Safety Related Issues of Spent Nuclear Fuel Storage
NATO Security through Science Series This Series presents the results of scientific meetings supported under the NATO Programme for Security through Science (STS). Meetings supported by the NATO STS Programme are in security-related priority areas of Defence Against Terrorism or Countering Other Threats to Security. The types of meeting supported are generally "Advanced Study Institutes" and "Advanced Research Workshops". The NATO STS Series collects together the results of these meetings. The meetings are co-organized by scientists from NATO countries and scientists from NATO's "Partner" or "Mediterranean Dialogue" countries. The observations and recommendations made at the meetings, as well as the contents of the volumes in the Series, reflect those of participants and contributors only; they should not necessarily be regarded as reflecting NATO views or policy. Advanced Study Institutes (ASI) are high-level tutorial courses to convey the latest developments in a subject to an advanced-level audience Advanced Research Workshops (ARW) are expert meetings where an intense but informal exchange of views at the frontiers of a subject aims at identifying directions for future action Following a transformation of the programme in 2004 the Series has been re-named and re-organised. Recent volumes on topics not related to security, which result from meetings supported under the programme earlier, may be found in the NATO Science Series. The Series is published by IOS Press, Amsterdam, and Springer, Dordrecht, in conjunction with the NATO Public Diplomacy Division. Sub-Series A. Chemistry and Biology B. Physics and Biophysics C. Environmental Security D. Information and Communication Security E. Human and Societal Dynamics http://www.nato.int/science http://www.springer.com http://www.iospress.nl
Series C: Environmental Security
Springer Springer Springer IOS Press IOS Press
Safety Related Issues of Spent Nuclear Fuel Storage
edited by
J. D. B. Lambert Argonne National Laboratory Chicago, U.S.A and
K. K. Kadyrzhanov Institute of Nuclear Physics NNC of Republic of Kazakhstan Almaty, Kazakhstan
Published in cooperation with NATO Public Diplomacy Division
Proceedings of the NATO Advanced Research Workshop on Safety Related Issues of Spent Nuclear Fuel Storage Almaty, Kazakhstan, 26–29 September 2005 A C.I.P. Catalogue record for this book is available from the Library of Congress.
ISBN 978-1-4020-5902-5 (PB) ISBN 978-1-4020-5901-8 (HB) ISBN 978-1-4020-5903-2 (e-book)
Published by Springer, P.O. Box 17, 3300 AA Dordrecht, The Netherlands. www.springer.com
Printed on acid-free paper
All Rights Reserved © 2007 Springer No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work.
TABLE OF CONTENTS Preface
ix
Acknowledgements
xi
STRATEGIES FOR SAFE STORAGE OF SPENT FUEL Spent Nuclear Fuel from Research Reactors: International Status and Perspectives P. Adlelfang, A. J. Soares, and I. N. Goldman
3
Regulatory Approach for Managing Radioactive Waste in the Republic of Kazakhstan T. Zhantikin and M. Sharipov
27
Spent Fuel Management in Poland S. Chwaszczewski An Overview of Spent Fuel Storage at Commercial Reactors in the United States J. D. B. Lambert and R. Lambert Managing Spent Nuclear Fuel at the Idaho National Laboratory T. J. Hill and D. L. Fillmore
37
55
71
RADIOLOGICAL PROBLEMS OF SPENT FUEL STORAGE Assessment of Environmental Impact of Reactor Facilities in Kazakhstan K. K. Kadyrzhanov, S. N. Lukashenko, and V. N. Lushchenko
87
Design and Manufacture of Fuel Assemblies for Russian Research Reactors 95 V. V. Rozhikov, A. A. Enin, A. B. Alexandrov, and A. A. Tkachev Strategy for Handling Spent BN-350 Cesium Traps in the Republic of Kazakhstan O. G. Romanenko, I. L. Tazhibaeva, D. Wells, A. Herrick, J. A. Michelbacher, C. Knight, V. I. Polyakov, U. Privalov, M. Sobolev, U. Shtynda, A. Gainullina, I. L. Yakovlev, U. P. Shirobokov, A. I. Ivanov, and G. P. Pugachev .
