This study was undertaken to clarify the alteration process of French LWR synthetic glass leached by distilled water at 250 and 300°C for 28 days. The present experiments emphasize “in situ” identification of the alteration products in small areas (20 × 10−4 mm2 ) along the alteration profile by a specific X-ray diffraction technique.

Experimental assembly: The sample is irradiated by an incident X-ray beam (Co k α, 40 mA, 40 Kv) focused to a 150 μ m diameter. The sample oscillates around a θ central selected position and the diffracted X-rays are detected simultaneously in a 14.5 (2 θ ) solid angle, then recorded by the linear position sensitive detector and stored in a multichannel analyzer.

A systematic study of the chemical durability of glass compositions developed for the immobilization of high-level radioactive wastes at West Valley is described. Most of the compositions studied here represent variations around the nonradioactive WV205 composition. The results from this study indicate that the effectiveness of the various oxides tested in increasing the durability of WV205, when considered in terms of 2% additions by weight to the central composition, decreases in the following order: A12O3>ZrO2>SiO2>Cr2O3>Fe2O3>TiO2>CuO>ZnO>NiO. The addition of 2–6% Cr2O3 or NiO to WV205 was observed to cause extensive crystallization. The addition of 2% or 6% ZnO or 2% TiO2, CuO or NiO to the glass actually enhances corrosion rates. A12O3 and ZrO2 are the best additives in terms of increasing the durability. The additions of ThO2 and U3O8 slightly increase the durability. The effects of various oxides on the durability of WV-205-based glasses with ThO2 and U3O8 were found to be very similar to the effects of these oxides on the durability of the corresponding glasses which did not contain ThO2 and U3O8. This study also showed that Fe2O3 has little effect on the durability. However, FeO greatly enhances the corrosion rates of WV205-based glasses. Much less pronounced effects of the redox state of iron were found in the case of more durable base glass compositions.

The aim of these measurements is to study how a silica saturated solution influences leaching from a HLW glass matrix. Cylindrical samples of I117 borosilicate glasses were suspended in polycarbonate vials containing a SiO2 saturated solution and leached at temperature of 50 or 80°C for different times (from 1/2 to 28 days). Mass loss, X-ray photoelectron spectroscopy (XPS) and nuclear reaction techniques were applied for post leaching tests on the glass samples.

Analysis of boron and sodium released from the glass were performed in the leachate. Sodium and boron release followed a square root time dependence at all the observed leaching times.

XPS measurements revealed a surface increase of silicon content with respect to unleached samples, with aluminium showing the same behaviour as silicon. The presence of the layer enriched in silicon and aluminium does not stop the release of sodium and boron, whereas the mass loss does slow down because of the formation of aluminio-silicate compounds.

Annealing kinetics of fission tracks (produced by induced fission of U235 in natural and artificial (doped) obsidian glass has been studied by etching microscopy as a function of annealing time and temperature. The observed annealing characteristics can be best decribed by a two term exponential function with an Arrhenius type temperature dependence of the annealing coefficients. The activation energies deduced are (0.94 ± 0.16)eV and (1.04 ± 0.15)eV for artificial obsidian and (0.83 ± O.13)eV and (0.60 ± 0.16)eV for natural (fossil) obsidian, respectively. In addition, leaching experiments of obsidian samples exposed to different radiation, doses and rates have been made at two different leachate temperatures. At a temperature of 60°C the leaching rate of irradiated samples is much larger than that of unirradiated samples, whereas at a temperature of 90°C observed leaching rates are similar.

Experiments have been performed on the aqueous dissolution of French LWR solution reference glass that have identified key dissolution mechanisms to be considered in modeling. The physical hypotheses on which the model relies are a dissolution rate governed by the silica concentration in solution according to a first order kinetic law, congruent release of the mobile elements at the gel-glass interface, and diffusion of silicic acid in the gel interstitial solution. A comparison of theoretical results obtained by a semi-analytical solution to the model and experimental results suggests some modifications to develop a more advanced model.

