Chemical and U-Th isotopic data on unsaturated zone waters from the Nopal I natural analog reveal effects of water-rock interaction and help constrain models of radionuclide release and transport at the site and, by analogy, at the proposed nuclear waste repository at Yucca Mountain. Geochemical reaction-path modeling indicates that, under oxidizing conditions, dissolution of uraninite (spent fuel analog) by these waters will lead to eventual schoepite precipitation regardless of initial silica concentration provided that groundwater is not continuously replenished. Thus, less soluble uranyl silicates may not dominate the initial alteration assemblage and keep dissolved U concentrations low. Uranium-series activity ratios are consistent with models of U transport at the site and display varying degrees of leaching versus recoil mobilization. Thorium concentrations may reflect the importance of colloidal transport of low-solubility radionuclides in the unsaturated zone.

Natural nuclear reactors provide unique evidence in helping to understand the processes that might occur over long timescales in radioactive waste disposal sites. In the presented work, the extent and kinetics of oxidation of core material from the Oklo-Bangombé natural reactors are investigated. The X-ray powder diffraction analysis shows that the uraninites core samples from the Bangombé Reactor and Oklo Reactor 2. and Oklo Reactor 13 have the same unit-cell parameters as synthetic UO2.25. A significant amount of fourmarierite, Pb(UO2)4O3(OH)4 4H2O, was identified in the core samples from two shallow reactors Bangombé and Oklo 2, but not in the deeper reactor Oklo 13. The results of U(IV)/U(IV) measurements indicate that the extent of oxidative weathering of shallow reactors (Bangombé and Oklo 2) is greater than for the deeper reactor Oklo 13. Evaporable organic compounds found in the uraninite inclusion containing “bitumen” at the edge of Okelobondo Reactor (400 °C) and in the black shale immediately above the Bangombé Reactor (260 °C) may work as a reducing buffer or/and a hydrophobic water shield to depress the oxidative dissolution of the uraninite cores.

The Hyrkkölä U-Cu mineralization located in south-western Finland is reassessed with reference to the corrosion mechanisms affecting the stability of native copper and the time-scales of corrosion processes. The mineral assemblage native copper – copper sulfide occurs in open fractures at several depth intervals within granite pegmatites (GP). The surfaces of these open fractures have accumulations of uranophane crystals and other unidentified uranyl compounds. The secondary uranium minerals are mainly distributed around copper sulfide grains. Microscopic intergrowths of copper sulfides and uranyl compounds also have been observed. Groundwater samples were collected from the vicinity of the Cu samples. The hydrogeochemical features of these samples indicate that the present conditions are oxidising. The minimum age of U(VI) transport and deposition is about 200 000 years. This age is indicated by 234U/238U and 230Th/234U activity ratios of uranophane. The age of the hexavalent uranium precipitation may be somewhat later than the last influxes and/or demobilisation of sulfur.

The mineral assemblage native copper – copper oxide (cuprite) occurs only at one depth interval within altered granite pegmatite. The fracture surface was coated by smectite. The content of uranium in smectite was 69–75 ppm U. The 234U/238U and 230Th/234U activity ratios of smectite showed that it has been exposed to recent groundwaters (e.g., during the last million years). The pH of the groundwater at this interval was near neutral (6.9). The copper grains present at this fracture surface were as large as 1 mm in diameter and had rims of cuprite of 0.01 to 0.1 mm thick. The smallest grains were totally oxidised.

