Olle Olsson
Swedish Nuclear Fuel and Waste Management Co. (SKB)
Pl 300, S-572 95 Figeholm, Sweden

In 1986 SKB decided to construct the Äspö Hard Rock Laboratory in order to provide an opportunity for research, development and demonstration in a realistic and undisturbed underground rock environment down to the depth planned for the future deep repository. The focus of current and future work is on development and testing of site characterization methods, verification of models describing the function of the natural and engineered barriers and the development, testing, and demonstration of repository technology. The program has been organized so that all important steps in development of a repository are covered, in other words the Äspö HRL constitutes a "dress rehearsal" for the Swedish deep geological repository for spent fuel and other long lived waste.

Geoscientific investigations on Äspö and nearby islands began in 1986. Construction of the facility, which reaches a depth of 460 m below the surface, began in 1990 and was completed in 1995. The Operating phase of the Äspö HRL began in 1995 and is expected to continue for 15-20 years until the first stage of the development of the deep geological repository for spent is expected to completed.

The implementation of the Swedish deep repository for spent nuclear fuel will be made in stages with well defined milestones where SKB will need approval by the regulatory authorities and the communities concerned in order to continue with the next stage (Fig. 1). Feasibility studies are currently in progress and will before this stage has been completed have been made in 5-10 communities. The objective of the feasibility studies is to study the environmental, economical, and social impact of locating a repository in the community and to identify potential repository sites within the communities based on currently available material. Based on an integrated evaluation of the feasibility studies two sites will be selected for comprehensive geological site investigations. Site investigations are expected to start year 2000 at the earliest and have a duration of 4-5 years. One of these sites will then be selected for detailed characterization and construction of tunnels to repository depth. Then a license application will be submitted to start initial operation of the deep repository. During the initial operation stage about 10% or 400 of the total of about 4500 canisters with spent nuclear fuel will be deposited. Experience from several years of initial operation of the repository will be evaluated and included in the application for regular operation of the repository including deposition of the total amount of spent fuel from the Swedish nuclear program. Regular operation of the deep repository is expected to begin approximately 20 years from now.

In this process the role of the Äspö Hard Rock Laboratory is to provide input to the performance assessments that have to be supplied as part of each license application and to develop, test, and evaluate methods for site investigations, detailed investigations, repository construction, and deposition before they are applied within the deep repository program. The Äspö HRL should also provide experience and train staff in performing the various tasks within the deep repository program. Äspö HRL also offers the opportunity to test various aspects of repository performance during a long time, up to 20 years, and will hence provide valuable input to the evaluation made in conjunction with application for regular operation of the deep repository.

The Äspö HRL program has been organized so that all important steps in development of a repository are covered, in other words the Äspö HRL constitutes a "dress rehearsal" for the Swedish deep geological repository for spent fuel and other long lived waste. The focus of current and future work is on development and testing of site characterization methods and their ability to provide data for performance assessment, verification of models describing the function of the natural and engineered barriers, and the development, testing, and demonstration of repository technology.

The decision to construct the Äspö Hard Rock Laboratory was taken by SKB in 1986. The main objective was to provide an opportunity for research, development and demonstration in a realistic and undisturbed underground rock environment down to the depth planned for the future deep repository.

Geoscientific investigations on Äspö and nearby islands began in 1986. The bedrock at Äspö consists of approximately 1.8 Gyear old medium-grained granite and diorite with intrusions of younger fine-grained granites. Construction of the facility, which reaches a depth of 460 m below the surface, began in 1990 and was completed in 1995 (Fig. 2). The characterization and construction phases of the laboratory were used as a prediction-validation exercise which demonstrated that the investigations on the ground surface and in boreholes could provide sufficient data on most of the essential safety-related properties of the rock at repository level. The work has also included significant development of groundwater flow and geochemical models. Considerable experience has also been gained in the integration of characterization, research, and construction work.

