THE PLANNED SWEDISH REPOSITORY FOR SPENT FUEL
AND LONG-LIVED LOW AND INTERMEDIATE LEVEL WASTE
Monica Hammarström and Fred Karlsson
Swedish Nuclear Fuel and Waste Management Co.
Box 5864
S-102 40 Stockholm
Sweden
ABSTRACT
The Swedish nuclear waste management programme aims at disposing spent nuclear fuel in a deep repository somewhere in Sweden. The fuel will be encapsulated in copper canisters embedded in bentonite. Long-lived low- and intermediate-level waste is planned to be disposed of in a special part of the deep repository.
Research is focusing on processes important for the safety of a deep repository. Large scale demonstration projects will be performed at the Äspö Hard Rock Laboratory to demonstrate and to test techniques for disposal. An encapsulation laboratory has been built for developing welding techniques for the sealing of the copper canisters.
The siting process of a deep repository has started. Feasibility studies are at present being performed in three municipalities in Sweden. These studies aim at identifying areas for further site investigations
INTRODUCTION
Sweden has been producing electricity with nuclear power since 1972. 12 nuclear power reactors produce about 70 Twh per year which accounts for about half of the electricity production in the country today. Up to the year 2010 the estimated amount of spent fuel will be 8000 tonnes.
According to Swedish law the producer of radioactive waste is responsible for taking care of it in a safe way. There are four power utilities in Sweden: Forsmarks Kraftgrupp AB, OKG Aktiebolag, Vattenfall AB and Barsebäck kraft AB. These four companies have formed SKB (The Swedish Nuclear Fuel and Waste Management Company) for the management and disposal of the radioactive waste from the power plants. SKB also arranges for the disposal of radioactive waste produced by hospitals, research institutes and industries.
The progress of the Swedish waste management programme is closely monitored by the government and by the pertinent authorities in Sweden. According to the law SKB has to every third year submit a program for research, development and all other neccessary measures in order to be able to handle and finally dispose of the spent fuel and other radioactive wastes. Such programmes have been submitted in 1986, 1989, 1992 and 1995; work is ongoing on the 1998 programme which will be submitted to the authorities at the end of September 1998.
FACILITIES
SKB is operating a transportation system (M/S) Sigyn, a final repository for short-lived operational waste (SFR) and an interim storage facility (CLAB) for spent nuclear fuel:
SFR is the central disposal facility for most of the short-lived low and intermediate level waste from the operation of the nuclear power plants and from industry, hospitals and research. The facility is located near the Forsmark nuclear power plant. It is built in bedrock at a depth of about 50 meters underneath the bottom of the Baltic Sea. By the end of 1996 about 21 000 m3 of low and medium level waste had been disposed in SFR. Recent estimates show that the present capacity of SFR - 60 000 m3 - will be sufficient for the waste arising from the Swedish programme if all units are operating up to 2010.
The central interim storage facility for spent fuel, CLAB, is located close to the Oskarshamn power plant. Operation of CLAB started in 1985. The facility consists of a receiving building at ground level and a storage unit in a rock cavern situated 25 meters below ground. By the end of 1996 about 2500 tonnes of spent fuel were stored at CLAB.The facility is planned to be expanded from 5000 tonnes to 8000 tonnes. Construction of a second storage will start in 1998 and should be finished by 2004.
All the nuclear power plants in Sweden are situated on the coast. The transportation of all spent fuel and operational waste to SFR and to CLAB is performed on the ship M/S Sigyn. Spent fuel is transported in casks and the operational waste in shielded containers.
In order to prepare for the siting and construction of a deep repository SKB decided to build the Äspö Hard Rock Laboratory (ref 1, 2). The laboratory is situated close to CLAB and to the Oskarshamn power plant. The main objective of the facility is to be able to perform research and development in a realistic environment at the depth which is planned for at future deep repository. The planning of the facility started in 1986 followed by the construction which started in 1990. The facility is presently at full operation. Several full-scale experiments and research projects have started and will be running in the next coming years.
STEPWISE PROCESS
The development of a system for handling the spent fuel started during the 1970s when a parliamentary committee proposed the construction of a central interim storage facility and research for disposal of high level radioactive wastes deep in the crystalline bedrock in Sweden. (AKA 1976). This first step was followed by a number of performance assessment studies and plans named KBS-1, KBS-2, and KBS-3. In 1992 SKB concluded that a method based on the KBS-3 concept would be the best suited. The implementation process started for the first steps towards building a deep repository. These comprises the siting and basic design of an encapsulation plant for the fuel and the deep repository. The first stage of the repository is planned for about 400 canisters or 800 tonnes of fuel. The encapsulation plant is planned for filling one canister per day.