107
vi
TABLE OF CONTENTS
Account and Control of Nuclear Materials at the WWR-SM Reactor in the Institute of Nuclear Physics, Tashkent B. S. Yuldashev, U. S. Salikbaev, S. A. Baytelesov, A. A. Dosimbaev, and U. A. Khalikov Activities of the Kharkov Institute Related to the Problem of Spent Nuclear Fuel Management V. M. Azhazha, I. M. Neklyudov, S. Y. Sayenko, and V. N. Voyevodin
143
147
TECHNICAL ISSUES OF WET AND DRY STORAGE Understanding and Managing the Aging of Spent Fuel and Facility Components in Wet Storage A. B. Johnson, Jr. Long-Term (100–300 Years) Interim Dry Storage for Spent Fuel: Package and Facilities Development Including Safety Aspects and Durability Assessment Program J.- P. Silvy, N. Moulin, and F. Laurent
167
181
Technical Issues of Wet and Dry Storage Facilities for Spent Nuclear Fuel 189 E. D. Federovich Problems of Nuclear and Radiation Safety of Casks with Spent Fuel during Long-Term Dry Storage A. Z. Aisabekov, S. A. Mukeneva, E. S. Tur, and V. M. Tsyngayev Trial of Storage Container Technology for Research Reactor Spent Nuclear Fuel Z. I. Chechetkina, Yu. V. Chechetkin, A. E. Novoselov, V. G. Bordachev, V. V. Maklakov, and I. Yu. Zhemkov Interim Storage and Long-Term Disposal of Research Reactor Spent Fuel in the United States N. C. Iyer, D. W. Vinson, R. L. Sindelar, J. E. Thomas, and T. M. Adams
209
211
225
MATERIALS STABILITY ISSUES OF SPENT FUEL STORAGE Managing Spent Fuel in Wet Storage at the Savannah River Site R. L. Sindelar, P. R. Vormelker, R. W. Deible, and J. E. Thomas
245
TABLE OF CONTENTS
Corrosion of Aluminium Alloy SAV-1 and Austenitic Stainless Steels 12Cr18Ni10Ti and 08Cr16Ni11Mo3—Core Structural Materials for WWR-K and BN-350 Reactors O. P. Maksimkin Corrosion of Fast-Reactor Claddings by Physical and Chemical Interaction with Fuel and Fission Products V. A. Tzykanov, V. N. Golovanov, V. K. Shamardin, F. N. Kryukov, and A. V. Povstyanko
vii
267
281
Corrosion of Research Reactor Aluminum Clad Spent Fuel in Wet Storage 295 L. V. Ramanathan, S. M. C. Fernandes, and O. V. Correa Influence of Neutron Irradiation on Mechanical and Dimensional Stability of Irradiated Stainless Steels and its Possible Impact on Spent Fuel Storage F. A. Garner Degradation in Mechanical Properties of Stainless Steels C0.12Cr18Ni10Ti and C0.08Cr16Ni11Mo3—Materials for Hexagonal Ducts of Spent Fuel Assemblies from the BN-350 Fast Neutron Reactor K. K. Kadyrzhanov, S. B. Kislitsin, O. P. Maksimkin, O. G. Romanenko, and T. E. Turkebaev
307
329
List of Authors
351
Workshop Photographs
357
PREFACE The NATO Advanced Research Workshop “Safety Related Issues of Spent Nuclear Fuel Storage” was held in Almaty from September 26 to 29, 2005. Experts from Brazil, France, Kazakhstan, Poland, Russia, Ukraine, USA, Uzbekistan, and the IAEA participated and gave presentations. The Workshop was organized parallel to the “5th. International Conference on Nuclear and Radiation Physics” held in Almaty from September 26 to 29, 2005, which was organized by the Institute of Nuclear Physics of the Republic of Kazakhstan. Some conference attendees sat in on Workshop discussions so that the number of participants was higher than expected. The NATO Advanced Research Workshop generated important interactions and provided an opportunity for scientists of different nations and of varied disciplines to discuss the challenges of spent nuclear fuel that confront operators of nuclear reactors around the world. The papers presented at the Workshop are published in this NATO Science Series. The general areas discussed at the Workshop included: the strategy and legal aspects of spent nuclear fuel storage; general radiological problems of spent nuclear fuel storage; the technical issues of wet and dry storage; and materials stability issues of spent nuclear fuel storage. Topics covered in these general areas were varied and ranged from the development of a pilot plant for the melt-dilute treatment of highenrichment fuel to the basics of Al corrosion in wet storage. No burning safety-related issue of spent fuel storage emerged from the total of the 22 papers presented at the Workshop. A common feature at many reactors, however, was the absence of space to store spent fuel, particularly in the original fuel storage pools. This has increased attention on using dry storage and all countries have research going on in the area. Participants agreed on the usefulness of the Workshop and the hope was expressed that the ARW was the first of several on this important topic.