The main task during the initial phases of the Äspö HRL has been to test the quality and reliability of different methods for site characterization and modeling with respect to conditions of importance for repository performance. The pre-investigation phase was divided into three stages:

• The siting stage (1986-1987) included aero geophysical surveys, regional gravimetric surveys, lineament studies, compilation of hydrogeological data in the Äspö region, and ground geophysical profiles to verify the location of major fracture zones.
• The site description stage (1987-1988) included detailed geological mapping, ground geophysical surveys, and drilling of three boreholes to a depth of about 1000 m and measurements in these boreholes.
• The prediction stage (1989-1990) included drilling of about ten boreholes down to a depth of about 500 m. A comprehensive measurement program was carried out in these boreholes including core logging, geophysical logging, borehole radar, cross hole seismic surveys, hydraulic testing, and groundwater sampling for geochemical analysis. Nearly all percussion and cored boreholes on Äspö and neighboring islands were instrumented for monitoring of pressure changes during interference tests and responses due to construction of the facility. Some large scale tracer tests were also performed. Based on all data collected predictive models were presented of geological structures, groundwater chemistry, groundwater flow and head distribution.

These predictions were then tested by comparison with data collected from the tunnels excavated during construction of the underground laboratory. During excavation of the tunnel, mapping of geological parameters was continuously performed, probe holes drilled ahead of the tunnel front to obtain data on the hydraulic properties of the rock, and geochemical sampling of groundwater. Head was continuously monitored in the surface boreholes and the inflow to the tunnel measured in 150 m long sections. This data was later compared with predictions of draw-down at different locations of the tunnel front. In addition, a number of cored boreholes have been drilled from the tunnel to obtain detailed information on rock volumes some distance away from the tunnel.

The comparison between predictions and tunnel data showed that predictions of the geometry and properties of major sub-vertical fracture zones (> 5 m wide) were generally good. The reliability of predictions of minor fracture zones (< 5 m wide) is relatively good with respect to existence and hydraulic character, but less reliable regarding extension and structural character at depth. Äspö is intersected by a number of narrow steeply dipping transmissive fractures or fracture zones trending NW. These existence of these zones was recognized during the pre-investigation phase but it was not possible to pin-point the location of these zones from surface investigations.

The results as a whole show that methods are available for investigating a repository site and collect data needed for construction and assessment of repository performance.

Site characterization work produces large volumes of data which are to be used in interpretation and modeling work. Hence, it is essential to have simple access to data, to have control over the origin, quality, and status of the data, and to be able to visualize the data in 3D. For this purpose a database system, SICADA, has been developed for archiving of activities and results from characterization, construction, and research. The Rock Visualization System (RVS) is developed to obtain a tool for interactive 3D interpretation of characterization data collected in boreholes, tunnels and on the ground surface. The RVS system is linked to the data base (SICADA) and it will hence make it possible to trace all data that has been used to build a model.

The development and testing of a wide range of characterization techniques and modeling approaches as well as the development of databases and visualization tools has provided SKB with the tools required for proceeding with site investigations at potential repository sites within the next few years. The site characterization program to be applied at the potential repository sites is under development. It is essentially based on the experiences gained at Äspö HRL. The program will be submitted to the Swedish authorities as part of the application to perform site investigations at two selected sites.

The Äspö HRL has been designed to meet the projected needs of the planned research, development and demonstration activities. The underground part takes the form of a tunnel from the Simpevarp Peninsula to the southern part of the island of Äspö. On Äspö, the tunnel runs in two turns down to a depth of 450 m (Fig. 2). The total length of the tunnel is 3,600 m. The last 400 meters were excavated with a tunnel boring machine (TBM) with a diameter of 5 meters. The first part of the tunnel was excavated by drill-and-blast. The underground excavations are connected with the surface facilities by a hoist shaft and two ventilation shafts. On the surface is the Äspö Research Village with offices, stores and hoist and ventilation building.

The design and construction work for the Äspö HRL has given valuable experience. The layout has successively been adopted to planning requirements as well as geology and construction methods. Conventional drill and blast techniques have been compared to full face excavation by a Tunnel Boring Machine. The cooperation between construction workers and scientists has been tested under real conditions and found to work well. Together with the experience from the operation of the facility in the coming years this will provide a good base for the coming work with design and construction for the deep repository.

The rock surrounding the repository constitutes a natural barrier to release of radionuclides from a deep repository. The most important function of the natural barrier is to provide protection for the engineered barriers through stable chemical and mechanical conditions and to limit transport of corrodants and radionuclides through slow and stable groundwater flux through the repository and reactions of radionuclides with the host rock.

