DECOMMISSIONING AND DECONTAMINATION
Selecting solutions12 December 2004
The Krsko plant is shared between Croatia and Slovenia – as is the responsibility for decommissioning and spent fuel management. An intergovernmental body examined the possible costs of handling these tasks so that an adequate joint fund could be created. By Nadja Zeleznik, Irena Mele, Tilen Jenko, Vladimir Lokner, Ivica Levanat and Andrea Rapic
A decommissioning programme for Krsko including low- and intermediate-level waste (LILW) and spent nuclear fuel (SNF) management has been prepared as required by the agreement between the governments of Slovenia and Croatia on issues related to investment, exploitation, and decommissioning of the Krsko plant.
An intergovernmental body (IGB) was formed by the ministers and representatives that coordinated the implementation of the agreement. It required that the programme should be an extensive revision of the existing 1996 programme, as one of several versions to be prepared before the final one, scheduled for the end of the plant’s lifetime.
The Krsko nuclear power plant, a 664MWe PWR
The purpose of the joint programme is to estimate the expenses of the future decommissioning and radioactive waste and SNF management at Krsko. Cost estimates with target sums required at the beginning of decommissioning will be the basis for establishing a special fund in Croatia and the correction of the rate for existing decommissioning funds in Slovenia.
Programme development was entrusted to Slovenia’s Agency for Radwaste Management (ARAO) and Croatian consultants APO, who formed the project team as the operative body. Also, the IGB nominated the advisory board with experts from Croatia and Slovenia. The role of the advisory board was supervising the activities of the project team and resolving issues raised by them. Consulting firms from Croatia and Slovenia were involved as well as experts from the International Atomic Energy Agency (IAEA) through short visits to Zagreb and Ljubljana for specialised fields such as economics, a pre-feasibility study for a SNF repository in crystalline rock, and so on.
Analysis was performed in several steps. First, to develop rational and feasible integral scenarios (strategies) of decommissioning and LILW and SNF management on the basis of detailed technical analyses and within defined boundary conditions. Each of the scenarios is a time sequence of interrelated and coordinated jobs on:
- The dismantling of the Krsko plant.
- Transport and storage of SNF.
- Export to or disposal of SNF in a geological repository.
- Disposal of LILW in near surface repository.
In the second step, financial analyses of scenarios were undertaken for the estimation of total discounted cost and related annuity for each of the scenarios. It was assumed that annuity is a good financial description of a scenario, so that by comparing them financially, affordable scenarios could be identified (a total of 19 payments must be made by both countries into the fund, established in 2004). Furthermore, some of the scenarios were eliminated as being less rational than others or as being inflexible in terms of time.
The third step involved additional analyses of the chosen scenarios’ technical or financial improvements – even at the cost of loosening some of the boundary conditions. Based on such rationalised scenarios, total discounted expenditure is determined along with its corresponding annuity. Using this abstract annuity, Croatia and Slovenia could determine real annuities for their national decommissioning funds. Values of annuities in Croatia and Slovenia could be different due to the existing asymmetrical situation, but accumulated sums in both of the funds by December 2022 should be the same or nearly same and, put together, sufficient for all future expenses.
The fourth step included financial comparison of the estimated expenses for the chosen scenario with the available data on decommissioning expenses for similar power plants or LILW and SNF repositories. Conclusions and recommendations were formulated.
The programme is divided in seven separate modules where:
- Previous work on decommissioning and waste or SNF management for Krsko is described. Boundary conditions are also presented here (modules 1 and 2).
- Technical solutions for decommissioning, dry storage, SNF management and disposing of LILW waste are explained (modules 3, 4 and 5).
- Scenarios are formulated respecting boundary conditions. Scenario analysis is done and the optimal scenario is chosen and improved (modules 6 and 7).
During 2000, Krsko undertook a modernisation project which included the exchange of steam generators. Output power was increased to 676MWe. The removed steam generators were temporarily stored on location in a specially prepared building. At the same time, SNF pool reracking was successfully finished, increasing capacity to that which will be required in 2023. A heat exchanger was replaced and the old one stored in the decontamination building. The modernisation was intended to extend the fuelling cycle from 12 to 18 months.
