Experiments on the sorption of uranium(6+) on clinoptilolite from solutions in equilibrium with atmospheric CO2(g) were conducted to understand the fundamental controls on uranium sorption on zeolite minerals, including the effects of pH, aqueous uranium speciation, and uranium concentration in solution. The results indicate that uranium(6+) species are strongly sorbed on the zeolite mineral clinoptilolite at near-neutral pH. The amount of uranium sorbed is strongly dependent on pH and, to some extent, on the total concentration of uranium. Uranium sorption on clinoptilolite is important in the pH range where UO2(OH)2°(aq) is the predominant uranium aqueous species, whereas sorption is inhibited at pH's where carbonateand hydroxy-carbonate-complexes are the primary uranyl species. Surface adsorption appearsto be the main sorption mechanism, although at pH<4 the results suggest ion exchange may occur between the UO2+2 ions in solution and the cations in the intracrystalline cation exchange sites of clinoptilolite.

The effectiveness of zeolite-rich horizons underneath Yucca Mountain, Nevada, as barriers to actinide transport through sorption processes will depend strongly on groundwater chemistry. Reliable predictions of radionuclide transport through these horizons will need to properly account for changes in solution chemistry.

The MINTEQA2 database has been updated and expanded to include radionuclide data from the most recent release of the EQ3/6 database. Comparison of U(VI)-speciation predicted using the old and new MINTEQA2 databases indicates several significant differences, including the introduction of neutral and anionic species at neutral to alkaline pH. In contrast, comparison of results calculated by EQ3 and MINTEQA2, both using Nuclear Energy Agency (NEA) uranium data, reveals only small differences that are likely due to differences in calculated activity coefficients.

With the new database, MINTEQA2 was used to model U(VI)-goethite sorption data from the literature with the Triple-Layer Model (TLM). Values were independently fixed forall but one of the model parameters. The parameter optimization code FITEQL was then used to determine binding constants for mononuclear uranium complexes (UO2(OH)n2−n). The surface complex MOH2-UO2(OH)4− produced a very good fit of the sorption data, which was not significantly improved by the use of two or more surface complexes.

An important question concerning the transport of radionuclides from nuclear waste repositories is whether the adsorption of metals by rocks and soils can be predicted from the properties of the constituent minerals. Attempts by previous researchers to use sorption models based on linear adsorption or weighted "sorptive additivity" have met with limited success. In this study, a “competitive-additivity” model based on surface complexation theory was used to model the pH-dependent adsorption of lead by goethite/Ca-montmorillonite mixtures using complexation constants obtained from single sorbent systems. Measurements of lead adsorption by goethite, Ca-montmorillonite, and goethite-Ca-montmorillonite mixtures (and similar studies of copper and zinc adsorption) demonstrate that the two adsorbents compete for adsorption of metals over wide ranges of pH and concentrations of adsorbents and metals. The adsorption behaviors of the mixtures are determined by the relative concentrations of the adsorbents and their respective affinities for the adsorbate metal. Particle-particle interactions such as heterocoagulation of the oxide and clay do not appear to be significant for the majority of the adsorption sites in this system.

Neptunium has a high solubility in groundwaters from Yucca Mountain [1]. Uranium in nuclear reactors produces 237Np which has a half-life of 2.14 × 106 years. Consequently, the transport of 237Np through tuffs is of major importance in assessing the performance of a high-level nuclear waste repository at Yucca Mountain. The objective of this work is to determine the amount of Np retardation that is provided by the minerals in Yucca Mountain tuffs as a function of groundwater chemistry.

Migration of water, chloride, sodium, and calcium in tonalite was studied, using dynamic column and static through-diffusion methods. Autoradiography of rocks impregnated with 14C-methylmethacrylate was introduced in order to determine the spatial porosity distribution, as well as to identify and visualize the migration pathways of non-sorbing radionuclides in tonalite matrix at the mm-cm scale. The migration routes of sorbing radionuclides and the sorptive minerals in tonalite were determined by autoradiographic methods, using 45Ca as a tracer. Transport of radionuclides was interpreted, using models for hydrodynamic dispersion with diffusion into the rock matrix. In tonalite, porous minerals were distributed homogeneously in matrix and, therefore, retardation capacity of the rock matrix was found to be high.

