The effects of waste composition and percent loading in a borosilicate glass designed for US defense high level wastes (HLW) have been evaluated. Three types of simulated wastes were investigated; high alumina, high iron and a composite representative of an average waste composition from Savannah River Plant (SRP) waste tanks. Corrosion resistance of the borosilicate glass is significantly enhanced by the presence of any of the three types of wastes. Additionally, corrosion resistance is improved as the % waste loading is increased in the glass. The best corrosion performance was obtained with the high alumina waste in deionized water.

Leachability is one of the most important properties of solidified nuclear waste forms because it provides information on the performance and the subsequent safety and reliability that the waste products will possess. One of the most important experimental findings in the leachability field has been the discovery and subsequent detailed characterization of protective surface layers that form on waste glass during leaching. These layers can have a beneficial effect on product performance while in storage by improving productdurability with time. As a result of surface layer formation and the effects on subsequent product leaching characteristics, new qualitative and quantitative leaching models have recently been proposed.

Remobilization of solidified high-level waste by corrosion and transport of radionuclides by groundwater are the only likely events for radioactivity to find its way back to the environment. The interaction of the waste form with aqueous solutions is the most important mechanism and must be known for safety and risk analysis. It has been shown that the activity release depends on a number of experimental and environmental parameters. The corrosion mechanisms of the waste forms control the leach rate of individual components. Of particular importance to the resistance against water attack of various glass compositions is the formation of layers on the glass surface. The results for leaching in deionized water reveafed that the mechanism was strongly influenced by the formation of a surface layer which determined the further attack on the glass. In this work the specific weight losses of the glass after leaching in rock salt and MgCl2–MgSO4-NaCl-KCl solutions were measured and surface layers investigated.

Volcanic glass of basaltic composition (approximately 50 wt.% SiO2) was studied as a natural analogue to proposed nuclear waste form glasses. Basaltic glass can resist devitrification and remain glassy for periods in excess of 10 m.y. Surface alteration of the glass occurs rapidly upon exposure to water or hydrothermal fluids, but is essentially absent if the glass remains dry. Alteration is restricted to a rind of clay-like, cryptocrystalline palagonite, generally less than 100 μm thick. Formation of this rind protects the underlying glass from further alteration. Similar rinds, or “gel layers”, have been produced in laboratory and field tests with borosilicate glasses. Major cations within the basaltic glass are selectively depleted or enriched relative to the glass in the alteration rind. Qualitatively-similar depletion and enrichment patterns characterize the short-term leaching behavior of candidate waste form glasses.

Safety and risk analyses for the isolation of radioactive waste in a repository must begin with a source term to quantify the amount of radioactivity released from the waste form under a specific set of conditions. The interaction of the waste form with aqueous solutions is the most important mechanism to consider, as any material released may be dissolved and reach the biosphere. In this regard the behaviour of heat generating high-level waste is of particular importance, because reaction rates are higher at elevated temperatures. A long-term leach rate was derived from previous and continuing experimental work. The purpose of this paper is not to describe the “real case” release but rather to provide guidelines for the design of leaching experiments and determine the required precision for the data. This can be derived from the relative sensitivity of extrapolated leach rates for various parameters measured in laboratory experiments.

A model has been developed to predict the long-term leach or release rates of various waste-form materials under repository conditions.

The effects of repository material and a tailored backfill or overpack on the leaching behavior of glass have been studied. Two types of glasses were used in this investigation: 1) Model glasses comprised of 33 mol% alkali oxide-67 mol% SiO2 , and 2) alkali borosilicate glasses with and without simulated wastes. Several types of repository material were placed in the same containers as the glasses to determine if their presence would alter the extent of leaching. A backfill material consisting of phosphate slime/sand mixtures was evaluated with the same procedure. The results indicate that the leaching behavior of the glass may be influenced by the presence of some materials. Preliminary results of a 16½ month burial experiment in Florida are also discussed.

The work reported here is part of a study of the leaching of individual elements from waste glasses. The first part of the paper reports on leaching of the major constituents at 30°C, and leaching of the alkalis at 2°C from glass UK189; the second part describes leaching of Sr, Cs, Tc and various actinides from glasses UK189 and UK209. Compositions are given in Table 1.

The leachability of solid nuclear waste forms has been studied by many investigators to evaluate the short-term kinetics of elemental release or to determine the effects of leachant composition or other system parameters. Some general observations from these studies have included: incongruent leaching; formation of reaction layers that contain rare-earth, alkaline-earth, or transition metal elements; and apparent saturation of some elements in solution while others continue to be leached. Before these observations can be used to predict long-term performance of waste glasses under repository conditions, the mechanisms controlling the release of elements from the solid must be understood.

When the leaching of glasses is taken into consideration in risk analysis models, it is necessary to make a distinction between accidental and normal conditions in the repository. In accidental conditions, it is assumed that the glass will be leached by ground-water which can also be renewed. The composition of this water will be related to the aquifer existing around the repository. Under normal repository conditions the leaching process has to be analyzed separately for each geological formation considered.

The aim of the work was to measure the resistance of simulated nuclear waste glasses to leaching by water and salt solution under the conditions of elevated temperature and pressure. Of particular interest was the behaviour of Am and Np.

