Bentonite colloid diffusion through the host rock of a deep geological repository

https://doi.org/10.1016/j.pce.2006.04.021Get rights and content

Abstract

The determination of experimental data on all processes that contribute to the radionuclide (RN) migration within the barriers of a deep geological radioactive waste repository is required to evaluate the long-term behavior of the waste repository. A novel approach for the experimental estimation of bentonite colloid diffusion coefficients in granite is proposed, since no experimental data on this issue are available. The Rutherford backscattering spectrometry (RBS) technique was selected, because it is suitable for the study and the detection of gradients of heavy elements in a well-defined matrix and allows measuring diffusion coefficients on a micrometer scale. To follow the trail of the bentonite clay colloids within the granite by RBS analysis the colloids were traced with a heavy element. Complementary batch sorption experiments guaranteed that the Eu was an adequate tracer. By RBS analysis, Eu concentration gradients within the granite were clearly observed, and it was then proven that traced bentonite colloids were able to diffuse within the granite. The bentonite colloids diffusion coefficient was estimated to be ≈10−17 m2/s, that is a value three orders of magnitude lower than that measured for the Eu as solute (10−14 m2/s).

Introduction

Compacted bentonite is the main engineering barrier considered in a high level radioactive waste repository (HLWR) in granite. Bentonite is able to retard the migration of radionuclides (RN) to the geosphere because it has very low permeability, limits the water income to the waste, and presents high sorption capability for many radionuclides. Nevertheless, for a long-term performance assessment of a HLWR, it is necessary to evaluate all the possible processes occurring at the bentonite barrier that may affect the RN transport. In particular, this work focuses on the possible contribution of bentonite colloids generated at the host rock/bentonite interface of the repository. Colloids are particles with diameters ranging from 1 nm to 1 μm that have high surface area that may affect contaminant transport in the subsurface (Abdel-Salam and Chrysikopoulos, 1995a, Abdel-Salam and Chrysikopoulos, 1995b, Tatalovich et al., 2000, James et al., 2005). Because of their high sorption capability colloid-borne migration can be particularly relevant for low solubility radionuclides, that otherwise could be immobilized in the rock surface (Swanton, 1995).

Prior to a study of the colloid transport mechanisms, the stability and mobility of colloids in a given environment are the first issues to be evaluated. In addition, the RN sorption irreversibility must be proved in order to assess the relevance of colloid-mediated radionuclide transport (Miller et al., 1994).

Recent studies have demonstrated that, in favorable conditions, colloids could be detached at this interface and that in low ionic strength and alkaline waters could be stable and mobile in the environment (Missana et al., 2003).

The transport of colloids in fractured rocks is expected to be controlled by advection and dispersion in water if a fracture network exists, and it is generally tackled by a theoretical point of view (Ibaraki and Sudicky, 1995, James and Chrysikopoulos, 1999, Skagius, 1992). The theoretical description of colloid transport in a fractured medium usually includes surface retention in fracture walls and matrix diffusion, as possible mechanisms that eliminate colloids from the aqueous phase. All phenomena that eliminate colloid from the aqueous phase are usually included in the term filtration. The scarce experimental colloid transport studies, performed by batch or column laboratory experiments (Ryan and Elimelech, 1996, Smith and Degueldre, 1993, Keller et al., 2004, Anders and Chrysikopoulos, 2005) or also in situ in a granitic massif (Möri et al., 2003), emphasized the need of identifying and quantifying the colloid filtration mechanisms to predict the colloid-mediated radionuclide transport, pointing out the strong lack of experimental data. Great uncertainties still exist on the actual contribution of these mechanisms (Ryan and Elimelech, 1996). In that sense, obtaining experimental parameters, such as colloid diffusion coefficients, for inclusion in theoretical transport codes can overload the main filtration mechanisms. One of these parameters is precisely the diffusion coefficient of colloid in the rock matrix.