Humic- and fulvic acids are able to complex polyvalent metal ions, e.g. radionuclides, leading to soluble complexes of significant strength, thereby decreasing the sorption of these compounds to soils and sediments. The interaction of humic materials with radionuclides may significantly influence the availability and transport of the latter in the environment. Typically, studies along these lines have focussed almost exclusively on the radionuclides, whereas the actual role of the humic material has been elucidated only indirectly. In order directly to study the behaviour of the naturally occurring organic macro-molecules in relation to the environmental fate of radionuclides, radio-labelled humic- and fulvic acids can advantageously be applied. Radio-labels such as 14C and 125I have successfully been covalently incorporated in humic- and fulvic acids. Labelling of humic substances as well as preliminary migration studies are discussed

Since colloids are present in all groundwaters, they are studied because of their potential role in the migration of relevant radionuclides in the geosphere. Colloid sampling and characterization campaigns have been carried out in the Grimsel area, Switzerland (Transitgas tunnel, Grimsel Test Site) and in the North Switzerland/Black Forest area (Menzenschwand, Bad Sickingen, Zurzach, Leuggem). On the basis of results obtained for 25 groundwaters a consistent picture is emerging. Colloids in granitic systems are basically composed of components of the aquifer materials (clay, silica). Under the hydrogeochemical conditions existing at depth in crystalline rock systems, colloid concentrations are not expected to exceed 100 ppb. However, under transient conditions when physical and/or chemical changes are occuring (e.g., hydrothermal activity), colloid generation may be enhanced. Colloid concentration of 10 ppm may be observed when the calcium concentration is low enough (<5 ppm).

Whereas reversible sorption on the host rock leads to conservative predictions for radionuclide migration rates, the reverse is true for sorption onto colloids. The most conservative assumptions are therefore reversible nuclide sorption and colloid attachment on the rock and irreversible sorption on colloids.

For calcium concentrations larger than 5 ppm, and sorption capacities at the nM level, their presence in granitic aquifer can be neglected in safety analyses when reversible sorption onto colloids is considered. This situation could change when sorption is followed by aggregation or incrustation processes. This study shows that for the colloid conservative size distributions found and for the granitic water conducting fissures considered, sorption-agglomeration reactions are unlikely to take place. Therefore, irreversible sorption by this mecanism is improbable.

In the safety assessment of nuclear waste repositories, sorption of radionuclides on the surfaces of colloids may significantly modify transport behaviour where colloid concentration is sufficiently high. In the case of fractured geological media, colloids may be excluded from matrix pores, in which case radionuclides bound to them are not subject to the retarding effects of matrix diffusion and sorption onto matrix pore surfaces.

A model is presented describing colloid facilitated transport through fractured media with non-linear sorption. A simple criterion is developed to predict when the presence of colloids will have a significant influence on transport and effects resulting from non-linearity of sorption are described. However, lack of comprehensive sorption data, as well as computational efficiency, mean that the use of a simplified transport model, with linear sorption both on pore surfaces and colloids, is desirable if it can be demonstrated to be conservative. A further criterion is developed to predict where such a model, with linear sorption calculated for the highest concentration encountered along the flow path, would be expected to yield conservative results.

The Waste Isolation Pilot Plant (WIPP) is a U.S. Department of Energy facility intended to demonstrate the safe disposal of transuranic nuclear waste. The WIPP is located in the thick halite beds of the Salado Formation in the Delaware Basin of southeastern New Mexico. Overlying the repository is the Culebra Dolomite Member of the Rustler Formation, a dolomitic unit containing accessory minerals such as calcite, gypsum, and clays. The Culebra is the predominant continuous water-bearing unit in the area. Part of the scientific demonstration of the safety of WIPP involves understanding and estimating what might happen should the WIPP inadvertently be intruded by exploration boreholes long after the WIPP has been decommissioned. Should such a breach occur, it is possible that dissolved transuranic waste could be transported up a borehole and into the Culebra, through which it could eventually reach the accessible environment. The rates and mechanisms for actinide transport in Culebra Dolomite are being investigated as a component of repository performance assessment.

