The neutralized high-level waste, stored at the Western New York Nuclear Service Center in West Valley, New York, was produced during the operation of the Nuclear Fuel Service, Inc. commercial nuclear fuel reprocessing plant. The supernatant is a highly concentrated salt solution (NaNO3 , NaOH, Na2SO4 and NaCl) containing essentially all of the dissolved cesium as the primary radioactive component. The sludge is primarily iron and aluminum hydroxides and contains strontium and the bulk of the long-lived isotopes. The supernatant will be treated to remove essentially all of the radioactivity and then be concentrated and disposed of as low level nuclear waste. The following supernatant treatment considerations have been evaluated on a laboratory scale using simulated West Valley waste: 1) Organic ion exchange resins; 2) Inorganic ion exchange media; 3) In-tank processing. These processes will be described and preliminary laboratory data will be presented.

Corrosion resistant canisters are an important component of the engineered barriers of a nuclear waste disposal system. In addition to providing containment for the waste form during transportation and emplacement, a durable canister can eliminate hydrothermal interactions with the waste form. The selection of a repository site and emplacement technique will affect canister design and could also impact material selection. While there are still many issues to be resolved, there are two different concepts being evaluated to provide durable canisters for waste disposal: (1) canisters fabricated out of extremely corrosion resistant materials such as Ti-base or Ni-base alloys, and (2) canisters fabricated out of less durable materials but designed with a corrosion allowance. Each of these types of canisters could fail to meet the design objectives through a variety of failure processes. The more important of these are discussed.

Progress in the Canadian program to develop durable containment systems for the direct disposal of irradiated fuel in a deep geologic vault is described. Possible containment lifetimes, and the factors that influence container design and materials selection, are discussed. Most effort is currently directed toward developing containers with a design lifetime of about 500 years, using a corrosion-resistant metal shell to provide the primary barrier to radionuclide release. The design details for various containers are described, together with the fabrication and testing programs that are in progress or planned. The container materials programs consist of both generic research, to enable predictions of container performance to be made, and engineering development, to evaluate methods of container fabrication and inspection.

Effects of microstructure and chemistry on corrosion and hydrogen embrittlement of TiCode-12 have been investigated. Heat treating mill-annealed TiCode-12 in the temperature range 500–750°C results in a sensitization effect, i.e., an increase in the uniform corrosion rate when tested in either boiling 1N HCl or a MgCl2 brine. This effect is caused by microstructural changes involving the precipitation of Ti2Ni. Electrochemical studies indicate that sensitization results from galvanic coupling between Ti2Ni, acting as cathode, and the α-Ti matrix with resultant shifts in corrosion potential. Increasing the Fe content of the alloy promotes the sensitization effect.

Hydrogen embrittlement of TiCode-12 was studied using the slow strain rate technique. Hydrogen concentrations to 130 wppm did not cause embrittlement. However, above 220 wppm hydrogen, degradation was observed as a decrease in mechanical behavior and the appearance of cleavage-like secondary cracks. The embrittlement was dependent on temperature and alloy chemistry but not environment (air or brine).

The influence of the main alloying elements of six different steels and high temperature resistant alloys for crucibles containing HAW glass on the homogeneity of the waste glass was investigated. The distributions of iron, chromium, nickel, and cobalt were studied by an electron beam microprobe. The results are given and discussed.

Three experiments performed on cement borehole seals in granite indicate flow through the seal is slightly lower than flow through the intact rock. The experiments were performed at stress states simulating depths of approximately 350, 700, and 1000 meters.

In order to predict the range of thermal conductivities to be expected in waste forms and in geologic media, an understanding of the pertinent phonon scattering processes is required. It has been shown that grain boundaries in polycrystalline media are unimportant at low temperatures relative to lamellae which arise from twinning, exsolution, or foreign inclusions within the grains. The possible role of lamellar structures on the conductivity at high temperatures will be discussed.

A mixture of bentonite clay and quartz sand is being considered for use as a waste-package backfill, the material placed between a radioactive-waste canister and the repository host rock. Compacts of bentonite/quartz with weight-percent ratios of 100/0, 70/30, 50/50 and 30/70 were made at room temperature under a pressure of 100 MPa (15 ksi). Upon initial heating, the thermal conductivity of the 70/30 compact rose from 1.20 W/m·K at 298 K to 1.32 W/m·K at 373 K. After further heating to 473 K, it fell to 1.10 W/m·K, reflecting the loss of interlamellar water from the bentonite. The conductivity of the now-dehydrated compact was reproducible through several heating and cooling cycles, varying from 1.15 W/m·K at 573 K to 1.03 W/m·K at 298 K. The other mixtures were qualitatively similar to the 70/30: the 100/0, 50/50 and 30/70 dehydrated compacts displayed conductivities of 0.59, 1.06 and 0.83 W/m.K,respectively, at 298 K. Measured densities ranged from 1.98 to 2.12 g/cc. Combined geometric-mean and Maxwell models for thermal conduction in composite systems predict the measured results reasonably well. An analysis of the impact of backfills on high-level waste (HLW) package design indicates that no significant thermal penalty is imposed.

