2003 National Meeting - International Perspectives: Reprocessing, Storage and Disposal

Charles McCombie

ARIUS: Association for Regional an International Underground Storage
5405 Baden/Switzerland



Introduction
In this symposium on technology and policy issues associated with nuclear wastes, I have been asked to address specifically three key issues: reprocessing, storage and disposal. Before doing so, I will give a brief, but quantitative, overview of the nuclear fuel cycle in order to illustrate which wastes arise from the activities involved therein.

Following this I will try to overview the status of technologies for storage, reprocessing and disposal and to address key policy issues associated with each of these technologies. The technical maturity and development potential differ strongly between the three fields, as do the political and societal issues currently being debated in each case.

The nuclear fuel cycle and its wastes
The conventional fuel cycle begins with mining or in-situ leaching of uranium ores (containing typically 0.15 to 0.2% uranium, but sometimes much higher - up to 20%) which are then concentrated and processed to produce uranium oxide (U3O8 "yellow cake").This oxide passes through a conversion plant, the product of which is uranium hexafluoride (UF6), which is fed to an enrichment plant. Here the content of the fissile isotope U235 is increased from its natural level of 0.7% to around 3-5%, with 85% of the feedstock being rejected as depleted uranium or tails. The enriched uranium goes to a fuel fabrication plant where it is converted to UO2 and incorporated into fuel assemblies. For a 1000 MWe, light water reactor (LWR), with around 80t of fuel is in the core at any time with a fraction being replaced each year. At a high burn-up of 50GWd/tU, this fuel will produce a total of around 3.3 GW.a of electricity.

During its years in the reactor each tonne of fuel will, from the U235 and U238 that are fissioned, produce around 60kg of fission products, 10kg of plutonium isotopes, and 8 kg of U236 and various other transuranics. If the spent fuel is reprocessed, then the fission products are solidified to yield around 50l of vitrified high level waste (HLW). A 100MWe nuclear power station therefore produces only 10-20 canisters of vitrified waste per year. The material flows and, importantly, the waste production are summarised in Figure 1.

Figure 1: Typical material flows in fuel cycles

All of the operations listed produce radioactive wastes in solid, liquid or gaseous forms. The greatest environmental challenges may actually be associated with wastes at the front end of this chain - namely managing safely the millions of tonnes of mining and milling tailings that remain on or near the land surface of uranium producing countries. However, most time, effort, resources and public attention are devoted to management of the low volume but highly hazardous wastes from the back-end of the fuel cycle. These are spent nuclear fuel (SNF), if this is regarded as waste, or the vitrified HLW ( and the accompanying transuranic wastes that are produced if a reprocessing route) is chosen. This paper also concentrates on the technical and policy issues associated with storing and disposing of these two waste streams. As far as disposal is concerned, this implies that the discussion will focus on deep geologic repositories, since these are recognised as being the only feasible method today of achieving safe permanent disposal.

Technical and policy issues
The title of this symposium highlights the fact that in the nuclear field there is an especially strong connection between technical and policy issues. Nuclear technology discussions in our society today tend to be highly polarised with committed exponents of extreme views on both the benefits and the drawbacks. There have been various explanations put forward for this: the military origins of advanced nuclear technology, the long-continued secretive approach of the civilian nuclear industry, the largely unprecedented scale of potential positive and negative consequences, an innate or inherent dread of unseen hazards, a reaction against all new and large scale technologies, etc. It is not the place here for detailed examination of such ideas. It is important, however, to be aware that the unusually controversial debate on nuclear matters can affect directly the judgement of debaters on both technical policy and issues associated with nuclear applications.

For the three areas proposed for comment, storage, reprocessing and disposal, Table 1 gives a concise summary of my assessment of the primary technical and policy points to be made. Those cognoscenti who are familiar with current debates on these activities can read from the table most of the messages I wish to present. For those less intimately involved in the technologies, each point is discussed individually below. Because of my personal experience and interests, but also because it is currently the most controversial issue, more discussion is devoted to waste disposal than to the other two topics.