The reprocessing of spent fuel from nuclear reactors and processing of fuels for defense purposes have generated large volumes of high-level liquid waste that need to be immobilized prior to final storage. For immobilization, the wastes must be converted to a less soluble solid, and, although other waste forms exist, glass currently appears to be the choice for the transuranic-containing portion of the reprocessed waste. Once produced, this glass will be sent in canisters to a geologic repository located some 200 to 500 m below the surface of the earth.

Both PNL76−68 and ABS-118 glasses show greater stability in seawater (SW) than in distilled and deionized water (DDW), in the MCC-1 90°C corrosion test. After 180 days of corrosion, the boron release from PNL76- 68 and ABS-118 glasses is about five and two times lower respectively in SW than in DDW. Seawater leachates show less increase in pH than DDW leachates. Glass corroded in SW shows little or no total weight loss. Significant amounts of Mg from SW have been added to the surface layers of both types of glass. Mg-silicates were found in surface layers formed in SW, whereas Zn- and Fe-silicates in surface layers formed in DDW. Shortterm leaching tests in D2O, NaCl and MgCl2 solutions suggested that high salt content in leachant can impede glass hydration. Both silica saturation and Mg diffusion may be responsible for greater stability of the glasses in SW than in DDW.

Analytical electron microscopy was used to characterize leached layers formed on a synthetic basalt glass after 300 days in deionized water, sodium chloride solutions, or sodium chloride solutions containing magnesium chloride. Ultramicrotomed cross-sections show that the leached layers consist of a mixture of amorphous and crystalline phases whose composition and microstructure are sensitive to solution chemistry. In general, Na, Ca and Mg are selectively leached in all solutions, leaving behind surface alteration products which are rich in Fe, Ti, and Al. In deionized water and the NaCl solution, gibbsite (AI(OH)3) is a major phase; in the Mg-containing solution, Si and Al are retained as smectite clays. Results are compared to the alteration of natural basalt glasses in marine environments, particularly as they relate to the use of basalt glass alteration as a means of verifying the long-term leaching behavior of nuclear waste form borosilicate glasses.

The environment during the operating period of a high-level nuclear waste repository in basalt is expected to be air saturated with steam. Liquid groundwater is not expected to be in contact with the container surface during that time.

The report presents corrosion findings from tests conducted for one to twenty-five months in an air-steam environment. Tests were carried out with bare metal specimens exposed to an air atmosphere containing 12% moisture in chambers maintained at temperatures between 150°C and 300°C. Cast carbon steel exhibited total penetrations less than 0.002 mm for exposures up to 25 months. A ferritic alloy steel, Fe9CrlMo, showed corrosion results very similar to cast carbon steel. Unalloyed copper materials showed essentially linear corrosion rates, with total penetrations between 0.002 mm at 150°C and 0.14 mm at 300°C in 25 months. Cupronickel 90−10 exhibited total penetrations between 0.001 mm at 150°C and 0.05 mm at 300°C in 25 months. There was a tendency for the corrosion rate to increase with time for cupronickel at 250°C and 300°C possibly because of a mid-test change in the corrosion mechanism. Limited testing of specimens surrounded with bentonite/basalt packing material indicated that the presence of packing has no strong effect on the corrosion of iron-base materials; however, copper-base and cupronickel materials corroded at higher rates in the presence of packing, with a possible shift towards the lower bare specimen corrosion rates with increasing time.

The U.S. Department of Energy's Salt Repository Project (SRP) is investigating the general corrosion resistance of cast mild steel as a candidate material for waste package containers. Evaluation of this material is being performed at the Pacific Northwest Laboratory in environments simulating expected repository conditions.

General corrosion studies of mild steel (ASTM A216 grade WCA) in the as-cast and normalized conditions were conducted in hydrothermal halitesaturated (saturated at ambient temperature) brine environments simulating a “dissolution” and an “inclusion” brine. Corrosion tests were also performed in brines similar to the inclusion brine but containing magnesium concentrations ranging from 1000 to 30,000 ppm to investigate the effect of magnesium on the corrosion behavior.

Corrosion rates of the cast mild steel were found to increase with increasing temperature and with increasing magnesium concentration. Some possible mechanisms that explain the observed behavior are presented.

