Synroc C waste form specimens prepared using the Australian-developed technology are uniaxially pressed in stainless steel bellows at 1200°C and 20MPa. This produces a material with high chemical and physical durability and with the radioactivity enclosed inside both the waste form and the bellows. An alternative method of producing the ceramic product is to use cold pressing of pellets followed by reactive sintering to provide densification and mineralization. Depending on the scale of waste form preparation required and on the activity level and nature of the waste streams, the cold press and sinter method may have advantages. To evaluate the effects of production method on waste form characteristics, especially resistance to dissolution or leaching of waste elements, we have prepared two simulated waste samples for evaluation. Both samples were prepared from liquid precursor materials (alkoxides, nitrates, and colloidal silica) and then doped with waste elements. The precursor material in each case corresponded to a basic phase assemblage of 60% zirconolite, 15% nepheline, 10% spinel, 10% perovskite, and 5% rutile. One sample was doped with 25% by weight of U; the other with 10% by weight each of U and Gd. Each sample was calcined at 750°C for 1 hr. in a 3.5% H2 in N2 atmosphere. Then one portion of each sample was hot pressed at temperatures ranging from 1120 to 1250°C and 20MPa pressure in steel bellows. A separate portion of each sample was formed into pellets, cold pressed, and sintered in various atmospheres at 1200°C to produce final products about 2/3 cm in diameter. Samples were then examined to determine density of the product, grain sizes of the phases, phase assemblage, and the location of the U and Gd in the final phases. Density data indicate that sintering gives good results provided that the samples are held at 200°C for long enough to allow trapped gases to escape.

Glass-ceramic materials based on sphene, pyroxenes, apatite as well as various SYNROC formulations have been produced through melting in a laboratory electrically-heated furnace and inductive-heated melter (“cold crucible”). Materials prepared were investigated by X-ray diffraction analysis, electron-probe microanalysis, infra-red and EPR spectroscopies. Basic properties such as leach rates of radionuclides, density and compressive strength have been determined. Melted SYNROC-B and C phase compositions produced by induction melting are similar to hot-pressed SYNROC.

Leaching experiments with deionised water at 70°C have been carried out to ascertain the effect of oxidising and anaerobic conditions and the presence of Boom clay on Np release from Synroc.

The normalised solution leach rates of Np measured in these experiments are about a factor of 10 lower under reducing conditions than oxidising conditions and the presence of Boom clay was found not to enhance the leaching of Np from Synroc. Neptunium was found to be predominantly in the mobile fraction and was not removed from the leaching solutions significantly by filtration or ultrafiltration.

Perovskite and zirconolite are two of the major phases of the Synroc titanate mineral assemblage. Their aqueous durability under a range of pH conditions at 90°C has been examined. Solution analysis, electron microscopy and X-ray diffraction have been used to investigate the dissolution behaviour of these phases, and a perovskite phase doped with Nd, Sr and Al, using buffered solutions at pH levels of 2.1, 3.7, 6.1, 7.9 and 12.9. After 43 days of leaching, Ca and Ti extractions from perovskite and zirconolite show only a weak pH-dependence.

SEM investigation of the samples leached at pH 2.1, 6.1 and 12.9 showed that a titanaceous surface layer formed on the perovskite specimens. XRD analysis of the perovskite samples showed that anatase formed on the leached surface at acidic and neutral pHs, but not under alkaline conditions, and that minor amounts of rutile also formed. In the leached perovskite specimens doped with Nd, Sr and Al, no rutile was found by XRD and anatase was only detected in the sample leached at pH 2.1. There were no detectable changes in the leached zirconolite samples examined by SEM and XRD.

Electron microscopy techniques were used to identify secondary phases on the surface of Synroc after exposure to doubly deionized water (DDW) at 150 ˚C for time periods up to 532 days. After one day, Ti02 develops on perovskite in the form of polycrystalline crusts of anatase or epitaxial crystals of brookite (± anatase), and as scattered single crystals on hollandite and rutile-Magnéli phases. With increasing exposure time the anatase crusts exhibit slow grain growth and development of preferred orientation; whereas, the epitaxial brookite crystals appear to grow more rapidly. Dissolution of perovskite and P-bearing intermetallic particles results indirectly in the precipitation of monazite after 84 days. Additional alteration products observed over a range of time include crystalline Al-O-H phases and amorphous to poorly crystalline Fe-O-H phases which form as a result of the dissolution of Al-rich oxides and Fe-bearing intermetallic grains, respectively. A poorly crystalline Ru-O-H phase was found as an alteration product of Ru-bearing intermetallic particles after exposure to DDW for 336 days.

