Elsevier

Chemosphere

Volume 269, April 2021, 128674
Chemosphere

In-situ X-ray fluorescence to investigate iodide diffusion in opalinus clay: Demonstration of a novel experimental approach

https://doi.org/10.1016/j.chemosphere.2020.128674Get rights and content

Highlights

  • Novel experimental set-up for the study of iodide ion diffusion in compacted clay.

  • In-situ downhole quantification of trace metal amounts.

  • Visualization of the impact of small-scale heterogeneities on transport properties.

  • Long-term and large-scale iodide diffusion data acquisition.

  • Merging of mineralogical, chemical and structural elements for data interpretation.

Abstract

During the last two decades, the Mont Terri rock laboratory has hosted an extensive experimental research campaign focusing on improving our understanding of radionuclide transport within Opalinus Clay. The latest diffusion experiment, the Diffusion and Retention experiment B (DR-B) has been designed based on an entirely different concept compared to all predecessor experiments. With its novel experimental methodology, which uses in-situ X-ray fluorescence (XRF) to monitor the progress of an iodide plume within the Opalinus Clay, this experiment enables large-scale and long-term data acquisition and provides an alternative method for the validation of previously acquired radionuclide transport parameters.

After briefly presenting conventional experimental methodologies used for field diffusion experiments and highlighting their limitations, this paper will focus on the pioneer experimental methodology developed for the DR-B experiment and give a preview of the results it has delivered thus far.

Introduction

Due to their very low hydraulic conductivities, large reactive surfaces available for radionuclide sorption and capacity to self-seal, argillaceous rocks are widely considered as potential host rocks for radioactive waste repositories. Several independent research programs dedicated to study the overall suitability of different indurated clayrocks for radioactive waste disposal are currently underway (e.g. Boom Clay in Belgium, Opalinus Clay (OPA) in Switzerland, Callovo-Oxfordian argillite in France and Boda claystone in Hungary) (Van Loon et al., 2012). These research programs focus on the study of the mechanisms governing solute transport through the porous structure of clayrocks. As a result of the micro-structure of clay minerals, molecular diffusion is expected to be the dominant solute transport mechanism in clayrocks under naturally occurring conditions (Van Loon and Soler, 2004; Mazurek et al., 2011; Altmann et al., 2012). In soft, plastic clays however some advection may occur (Aertsens et al., 2013). Radionuclide diffusion within clays has been extensively studied both in laboratory and in field experiments (Aertsens et al., 2013; Van Loon and Müller, 2014; Van Loon et al., 2004a, 2004b, 2005, 2007; Gimmi et al., 2007, 2014; Bazer-Bachi et al., 2006; Tevissen et al., 2004; Glaus et al., 2008; Wigger and Van Loon, 2017; Wersin et al., 2004). These experiments, additionally to improving our understanding of the processes governing solute diffusion and the influential parameters, have led to the creation of a unique database regrouping the transport parameters of safety-relevant radionuclides within Opalinus Clay (Leupin et al., 2017) as well as in other clayrocks. These parameters are of paramount importance as they are essential inputs for safety and performance assessments of deep geological repositories for radioactive waste.

Laboratory experiments, where solute diffusion is performed on small-scale samples, have generally been the method of choice for the determination of relevant radionuclide transport parameters due to their relative simplicity to execute and the fact that they can be performed under controlled conditions. Thereby, diffusion processes can be carefully studied and values for the diffusion coefficient and the solute/rock interactions can be determined. However, laboratory diffusion experiments have their drawbacks, such as the limited clay sample size (which might not be representative of the whole lithological unit), the relatively short time-scales which can be studied and, most notably, the problems related to sample preparation. Indeed, overburden pressure release when the sample is extracted from the original bedrock may cause irreversible changes to the micro-structure of the clay sample (e.g. microfracturing along bedding, increase of porosity), which in turn can affect the transport-relevant porosity and, consequentially, the measured diffusion coefficients. Furthermore, sudden exposure of the sample to Earth’s atmosphere, a strongly oxic environment, may alter the sample. Finally, during so-called cell diffusion experiments, which is the most common laboratory technique employed to determine transport parameters, the high water-to-rock ratio employed can lead to a “bleaching” of the sample and thus cause significant changes in the sample’s original chemical composition (Van Loon et al., 2012; Leupin et al., 2017). For these reasons, transport parameters obtained from laboratory experiments may not be perfectly representative of those existing under actual in-situ conditions. Therefore, field experiments, which are more representative of the actual in-situ conditions, are necessary complements to laboratory diffusion experiments.

The Mont Terri rock laboratory situated in the Swiss Jura mountains is a unique platform for improving the understanding of the transport and retention properties of indurated clays. During the past two decades, it has hosted a large number of different diffusion experiments. All of them, apart from DR-B, were based on similar experimental methodologies (Leupin et al., 2017).

The conventional experimental procedure followed by field diffusion experiments consists of the following steps: a borehole is drilled, a packer is installed in order to hydraulically isolate a section of the freshly drilled borehole and a circulation loop connecting that packed-off section to the surface is installed. A solution containing one or several radio-tracers (e.g. tritiated water (HTO), 125I, 22Na+, 137Cs+, 133Ba+, …) is then circulated through this hydraulically packed-off section of the borehole from where the tracers diffuse into the surrounding rock. In order to avoid advective transport, the pressure head of the circulating fluid is always kept close to the pore pressure. As the cocktail solution is pumped through the circulation loop, the tracer concentrations are constantly being monitored. When the experiment reaches the end of its designed operation time, the whole original borehole is over-cored. Thin slices of that drill-core are then prepared and the amount of radionuclides analysed using different analytical techniques (e.g. gamma-spectrometry, liquid scintillation counting, ICP-OES, …).