After the first stage of the repository has been completed a thorough evaluation will be made both of the experiences gained from the first stage and from development of other, alternative treatment and disposal methods studied and/or applied in Sweden or elsewhere. The opportunity to change route or even to retrieve the canisters that have been deposited will be available. This strategy thus provides an approach where irrevocable decisions must not be made until all aspects of the repository implementation have been fully demonstrated.
The implementation of the first stage will also proceed in steps with siting, basic design and supplementary R&D during the 1990s, with construction during the main part of the next decade and first stage operation and evaluation during the 2010s. The stepwise approach is thus a key element in the planning and implementation of a repository.
SAFETY OF A DEEP REPOSITORY
The safety of a deep repository is dependent on the radiotoxicity and on the accessibility of the waste. Both these properties are time functions. Thus the safety of the repository has to be assessed as a function of time. There will always be a fundamental uncertainty in the prediction of the future behaviour of any system and the uncertainty may increase with time.
To achieve the desired safety during construction of a deep repository, during the operating phase and during the long-term containment phase, requirements are put on the function of the repository and its components. The composite performance of all the components must together provide adequate safety.
In order to achieve long-term safety, the disposal system is designed to isolate the spent nuclear fuel from the biosphere. This isolation is achieved by encapsulating the spent nuclear fuel in impervious canisters which are deposited deep in the crystalline bedrock on a selected repository site. If the isolation should be broken, the repository has the function of retaining the radionuclides and retarding their transport. Furthermore, transport pathways in the geosphere and dilution conditions in the biosphere can be selected so that any radionuclides that escape will only reach man in very small quantities.
The materials used in the repository have been selected with a view to the possibility of verifying their long-term stability and safety performance in the repository with experience from nature. For the same reason, the thermal and chemical disturbance which the repository is allowed to cause in its surroundings is limited. The safety philosophy for the deep repository is based on the multibarrier principle, i.e. safety must not be dependent on the satisfactory performance of only one single barrier.
The safety functions can be divided into three levels
Level 1 - Isolation
Isolation enables the radionuclides to decay without coming into contact with man and his environment.
Level 2 - Retardation
If the isolation is broken, the quantity of radionuclides that can reach the biosphere is limited by:
Level 3 - Recipient Conditions
The transport pathways along which any released radionuclides can reach man are controlled to a great extent by the conditions where the deep groundwater first reaches the biosphere (dilution, water use, land use and other exploitation of natural resources). A favorable recipient means that the radiation dose to man and the environment is limited.
The safety functions at levels 1 and 2 are the most important, and are achieved by requirements on the properties and performance of both engineered and natural barriers and on the design of the deep repository. Within the frames otherwise defined, good safety function at level 3 is also striven for by a suitable placement and configuration of the deep repository.
DEEP REPOSITORY
The isolation of the spent nuclear fuel from the biosphere is achieved by encapsulating the fuel in a canister with good mechanical strength and very high resistance against corrosion. The conceptual design adopted is a copper canister with a steel insert. The copper provides a very good corrosion resistance in the geochemical environment foreseen in a deep repository in Sweden. The steel insert provides the mechanical protection needed. Each canister contains about 2 tonnes of spent fuel. The total weight of the each canister will be 25 tonnes. The canisters are placed in deposition holes drilled from the floors of tunnels at about 500 m depth in the crystalline, granitic bedrock. Each canister is surrounded by blocks of compressed bentonite. When the bentonite absorbs water from the surrounding bedrock it will excert an intense swelling pressure and completely fill all void space in the near vicinity of the canister with bentonite clay. The clay barrier will contribute to the isolation by preventing or delaying dissolved corrosive species that may exist in minor amounts in the ground water to reach the canister. The clay will also provide some mechanical protection for the canister. The tunnels will eventually be backfilled by some material like a mixture of crushed rock and bentonite.
A repository for all spent fuel for the present Swedish programme should have a capacity of about 8000 tonnes or 4000 canisters. In addition it should be able to deposit other types of long-lived wastes at the same site. This means that the underground facilities will need some 30-40 km of tunnels and cover an area of about 1-2 km2. The surface facilities at the repository will require an area of about 0.2 km2.
