Adsorption processes are important in controlling U concentrations in ground water. Quantifying such processes is extremely difficult in that in situ conditions cannot be directly measured. One rock characteristic that must be known to quantify adsorption is the specific surface area of reactive minerals exposed to the ground water. We evaluate here three methods for estimating specific surface area in situ. The first is based on the dissolution kinetics of sodium feldspars, the second on emanation of radon-222 and the third on adsorption of naturally-occurring U. The radon-222 method yields estimates 5 to 8 orders of magnitude greater than those obtained via the other two methods; too large probably because of effects related to fracture geometry. Estimates of specific surface area based on modelling adsorption of natural U by aquifer materials are of comparable magnitude to those from the feldspar-dissolution kinetics approach. These conclusions are based on analyses of water from 145 wells in crystalline-rock aquifers from Pennsylvania, New Jersey, Maryland, and Colorado. Computer modelling of the chemical data using PHREEQE [1] showed that uraninite or coffinite approach saturation in reducing water, limiting total U to <2 × 10−9 m. Generally, U minerals are below saturation in oxidizing ground water, where uranyl-carbonate complexes are the dominant dissolved U species. Autoradioluxographs of thin sections show areas of concentration of radioactivity in the rocks and establish that U is concentrated along fracture boundaries and on ferric oxyhydroxide grain coatings. Because U minerals generally are undersaturated, U mobility is limited by adsorption onto ferric oxyhydroxides and other mineral surfaces. Calculations of uranyl adsorption from the ground water onto goethite using the program M1NTEQ [2] show that adsorption decreases with increased carbonate concentrations due to the formation of uranyl-carbonate complexes. Results of this paper improve our understanding of the mobility of U that might be released into oxidized ground water in crystalline rock from a breached radioactive-waste repository.

Transport by groundwater within the Culebra Dolomite, an aquifer above the Waste Isolation Pilot Plant (WIPP), is the most probable mechanism for long-term release of radionuclides to the accessible environment. Radionuclides could be retarded by sorption if the groundwater is exposed to sufficient amounts of fracture-lining clays. In this natural-analog study, distributions of U and trace metals have been examined to constrain the strength of clay/solute interactions within the Culebra.

Uranium solid/liquid distribution ratios, calculated from U concentrations of groundwaters and consanguineous fracture-filling clays, range from ∼80 to 800 mℓ/g and imply retardation factors of 60 to 500 using a fracture-flow model. Retardation factors inferred from uranium-series disequilibria and 14C ages in Culebran groundwaters alone are much lower (∼10), implying that clays may contain a significant unreactive component of U. Such a possibility is corroborated by Rb/Sr ages; these imply long-term stability of the clays, with resetting occurring more than 250 Ma ago. Factor analysis and mass-balance calculations suggest, however, that Mg-rich clays are dissolving in Pleistocene-age groundwaters and/or are converting to Na-rich smectites, and that B and Li are taken up from the water by the clays. Apparently, the solution chemistry reflects gradual equilibration of clays with groundwater, but thus far the bulk of the clays remain structurally intact. Measurements of the distribution of U in the Culebra will be more meaningful if the inert and exchangeable components of the U content of the clays can be quantified.

Hydrothermal systems in tuffaceous and older sedimentary rocks contain evidence of the interaction of radionuclides in fluids with rock matrix minerals and with materials lining fractures, in settings somewhat analogous to the candidate repository site at Yucca Mountain, NV. Earlier studies encompassed the occurrences of U and Th in a “fossil” hydrothermal system in tuffaceous rock of the San Juan Mountains volcanic field, CO. More recent and ongoing studies examine active hydrothermal systems in calderas at Long Valley, CA and Valles, NM. At the Nevada Test Site, occurrences of U and Th in fractured and unfractured rhyolitic tuff that was heated to simulate the introduction of radioactive waste are also under investigation. Observations to date suggest that U is mobile in hydrothermal systems, but that localized reducing environments provided by Fe-rich minerals and/or carbonaceous material concentrate U and thus attenuate its migration.

Several radionuclides of the naturally occurring 238U, 235U and 232Th decay series have been measured in high-temperature (300–350 °C) brines from several flow tests of the Sal ton Sea Scientific Drilling Project (SSSDP) well. Activities (dpm/kg) at the initial flow test of the well were 222Rn (2130), 226Ra (2200), 228Ra (1120), 224Ra (1010), 223Ra (6), 210Pb (3260), and 212Pb (2250). In contrast, activities of U and Th isotopes were much lower (< 1 dpm/kg). Following the first flow test (12/85), the well was deepened and sampled again during a short flow test (3/86). Radium and uranium concentrations were about two times lower during the second flow test, while other elements had similar concentrations. During a subsequent 20 day flow test of the well (6/88), flow rates were regulated at the wellhead to investigate reservoir characteristics, and 226Ra and 210Pb activities were found to be negatively correlated with the flow rate and positively correlated with the wellhead pressure. The 226Ra activity varied by a factor of five, roughly the variation in total flow rate, while well-head pressure varied only by 25% and 210Pb activity varied by a factor of two. Ratios of short-lived daughters to longer-lived parents increased in these flow tests. Results from mass balance calculations incorporating reaction kinetics suggest that differences in the degree of radium adsorption, rates of precipitation and dissolution of reservoir minerals exist in different flow zones.