Storage of nuclear wastes
I begin by looking at the technological and policy issues associated with storage because this is the simplest of the three activities to be addressed. Storage of wastes is a long-established technology; it is widely practiced today and will continue to be necessary for many decades into the future, since no subsequent technology that removes the need for continuing storage will be implemented for a long time.

The HLW and SNF upon which we are concentrating are extremely radioactive for many years. The SNF contains fissile materials that could potentially be misused in nuclear weapons. The radiological hazards associated with both waste types are very high, so that reliable measures are needed to ensure that there is no inadvertent entry of radionuclides into the environment and that deliberate releases by malevolent groups are excluded. In short, safety and security of stored HLW and SNF must be guaranteed.

The technologies for safe storage are well proven. Early storage facilities for SNF were primarily in water pools in well protected nuclear facilities (wet storage). Increasingly, spent fuel and HLW is being stored in strong sealed and shielded containers (dry storage) or in vaults. Both wet and dry approaches can provide safety and security. The dry storage approach is better suited for very long storage periods, although some wastes (e.g. SNF from Magnox reactors) must be kept under water for safety reasons. No technical developments are necessary for ensuring the safety of storage (although work is still going on examining fuel creep, hydrogen migration etc.). The recent rise in terrorist acts has led to some re-examination of security aspects, such as susceptibility to missile attack. In most countries, spent fuel is stored in pools at the reactor site. In some, e.g. Sweden and Finland, centralised pool storage has been implemented. Dry storage casks have been licensed at reactor sites in the USA and Canada, centralised cask stores are in operation in Germany and Switzerland, and dry storage in vaults is practised in the UK.

Storage is technically unproblematic; but there are some important policy issues to be addressed. These concern the duration and the location of storage activities. When reactors or reprocessing plants were built it was believed that the SNF or HLW would be removed (after limited storage to allow cooling) and shipped to a disposal facility. The huge delays in implementing such facilities have meant that storage periods are drastically extended; in some countries geologic repositories are so distant that no date is predictable; in a few countries there is insufficient acceptance of the feasibility off geologic disposal so that storage becomes "indefinite". These policy positions have some feedback to technical issues such as long-term maintenance, replacement of obsolete facilities, etc.

However, the key issues are societal. Communities currently hosting SNF or HLW stores or being proposed as hosts do not wish to become de-facto final repositories. This controversy is growing with time, since the delays in disposal are leading to increasing requirements for storage. The additional storage can be at reactor sites., providing licenses can be amended, or at new centralised storage facilities, providing these can be sited.

Extending storage at operating sites (by introducing compact storage in existing pools or by expanding facilities) is proving feasible in various countries; in others, for example the USA and Taiwan, reactor operators are having increasing problems. What will happen if the need for storage outlasts the operational lifetime of the facility is unclear. In Germany the anti-nuclear government has introduced a deliberate policy of implementing at-reactor storage, although existing centralized facilities are not been utilized. This is based on the unjustified assertion that transport risks rule out transfer to the existing stores. In practice, German transports have been made enormously complex and expensive because of the measures needed to control the extensive public opposition (which is targeted on specific transfers to the Gorleben site)). In the USA, because some reactor operators cannot implement sufficient on-site storage of SNF, contentious litigation has been initiated with the USDOE, which failed to meet its 1998 target date for acceptance of such fuel. This has also led to proposals for storage facilities run as private enterprises.

The situation in the USA concerning storage and disposal illustrates well how policy issues can directly affect technical programmes. To run disposal operations as efficiently as possible it would be advantageous to have a large storage capacity close to the repository so that waste emplacement could be decoupled from acceptance and could be planned to minimise thermal loading problems. The newly published study of repository staging by the National Research Council also makes this point for the Yucca Mountain Programme. Because of the political sensitivities associated with implementing large stores before repositories are available, this is currently not a feasible policy option in most countries. One solution would be that the host community agrees contractually with the operator of the storage facility that the wastes must be removed by a certain date. This is the case with the Swiss ZWILAG storage facility - however, long-term contracts of this nature require a level of mutual trust that is not always in evidence between communities and implementers.

