The active range of Desulfovibrio desulfuricans. a species of sulfate-reducing bacteria, was examined in terms of pH and Eh using a fermenter at controlled pH and Eh. Such research is important because sulfate-reducing bacteria (SRB) are thought to exist underground at depths equal to those of supposed repositories for high-level radioactive wastes and to be capable of inducing corrosion of the metals used in containment vessels.

SRB activity was estimated at 35°C, with lactate as an electron donor, at a pH range from 7 to 11 and Eh range from 0 to -380 mV. Activity increased as pH approached neutral and Eh declined. The upper pH limit for activity was between 9.9 and 10.3, at Eh of -360 to -384 mV. The upper Eh limit for activity was between -68 and -3 mV, at pH 7.1. These results show that SRB can be made active at higher pH by decreasing Eh, and that the higher pH levels of 8 to 10 produced by use of the buffer material bentonite does not suppress SRB completely.

A chart was obtained showing the active range of Desulfovibrio desulfuricans in terms of pH and Eh. Such charts can be used to estimate the viability of SRB and other microorganisms when the environmental conditions of a repository are specified.

The Maqarin site, Jordan is being studied as a natural analogue of a cementitious radioactive waste repository. The microbiology has been studied and diverse microbial populations capable of tolerating alkaline pH were detected at all sampling localities. Dissolved organic carbon was identified as the potentially most important reductant with sulphate identified as the main oxidant, both supplying energy for microbial life. Calculations on upper limits of microbial numbers were made with a microbiology code (MGSE) using existing information but the results are overestimates when compared with field observations. This indicates that the model is very conservative and that more information on, for example, carbon sources is required.

Apparent stability constants of Eu(III)- and Am(III)-humates determined at various pH (4.8 to 8) and supporting electrolyte concentration (0.02 to 1 mol/1), indicate that the humate complexes may be dominant species in the aquifer. Distribution coefficients of Eu(III) and Am(III) between kaolinite and solution phase were affected strongly by the presence of humic acid in the solution. Adsorption of humic acid may be an important factor controlling fixation of actinides(III) on mineral surfaces.

We have studied the interaction of a low molecular weight alginic acid (ca. 7,000 daltons) with Ca2+, Mg2+, Eu3+ and La3+, using lanthanide luminescence spectroscopy (LLS) and 13C NMR. LLS results imply that nine-coordinate lanthanide ions lose six water molecules upon complexation with alginate, forming inner-sphere alginate complexes. In general, the 13C resonances in the NMR studies decreased in intensity as metal ion concentration increased, but the preference of a cation for repeating mannuronate and guluronate units (MM and GG blocks) varied greatly. The La3+ ion binds preferentially to the GG blocks, though some binding to the MM blocks is evident, while Eu3+ binds equally well to GG and MM blocks. Ca2+ shows a strong preference for the GG blocks only, while Mg2+ shows no evidence of binding to alginate at all. These preferences may be explained in terms of cation size, charge and coordination number. The diequatorial linkage pattern of sequential mannuronate residues leads to a flat ribbon-like structure with shallow cavities for the metal ions while the diaxial linkages of the guluronates allow a deeper, more size-specific cavity. Calcium has the optimum ionic radius for the GG cavities compared to the larger Eu3+ and La3+ ions. The smaller Mg2+ ion does not bind at all over the concentration range studied. This could be due to its large charge density to coordination number ratio that would make dehydration of this ion and subsequent complexation by alginate thermodynamically unfavorable.

Natural organic compounds can play an important role in the transport of radionuclides through the geosphere, but the inclusion of polyelectrolytic fulvic and humic acids into geochemical speciation codes is not straightforward. Schubert ion-exchange experiments have been used to parameterise an equilibrium-based, discrete site-electrostatic model, Model V, which is concerned with predicting trace metal-humic interactions in solution. However, before such a model can be applied to natural groundwaters, its ability to deal with competition between trace cations and anionic ligands has to be tested.