The two primary mechanisms for delaying solute release from a double porosity system are called physical and chemical retardation, both of which retard solute migration relative to bulk water flow. Physical retardation occurs when solutes diffuse out of the advective transport regions (fractures) and into diffusive transport regions (matrix). Results of hydrologic field tests suggest that the Culebra behaves as a double porosity medium. Chemical retardation occurs when chemical interactions between dissolved species and mineral surfaces remove solutes from the aqueous phase. Some fracture surfaces in the Culebra are clay-lined, and, because clays can have a large potential for chemical interactions with solutes (e.g., ion exchange or adsorption), it is possible that fracture linings alone may cause significant retardation of any actinides that could reach the Culebra. This report examines these two features in Culebra transport: delimiting the parameter space where a fully two-dimensional model is needed, and examining the importance of clay linings during transport.

Numerical simulations with the FMT (Fracture Matrix Transport) finite difference model in a finite, idealized double porosity system were performed to: 1) delineate parameter values for which a set of two coupled one-dimensional equations is insufficient to describe behavior and a fully two-dimensional representation is needed, 2) compare chemical retardation by ion exchange for fractures with and without a clay lining, and 3) compare physical and chemical retardation by ion exchange as a function of fracture velocity. The simulations differ from others in that the matrix is finite, and thus can become saturated with tracers introduced into the domain. The modeling suggests that the chemical retardation caused by clay linings may not be important in a double porosity system. Over a wide range, a single parameter group essentially determines the complexity of the model.

A new flow technology has been developed that significantly decreases the time required to obtain transport data on saturated and unsaturated porous/fractured media. This technique is based on open-flow centrifugation and was developed to measure steady-state transport properties in most geologic materials within a matter of hours. Centripetal acceleration does not induce artificial effects in samples i.e., fracturing, collapse of interlayer structures, structural dewatering, compaction, chemical changes, etc., that occur with high-pressure methods. Using this technique, hydraulic conductivities (K) and diffusion coefficients (D) for compacted bentonite and four host rocks have been measured and re-interpreted. Based on these new data, K for compacted bentonite is less than 10−14 m/s, a factor of 1000 lower than previous pressure-gradient measurements, providing further assurance that radionuclide transport through bentonite backfill will be diffusion limited. Measured K for mudstone (1.8 × 10−12 m/s) indicates diffusion-limited far-field transport, while advective transport should occur for granite, basalt, and tuff, with expected matrix diffusion coefficients (correlated to measured D values) of 8.3 × 10−13 and 2.5 × 10−12 m2/s for fractured granite and basalt, respectively.

The nature of diffusivity and porosity in crystalline rock was studied by electrical conductivity measurements, steady-state diffusion experiments, saturation-leaching of tracers with cylindrical rock samples and analysis of the concentrations of different elements from core samples or pore water near fractures. The phenomena of main interest were dead-end porosity, ion-exclusion, sorption, and the continuity of pore networks. The modelling of experimental results was based on a modified Fick's second law for diffusion, which was solved either by analytical or numerical methods. The measured De and ε were found to statistically follow an exponential presentation: Archie's law. The existence of ion-exclusion for anions was confirmed. The connectivity of the pore network extended in the laboratory experiments at least six centimetres, in coarse-grained granite in nature several metres but in fine-grained rock samples of a uranium deposit the element mobilization effects could be seen only to the depth of 2–3 centimetres.

A gas method for fast measurements of diffusion properties of porous materials has been developed. Diffusion coefficients in the gas phase are typically four orders of magnitude larger than those in the liquid phase. For samples whose structures do not change much upon drying it is possible to estimate the diffpision properties of the liquid phase when the properties of the gas phase are known. Advantages of the gas method are quick and easy measurements and therefore they can be used to optimize the liquid-phase measurements which may last months or years. For materials with good correlation between the gas and liquid-phase diffusion, the number of liquid phase measurements can be greatly reduced.

