A simulator of reactive chemical transport has been constructed with the capabilities of treating variable temperatures and variable oxidation potentials within a single simulation. Homogeneous and heterogeneous chemical reactions are simulated at temperature-dependent equilibrium, and changes of oxidation states of multivalent elements can be simulated during transport. Chemical mass action relations for formation of complexes in the fluid phase are included explicitly within the partial differential equations of transport, and a special algorithm greatly simplifies treatment of reversible precipitation of solid phases. This approach allows direct solution of the complete set of governing equations for concentrations of all aqueous species and solids affected simultaneously by chemical and physical processes. Results of example simulations of transport, along a temperature gradient, of uranium solution species under conditions of varying pH and oxidation potential and with reversible precipitation of uraninite and coffinite are presented. The examples illustrate how inclusion of variable temperature and oxidation potential in numerical simulators can enhance understanding of the chemical mechanisms affecting migration of multivalent waste elements.

Synthetic basaltic ground water was spiked with different ratios of high-purity solium selenate and sodium selenite, to yield ratios of aqueous SeO42to SeO 42from 2100/1 to 1/44000, at temperatures of 25°C and 75°C. The apparent 2h of the solutions was measured by means of a platinum electrode. The time series Eh data were extrapolated to infinite time using the method of Muller [1,2]. Results show no relationship between the analyzed ratios of Se(VI)/Se(IV) in solution and the observed+Eh plues at the platinum electrode. In contrast, acidic solutions of Fe /Fe show nearly perfect Nernstian behavior and rapid electron transfer at the platinum electrode. Based on the results of this study, it is probably invalid to use Eh measurements as input to any equilibrium computer model for purposes of predicting the behavior of a pure selenium system.

Technetium and plutonium in unprocessed nuclear reactor wastes are major concerns with regard to their containment in the geologic environment. Both nuclides have long half-lives; therefore, they will exist long after engineered barriers can be considered reliable. Consequently, strategies for the containment of these two elements depend on their retention in the geologic barrier until they have decayed to innocuous levels. Because these are the rarest elements in nature, there have been few direct observations of their geochemical behavior; predictions con- cerning their fate in the repository are based on properties that can be observed in the laboratory. We are attempting to complement the laboratory work by studying the geochemistry of natural plutonium and technetium.

In support of the Nevada Nuclear Waste Storage Investigations Project experiments were conducted to study the effects of heat generated by a nuclear waste repository in densely welded, devitrified tuff on the underlying, compositionally-equivalent glassy tuff at Yucca Mtn. Solid wafers of glassy tuff were reacted with a dilute ground water for several months at 150°C and 250°C at 100 bars pressure in Dickson-type, gold-bag rocking autoclaves. The in-situ chemistry of the hydrothermal fluids was modeled and the chemical affinities for all possible mineral precipitation reactions (contained within the extensive database) were calculated using the EQ3/6 program.

In the 250°C experiment the calculations suggest that a zeolite mineral would be expected to form. Analyses of the run products showed that not only had the wafer been extensively corroded and the glass shards replaced by clinoptilolite, but pure clinoptilolite had precipitated directly from solution. In the 150°C experiment, although clay minerals were thermodynamically favored to form in the first half of the experiment, by the end of the run a zeolite mineral was predicted to form. Analyses of the run products showed no well-formed secondary minerals (clays or zeolites) had formed. At the lower temperature the effects of precipitation kinetics may preclude the formation of the zeolite within the time span of this experiment. In general the observations are in relatively good agreement with the geochemical model calculations. This type of study demonstrates the interpretive/predictive capabilities of a combined experimental/geochemical modeling approach to studies of nuclear waste isolation. This combined approach will aid in satisfying licensing requirements to assure long-term performance.

The generation of humic colloids of Am(III) has been investigated in Gorleben groundwaters containing different amounts of humic substances. Dissolved organic carbon (DOC) in these groundwaters consists mainly of humic acid and fulvic acid, which is present in a colloidal form through aggregation with trace heavy metal ions of groundwater constituents. Concentrations of these heavy metal ions are proportional to the DOC concentration. The generation of Am(III) pseudocolloids through geochemical interactions with humic colloids in different groundwaters is quantified by ultrafiltration as well as ultracentrifugation by the aid of radiometric concentration measurements. The speciation of dissolved Am(III) species in groundwaters is carried out by laser induced photoacoustic spectroscopy (LPAS).

This work concerns itself with the study of effects of soil aggregation and high salt concentrations on the transport of radionuclides by concentrated brine flowing through an aggregated porous medium. The medium is considered to be composed of porous rock aggregates separated by macropores through which the brine flows and transport of salt and radionuclides takes place. The aggregates contain dead-end pores, cracks, and stationary pockets collectively called micropores. The micropore space does not contribute to the flow, but it serves as a storage for salt and radionuclides. Adsorption of radionuclides takes place at internal surfaces of aggregates where we assume that a linear equilibrium isotherm describes the process.