When a canister containing a borosilicate glass incorporating highly radioactive waste is put into an underground repository, care must be taken in favouring the dispersion of the generated decay heat by conduction in the surrounding rock. For this purpose the space existing between the glass container and the repository walls is filled with a material having an acceptable thermal conductivity. If the backfilling material is highly compacted, it can form a zone of very low permeability reducing in this way the flow of water around the canister.

This is one of two papers discussing the findings in an in situ-burial experiment, presently being performed in the Stripa mine in Sweden. The purpose of the experiment is to evaluate the effects of various components in the SKBF/KBS waste storage system on the leaching of the vitreous waste form. Two configurations of glass, canister, overpack and buffer/backfill materials were designed (Figs. 1 and 2). Both configurations were inserted into 56 mm diameter boreholes in the Stripa mine and maintained at 90°C. One of the configurations (Fig. 2) was also kept at ambient temperature, 8°C. In the experiments two glass types were used, ABS 39 and ABS 41 (Table 1). These glasses, developed by Dr. T. Lakatos of the Swedish Glass Research Institute, contain 9% simulated fission products by weight and are compatible with the French AVM process.

An essential component of any assessment of HLRW geological disposal options is the quantitative prediction of radionuclide release rates from the near-field over time spans of the order of 103-106 years. Fundamental to this assessment is the investigation of the interaction of potential wasteforms with groundwater under repository conditions of temperature, pressure, and groundwater flow-rate. Consequently, many studies world-wide have been initiated to examine the kinetics of wasteform dissolution over a wide range of physical and chemical conditions. Although these studies have provided a considerable amount of invaluable data on wasteform-fluid interactions, they have tended to focus on breakdown of the wasteform itself, and not on the fate of released waste components in the nearfield. For example, effects of saturation of species in solution, precipitation of secondary minerals or amorphous gels, and the effect of host rock chemistry on the products (solid and fluid) of waste-fluid interaction have largely been ignored or even specifically excluded in laboratory experiments. This is despite growing evidence from source term modelling studies which suggest that the above processes may well be the chief factors in governing rates of radionuclide release from the near-field, bearing in mind the limited availability of ground

The purpose of this experiment was to evaluate the effects of various components of the SKBF/KBS nuclear waste storage system on the leaching of the vitreous waste form. Two cohfigurations of nuclear waste glasses, canisters, overpacks, and backfill mate-7ial were inserted into 5.6 cm x 3 m deep boreholes located at the 350 m level in the STRIPA mine. Some were maintained at 90°C. The others were allowed to equilibrate at the ambient temperature of the mine, approximately 8°C. Two borosilicate nuclear waste glass compositions (termed ABS 39 and ABS 41) compatible with the French AVM process containing 9 percent by weight of simulated fission products were compared. The two compositions ( Table 1) bracket the range of Si02/Na2O/B203 ratios likely to be selected for commercial vitrification operations at La Hague.

The time dependence of leaching and dissolution under hydrothermal conditions at 250°C has been investigated in an autoclave designed for continuous sampling; it is found that the rates are in accord with those obtained by the commonly used procedures. The effect of pressure by itself has been studied in an ultracentrifuge; ambient rates for Si and Na loss remain unchanged up to about 40 MPa followed by a monotonic increase to about 1.5 times the ambient rates at 160 MPa. The surface layers formed as a result of chemical attack have been examined (XRD, IR, XPS and PIXE). It is found that Zn present in the glass or in the solution is retained in the siliceous layer which in some circumstances contains a crystalline zinchydroxy- silicate phase. A method for enhancing rates of displacement damage by a factor of ∼106 has been developed. Powdered borosilicate glass and glass doped with UO2 have been neutron-irradiated in a high-flux reactor. Damage is thereby introduced in the bulk as a result of 10B(n,α)7Li and (n,fission) events and subsequent slowing-down of energetic charged ions. Irradiation doses equivalent to 106 years HLW storage have been achieved. The resultant stored energy of ∼102 kJ kg−1 is released over the range 300 to 800 K. The initial leaching/dissolution rates are found to be increased by a factor of about three by irradiation; the changes in longer-term rates are difficult to determine due to the effects of back-reactions and changing particle size geometry and size distribution, but significant enhancement is found. Rates for fission and activation products are similar to those for bulk species.

The German concept of high level waste final storage provides the use of certain glasses containing radioelement oxides as glass components. These waste forms are to be stored in rock salt formations in order to isolate the waste from the biosphere. The efficiency of this isolation is a most important question. The aim is to achieve a high safety standard that remains valid under extreme conditions such as the uncontrolled water entrance to the deposit.

The Nuclear Waste Materials Characterization Center at Pacific Northwest Laboratory is developing standard tests to obtain data on nuclear waste forms, barriers, and backfills. These tests include performance measurements of thermal, radiation, mechanical, and chemical properties. Five tests have been developed to determine the chemical durability of waste forms under either static (MCC-IP and MCC-2P) or flowing (MCC-4S and MCC-5S) leaching environments. Maximum credible release by waste forms is determined using powders and stirred solutions (MCC-3S).

The Rutherford backscattering technique is useful for the determination of the concentration profiles of some heavy elements in the near surface region of glasses, but is not able to provide chemical information on the elements detected.

Cm-sized blocks of a simple borosilicate glass have been leached in hydrothermal conditions with various rates of water renewal. Weight loss measurements reveal that restricted water access does not slow down the dissolution rate of this particular glass as would be expected from solubility limits, but instead accelerates its destruction and its transformation into a secondary solid phase still ill-defined.