The modelling of radionuclide release from waste forms is an important part of the performance assessment of a potential, high-level radioactive waste repository. Since spent fuel consists of UO2 containing actinide elements and other fission products, it is necessary to determine the principal parameters affecting UO2 dissolution and quantify their effects on the dissolution rate before any prediction of long term release rates of radionuclides from the spent fuel can be made.

Assessing the concept of direct disposal of used nuclear fuel in a geological vault requires a model to predict the dissolution rate of UO2 in groundwater. A solubility-limited model can be used to calculate the dissolution rate of UO2 fuel under non-oxidizing conditions. When the oxidative dissolution of UO2 is an irreversible process, a kinetic model is more suitable to describe the dissolution of UO2 under oxidizing conditions. Experimental studies were carried out using electrochemical techniques and X-ray photoelectron spectroscopy, XPS, to determine criteria for selecting the appropriate model for estimating used-fuel dissolution rates as a function of the redox conditions in the vault at the time of container failure. UO2 electrodes were subjected to prolonged (>1000 min) potentiostatic oxidation, and the rate of oxidation and dissolution of UO2 fuel was investigated as a function of the applied potential. UO2 oxidation was also carried out by the products of water radiolysis and studied as a function of dose rate, total dose and solution chemistry.

These studies show that significant oxidative dissolution of UO2 appears possible for potentials more positive than -100 mV vs SCE in solutions with a pH close to that of the deep groundwaters, i.e., from 6 to 10. A kinetic model, which takes into account the mechanism of UO2 oxidation, is more appropriate to estimate dissolution rates of UO2 fuel for redox conditions more oxidizing than -100 mV vs SCE.

Soluble radionuclides concentrated within the gap and grain-boundary regions of spent fuel could dissolve relatively rapidly were the waste container to fail and the fuel to be contacted by water in a geologic repository. To provide an estimate of the quantities of radionuclides that may be rapidly released, fractional inventories of Cs, Tc, and Sr concentrated within the fuel/cladding gap region have been measured for U.S. LWR spent fuels with fission gas release (FGR) values ranging from 0.25% to 18%. Separate measurements of the grain-boundary inventories of Cs, Tc, and Sr have been made for the same fuels. The Cs gap inventories were generally found to be about one fourth of the FGR values. The Cs grain-boundary inventories were generally less than 1% of the total Cs inventories and were not strongly correlated with FGR. Both the gap and grain-boundary inventories of Tc and Sr were near the detection limits of the methods used, less than 0.2% of the total inventories of these elements. However, some of the Tc may reside at the grain boundaries in the form of relatively insoluble metallic particles and not be detected by these experiments. Data obtained by comparing the dissolution behavior of fuel fragments with that of fuel grains were used to estimate the dissolution rate of Cs from the grain boundaries of one of the fuels. Surface-area normalized dissolution rates determined for fuel fragments in these same tests exceeded those determined for grains. A likely explanation is that the estimated fragment surface area did not take into account the “effective” grain-boundary surfaces.

Fission products in nuclear fuels exhibit a significant variation in solubility and stability. To understand better the reasons for this variation, the Mott-Littleton simulation technique has been used to calculate solution energies of the fission products Br, Kr, Rb, Sr, Y, Zr, Te, I, Xe, Cs, Ba, La and Ce. Nonstoichiometry and variable charge state are important components of the model.

Instant-release fractions were measured for release of 14C, 99Tc, 90Sr and 137Cs from segments of clad used CANDU fuels in distilled deionized water at pH 8.5 and 25°C. Measured–instant release values, 0.002 to 0.07% for 14C, 0.001 to 0.23% for 99Tc and 0.0005 to 0.04% for 9 0 Sr, were one to two orders of magnitude lower than the estimated values currently used in the safety assessment of the Canadian concept of geological disposal of intact used CANDU fuel. The values for 14C were also lower than values reported in the literature for PWR fuels.

Recent results on the various methods for the recovery and the mutual separation of Mo, Ru, Rh and Pd from the simulated insoluble residue in the reprocessing plant, such as liquid lead and zinc extractions, nitric acid dissolution and preferential oxidative vaporization in various atmospheres by the authors are briefly reviewed. By combining these results, four kinds of processes for the recovery and the mutual separation of Mo, Ru, Rh and Pd (and Tc) i.e. two processes composed of Pb (and Zn) extraction(s) and subsequent nitric acid dissolution, and two processes using vaporization in various oxidizing atmospheres at various temperatures are proposed.