ix
ACKNOWLEDGEMENTS
The NATO Advanced Research Workshop “Safety Related Issues of Spent Nuclear Fuel Storage” organized in Almaty from September 26 to 29, 2005, generated important interactions and provided an opportunity for scientists of different nations and of varied disciplines to discuss the challenges of spent nuclear fuel that confront operators of nuclear reactors around the world. The co-organizers of the ARW, J. D. B. Lambert, USA, and K. K. Kadyrzhanov, Kazakhstan, are grateful for the financial support from the North Atlantic Treaty Organization (NATO) under the Science Program in the area of Security-Related Civil Science and Technology (EAP.ARW.981019), with special thanks going to K. Gardner and F. Carvalho Rodrigues of the Public Diplomacy Division. The co-organizers are also grateful to the International Atomic Energy Agency (IAEA) for its financial support of experts from non-NATO countries, with special thanks to M. Samiei of the Department of Technical Cooperation. Without such support, this ARW and the exchanges among the Workshop participants would not have been possible. Similarly, the co-organizers must thank the NATO Scientific Affairs Division for permission to publish the proceedings in the NATO Science Series, because only with publication will the Workshop be known to the wider audience it deserves. The co-organizers would like to express their gratitude to the speakers from Brazil, France, Kazakhstan, Poland, Russia, Ukraine, USA, Uzbekistan, and the IAEA for writing their technical papers and presentations, and to everyone who attended the ARW and contributed to the discussions. The Institute of Nuclear Physics near Almaty provided excellent meeting facilities and arrangements. The staff of the Institute of Physics—ably led by Oksana Tivanova—who handled all the arrangements for the meeting, transportation, and lodging in Almaty, contributed greatly to the success of the ARW. The workshop presentations and discussions were much enhanced by the presence of the professional interpreters Artem Yermilov and Ljudmila Trautman, whose work was quite outstanding.
xi
STRATEGIES FOR SAFE STORAGE OF SPENT FUEL
SPENT NUCLEAR FUEL FROM RESEARCH REACTORS: INTERNATIONAL STATUS AND PERSPECTIVES
P. ADELFANG∗, A. J. SOARES, AND I. N. GOLDMAN Division of Nuclear Fuel Cycle and Waste Technology International Atomic Energy Agency Vienna, Austria Abstract: The back-end of the research reactor (RR) nuclear fuel cycle is not only a technical issue. Non-proliferation, physical security, and environmental concerns are just as important, if not more so, as technical concerns such as: safe management of spent nuclear fuel (SNF), storage capacity, availability of qualified high-density reprocessable fuel, and national self-sufficiency to deal with the domestic turnover of the research reactor’s spent nuclear fuel (RRSNF). International activities in the back-end of the RR nuclear fuel cycle are dominated by two important undertakings. The first is the Reduced Enrichment for Research and Test Reactors (RERTR) programme, and the second is the acceptance of RRSNF by the country where it was originally enriched. Both programmes aim to eliminate the utilization of highly enriched uranium (HEU) in RR. However, when these programmes have achieved their goals and there are no more HEU inventories at RRs and no more commerce in HEU for RRs, it is almost certain that the take-back programmes will cease. Many countries with one or more RRs and no nuclear power programme will have to face the problem of final disposition for relatively small amounts of spent fuel or permanently shut down their RRs before the termination of the take-back programmes. Regional or international solutions would seem to be the only chance of survival for the RRs in those countries. Access to a multinational long-term interim storage facility and eventually a multinational repository is an ideal and acceptable solution. Keywords: research reactors, spent fuel, RRDB, RRSFDB, RR fuel cycle
______ ∗
To whom correspondence should be addressed: P. Adelfang, Nuclear Fuel Cycle and Materials Section, Division of Nuclear Fuel Cycle and Waste Technology, International Atomic Energy Agency, Wagramer Strasse 5, P.O. Box 100, A-1400 Vienna, Austria; e-mail:
[email protected] 3 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 3–25. © 2007 Springer.