The distribution of 238U, 234U, 230Th and 236Ra in rocks from the Koongarra uranium deposit has been examined. Alpha-spectrometry and ICP-MS were used to estimate elements in “accessible” and “inaccessible” phases, separated by a two-phase extraction scheme; and bulk analyses of visibly distinct zones were made using PIXE-PIGME, DNAA and XRF.

The secondary U dispersion fan of the Koongarra site is characterised by 234U/238U, 230Th/234U and 236Ra/230Th activity ratios (ARs) below unity in the extractable (accessible) phases, and above unity in the non-extractable (inaccessible) phases. The patterns of the ARs suggest zones of different U deposition and leaching rates with the most rapid accumulation at the base of weathering, especially at the extremities of the dispersion fan.

The role of 234Th, a daughter nuclide of 238U having a half life of 24 days, in the migration of uranium series nuclides has been studied to understand the mechanism which gives a higher migration velocity of 238U than that of 234U. We assume that 234Th is adsorbed at two different sites; one, where 234Th is reversibly adsorbed, and the other, where 234Th is irreversibly fixed. Calculations have shown that the fixation of ^Th to a rock can cause the apparent migration velocity of 234U to be reduced compared to that of 238U and this agrees with observed field data. Changes in the fixation rate constants affect the relative mobility of 238U and 234U, and this can account for the different migration behavior at different depths in the Koongarra ore deposit.

The relevance of alteration of chlorite, one of the major constituent minerals of the host rock to uranium ore at Koongarra, to the redistribution of uranium in the vicinity of the ore deposit has been examined. The chlorite alteration is produced by weathering; chlorite is transformed to vermiculite, through regularly interstratified chlorite/vermiculite. The Fe released from chlorite reprecipitates to form iron minerals, possibly ferrihydrite. Vermiculite is then replaced by kaolinite (and possibly smectite), further releasing Fe. On the millimeter scale, an alpha autoradiography study has shown that uranium concentrations are qualitatively proportional to the extent of the alteration; altered chlorite grains having higher uranium concentration. On the meter scale, the X-ray diffraction study has revealed that the abundances of chlorite, vermiculite, and kaolinite correspond well to the low, intermediate, and high uranium concentration zones, respectively. These suggest that the interaction of the uraniferous solution with chlorite causes the alteration of chlorite and the precipitation and sorption of uranium in the alteration products from the solution, and thus, the uranium migration is retarded at Koongarra.

A U-deposit located in hydrothermally altered tuff in Mexico, and two tropical weathering systems from Brazil and Cameroon were studied as natural analogues of radionuclide release and migration in the geosphere. In these alteration systems, kaolinite is the main secondary phyllosilicate associated with variable amounts of iron oxides. Defect centers are detected in natural kaolinites by means of electron paramagnetic resonance. The origin of the paramagnetic defect centers (PDC) is from radionuclide irradiation. This interpretation is supported by laboratory experiments on kaolinite using four different types of radiation: x-rays, γ-rays, and He+ and Pb2+ ion beam implantations. Several types of PDC (designated A-, A‵- and B-centers) are clearly identified and can be distinguished with respect to their relative stability. The A-center is stable over geological periods (on the order of 109 years) at surface temperatures. Moreover, it is shown that PDC’s in kaolinites reflect the cumulative radiation dose.

PDC’s in kaolinite have been used to trace radionuclide migration in the three natural alteration systems studied. The kaolinites investigated show a significant concentration of PDC’s which corresponds to background natural irradiation. The different types of PDC detected may record successive irradiations of kaolinite directly linked to geochemical conditions prevailing during and after kaolinite growth. Because of their different thermal stability, these different PDC–s may be considered to have formed at different time periods. This makes a unique tool for tracing the dynamics of the transfer of radionuclides and allows the use of kaolinite as an in-situ dosimeter. Because some of the geological settings chosen for nuclear wastes are located in kaolinite-bearing strata, PDC dosimetry could be used in a safety assessment of these wastes.

Halogen ion migration through clay-rich sediments is particularly well documented in cores from Loch Lomond, Scotland, where distinctive patterns for CI, Br and I have been observed. The sediments are a sequence of marine and limnic silts and clays which cover glacial deposits. The marine band, resulting from a transgression 6900 y B.P. to 5400 y B.P., acts as a source for the halogens.