Reprocessing of spent fuel
Civil reprocessing technology was developed in order to obtain unused uranium and plutonium for use in fast reactors. The technical challenges associated with the process have been solved and reprocessing plants have run on a large scale in France, the UK, Russia, the USA, and in other countries (e.g. Belgium and Italy) as research operations. France, the UK and Russia offer commercial reprocessing services that a number of countries are using or have used (e.g. Japan, Germany, Switzerland, Italy, Spain and Sweden). Japan is currently building its own commercial plant.

The technology of reprocessing spent fuel with separation of plutonium and recovery of unused uranium is established. Technical improvements could increase separation efficiencies or could reduce emissions of radioactive substances during operation. There are bigger technical challenges, however. One is to develop an advanced reprocessing and fuel fabrication cycle that avoids having segregated plutonium at any stage. This could greatly reduce the proliferation risks associated with reprocessing. Another technical challenge might be to develop a "second generation" waste form to replace the borosilicate glasses in which the fission products are included at present. Much work has been done on development of ceramics and synthetic minerals. Whether the improved performance that these can show is needed in order to further increase the high levels of safety already predicted for repositories is the main question.

In fact, the main technical challenge associated with reprocessing may actually be associated with the mundane goal of reducing costs. Currently the costs of reprocessing are so high that they cannot be compensated by the value of the recovered materials nor by the savings on spent fuel storage and disposal costs. Using estimated figures for reprocessing costs of around USD 1000/kg and mixed oxide fuel fabrication costs of USD 2000/kg give, for the production of 1 kg of mixed oxide (MOX) fuel, a cost of USD 7000, since recovered material from 5 kg of uranium fuel is needed. Savings are achieved by not having to dispose of these 5 kg of uranium but these are not sufficient to offset the price differential to uranium fuel, which costs around USD 1400/kg. Commercial break-even would happen only if the price of the 8 kg of uranium feed increased far beyond the current typical uranium prices of 20-40 USD/kg. Another driver that could make MOX fuel commercially more attractive would be if subsidies could be given for MOX fuel use because of its potential value in eliminating excess weapons plutonium.

This major technical challenge of reducing reprocessing costs is intimately connected with the policy reasons that originally led to development of the technology. A once through fuel cycle burning U235 utilises about 3.5% of the fission energy in the uranium; reprocessing and use of plutonium in fast reactors could allow 60-90% to be used. The policy decisions that led to development of commercial reprocessing were, therefore, based upon the desire to conserve uranium resources by providing a plutonium supply for fast reactors. This argument has been severely weakened over time by the fact that the slow growth of nuclear power has meant that uranium has remained cheap. Moreover, postponed deployment of breeders has reduced demand for, and therefore value of, plutonium. In fact, the excess plutonium already produced from reprocessing has become a liability (it is expensive to store and degrades in storage) and represents a potential proliferation risk. Today the nuclear industry is using some plutonium in MOX fuels, although the costs are high (as mentioned above) and the uranium savings only around 20%. Serious consideration is therefore being given to reactors designed for deliberately burning excess plutonium and even to directly disposing of plutonium as a waste.

In summary the key policy issues associated with reprocessing are the weakened resource conservation arguments, the increasing concerns about nuclear proliferation and the non-viable economics when fresh uranium can be extracted at low prices. Indeed, it has been argued that the prices of extracted uranium would have to rise to some hundreds of USD/kg to make reprocessing commercially competitive - and at such prices it may even be feasible to extract effectively unlimited quantities of uranium from sea water.
Countries that currently follow a reprocessing strategy do so for one or more of the following reasons:

  • they believe that uranium resources may yet become scarce, or access to uranium become difficult for countries with no indigenous resources, so that plutonium becomes a valuable fuel
  • they wish to move spent fuel off reactor sites and there are neither storage nor disposal facilities available
  • reprocessing gives reduced volumes of a high quality waste product (vitrified HLW) that can be stored and disposed at lower cost
  • they have invested large sums in reprocessing technology; these must in any case be amortised so that only marginal costs are relevant.

