Management with CARE6 November 2004
Disposal of long-lived radioactive waste using the cavern retrievable (CARE) method would require future generations to maintain the facility during a period of long-term storage. However, this would seem to be socially acceptable in countries such as Japan with a long history of well-established institutional control by government and religious institutions. By Sumio Masuda, Hiroyuki Umeki, Ian McKinley and Hideki Kawamura
Within Japan there is currently active debate on the optimal management option for spent nuclear fuel (SNF). Although the country has a large nuclear power sector (installed capacity of about 45GWe), it has no significant uranium resources. Therefore, present government policy has established reprocessing as the chosen option in order to maintain a secure, strategic reserve of fuel uninfluenced by potential external fluctuations in supply or price. The policy of recycling of spent fuel is also compatible with the principle of sustainability, which encourages optimal use of natural resources.
Past economic analyses have also supported reprocessing as being commercially viable. In recent years, however, the price of fuel has not increased as expected due to a number of factors including slow international growth of nuclear power, discovery of larger, richer and more easily exploited resources of uranium ore and input to the market of fissile material from the decommissioning of nuclear weapons. Further, the cost of reprocessing has turned out to be higher than originally expected, there has been resistance to the use of reprocessed fuel (MOX) in some areas and the fast reactor programme has developed much more slowly than envisaged. Taken together, there are arguments that — at present — reprocessing of SNF and disposing of vitrified high-level waste (HLW) is associated with cost penalties compared to an option of direct disposal of SNF. This is also the case in other countries with similar boundary conditions (such as Switzerland), which have not extended reprocessing contracts and are now focusing on direct disposal of SNF.
In the development of a management strategy for SNF, it should be taken into consideration that the entire cost comparison could change markedly due to international socio-political factors (for example, expanding nuclear power due to concern about global warming, or a move to nuclear-based hydrogen production systems to replace oil and gas). Nevertheless, the balance may not shift in favour of reprocessing for several decades. With this background, continuation of a strict policy of reprocessing is being intensely debated.
To support a decision to consider range of scenarios with complete, limited or no further reprocessing, a waste management strategy that maximises flexibility is desirable. From international experience, the basic concepts developed for HLW disposal can probably be adapted for both SNF and HLW — although there may be tougher requirements on the site geological environment, the engineered barrier system (EBS) and the associated performance assessment for the SNF case. Nevertheless, there are also strategic differences between disposing of HLW — which has no conceivable future use — and SNF, which is potentially a valuable resource.
The Specified Radioactive Waste Disposal Act was promulgated in June 2000 to define requirements for the safe management of 'specified' radioactive waste, which is currently defined to include only vitrified waste from the reprocessing of power reactor fuel. The act is prescriptive in terms of the management options to be followed. It specifies deep geological disposal in Japan at depths of greater than 300m below the surface and a capacity of at least 40,000 waste packages — equivalent to the arisings from reprocessing all fuel used by Japanese power plants until around 2020. Also specified are requirements for geological stability and avoiding conflict with natural resources. The phased process of developing a siting project, involving more detailed characterisation as options are narrowed down, is also outlined in the act, which led to the Nuclear Waste Management Organization of Japan (NUMO) being established in October 2000.
NUMO has decided to use a siting approach for a HLW repository based on a call for volunteers sent to all municipalities in Japan. This decision was based on international experience which has shown acceptance by local communities to be one of the key factors determining the success or otherwise of such projects. To facilitate this process, it is required that NUMO must not only demonstrate openness and transparency in its work, but also strive to actively involve the local communities in the development of concepts for repository implementation.
The key technical components of the Japanese HLW disposal concept are based on work carried out over two decades by a range of R&D organisations, led by the Japan Nuclear Cycle Development Institute (JNC), before NUMO was established. This work, most recently summarised in the H12 reports, involved a generic evaluation of the requirements for a safe repository in the types of rocks and geological environments expected to be found in Japan. The four main reports summarising this work are available in English on the JNC website.
The basic repository design developed involved sealing vitrified waste in a thick steel container (the 'overpack'), which is surrounded by a highly compacted bentonite layer (the 'buffer'). A number of variants of the emplacement layout and the materials involved were examined in the H12 study, but there was no attempt at optimisation with regard to operational practicality or the exact conditions to be expected in site-specific host rock environments (see repository layout and emplacement strategies).