The Nopal I uranium (U) deposit, Peñia Blanca, Mexico is a source term and contaminant transport natural analog to the proposed high-level nuclear waste repository at Yucca Mountain, Nevada. In an attempt to characterize the mobilization and potential transport of adionuclides in the unsaturated zone at the Nopal I deposit, vegetation growing on ore piles was analyzed for 238U, 235U, and 232Th decay-series isotopes. Specimens of Phacelia robusta growing on highgrade piles of U ore were collected and analyzed by alpha autoradiography, and by alpha and gamma spectrometry. Activities for U, thorium (Th), and radium (Ra) isotopes (Bq/kg dried plant) were 300, 1000, and 7000 for 238U, 230Th, and 226Ra, respectively. The 226Ra activities in these specimens are among the highest ever measured for plants; furthermore, the plant-to-soil 226Ra concentration ratio is higher than expected. These results demonstrate the large mobility and bio-availability of Ra in the Nopal I environment, and support previous indications of recent loss of 226Ra from the ore body. Comparison between the activities of 238U and 232Th decay-chain Th isotopes in the plants and in the ore substrate indicate that relative mobilization into pore solutions of 228Th > 230Th > 232Th, in a ratio of about 50 – 25:4:1, respectively. The similarity of the plant's 234U/238U activity ratio (˜1.2) to that of a caliche deposit that formed adjacent to the Nopal ore body around 54 ka suggests the 234U/238U activity ratio of U released from the ore is approximately 1.2. The U and 236Ra isotope activities of the plants and ore substrate, and solubility considerations, are used to assess a source term model of the potential Yucca Mountain repository. These results suggest the use of a natural analog source term model in performance assessments may be non-conservative.

The corrosion performance of spent nuclear fuel waste packages is becoming increasingly important in establishing the viability of the proposed Yucca Mountain repository system. Current package concepts propose the use of a 2 cm thick nickel-base superalloy (Alloy 22) shell as the main barrier to prevent corrosion penetration over many thousands of years. The expected package service conditions, as well as their variability and uncertainty, are discussed. The electrochemical conditions known to be responsible for passive behavior and its breakdown in Alloy 22 and similar alloys are examined in the light of the predicted repository environment. Durability prediction approaches and their conclusions are considered. Efforts to determine the relative impact of localized modes of failure and uniform passive dissolution on package durability are reviewed, along with open issues in need of resolution and alternative package designs. The basic question of the validity of extrapolating corrosion behavior over many times the duration of the present base of experience is addressed.

Alloy 22 [UNS N06022] is now being considered for construction of high level waste containers to be emplaced at Yucca Mountain and elsewhere. In essence, this alloy is 20.0–22.5% Cr, 12.5–14.5% Mo, 2.0–6.0% Fe, 2.5–3.5% W, with the balance being Ni. Other impurity elements include P, Si, S, Mn, Co and V. Cobalt may be present at a maximum concentration of 2.5%. Detailed mechanistic models have been developed to account for the corrosion of Alloy 22 surfaces in crevices that will inevitably form. Such occluded areas experience substantial decreases in pH, with corresponding elevations in chloride concentration. Experimental work has been undertaken to validate the crevice corrosion model, including parallel studies with 304 stainless steel.

Coupons of titanium alloys under consideration as components of the Engineered Barrier System in the proposed repository at Yucca Mountain have been evaluatedfor their passive film composition and stability. Oxide depths and compositions on specimens exposed in long-term corrosion testing for one year were determined with x-ray photoemission spectroscopy. The specimens removed from long-term testing, as well as separate coupons polarized cathodically in an electrochemical cell, exhibited blistering associated with hydride formation in both scanning electron microscopy and atomic force microscopy.

Mild steel has been investigated as one of the overpack materials for geological disposal of high-level nuclear waste. The present paper discussed the passivation behavior of mild steel using an electrochemical approach to evaluate the critical condition for passivation. The critical current density for passivation, icris , of mild steel in 0.1 mol /L [HCO3 + CO3 2−] solution at room temperature was found to decrease with increasing pH of the solution. The depassivation pH, pHd , at room temperature was determined to be about 7.5 by comparisons between icris and the diffusion-limiting current density, iL02 , for dissolved oxygen of 8.1 ppm in equilibrium with atmospheric oxygen. The anodic current density of mild steel in the solution of pH = 8 at 90°C was also measured and was found to decrease rapidly with time, leading to the absence of a critical current density for passivation. Thus, mild steel can be self-passivated in the system. An empirically determined corrosion phase diagram is provided for mild steel in 0.1 mol/L [HCO3 − + CO3 2−] solution at 90°C, based on these findings and some published data.