Recent studies have indicated that localized corrosion of a relatively small area of a waste container may impair the containment function to such an extent that larger releases may be possible than from the bare waste form. This would take place when a large number of holes coexist on the container while their concentration fields do not interact significantly with each other. After performing a steady state analysis of the release from a hole, it is shown that much fewer independent holes can coexist on a container surface than previously estimated. The calculated radionuclide release from multiple independent holes must be changed accordingly. Previous analyses did not proceed to a correct application of the linear superposition principle. This resulted in unacceptable physical conclusions and undue strain on the performance assessment necessary for a container licensing procedure. The paper also analyzes the steady state release from penetrations of finite length and whose concentration fields interact with one another. The predicted release from these penetrations is lower than the previously calculated release from holes of zero thickness. It is concluded here that the steady-state release from multiple holes on a waste container can not exceed the release from the bare waste form and that multiple perforations need not be a serious liability to container performance.

A series of corrosion tests were sponsored by the International Lead Zinc Research Organization, Inc. and conducted by Battelle Northwest to obtain information regarding the suitability of lead for applications in the containment of high-level nuclear waste. Two grades of unalloyed lead and two lead alloys were exposed to basalt, salt, and tuff simulated repository environments for times up to 1 year and temperatures from 55 to 200°C. Information was obtained on the corrosion rates, corrosion products, and concentration of lead in the test liquids under these conditions, which were intended to simulate the anticipated environments in candidate U.S. nuclear waste repositories. Leachates from the 12-month tests were analyzed in order to prepare synthetic leachates that were used to conduct scoping evaluations of the transport of lead in basalt and tuff environments.

The Nuclear Waste Management Program at Lawrence Livermore National Laboratory is responsible for the development of the waste package design to meet the Nuclear Regulatory Commission licensing requirements for the Nevada Nuclear Waste Storage Investigations (NNWSI) Project. The metallic container component of the waste package is required to assist in providing substantially complete containment of the waste for a period of up to 1000 years. Long term phase stability of the austenitic candidate materials (304L and 316L stainless steels and alloy 825) over this time period at moderate temperatures (100–250°C) can impact the mechanical and corrosion behavior of the metal barrier.

A review of the technical literature with respect to phase stability of 304L, 316L and 825 is presented. The impact of martensitic transformations, carbide precipitation and intermediate (σ. χ, and η) phase formation on the mechanical properties and corrosion behavior of these alloys at repository relevant conditions is discussed. The effect of sensitization on intergranular stress corrosion cracking (IGSCC) of each alloy is also addressed. A summary of the impact of phase stability on the degradation of each alloy in the proposed repository environment is included.

The hydrothermal stability of copper has been studied to assess its suitability as a container material for disposal of nuclear waste in the proposed repository site at Hanford, Wa. The experiments (Cohassett basalt, synthetic Grande Ronde #4 groundwater, and copper powder with a water:rock:Cu-powder mass ratio of 20:1:1) were conducted at 300°C, 30 MPa, using Dickson rocking autoclaves for periods up to 3000 hours. Redox was calculated from dissolved H2 measured by gas chromatograph and He-ionization detector.

After ca. 100 hours the solution Cu concentration stabilized at 2–3 ppm, near Cu saturation, and did not vary significantly during the remainder of the experiment. The in situ solution pH was slightly alkaline. The copper powder showed little evidence of etching. The Cu concentration did not reflect oxide-coating spallation effects such as those described by Johnston et al.[15]. Within 48 hours, the log fO2 values decreased rapidly toward the magnetite-hematite phase boundary (−31 at 300°C); this is well within the stability field of native copper. SEM and EDX analysis of the reaction products revealed a copper-iron sulfide. One experiment used copper powder containing ca. 5% cuprite. Oxygen released by the cuprite overwhelmed the buffering capacity of the basalt. The resulting log fO2 values stabilized near the copper-cuprite phase boundary (−23.1 at 300°C) with the solution remaining within the copper stability field. Copper purity is important as oxygen contamination or oxidation of the copper containers may strongly affect the repository redox and mobility of radionuclides.