kW, high level waste canister could be converted entirely to colloidal Na (and presumably chlorine gas) in 200-400 years. The total Na metal formed will be 250-300 kg. A heavily shielded, 3.3 kW, spent fuel canister will convert roughly 0.3 percent of the salt at the canister surface to colloidal Na and the total sodium metal will be roughly 0.5 kg. Even at the lowest colloid levels the Na metal formed should greatly influence interactions between canisters and the surrounding salt, particularly if brine enters.

Surface analysis has been conducted on samples leached in a Sochlet apparatus at 100°C in the presence and in the absence of air. The XPS technique was applied to analyse the content of iron and silicon while the nuclear reaction method was utilized to analyse the content of hydrogen. Samples leached in argon atmosphere have shown a smaller content of iron and silicon with respect to the samples leached in air atmosphere. The H concentration has shown the same behaviour. The results are discussed in terms of possible formation of iron compounds in the different redox condition and of their different stabilities and in terms of their efficiency in reducing exchange between Na+ and H+ ions.

A model has been developed for the elemental leach rates as a function of time in the high-dilution limit for a two-phase composite glass material in which one phase exists as unconnected inclusions embedded within a major continuous phase. For those elements that are preferentially concentrated in this minor and presumed less durable phase, the leach rates are strongly time-dependent, decreasing significantly from the initial to the final steady-state value.

The ion-induced modifications of aqueous dissolution have been thoroughly investigated in several silicates (amorphous silica, quartz, soda-lime glass and albite) by applying depth profiling techniques (RBS and resonant nuclear reactions) for the characterization of leached surfaces. We show in particular that the H penetration after implantation and leaching is markedly increased when the ion dose exceeds a critical value comprised between 1013 and 1014 Pb.cm-2. This feature supports the radiation damage origin of the observed effects. Such experiments provide interesting clues about the aqueous dissolution of silicates and show that no important effect of α-decay is expected on the hydration of HLW glasses during disposal.

Assessment of the geologic performance of radioactive waste glasses requires extrapolation of finite tests to very long times. Hydration thermodynamics provides a means to compare the stability of waste glasses to natural analogues and to ancient synthetic glasses. The glass composition is separated into structural components of known free energy of hydration. These are then summed to provide a discrete measure of the stability of a given glass to aqueous attack. Hydration thermodynamics can be used to extend the results of laboratory tests of Savannah River waste glass to the repository environment.

The Nevada Nuclear Waste Storage Investigations (NNWSI) project under the Civilian Radioactive Waste Management Program is planning a repository at Yucca Mountain at the Nevada Test Site for isolation of high-level nuclear waste. Lawrence Livermore National Laboratory is developing designs for an engineered barrier system containing several barriers such as the waste form, a canister and/or an overpack, packing, and near field host rock. In this paper we address the selection of metal containment barriers.

Simulated radioactive waste glass was irradiated in a nuclear reactor and radiation damage was induced by 10B(n,0)7Li reaction up to 0.14 displacement per atom. Static leach tests of the activated waste glass were carried out at 40°C, up to 38 days. Concentration of elements in leachate was determined by gamma-spectrometry. The leach test of non-irradiated glass was also performed at 40°C up to 82 days in which the solution concentration of elements was determined by an atomic absorption analysis. Leach results are given in terms of normalized elemental mass loss for the seven elements Na, K, Cs, Ca, Zn, La and Ce. Normalized elemental mass loss of alkali ions was proportional to square root of time and then, was proportional to time. Normalized elemental mass loss of alkali earth ions did not show such a simple tendency. Build up of zinc on the waste glass surface was observed following its initial leaching. It was also found that theleach rate of Cs, Zn and Ce decreased with increase of irradiation exposure of waste glass.

The corrosion of iron overpack materials for nuclear waste forms in salt is highly dependent upon the species produced by the radiolysis of brines which may contact the metal. A computer code has been developed to predict the concentrations of the radiolytically produced species as a function of time using published values for radiation yields of species, chemical reaction rate constants, and parameters derived from experimental evidence.

