Surface examination of sintered UO2 pellets, leached in groundwater at 90°C, have been conducted and the main results are described. They show the presence of a hydrated layer in which the uranium content decreases with leaching time. The oxidation state of uranium seems to remain (IV). The major components of the layer are magnesium and silicon. The growth of the surface layer in mineral water involves a decrease of uranium release in opposition to its continuing increase i n deionized water.

X-rays photoelectron spectroscopy (XPS) analysis has been carried out on uranium dioxide single crystals placed in both deionised or mineral water at 60 °C and 90 °C for a few months. The surface layer on the sample immersed in mineral water at 90 °C was sufficiently thick to mask the presence of uranium in the recorded XPS spectrum. Depth profiling experiments showed that the deposit was ∼ 100 nm in thickness and revealed the formation of a mixed magnesium-uranium oxide region between the bulk UO2 and the outer magnesium- rich layer. Analysis of the aqueous medium following immersion of the UO2 samples showed that the concentration of uranium released to the mineral water was 10 to 100 times lower than that in deionised water.

Three iron-based to nickel-based austenitic alloys and three copper-based alloys are being considered in the United States of America as candidate materials for the fabrication of high-level radioactive waste containers. The austenitic alloys are Types 304L and 316L stainless steels as well as the high-nickel material Alloy 825. The copper-based alloys are CDA 102 (oxygen-free copper), CDA 613 (Cu-7A1), and CDA 715 (Cu-3ONi). Waste in the forms of spent fuel assemblies from reactors and borosilicate glass will be sent to a proposed repository at Yucca Mountain, Nevada. The decay of radionuclides will result in the generation of substantial heat and in gamma radiation.

In this paper, we describe the preliminary stages of the development of a mathematical model of the evolution of the solution chemistry within a corroding crevice on passive stainless steel. It is based on a formulation by Oldfield and Sutton [1], but models the physical and chemical processes which determine the crevice solution in a more rigorous manner. The model will eventually be used to assess whether a ‘critical solution composition’, which results in the depassivation of the crevice and the onset of localised corrosion, is attainable for a range of repository conditions, steel types and canister designs etc. We also describe experiments that provide input data in the form of passive currents for this model. Preliminary sensitivity tests with the model have indicated a need for accurate thermodynamic data for the chemical equilibria constants (particularly those for the chromium reactions). These tests also suggest that there are certain critical relationships between various parameters in the system (such as crevice dimensions, the composition of solution outside the crevice and the passive current) that mark different behaviour in the evolution of the solution composition. Further experiments will be performed, as part of this work, both to validate the predictions of the model and to determine whether the predicted compositions of the crevice solutions are sufficiently aggressive to initiate crevice corrosion.

In situ corrosion testing and in situ monitoring of the near field chemistry are performed as part of the Belgian experimental program for the qualification of an overpack material for long term HLW containment in deep clay formations. Several test facilities are now in operation and some of the first experimental data are presented and discussed in terms of the program objectives.

Hydrogen evolution from anoxic corrosion of cast steel overpacks in high-level waste repositories is an important issue for design if, as has been estimated, the hydrogen is prevented from escaping by diffusion by a low permeability compacted bentonite backfill.

Evaluation of the corrosion results showed three basic types of corrosion behaviour: general corrosion with oxide layer formation, unstable corrosion behaviour with pitting or macro-element formation and stable passive behaviour.

Cast steel containers under Swiss repository conditions are expected to suffer general corrosion with oxide layer formation. This behaviour gives the highest long term corrosion rates (2–5 μm/a) without local attack, above the 0.03–0.8 μm/a tolerated for hydrogen escape by diffusion but below the 20 μm/a assumed for overpack design.

Copper canisters for nuclear waste containment can be expected to be exposed to temperatures up to 100°C. Since the material is pure copper, creep properties must be taken into account in particular for the welded joints in the canisters. In the paper creep rupture properties of parent metal, weld metal, and simulated heat affected zone are presented for 110°C. About ten times shorter rupture times were found for the weld metal in comparison to the parent metal. Cross weld specimens showed even shorter rupture times.

Copper is being considered as a possible container material for the disposal of nuclear fuel waste in Canada. The mechanism of the uniform corrosion of copper is being investigated in aerated 1 mol·dm-3 NaCl and under simulated disposal conditions. In NaCl solution, the rate of corrosion of a copper rotating disc electrode (RDE) is controlled partly by the rate of transport of copper complex ions away from the electrode surface. Similarly, under simulated disposal conditions, the rate of corrosion may be controlled by the transport of copper through compacted buffer material. However, longer-term results are needed to prove this latter statement.