Thus, two independent sets of data are collected: (1) the temporal evolution of the tracer concentrations in the circulation loop and (2) the spatial distribution of the tracer profiles in the drill-core slices. These data sets enable the determination through modelling of (a) an effective diffusion coefficient and (b) an accessible porosity (for conservative tracers) or a sorption coefficient (for sorbing tracers).

In general, the models used to interpret the collected data solve the diffusion-sorption equation:Ctott=(DeC)where Ctot is the total solute concentration (including the sorbed fraction), C is the aqueous solute concentration, t is the time and De is the effective diffusion coefficient tensor. Models for sorbing tracers often assume a linear sorption isotherm (Van Loon and Soler, 2004; Gimmi et al., 2014), leading to a constant distribution coefficient Kd and a constant rock capacity factor α according to:α=ε+ρdKdandCtot=αCwith ε being the accessible porosity and ρd the dry bulk density of the rock. In this case, the decrease in tracer concentration in the injection interval depends directly on both De and α (ε for conservative tracers, i.e. Kd = 0), while transport distances in the rock depend on the magnitude of the apparent diffusion coefficient Da = De/α. Thus, both De and α can be determined for each tracer by simultaneously adjusting the model to both independent data sets. If sorption is non-linear or irreversible, the rock capacity factor α, and thus Da, is not a constant, but varies with time and space.

The transport parameters obtained from previously performed field experiments at Mont Terri, i.e. experiments FM-C, DI, DI-A1, DI-A2, DI-B, DR and DR-A were generally in good agreement with transport parameters obtained from laboratory experiments (Van Loon et al., 2012; Leupin et al., 2017). However, the experimental methodology employed for these experiments is subject to a few limitations:

  • The clay present in the immediate vicinity of the injection borehole has an altered structure as a result of the stresses and heat produced by the drilling procedure as well as the unavoidable exposure to oxygen. This region, commonly referred to as the borehole disturbed zone (BdZ), may therefore not have the same properties than virgin OPA regarding nuclide transport.

  • Due to the fault-prone nature of circulation loops, active maintenance is required.

  • Retrieving intact drill-cores represents a serious technical challenge as Opalinus Clay is a brittle material. There are therefore practical limits on the maximal volume of rock that can be safely over-cored. This limits the volume of rock that can be studied.

  • As radiotracers have to be sufficiently long-lived to be detectable over the experiment’s entire duration and radiological interferences have to be avoided when using a “cocktail” of radioactive tracers, the number of adequate radiotracers available for field experiments is limited. Furthermore, comparably high activities are required for experiments with durations of several years which may in turn lead to cumbersome licensing procedures.

These issues led to the development of an alternative experimental methodology which would not suffer from the same limitations and which would, ultimately, provide another mean to validate radionuclide transport parameters at the field scale.

Section snippets

The diffusion and retention experiment B (DR-B)

The aim of the DR-B experiment is to provide a mean to validate current diffusion process understanding as well as previously acquired transport parameters by using a novel methodology which enables longer diffusion periods ( 10 years) and larger rock volumes to be investigated.

Results & discussion

This section exclusively treats data collected in observation borehole BDR-B8. The decision to omit the XRF scans related to the other two observation boreholes in the present analysis boils down to the fact that, as of this date, the iodide breakthrough has only been observed in BDR-B8. This is a consequence of BDR-B5 and BDR-B6 being situated further from the injection interval.

Credit author statement

Max Jaquenoud: Investigation, Formal analysis, Data curation, Visualization, Writing, William T. Elam: Resources, Software, Tim Grundl: Resources, Software, Thomas Gimmi: Formal analysis, Andreas Jakob: Formal analysis, Senecio Schefer: Resources, Veerle Cloet: Investigation, Pierre De Cannière: Methodology, Luc R. Van Loon: Methodology, Resources, Olivier X. Leupin: Conceptualization, Methodology, Supervision, Project administration, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors would like to express their gratitude towards the Federal Agency for Nuclear Control (FANC), the National Cooperative for the Disposal of Radioactive Waste (NAGRA), the Nuclear Waste Management Organization (NWMO), the Radioactive Waste Management (RWM) and the Federal Office of Topography Swisstopo for their funding. We would also like to thank the Swisstopo personnel at St-Ursanne for providing excellent working conditions in the Mont Terri rock laboratory as well as the many

References (29)

  • L.R. Van Loon et al.

    Studying radionuclide migration on different scales: the complementary roles of laboratory and in situ experiments

  • M. Mazurek et al.

    Natural tracer profiles across argillaceous formations

    Appl. Geochem.

    (2011)
  • P. Wersin et al.

    Long-term diffusion experiment at Mont Terri: first results from field and laboratory data

    Appl. Clay Sci.

    (2004)
  • M. Aertsens et al.

    Overview of radionuclide migration experiments in the HADES underground research facility at mol (Belgium)

    Clay Miner.

    (2013)
  • Cited by (5)

    • Immobile crystallization of radioactive iodide by redox transformation of a low crystalline copper phase

      2022, Chemosphere
      Citation Excerpt :

      With the increase of nuclear power plants and medical treatments, the potential risks of radionuclide contamination within the environment are now increasing. Among growing risks, radioiodine contamination has been particularly concerning for its great impacts on environment and human beings (Sato et al., 2011; He et al., 2014; Xing et al., 2015; Rose et al., 2015; Ohta et al., 2012; Braghirolli et al., 2014; Jaquenoud et al., 2021; Roulier et al., 2021). For example, some facilities that treat spent nuclear fuel in northern Europe are the main global sources of releasing 129I into seawater (He et al., 2014).

    View full text