The aim of the present work was, thus, to assess whether bentonite colloids are able to diffuse within the granite. A methodology based on the ion nuclear beam technique Rutherford backscattering spectrometry (RBS) was selected to perform these colloid diffusion studies. The RBS is a technique widely applied in material science to study diffusion processes at the micrometer scale. RBS is not usually applied to study geological materials because of their high heterogeneity, but some applications in earth sciences are reported in the literature (Dran et al., 1988, Ryan, 2004, Toulhoat et al., 1996, Trocellier et al., 1999). In previous works an adequate characterization of granite was carried out to study diffusion processes within the rock (Alonso et al., 2003a, Alonso et al., 2003b).

One should note that in order to determine bentonite colloid diffusion within the granite by the RBS, tracing the colloids was necessary. The clay and the granite matrix are, in fact, mainly composed of light elements. Europium was examined as a possible tracer because of being a heavy element, thus imparting RBS sensitivity to the system. Complementary batch sorption experiments of Eu on bentonite colloids were required previous to diffusion studies. Eu presented almost 100% sorption onto the clay colloids and practically no desorption in the presence of granite was observed within the time scales of diffusion experiments. Thus Eu was considered a suitable tracer for these experiments.

Then RBS spectra performed on granite samples, with Eu as a solute and adsorbed onto bentonite colloids, were compared. Undoubtedly, Eu diffusion profiles were detected. The demonstration of Eu as a suitable tracer of bentonite colloids enables one to guarantee that the determined Eu diffusion coefficients can be taken as an estimation of a colloid diffusion coefficient within a crystalline rock. This is the first time that such a parameter was experimentally determined.

Section snippets

Materials

A spanish granite was selected for these studies. It is granodiorite type and its mineral composition is basically quartz (33–35%), plagioclase (29–32%), K-feldspars (26–28%), muscovite (5–6%) and biotite (2–3%). Additionally it presents corderite, andalucite, illmenite and zircon (Buil, 2002).

The clay used to prepare the bentonite colloids is a Spanish clay (FEBEX bentonite), basically comprised of smectite (93 ± 2%), with small percentages of quartz (2 ± 1%), plagioclase (3 ± 1%), cristobalite (2 ± 

Sorption studies

Fig. 2 shows the logarithm of distribution coefficient (in ml/g) of the Eu adsorbed onto the bentonite colloids as a function of time ([Eu] = 4.6 × 10−6 M, pH 5 and ionic strength 10−3 M). It can be appreciated that, within experimental error, after only 5 min the log (Kd) value measured was 6.0 ± 0.4 ml/g, which is equivalent to a percentage of Eu adsorbed onto the colloids up to 98 ± 2% as can be seen in the inset of Fig. 2. The Eu adsorption on the tube surfaces, under these conditions, was accounted for

Conclusions

The diffusion of bentonite colloids within granite was experimentally studied by means of a novel approach applying the RBS technique, and compared with the diffusion of Eu as solute. The methodology carried out in this work allowed the observation of Eu profiles within the granite in both cases that presented time dependence, in agreement with a diffusion process. The diffusion coefficient calculated for the Eu as solute (10−14 m2/s) is in the same range as those reported in the literature,

Acknowledgements

This work was partially supported by the EU within the FUNMIG (Fundamental Processes of Radionuclide Migration) project (Ref: FP6-516514), by the INFN-LNL within the MIRACOL (Migration of Radionuclides and Colloids) project (USP Ref. 0003446/04) and in the framework of the CIEMAT-ENRESA association. The authors are grateful to Prof. Stucki and D. Ghosh for their helpful revisions that significantly improved the final manuscript.

References (40)

  • G. Plancque et al.

    Europium speciation by time-resolved laser-induced fluorescence

    Analytica Chimica Acta

    (2003)
  • C.G. Ryan

    Ion beam microanalysis in geoscience research

    Nuclear Instruments and Methods in Physics Research B

    (2004)
  • J.N. Ryan et al.

    Review: Colloid mobilization and transport in groundwater

    Colloids Surfaces A

    (1996)
  • P.A. Smith et al.