4
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
1 Introduction For over 60 years, research reactors (RRs) have made valuable contributions to the development of nuclear power, basic science, materials development, radioisotope production for medicine and industry, and education and training. However, during the last two decades RRs have been under strong pressure from the governments and/or their operating organizations to improve utilization and be able to obtain at least a fraction of their operating budget from selling services to external customers. Consequently, many facilities are challenged to find users for their services, or to permanently shut down and to eventually decommission. Unfortunately, many RRs worldwide are underutilized. Only reactors with special attributes, such as a high neutron flux, a cold source, in-core loops to simulate power reactor conditions, special features for training and education, or with commercial customers, such as radioisotope production or silicon doping, are efficiently utilized. During the last two decades many more reactors have been shut down (but only a fraction really decommissioned) than built. If no action is taken in the appropriate time, some countries with RRs and no nuclear power programme will face problems with spent fuel management after the decommissioning phase. Activities in the area of management of spent nuclear fuel (SNF) from research and test reactors are dominated by two important programmes. The first is the Reduced Enrichment for Research and Test Reactors (RERTR) programme, and the other is the acceptance of RR spent fuel by the country where it was originally enriched. Both programmes aim to eliminate the utilization of highly enriched uranium (HEU) in RRs. However, it is recognized that after 25 years of the RERTR programme, over one-third of all stored spent fuel assemblies are still HEU. It is also recognized that the acceptance programmes will not continue indefinitely, and it is not ethical to perpetually postpone a final decision for the RR spent fuel. Thus the time is ripe for a serious discussion about the options for dealing with the research reactors’ spent fuel (RRSF), in national, regional, and international scenarios. 2 General Status of RRs Most of the information presented in this section is taken from the International Atomic Energy Agency’s (IAEA) RR Database (RRDB). The RRDB was last updated in 2004. The IAEA is currently distributing the questionnaires to the owners and operators of RRs requesting an update on information relating to each facility.
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
5
The RRDB contains information on 657 RRs that have been built since the Chicago graphite pile CP-1 went critical in December 1942. From these, 274 are operational in 57 countries, 214 are shut down, and 168 have been decommissioned. Table 1 presents the distribution of RRs, categorized by operational status, between developed and developing countries. TABLE 1. Operational status of research reactors Country United States Russian Federation Other developed countries Developing countries Total
Operational
Shutdown
Decommissioned
Under construction
Planned
Unverified
52
107
68
0
0
57
28
11
1
0
79
52
75
3
2
86
27
14
6
4
1
274
214
168
10
6
1
The breakdown of the 657 reactors by operational status reveals that 382 have been shut down, but only 168 have been decommissioned. It is a serious concern that many of the shut-down but not decommissioned reactors still have fuel, both fresh and spent, at the sites. An extended delay between final shutdown and decommissioning will certainly affect both cost and safety at the time of decommissioning, mainly due to the loss of experienced staff (already ageing at the time of shutdown) necessary to participate in decommissioning activities. The distribution of the number of countries with at least one operational RR, as shown in Figure 1, peaked in 60 countries in the mid-1980s, in coincidence with the peak at 41 for developing countries. The number of countries with at least one RR remained almost constant for industrialized countries from 1965 and for developing countries from 1985 to the present.
Figure 1. Countries with operational research reactors
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
6
Figure 2 indicates that the number of RRs in industrialized countries peaked in 1975 and has declined since then. The number in developing countries has gradually increased, but changed little since the mid-1980s. About 70% of all operating RRs are in industrialized countries. As we can see in Figure 2, from 1975 until 2000, many more reactors have been shut down than have been commissioned. At present, 10 new RRs are under construction and 6 more are planned. It is understood that these new reactors are mostly innovative, multipurpose reactors, and/or with high neutron fluxes, which can address many present research and development (R&D) needs. developing countries
industrialized countries
400
393
383
# of reactores
375
350 326 300
328
328
total
346 314
303
289
285
257
250 173 155
150 100
0
274
188
188
85
87
219
200
50
273
55
73
79
86
89
84
41 39 38 18 15 14 76 1 1 1 Year 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Figure 2. Number of research reactors during the last 60 years
The distribution of RRs among IAEA member states is displayed in Table 2. The USA and the Russian Federation are member states with the largest numbers of RRs, with roughly equal percentages of all operational reactors. TABLE 2. Operational research reactors in IAEA member states Member state USA Russia Japan France Germany
Reactors Number % 52 57 16 15 13
19 21 6 5 5
Member state China Canada UK Other industrialized Other developing
Reactors Number % 14 8 3 26 70
5 3 1 9 26
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
7
The evolution of the number of RRs and their steady thermal power is presented in Figure 3, where we can see that the integral steady thermal power peaked in 1980. Analysing Figures 2 and 3, we conclude that the significant number of reactors permanently shut down in recent years is mainly in the low power range, and in industrialized countries.
Figure 3. Growth in research reactors
The age distribution of operational RRs peaks in the range of 40 years, as shown in Figure 4 and Table 3. They show that about 65% of operating reactors are over 30 years old since commissioning. It is well recognized that although a few of these old reactors invoke safety concerns, most of them have been refurbished, so that the key components meet or even exceed modern safety standards.
Figure 4. Age distribution of research reactors
8
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
TABLE 3. Age distribution of operational research reactors Reactor age
Reactors Number %
Reactor age
Reactors Number %
0–10 years
14
5
31–40 years
90
33
11–20 years
30
11
41–50 years
85
31
21–30 years
52
19
>50 years
3
1
The thermal power distribution of operating RRs, shown in Figure 5 and Table 4, indicates that a large fraction of RRs (about 77%), have thermal power lower than 5 MW, and for 50% of them the thermal power is lower than 100 kW, indicating that they operate with a lifetime core and no spent fuel problems are expected to arise until these reactors are permanently shut down.