Over the past few years, there has been an increased awareness of the potential hazard of energetic chemical reactions in high-level radioactive waste tanks at the Hanford tank farm. In particular, a mixture of Na2NiFe(CN)6 with NaNO3 and NaNO2 in several high-level waste tanks has caused concern. The problem of the FeCN tanks is fundamentally one of a potentially unstable mixture of fuel (the CN- moiety) and oxidizer (NO3- or NO2-).

At Los Alamos National Laboratory, we have performed an extensive reanalysis of the safety problems associated with the presence of Na2NiFe(CN)6 mixed with NaNO3/NO2 for a particular tank (104-BY) that contains by far the largest amount of the nickel ferrocyanide salt (∼2E5 mol). Our approach is to use conservative assumptions to bound both the energy density for a potential runaway reaction and the mass that could participate if we assume ignition as the result of bounding radionuclide concentrations. The subsequent progress of the accident is analyzed using an advanced hydrodynamics computer code called MESA to evaluate the loads on the structure and the generation of aerosols. The subsequent doses are shown to be low both on and off the site. The conservatism in the analysis is quite large, and the expected results using more realistic assumptions are discussed.

In previous corrosion studies, carbon steels, especially the fine-grained steel TStE355, were identified as promising materials for heat-generating nuclear waste containers acting as a barrier in a rock-salt repository. In the present study detailed investigations have been performed on fine-grained steel to determine the influence of important parameters on its corrosion behaviour in disposal-relevant salt brines. These parameters are: brine composition (Mg Cl2-rich and NaCI-rich brines), temperature (90°C, 170°C), and salt impurities, such as H2S concentrations of 25 mg/I-200 mg/I salt brine.

Under the conditions of the tests used here, carbon steel was subjected to general corrosion. Pitting and crevice corrosion or stress-corrosion cracking were not observed. The increase in temperature from 90°C to 170°C strongly enhanced the corrosion rate of the steel. In the MgCl2-rich brines, considerably higher rates (37-70 μm/a at 90°C, 200-300 °m/a at 170°C) were observed than in the NaCI-rich brine (5 μm/a at 90°C, 46 μm/a at 170°C). H2S concentrations in the MgCl2-rich Qbrine of up to 200 mg/l did not influence significantly the corrosion rate of the steel. The corrosion rates determined imply corrosion allowances that are technically acceptable for thick-walled containers. In view of these results, fine-grained steel continues to be considered as a promising material for long-lived HLW containers.

The published information on the effect of radiation on the corrosion of potential waste container materials is reviewed. The materials discussed are irons and carbon steels, copper and copper alloys, stainless steels, nickel-based alloys, and titanium alloys. In general, a dose rate of >5 Gy·h-1 is required to accelerate the general corrosion of corrosionallowance materials. The effects of radiation on the localized corrosion of corrosion-resistant materials appear to be contradictory. In some cases, localized corrosion is induced, whereas in others it is inhibited.

It has been established that the mild steels which undergo the general corrosion in the acidic to neutral environments, attain the passivity status in alkaline environments, thereby becoming liable to the localized corrosion, such as pitting corrosion and crevice corrosion. Now, for the case of using bentonite as a buffer to stand between the hostrock and the geological disposal packages of high level nuclear waste, localized corrosion behaviors of mild steel as a candidate for such a package has been studied quantitatively for environments where the the otherwise neutral ground water would be turned slightly alkaline with pH = 9.5 - 10.0.

In view of the lack of quantitative data on the passivity-to-localized corrosion of mild steel in natural water environments of weak alkalinity, the present authors have previously determined an empirical E-pH diagram for mild steel with a 20 °C, 1 m mol/L [HC03-], 10 ppm [CI -] solution simulating the natural water environment concerned; it has been shown that the general corrosion-to-passivity transition condition was determined to be pHd =9.4, and the mild steel was shown to be liable to localized corrosion over a large portion of the passivity domain. The present paper discusses behaviors, mechanisms, and critical conditions for initiation of localized corrosion in mild steel placed in bentonitesuspending natural water environment, in terms of the critical potentials for pitting (Vc), and crevice corrosion (ER,CREV). Bentonite was addid to the solution in varying amounts to give bentonite-to-solution ratios up to 0.1, while the pHvalue was adjusted appropriately with sodium carbonate, always keeping the bentonite particles in suspension.