Nitrogen was used as the carrier gas and helium as the tracer gas. Helium concentrations were measured by using standard techniques. The equipment was tested for two sample geometries, a column made of a polished granite slab and an acrylic slab with a flow channel in between, and a hollow bore core where diffusion occurs through the sample. Both types of measurement were modelled and apparent diffusion coefficients in the gas phase were determined.

British Nuclear Fuels ple has initiated an experimental programme to measure the leaching behaviour of radionuclides from various low level radioactive waste (LLW) mareials using Lysimeters. The programme commenced in 1986 and to date 10 lysimeters have been commissioned. These have concentrated on simulating shallow trench conditions but a further programme is now planned to study concrete vault environments. The aim of the study is to provide information on leaching processes as part of the ongoing Drigg Near Field Programme, and also to yield input data for radiological assessment purposes. Towards this end, data have been gained from the lysimeters on basic chemistry, gas generation and radionuclide Release Coefficients. This paper concentrates on one of the lysimeters which has recently been decommissioned and for which interim analytical data are available. Some general comments are given on BNFL's experience using lysimeters and their applicability as a rapid and effective technique for studying near field degradation processes.

Field- and pilot-scale uranium leaching facilities (lysimeters) have been constructed at the Oak Ridge Y-12 Plant to investigate the rate of uranium (U) leaching and the residual U concentration that can be expected when low level radioactive waste (LLW) is buried in landfills. Current plans are to load three large standard column lysimeters with compacted production trash (blotter, waste wipes, paper etc.) and apply groundwater to the surface of the lysimeters.

Several issues had to be addressed before the field lysimeters could be loaded with the LLW. Predominate among these concerns was how to adequately characterize the waste placed in the lysimeters (∼ 10 yd3 waste per lysimeter) and how to effectively collect in situ leachate samples from compacted wasteforms in the absence of soil. Pilot scale studies revealed that monitoring individual bags (30 gal) of LLW with a waste curie monitor would be sufficient to determine the initial U loading and that conventional suction candles would perform adequately within the compacted waste for extended time periods if they are installed properly. In the pilot leaching studies, the highest U concentration (∼ 0.2 mg/l) in the leachate was observed during the first month of leaching, before active biological decomposition processes became dominant making the lysimeter anoxic. After the initial oxic flushing, U concentrations in the leachate became very low (mean = 0.021 mg/l).

The regulations that currently govern repositories for spent fuel and high-level waste require demonstrations that are sometimes described as impossible to make. To make them will require an understanding of the current and the future phenomena at repository sites; it will also require credible estimates of the probabilities that the phenomena will occur in the distant future. Experts in many fields—earth sciences, statistics, numerical modeling, and the law—have questioned whether any amount of data collection can allow modelers to meet these requirements with enough confidence to satisfy the regulators. In recent years some performance assessments have begun to shed light on this question because they use results of actual site investigations. Although these studies do not settle the question definitively, a review of a recent totalsystem assessment suggests that compliance may be possible to demonstrate. The review also suggests, however, that the demonstration can be only at the “reasonable” levels of assurance mentioned, but not defined, in the regulations.

The Waste Isolation Pilot Plant (WIPP) in southeastern New Mexico is being developed by the U.S. Department of Energy as a disposal facility for trans-uranic waste. In support of this project, Sandia National Laboratories is conducting an ongoing performance assessment (PA) for the WIPP. The ordered triple representation for risk proposed by Kaplan and Garrick is used to provide a clear conceptual structure for this PA. This presentation describes how the preceding representation provides a basis in the WIPP PA for (1) the definition of scenarios and the calculation of scenario probabilities and consequences, (2) the separation of subjective and stochastic uncertainties, (3) the construction of the complementary cumulative distribution functions required in comparisons with the U.S. Environmental Protection Agency's standard for the geologic disposal of radioactive waste (i.e., 40 CFR Part 191, Subpart B), and (4) the performance of uncertainty and sensitivity studies. Results obtained in a preliminary PA for the WIPP completed in December of 1991 are used for illustration.