PHREEQEV, which incorporates Model V with the inorganic speciation code PHREEQE, has been used to guide experiments based on a modified Hummel-Dreyer method. To improve speed and precision, these experiments used a HPLC method rather than Sephadex gel, and measured the effect of Ca on Co and Ni binding to fulvic acid. The results were compared with the predictions made by PHREEQEV. Although the system was further complicated by the need to add citric acid to prevent binding of metals to the HPLC column, predictions at pH 5 were good for both metals. At pH 7, the Co binding results were not well predicted and the reasons for this are still unclear. Comparisons between the Schubert and Hummel-Dreyer-type experiments show that the two methods are compatible.

Drink is a 2D research code, which simulates the long term evolution of shallow, trenched, LLW disposal sites with significant groundwater flow. It employs a finite difference solver and is built around a geochemical transport model into which various functional units are interfaced. These units describe sorption, corrosion, microbiology, radionuclide decay, colloids, mineral precipitation and dissolution and gaseous release. Each of these units is described here, along with a 1 D simulation of uranium migration.

The potential importance of backfilling and plugging in underground radioactive waste repositories has led differents research institutions around the world (SKB in Sweden, CEA in France, AECL in Canada, etc.) to carry out extensive studies of swelling clay materials for the development of engineered barriers. These materials, which have to be emplaced in underground conditions, should combine a variety of complementary properties from both the hydro-thermo-mechanical and geochemical viewpoints: impermeability, swelling ability in order to fill all void space, heat transfer and retention capacity for the most noxious radionuclides. For years, the scientific community has acknowledged the fact that smectite clays best exhibit these properties and, thus, most of the research effort has been devoted to this type of materials. The aim of such studies is to try and link the microscopic characteristics of the material (mineralogy, geochemical properties, microstructure, etc.) to its macroscopic behaviour (swelling properties, etc…).

Effect of smectite illitization on mass release rate from the bentonite-filled buffer of the engineered barriers is analyzed based on experimental results of sorption of americium and neptunium onto illite-smectite mixtures, and on mathematical models for smectite illitization by K+ diffusion, water flow through the engineered barriers, waste glass dissolution, and transport of americium and neptunium including solubility sharing with two major americium isotopes and moving boundary effect of the neptunium precipitation region. Numerical results for mass release rates of americium and neptunium as well as time-dependent pore velocity and glass dissolution rate are shown. Illitization effect is negligible with parameter values assumed here.

A series of hydrothermal experiments was performed to determine the effect of layer charge of starting materials on the smectite to illite reaction rate that might be applied to nuclear-waste repository design. The experiments were conducted on K-saturated <2μm fractions of Wyoming smectite (SWy-1) and Tsukinuno smectite (SKu-F, commercially, Kunipia-F) in a closed system at temperatures of 95, 150, 200, 250, 300°C for run durations of up to 477 days with a 1:20 mass ratio of solid to deionized water. The mean layer charge and tetrahedral charge of SKu-F are larger than those of SWy-1. The proportion of smectite layers in illite/smectite interstratified minerals rapidly decreases, and then slowly decreases with increase in reaction time; a plot of In (100/% smectite) vs. time produces two distinct straight lines in all experiments. These lines are suggestive of two first-order kinetic processes with different rates for this reaction; the first process has a greater rate than the second one. An Arrhenius plot of the reaction rates for each process produces a folding and straight lines for the first and second processes, respectively, suggesting that there are at least two parallel processes in the first process, and a dominant process is different between high- and low-temperature reactions. The activation energies of the first and second processes determined from the plots are the same for the two starting materials, meaning that the reaction mechanisms for the two starting materials are the same. However, the rate of the first process is different between the two starting materials, although that of the second process is similar. The difference in the rate of the first process results possibly from the difference in the amount of layer charge between the two starting smectites.

Bentonite/sand mixture is expected to be an effective component of barriers for radioactive waste disposal facility. Impermeable mixture can be obtained through enrichment of the bentonite content,but it is difficult to estimate a permeability of bentonite/sand mixture before permeability test. At designing of bentonite/sand mixture, we have to decide which sand to be used, and what bentonite content is. But.this means that designing will be a time-consuming task. In this report, we made many kinds of bentonite/sand mixtures using some kinds of sand with different bentonite content, and permeability tests were conducted for these mixtures. After examination of the permeability tests results, the effective void ratio of bentonite/sand mixture which was obtained by compaction test, is an efficient parameter for estimation of the permeability. Last of all, we propose the method of designing bentonite/sand mixture using the effective void ratio.