Fig 1. Schematic drawing of the deep repository
The planning work for a deep repository is ongoing by preparing plant descriptions. These gives examples of possible ways to design the repository with its buildings, land areas, rock caverns, tunnels and shafts. They also contain requirements on and principles for the various functions of the repository. The construction and operation of the facility can, to a large extent, be based on experience and proven technology from nuclear installations and underground rock facilities. Special attention is given to e.g. the impact of the excavation work on surrounding rock, on methods for preparation and installation of the buffer bentonite blocks and on technology for backfilling and sealing. The studies of the deep repository design will continue in increasing detail during the coming years and construction criteria are being developed for the different parts of a repository system.
ENCAPSULATION OF SPENT NUCLEAR FUEL
Another necessary facility where the planning work has started is a plant for encapsulating the spent nuclear fuel. The intention is to expand the existing interim storage facility, CLAB, at Oskarshamn with such a plant. The plant will take fuel assemblies from the underground storage pools, dry them, transfer them to canisters made of copper with a steel insert, change the atmosphere to inert gas, put lids on the canister and seal the lids by electron beam welding. The quality of the filled and sealed canisters will be inspected by non-destructive examination (NDE) methods - ultrasonic and radiographic - before shipping to the repository.
Each canister will contain 12 BWR fuel assemblies or 4 PWR assemblies. The copper thickness will be about 50 mm and the steel thickness also about 50 mm. The copper thickness shall be enough to prevent that corrosion will penetrate the canister during the time when the spent fuel radiotoxicity exceeds what you could find in a rich uranium ore. The combined thickness should be enough to prevent any significant radiolysis of water outside the canister after deposition in wet bentonite clay. The steel insert is designed to withstand the normal mechanical loads that will prevail on the canister in the repository such as hydrostatic pressure and the bentonite swelling pressure. The total weight of a canister with fuel will be about 25 tonnes. In total some 4000 canisters will be required for the spent fuel arising from the Swedish reactors up to 2010.
The design of the steel insert is still under evaluation. The present reference concept is a cast insert with thick steel walls between each fuel assembly. This gives a good mechanical stability as well as it provides adequate protection against criticality in the unlikely case that the canister at some unspecified future time should be filled with water.
The fabrication of copper canisters of the size needed is by no means an industrially available technology. The seal welding technology has recently been demonstrated on a laboratory scale in work sponsored by SKB at the Welding Institute in UK. Full scale canisters have also been fabricated on laboratory scale. In order to make key technology more mature SKB has built a pilot plant - a laboratory for encapsulation technology at the former shipyard in Oskarshamn (ref 3). The main parts of the laboratory are the electron beam welding and non-destructive testing operations. Demonstration of other parts of the encapsulation process (e.g. fuel drying, transfer of fuel to he disposal canister and sealing of the cast insert is also planned.
DESIGN, CONSTRUCTION OPERATION AND CLOSURE OF A DEEP
REPOSITORY FOR SPENT NUCLEAR FUEL AND
OTHER LONG LIVED WASTE
A deep repository is a medium-sized industry with facilities both above and below ground. How these facilities are to be designed so that they provide optimal performance from a technical, environmental and safety point of view requires long-term and careful planning.
Key issues for achieving a well-functioning repository are:
Facility Design
The central operation at the deep repository is to receive canisters with spent nuclear fuel and to deposit them at selected positions 500 m down in the rock. Other types of radioactive waste will also be deposited in the deep repository during regular operation of the facility. A number of tasks will be required during normal operation:
Requirements and principles for the different functions of the deep repository have been studied and compiled in Facility Descriptions. These are based on general assumptions regarding the conditions about a future repository site and on general rock engineering assumptions based on the results of the investigations conducted by SKB on study sites. The design and layout of buildings, land areas, rock caverns, tunnels, shafts etc. above and below ground have been exemplified. The deposition of spent fuel is exemplified for the KBS-3 reference concept, while the deposition of other long-lived waste is exemplified with an SFR-like method. Three different descriptions have been worked out for the three fundamentally different access systems:
The most suitable access system will depend on technical factors as well as local conditions.
Facility Descriptions also assume that the entire deep repository is located on a single level. During the 1980s, a scheme was outlined with deposition galleries on two levels separated by 100 m. The option of planning for more than one level remains open, but will probably only be considered in special cases when the horizontal extent of good rock is limited.