The Tracer Retention Understanding Experiments (TRUE) are performed to gain a better understanding of radionuclide retention in the rock and create confidence in the radionuclide transport models that are intended to be used in the licensing of a deep repository for spent fuel. The basic idea is to perform a series of tracer tests with progressively increasing complexity. In principle, each tracer experiment will consist of a cycle of activities beginning with geological characterization of the selected site, followed by hydraulic and tracer tests, after which resin will be injected. Subsequently the tested rock volume will be excavated and analyzed with regards to flow path geometry, and tracer concentration.

The first test cycle, TRUE-1, which is ongoing, is performed on a small scale, is of limited duration in time, and is primarily aimed at technology development. A series of tracer experiments in radially converging (Fig. 3) and dipole flow configuration have been performed over distances of 5-10 m using both conservative and sorbing tracers. The first tracer test with sorbing radioactive tracers, STT-1, was started mid July 1997 with injection of two conservative (Uranine and HTO) and six weakly/moderately sorbing tracers (²²Na, ⁸⁵Sr, ⁸⁶Rb, ⁴⁷Ca, ¹³³Ba, ¹³⁷Cs).The test will continue through March 1998. These tests have been subject to blind predictions by the Äspö Task Force on groundwater flow and transports of solutes.

A subproject within TRUE has the aim to develop and test sorbing radioactive tracers suitable for use in studying sorption processes at the Äspö HRL. An initial study indicated that the cations of the alkali metal and alkaline earth metal groups have suitable characteristics. During 1994-1997, laboratory experiments have been underway on both generic Äspö material (Äspö diorite and Fine-grained granite). The experiments comprise both through diffusion and batch sorption experiments, the latter on different size fractions. One component of the investigations is a geochemical analysis of the size fractions. The experiment has provided data on K_d and diffusivities for the tested cations.

The TRUE Block Scale Experiment has been initiated to increase understanding and our ability to predict tracer transport in a fracture network over spatial scales of 10 to 50 m. The project is currently in the site characterization and preparation stage. Three boreholes aimed at a target volume located approximately 100 m from existing drifts at the 450 m level below ground have been drilled and characterized.

Most radionuclides have a strong affinity for adhering to different surfaces, i.e. a high K_d value. Numerical values that can be used in the safety assessments have been arrived at via laboratory measurements. However, it is difficult in the laboratory to simulate the natural groundwater conditions in the rock when it comes to redox status and concentrations of colloids, dissolved gases and organic matter. A special borehole probe, CHEMLAB, has been designed for different kinds of retention experiments where data can be obtained representative for the in situ properties of groundwater at repository depth. The results of experiments in the CHEMLAB probe will be used to validate models and check constants used to describe radionuclide dissolution in groundwater, the influence of radiolysis, fuel corrosion, sorption on mineral surfaces, diffusion in the rock matrix, diffusion in buffer material, transport out of a damaged canister and transport in an individual fracture. In addition, the influence of naturally reducing conditions on solubility and sorption of radionuclides will be tested. Tests of leaching of spent fuel under natural conditions are also planned. A great advantage of the CHEMLAB probe is the possibility to perform experiments with radioactive material without risking contamination of the environment. The CHEMLAB probe was put into operation at Äspö late 1996. A set of experiments to study diffusion of in bentonite of Cs, Sr, Tc, Co, and I have been performed so far.

The chemical properties of the groundwater affect the canister and buffer stability and the dissolution and transport of radionuclides. It is therefore important to make evaluations of the possible changes and evolution of the groundwater chemistry during the repository life time. Important questions concern the understanding of the processes which influence and control the salinity, occurrence, character and stability of both saline and non-saline groundwater.

Geochemical interpretation of groundwater-rock interaction along flow paths makes use of the results from groundwater chemical investigations, i.e. chemical constituents, isotopes and master variables pH and Eh in combination with the existing mineralogy, petrology and thermodynamic data. Useful tools for these calculations are reaction path codes like NETPATH and equilibrium-mass balance codes like EQ 3/6. These codes are frequently used in hydrochemical studies.

