Bentonite rock interaction experiment: A hydro-structural-mechanical approach
Introduction
The description, characterisation, conceptualisation and modelling of different geological systems is of uttermost importance to our sustainable development because geological systems play a fundamental role in our society and may impact significantly upon it. Examples are the potential of a geological system to act as reservoir for geofluids (e.g. Rutqvist, 2012), or to be used as a repository for nuclear waste (SKB, 2011; Finsterle et al., 2019). Geological systems are generally complex and their understanding thus requires a multidisciplinary, holistic approach that takes account of all their components, including the composition, physical properties and reactivity of fluids therein (e.g. Zhao et al., 2014) and intrinsic or external factors that steer the bulk mechanical behaviour of the system under study (Olsson and Barton, 2001; Cosgrove et al., 2006; Lönnqvist and Hökmark, 2015).
Following the identification and first-order description of the geological system of interest, a detailed characterisation of the fundamental processes regulating it and steering its behaviour through time and in space has therefore to be performed. This has to be done in conjunction with the description of its geometrical framework, such as its size and shape, as well as the presence and characteristics of any anisotropy it may contain, like, for example, textural and lithological variations, diffuse deformation zones or localised discontinuities such as fractures and faults (e.g. Viola et al., 2009; Saintot et al., 2011). For fractures and faults, size (Cosgrove et al., 2006; Munier, 2010), surface geometry (Lönnqvist and Hökmark, 2015) and related properties (hydraulic, e.g. Chesnaux et al., 2009; Benedek and Molnár, 2013; mechanical, Olsson and Barton, 2001, and thermal, Tsang et al., 2009) have to be defined and accounted for. Only such an integrated approach can lead to the comprehensive definition of a well-constrained conceptual and deterministic model.
In this paper we describe the experimental test site of the now completed experiment “BRIE” - Bentonite Rock Interaction Experiment (see e.g. Fransson et al., 2017; Finsterle et al., 2019) and document the detailed characterisation of the investigated geological system carried out by integrating structural geology, hydrogeology and rock mechanical studies. This integration was necessary to produce full geometrical descriptions of the physical system and to increase our understanding of the chosen experimental site, while aiming at the reliable upscaling of the results for a meaningful modelling phase. The quality of the produced model stresses the importance of this holistic approach and the benefits that derive from a multifaceted input. The large range of possible values to be used during modelling (e.g., the mechanical properties to be fed into the modelling of the mechanical response of a rock mass to applied stresses, such as fracture normal stiffness) points to the great need for investigations and standard procedures that can help researchers to better define what is to be considered as sound and reasonable numerical inputs.
BRIE was performed between 2010 and 2014 at the Äspö Hard Rock Laboratory (Äspö HRL) in Sweden (Fig. 1). The purpose of the experiment was to investigate the hydraulics of the "crystalline basement rock - bentonite" system. The experiment was conceived in the framework of the research efforts for a future deep underground repository for high-grade nuclear waste. The multibarrier Swedish concept for underground nuclear waste storage (e.g. SKB, 2011) includes a copper canister and a bentonite buffer (engineered barriers) and the crystalline rock (the natural barrier).
The main focus of this paper is to describe the multidisciplinary scientific approach that was implemented to study in detail the Äspö HRL tunnel that was dedicated to the BRIE experiment (TASO tunnel; Fig. 1a). In addition to outcrop mapping, we relied on the analysis and logging of two fully retrieved vertical boreholes, between 3 and 3.5 m long and 300 mm in diameter, that were core-drilled into the tunnel floor. Following the retrieval of the cores, bentonite parcels were installed in the two boreholes as part of the experiment (Fig. 2a), reproducing the set-up of the final disposal configuration. The system was constantly monitored by means of a range of sensors. The parcels were eventually extracted and studied in detail together with the surrounding rock (Fig. 2b).
The performed geological and hydrogeological investigations have resulted in the conceptualisation of the geometric framework of the experimental site. The latter accounts for the local structural framework as well as for the main water-bearing features identified in the tunnel and in the lower sections of the boreholes (Fransson et al., 2017). The TASO tunnel was studied in great detail by means of a multidisciplinary approach specifically focusing on structural geological mapping, aiming to provide as many constraints as possible to the subsequent phase of rock mechanical modelling (Fransson et al., 2019). This approach was proven to be necessary since the hydrogeological observations on both rock and bentonite parcels indicated localised, transient changes of hydraulic flow during the execution of the experiment. Structural mapping was therefore done aiming at strengthening the cross-linking between hydrogeological observations and the overall rock mass structural framework, whereas rock mechanical modelling assisted the semiquantitative analysis of the studied system.
The hypothesis underpinning our study was that the changes of hydraulic flow observed between the host rock and the bentonite parcels inserted within the boreholes are due to the presence of specific natural geological structures in addition to the effects caused by induced stresses, such as those due to the excavation, drilling of boreholes and swelling of the inserted bentonite parcels. Natural geological structures, when favourably oriented and characterised by specific parameters or characteristics, would increase the likelihood of mechanical deformation, for example by amplifying its effects.