Figure 5. Power distribution of operational research reactors
TABLE 4. Power distribution of operational research reactors Power range <1 kW 1–100 kW 0.1–1 MW 1–5 MW
Reactors Number % 74 27 66 24 41 15 32 12
Power range 5–20 MW 20–100 MW >100 MW
Reactors Number % 33 12 24 9 4 1
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
9
3 The IAEA Research Reactor Spent Fuel Database Most of the information presented in this section is taken from the IAEA’s RR Spent Fuel Data Base (RRSFDB), as of September 2003—date of last updating. The IAEA is currently distributing the questionnaires to the owners and operators of RRs requesting an update on information of spent fuel relating to each facility. RRSFDB contains 210 entries. Of these RRs, 44 are permanently shut down, 14 are temporarily shut down for refurbishment, 4 are planning to shut down, 3 have unverified information on status, and the remaining 145 are operational. Spent fuel is usually an ongoing liability after a reactor is shut down and the IAEA would like to include details of spent fuel, if it has not been reprocessed, from all of the known 214 shut-down reactors not yet decommissioned reported in RRDB. In addition, there is a large discrepancy between the 274 operational reactors in RRDB and the 145 reactors that have so far responded to the questionnaires for RRSFDB. Fortunately, most research and test reactors with substantial turnover of fuel and, hence, significant inventories of spent fuel, are included in RRSFDB. Nevertheless, it is essential for the IAEA to get a clear and accurate picture of the problems faced by RR operators and their concerns about management, storage, and ultimate disposal of spent fuel, in order to be able to address them and to exert pressure internationally for the implementation of spent fuel take-back programmes by supplier countries, and to begin a dialogue about possible regional repositories as an ultimate solution for countries with no nuclear power programme. The next section deals with numbers of fuel assemblies, their types, enrichment, origin of enrichment, and geographical distribution among the industrialized and developed countries of the world. All values are based on the information actually available in the IAEA RRSFDB. 4 Accumulated Spent Fuel Most RR fuels are shipped in assembly form. For this reason, spent fuel numbers in RRSFDB are recorded in assemblies, where a fuel assembly is defined as “the smallest fuel unit that can be moved during normal reactor operation or storage”. Even so, questions regarding numbers of fuel assemblies obviously caused confusion to respondents to the questionnaires. Consequently, the data received has been reviewed and corrected by a panel of experts who know the details of the various fuel assembly designs. At any particular facility, several different spent fuel types or spent fuels of different enrichments are usually stored. For example, the store may contain one or more types of HEU from prior to core conversion and one or more types of LEU following conversion. Several facilities report more than three types of spent fuel and for this reason the records in RRSFDB store up to ten fuel types per facility. Strictly
10
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
speaking, fuels enriched to ≥20% 235U are classified as HEU. Since many facilities with LEU cite a nominal enrichment of 20%, we have modified the definition of LEU to be ≤20% 235U for the purposes of RRSFDB. Since any fuel with exactly 20% enrichment before irradiation will have <20% enrichment after significant burn-up, this does not violate the accepted definition. The distribution of fuel types among the reactors in the RRSFDB is shown in Table 5. Although the majority are of MTR, TRIGA, or standard Russian types, a significant percentage (28%) are classified as other types, which underlines the fact that many experimental and exotic fuels exist at RRs around the world, posing problems for their continued storage, transportation, and ultimate disposal. TABLE 5. Distribution of reactors by fuel type Fuel type
Reactors using fuel type Number
Percentage
MTR
67
32
TRIGA
40
19
Russian
43
21
Other
58
28
According to the questionnaires received, the RR spent fuel inventories worldwide can be summarized as follows1,2: • 62,027 fuel assemblies in storage • 45,108 in industrialized countries • 16,919 in developing countries • 21,732 HEU assemblies • 40,295 LEU assemblies The regional distribution of spent fuel inventory, with the origin of the enrichment broken down into fuel from the USA, Russia, and others is shown in Table 6. In this case, “others” include China, France, the UK, and South Africa, natural uranium fuels and those cases where the origin of enrichment is not known or simply left blank on the questionnaire. As expected, the USA supplied most of the enriched fuel to RRs in North America and Asia-Pacific, while Russia (or the former Soviet Union) supplied most of the enriched fuel to RRs in Eastern Europe. It is important to notice that according to the numbers shown in Table 6, more than 17,000 fuel elements may not be eligible by the existing acceptance programmes.