Previous corrosion studies performed on a number of materials have shown that unalloyed steels are promising materials for long-term resistant packagings to be used in disposal of heat-generating wastes in rock salt formations. This is the reason why those steels are the subject of more detailed investigations. This paper reports an in-situ experiment conducted in the Asse salt mine in which the influence of selected characteristics (welding, shape) of containers on the corrosion behaviour of cast steel was studied. The material was tested in NaCl brine which might intrude into an HLW borehole in an accident scenario. For this, an electron beam welded cast-steel tube was stored for 18 months in a 2-m deep heated borehole and the annular gap between the tube and the borehole wall was filled with saturated NaCl brine. The vertical temperature profile in the borehole was in the range from 90°C to 200°C; the maximum temperature occurred in the center of the heated zone and the minimum temperature in the upper parts of tube.

Under the testing conditions cast steel was subjected to general corrosion at a maximum corrosion rate of 120 μm/a. Considering this magnitude of the corrosion rates, the resulting corrosion allowances are technically acceptable for a packaging having long service-lives. Pitting and crevice corrosion as well as stress-corrosion cracking did not occur in cast steel, and electron beam welding did not exert a noticeable influence on cast-steel corrosion. With these results available, cast steel continues to be considered as a promising HLW packaging material.

Research at Oak Ridge National Laboratory (ORNL) has shown that cement-based waste forms can be tailored to tolerate wide fluctuations in waste-feed compositions and still maintain properties that are compatible with standard equipment and produce a product that meets regulatory requirements. The 20-year operational history of ORNL's grouting program has demonstrated this resilience and reliability on a spectrum of waste feeds encompassing the developmental history of the nuclear fuel cycle.

Cement-based materials are the most frequently used waste forms for solidifying non-high-level waste. These materials are also perhaps the most misused for immobilizing wastes. This paper discusses the scientific basis for selecting a suitable cement-based waste form to solidify and immobilize radioactive waste. For instance, when a given amount of cement (the source of calcium) is replaced with class F fly ash, the strontium leachability is reduced proportionally.

This paper discusses the use of gel clays such as bentonite and attapulgite to absorb excess fluids and thus increase waste loading. Also, cement-compatible natural materials such as nonswelling clays are used to ion exchange or absorb soluble ions such as cesium and thus reduce leachability.

Recently, blast furnace slag has been the subject of much interest in cement-based immobilization technology. Data are presented to show that a waste form containing blast furnace slag and other cementing materials is very effective in reducing the release rates of 99Tc.

The confinement ability of a waste package is one of the major safety characteristics to consider in shallow land burial. In order to determine if the confinement is acceptable, in accordance with local policy, one way is to proceed to leaching tests. The practical method, for sake of simplicity, cost and time limit, is to carry out the leaching tests on laboratory samples which are easier to prepare than full-scale blocks, but the representativity of which needs to be treated with caution; it is in this context, that one of the aspect of our work concerns what is known as the “scale effect”.

This study has been conducted using blocks the volumes of which are respectively of 200, 20, 2 and 0, 2 1, and made with the same cement-waste form (13 Cs) system.

The pH of a cementitious radioactive waste repository is expected to decrease with time as water leaches out soluble ions or as a result of reactions between repository constituents. These processes have been simulated in the laboratory and the results compared with behaviour anticipated from the known chemistry of cement compounds.

The studies indicate that the pH in Portland cements and Blast Furnace Slag cements (BFS) is likely to remain above 10.5 for very long periods of time, of the order of several hundred thousand years for a typical repository. They also show that extensive use of Pulverized Fuel Ash (PFA) leads to uncertainty in long term pH which could be as low as 9 to 10 in some cases.

The current UK concept for a low- or intermediate-level nuclear waste repository includes a largely cementitious backfill. The cement provides a high pH environment in which the general corrosion rate of the metal canisters is reduced and the solubilities of many nuclides low. It has previously been assumed that this high pH will exist for a period of 107 years, however cement will degrade due to leaching of the solid components and attack from aqueous species in groundwater. In this paper we describe the preliminary stages of a model of the degradation of cement in a repository. The modelling involves the incorporation of a thermodynamic description of cement[2] into the static code PHREEQE[5]. This is then used in a coupled chemistry-transport model of simple leaching of cement using the code CHEQMATE[4]. This preliminary modelling also provides a useful verification of CHEQMATE as a direct comparison with a THCCDM (a coupled code based on CHEMTRN) model is possible. Results from this preliminary model suggest that the fall in pH due to leaching is slow. The model is sufficiently flexible to form the basis of more detailed investigations of the effect of groundwater interactions on the degradation of cement.

Considerable water uptake, formation of free liquid and leaching of activity may take place if cemented waste materials containing soluble salts are exposed to high humidity air. Another feature of importance for the structural stability of some materials is cyclic variations between water saturation and dry-out periods. The phenomena have been investigated on a laboratory scale using simulated waste in the form of cemented sodium nitrate or ion-exchange resins. Spherical samples are found to be valuable for simultaneous determination of the leaching of major components, mainly sodium, and the water uptake and swelling obtained from weight-increase measurements for the samples weighed in air and immersed in water.