This study contributes to the field of engineering geology by demonstrating that an integrated (hydro-structural-mechanical) parametrisation and modelling approach can describe complex behaviours that would otherwise be impossible to characterise with more limited data sets (e.g. hydro-structural or structural-mechanical input only). We describe fracture sets, their scale-independent, self-similar behaviour, and their properties in relation to rock mechanical modelling and provide a semiquantitative analysis linking rock mechanical modelling and hydrogeological characterisation. These are all key aspects for a comprehensive description of a rock volume and its mechanical modelling and conceptualisation. Details on the rock mechanical modelling that was based on the parametrisation presented herein is found in Fransson et al. (2019).
The TASO tunnel is located at a depth of approximately 420 m b.s.l. (Fig. 1b, red circle). The bedrock at the Äspö HRL dates back to the Svecokarelian orogeny, 1.9–1.8 Ga ago, and is part of the Fennoscandian Shield (Viola et al., 2009). The main rock types at the Äspö HRL are gneisses (granodioritic and quartz monzodioritic). The rock also contains veins and dikes as well as fine-grained granite (minor bodies), composite intrusions and pegmatite.
The first step of the BRIE experiment was the identification of a suitable site. The TASO tunnel, which is a several meter long experimental niche (Fig. 1a), was selected based on the information from five vertical and 3 m long boreholes (KO0014G01, KO0015G01, KO0017G01, KO0018G01 and KO0020G01, all 76 mm in diameter, Fig. 1c). In the next step, 14 vertical boreholes (same diameter) were drilled and investigated (Fig. 1c). Ten of these boreholes reached a depth between 3.0 and 3.5 m from the tunnel floor. Due to expected bad quality rock close to a deformation zone (see the structural features perpendicular to the tunnel in Domain A, Fig. 4), four of the boreholes were drilled to a depth of only 1.5 m. Fig. 1a shows the four additional horizontal boreholes that were drilled in the walls of the tunnel (all the same diameter and a length of 10 m). During a later investigation stage, the two central floor boreholes, 17G01 and 18G01, Fig. 4, were overcored leading to two 300 mm boreholes. These were further investigated by geological mapping, structural logging and hydraulic testing.
After final overcoring, bentonite parcels were placed into the 3.5 m long borehole 17G01 and 3 m long 18G01 (see the parcels in Fig. 2a and their locations in the tunnel in Fig. 4). Each parcel had a cylindrical shape (298 mm in diameter) and was approximately 3.4 m and 2.9 m in length for 17G01 and 18G01, respectively. For the installation (Fig. 2a), bentonite blocks (including instruments for monitoring) were placed on a central tube with a bottom plate. A pillar was subsequently placed between the tunnel roof and the upper part of the parcel (Fig. 2a).
The last two stages of the experiment consisted in the continuous monitoring of the installed sensors and final extraction and analysis of the parcels and of the physically matching surrounding rock. Extraction of bentonite and rock blocks was done by using a combination of stitch drilling (at the periphery of the blocks) and wire-line sawing (at the base) resulting in packages having a diameter of approximately 0.7 m and a depth similar to the depth of central boreholes (Fig. 2b). After extraction, the parcels were carefully inspected and analysed for their water content expressed as mass of water per mass of dry substance (see details in Fransson et al., 2017). In addition to water content plots, photos of the bentonite parcel surfaces were systematically taken to continuously document the spatial distribution of the wetting pattern of bentonite (Fig. 7 and Fig. 8).
Section snippets
Materials and methods
We used a three-pronged approach in this study (Fig. 3):
- 1)
Structural mapping of the TASO-tunnel and of the recovered 300 mm cores: This was necessary to produce geometrical and kinematic descriptions of the analysed deformation zones, to identify all existing fracture sets and to generate diagnostic structural tools to assist with the characterisation of individual fracture properties. As an example, finite displacement fractures whose trace can be followed all around the tunnel perimeter (also
Discussion
Deformation zones and fractures are of major importance to engineering and to the characterisation of the mechanical- and hydraulic behaviour of rock. Fracture sets and conformity between scales (Section 3.1) and, fracture properties in relation to rock mechanical modelling (Section 3.2) are therefore considered to be key topics. To highlight the usefulness of a hydro-structural-mechanical approach, the last section (Section 3.3) summarises the semiquantitative analysis that allows to
Concluding remarks
This paper presents the multidisciplinary and trans-stadial investigation and modelling approach used during the implementation of experiment BRIE (Fransson et al., 2017). Construction- and experimental stages related to the progressive excavation of the TASO-tunnel and execution of BRIE provided inputs for rock mechanical modelling (Fransson et al., 2019) (modelling sequence) of induced stresses and step-wise documentation of hydrogeological observations. Detailed structural mapping was of
CRediT author statement
Åsa Fransson: Conceptualization, Methodology, Investigation (focus on hydrogeology), Writing- Original draft preparation (focus on hydrogeology, engineering geology and parametrization), Writing- Reviewing and Editing (in general and focusing on engineering geology and parametrization). Giulio Viola: Conceptualization, Investigation (focus on structural mapping), Visualization, Writing- Original draft preparation (focus on structural mapping), Writing- Reviewing and Editing (in general).
