Technical Report NTB 03-01

Grimsel Test SiteInvestigation Phase VThe CRR final project report series I: Description of the Field Phase – Methodologies and Raw Data

The Colloid and Radionuclide Retardation Experiment (CRR) is dedicated to improve the understanding of the in situ retardation of colloid-associated, safety-relevant actinides and fission products in the vicinity of the Engineered Barrier System (EBS)/host rock interface. In addition to a series of in situ dipole experiments that were carried out at the Grimsel Test Site (GTS), the project partners, namely ANDRA (F), ENRESA (E), FZK-INE (D), JNC (J), USDOE/Sandia (USA) and Nagra (CH), funded an extensive programme of laboratory and modelling investigations. The aims of CRR were: examination of the in situ migration of bentonite colloids in fractured rocks, investigation of the interactions between safety relevant radionuclides and bentonite colloids in the laboratory and in situ and, in addition, testing of the applicability of numerical codes for representing colloid-mediated radionuclide transport.

The present report is the first of a quadruplet of final project reports that summarise the findings of the CRR project. In addition to this field report, the series includes laboratory and modelling reports along with a final, synthesis report. This report summarises and discusses the results of the field investigations that were carried out in 2001 and 2002 as part of the overall CRR project.

The overall concept behind CRR is based on the fact that, in most high-level radioactive waste repository designs, the waste is packed in massive metal canisters which are surrounded by a large volume of bentonite clay (collectively known as the Engineered Barrier System, or EBS). The canisters will slowly degrade and eventually fail, releasing some radionuclides, most of which are expected to be retained and to decay within the bentonite. However, it is conceivable that erosion of the bentonite at the EBS/host rock interface will produce bentonite colloids and that a limited amount of radionuclides escaping the EBS may become associated with these colloids and migrate through water conducting features in the geosphere towards the biosphere.

The central part of the CRR project was a series of dipole tracer tests that were carried out in a well-defined shear zone, in which dipole flow fields of 2.2 and 5 m length were generated. Preliminary tracer tests were performed with uranine, followed by tests with bentonite colloids and homologue elements for the tri- and tetravalent actinides (Tb for Am, Hf and Th for Pu, respectively). The tests culminated in the injection of the final tracer cocktails containing different isotopes of Am, Np, Pu, U, Tc, Th, Cs, Sr and I in the absence and presence of bentonite colloids.

The field installations consisted of several on-line measurement devices such as a downhole uranine detection device for the determination of the tracer input functions, a High Purity Germanium (HPGe) detector for γ-spectrometric measurements as well as a Laser Induced Breakdown Detector (LIBD) and a Photon Correlation Spectroscopy (PCS) apparatus for on-site colloid detection. The analytical techniques that were used off-site consisted of α-/γ-spectrometry and ICP-MS (Inductively Coupled Plasma-Mass Spectrometry) measurements for radionuclide detection as well as of Single Particle Counting (SPC) for the determination of the different colloid size classes. The interaction of the strongly sorbing tri- and tetravalent actinides with the equipment was avoided by producing as many as possible of those parts of the in situ equipment that were in direct contact with the tracers in PEEK (an inert plastic).

The natural colloid background of the groundwater in the experimental shear zone showed an average colloid diameter around 200 nm and a stable colloid concentration around 5 μgL-1. Increased colloid concentrations observed temporarily at the beginning of the experiments were most likely due to mechanical stress induced by pressure pulses generated during installation of the test setup. The four different colloid detection techniques, namely LIBD, ICP-MS1 , PCS and SPC, produced internally consistent breakthrough data of the injected bentonite colloids. The bentonite colloids arrived slightly earlier than did the conservative dye uranine and the recovery was about 90%. Filtration effects varied depending on the colloid size and measurement technique employed and, as such, require further investigation.

Homologue pre-tests proved to be very useful for the prediction of the in situ behaviour of triand tetravalent actinides. In the absence of bentonite colloids, a clearly lower recovery was found for the homologues than when injected together with bentonite colloids and the peak maxima of the homologue breakthrough were slightly shifted to earlier arrival times compared to that of the uranine.

The tracer cocktail composition for the final tracer injections covered the entire range of oxidation states from -I to VI and was decided based on the results of laboratory experiments, the kinetics of redox reactions and practical constraints on the in situ use of these elements. The preparation of an injection cocktail which contains tri- and tetravalent actinides proved to be problematic, as shown by the presence of a variable colloidal fraction for Am, Pu and Th, even in the absence of bentonite colloids. However, the injection cocktail, which included bentonite colloids, showed high colloid association and long term stability for the tri- and tetravalent actinides with the bentonite colloids, indicating that a significant proportion of the radionuclides were associated with the added bentonite colloids.

In the first run (without bentonite colloids), the tri- and tetravalent actinides Am, Th and Pu displayed lower recovery, less tailing and a peak time which was about 10 minutes earlier than U, Np and I (which is assumed to behave in a conservative fashion), indicating that a fraction of these actinides was transported in a colloidal state. With regard to the varying colloid content in the injection cocktail, the source of these colloids cannot yet be uniquely defined (homogeneous- or heterogeneous radiocolloids) and artifacts, for example, during cocktail preparation, cannot yet be ruled out completely.

With the addition of bentonite colloids, an increased recovery of Am, Pu and Th compared to the first run was observed. The shape of the breakthrough curves did not change significantly as the peak in the first experiment was also affected by a colloidal fraction. Only about 1% of the Cs was colloidally transported which implies that 90% of the initially colloid bound Cs in the injection cocktail (10% Cs in the injection cocktail was in colloidal form) desorbed during migration.

Finally it should be noted that the field experiments constitute only a part of the overall CRR project and interpretation and transfer of these data needs to be carried out taking into account the results of the laboratory experiments along with the effects of site groundwater chemistry, very short test duration and other technical constraints.