Countries that have an on-going commercial reprocessing strategy, despite the high costs involved are France, the UK, Japan and Russia. Other countries that have historical commitments to reprocess include Belgium, Switzerland, Italy and Germany.

Geologic disposal: technical and policy issues
Of the three nuclear technologies discussed in this paper, geologic disposal is the least tried. In fact, although the concept of disposal in deep geological formations was recognised by the US National Academy of Sciences as early as 1957 to be the most promising form of confinement for long-lived wastes from the nuclear fuel cycle, there is not yet any deep geologic repository for SNF or HLW in operation. Geologic disposal was certainly not (despite the view held today by many opponents) chosen as a cheap and dirty option to get the radioactive waste "out of sight and out of mind". The concept of geologic disposal is a logical consequence of the easily observable decay of radioactivity with time which leads to a continuous reduction in toxicity of these wastes. Finite hazardous lifetimes (and low volumes) led to the development of concepts where environmental protection could be aimed at by isolating wastes from man's surroundings for long enough to allow such decay to occur. This led, in turn, to the search for environments which showed sufficient stability for the time periods involved. There are not many environments for which we have evidence of their evolution and their stability over the timescales appropriate for long-lived wastes - namely hundreds of thousands of years. Old, deep geological formations are the most obvious candidates which can be accessed with today's technology. Consequently, concepts for geological disposal under the continental earth's crust have been developed over many years.

However, virtually every waste disposal programme in the world has experienced delays - often very significant - in its originally proposed schedules for disposal of spent nuclear fuel or high level wastes. The USA has always been the country with the earliest proposed operational dates for a geological repository for HLW/SNF. This is partly explained by the US intention to dispose of relatively young spent fuel, as opposed to many other countries, which plan for a cooling period of 30-50 years. Following years of fieldwork and expenditure of several billion dollars, a license application is now being prepared by USDOE for submission to USNRC. The DOE aims at disposal by 2010, but the delays that may result, in particular from lawsuits, are not calculable at present.

The next most ambitious timing has been in Sweden, where an early decision to close down nuclear power meant that a definite final waste inventory could be planned for. Finland has caught up with, or overtaken, its neighbour and has successfully nominated a site for spent fuel disposal. Today, it still appears that the USA may be the lead nation, if the Yucca Mountain project passes its current hurdles. Finland and Sweden may follow closely, so that there could be three operating repositories by around 2020. In other countries, e.g. Japan and Switzerland, there is no need to, and no intention to, implement deep disposal of SNF/HLW before about 2030-50. In some countries, (e.g. UK, Canada, Spain, Netherlands) decisions on geological disposal are wide open and implementation may be a hundred years off. In fact, in both of the first two countries mentioned, the government has - after catastrophic failure of established repository development programmes - officially declared that geologic disposal is no longer established policy and must be considered along with other waste management options.

All of the countries mentioned thus far have run waste disposal research programmes of various intensities, for many years. Other countries that rely strongly on nuclear power, e.g. Taiwan and South Korea, have initiated disposal studies only quite recently, and numerous smaller countries, with limited nuclear facilities, will have problems finding sufficient resources to develop appropriately scaled national R&D programmes and, even more so, to implement repositories. For these, the solution may be the implementation of international or regional repositories. This is a very topical, but still a relatively controversial, issue in many countries.

What are the main technical challenges associated today with geological disposal? Implementing a deep repository is a major technical task involving designing or selecting engineered and natural safety barriers, major underground construction, building and operating equipment and facilities for transporting, encapsulating and emplacing the SNF or HLW, running the emplacement operations, and backfilling and sealing the facility after many decades. There is general agreement amongst waste management organisations that a staged or stepwise programme is needed for such major and long-lasting projects. Despite the scale and complexity of the engineering and scientific work involved, however, the only technical issue that remains truly controversial is the scientific credibility of predicting repository system behaviour for tens or hundreds of thousand of years into the future.