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
11
TABLE 6. Distribution of spent fuel by supplier country Region Africa and Middle East Asia-Pacific Eastern Europe
USA
Russia
Other
247
37
383
5,327
372
2,926
11
19,137
0
Region Latin America North America Western Europe
USA
Russia
Other
217
5
29
2,974
0
542
5,450
10,636
13,734
The regional breakdown of US-origin and Russian-origin fuel, classified as HEU or low enriched uranium (LEU), is shown in Table 7. Of interest in this table is the fact that HEU outweighs LEU in North America, whereas the reverse is true in Western Europe. To some extent this is because more RRs in Western Europe have undergone core conversion than is the case in North America. It is worth noting that a significant fraction of Russian-origin HEU was originally enriched to only 36%, while most US-origin HEU was originally enriched to ≥90%. Also, the inventory of spent fuel assemblies of Russian origin contains many thousands of small, spent fuel slugs and EK-10 elements, which considerably inflates the numbers of spent fuel assemblies of Russian origin compared with those of US origin. In fact, the amounts of 235U involved are very much bigger in the US inventories at foreign RRs compared with the Russian equivalents. This is particularly true of the respective amounts of HEU. TABLE 7. Distribution of US- and Russian-origin spent fuel by enrichment* Region
US
Russian
Region
US
Russian
Africa and Middle East
189 58 1,527 3,800 11 0
0 37 94 278 11,036 8,101
Latin America
109 108
North America
1,639 1,335
Western Europe
2,442 3,008
0 5 0 0 2,208 8,428
Asia-Pacific Eastern Europe *HEU LEU
At present 12,850 spent fuel assemblies of US-origin are located at foreign RRs, while the equivalent number of Russian-origin is 24,803.
12
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
5 Need for Multinational Solutions to the Back-end of the RR Nuclear Fuel Cycle After discharge from the core, RRSNF is stored under water, usually in at reactor (AR) facilities, for a period of time to allow for cooling. This wet storage can be extended, in AR or away-from-reactor (AFR) facilities, in some cases over long periods (~50 years), or the RRSNF may be transferred to dry storage (AR or AFR) sites and stored dry for even longer periods. None of the above-mentioned strategies for RRSNF management can be considered as the end point of the RR fuel cycle. Regardless of how long this extended interim storage is drawn out; the resolution of the back-end problem will remain, and proliferation, safety, and physical security concerns will continue. Considering current technologies, the end point of the RR nuclear fuel cycle is attained (at least for the generating country) when the RRSNF is: (a) Returned to the country of origin (b) Reprocessed, the HL and long lived wastes disposed of in a geological repository, and the useful isotopes reused (c) Directly disposed, or disposed of after conditioning, in a geological repository We shall point out that the main characteristic of options (b) and (c) is the irretrievability of the disposed materials Perpetual postponement of a final decision regarding the end point for the RR spent fuel produced nowadays is not ethical or technically reasonable considering that after achieving their goals the take-back programmes will certainly cease. Then every country with a RR will face the necessity to develop a national strategy for disposal of RRSNF, and for countries with small nuclear power programmes (with one or two RRs and no nuclear power plant) the expensive construction of away-from reactor extended interim storage facilities and/or geological repositories for the relatively small amounts of spent fuel accumulated is obviously not practicable. Unfortunately, for many of these countries the option of reprocessing the fuel abroad is unlikely to be affordable. Moreover, if the fuel is shipped abroad for reprocessing, the problem of the final disposal of any returned high-level waste (HLW) will have to be addressed anyway. Therefore, sooner or later, every country with at least one RR, which continues to operate beyond the termination of acceptance programmes of the countries of origin, will need to work on a final solution for spent fuel and/or HLW. Clearly, access to a multinational long-term interim storage facility and eventually a multinational repository is an ideal and acceptable solution. International interest in multinational solutions has been expressed in two major events:
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
13
1. The International Conference on Storage of Spent Fuel from Power Reactors, held in Vienna, 2–6 June 2003, organized by the IAEA in cooperation with the OECD Nuclear Energy Agency, in which, one of the conclusions was: “Representatives of Member States with smaller nuclear programmes informally expressed continued interest in regional storage initiatives, as well as topic specific workshops and training courses”. 2. The statement of the Director General of IAEA Dr. Mohamed ElBaradei during the Forty-seventh Regular Session of the IAEA General Conference 2003: Our consideration should also include the merits of multinational approaches to the management and disposal of spent fuel and radioactive waste. Not all countries have the appropriate conditions for geologic disposal and, for many countries with small nuclear programmes for electricity generation or for research, the financial and human resource investments required for research, construction and operation of a geologic disposal facility are daunting. Considerable economic, safety, security and non-proliferation advantages may therefore accrue from international co-operation on the construction and operation of international waste repositories. In my view, the merits and feasibility of these and other approaches to the design and management of the nuclear fuel cycle should be given in-depth consideration. The convening of an Agency group of experts could be a useful first step. The possibility of a regional and multinational repository is also addressed by the Joint Convention on the Safety of Spent Fuel Management and on the Safety of Radioactive Waste Management, which states in its article (xi): “Convinced that radioactive waste should, as far as is compatible with the safety of the management of such material, be disposed of in the State in which it was generated, whilst recognizing that, in certain circumstances, safe and efficient management of spent fuel and radioactive waste might be fostered through agreements among contracting parties to use facilities in one of them for the benefit of the other Parties, particularly where waste originates from joint projects”. Many issues have to be discussed before a regional or multinational repository is established. The implementation of this option requires a significant political effort to resolve the complex array of agreements needed. However, since an international repository would resolve some of the problems being tackled by the Global Threat Reduction Initiative (GTRI), in the areas of nonproliferation and securing materials, it is understood that it could be pursued jointly by GTRI and IAEA.