A 1-D conceptual model and code has been developed taking into account variations in the concentration boundary conditions and the accumulation of sediments. The model has been calibrated against the CI concentration distributions for which boundary conditions can be estimated readily.

The model has been applied to Br and I profiles. Deviations between the Cl-calibrated model and the measured profiles were attributed to reactive processes. Release rates from the marine band for Br and I are deduced. Their sensitivity with respect to system properties, assumptions over boundary and initial conditions and reaction mechanisms is discussed.

An assessment of the fundamental research needs in understanding and predicting the migration of radionuclides in the subsurface is provided. Emphasis is on the following three technical areas: (1) aqueous speciation of radionuclides, (2) the interaction of radionuclides with substrates, and (3) intermediate-scale interaction studies. This research relates to important issues associated with environmental restoration and remediation of DOE sites contaminated with mixed radionuclide-organic wastes.

To help evaluate the concept of nuclear fuel waste disposal in plutonic rock of the Canadian Shield, the long-term performance of a hypothetical disposal facility is being studied. This case study uses information from a Shield research area, laboratories, and conceptual engineering. It illustrates the use of performance assessment to derive constraints on the siting, construction, and operation of a disposal facility, and to demonstrate the feasibility of safe disposal of nuclear fuel waste.

The advective diffusion analyses have been applied to many fields of science and engineering, such as dispersion for chemically reacting(first-order reaction) substance, thermal transport in fluid, analysis of electromagnetic field caused by a moving magnet. electron transport in semiconductors, underground migration of radioactive waste, and so on. The boundary element method (BEM) has been developed extensively for the last decade to solve these transient advective diffusion equation. The time integrations of the fundamental solution in the boundary integral equation, however, make the BEM application to advective diffusion problems difficult. Therefore, the time integration has been approximated in the past relevant publications. This paper describes the BEM in which the time integration is done analytically, and technique is demonstrated with several examples. Good results have been obtained in the example calculations, where comparisons are made with the results from other numerical codes.

Computer graphics systems that provide interactive display and manipulation of three-dimensional data are powerful tools for the analysis and communication of technical information required for characterization and design of a geologic repository for nuclear waste. Greater understanding of site performance and repository design information is possible when performance-assessment modeling results can be visually analyzed in relation to site geologic and hydrologic information and engineering data for surface and subsurface facilities. In turn, this enhanced visualization capability provides better communication between technical staff and program management with respect to analysis of available information and prioritization of program planning.

A commercially-available computer system was used to demonstrate some of the current technology for three-dimensional visualization within the architecture of information systems for nuclear waste management. This computer system was used to interactively visualize and analyze the information for two examples: 1) site-characterization and engineering data for a potential geologic repository at Yucca Mountain, Nevada; and 2) three-dimensional simulations of a hypothetical release and transport of contaminants from a source of radionuclides to the vadose zone. Users may assess the three-dimensional distribution of data and modeling results by interactive zooming, rotating, slicing, and peeling operations. For those parts of the database where information is sparse or not available, the software incorporates models for the interpolation and extrapolation of data over the three-dimensional space of interest.

Limiting concentrations of radionuclides (RN) in a granitic environment are determined by computer modelling of a disposal site with quartz sand and clay engineered barriers. Six RN (Sr, U, Pu, Np, Th and Tc) are studied. Temperature is varied with distance from the host rock to waste from 25 to 200°C, The behaviour of these elements appears to be iindependent, of the barrier material and strongly dependent on temperature effect on mineral solubilities.

The geochemical modelling in support of the performance assessment of radioactive waste disposal systems calls for a large number of chemical thermodynamic data. For realistic modelling it is essential that the data used are fully consistent. The verification of consistency of existing data bases is complicated by the fact that it requires the knowledge of a considerable amount of primary information to ensure:

1. 1. Consistency with the fundamental laws of thermodynamics

2. 2. Consistency within a chemical model

3. 3. Consistency with auxiliary data

4. 4. Consistency in the data correction procedures

This paper includes selected examples for each of the four items to visualize the problems. It should be noted that there are numerous other systems that could serve as examples as well, and the range of cases reported here is far from being exhaustive.

Realistic geochemical modelling depends not only on the quality of the data base, but also on the quality of the chemical model used. For the establishment of a chemical model, additional information is needed that is not contained in thermodynamic data bases.

The safety assessment of the repository for high level radioactive waste in the Boom clay formation requires reliable data for the migration parameters. The experimental set-up and the interpretation method is briefly described for percolation experiments on undisturbed clay cores drilled from the formation and results are reported. The undisturbed clay cores are drilled perpendicular and parallel to the stratification of the formation to study the isotropy of the formation. To represent the real situation as close as possible, in-situ interstitial clay pore water is used as percolating liquid.