Even this technical issue is not as controversial as one might be led to believe when observing the emotional debates that take place in virtually all countries with waste disposal intentions. The intensity of the debate is increased, on one hand, by opponents of nuclear power who fear that acceptance of geologic disposal as a safe endpoint might encourage wider use of nuclear technology. On the other hand, proponents of geologic disposal have often overstated the certainty of their arguments, failed to make sufficiently clear that absolute certainty and zero risk are non-attainable wishes, and concentrated on talking at the concerned public instead of listening.

One interesting technical issue concerns repository design strategies, i.e. the choice of the engineered and natural (i.e. geological) barriers. In Sweden and Finland, where disposal is planned in crystalline rocks below the groundwater table, an extremely long-lived copper overpack is a key feature. In countries like Switzerland and Japan (which have looked at both clays and crystalline rocks), steel overpacks in combination with a massive buffer of low permeability clay are extremely important components of the overall safety system. In a salt host rock, with its potentially very good isolation properties, a high integrity, long-lived overpack may not be necessary. The repository layout and the emplacement geometry selected are a function of the mechanical and thermal properties of the rock, the physical and thermal properties of the wastes and the operation strategies determining requirements for accessibility and/or retrievability of the emplaced waste packages. The sophisticated engineering designs currently proposed in the Yucca Mountain project result from the requirements to emplace large SNF packages with relatively high heat loadings in horizontal tunnels in an unsaturated oxidising environment. The US system is perhaps the most complex proposed to date. Specific unusual features are the planned use of expensive titanium drip shields and the reliance on long-lasting fuel cladding as a significant safety barrier. One advantage of the Yucca Mountain system is that accessibility and retrievability of the waste packages over long times are relatively easily achieved in such a case.

Although such interesting technical issues are still under discussion, I believe that the scientific or technical challenges associated with geologic disposal do not involve devising new or better techniques. They lie in improving our understanding of the long-term behaviour of those system components ensuring safety. Challenges of equal or greater magnitude lie in communicating to a wider audience exactly what we do and do not know about future system evolution. An exact prediction of a single evolution is not necessary. If all conceivable future scenarios lead to acceptably safe repositories, then it is not essential to predict which one will actually occur. Precise predictions of component behaviour are not essential if pessimistic bounding estimates lead to acceptably safe repositories.

These opinions lead to the conclusion that there are no strictly technical issues preventing implementation of geologic repositories, although the task is by no means trivial. In fact, this opinion is held by the majority of the scientific and technical community. Even many of those technical experts who are still concerned about remaining uncertainties would be prepared to initiate the disposal process, if reversibility can be maintained over coming decades. The common impression that a great technical controversy exists is due to a large extent by the tendency of the media to give equal coverage to both sides of an argument, independently of where the weight of opinion lies.

Most of the controversial issues associated with geologic disposal are, therefore, really of a policy rather than a technical nature. Are there credible alternatives to geologic disposal? When should disposal projects be implemented? Where, if anywhere, can repositories be equitably sited, i.e. in a process aimed at satisfying technical safety criteria and also winning the support of a sufficient fraction of the public? Can one reverse the disposal strategy, should unforeseen problems arise? These are the key questions being repeatedly posed about geologic disposal over the past years.

Concerning alternatives to geologic disposal, a 2001 report of the National Research Council recorded the most common scientific view - namely that there are no other approaches that offer long-term safety without placing a continuing burden on future generations. Isolation can be achieved for long times by building, maintaining and guarding strong and secure surface storage facilities. This is a tempting option for many players in the game. Political problems of siting disposal facilities are pushed into the future; large expenditures by waste producers for constructing and operating repositories are postponed; there are no immediate safety concerns since monitored storage is a proven technology; believers in technological break-through can wait for the "perfect" solution; researchers have more time and money to research; nuclear opponents can continue to point out that the problem is still unsolved. However, indefinite storage passes on a legacy to others who then must continue to commit resources and to maintain institutions to care for the storage facilities. The ideal solutions might be to permanently remove the material from our earth (e.g. by ejection into space) or to change it to a less harmful form. The former option has been considered periodically since the 1970’s and always it has been found to be too risky and too costly. Transmutation of radioactive materials to more benign forms has also been studied for 30 years. The current consensus is that transmutation is a complex and costly process, which could reduce the quantities of long-lived wastes - but it cannot get rid of all problematic radionuclides.