Studies have indicated alternative layouts that make better use of the tunnels than is possible with one canister in each hole in vertical holes beneath the tunnel floor. One alternative would be to position the canisters horizontally in holes bored in the tunnel wall. A further development is a system with two canisters in each hole.
Final disposal of long-lived low- and intermediate-level waste is planned to take place in a special part of the deep repository located separate from the repository sections with high-level waste. Since there are different types of low- and intermediate-level waste, three disposal caverns are needed (named SFL 3-5), see Figure 2. These caverns are designed so that the different needs of the different waste types for handling and disposal are taken into account.
Fig 2. Overview of caverns for the disposal of long-lived low- and intermediate-level waste
One waste type is long-lived waste from Studsvik. This includes some of the waste which Studsvik collects from research (internal and external), industry and medicine. This waste is conditioned and packaged at Studsvik. It also includes some older, already packaged waste. This type of waste will be lowered into separate compartments in an underground concrete structure. The remaining spaces in the compartments will probably be filled with concrete. The backfill material in the space between the rock and the concrete structures may, for example, be sand, crushed rock or bentonite. Operational waste from CLAB and the encapsulation plant that arises after SFR has been closed will also be deposited in the concrete structure. Accordingly, only some of this waste can be called long-lived. A large portion consists of waste of the type that is deposited today in the final repository for radioactive operational waste (SFR).
Another waste type consists of core components and reactor internals from the power-generating reactors. It consists for the most part of stainless steel. The intention is to package the components in concrete containers which are also backfilled with concrete. The waste will then be deposited in rock caverns with floors and walls of concrete. The rock caverns will probably be backfilled with sand or crushed rock.
Decommissioning waste from CLAB and the encapsulation plant, as well as CLAB interim storage canisters (if they have not been decontaminated) and transport casks, are types of waste that arise very late in the programme. The deposition chambers for these types of waste consist of the transport tunnels and cavities that are left after closure and sealing of other repository sections for low- and intermediate-level waste.
The total volume of low- and intermediate-level waste that is planned to be deposited in the deep repository is estimated to about 25,000 m3. Reactor components and much of the Studsvik waste are counted as long-lived waste. More than half the volume consists of operational waste and decommissioning waste, which in principle could be deposited in the SFR repository in Forsmark.
Closure, Retrieval and Monitoring
The repository must be designed to ensure safety over a very long time period. Backfilling and plugging of tunnels and shafts will therefore be very important. After closure the site will be restored to nearly initial conditions.
Closure
After deposition of the canisters in the deposition tunnel the tunnels will be backfilled with a mixture of bentonite and ballast material. Temporary walls will be constructed at tunnel entrances. When all canisters have been deposited transport tunnels will also be backfilled to limit possible transport pathways for the groundwater at the site.
Fracture zones and shafts will be injected or plugged to further prevent water flow in the tunnels and also in the rock volume close to the tunnels (the excavation-disturbed zone). Boreholes must also be plugged. Solutions for how the plug should be designed have been presented. Methods for plugging of tunnels and boreholes have been tested at Stripa (former research mine), at the Ranstad plant and in SFR (boreholes).
Retrieval
Methods and equipment are also being developed for retrieval of canisters after deposition and backfilling if this should prove necessary. The possibility of reversing the deposition process is therefore one of several functions included in the design and testing of the deposition system.
Monitoring
The monitoring during operation will be organized in a similar manner as other nuclear facilities. For canisters deposited during initial operation there will be systems for measuring different parameters (pressure, temperature, moisture content, radiation level, etc.) in deposition holes and deposition tunnels. The details of such a systems will be worked out during the planning work and will be based on experience obtained during the experiments in the Äspö HRL. Shafts, ramp, access tunnels and maintenance areas in the deep repository will naturally be kept open and accessible throughout the operating period. When the time comes to the closing of the repository the initially deposited canisters will have been observed for several decades. The performance of the canisters will have been registered carefully to enable changes during operation, further inspections and additional verifications.
The scope of future monitoring is decided by each generation for itself. What can be done now is to stipulate and describe the technical possibilities for monitoring of the site and the conditions in and near the deep repository. In this context it is necessary to analyze what effect monitoring measures can have on both short-term and long-term safety.
REFERENCES