For the geologic disposal of the high level nuclear waste (HLW), aqueous corrosion is considered to be the most important factor in the long-term performance of containers, which are the main components of the engineered barrier subsystem. Container life, in turn, is important to the overall performance of the repository system. The proposed container designs and materials have evolved to include multiple barriers and highly corrosion resistant Ni-Cr-Mo alloys, such as Alloys 625 and C-22. Calculations of container life require knowledge of the initiation time and growth rate of localized corrosion. In the absence of localized corrosion, the rate of general or uniform dissolution, given by the passive current density of these materials, is needed. The onset of localized corrosion may be predicted by using the repassivation and corrosion potentials of the candidate container materials in the range of expected repository environments. In initial corrosion tests, chloride was identified as the most detrimental anionic species to the performance of Ni-Cr-Mo alloys. Repassivation potential measurements for Alloys 825, 625, and C-22, conducted over a wide range of chloride concentrations and temperatures, are reported. In addition, steady state passive current density, which will determine the container lifetime in the absence of localized corrosion, was measured for Alloy C-22 under various environmental conditions.

The extent of stress corrosion cracking (SCC) of copper nuclear waste containers is being predicted on the basis of a “limited propagation” argument. In this argument, it is accepted that crack initiation may occur, but it is argued that the environmental conditions and material properties required for a through-wall crack to propagate will not be present.

In this paper, the effect of one environmental parameter, the supply of oxidant (Jox), on the crack growth rate is examined. Experiments have been conducted on two grades of Cu in NaNO2 environments using two loading techniques. The supply of oxidant has been varied either electrochemically in bulk solution using different applied current densities or by embedding the loaded test specimens in compacted buffer material containing O2 as the oxidant. Measured and theoretical crack growth rates as a function of Jox are compared with the predicted oxidant flux to the containers in a disposal vault and an estimate of the maximum crack depth on a container obtained.

Application of pure, oxygen-free copper as a construction material having excllent corrosion resistance is limited because of the effect of intergranular emnibrittleinent at temperatures above 100–150°C. Dilute copper alloys containing S, P, and Ag inpirities and vacancies are studied theoretically on the basis of total energy calculations. The dissolution energies, volunie misfits, and defect interaction energies are calculated and used to study the microscopic miechanisin behind the effect of these impurities on the eiibrittlenieiit of colpper at interinedia temiperatures. A large linding energy of a sulfur-vacancy defect pair (–0.46 eV) is found. The sulfur-vacancy and sulftir-sulftir interactions in the copper matrix seeni to favor precipitation of copper sulphide Cu2S which is the most probable cause of the einhrittleiieiit. The effect of phosphorus and silver ilmpurities on the einbrittlement of sulfiir-contaminated copper can he related to their competition with sulfuir to attract to vacancies as well as to other lattice defects.

Testing of waste containers under open field conditions is a component part of the research program that is being carried out at SIA “Radon”for more than 20 years to understand the long-term behaviour of radioactive waste forms and waste packages. This paper presents the preliminary results of these ongoing studies. We used a typical NPP operational waste, containing 137Cs, 134Cs, and 60Co as the dominant radioactive constituents. Bituminized and vitrified waste samples with 30–50 wt.% waste loading were prepared. Combined effects of climatic factors on corrosion behaviour of carbon steel containers were estimated using gravimetric and chemical analyses. The observations suggest that uniform corrosion of containers prevails under open field conditions. The upper limits for the lifetime of containers were derived from calculations based on the model of atmospheric steel corrosion. Estimated lifetime values range from 300 to 600 years for carbon steel containers with the wall thickness of 2 mm containing vitrified waste, and from 450 to 500 years for containers with the wall thickness of 2.5 mm that were used for bituminized waste. However, following the most conservative method, pitting corrosion may cause container integrity failure after 60 to 90 years of exposure.

This paper describes a modeling study for general corrosion of copper which is a candidate material for high-level radioactive waste overpacks. The model is a mixed-potential model combined with diffusive transport of reactants and reaction products. The rest potential and corrosion rate of copper in aerated solution were measured while controlling the thickness of a diffusive solution layer on the copper surface using a rotating-disk electrode. Experimental data were used for validation of the model.

