Methods for disposal of nuclear waste1 February 2002
There are two strategies for the management of spent fuel: reprocessing (with or without subsequent transmutation) and direct disposal. The choice between these is determined by whether a country regards fuel as a waste or a resource.
SKB produced a report: "RD&D Programme 2001, Programme for research, development and demonstration of methods for the management and disposal of nuclear waste." This report was required by the Swedish Nuclear Activities Act. The report presents SKB's plans for 2002-2007.
SKB's plan is to implement deep disposal of the spent fuel in accordance with the KBS-3 method. The RD&D programme describes SKB's activities and planning for this line of action and the work that is being conducted on alternate methods.
In December 2000, SKB presented proposals for sitings of the deep repository and the background material on which the choice was based, as well as programmes for site investigations. In June 2001, SKB and Kasam submitted their statements of opinion to the government following an extensive review.
The RD&D report concentrates on questions relating to research and technology development. The report links regulatory requirements on long-term safety and development of safety assessment methodology and research on long-term processes in the repository. Programmes for safety and research are then linked together with programmes to develop methods and instruments for site investigations and design of the deep repository, encapsulation plant and canister. Programmes for alternative methods, decommissioning and other long-lived waste are also explained.
An overall goal for SKB is to start the initial operation of a deep repository for spent fuel in 2015. This presumes that site investigations have commenced at the start of 2002, and that the different phases have been executed without major changes. Regular operation should then be able to commence in the early 2020s before the storage pools in CLAB are full, thus avoiding further expansion. The encapsulation plant should be ready to start roughly one year before the deep repository is finished.
Future RD&D programmes will probably place the emphasis slightly differently depending on different phases and permit applications. RD&D Programme 2004 is expected to give a central role to the canister and the encapsulation technology. In RD&D Programme 2007, the deep disposal technology and continued work on alternative disposal methods may be important topics.
Over the last few decades, SKB has built up a system for management and disposal of different types of radioactive waste. The system includes a specially-built ship for transport, a final repository for different types of operational waste (SFR) and a central interim storage facility for spent nuclear fuel (CLAB). However, three important components are lacking to manage the spent nuclear fuel: an encapsulation plant for encapsulating the spent fuel in copper canisters; a canister factory for fabricating the copper canisters; and a deep repository where the encapsulated waste can be disposed of in a safe, long-term manner. A transportation system designed to serve these facilities is also needed. Moreover, repositories are needed for disposal of the waste that arises when the nuclear installations are decommissioned and for other types of long-lived waste than nuclear fuel, such as core components.
The overall goal of SKB's work of managing and disposing of the spent nuclear fuel is that the first stage of the deep repository should be finished in 2015 and that initial operation can be started then. This goal presumes that all decisions and consents from national and local authorities have been made and given, so that the site investigations can be initiated at the start of 2002. The timetable that spans such a long period of time naturally contains many uncertainties. However, it also comprises a basis for decisions concerning strategies and priorities for the years to come, and is continuously updated as better data becomes available.
The reason for the overall goal of starting deposition in 2015 is so that responsibility for the final disposal of the spent fuel will not be shifted to future generations. The goal is also that the regular operation of the deep repository should be able to be commenced before the storage chambers in CLAB are full, thus avoiding the necessity of a further expansion of CLAB.
The coming work with the deep repository will require large resources. The competence that exists in Sweden when it comes to site investigations, analysis of investigation results and safety assessments will be fully engaged in this effort. SKB intends to prioritise the work with safety assessments and design of the deep repository and the encapsulation plant in relation to corresponding work with the final repository for other long-lived waste.
Such prioritisation is possible because the final repository for other long-lived waste is not needed before decommissioning of the nuclear power plants is in its final phase. Siting and construction will therefore not take place until after 2035, according to the current timetable.
Encapsulation of the fuel should start approximately one year before the deep repository is put into operation. With reference to the timetable for execution, construction of the encapsulation plant should therefore commence before the deep repository. However, it is not appropriate to start construction before the safety assessment for the deep repository has been completed and reviewed.
Adaptation of deep disposal and encapsulation
SKB's principal task during the next few years is to adapt deep disposal, encapsulation and canister fabrication to the sites where site investigations will be conducted, and later also to the site where a detailed characterisation will be conducted. Work on this has already come quite far, and is being pursued for different parts with the following basic subdivision into stages:
• Overall choice of system (finished).