A chemical equilibrium model has been developed for ion-exchange and to a limited extent for other reactions, such as precipitation or dissolution of calcite or gypsum, in compacted bentonite water systems. The model was successfully applied to some bentonite experiments, especially as far as monovalent ions were concerned. The fitted log-binding constants for the exchange of sodium for potassium, magnesium, and calcium were 0.27, 1.50, and 2.10, respectively. In addition, a coupled chemical and diffusion model has been developed to take account of diffusion in pore water, surface diffusion and ion-exchange. The model was applied to the same experiments as the chemical equilibrium model, and its validation was found pardy successful. The above values for binding constants were used also in the coupled model. The apparent (both for anions and cations) and surface diffusion (only for cations) constants yielding the best agreement between calculated and experimental data were 3.0×10-11 m2/s and 6.0×10-12m2/s, respectively. These values are questionable, however, as experimental results good enough for fitting are currently not available.

The objective of this study is to investigate the permeability variation of the compacted bento-nite-sand mixture, a candidate material to be used in low level radioactive nuclear waste geotechnical disposal facility, due to shear deformation which might occur in the mixture caused by the earthquake and/or the nonuniform deformation in its long term performance. To execute the study systematically, a coupled shear and permeability testing apparatus was firstly developed in which a hollow cylindric specimen and the constant flow pump was adopted. It enabled us to measure the permeability along the shear plane in the specimen successfully and rapidly. The result of this study revealed that the permeability of the compacted bentonite-sand mixture with 15% of bentonite was not significantly influenced by the shear deformation. This has tentatively shown the effectiveness of both the new coupled shear and permeability testing apparatus and the bentonite-sand mixture to be used for retarding the radionuclide migration around the radioactive waste repository accompanying water movement.

Apparent diffusion coefficients for Se and Zr in bentonite were measured by in-diffusion method at room temperature using water-saturated sodium-bentonite. KunigelVl® * containing 50wt% Na-smectite as a major mineral was used as the bentonite material. The experiments were carried out in the dry density range of 400–1800kg/m3. Bentonite samples were immersed with distilled water and saturated before the experiments. The experiments for Se were carried out under N2 atmospheric condition (O2: 2.5ppm). Those for Zr were carried out under aerobic condition. The apparent diffusion coefficients decrease with increasing density of the bentonite. Since dominant species of Se in the pore water is predicted SeO32-, Se may be retarded by anion-exclusion because of negative charge on the surface of the bentonite and little sorption. The dominant species of Zr in the porewater is predicted Zr(OH)5- or HZrO3-. Distribution coefficient measured for Zr on the bentonite was about 1.0m3/kg from batch experiment. Therefore, the retardation may be caused by combination of the sorption and the anion-exclusion. A modelling for the diffusion mechanisms in the bentonite were discussed based on an electric double layer theory. Comparison between the apparent diffusion coefficients predicted by the model and the measured ones shows a good agreement.

Effective diffusivities of radioactive nuclides in compacted bentonite were calculated theoretically by using an electric double layer theory. Comparison between calculated diffusivities and measured ones show good agreements.

The effective diffusivity is dominated by pore structure and pore diffusivity Dp. The pore structure can be characterized by effective porosity ε eff, constrictivity δ, and tortuosity Γ. The δ was assumed to be unity. The ε eff and the Γ were determined experimentally. The Dp was estimated by means of the electric double layer theory. In the estimation, smectite interlayer was assumed the space between parallel plane sheets of smectite crystal lattice.

Diffusion experiments were carried out by using Cs+ for monovalent cation, C1- and Tc04- for monovalent anion, and tritiated water for neutral molecule. The measured and calculated effective diffusivities in different densities showed the same tendency of cation > neutral > anion. The dry density of bentonite became higher, the discrepancy between the estimated and the measured diffusivities became larger. The calculation was limited by the applicability of the electric double layer theory in the near surface region of smectite.

Montmorillonite purified from bentonite was homoionized by leaching with a chloride solution of sodium, potassium, cobalt or strontium, and then leaching with ethanol until the leachate contained only a negligible amount of the cation.