Hydrogen production by radiolysis of water in cement waste forms has to be limited due to safety requirements. Specific generation rates (g-values) are derived from data obtained in laboratory scale experiments and measurements on 200 1 drums containing real waste. To interpret the data model calculations are carried out including those for PH, dose rate, nitrate concentration and diffusion. An OH depleting reaction is added to the radiolytical reaction mechanism of water to model the hydrogen production rate observed for γ-irradiated cement.

A ‘thin-plate’ technique, which is well suited for laboratory studies on cements, has been used in equilibrium experiments aimed at studying material transport between concretes of different composition. The extent of material transport is small, but there appear to be a movement of calcium towards thin plates containing excess silica fume, where supposedly the pozzolanic reaction takes place, an aspect which can be of interest for crack-healing of cements. A model concerned with crack healing of cement is discussed. Preliminary data have been obtained in a study where thin plates have been employed to investigate the effect of leaching on the microstructure of cements using the technique of Small Angle Neutron Scattering (SANS).

Blast furnace slag (BPS) cement blends have certain advantages for the encapsulation of low and intermediate radioactive wastes. However, their hydration reactions are more complex than for Portland cement because Portland cements fully react within several years whereas slag hydration takes considerably longer. Empirical testing of blends matured for short periods, up to several years, are not therefore adequate to characterise the chemical immobilisation potential of such systems. A modelling approach is required to predict long—term properties.

This paper is concerned with defining the internal environment within slag-cement matrices. The important parameters for definition are pH, Eh, solubilities and speciation in the aqueous phase, and solid phase equilibria. Compatibility studies in the CaO-CaSO4-Al2O-SiO2-MgO-H2O system are described and compared with the results of X-ray diffraction on pastes aged up to 2 years. In addition, pore water compositions for slag-rich cements are given, and the potential for predicting their redox level(Eh) by extending the present model is discussed.

The leaching behaviour of Cl-36, Cs-137, Sr-90 and Tc-99 from cement-fixed intermediate level liquid waste was investigated. Bentonite, MicroSili-ca®, acrylic resin and sodium sulfide were used as additives to a blast furnace cement matrix. In some cases portland cement was used. The liquid (water or waste solution) to cement ratio applied was mostly 0.47. Samples were leached with water or quinary brine (Q-brine).

After 250 days the sequence of leachability from additive-free specimen in Q-brine was Cl>Cs>Sr>Tc with little modification in water: Cs>Cl>Sr>Tc. Cl and Cs leaching are diffusion-controlled, Sr and Tc are chemically fixed.

Bentonite improves the retardation of Cs significantly, but has no significant effect on Cl, Sr and Tc. MicroSilica® deteriorates the retardation of Cs, but has no effect on Cl, Sr and Tc release.

Sulfide was expected to reduce the leachability of Tc by forming insoluble TC2S7. No significant effect was observed, however, because either TC2S7 does not form at the pH value of the cement slurry or, due to a solubility competition, insoluble Tc species also form in the absence of sulfide ions.

Acrylic resin does not reduce the leachability of Cs significantly. Irradiation with doses up to 4.4–105 Gy slightly deteriorates Cs retardation.

Applying a liquid to cement ratio of 0.35 instead of 0.47 improves the retardation of Cs by a factor of 1.5–4, which may be due to a reduction of porosity in the specimen.

Cement samples containing variable amounts of Cr(III) or Ni(II) or Cd(II) as chloride admixtures present in a MLW stream from reprocessing of spent fuel were prepared under established procedure. Due to the high pH generated in the cement clinker hydration, respective metal hydroxides are formed and improve the physical characteristics of the cement matrix. An increase in the value of the compressive strength was observed when about 7% of the above elements were blended in cement. The increase depends on the nature of the element in the sequence Cr >Ni >Cd.

The leach tests confirm the formation of insoluble hydroxide presenting a low concentration in the leachate corresponding to 0.05, 0.004 and 0.052 mg.L−1 for Cr, Ni and Cd, respectively. On the other hand, the hydro-lyzed chloride ions seem to react with calcium with consequent increased degradation of the cement matrix.

The leaching rate of cesium has been studied in cement. The solidification matrix was a standard Portland cement with increased sulphate resisting properties mixed with 300 kg/m3 or 350 kg/m3 powdered ion exchange resins, and with or without 5 weight-% zeolite added to the grout. The leaching rates from an ordinary cement matrix and a cement matrix mixed with zeolite were measured to be (1.3 to 2.6)×10–7 kg/m2s and (1.6 to 2.1)×10−8 kg/m2s, respectively. From the leaching data the apparent diffusivity of cesium in an ordinary cement matrix with a waste load of 300–350 kg/m3 powdered ion exchange resins was measured to be (0.6 to 1.3)×10−10 m2/s, and in a cement matrix mixed with 5 weight-% zeolite and the same waste load, the diffusivity was measured to be (0.8 to 1.0)×10−12m2/s.