The influence of bentonite colloids on neptunium(V) migration in granitic rock

In Finland, the repository for spent nuclear fuel (SNF) will be excavated at a depth of about 500 meters in the fractured crystalline bedrock in Olkiluoto at Eurajoki implemented by Posiva Oy. The engineered barrier systems (EBS), consisting of a solid fuel capsule, a copper-iron canister and the bentonite buffer should prevent the migration of radionuclides to the biosphere. Montmorillonite, the main mineral of bentonite, is like other aluminosilicates known to retain radionuclides, thus, preventing them from migrating from the repository with the groundwater. Bentonite erosion resulting in the formation of colloids may have a direct impact on the overall performance of the bentonite buffer. The potential relevance of colloids for radionuclide transport is highly dependent on the formation of colloids, the stability and mobility of colloids in different chemical environments, and their interaction with radionuclides [1]. Stable and mobile bentonite colloids can be formed when the glacial meltwater dilutes the groundwater. In these mildly oxic conditions, neptunium(V) will be present in its pentavalent oxidation state as the neptunyl cation (NpO2+), which is rather soluble, highly mobile and poorly adsorbed. Due to the long half-life of Np-237 (2.144·106 y), it will be a major dose contributor after 100,000 years in the SNF repository.
In our previous study, the interaction of Np(V) with Na-montmorillonite purified from MX-80 bentonite and corundum was investigated [2]. Corundum was used as a reference mineral in order to study the aluminol surface sites present on clay minerals, which are regarded as the main adsorption sites for radionuclide attachment [3]. This study aimed at investigating two processes: retardation of Np(V) on the bentonite colloids and granitic rock and the effect of the stable and mobile bentonite colloids on the migration of Np(V) in intact and crushed granitic rock columns.
The materials used in this study were colloids prepared from MX-80 Volclay type bentonite (76% montmorillonite) and Kuru Grey granite. Np(V) sorption on these materials under stagnant conditions was studied as a function of pH, solid concentration, time, and Np(V) concentration. The sorption experiments as a function of pH (3-11), were performed at a constant Np(V) concentration of 10-6 M. The sorption isotherms as a function of Np(V) concentration were conducted at concentration from 10-9 to 5·10-6 M at pH 8, 9, and 10. Solid concentrations were 0.08 g/L and 0.8 g/L for colloids and 40 g/L for granite. The samples were prepared by adding a small aliquot of colloid stock solution or crushed granite, Np-237 tracer and the background electrolyte in 20 ml polypropylene vials. The solution was buffered to the desired pH and after one week equilibration time the solid phase was separated from the liquid by centrifugation and 1 ml aliquots were taken immediately for liquid scintillation counting (Perkin Elmer Tri-Carb 3100 TR or Quantulus liquid scintillation counter). All the batch sorption studies were conducted in 10 mM NaClO4 either in carbonate-free N2-atmosphere (bentonite colloids, 0.08 g/L) or under ambient air conditions (granite and bentonite colloids 0.8 g/L).
The effect of bentonite colloids on Np(V) migration was studied in column experiments, where the column material was either crushed granite (grain size 0.01-0.1 mm) or an intact drill core of the Kuru Grey granite. The crushed granite column diameter was 1.5 cm and the length 15 cm. Drill core columns were constructed from Kuru grey granite cores which were placed inside a tube to form a flow channel (L = 28 cm, w = 4.4 cm) representing an artificial fracture formed by the 0.5 mm gap between the core and the tube [3]. In the experiments, colloid solution was injected into the water flow and the colloid breakthrough was detected by photon correlation spectroscopy (PCS) measurements. The column experiments were performed under ambient air conditions in 10 mM NaClO4 solution using flowrates of 1.5 mL/h, 0.8 mL/h, and 0.3 mL/h. The Np-tracer was injected into the flow, through an injection loop of known volume. The flow conditions in the columns were determined using chloride (36Cl-) as a conservative tracer. The effect of bentonite colloids on Np(V) transport at pH 8 and pH 10 was determined in the absence and presence of colloids (0.7 and 0.9 g/L). The colloid concentration in the collected fractions was determined by PCS and the Np(V) concentration was determined after PCS measurements from the same samples by liquid scintillation counting.
Np(V) adsorption onto MX-80 bentonite colloids and crushed Kuru Grey granite in 10 mM NaClO4 is shown as a function of pH in Figure 1a and as a function of Np concentration in Figure 1b. Sorption onto colloids was rather weak (20%) at pH 8 and higher adsorption occurred only above pH 10. According to the pH-edge results, the sorption isotherms for bentonite colloids are as expected, linear and the slopes are close to one another. The weak sorption of Np(V) on the colloids indicates that Np(V) will be mobilized as a neptunyl cation in solution. Despite the low uptake of Np(V) by the bentonite colloids, the obtained column results show that Np(V) breakthrough from the granite columns is enhanced in the presence of colloids (Figure 2).