The surface alteration of strontium aluminum silicate has been studied under hydrothermal conditions in connection with the leachability of Sr from solidified radioactive wastes. Samples exposed to a 3 M NaCl solution at 300°C produced a precipitated layer of analcime (NaAISi2O6) with high crystallinity. The normalized leach rates for Sr in 3 M NaCl and 3 M KC1 solutions were found to be less than 10-3 g/cm2.d, yielding high resistance to leaching in these solutions.

On exposure to the CaCl2 solution at 300°C, anorthite (CaAl2Si2O8) was formed on the surface, indicating the replacement of Sr with Ca through isomorphous substitution. In the case of MgCl2 solution, a thick precipitated layer of clinochlore (Mg5Al(Si3Al)O10(OH)8) built up on the outermost surface. This layer serves to depress the migration of Sr, resulting in the deposition of SrCl2 salts in the altered layer.

A quantitative powder X-ray diffraction study was made of actinide doped perovskite of bulk composition Ca0.98919An0.01081Ti0.98919Al0.01081O3 where An corresponded to approximately equimolar proportions of Cm-244 and Pu-240. Sections of this sample accumulated irradiation doses up to 7.51 × 1017 alpha decay events per gram (± g−1). The damaged samples were treated in two ways. First, to establish the critical temperature for structural recovery under the reducing conditions of geological repositories, isochronal annealing was carried out at 600, 800, 1000 and 1100˚C for 12 hours in graphite crucibles. Two groups of perovskites that had previously sustained doses of 4.5 × 1017 and 7.4 × 1017 ± g−1 were tested in this way. For the former group, these conditions resulted in up to 9 weight percent (wt%) of available actinide separating as a fluorite-type dioxide near the perovskite surface. In the latter group, calcium was reduced to the metal which vaporised, leaving an excess of refractory titanium that crystalised as rutile. Second, material which had sustained doses of 1.6–4.0 × 1017 alpha decays per gram was subjected to an MCC-1 leach test for two months at 90˚C using a pH ∼ 2 solution. Under these conditions surficial perovskite dissolved congruently to release calcium into solution while the titanium reprecipitated as anatase. The implications of these results for the ultimate disposal of perovskite-bearing polyphase nuclear waste ceramics are considered.

This paper describes an integrated approach to developing a predictive computer model for long-term performance of concrete engineered barriers utilized in LLRW and ILRW disposal facilities. The model development concept consists of three major modeling schemes: hydration modeling of the binder phase, pore solution speciation, and transport modeling in the concrete barrier and service environment. Although still in its inception, the model development approach demonstrated that the chemical and physical properties of complex cementitious materials and their interactions with service environments can be described quantitatively.

Applying the integrated model development approach to modeling alkali (Na and K) leaching from a concrete pad barrier in an above-grade tumulus disposal unit, it is predicted that, in a near-surface land disposal facility where water infiltration through the facility is normally minimal, the alkalis control the pore solution pH of the concrete barriers for much longer than most previous concrete barrier degradation studies assumed. The results also imply that a highly alkaline condition created by the alkali leaching will result in alteration of the soil mineralogy in the vicinity of the disposal facility.

It is currently proposed to site the UK repository for intermediate-level radioactive waste (ILW) underground at a depth of several hundred metres, and to use a cementitious backfill to fill spaces between disposal packages within the repository vaults. Flow of water through the repository could be concentrated within cracks that may form in the backfill. This paper describes the CRACK and CRACK2 computer programs. These are used to predict radionuclide release via cracks, and radionuclide distributions within the vault. The programs are applied to an assessment of the release of inventory- and solubility-limited radionuclides from hypothetical ILW vaults. Calculations are reported for different densities of cracking, and for different repository designs, and the dominant modes of transport within cracks in repository vaults are discussed.