The Department of Energy's Nevada Operations Office has disposed of a small quantity of transuranic waste at the Greater Confinement Disposal (GCD) facility in Area 5 of the Nevada Test Site. The Environmental Protection Agency standards for disposal of transuranic waste (40 CFR 191) require that a performance assessment be conducted and an iterative approach has been adopted. This approach consists of using performance assessment results to guide the collection of site characterization data, and in turn, new site characterization data is incorporated into each successive iteration of the performance assessment. The results of the first iteration indicated that the GCD site should be able to comply with 40 CFR 191. This paper presents results of the second iteration based on results from site-specific environmental tracer studies. These studies indicate significantly lower recharge values then those used in the first performance assessment iteration. For performance assessment these results imply that the downward ground-water advection pathway may be negligible, and therefore, the upward liquid diffusion pathway becomes more significant. The models and the model parameters that were used to simulate this upward pathway have been refined. The preliminary results from these models still indicate that the GCD facility should be able to comply with the Containment Requirements in 40 CFR 191.

An approach to the validation of ground-water flow and transport models that are used in support of licensing a radioactive waste site is proposed and demonstrated. The approach attempts to interpret modeling results in light of their intended purpose. Instead of trying to find the ‘best’ fit model to the experimental data, we attempt to determine the conservativeness or bias invoked by applying different models to simulate the same experiment. In this way we are able to provide guidance to the regulatory community on where they should expect model results to fall relative to actual data and provide some evidence that certain model approaches lead to conservative results while other modeling approaches lead to non-conservative results. Applying this approach to the Twin Lakes Tracer Test demonstrated that a simple one-dimensional flow and one-dimensional dispersion model consistently over-predicts the maximum concentration. A one-dimensional flow with three-dimensional dispersion and two dimensional flow with two dimensional advection-dispersion are conservative only if they employ laboratory scale dispersivities. Therefore, if a decision-maker were to accept a similar site based on results from these types of models, there would be a relatively low chance that he has accepted an unsafe site. However, if decision-makers were to reject a similar site based on these types of results, there would be a relatively large chance that they rejected a safe site.

This paper presents a method for sensitivity and uncertainty analyses of a hypothetical nuclear waste repository located in a layered and fractured unconfined aquifer. Groundwater travel time (GWTlT) has been selected as the performance measure. The repository is located in the unsaturated zone, and the source of aquifer recharge is due solely to steady infiltration impinging uniformly over the surface area that is to be modeled. The equivalent porous media concept is adopted to model the fractured zone in the flow field. The evaluation of pathlines and travel time of water particles in the flow domain is performed based on a Lagrangian concept. The Bubnov-Galerkin finite-element method is employed to solve the primary flow problem (non-linear), the equation of motion, and the adjoint sensitivity equations. The matrix equations are solved with a Gaussian elimination technique using sparse matrix solvers. The sensitivity measure corresponds to the first derivative of the performance measure (GWTT) with respect to the parameters of the system. The uncertainty in the computed GWTT is quantified by using the first-order second-moment (FOSM) approach, a probabilistic method that relies on the mean and variance of the system parameters and the sensitivity of the performance measure with respect to these parameters. A test case corresponding to a layered and fractured, unconfined aquifer is then presented to illustrate the various features of the method.

This paper presents results of unsaturated flow simulations undertaken as an auxiliary analysis for the Iterative Performance Assessment (IPA) project, one of the approaches adopted' by the U.S. NRC to develop repository license application review capabilities. The effects on flow of common geological features, such as nonhorizontal stratification and vertical or near-vertical fault zones intersecting the strata, in a two-dimensional (2D) domain are studied. Results indicate that the presence of layers and crosscutting fault zones tend to induce three-dimensional (3D) unstable flows in the unsaturated zone. The instability is manifested in our simulations by an oscillatory behavior of steady state. This numerical instability imposes extremely stringent criteria on the time step used in the simulation. Finally, once stable steady-state solutions are attained, the effect of the crossing point in the matrix-fault unsaturated hydraulic conductivity curve on groundwater flux vectors and moisture content distributions is studied.