Internationally, there are two principal strategies for management of the spent fuel: reprocessing (with or without subsequent transmutation) and direct disposal. The choice between these alternatives is determined by whether a country chooses to regard the fuel as waste or as a resource. In several states with large nuclear power programmes, reprocessing has been viewed as a necessity for sustainable management of the earth's resources. Regardless of whether the spent fuel is considered to be a resource or waste, the idea is that it should be deposited in some type of geological formation. Multiple barriers will then prevent the radionuclides from reaching the biosphere.
SKB carried out a comparative system analysis to compare different strategies for disposal of spent nuclear fuel. The methods included in the comparison are the reference method KBS-3, very long holes (VLH), WP-Cave and very deep holes (VDH). KBS-3 was deemed to be the best alternative. One argument in favour of KBS-3 is that the operating phase provides individual control over both canisters and buffer.
• System analysis and optimisation (under way).
The technical solutions for managing and disposing of different types of radioactive waste can be designed in various ways. The capacity of the facilities has previously been calculated based on the assumption that nuclear power in Sweden will be phased out by 2010. SKB is now basing its long-term plans and cost calculations on the assumption that the nuclear power plants will be operated for 40 years. This means that an average of 165 canisters per year will be depositing over a 25-year period. In the light of this, it has been judged that a capacity of 200 canisters per year is still a reasonable design basis for the different parts of the system.
The building of the KBS-3 system will proceed in steps. This means that decisions on individual facilities or system parts will be made progressively. This in turn means that the freedom of choice for each individual facility, and thereby for the whole system, will be gradually narrowed. This narrowing of the freedom of choice will proceed at different paces for different facilities. Different designs of the deep repository may, for example, be retained even after decisions on the design and siting of the encapsulation plant, while the main features in the design of the transportation system are fixed when the decision on the encapsulation plant is taken.
Variants of the KBS-3 method will be evaluated over the coming years. MLH (Medium Long Holes) where the copper canisters are deposited horizontally, is one such variant. Deposition of two canisters in vertical holes that are made deeper than in the reference design is another. The final choice of deposition method will be based on an evaluation of safety, technology, costs and environmental aspects.
• Siting (under way).
The work of finding a suitable site for the repository is proceeding. Feasibility studies have been carried out in eight municipalities: Storuman, Malå, Östhammar, Nyköping, Oskarshamn, Tierp, Hultsfred and Älvkarleby. SKB has also conducted county-specific general siting studies, and a survey of the advantages and disadvantages of a siting in northern versus southern Sweden.
The encapsulation plant can be sited either at CLAB, at the deep repository, at an existing nuclear installation or somewhere else. It is an advantage if the facility can be coordinated with an existing activity and if it has access to personnel with the necessary competence and experience. SKB's main alternative is to build the encapsulation plant adjacent to CLAB.
The site of the canister factory has not yet been determined. Questions that must be taken into consideration in siting include transportation to and from the factory and access to labour
A considerable portion of SKB's programme for research, development and demonstration concerns the processes that are of importance for the long-term safety of a deep repository for spent fuel.
If a canister is damaged and groundwater comes into contact with the spent fuel, the fuel will eventually be dissolved. Fuel dissolution was pointed out in SR97, and its review as an important area where knowledge needs to be improved by further research.
Part of SKB's fuel programme during 2001-2004 will be carried out within the framework of two EU projects. The project: "Rates and mechanisms of radioactive release and retention inside a waste disposal canister" has been underway since the autumn of 2000, where SKB is coordinating the work on thermodynamic and quantum mechanical modelling of the chemical reactions in the canister cavity. The project "Spent fuel stability under repository conditions" commenced during 2001.
The implications of the research to date are that highly reducing conditions are expected to prevail in the deep repository. If a canister is damaged and water enters, the cast iron insert will probably corrode, producing hydrogen. The hydrogen gas pressure can be expected to be at least equal to the water pressure at repository level (around 5MPa) for a very long time.
This means that the fuel dissolution rates that are observed in autoclave experiments in the presence of 5MPa hydrogen today are deemed to be the most realistic for the deep repository, as long as hydrogen production continues. The results from the first experiment under such conditions point towards a very slow or non-existent dissolution.
When hydrogen can no longer be expected to be produced or remain in the canister, the experimental observations for anoxic conditions with flow reactors will be those that are judged to be able to provide an upper limit for the fuel dissolution rate. A thorough evaluation of experimental results is required prior to future safety assessments for the purpose of finding defensible descriptions of fuel dissolution under different conditions in the interior of the canister.
For an initially intact canister, it was concluded in SR97 that the uncertainties all relate to strength at loads during a glaciation and during earthquakes. Strength under isostatic loads need to be calculated with realistic material data for the cast iron insert. Strength in connection with rock movements has only been calculated for a weaker canister than SKB's current reference canister, and with incomplete mechanical data for the buffer. The study needs to be updated. Great resources will also be devoted to studies of copper corrosion and stress corrosion cracking in the copper canister.