The electrical conductances of thus obtained Na-, K-, Co- and Sr- homoionized montmorillonite were measured. Applying the Nernst- Einstein equation to the electrical conductance, the diffusion coefficient for cations attached to the solid phase (montmorillonite) Ds, and that for cations in the bulk water phase Dw, were evaluated independently. As predicted, Ds was almost independent of the solid/liquid ratio, while Dw increased rapidly with the decrease of the ratio towards the diffusion coefficient in free water.

Diffusion experiments of radionuclides in compacted sodium bentonite with a dry density of 1.0 g/cm3 were performed in nitrogen gas atmosphere at 90 °C for 208 d and 375 d. The corrosion experiments of crushed radioactive glass, JSS-A, carried out simultaneously to provide the source of the radionuclides for the diffusion experiments. The normalized elemental mass losses of cesium isotopes and 238Pu were lower than those of boron (ca. 10 g/m2) probably because of the difference of sorption and/or precipitation. The apparent diffusion coefficients of 238Pu, 234U and 125Sb were determined to be 2x 10-14 m2/s, 5x 10-12 m2/s and 2x 10-12 m2/s, respectively. The distribution coefficient of Pu estimated from the diffusion data was of the same order as that from batch sorption experiments. The glass corrosion and the plutonium diffusion were described by the geochemical codes PHREEQE, STRAG4 and GESPER. The calculation results well fitted the observed data.

The results of experimental studies performed to determine the radionuclide diffusion coefficients in a compacted clay and the hydraulic conductivities of clay/crushed granite mixtures with various clay contents are presented. Clay used in the experiments is a natural clay from the southeastern part of Korea, and it contains mainly calcium bentonite. The hydraulic conductivities of clay/crushed granite mixtures decreased with increasing clay content. In case of clay content of 50 wt.%, they maintain the considerably lower values even at the dry density of 1.5 Mg/m3. The diffusion coefficients for 90Sr, 137Cs, 60Co and 125I in water saturated clay at a dry density of 1.4 Mg/m3 were measured at room temperature. The average apparent diffusion coefficients obtained are 4.5 × 10−12 m2/s, 9.0 ×10−13 m2/s, 3.4 × 10−13 m2/s and 6.7 × 10−11 m2/s for 90Sr, 137Cs, 60Co, and 125I, respectively.

A series of triaxial compression tests is being conducted under the drained condition on bentonite and sand mixtures, known as buffer, in saturated and optimum water content states to clarify the mechanical properties of the buffer.

It was found that the mechanical properties of bentonite and sand mixtures are strongly influenced by water and bentonite contents: shear strength in a saturated state is less than that in an optimum water content state; shear strength decreases rapidly with increasing bentonite content. Strength properties are much dependent on confining pressure.

The Low-Level Radioactive Disposal facility [1] is planned to cover the vaults with impermeable bentonite-sand mixture in order to minimize groundwater flux through the facility. In case that metal wastes are to be disposed of in the burial facility in the future, it is essential to confirm that bentonite mixture should have gas permeability enough for the gas volume which would be expected to be generated due to an aerobic corrosion of metals[2]. This report performed a preliminary test with a bentonite-sand mixture whose aggregate was Japanese Standard sand. Results are the followings;(l)Gas breakthrough pressure of the mixture was less than the overburden pressure. The gas permeability was estimated to be of order of 10-2 md. (2)The mixture did not deteriorate in hydraulic conductivity after gas breakthrough.

In an attempt to determine the thermodynamic properties of water in bentonite, the vapor pressure of water in compacted bentonite was measured as functions of water content and temperature, under external pressure-free conditions. The relative partial molar Gibbs free energy ΔGH2O, enthalpy ΔHH2Oand entropy ΔSH2O of tne waler in bentonite were determined at temperature of 298.15K. The interlayer distance of montmorillonite in bentonite was also measured by X-ray diffraction.

It is probable that one fourth of the total water included in the bentonite at water content of 20.3wt% and dry density of 1.76 × 103kg/m3 is nearly free water; the water is not regarded as dilute electrolytic solution but the solution with higher ionic strength. Another one fourth of the water in the bentonite at the water content is bound water; the partial molar entropy of the bound water referred to pure water is from a half to whole of solidification entropy of pure water. The remainder is regarded as intermediately bound water.