Accelerated aging tests have been conducted to evaluate the long-term durability of cementitious material against aggressive ions. In tests, cementitious specimens were immersed in the solutions containing concentrated aggressive ions at high temperature and it promoted diffusion of the ions in the specimen. This method would be suitable for the evaluation on the aging as the alteration of the specimen would be expected to resemble the natural behavior. This paper presents a classification of alteration mechanism in the immersion tests using MgCl2 and Na2SO4 solution. This classification was done by relating the changes of compressive strength to microscopic and mineralogical changes.

Mortar was used as a matrix material in an immobilization process of radioactive waste material, 137CsCl solution, where the final product had a water-to-cement ratio of 0.36. Solidified orthocylinder shaped samples, H=4.5 cm, with the exposed surface area completely open to contact leachant medium, were used in prospecting physico-chemical and mechanical characteristics of mortar-radwaste mixture forms. After curing period of 28 days in an atmosphere of defined parameters, samples were introduced to leaching in the specific leachant media. Leachants were chosen up to their aggressive influence on matrix material deterioration and decomposition: distilled water, chloride and sulphate solutions. Experiments were performed for 730 days and the 137Cs cumulative leached fraction for each leachant was calculated. Leach-rate steady-state values were obtained after approximately 290 days. The same trend in obtaining saturation value was noticed in an investigations of pH value and Ca2+ concentration in the three groups of leachants, due to a Ca2+ depletion from the matrix structure. Mortar-radwaste mixture samples were withdrawn from the leaching experiments and investigated on their mechanical strength. Mechanical strength characteristics adopted a decreasing trend for the samples treated in leachants longer than 100 days, which continued in the two years period of experimental duration.

Radioactive waste repositories are constructed underground below ground water level. Main engineered barrier of the repository consists of structural walls, backfill mortar, and consolidated wastes. Cementitious materials are mainly adopted for the engineered barrier. Gas (hydrogen or methane) generation due to corrosion and microbial degradation is predicted, since the waste consists of metals like steel and organic material. It is important issue to define gas transport mechanism in cementitious materials in order to assess the gas release influence on the engineered barrier and pressure rise in the repository. Although cementitious materials have enough permeability to transport gas in dry condition, it is expected to be impermeable or so in saturated condition, in which repository is placed. In this paper, water and gas permeability both in dry and saturated conditions are investigated. The pressure in the vault, and gas transport mechanism is predicted. Certain threshold pressure exists, and it can be recognized that gas permeability is dependent on water saturation degree at a pressure greater than the threshold pressure. Loading pressure of 0.7 MN / m2 is needed for the flow of gas in saturated concrete. Intrinsic gas permeability of 6.5 × 10−19 m2 is obtained, which is very small value to keep the low pressure inside the vault. Consequently, it is predicted that gas cannot be transported through the intact concrete, however, it can be transported through defective zone which is developed during construction, and so on.

Hydrogen gas is generated from cementitious waste forms by radiolysis of water. In the case of low level radioactive waste, gas yields have been confirmed to be sufficiently low by irradiation experiments. However, studies have suggested that the hydrogen generation rate in cementitious waste forms is larger than the rate calculated from the g-value (H2 yields for 100eV absorbed). In this paper, the factors that increase the gas generation were investigated quantitatively. Two factors were identified, the effect of an organic diethylene glycol which reacts with hydrogen radicals to produce hydrogen, and the effect of electrons generated in the cementitious matrix which decompose water to hydrogen. The hydrogen generation rate was confirmed to drop less than the rate calculated from the g-value when these factors were eliminated.

Within the framework of the Belarus National Programme, IPEP is performing research on the immobilization of low level radioactive waste after decontamination and remediation of Chernobyl Zone. The goal of the R&D programme is to develop a composite matrix for embedding these wastes using special encapsulation technology and reinforced and hydro-resisting polymer-concrete layer on solidified waste blocks to minimise the amount of cement for conditioning of waste.

The use of mechanistic (surface-complexation, electric-double layer) and semi-empirical (affinity spectrum) models for representation of the association of metal ions with heterogeneous environmental materials, such as humic acids and soil particle surfaces, is compared. It is seen that mechanistic models are not nearly as mechanistic as one generally assumes, and that semi-empirical models are much more valuable than one might assume by comparison to simple Kd, models. A semi-empirical discrete-log-K-spectrum model was used to describe the binding of Co(II), as a function of pH and NaClO4 concentration, to two environmental substrates: leonardite humic acid and a kaolinitic subsoil. Excellent agreement of the model and the data was obtained over a wide range of solution composition. These models appear to be the most promising among several alternatives for modeling interactions of metal ions with complex heterogeneous environmental materials over a wide range of solution composition.