Parameter values determined in the percolation experiments are presented for bromine, iodine and tritiated water. Darcy velocity and effective stress are used as variables. Anisotropy of the formation is demonstrated for the hydraulic conductivity, but is found to be trivial for the dispersion parameters. A general relation between the apparent dispersion constant, the diffusion accessible porosity and the Darcy velocity is also given.

In crystalline rock groundwater flows predominantly in fractures and fissures. Strongly varying fracture aperture guides the flow preferentially in some parts of a fracture plane, in so called channels. In our hydraulic model the degree of channeling together with the aperture variation along a channel is included as a factor which is the ratio of the aperture from transmissivity measurements and the aperture from the tracer tests.

The developed transport model takes into account the coupling of molecular diffusion and advection in a velocity field varying linearly over a characteristic width. Various flow velocities in different parts of a channel cause a transient phase with non-Fickian behavior of dispersion. This might erroneously be attributed to other processes e.g. matrix diffusion when not taken into account in the migration modeling of tracers. Molecular diffusion across the flow field, however, tends to smooth out the transport time differences. With time the dispersion diminishes and becomes more symmetric in confined channels.

The concept and models have been applied to predict and interpret field experiments aimed to investigate transport over long distances in highly conductive fracture zones. The analyzed experiments have been performed at the Finnsjön research area in Sweden and they belong to the test case 5 of the INTRAVAL project.

A procedure for the measurements of the redox capacity of rocks and minerals is presented. Both solid and pulverized samples are used. In a preliminary test described in this paper, an oxygen loss of an aqueous solution in contact with pulverized granodiorite was measured directly on-line, with an oxygen sensor.

The increase of Fetot in the solution responded relatively well to the oxygen consumption. The maximum rate varied from 3.0 × 10−4 to 9.0 × 10−11 mol l−1 s−1 correspondingly. Oxygen loss of 0.04 mmol (O2) and iron increase of 0.1 mmol were recorded in the solution. Fe(II) reduction in the solid phase was also examined. Fe(II) was inferred to have partially dissolved during the experiment. When the system was equilibrated at the ambient conditions, the vessel was pressurized upto 100 bar with oxygen gas. Three weeks later, the additional decrease of Fe(II) in the solid phase was minimal, only 0.07 % Fe(II) was dissolved or oxidized on the rock surfaces. At this point, the Fetot in the solution, had decreased down to the background level. According to the total oxygen loss at the ambient coditions, the maximum redox capacity of the examined granodiorite (BET 1.0 m2/g) is 0.002 mmol/m2.

Systematic differences in pH, cation/proton ion activity ratios, and redox have been observed between solutions produced in rock-water hydrothermal experiments with tuff, granite, and basalt. Stable pH values in tuff-water experiments may be as much as 1.5 pH units more acidic than basalt-water experiments at the same temperature and ionic strength. Redox (log fO2) values in 300°C tuff experiments are 4–7 orders of magnitude more oxidizing than basalt experiments and ca. 4 log units more oxidizing than the magnetite-hematite buffer. Such fluid differences could significantly affect the performance of a high-level nuclear waste repository and should be considered in repository design and siting.

We present analytic studies of the transport of gaseous species released from a spent-fuel waste package, as affected by a heat-pipe of counterflowing liquid and vaporized water in the surrounding rock. A heat-pipe is caused by heating which vaporizes pore water near the waste, releasing vapor into the fractures. Driven by its pressure gradient, the vapor flows away from the waste and condenses where the rock is cooler. Because of capillary pressure gradient due to non-uniform liquid saturation, condensate flows towards the waste through the porous rock. We first develop analytic solutions for the time-dependent transport of energy and fluid from the waste container to the surrounding fractured porous rock. From the mass fluxes of liquid and vapor, we solve the advective-diffusive transport of a gaseous species released from the waste. The major assumptions are quasi-steady-state, local thermodynamic equilibrium, no noncondensable gases, and no sorption. Our results include the extent of the heat-pipe zone as function of time, the vapor velocity distribution in the heat-pipe zone, radionuclide concentration in water vapor, and the flux of radionuclide at the waste surface normalized to the surface concentration. We find that the vapor velocity in the heat-pipe zone is 1000-fold greater than the local air velocity if there were no heat pipe. If the gaseous species release mechanism maintains a near-constant concentration of gaseous species in the gas outside and near the waste container surface, the mass rate of transport of that species would be increased 1.3 to 7 times greater than if there were no heat pipe. However, if the release rate of the gaseous species is affected little by the concentration of that species outside the container, the heat-pipe can have little affect on the transport rate of that species.