14
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
6 Conceptual Multinational Solutions The term “multinational facility” means a facility in a country (host country or host), which serves several countries (partner countries or partners). This definition applies to storage facilities, geological repositories, and reprocessing plants. Different options for the services provided by the multinational facilities might be conceived. The one presented in Figure 6 is a fully integrated approach that considers the classical alternatives for the RRSNF back-end (reprocessing or direct disposal). After transportation from the partner country, the spent fuel is stored for a specified period in the multinational storage facility, after which it is transferred either to a multinational disposal facility (after appropriate conditioning) in the hosting country or to a reprocessing plant. The HLW from the reprocessing plant is finally disposed of in a multinational repository. Needless to say, the convenience of locating all the necessary facilities (storage, reprocessing, and repository) in the same host country and even in the same site would be logical and economically desirable.
Figure 6. Multinational back-end system
Typical situations from which a multinational approach might develop are: • • •
Several industrialized countries with relatively small nuclear energy programmes decide to cooperate in the management of the back-end of the nuclear fuel cycle. A country with a large nuclear energy programme offers back-end services to other countries with a limited production of RRSNF. Countries with small nuclear energy programmes in varying stages of development seek assistance from each other. Among other issues would be to develop a suitable and common back-end option.
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
15
Countries likely to function as hosts are those with existing reprocessing facilities, advanced disposal programmes, and/or favourable geological sites that see the tremendous commercial opportunities. Partners would be countries with small nuclear power programmes or just RRs that cannot realistically develop a national final solution and countries that see an economic and/or political advantage in joining a multi-national undertaking. 7 Important Issues to be Considered in a Multinational Approach 7.1 LEGAL ASPECTS The legal and regulatory situation in countries willing to consider a multinational solution should be harmonized among the partners. Mature and stable regulatory frameworks should be developed and, where existing regulations are inadequate or insufficient, the use of existing international conventions, e.g., the Joint Convention on the Safety of Spent Fuel Management and on the Safety of Radioactive Waste Management3 would help in bridging the gaps. 7.2 SAFETY PRINCIPLES Safety criteria should comply with international standards. A key advantage of the concept of the multinational approach is to reduce the number of locations at which radioactive materials are stored and disposed of. A thorough risk assessment has to be performed. 7.3 TECHNICAL ISSUES Current inventories of all RRNSF available for management must be established before serious consideration can be given to establishing a multinational undertaking. There should be agreement between the host country and its partners as to acceptance criteria and quality assurance and quality control (QA/QC). The characterization and selection of a site as well as the design, construction, and operation of the required facilities should be agreed upon. The application of well-established technologies should be preferred. 7.4 ECONOMIC ISSUES Multinational approaches will have a main economic impact in reducing the expenditures of the partners and increasing the resources of the host country. Many particular concerns should be taken into account. The costs and liabilities
16
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
to all affected partners must be weighed against the benefits. As cost sharing will extend over many years, long lasting financial arrangements are thus unavoidable whether the project will be run and financed jointly or whether the non-host partners only play the role of customers. Financial provisions for future liability of the host country have to be considered seriously. 7.5 INSTITUTIONAL ASPECTS AND POLITICAL CONTINUITY The considerations about cost, liability, safety regulations, etc. are closely linked to the institutional character of the project that involves national and multinational relations among regulatory and licensing bodies, as well as with contractual partners. Since a multinational undertaking may extend over decades or centuries, it may be run under an international convention or agreement. The political stability of the host and the partners is again a vital element. 7.6 OWNERSHIP OF RRSNF Ownership of RRSNF requires early negotiations between the countries participating in a multinational project. There is a strong interrelation between ownership of RRSNF and liability. Partners involved have to agree when (if ever), in the process, ownership is transferred to the host country operating the multinational facilities and on the full implications of the transfer. 7.7 ETHICAL ASPECTS The ethical considerations are embodied in the IAEA’s Safety Series publication No. 111-F “The Principles of Radioactive Waste Management”, in particular with regard to the protection of human health and the environment, with emphasis for the protection of future generations, the protection of third countries/parties beyond national borders, and the principle of avoiding undue burdens on future generations. Equity must apply amongst the partner countries, that is, a fair balance must exist between the burden transferred and the compensation received through the multinational agreement. 7.8 PUBLIC ACCEPTANCE The public acceptance issue is inevitable and crucially important for multinational projects, serving several countries or communities. High safety standards, quality assurance on conditioning and disposal, cost sharing, transparency with regard to coverage of potential future costs, clear and convincing answers with regard to ethical concerns, etc. are thus essential in the process of obtaining public acceptance of a multinational project.