NUMO has built on the H12 foundation to develop a series of design options which emphasise practicality of implementation in a safe and quality-assured manner and which may be applicable for the expected conditions at volunteer sites — for example, the availability of sufficient disposal space (a particular concern that arises from the small size of many Japanese municipalities), the need to keep appropriate distances from volcanoes and active faults, and the complicated geological structures found in many regions of Japan. Additionally, repository design options emphasising ease of inspection and waste retrieval over long time periods have been developed as a possible response to desires expressed by some stakeholders. These factors, taken together, led to a particular interest in cavern disposal designs. Such designs — embodied in the cavern retrievable (CARE) concept — are particularly attractive when a flexible management programme is desired.
The basic principles behind the CARE option are illustrated in Figure 2. The repository is implemented in two distinct stages; in the first, ventilated underground caverns are operated like a surface storage facility, with waste being fully accessible. When a decision is made to close the repository (provisionally after around 300 years), the caverns are then backfilled and sealed. A technical advantage of this option is the high emplacement density (small footprint) — thermal loading is decreased by the decay of thermal output during the storage or 'institutional control' (IC) period. In terms of long-term safety, the basic engineered barrier components after closure are basically the same as in H12; for vitrified HLW, the stainless steel pour-containers are encapsulated in a massive steel overpack and surrounded by a bentonite-sand buffer. Indeed, as many key radionuclides are solubility-limited, this high density option tends to reduce the total flux of radionuclides from the engineered barriers and hence improves performance.
Figure 2: The CARE concept: (a) the caverns at a depth >300m are accessed by a ramp and a number of shafts; (b) the large transport/storage/disposal casks are transported down the ramp by a road or rail system; (c) the casks are emplaced upright in large caverns with strong ventilation to remove radiogenic heat; (d) during the storage period, waste is fully inspectable and can be easily retrieved at any time; (e) after a decision is made to close the facility, caverns are backfilled and sealed.
In terms of public acceptance, the long period of active IC may allow more opportunity for communities to become comfortable with the idea of taking the decision to close and seal the repository. Unlike surface storage, however, the commitment is made for a permanent solution to the waste issue. Deep storage also provides extremely high levels of security and physical protection against any kind of natural or anthropogenic perturbations.
To allow for ease of inspection and possible retrieval, extremely massive, multi-purpose (transport/storage/disposal) overpacks are envisaged. Detailed designs have not yet been developed, but these can be considered to be similar to the Castor/Pollux system developed by DBE of Germany or the multi-purpose canister design for the Yucca Mountain project. These designs are effectively self-shielding, can be readily transported with well-established technology and are extremely robust in terms of possible operational perturbations.
The 'cask' overpacks are stored in caverns excavated at a depth of at least 300m below surface. Preliminary design studies have indicated that construction in hard rock would be possible at depths of 500-1000m. The lined tunnels would be fully equipped with drainage, ventilation and services, allowing full remote inspection and even direct access by inspectors, if required. This means that the IC period is seen as being an active phase with ongoing development, which allows continual improvement of the disposal system, rather than simply passive storage.
As with transport casks, such casks can readily be designed to hold either HLW or SNF. The challenges with the latter are predominately associated with higher thermal output and radiation field. For the storage period, external radiation shielding could be included, which is removed before the facility is backfilled and sealed.
For SNF, in particular, easy retrieval during extended storage clearly maximises programme flexibility. At any time before final backfilling, casks containing SNF can be easily recovered and fuel reprocessed if this becomes desirable. For such SNF, ease of inspection is also very important (due to safeguards requirements).
A cavern design developed for HLW envisaged relatively dense emplacement: for example 2000 casks, holding 20 HLW packages each, emplaced within 10 caverns each 200m long. Naturally, dimensions and layout would be tailored to the geological environment considered. In any case, thermal loading is a key concern, with initial heat output being of the order of 2MW/cavern for the case above. Active ventilation (possibly with a natural convection back-up) would be needed for most, if not all, of the IC period of 300 years. By this time, thermal output will have decayed by a factor of about 20 and hence the heat output per overpack is similar to that in the H12 studies and the caverns can then be backfilled and sealed.
For SNF, the inventory equivalent to the 40,000 waste packages of HLW (approximately 30,000t) could be emplaced in the same number of casks (15t/cask), the main difference is the thermal loading, which would be considerably higher (depending on fuel type and burnup).
It may be noted that the layout includes additional empty caverns. These reflect the realisation that extensive refurbishment of the emplacement caverns would be needed over the very long time period involved. Extra caverns allow casks to be moved to temporary storage locations while such maintenance work is carried out. It is clear, therefore, that this facility is not a simple store but a more active nuclear facility and its influence on the local community needs to be considered.