A one-dimensional numerical model is developed which is based on two sets of equations: one set for the flow and transport of salt and another set for transport of radionuclides. Results of numerical experiments clearly indicate that that. the existence of high salt concentrations markedly reduces the peak of nuclides concentration and slows down their movement. Also, it is found that diffusive mass exchange between macropores and aggregates results in a pronounced lowering of the radionuclides concentration peaks.

We have attempted radiometric dating of halide-sulfate salts and clay minerals from the Delaware Basin, New Mexico, USA, as part of geochemical study of the stability of the evaporite sequence at the WIPP (Waste Isolation Pilot Plant--a US DOE facility) site. We undertook this dating to determine: (1) primary age of evaporite genesis or time(s) of recrystallization, (2) if previously undated evaporite minerals (leonite, polyhalite, kieserite) give useful data, and (3) if the detrital clay minerals have been radiometrically reset at any time following their incorporation into the evaporite medium. We have shown earlier that polyhalites can indeed be successfully dated by the K-Ar method, and once corrections are applied for admixed halide minerals, dates of 210-230 Ma for the Delaware Basin are obtained. Rb-Sr isochrons from early stage sylvites-polyhalites- anhydrites yield 220 ± 10 Ma, even when some sylvites yield lower K-Ar dates due to l1s of 40Ar*. K-Ar dates on leonites and kieserites are also low due to 40Ar* loss, but their Rb-Sr dates are higher. Detrital clay minerals from the Delaware Basin collectively yield a highly scattered isochron (390 ± 77 Ma), but samples from a local area, such as the WIPP Site, give a much better age of 428 ± 7 Ma. These dates show that the interaction between the clay minerals and the evaporitic brines was insufficient to reset the clay minerals Rb-Sr systematics. In a related study, we note that a dike emplaced into the evaporite at 34 Ma had only very limited effect on the intruded rocks; contact phenomena were all within 2 m of the dike. All of our geochemical (radio-metric and trace element) studies of the WIPP site argue for preservation of the isotopic and chemical integrity of the major minerals for the past 200 Ma.

According to the present Finnish concept sodium bentonite will be used as a buffer material in the repository for high-level waste. Experimental and theoretical studies treating the effect of bentonite upon the chemical conditions in a repository have been initiated with the object of specifying the chemistry of the near field.

Sodium bentonite was let react with water under anaerobic conditions at 25°C for 540 days, during which time six fluid samples were extracted for the chemical analysis of 15 chemical species. The generated fluid phase was alkaline (PH = 9…10) and contained a high amount of bicarbonate. Also a low redox-potential was measured. The fluid phase chemistry was investigated using the geochemical code PHREEM. Calcite saturation was observed in all fluid samples.

A modelling of sodium bentonite interaction with water based on the main mineral components of bentonite was also performed with PHREEQE. A fairly good agreement between experimental results and model calculations was observed.

The basic process in thermal transformation of smectite to hydrous mica is generally thought to be a lattice charge change caused by partial replacement of tetrahedral silica by aluminum. This is assumed to yield preferential uptake of potassium with concomitant contraction and loss of expandability of smectite aggregates as well as release of silica that migrates by diffusion from interlayer space into interaggregate voids, where it may form amorphous silica hydrogels and possibly quartz.

The latter process was investigated by use of hydrothermal tests of fully water saturated, as well as partly saturated Na-montmorillonite gels, the intention being to identify possible heat-induced changes in expandability on a molecular scale by applying electron microscopy. The gels were exposed to a temperature of 225°C for 18 days in a first test series and the microstructural patterns compared with those of non-heated material.

A clear tendency of lattice contraction was observed in the heated clay gels, particularly in the non-saturated one. The microstructure had the form of networks of virtually non-expandable, interwoven dense stacks.

A possible physical explanation of the contraction is that the heat caused instability of the interlayer water lattices, yielding dominant interatomic mass forces which caused contraction of the stacks. In connection herewith, silica was released from the smectite lattices and precipitated at the edges of the stacks, which reduced or eliminated the expandability. Minute, precipitated silicic bodies, amorphous as well as crystal-line, appeared in both clays.

In this research the diffusion of strontium and cesium through bento- nite slabs in stationary state has been measured. The measurement gives directly the transport through the sample and the corresponding effective diffusivity. The apparent diffusivity, diffusivity of the sorbed ions as well as the sorption factors can also be evaluated on the basis of the same diffusion measurement. The sorption on the coxmpacted bentonite has also been measured in a special sorption test where a radioactive tracer has been allowed to diffuse into compacted bentonite until equilibrium was reached.

The measured effective diffusivities were clearly higher than they would be in the case of pure pore diffusion. The effective diffusivities also increase with K and the apparent diffusivities are almost independent on KD. This supports the theory that also the sorbed ions diffuse in their own phase. The sorption factors of the compacted bentonite correspond to the values of the batch experiments.

Results are presented of physical measurements on candidate buffer materials for use in nuclear fuel waste disposal. The materials being considered as constituent elements of engineered barriers are essentially calcium smectite clays, in other terms swelling clays, coming from fourteen french deposits. The criteria for good candidates are mainly: smectite content in the clay materials, carbonate and organic material content and bulk density of the material, compacted under a pressure of 100 MPa.