At the end of 2002, there were 2.208m3 of operational solid LILW in the Krsko storage location. Most of the waste is short-lived with very low content of alpha emitters. With current conditioning and packaging technology it was estimated that 3615m3 of waste will be generated before the end of plant life. The total quantity of LILW to be generated by decommissioning and replacement of major components is estimated at 17,500m3. Approximately 1% of that volume will be the long-lived LILW.
The establishment of a finite number of rational scenarios (strategies) integrating decommissioning and waste management for Krsko requires choosing some assumptions – boundary conditions on the processes and their time limitations. Since this part of the programme is required to be a limited revision of the study carried out by NIS Ingenieurgesellschaft (NIS) of Hanau, Germany, a significant number of the boundary conditions were kept. However, some of the boundary conditions were changed according to suggestions formulated at the Ljubljana 2000 and Ljubljana 2001 workshops which reviewed the NIS study. The project’s advisory board made a revision of the entire set of boundary conditions as well as the formulation of some new ones chosen particularly for this iteration of the programme. The most important boundary conditions in this study are:
- Krsko will operate until 2023.
- Only variations of the Strategy Immediate Dismantling (SID) strategy introduced by the NIS study will be evaluated. In particular, SID-15 and the original version of SID-96, while SID-30 will be examined in sensitivity analysis.
- Financial results should be expressed in 2002r as estimations of nominal and discounted costs and cashflows of accumulation and expenditure on a timescale.
- One near-surface LILW repository will be operational in either Slovenia or Croatia from 2013.
- One geological repository for SNF, either in Slovenia or Croatia, will be operational from 2030, but permanent export will be analysed.
- The programme will address dry storage of SNF.
- Discounting is performed with an annual inflation factor of 1.0073, an interest factor of 1.0429 (with a corresponding discounting factor of 1.035). Annuity for the decommissioning fund is calculated from the total discounted expenses, assuming 19 equal payments into the presently empty fund.
Technology and expenses
The quality of data used for evaluation of decommissioning expenses as well as expenses for waste and SNF management is variable:
- LILW management and disposal is well known, based on several good feasibility studies.
- SNF management and disposal is less well known: export expenses are known only as the first and preliminary offers; only one pre-feasibility study was available transposing Swedish disposal technology for SNF into local (very different) social and economical circumstances. The calculation of expenses was generic and quite simplified.
- Overall dismantling costs are known with less accuracy than in the NIS study, the reason being additional corrections and recalculations for the new circumstances.
Based on a thorough analysis of the different approaches taken by various countries, a near-surface repository was chosen as most appropriate. This type of repository could be constructed in one of two forms: a network of tunnels or a surface vault. This programme primarily uses analysis for establishing a tunnel type of repository for 17,500m3 of Krsko LILW – enough for all the waste generated during operation and decommissioning.
For construction, operation and closure of the repository the estimate of expenses was mostly derived from existing feasibility studies. However some of the expenses were internally assessed with the assistance of IAEA experts. Most of the expenses are related to the construction of a repository and infrastructure. The construction of the underground objects to be finished from 2011 to 2013 is the biggest single expense.
The methodology used differentiates two periods: the first, development and construction of repository, covers expenses for site selection, negotiation with local communities, preparation of requests and obtaining various licences, construction of repository and infrastructure, disposal technology and safety assessments as well as incentives; the second period, starting from 2014, covers the expenses for routine operation of the repository, closure expenses and incentives for the local community.
Disposal in deep geological formations is considered to be the only technically feasible and safe long-term solution for SNF and high-level waste (HLW). The Swedish model, developed by SKB, the Swedish agency for waste management as the KBS-3 concept, was adopted for the evaluation of the expenses related to development and construction of such a repository 500m underground in hard rock. This is a logical choice since KBS-3 is the most developed disposal concept, which is also going to be used in Finland in the first operational repository in Europe. Basic characteristics of the concept are:
- Direct disposal of SNF (no reprocessing).