The diffusion of technetium, iodine, uranium and neptunium in compacted bentonite has been studied. The possible reduction of the transport rate of these elements (i.e. redoxsensitive elements) by mixing the clay with metallic iron (for technetium, uranium and neptunium) or by adding a chemisorbent (for iodine) to the clay is reported. Technetium has an apparent diffusivity about 5 times higher in the heptavalent state (TcO4−) than in the tetravalent state (TcO(OH)2 or TcO2), uranium and neptunium in their higher oxidation state (VI and V) have apparent diffusivities about 6 and 50 times higher, respectively, than in the tetravalent state. Iodine, as I− (or 103−), has a transport rate more than one order of magnitude lower than TcO4−.

Two approaches for modeling radiation damage growth are presented and evaluated in the light of existing data. It results that purely geometric and statistical considerations about the distribution and overlap of damage are insufficient for accounting for the differences observed in leaching data from ion implantation and actinide doping experiments. It is shown, however, that the incorporation of effects such as those induced by the proximity of a free surface and the directionality of the irradiation into the kinetics of radiation damage ingrowth has the capability for accounting for the differences.

Solubilities of key solid compounds that are either present in the waste form or can readily precipitate from waste package leachates under repository conditions can be used to set maximum limits on radionuclide concentrations expected in groundwater. This is because the solubility limited concentrations are independent of the release scenarios, hydrologic transport characteristics, and adsorption/desorption reactions. Some of the important factors that control solubilities are pH, pe, type of solid phase, and nature of complexing ligands in the ground waters. Most of the above factors are affected by radiolysis due to the inherent radiation field of the waste form. Experimental results pertaining to the solubilities of selected Am, U, Np, and Pu compounds and the effects of radiolysis are discussed. These results show that: 1) at expected repository pH and reducing conditions, solubility controlled concentrations of several actinides are low and near acceptable limits, 2) the redox conditions at the waste form-water interface may be very oxidizing due to radiolyticeffects, despite the fact that normal repository conditions are assumed to be reducing, 3) additional data on solubility limits and key thermodynamic parameters are needed.

To give due credit to the barrier of the far field geologic environment in many host media it is necessary to account for sorption processes. The ultimate impact of sorption modelling will be in helping to define the degree of confidence that may be placed on geochemical retardation occurring in this barrier. For those involved in site characterisation, systems' design, design implementation and regulation, this confidence is best derived from a combination of a fundamental phenomenological understanding of the sorption process with empirical observations of sorption in natural environments. Neither alone is adequate. By performing a few additional measurements during classical Ko experiments, the data necessary for the more fundamental models, such as that of double-layer or surface ionization and complexation, may also be provided. The basis of these models and their integration into broader performance analysis are outlined in the context of how this maximises confidence in the geologic barrier of critically concerned groups.

The sorption of simple cations in tuff is dominated by adsorption on aluminosilicates that have charged surfaces, suc as zeolites and clays. The most significant sorbing minerals present in Nevada tuff are clinoptilolite, heulandite, mordenite, and montmorillonite. The kinetics of sorption on tuffs containing the minerals clinoptilolite and montmorillonite has been determined by studying the uptake of strontium, cesium, and barium on thin tuff wafers. The rate constants for uptake of these elements on tuff are consistent with a model of sorption that is diffusion limited but where diffusion occurs in two stages. First the cations diffuse into the rock through the water-filled pore space. Next, the cations must diffuse into the much narrower channels within the aluminosilicate crystals. After they are within the zeolite framework o between the clay planes, the cations may rapidly sorb on the negatively charged surfaces. The actinide elements have a time constant for the apparent sorption that is inconsistent with this model and may have a radically different mechanism o removal from solution.

The diffusion of non-sorbing species in different rock materials and fissure coating materials have been studied. The results show that the effective diffusivity of iodide in granites with fissure coating materials is of the same magnitude or higher as the effective diffusivity for iodide in granites without fissure coating material. Also the porosities of the granites with fissure coating material were higher. The effective diffusivity for iodide in rock materials without fissure coating material was determined to be from 1.i0-14 m2/s to about 7.10-13 m2/s.To simulate the stress that exists in the bedrock at large depths diffusion experiments with iodide in rock materials under mechanical stress, and electrical resistivity measurements in salt-water-saturated rock cores under stress have been started. Electrical resistivity measurements is an indirect method to determine the diffusivity.Results from some initial measurements show that the effective diffusivity is reduced to about 40 % of the value for unstressed samples at pressures of 230-280 bar. This is the expected stress at repository depths.