If there is no decision to make use of the spent fuel, the repository would be backfilled and sealed after 300 years. For the specific case of SNF, the EBS design might be modified to improve performance — and simultaneously dispose of another problematic waste — by including depleted uranium (DU) in the cavern backfill. Alternatively, DU could be disposed of in the extra caverns. In the case that a decision is made to recover the SNF in the distant future (after closure and sealing), this should be possible and even the DU could be a possible resource (for example, for use in breeder reactors).
If the fuel is removed for reprocessing, the caverns can then be used for disposal of resulting HLW (plus associated transuranics). Ideally, the casks can be designed so that they can be re-used for this waste.
From the community perspective, a guaranteed source of employment in a facility with relatively low environmental impact could be an attractive option. The workforce is, however, envisaged to be technical staff — not the kind of 'atomic priesthood' suggested in some more science fiction based waste storage proposals. The long periods of low activity between major refurbishments would make it difficult to maintain an experienced staff and hence the repository could form the centre of an R&D facility (possibly associated with a local university), which would maintain state-of-the-art expertise in key areas. Examples of topics studied could include:
- Development of corrosion-resistant coating technology.
- Ventilation technology (including utilisation of waste heat).
- Sensor development and monitoring technology.
- Cavern maintenance technology (stability of linings and rock).
- Backfilling materials and emplacement technology.
It is to be expected that significant developments will occur in these areas and these can be implemented during maintenance of the facility. Thus, the design does not stagnate, but is continually kept to the highest technical level available.
Regardless of whether the design is modified or not, a decision might be made to close the repository in less than 300 years. A major focus of support work will thus be to develop concepts that allow this possibility, based on existing technology. The main concern is the impact of higher temperatures on the long-term buffer performance. Nevertheless, the particular design allows for the loading to be reduced by the simple process of moving some casks to the reserve storage tunnels before backfilling and sealing.
In Japanese society, decision-making is generally a gradual, stepwise process. Reversing a decision can be difficult and hence there is a desire for staged processes where a very high degree of confidence can be built up before each individual decision step. This is especially the case if it involves setting a precedent, like a first-of-a-kind facility.
An extended process of implementation inevitably involves a longer-term commitment to an active repository programme. Resultant delays in implementing a final solution to the waste management problem are sometimes regarded in a negative manner. Emphasis is placed on ensuring that the generation benefiting from nuclear power should take full responsibility for safe disposal of wastes and not pass any burdens to future generations. This leads to an emphasis on repositories with passive safety, which can be 'walked away from' soon after waste is emplaced (actually specified in regulations in Switzerland, for example). Despite this general aim, multi-century site management is already a key component of the safety case for near-surface low-level waste disposal facilities around the world.
In Japan (and some other Eastern societies), IC is generally seen as a positive attribute as there is a tradition of carrying responsibilities over many generations. The social acceptability of long-term IC is clearly related to the local cultural and historical setting — for example, the USA has few organisations more than a hundred years old and the political map of Europe has been repeatedly transformed over the last few decades. In contrast, Japanese society has been relatively stable over the last millennium and has imperial and religious traditions which are significantly older.
A good analogue might be temples and places of worship. In the West, the few very old examples tend to be either ruins or very massive structures that have been repeatedly abandoned and transformed over the centuries. These can be compared to Japanese wooden temples that are over 1000 years old — continuously maintained with rebuilding work constrained to original specifications and still used for their original purpose (for example, Horyuji temple). Such structures — and the social systems which support them — indicate that the extended commitment required by the CARE concept would not be a novelty in this environment.
Horyuji temple: this UNESCO (United Nations Educational, Scientific and Cultural Organization) world heritage site contains a range of wooden buildings which have been continuously maintained for 14 centuries
The authors worked together on the H12 project and its predecessor H3. Sumio Masuda, Executive Technical Counselor, Obayashi Corporation Tokyo Head Office, Technology Department No. 4, Civil Engineering Technical Division, Shinagawa Intercity Tower B, 2-15-2 Konan, Minato-ku, Tokyo 108-8502, Japan; Hiroyuki Umeki, Nuclear Waste Management Organisation of Japan (NUMO), Mita NN Bldg, 1-23, Shiba 4-Chome, Minato-ku, Tokyo 108-0014, Japan; Ian G McKinley, Nagra, Hardstrasse 73, CH-5430, Wettingen, Switzerland; Hideki Kawamura, Obayashi Corporation, Design Department #2, Civil Engineering Technology Division, Shinagawa Intercity Tower B2-15-2 Konan, Minato-ku, Tokyo 10808502, Japan. The authors wish to point out that, as with all articles published in NEI, the views expressed here are those of the authors and not necessarily those of the organizations they representRelated ArticlesGerman minister wants to separate utilities from repository company