- Capacity for 1531 fuel elements or 620 metric tons of metallic uranium with a small additional volume of HLW (~16m3).
- Research and development.
- Site selection and characterisation.
- Design and construction.
Two variations on the reference scenario were analysed: repository operation from 2030 with SNF store in pools; repository operation from 2050 with SNF in dry storage.
Since the Russian Federation put into force three laws opening the possibility of SNF import to Russia for reprocessing and/or permanent disposal, a prospect for export of the Krsko SNF has opened. The programme examined circumstances in which such an export could be conceived. Assuming that social and political conditions for agreement on the issue could be met, export of SNF was analysed as a parallel option to disposal in the local repository in scenarios where this was technically possible.
It could be seen from comparison with other decommissioning programmes that fixed expenses are the biggest part in establishing deep geological repositories. Because of that, the price per disposed kg of SNF will be high where the quantities of SNF are comparatively small as is the case with the Krsko wastes.
According to boundary conditions, the NIS study was used as the only source of data for estimation of dismantling costs. Expenses were ‘deconstructed’ on basic activities, their revaluation was done and then they were ‘reconstructed’ in new entities (with some modifications or added activities), and distributed in time accordingly. This was accomplished with the support of IAEA experts. Analysis of NIS expenses was used to separate expenses for the establishment and operation of LILW repository and SNF disposal facilities originally integrated in the total NIS price. In particular, workforce expenses were scrutinised since the NIS study assumed that workforce costs compose 60% of all expenses. Furthermore, average salary was set to be based on 16 1995 German Deutschmarks per hour, which is rather low. Also, it was assumed that 75% of the total workforce expenses are for the plant personnel and 25% for the local companies.
Based on this, revision of original expenses was done: per-hour salary was doubled for all the workers and then a contingency was added (50% for technological expenses and 20% for expenses related to SNF management). Recalculated expenses were then converted into 2002€.
Appropriate modifications with different dismantling options were introduced in the SID-96 original technological sequence to be used in integral scenarios which differ from the scenarios analysed in the NIS study.
By variation of the original SID strategy, seven rational decommissioning scenarios were formulated respecting boundary conditions and other limitations. Three of them end up with local disposal of SNF and four of them are assuming permanent SNF export.
One original SID-96 scenario could be adapted to new boundary conditions without variations in technology and without changes in sequence of dismantling activities: SID-96 – disposal. The same sequence of dismantling activities but with permanent export of SNF is a basis for the parallel scenario: SID-96 – export.
Since it is impossible to dispose of SNF immediately after plant shutdown, a scenario of fast decommissioning could not be made as a simple shortening of the original SID-96, except in the case of permanent export of SNF, SID-15 – export. For this scenario, and the rest of the scenarios considered here (apart from two aforementioned SID-96 scenarios), 80 years of on site storage for the main components and reactor vessel is cancelled and technological modifications are introduced to enable their dismantling, cutting and disposal prior to the rest of decommissioning operations.
To achieve fast decommissioning without the export of SNF, the original SID scenario should be modified enabling dismantling while SNF is still cooling down in the pool. If wet storage is introduced, the scenario could be finished in less than 15 years even if disposal of SNF begins from 2031 (SID-15WS – disposal). SNF could be kept in wet storage for the same number of years prior to export in symmetrical scenario, SID-15WS – export.
If it would be necessary to store SNF for more than 10 years (as is the case with SID-15WS scenarios) dry storage is indicated as a better solution since it is cheaper for longer periods than wet storage. Corresponding scenarios lasting approximately 30 years (SID-30 – disposal and SID-30 – export) were derived from technological operations of the original SID in the similar way it was done for SID-15 with export. Location for dry storage was not defined in this programme: it could be on the Krsko site or elsewhere.