Recent documentation from various organisations have confirmed the confidence of the scientific community in the geological disposal option; there is, however, a significant fraction of the public which does not share this confidence. This is often related directly to the controversial issue of siting nuclear facilities. Siting efforts have a troubled history that has led to the technical and societal processes involved in selecting disposal sites being amended over the years. Very early, some sites were chosen purely by experts and officials in a closed process. The selection of the Gorleben site in Germany in the 70’s is a prime example. In the 1980’s, international bodies, primarily the IAEA, mapped out a top-down, technical procedure intended to allow objective narrowing in to a single site that would hopefully be recognized by all stakeholders as being the most suited. Experience in various countries (e.g. France, UK, and Switzerland) showed that this decide, announce and defend (DAD) strategy could lead to controversy, delays or failures in siting. Increased weight was then placed on societal criteria, in particular the degree of acceptance in potential host communities, although the capability of a site to provide long-term isolation remains a condition sine qua non. Recently this approach has proven successful, particularly in the Scandinavian countries, Finland and Sweden. In the former, agreement has been reached on the favoured geological repository site by the implementer, the local community, the parliament and the government. In the latter, local communities at two potential sites have agreed to investigation programmes that could lead to implementation. An interesting new initiative is in Japan, where - as an initial siting stage - requests for volunteers have gone out to all municipalities (over 3000).

The US selection process leading to selection of 3 preferred sites, by using a multi-attribute utility analysis, was originally one of the most structured and most transparent approaches. A similar formal approach was used in the UK for siting a L/ILW repository; this also led to 3 candidates (although without revealing the identities of all those that were dropped). In both cases, however, a highly political process was subsequently used to select the final single candidate, and in both cases (Yucca Mountain and Sellafield) long-lasting controversy resulted.

An interesting siting issue is that, if geologic disposal is the chosen option, are no ethical, technical or other reasons that compel countries to implement national solutions. The nuclear fuel cycle is already international, with mining, enrichment, fuel fabrication and reactor construction all being carried out by relatively few nations for the dozens of countries using nuclear power. Other toxic wastes are imported and exported when better environmental results can be achieved by doing so. International agreements, e.g. the IAEA Radioactive Waste and Spent Nuclear Fuel Convention, recognise that such transfers can take place - without relieving the waste producing country of its prime responsibility for safe and secure management. Some individual countries have, nevertheless, chosen to legislate against waste import. This is a national prerogative that must be respected. It is based, however, more on considerations of public acceptance and political feasibility than on ethical considerations.

Achieving adequate acceptance of a proposed geologic repository is also associated with the policy towards reversibility of the process, up to and including retrievability of the wastes. The concept of deep geological disposal was developed in order to permanently remove radioactive wastes from the human environment. The very foundation of the concept is that wastes deep underground will be contained until they present no significant hazard; retrievability was therefore not a significant issue during concept development. Retrieval of wastes for safety reasons was reckoned to be a scenario of such low probability that little effort was devoted to its study. Retrieval for other reasons, such as recovery of usable raw materials (fissile isotopes, precious metals etc.) was treated under the heading of deliberate human intrusion. The philosophy which was commonly followed was that no measures should be taken to ease such retrieval and that any future society deliberately embarking on this course is itself responsible for any risks arising. The responsibility of today's society is to maximise the safety of future generations whilst imposing minimum future burdens.