Sodium bentonite is a candidate material for the waste package backfill component in a repository in basalt at the Hanford Site. Preliminary hydrothernmal experiments have been conducted under near-field geochemical conditions expected to occur in the reference repository location in the Grande Ronde Basalt. Experiments have been conducted in the basalt/groundwater, bentonite/groundwater, and basalt/bentonite/groundwater systems. The experiments have been conducted at 300°C using a simulated Grande Ronde groundwater, reference Umtanum basalt, and sodium bentonite. Key data generated by the experiments include experimental solution analyses as a function of time and preliminary solids analysis by scanning transmission electron microscopy and X-ray diffraction. Solution trends of the major aqueous species were similar in the three systems and are characterized by: (1) the gradual reduction of the pH value from ∼9.75 to a steady-state value of ∼6, (2) an initial rapid increase followed by a gradual decrease in silica concentration, and (3) a slight or negligible increase in sodium, sulfate, and chloride concentrations. In the bentonite/groundwater experiment, small amounts (<1%) of an albite reaction product were observed. Conversely, the formation of illite, a common bentonite alteration product, was not observed. These results indicate that sodium bentonite will remain sufficiently stable at 300°C under hydrothermal conditions in basalt to permit its use as a backfill material.

The diffusion of cesium(I), strontium(II), pertechnetate and europium in brine-saturated backfill materials was measured. Plastic diffusion cells containing cylindrical diffusion columns were used for low density backfill materials. The diffusion of gamma-emitters was followed by a gamma scanning technique. Metal diffusion cells constructed entirely from Hastelloy C-276 were used for the diffusion of pertechnetate in highly compacted bentonite. Apparent distribution coefficients calculated from diffusion data are (a) 0.02 m3 /kg for cesium(I) in 40 wt.% mordenite and 60 wt.% bentonite; (b) 0.04 m3/kg for strontium(II) in 10 wt.% sodium titanate and 90 wt.% bentonite; (c) 0.5 m3/kg for pertechnetate in 70 wt.% charcoal and 30 wt.% bentonite; and (d) 3 m3/kg for europium in 100% bentonite. Backfill effectiveness estimates based on batch sorption measurements were supported by these results;however, the diffusion results for europium did not agree well with a model for diffusion retarded by linear sorption. First measurements of pertechnetate diffusion in highly compacted bentonite suggest that anion exclusion may play a role in reducing mass transport rates of anions in this material. Needs for diffusion measurements that take into account site-specific materials interactions are described.

Addition of an inertial term to the constitutive relation for mass flux leads to a mass transport equation which is hyperbolic and describes propagation of a distorted wave with a finite velocity. This approach eliminates the “instantaneous propagation paradox” inherent in parabolic transport equations based on Fick's law. Analytical solutions of the wave transport equation have been derived and have the properties that the leading edge of a solute front propagates at a finite, predictable velocity and is truncated by a step function which decreases in magnitude exponentially with time. The inertial effects on computed solute fronts are most evident near the leading edge, and have potential significance in the prediction of engineered barrier performance.

The effects of internal hydrogen on the mechanical properties of Ti Code 12 were studied by gas phase charging smooth bar tensile samples. Tensile tests showed no change in yield strength with increasing hydrogen concentration. A slight decrease in ductility and fracture stress at hydrogen concentrations above 200 wppm accompanied by secondary cracking along grain boundaries indicated embrittlement. At hydrogen concentrations near and above 400 wppm, cleavage fracture associated with strain-induced hydrides was also observed.

The corrosion behavior of TiCode-12 (Ti-0.3 Mo-0.8 Ni) high level nuclear waste container alloy has been studied for a simulated WIPP brine at a temperature of 150°C or below. Crevice corrosion was identified as a potentially important failure mode for this material. Within a mechanical crevice, a thick oxide film was found and shown to be the rutile form of TiO2, with a trace of lower oxide also present. Acidic conditions were found to cause a breakdown of the passive oxide layer. Solution aeration and increased acidity accelerate the corrosion rate. In hydrogen embrittlement studies, it was found that hydrogen causes a significant decrease in the apparent stress intensity level in fracture mechanics samples. Hydride formation is thought to be responsible for crack initiation. Stress corrosion cracking under static loads was not observed. Attention has also been given to methods for extrapolating short term uniform corrosion rate data to extended times.

Argillaceous rocks, used either as a repository rock or as a backfill material provide a reactive medium with which escaped radionuclides can react. Solutions containing U, Nd Gd, Cs, and Sr were reacted with bulk specimens of two argillites with the objective of determining penetration of heavy metals into the rock, and the reaction products. The two Devonian argillites were chosen to represent a highly reducing rock containing pyrite and carbonaceous material as accessory minerals, and a clean, alumino-silicate rock containing mainly quartz and layer silicates. Cylinders of rock were reacted with solutions at saturated steam pressures and 200°C.

At the end of the runs, both rocks contained reduced uranium in the form of U02.25. The rare earths were associated with the uranium. There was a precipitated coating of uranium-rich material on the outside of the reducing shale and few of the radionuclides had penetrated the interior. Uranium also formed a precipitate within 0.5 mm of the surface of the oxidized shale leaving a depleted zone near the surface. Cesium and strontium did not penetrate the reducing shale but were uniformly distributed throughout the oxidized shale. Chlorite, apatite and pyrite were found to be the most reactive phases in the shales.

A laboratory method for measurement of diffusion of radionuclides inconcretes and compacted bentonite has been developed. In this method a tracer is introduced through one end into the cylindrical sample closed in a tube and preequilibrated with water. After the introduction period of the tracer the tube is sealed hermetically and the concentration profiles of the radionuclides are measured periodically from the outside of the sample using a collimated detector. Diffusivities are calculated from the activity profiles by fitting the theoretical curve with the measured profile. The paper will give diffusion data for Co, Sr and Cs in some concrete products and for Co and Cs in compacted sodium bentonite.