Each of the seven scenarios was described with a temporal distribution of expenses for all of the central activities (dismantling, transport and storage of SNF, disposal of LILW and disposal or export of SNF). Financial analysis for which all of the needed tools were developed (financial models and computer programs) produced a discounted total cost for every scenario and a corresponding annuity for the hypothetical decommissioning fund.
A review of the most relevant financial characteristics of all scenarios investigated in this programme is presented in the Table.
After financial analysis, the most expensive scenarios were eliminated (SID-96 – export, SID-15 – export and SID-15WS – export). Among the financially superior scenarios (SID-96 – disposal, SID-30 – export, SID-30 – disposal and SID-15WS – disposal) some are technically better then others: SID-96 – disposal (inherited from the NIS study) is based on rather complicated and expensive solutions, related particularly to SNF disposal; SID-15WS – disposal was introduced as a realistic solution in response to a boundary condition requiring fast decommissioning. Both of them are technologically weak in the same way: they are inflexible to changes in planned operations with SNF (such as extended site selection process or late approval of licences, consequently causing considerable expenses for extended wet storage).
Two SID-30 scenarios could be singled out. They have both of the required properties: low expenses and adaptability to possible changes. In addition, since SID-30 scenarios are based on dry storage of SNF, some changes were scrutinised in order to make scenarios cheaper and more realistic. Lower expenses could be accomplished by slight modifications of boundary conditions. Obviously, lowering nominal costs is one possibility and shifting the costs in time is the other, if discounting is considered. In our case, three options are possible:
- Later disposal or export of SNF.
- Later opening and shorter lifespan of LILW repository.
- Using a surface-type LILW repository.
Later disposal or export of SNF and a shorter lifetime for the LILW repository were chosen for the optimisation. This was done primarily to make total expenses for export and disposal comparable, and to enable postponement of the final decision on disposal or export of SNF. Introduced extensions in timing are formally reflected in the names of new scenarios which we labelled as SID-45 – disposal (see the Figure above) and SID-45 – export. SID-45 annuities and financial indicators are also shown in the Table. Both scenarios have comparable expenses. The choice of the one to be finally executed is up to the stakeholders deciding on each scenario’s social acceptability.
Modifications made to SID-30, transforming it into SID-45, improved it financially and technologically. Both of the modified scenarios (SID-45 with disposal and SID-45 with export) are structurally quite similar, with almost identical discounted expenses. This enables simple switching from one scenario to another and back for several decades from now. Furthermore, dry storage which is an important part of the SID-45 scenarios enables simple adjustment to changes of circumstances over time and the simple translation of solutions for several years (for example, the opening of the SNF repository later than planned or changes to the schedule of SNF export) will not significantly influence the financial plan.
Due to specific circumstances, several limitations were imposed by the terms of reference to the process of the programme preparation. The next revision of the programme should be prepared differently. In particular, the NIS study should be dropped altogether as a decommissioning foundation, and new decommissioning, dry storage and SNF transport programmes should be started using the new European unified nomenclature of decommissioning jobs. The next version should respect Krsko’s particularities as well as its specific solutions for SNF and LILW, with as few generic solutions as possible in the four project stages. Some of the technical solutions could be based on Slovenian or Croatian industries. The next version should be started as soon as possible: two to three years of study will be required.
It is recommended that the basis for collecting financial resources into the decommissioning funds in Croatia and Slovenia should be the total discounted cost of the programme (in 2002r) rounded to r350 million. The corresponding value of 19 equal annual installments (deposited from 2004 through 2022) to meet that figure is €28.5 million, calculated for one joint fund assumed empty at the beginning of 2004.
Nadja Zeleznik, Irena Mele and Tilen Jenko, ARAO, Parmova 53, Ljubljana, Slovenia; Vladimir Lokner, Ivica Levanat and Andrea Rapic, APO d.o.o., Savska 41/IV, Zagreb, CroatiaFilesSID-45 – disposal timelineTablesSignificant financial indicators for all investigated scenarios Significant financial indicators for all investigated scenarios