In recent years, however, there has been an increasingly active debate on what exactly are the prime responsibilities towards future generations by the current one. Do we want to minimise the burdens or maximise the choices of options - or can both aims be fulfilled at the same time? Should one plan for enhanced future accessibility in order to offer wider choices or should one emphasise passive safety systems which may make access more difficult, but will thereby minimise future burdens? For HLW without significant content of fissile materials retrievability arguments are related mainly to the confidence of different groups in the long-term safety performance of the repository. For fissile materials, the prime arguments for and against retrievability concern resource conservation and weapons safeguards. However, the public desire to have reversibility as such - without specifying the reason or giving any justification - needs to be acknowledged. There is a growing recognition that many societies are uncomfortable with the concept of perceived irretrievable disposal; bitter lessons from the past have too often revealed that technical or societal developments have not always progressed as expected. We have, thus, a perceived potential conflict. Technologists are dedicated to avoiding any compromise of safety by introduction of intrusive, post-closure monitoring or of retrieval measures which might be counter-productive; society at large has less confidence in technology and a stronger desire to keep options open.

The answer offered by many national disposal programmes to these questions is to implement repository projects in a staged manner, maintaining reversibility in the process for as long as feasible. In fact, a period during which the wastes in their final configuration can be observed, monitored and if necessary retrieved has, in fact, been a feature of regulations in some national programmes (e.g. US requirements for an initial 50-year retrieval period). The feasible timescales, however, were judged earlier to be only some decades; whilst this is long for human activities, it covers only a negligible portion of the relevant containment timescales for a geologic repository. Today, some programmes offer simple retrievability for much longer (up to hundreds of years) and argue that retrievability remains in principle feasible for much longer. This conclusion must be demonstrated to the public on the basis of specific studies on retrieval concepts and techniques.

A final policy related issue in waste disposal that has recently been topical concerns the security implications of moving wastes and spent fuel from distributed surface storage facilities to centralised underground repositories. It seems clear that a higher level of nuclear safeguards (guarding against misuse by States) and of physical protection (guarding against terrorists) can be offered at repositories. Counterarguments are the risks involved in transport to the repository and the potentially greater attraction of concentrated materials for terrorist groups. The debate is, however, of little immediate relevance since there is no way of accelerating repository programmes enough to have them make a major impact in security over the next ten years and more.

Conclusions
The broad conclusion that can be built upon the above discussions can be summarised as follows:
  • The technologies for storage of HLW and spent fuel, for reprocessing and for disposal have all been developed to the implementation stage. The challenges facing these activities are in all cases more societal than technical.
  • Storage technologies are well tried and long tested; they present no technical problems. Siting of the centralised storage facilities that are needed as reactor stores fill up is a serious societal challenge.
  • Reprocessing technology has been developed and implemented in various countries; improvements could be made, but there is little incentive at present, given the declining interest being shown primarily for economic reasons. The intensive debate on the proliferation hazards potentially associated with reprocessing has abated; should it re-emerge, new technologies that avoid segregating plutonium could be developed.
  • The technology for geologic disposal is developed and could be implemented today, although significant optimisation of designs is possible. The chief obstacle has been obtaining the required level of societal acceptance. Some countries, however, are moving ahead now in a way that promises the operation within 10-20 years of repositories that could act as reference facilities.
  • The USA, through implementation of the project at Yucca Mountain, could become the first example of a country that has implemented deep geologic disposal of spent fuel, thus following upon the long-delayed success of the WIPP project.
  • Not all of the lessons that can be drawn from the US programmes are positive. The costs involved are horrific examples for smaller nations. The exemplary transparency of programme progress is somewhat tarnished by the prominence of political bargaining and adversarial legal wrangling.
  • The role of the National Academies in influencing waste management strategies has been long-lasting and important. This is illustrated by the overarching reports produced at regular intervals from the landmark report of 1957 through to the staging report released on the day of this symposium. It is emphasised further by the numerous more technical reports produced by scientists and technologists working within the framework of the extensive committee system set up by the National Research Council to give unbiased input on issues vital to ensuring safe management of all radioactive wastes.