Declaration of Competing Interest
None.
Acknowledgements
The authors express their gratitude to the Swedish Nuclear Fuel and Waste Management Co., SKB and Äspö HRL for funding and hosting the previously performed experiment, BRIE. Special thanks go to Mattias Åkesson, Clay Technology, Margareta Lönnqvist, Clay Technology and Johan Thörn, Bergab and others involved in the experiment. Financial support (grant number 2017-00360) from the Swedish Research Council, Formas, to the study presented in this paper is also greatly appreciated. The Editor, Prof.
References (29)
- et al.
Combining structural and hydrogeological data: Conceptualization of a fracture system
Eng. Geol.
(2013) - et al.
Regional fracture network permeability using outcrop scale measurements
Eng. Geol.
(2009) - et al.
Distinguished multiple events of fracture mineralisation related to far-field orogenic effects in Paleoproterozoic crystalline rocks, Simpevarp area
SE Sweden. Lithos.
(2009) - et al.
Rock mechanical modelling of the Bentonite Rock Interaction Experiment, Äspö Hard Rock Laboratory
Sweden. Int. J. Rock Mech. Min. Sci.
(2019) - et al.
Multi-stage structural and kinematic analysis of a retrogressive rock slope instability complex (Preonzo, Switzerland)
Eng. Geol.
(2019) - et al.
Seismic response of the Lengzhuguan slope caused by topographic and geological effects
Eng. Geol.
(2020) - et al.
In situ stress states at KURT, an underground research laboratory in South Korea for the study of high-level radioactive waste disposal
Eng. Geol.
(2019) - et al.
New constraints on 1.7 Gyr of brittle tectonic evolution in southwestern Finland derived from a structural study at the site of a potential nuclear waste repository (Olkiluoto Island)
J. Structural Geol.
(2014) - et al.
An improved model for hydromechanical coupling during shearing of rock joints
Int. J. Rock Mech. Min. Sci.
(2001) - et al.
Three-dimensional analyses of slip distributions on normal fault arrays with consequences for fault scaling
J. Structural Geol.
(1996)
Solute transport in a single fracture with time-dependent aperture due to chemically medicated changes
Int. J. Rock Mech. Min. Sci.
Geological characteristics of deformation zones and a strategy for their detection in a repository
Conceptual uncertainties in modelling the interaction between engineered and natural barriers of nuclear waste repositories in crystalline rocks
Geol. Soc. Lond. Special Pub.
Geological and hydrogeological reference conditions for rock engineering: an example for grouting design at Äspö Hard Rock Laboratory, Sweden. In ISRM AfriRock-Rock Mechanics for Africa, Cape Town, South Africa
Int. Soc. Rock Mech. Rock Eng.
Cited by (10)
Experimental analysis on applicability degree and influence factors of MWD parameters for anchor drilling
2023, Measurement: Journal of the International Measurement ConfederationAnalysis and selection of measurement indexes of MWD in rock lithology identification
2023, Measurement: Journal of the International Measurement ConfederationPrecambrian fault reactivation revealed by structural and K-Ar geochronological data from the spent nuclear fuel repository in Olkiluoto, southwestern Finland
2022, TectonophysicsCitation Excerpt :The likelihood of reactivations to occur is dependent on the prevailing stress field, the orientation of the fault and the strength of the fault surface (Sibson, 1985; Morris et al., 1996). Consequently, the geometry and the mechanical properties of geological brittle structures need to be systematically studied and taken into account as they may directly impact the planning and designing of excavations and reinforcements in underground construction projects (Hoek and Brown, 1980; Hudson et al., 2011; Fransson et al., 2019; Fransson and Viola, 2021). Underground spent nuclear fuel repositories, for example, require long-term tectonic stability and thus need to be built avoiding large brittle structures, since structural reactivation along a fault or a large fracture that intersects a spent nuclear fuel canister might jeopardize its integrity (e.g., Munier, 2006; McEwen et al., 2012).
Complexity Feedback of Individual Indexes and Composition Analysis of Specific Energy Index for Measurement-While-Drilling Technology of Roof Bolter
2024, IEEE Transactions on Instrumentation and MeasurementA New Close-Form Solution for Elastoplastic Seepage-Induced Stresses to Circular Tunnel with Considering Intermediate Principal Stress
2023, Rock Mechanics and Rock Engineering