It is demonstrated that bentonite particles suspended in water will deposit upon the steel on receipt of Fe2 + ions, thereby promoting pitting corrosion by preventing repassivation and promoting crevice corrosion by acting as an effective crevice, once the environment conditions become favorable for localized corrosion.

We conclude therefore that disposal package made of mild steel and placed in an underground water environment with bentonite as buffer will be liable to localized corrosion.

Previous corrosion studies have shown that the passive alloys Ti 99.8-Pd and Hastelloy C4 are the most promising corrosion protection materials for long-term resistant carbon-steel containers for the disposal of heat-generating wastes in rock salt formations. Consequently, these materials are subject to more detailed investigations. The influence of selected container manufacturing characteristics (sealing technique, application mode of the corrosion protecting layer on the steel) on the corrosion behaviour of Ti 99.8-Pd and Hastelloy C4was studied using in-situ experiments. The corrosion resistance of the materials was examined both in NaCI brine and MgCl2-rich brine, which might enter into an HLW borehole under certain disposal conditions. Four electron-beam welded cast-steel tubes plated with corrosion protection layers made of Ti 99.8-Pd and Hastelloy C4, respectively, were stored for 18 months in 2-m deep heated boreholes in the Asse salt mine. The annular gap between the tubes and the borehole wall was filled with saturated NaCl-brine or Qbrine (MgCl2-rich). The vertical temperature profile in the boreholes ranged from 90°C to 200°C

In the cast-steel tube with a corrosion protection of Ti 99.8-Pd, no indication of corrosion was observed on the EB-welds or on the explosion plated material in both brines. The cast-steel tube with a corrosion protection of Hastelloy C4 also was found to be resistant to any kind of corrosion in the NaCl-brine. In the MgCl2-rich brine, however, small pitting corrosion of 15 pm was observed after 1.5 years on the passive surface layer that was formed.

On the basis of these results, Ti 99.8-Pd continues to be considered as a promising corrosion protection material for long-lived HLW containers.

This paper describes experiments and analyses conducted to determine the range of apparent diffusion coefficients of lead diffusing from an intentionally perforated lead-matrix titanium-shell container into a compacted 1:1 (by dry mass) silica-sand/sodium-bentonite buffer mixture saturated with Standard Canadian Shield Saline Solution at 363 K. Analysis of the experimental data using a single apparent diffusion coefficient could not explain the findings. A possible explanation of the behaviour is presented here. It uses a 2–D finite-element model with six lead species having six different apparent diffusion coefficients. The model can explain the data satisfactorily. Sixty-three percent of the source concentration consisted of slow-moving species, with an apparent diffusion coefficient of 10-15 m2 /s, whereas the fastest species, with an apparent diffusion coefficient of 10-10 m2 /s, constituted only three percent of the source concentration.

Titanium and copper have been proposed as suitable container materials for disposal of nuclear fuel waste in plutonic rock of the Canadian Shield. Studies of the corrosion of these materials have led to the development of container failure models to predict long-term performance. Crevice corrosion and hydrogen-induced cracking of titanium have been identified as potential failure mechanisms, and these two processes have been studied in detail. Using data from these studies as well as a number of conservative assumptions, titanium container lifetimes of 1200 to 7000 a have been estimated. For copper, general corrosion has been studied in detail in bulk solution and in compacted clay-based buffer material. Results indicate that the copper corrosion rate is likely to be controlled by the rate of transport of copper species away from the container surface. An assessment of copper pitting data suggests that pitting is an extremely improbable failure mechanism. The copper container failure model predicts minimum container lifetimes of 30 000 a. The results demonstrate that long lifetime containment can be provided, should performance assessment studies indicate the need for such an option.

X-ray photoelectron spectroscopy (XPS) was used to study samples from the Cigar Lake uranium deposit in northern Saskatchewan. Two uranium-rich sections, CS-615 and CS-620 were studied. Peaks due to U, 0, C, Pb, Si, Ca, Al, S, Cu and Th were seen in the XPS spectra. Concentrations of Pb of up to 14 wt% were measured by XPS and correspond to the Pb ingrowth by radioactive decay during the 1.3 billion years since the ore was formed. High-resolution spectra were recorded for the U, Pb, 0, and C bands. Lead was in the +2 oxidation state in all samples. The carbon signal indicated the presence of organic carbon, while the oxygen bands indicated the presence of significant amounts of water in the samples. The samples from the CS-615 section had U6+/U4+ ratios between 0.16 and 0.29. However, the U6+/U4+ ratios for section CS-620 were about 0.7 but this is attributed to oxidation of the ore by water during the drilling and cutting processes. The significance of the results for disposal concepts in nuclear fuel waste management programs is discussed.