The studies of microbial corrosion of copper will continue for the next three years. They will include laboratory experiments as well as field studies in the Äspö HRL. The question is can naturally occurring populations of sulphate-reducing bacteria in the groundwater and in the bentonite survive under repository conditions and reduce sulphate to sulphide in such quantities that it can threaten the canister's integrity?
Field experiments at the Äspö HRL involving exposure of copper for several years to both the atmosphere in the underground laboratory and different groundwaters have been started and will proceed over the next three years. Moreover, copper corrosion in compacted bentonite will be studied in-situ in the Äspö HRL. In these experiments, possible local corrosion will be followed by monitoring and analysing the electrochemical noise from the copper specimens. These field experiments will be augmented by laboratory studies of anaerobic copper corrosion in waters with high salinity.
The research on stress corrosion cracking of copper in the repository environment will continue with both laboratory experiments and field experiments in the Äspö HRL. At present, a pilot study is in progress on the prospects of studying SCC with measurements of electrochemical noise in cooperation with Canada. If the results are positive, the method will be installed and used in the Äspö HRL for studying in-situ the prospects of stress corrosion cracking of copper in a realistic repository environment. These investigations will be supplemented by calculations of the states of stress in the canister at different times after deposition.
As far as the canister's initial integrity is concerned, it is urgent to obtain greater knowledge concerning types and probabilities of leaks or defects, particularly in the canisters' lid weld.
For canisters with an assumed initial defect, it is urgent to further explore the coupled chemical, hydraulic and mechanical evolution that results if water enters the canister.
A step in the optimisation of the canister, long-term safety will also be examined for a variant where the thickness of the copper shell has been reduced from 50 to 30mm, and the thickness of the insert has been increased accordingly. The basis of this study is that the canister's corrosion protection (copper thickness) is judged to be much more overrated than its mechanical strength, which is determined by (among other things) the thickness of the insert.
After deposition, the buffer undergoes a complicated thermo-hydro-mechanical (THM) evolution when it becomes saturated with water. The process proceeds for about ten years, and our knowledge of the coupled processes during this time is not complete. Nor is such knowledge necessary for the safety judgements. It is, however, important to understand and be able to predict the state of the buffer after completed water saturation, since this comprises the point of departure for the analysis of its long-term evolution. That is why the buffer's initial THM evolution is studied in the field and with models.
The evolution of a water-saturated buffer is much slower and the couplings are simpler to handle. Studies are being conducted here as well, both in the field and with models. Individual processes where further research is urgent are the effects of saline groundwater and colloid release/erosion. As far as radionuclide transport is concerned, the phenomenon of surface diffusion may require further studies.
Swelling pressure in compacted bentonite at high salinity is being examined in an ongoing joint project with Posiva, Finland. The project is aimed at further refining the description of the bentonite/water/salt system with respect to ion concentration in the bentonite under given groundwater conditions. Effects of elevated temperature and other types of ions than sodium need to be ascertained with respect to both swelling pressure and hydraulic conductivity.
Bentonite erosion is being investigated in a recently started project, COLLOID. First a pilot experiment is being performed so that the experimental procedure can then be optimised. The intention is to investigate whether bentonite clay in contact with various electrolyte solutions forms colloids. The basic assumption is that colloid formation is dependent on the ionic strength of the solution, temperature and bentonite type (Na or Ca). The solutions in contact with bentonite clay are analysed with respect to the size distribution of the colloids, and as far as possible, the concentration of colloids. When the pilot experiment has been evaluated, an optimised laboratory experiment will be started. This will lead to experiments in the Aspo HRL and in the Äspö fracture zone.
The transport of radionuclides through the buffer is mediated by
different diffusion mechanisms. It has been established that certain cations can have high diffusivities. One possible explanation for this phenomenon is the theory of surface diffusion. The process is handled in the safety assessment by assigning higher diffusivity values to cesium, strontium and radium.
When bentonite has such high density that the electrical double layers between two planes are superimposed, a phenomenon known as annion exclusion occurs. Anions cannot penetrate into the interlamellar pores due to the electrostatic forces between the negatively charged surfaces and the anion. Anion exclusion significantly reduces the porosity available for diffusion. The effect of anion exclusion reduces at high salinities, and in crushed rock/bentonite mixtures it is negligible.
The results of the experiments at the Äspö HRL are in part unexpected for substances that exhibit surface diffusion. This will be further investigated via laboratory experiments.
The process is included in the computer program NUCTRAN/COMP23, which calculates radionuclide transport in the near field.