An important issue for assessing the feasibility of a nuclear waste repository located in basalt at the Hanford Site, Washington, is whether or not waters can percolate vertically through the rocks at the BWIP site. Flow in well defined horizontal, interbed units is well known, but vertical flow has not been fully investigated. Because vertical fractures are common, it is possible that some waters could penetrate the basalts and move normal to the horizontal flow layers. To investigate this, we have chosen to examine the chemistry and isotopics of minerals found in the fractures. These minerals may have originated earlier in the basalt history, i.e. deuteric, in which case their chemistry and isotopic composition should reflect that of the basalt. Alternately, they may have formed late in the basalt's history, in which case their chemistry may be a reflection of material deposited from percolating waters. Thirdly, they may represent a mixing of the basalt and groundwater reservoirs. We examined this issue by determining the strontium isotopic composition of the various reservoirs, supported by other chemical studies. In this fashion we have carefully investigated the origin of the strontium and other trace elements in the fracture filling minerals. To date, waters from surface or subsurface reservoirs at Hanford have not been provided for isotopic analysis, although we plan to run these as soon as they are available. Our conclusions on the strontium isotopic work are therefore qualitative to semi-quantitative at best, but the results can be directly applied to the source of the strontium in the fracture filling minerals.

The use of sediment laid down in a Scottish loch during the Flandrian marine transgression as a source term for a “natural analogue” study of elemental migration in clays over a timescale of 103 − 104 y has been investigated. Depth profiles of over 20 elements have been measured which illustrate varying post depositional behaviour ranging from effectively total immobility to complete redistribution throughout the sediment profile. These results are contrasted with laboratory batch sorption data for a range of nuclides on samples of this sediment.

The formation of colloidal particle fractions of triand tetravalent actinides (Am(III), Th(IV), Pu) has been studied in aqueous solutions under various conditions. (Variations of nuclide concentration 10-7-10-11 M; pH 3-12; salt medium 0.01-1.0 M NaCl04; storage time 6 h-6 months; temperature 5°C, 25°C, 70°C). A large fraction of the nuclides is sorbed on the vessel walls, but there is also a formation of centrifugable particles. For americium this fraction has two maxima, at pH 5-8 and at pH above 12. For thorium and plutonium this fraction is very small except at pH about 12. In the pH-range 7-9, which is expected in underground waste repositories, the particle fractions do not seem to be highly mobile. Plutonium behaves very similar to thorium in all studied systems, so it is probably tetravalent to a large extent.

The rates of polymerization and disproportionation of Pu(IV) have been studied using low concentrations: (1.7 − 10) × 10−6M Pu, (0.8 − 12) × 10−4M HCI and 0.01M ionic strength. Osmium(II) complexes such as the tris−4,41−2,21−bipyridine complex were found to react rapidly with Pu(IV) but very slowly, if at all, with Pu(IV) polymer, Pu(lll), or Pu(V). Thus, it is possible to determine unreacted Pu(IV) in the presence of reaction products by using Os(II) complexes. Disproportionation reaction products, Pu(IlI) and Pu(V), were determined using their reactions with Ce(IV) sulfate. We find −d[Pu(IV)]/dt = k'[Pu(IV)]2 at constant pH. Log k1 varies from about 4.25 at pH 3 to about 7.0 at pH 4.1 (units for k1 are M−1min−1). The [H+] dependence varies from about −2 to −3 over the pH range studied. The measured rate is the sum of those for polymerization and disproportionation; the latter reaction amounts to about 75% of the total at pH 3 and 20% at pH 4. The second-order rate constants for disproportionation are very much larger than expected on the basis of extrapolation from 0.2 to 1.OM HClO, solutions. The products of the reaction do not affect the rate, but U(VI), aged Pu(IV) polymer, and CO2 increase the rate.

The solubilities of crystalline Am(OH)3 and AmOHCO3 were measured at 25° C in aqueous solutions of 0.1 M NaCl04 by determination of the solution concentrations of Am. Prior to use in the measurements, the solid materials were characterized by their x-ray powder diffraction patterns. The solubility product quotients were calculated from the experimental data. The hydrolysis quotients of Am3+ were also estimated from the hydroxide solubility data. Using the thermodynamic data derived from these experiments and the recently reported formation constants for the Am3+ carbonate complexes, the solid phases and concentrations of solution species of americium in several aqueous carbonate systems were calculated using the computer code MINEQL.