The phase stability of C-22 alloy (UNS #N06022) was studied by aging samples at 593, 649, 704 and 760°C for 2000 h (2.7 mo) and 16,000 h (1.8 yr). The tensile properties and the Charpy impact toughness of these samples were measured in the mill annealed condition as well as after aging. The microstructures of samples aged 16,000 hours were examined using scanning and transmission electron microscopy (SEM and TEM). Preliminary TEM results suggest that μse forms at all temperatures investigated. Discrete carbide particles in addition to a film with very uniform thickness which appears to be μ phase formed on grain boundaries in the sample aged at 593°C. The ordered Ni2(Cr, Mo) phase was also seen in this sample. At the higher aging temperatures, mainly μ phase forms covering all the grain boundaries and also distributed throughout the bulk. Although strength increased somewhat with aging. the ductility decreased due to the formation of these grain boundary precipitates and brittle intermetallics.

Ceramic coated carbon steel coupons were corrosion tested in water with dissolved salts to simulate exposure to evaporation concentrated groundwater in an underground nuclear repository. Metallography revealed no corrosion at the ceramic metal interface of dense coatings, even though electrical measurements demonstrated that the coatings were slightly porous. Experimental results and a model to predict corrosion rates influenced by a porous ceramic coating and coating lifetimes are presented.

Assessments are made of the corrosion characteristics of spent nuclear fuel Zircaloy cladding in a Yucca mountain repository environment and the potential for the cladding to provide protection against radionuclide release following waste package failure. Considerations and assumptions includes a waste package life near 10,000 years and air-saturated water contacted with waste package corrosion product goethite, based on the near-field geochemical environment evaluated in the Yucca Mountain Viability Assessment [3]. Literature corrosion data (general, pitting, and localized crevice attack) are evaluated on the basis of these conditions and the expected chemical environments that can result on the surface of the fuel. General corrosion of Zircaloy is expected to be negligible and result in a lifetime of the SNF cladding of several hundred thousand years, approaching a million years. General surface pitting is not expected. Effects of crevice localized corrosion for periods beyond 10,000 years are uncertain and require modeling development and experimental characterization. Details of the evaluations that provide the basis for the conclusions are presented.

Stainless steel-zirconium (SS-Zr) alloys have been developed as waste forms for the disposal of metallic waste generated during the electrometallurgical treatment of spent nuclear fuel. The waste forms incorporate irradiated cladding hulls, components of the alloy fuel, noble metal fission products, and actinide elements. The baseline waste form is a stainless steel- 15 wt% zirconium (SS-15Zr) alloy. This article presents microstructures and some of the corrosion studies being conducted on the waste form alloys. Electrochemical corrosio n, immersion corrosion, and vapor hydration tests have been performed on various alloy compositions to evaluate corrosion behavior and resistance to selective leaching of simulated fission products. The SS-Zr waste forms immobilize and retain fission products very effectively and show potential for acceptance as high-level nuclear waste forms.

Argonne National Laboratory is developing an electrometallurgical treatment for spent fuel from the experimental breeder reactor II. A product of this treatment process is a metal waste form that incorporates the stainless steel cladding hulls, zirconium from the fuel and the fission products that are noble to the process, i.e., Tc, Ru, Pd, Rh, Ag. The nominal composition of this waste form is stainless steel/15 wt% zirconium/ 1–4 wt% noble metal fission products. The behavior of technetium is of particular importance from a disposal point of view for this waste form due to its long half life, 2.14E5 years, and its mobility in groundwater. To address these concerns a limited number of spiked metal waste forms were produced containing Tc. These surrogate waste forms were then studied using scanning electron microscopy (SEM) and selected leaching tests.

Metals and alloys with relatively low melting temperatures such as lead and lead-based alloys are considered in Russia as prospective matrices for encapsulation of spent nuclear fuel in containers in preparation for final disposal in underground repositories. Now lead and leadbased alloys are being used for conditioning spent sealed radioactive sources at radioactive waste disposal facilities.

The solubility of nickel under anoxic conditions was investigated in groundwater and in cement-conditioned water. The solubility was studied at 30 °C using the over-saturation method. The nickel concentrations were determined using 63Ni and liquid scintillation counting. The results indicate a nickel solubility of 1–10−4 mol/L in the cement water. The same solubility was also seen in groundwaters that had a pH of 9 or higher. No solubility limit was determined for groundwaters with pH values in the range of 7 to 9.