The backfill has been treated less thoroughly than other parts of the repository in SR97 and earlier safety assessments, in part because its properties have
not been definitively specified in the KBS-3 concept.
Different compositions of the backfill (smectite clays mixed with crushed rock in varying proportions) will be tested, in terms of both practical
usefulness and long-term properties. The research on backfilling with mixtures of bentonite and crushed rock is being done within the framework of the Backfill and Plug test in the Aspo HRL.
The objectives of the test are to develop and test different materials and compaction techniques for backfilling of drill-and-blast tunnels; to test the function of the backfill and its interaction with the surrounding rock on a full scale in drill-and-blast tunnels; and to develop technology for building plugs and testing their function.
The following tests are performed after the backfill has been saturated with water. Water transport under saturated conditions is studied by flow tests. Water transport is studied through the central portions of the backfill, in the contact zone between backfill and floor, and in the contact zone between backfill and roof.
Swelling and mechanical interaction between backfill and near-field rock is being studied by means of swelling pressure measurement and pressure cylinders. The swelling pressure
during and after water saturation can be measured with the total pressure cells placed in the roof and on the floor. The mechanical properties of the backfill in the contact zone with the rock can be studied by gradually pressurising the pressure cylinders and measuring pressure and deformation.
After the tests are finished, the plug will be demolished and the backfill dug out at the same time as extensive
sampling is performed.
Studies will be carried out on the possibility of using natural clays as a backfill material in the form of both continued field tests on Friedland clay, where above all compaction and compression properties will be tested and evaluated, and tests on bentonite. The latter clay is known for its good hydraulic properties, but needs to be tested with regard to both compaction properties and compressibility.
The long-term evolution of the backfill is controlled largely by the same processes may be different in the backfill compared to the buffer.
The modelling of earthquakes was not fully developed in SR97. SKB concluded that this development work needs to be carried further and a number of points for further study were mentioned.
The goal is that the earthquake analyses should be used in the deployment of canister positions so that earthquake-induced canister damage can be avoided completely. Another intention is that earthquakes should not be studied in a separate scenario as in SR97, but integrated in all scenarios.
The purpose of model development for groundwater flow is to make the main alternatives, the continuum models, more flexible and to further develop alternative, discrete models.
SR97, like previous assessments, shows that the transport resistance, which is dependent on flow conditions and geometric properties of water-bearing fractures, is decisive for how radionuclides are transported in the geosphere. It is also observed in SR97 and its review that a better understanding is needed of the transport resistance and how it should be handled in the safety assessment.
The most essential questions concerning future hydrochemical conditions have largely been clarified. The single major area that remains an area of concern is that of microbial processes, where research started relatively late. It is known today that microbes occur at repository depth and much deeper. The microbes affect redox conditions and can be of great importance as a redox buffer if oxygenated glacial water should penetrate down into the bedrock.
Long-term interaction between groundwater and bentonite can affect the groundwater chemistry in the near field. Basic research continues to be pursued in this field.
A methodical handling of all geosphere information from the site investigations will also require considerable efforts during the period.
The evolution of the biosphere will always comprise an essential part of a safety assessment, mainly because the consequences of a possible release from a deep repository become manifest in the biosphere. Unlike the geosphere, the biosphere can be expected to change greatly during the time the safety of the repository is to be judged, above all as a result of future climate change. This means that a series of different biospheres can be expected to exist in the future on a given site. The consequences of possible releases from the repository must be acceptable in all of these if the repository is to be considered safe.
Radionuclide turnover in the biosphere is usually described in the safety assessment with component models, where the biosphere is divided into a number of relatively homogeneous subunits between which radionuclide transfers are calculated. In previous assessments, the description of these transfers has been based on rough empirical measurements and estimates. At present, there is a trend towards more realistic so-called process-oriented descriptions, where the calculation of transfers is based on knowledge concerning the underlying mechanisms. Among the ecosystems being studied and modelled are forest, mire and sediment.
Long-term variations in climate, land uplift and salinity are crucial for the evolution of the biosphere. The site investigations require extensive require extensive methodology development as regards the biosphere.
Climate change can undoubtedly be expected to occur in the more than 100,000 year timespan over which the safety of the deep repository is to be assessed. A description of climate change and the impact it may have on the repository is therefore necessary in a safety assessment. The uncertainties surrounding future climate change are great, and any description of how different climate states succeed each other will necessarily be sketchy. On the other hand, it can be said with greater certainty that certain climate states will prevail during at least some era in the future, even though the era cannot be exactly pinpointed. The safety assessment is therefore focused on describing a number of climate states and the effect they have on the repository.