Small polished blocks of granite, diorite and dolerite were immersed in solutions containing uranium and plutonium at equal initial concentration. The samples were analysed by the advanced surface analytical techniques of secondary ion mass spectrometry (SIMS) and nuclear microprobe analysis. The results show that both actinides sorb onto the same minerals in the three rocks. However, SIMS data show that significantly more uranium was sorbed than plutonium.

An integrated programme of batch sorption experiments and mathematical modelling has been carried out to study the sorption of plutonium and americium on a series of repository, backfill and geological materials relevant to the JNFL low-level radioactive waste repository at Rokkasho-Mura.

The sorption of plutonium and americium on samples of concrete, mortar, sand/bentonite, tuff, sandstone and cover soil has been investigated. In addition, specimens of bitumen, cation and anion exchange resins, and polyester were chemically degraded. The resulting degradation product solutions, alongside solutions of humic and iso-saccharinic acids were used to study the effects on plutonium sorption onto concrete, sand/bentonite and sandstone.

The sorption behaviour of plutonium and americium has been modelled using the geochemical speciation program HARPHRQ in conjunction with the HATCHES database.

Sorption behaviors of neptunium (V) on naturally-occurring magnetite (Fe3O4) and goethite (α-FeOOH) in 0.1M NaN03 electrolyte solution under aerobic conditions were interpreted using the surface complexation model (SCM). The surface properties of these materials were experimentally investigated by C02-free potentiometric titration, and SCM parameters for the constant capacitance model, such as protonation/deprotonation constants of the surface hydroxyl group, were determined. The number of negatively charged sorption sites of goethite rapidly increased with the increase of the bulk solution pH compared with that of magnetite and this tendency was similar to the pH dependence of neptunium sorption. This implies that the neptunyl cation, NpO2+, plays a dominant role in possible sorption reactions. Assuming that the dominant surface complex is XO-NpO2, modeling by means of SCM was carried out, and the results were found to agree with experimental data.

The behaviour of Tc in soil depends upon its chemical forms. Under aerobic conditions, Tc is present as pertechnetate Tc+7 (TcO4–), which has a high geochemical mobility and bioavailability. However, the form changes through a combination of factors such as redox conditions and microbial activity in soils. The anaerobic condition can be supplied in waterlogged rice paddy soil, which is common to Japan and other Southeast Asian countries.

In this study, we focused on the influence of microbial activity on Tc behaviour in soil under the waterlogged condition. The microbial activity is expected to be controlled by addition of glucose and the activity causes a change in the soil redox condition. Air-dried soil with increasing glucose contents and sterile soil were used to compare the effect of microbial activity. The soils were waterlogged with 99Tc solution. Although the concentrations of 99Tc in the surface and bottom solutions of air-dried soils decreased over time with increasing glucose concentration, those of sterile soils were almost constant during the experiment. From these results, it was assumed that the activity of microorganisms influenced Tc adsorption onto the soil.

Biotite has the most important role for the sorption of radionuclides in granitic rocks. Experiments on the sorption of uranium (VI) on biotite were conducted to understand the fundamental controls on uranium sorption on biotite mineral, including the effects of pH and uranium concentration in solution. Biotite powder (mesh 32 – 60) were washed with 1N HC1 for a week and were rinsed twice with deionized water for a week. This HC1 treatment was necessary to avoid the effects by other minerals. The agreement between surface adsorption coefficient, Ka, of both biotites with and without HC1 treatment was within one order of magnitude. The peak Ka value was in the range of 0.1 to 0.01 cm3/cm2 around pH 6. A comparison of aqueous uranium speciations and sorption results indicates that neutral uranyl hydroxide could be an important species sorbed on the biotite.

Sequential desorption experiments with KC1 and HC1 solutions were also carried out after sorption experiments to investigate sorption forms of uranium. Approximately 20 % of uranium in solution were sorbed on the biotite as an exchangeable ion. The fraction of exchangeable uranium had a little dependence on pH. The other uranium could not be extracted even by 6N HC1 solution. It is possible that most of the uranium could be precipitated as U(IV) via Fe(II) reduction on the biotite surface.