In an open pit uranium mine at Pogos de Caldas (PDC) in Brazil, uranium oxide has accumulated at the redox front. The redox front forms and moves by the oxidation of pyrite in the rock by oxygen infiltrating with rainwater. The rock was formed nearly 90 million years ago by volcanic activity. Between 3 and 12 km of rock has eroded away since the formation of the rock. The redox front lies between tens and several hundred meters below the ground surface. The front is uneven and locally there are redox “fingers” which extend to depths of hundreds of meters. The redox fingers are often associated with fractures or fracture zones.

The uranyl oxide hydrates are common initial corrosion products of uraninite (nominally U02+x) during weathering. In the presence of dissolved silica these early-formed phases alter to uranyl silicates (most commonly soddyite, U2SiO8-2H2O, and uranophane, CaU2Si2O11·6H2O). Uraninite, however, usually contains radiogenic Pb, and the earlyformed Pb-poor uranyl oxide hydrates alter incongruously to uranyl silicates plus Pb-enriched uranyl oxide hydrates such as curite. Similar to dissolved silica, radiogenic Pb may also serve to limit the mobility of U in nature by fixing U in solid phases. Curite may also play an important role in the formation of uranyl phosphates, which are significantly less soluble than the uranyl silicates, and control U solubility in many groundwaters associated with altered U ore.

The chemical forms of iron minerals and uranyl species have been examined non-destructively by visible microspectroscopy and electron microprobe analysis in order to obtain a better understanding of uranium redistribution and fixation mechanisms. The uranyl species are associated with sub-micron sized iron minerals, such as hematite (Fe2O3), goethite (FeOOH), and ferrihydrite (Fe4-5(OH,O)12). Additionally, some uranium is fixed to goethite in weathered chlorite grains. Sub-micron sized saleeite, Mg(UO2)2(PO4)2. 1OH2O, is the most probable uranyl phase associated with the iron minerals. This suggests that the uranium fixation in the vicinity of the Koongarra uranium ore deposits is the result of the coprecipitation and sorption of microcrystals of saleeite on to the iron minerals released during weathering of chlorite.

Groundwater and entrained particles were sampled from several boreholes at the Koongarra uranium deposit in the Northern Territory of Australia. Conventional filtration techniques, hollow fibre ultrafiltration, and stirred-cell ultrafiltration were used to separate dissolved species, large particles (> 1 μm) and colloids (< 1 μm). The colloids and particles included clay minerals, particularly kaolinite, and chlorite, together with fine quartz grains. Iron was present as particle coatings, and in a separate colloidal form. The amount of 238U associated with colloids (expressed as a percentage of the total 238U which passed through the 1.0 μm filter) ranged up to 6.5%. The corresponding figures for Th were 10-85%. However, the amount of 230Th which passed through the 1.0 μm filter was extremely small, and 230Th was associated to a much greater extent with larger particles, which are unlikely to be mobile in natural groundwaters. In some fine particle and colloidal fractions, the 227Th/230Th activity ratio in the thorium alpha spectrum was unusually high, indicating the presence of substantial quantities of 227Ac. This suggested that actinium could be present as a mobile colloid phase. Overall, there was very little colloidal material in these groundwaters, with only iron, uranium, actinium, and thorium showing a significant association with colloids.

New isotopic data on daughters of transuranium elements and fission products from the Oklo reactor 10 and its surrounding provide evidences of Nd and U migrations. Whereas Nd was released only along carbonated fissures (NdFp as high as 1.048 in the fissures compared to 1.007 in the neighbouring sandstone), U has migrated through the sandstone as well (abnormal negative 235U anomalies as low as 0.714 in both fissures and surrounding sandstone). Nevertheless the released amount of radioelements is very small and visible in the drop of the concentrations between the reactor and the surrounding sandstone [8]. As consequence, a retention effect is evidenced at the reactor-sandstone boundary.

A structural and chemical analogue of UO2 in spent fuel, uraninite, UO2+x, is unstable in the presence of dissolved silica (>10-3.6 mg/L) under reducing conditions and transforms into coffinite, USiO4 nH20. Coffinite may incorporate numerous elements into its structure including actinides and fission products. Conditions favorable for coffinitization of UO2 in spent fuel may occur in repositories in granitic, basaltic, and tuffaceous rocks.