    Colloid-facilitated transport of radionuclides through fractured media

    Journal of Contaminant Hydrology

    (1993)
  • S.W. Swanton

    Modelling colloid transport in groundwater. The prediction of colloid stability and retention behavior

    Advances in Colloid and Interface Science

    (1995)
  • P. Trocellier et al.

    Nuclear microprobe applications to radioactive waste management basic research

    Nuclear Instruments and Methods in Physics Research B

    (1999)
  • J. Van der Lee et al.

    Module-oriented modeling of reactive transport with HYTEC

    Computers and Geosciences

    (2003)
  • T. Yamaguchi et al.

    Effective diffusivity of the uranyl ion in a granite from Inada, Ibaraki, Japan

    Journal of Contaminant Hydrology

    (1997)
  • A. Abdel-Salam et al.

    Analysis of a model for contaminant transport in fractured media in the presence of colloids

    Journal of Hydrology

    (1995)
  • A. Abdel-Salam et al.

    Modeling of colloid and colloid-facilitated contaminant transport in a two-dimensional fracture with spatially variable aperture

    Transport in Porous Media

    (1995)
  • Cited by (21)

    • Dynamics and sources of colloids in shallow groundwater in lowland wells and fracture flow in sloping farmland

      2019, Water Research
      Citation Excerpt :

      Low percentages of colloid in the fracture flow and lowland wells were ascribed to colloid diffusion into dead-end voids, such as the soil matrix or fracture apertures, during the subsurface transport process. This matched the observations from other studies, e.g., Alonso et al. (2007), Mondal and Sleep (2012), Zvikelsky and Weisbrod (2006). The much smaller maximum size of suspended particles and the higher cumulative particle percentage from colloids to coarse particles (2–10 μm) in the well water relative to the fracture flow implies a greater retardation of large particles (>10 μm) during their transport in the catchment.

    • The effect of colloids on nanofluidic power generation

      2018, Energy
      Citation Excerpt :

      Recently, Feng et al. exploited a single-layer MoS2 membrane with a homogeneous pore size of 10 nm, and the optimistic estimated power density could reach 106 W/m2 under parallelization at large scale [23]. In the application environment, colloidal particles can serve as transport vectors of diverse contaminants in surface water, such as seawater, lakes, rivers, and freshwater bodies [24–27]. The diameter of colloidal particles ranges from 1 to 100 nm and they have a large surface/volume ratio.

    • <sup>137</sup>Cs transport in crushed granitic rock: The effect of bentonite colloids

      2018, Applied Geochemistry
      Citation Excerpt :

      The transport of radionuclides through the rock matrix in the presence of bentonite colloids has been studied extensively with experimentation tending to focus on the role of bentonite colloids and radionuclide transport in the natural barrier (Missana et al., 2008; Delos et al., 2008). It has been established that the presence of bentonite colloids may enhance the mobility of radionuclides; the radionuclides form radiocolloid complexes with the colloid particles which diffuse into the rock matrix and mediate the migration of formerly immobile contaminants (Alonso et al., 2007, Kretzschmar and Schäfer, 2005). Missana et al. (2008) observed that bentonite colloids significantly influence the retardation behaviour of multivalent actinides; the migration of radionuclides (152Eu, 239Pu) was found to be strongly mediated by bentonite colloids and, in the presence of colloids, the recovery of Eu and Pu was greater than that in systems without the presence of bentonite colloids.

    • Parallel-plate fracture transport experiments of nanoparticulate illite in the ultra-trace concentration range investigated by Laser-Induced Breakdown Detection (LIBD)

      2017, Colloids and Surfaces A: Physicochemical and Engineering Aspects
      Citation Excerpt :

      Especially fractures serve as preferential conduits for colloids. This is due to (i) large apertures that enable high flow velocities compared to the surrounding rock matrix, (ii) mostly unfavorable conditions for attachment due to negatively charged surfaces of colloids and fracture walls under the circumneutural pH range expected in natural groundwater systems and (iii) negligible matrix diffusion of colloids in contrast to solutes [3,15–17]. Colloid transport through both artificial and natural fractures has been studied extensively in the past [18–20].

    View all citing articles on Scopus
    View full text