A newly developed code M3 assumes a complete and complex mixing of the water in the investigated system. Identified end-members or reference waters are mixed in proportions to describe all other observations. The principal assumptions behind this concept are that the varying hydraulic conditions of the past have caused the complex mixing pattern presently observed. Mass balance calculations are then made to explain the difference between the ideal mixing and the observations. These calculations will be used to describe the possible scenarios for hydrochemical evolution at Äspö over the next 100.000 years, separated into time slabs of 0-100, 100-1000, 1000-10000 and 10.000-100.000 years.

It is important that development, testing and demonstration of methods and procedures, as well as testing and demonstration of repository system performance, is conducted under realistic conditions and at appropriate scale. A number of large-scale field experiments and supporting activities are therefore planned to be conducted at Äspö HRL. The experiments focuses on different aspects of engineering technology and performance testing, and will together form a major experimental program.

A comprehensive program has been initiated for the development, testing, and demonstration of the engineered barriers and their interaction with the host rock which includes the following main projects:

The Prototype Repository will be a test of full scale replica of a repository with four to six deposition holes located 450 m below the ground surface (Fig. 4). It is aiming at a demonstration of the integrated function of the repository components including their interaction with the host rock and a comparison of the results with models and assumptions. It will include the testing of characterization methods in the deposition tunnel, boring of deposition holes, emplacement of buffer, canister and backfill, construction of plug and instrument installations. The modeling is carried out as integral parts of the tests. The plan is to decommission the Prototype Repository in two steps, the first after about 5 years and the second part after more than 10 years.

The Technology Demonstration will be a full size copy of a deposition tunnel with four deposition holes. It is aiming at developing and testing technology and equipment as well as demonstration of the different steps required for deposition of spent nuclear fuel in a deep geologic repository. The test consists of construction of equipment for emplacement in full scale size of buffer and canister in deposition hole under radiation shielded conditions. Characterization of the Demonstration Tunnel and boring of full scale deposition holes are included as test objects but only to the extent required for the Demonstration Test.

The Retrieval Test will be a full size copy of a repository with two deposition holes but no backfilling of tunnel. It aims at testing methods for canister retrieval from a saturated buffer. Characterization of the Retrieval Tunnel and boring of full scale deposition holes are included as test objects but only to the extent required for the Retrieval Test.

The Backfill and Plug Test will be a full size copy of a repository with a backfilled tunnel and an ending plug. It aims at testing different backfill materials and techniques for backfilling and plugging, and studying of the integrated function of rock, backfill and plugs. The test is partly a preparation for the Prototype Repository.

The Long Term Tests and Tests of Adverse Conditions will be carried out in both scaled down and full scale geometries, in single boreholes. They aim at either simulating conditions that are unlikely to occur but still possible, or exaggerate conditions in order to study certain phenomena. The Long Term Tests of Buffer Material aim to validate models of buffer performance at expected repository conditions (i.e. 90° C maximum buffer temperature, moderately saline groundwater), and at quantifying clay buffer alteration processes at adverse conditions. Two tests holes have been instrumented and the temperature raised to 90 and 130°C, respectively. Additional tests of this type will be initiated with a planned duration of 5-10 or maybe up to 20 years.

The comprehensive research, development, and demonstration work carried out and planned at the Äspö HRL provides a solid basis for SKB’s continued implementation of a deep repository for spent nuclear fuel in Sweden. Äspö HRL has provided the experience needed for characterizing a repository site and constructing a facility. Hence, siting, characterization, and construction of a repository can begin and continue in parallel with the experiments at the Äspö HRL. These experiments will successively improve the scientific basis for performance assessments and allow development and testing of the engineered barriers for a long time before they will be implemented in the real repository. Some experiments, like the Prototype Repository, are tentatively planned to run in parallel with the first phase of operation of the deep repository which includes deposition of some 5-10% of Sweden’s spent nuclear fuel. In this way Äspö HRL can provide some 15-20 years operating experience before the decision is taken to put the deep repository into regular operation.

The Äspö HRL has become a center of comprehensive international cooperation in the field of nuclear waste management. Currently (January 1998) nine organizations from eight countries are participating in the Äspö Hard Rock Laboratory in addition to SKB. This provides opportunities for exchange of ideas, experience, and know-how among top experts in different areas.

BACK