Safety, the Swedish way - KBS-3 model for disposal of spent nuclear fuel
The Swedish KBS-3 model for geological disposal of spent nuclear fuel is explored by Lena Z Evins, from the Swedish Nuclear Fuel and Waste Management Company in Stockholm, and Neil Hyatt of The University of Sheffield, UK.
Nuclear power has provided Sweden with energy for many years and will continue to do so for the foreseeable future. This results in nuclear waste in the form of spent nuclear fuel, as well as other radioactive waste. SKB, the Swedish Nuclear Fuel and Waste Management Company based in Stockholm, has the task of assuring safe handling and disposal. Sweden and SKB have devoted three decades of R&D to this assignment, which has now led up to the submission of a licence application for a spent nuclear fuel repository. Since SKB is one of the first in the world to have come so far in this process, many countries now turn to SKB for advice and collaboration.
The Swedish concept
The Swedish method, called KBS-3, is based on two safety functions – waste isolation and radionuclides retardation, if (or when) the containment is breached. Together these ensure the risk from nuclear waste is kept lower than the regulatory limit. Spent nuclear fuel is encapsulated in an isolating canister made of a cast iron insert clad with five centimetres of copper. This canister, about five metres high, and about one-and-a-half metres in diameter, is then transported to the repository, located approximately 450m down in crystalline bedrock. There, the canister is lowered into a vertical hole in the bedrock, which has been pre-lined with 50cm-thick bentonite blocks, called the buffer. The hole is covered with more bentonite, and the tunnel above is backfilled and eventually sealed.
The KBS-3 concept, therefore, is based on the long-term safety provided by a multiple barrier system. The canister is a totally sealed container which isolates the waste, the buffer protects the canister and retards any radionuclides that may have escaped a failed canister, the backfill is designed to make sure the buffer swelling is kept within an acceptable range and the geosphere, in which radionuclides are retarded in the case of release.
The KBS-3 concept has been developed, tested and improved for many years. Since the risk from the waste will last as long as the long-lived radionuclides it contains, the safety assessment covers the time up to one million years from now. The latest assessment is a major and central piece of the licence application, which brings together and applies the knowledge gained from years of research in different scientific fields.
In the repository
SKB has reached the conclusion that the proposed repository fulfils the long term safety requirements. This has been achieved by significant and focused efforts to understand the behaviours of different materials under the conditions expected in the repository. It includes research on the bedrock properties and the expected developments during the coming one million years. It also includes investigation of the behaviour of the bentonite buffer and copper canister during the changing chemical and physical environment, as well as research on the stability of the spent nuclear fuel.
The behaviour of water in all the parts of the repository and its role in far-field transport of radionuclides is of key concern. The flow and chemistry of water is central to the safety assessment and, for this, much knowledge has been gained through hydrogeological and hydrochemical research during site investigations, combined with results of climate modelling.
The bentonite buffer
Much of the R&D on the buffer has involved defining the required initial state and reference design, considering, for example, the acceptable material specifications (chemical composition, density and so on). The description of the reference design was only possible with substantial research on the processes expected in the buffer after deposition, for example swelling, gas transport and ion exchange reactions. The impact of thermal, hydraulic and mechanical processes have been described using finite element modelling, however, there are some remaining uncertainties. One of these relates to the buffer’s erosion through colloid release and, in particular, how this is affected by the montmorillonite’s chemical composition. Therefore, much of the efforts in the near future will aim to provide a more detailed knowledge of buffer erosion.
The durable canister
Copper is a durable metal, which provides a corrosion-resistant shell for the mechanically robust cast iron insert, ensuring both chemical and mechanical protection of the encapsulated fuel. Again, much effort has been spent on defining the reference design and developing methods and techniques to achieve a sealed container, such as by using friction stir welding.
The limits on the acceptable levels of impurities have been investigated due to the effect of some impurities on copper. And a number of processes, such as creep and stress corrosion cracking, have also been explored.
The finite element method has been used to model material behaviour during different situations and loads. Research concerning chemical reactions between copper and water in contact with the canister has been focused around copper corrosion. It has become clear that this is a question that brings about much interest and discussion – one challenge for SKB is to try to meet the demands of the stakeholders and to present results from on-going corrosion studies.
Spent nuclear fuel
Spent nuclear fuel is a complex material made up of a mix of elements produced during fission reactions (fission products, actinides and activation products) in an uranium oxide matrix. The microstructure of a spent fuel pellet is also quite different from that of an unirradiated fuel pellet, as an effect of in-reactor processes. It is radioactive and, therefore, a material that is difficult and expensive to study, and is handled in specialised laboratories equipped with hot cells.
Many studies aim to understand how to stabilise spent fuel, and have been performed on analogue materials simulating the properties of spent fuel. Importantly, dissolution studies under repository-like conditions have been conducted on uranium oxide doped with alpha-emitting U-233, simulating the radiation field of ancient spent fuel. Through these studies, it has been shown that hydrogen gas produced through anaerobic corrosion of iron suppresses radiolytically-induced spent fuel dissolution. However, the mechanism behind it is still being investigated.
The question of what to do with radioactive waste is of international concern. This is especially true for high-level waste such as spent nuclear fuel. Therefore, many R&D projects in this area are international. Currently, 11 European waste management organisations, and about 60 other organisations from universities, research institutes and research centres, are working together in the Implementing Geological Disposal of Radioactive Waste Technology Platform (IGD-TP). The vision is that, by 2025, the first geological disposal facilities for spent fuel, high-level waste and other long-lived radioactive waste will be operating safely in Europe.
Some of the IGD-TP participants are also conducting individual research projects and collaborations under the European FP7 programme, exemplified in the UK by the Euratom project – Reducing Uncertainty in Performance Prediction (REDUPP). It involves SKB and Uppsala University in Sweden, Posiva and VTT in Finland, and the University of Sheffield in the UK. The questions explored in REDUPP programme concern surface changes during dissolution of materials analogue to the uranium oxide matrix of spent nuclear fuel. One remaining uncertainty that will be investigated is the effect of high-energy sites on the surface of these materials. The hypothesis is that these high-energy sites dissolve faster than the surrounding surface, and that this speeds up measured bulk dissolution over the duration of the laboratory experiment. In a natural setting and a long-term perspective, the high-energy sites would be removed, and bulk dissolution would take longer.
Another uncertainty, which also concerns the extrapolation of laboratory experiments to a natural setting, is how the complexity of natural groundwater affects the dissolution rate. In the laboratory, a synthetic groundwater, with a controlled mix of elements, similar to what is found in nature, is often used. However, in real groundwater there is also a mix of trace elements, adding complexity to the chemical environment. Therefore, experiments using three different natural groundwaters will be performed in the REDUPP project.
A continuous challenge for spent nuclear fuel research is the dialogue required with nuclear power plants. The changes implemented in power plants affect the waste that needs to be handled and disposed of. The trend to increase the effect and burn-up of the fuel results in a spent fuel with different properties, which may, or may not, be of concern to the long term safety assessment. These questions need to be continuously asked and addressed, and collaboration and communication between the groups involved is essential in the years to come.
The world of geological disposal is a world of material challenges. The unique topic of a man-made construct that needs to stand the test of time in a geological environment for one million years is what drives the R&D, and will continue to do so for many years.