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
17
7.9 SAFEGUARDS RRSNF contain fissile materials that according to the terms of the Treaty on Non-proliferation of Nuclear Weapons, are subject to national and international safeguard regulations. Therefore, well-defined national and international safeguards regulations will have to be applied. Maintenance of long-term controls should be assured. One obvious advantage is that there will be fewer facilities to be safeguarded and physically secured. 8 A Success Story: The Regional Project IAEA-RLA/4/018-Management of Spent Fuel from Research Reactors in Latin America The countries of Latin America, including Argentina, Brazil, Chile, Mexico, and Peru, recognize their responsibility of safely and securely managing the spent fuel from their RRs that have been in operation for several decades. These countries share the view that it is essential and opportune to begin the evaluation of options for interim storage and disposal of their spent fuel, including any derivatives, if processing is one of the chosen options. These facts were the driving force for initiating the IAEA Technical Cooperation Regional Project RLA/4/018 “Management of Spent Fuel from Research Reactors”. The main objective of the Regional Project RLA/4/018 was to define the basic conditions for a regional strategy for managing RRSF, providing solutions within the economic and technological realities of the countries involved, and in particular, to determine the needs for the temporary storage of spent fuel from the RRs in the countries that participated in the Project. Table 8 shows the main characteristics of each of the reactors built in the countries that participated in the regional project RLA/4/018.4 TABLE 8. Research reactors and countries in IAEA Regional Project RLA/4/018 Country Argentina
Facility name RA-0
Power (kW) 0.01
Argentina Argentina Argentina Argentina
RA-1 RA-2 RA-3 RA-4
40 0.03 5,000 0.00
Argentina
RA-6
500
Reactor type ZPR Tank Tank ZPR MTR Pool ZPR Homog. MTR Pool
Reactor status Operational
Fuel type Pin
Critical (mm/yy) 01/65
Operational Shut-down Operational Operational
Pin MTR MTR Disk
01/58 07/66 05/67 01/72
Operational
MTR
09/82
(Continued)
18
Argentina Brazil Brazil Brazil Brazil Chile Chile Mexico Mexico Mexico Mexico Peru Peru
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
RA-8 IEA-R1 IPR-R1
0.01 5,000 250
ARGONAUTA IPEN-MB RECH-1 RECH-2 CHIMod 9000 CHIMod 2000 SUR-100 TRIGA Mark III RP-0 RP-10
Operational Operational Operational
Pin MTR Rods
06/97 09/57 11/60
0.50
ZPR Tank MTR Pool TRIGA Mark 1 ARG
Operational
MTR
02/65
0.10 5,000 2,000 0.00
ZPR Tank MTR Pool MTR Pool Subcr.
Operational Operational Shut-down Operational
Pin MTR MTR Pin
11/88 10/74 02/77 05/69
0.00
Subcr.
Operational
Pin
01/69
Decommissioned
Disk
09/72
Operational
Rods
11/68
Operational Operational
MTR MTR
07/78 11/88
0.00
ZPR Homog. 1,000 TRIGA Mark III 0.00 ZPR Tank 10,000 MTR Pool
To meet the proposed objectives, the project was organized into 4 main areas: Spent Fuel Characterization, Nuclear Safety and Regulation, Public Communication Strategy, and Options for Spent Fuel Management. The activities developed in the area of “Spent Fuel Characterization” aimed at the implementation of a surveillance programme for RRSF in interim storage in every participant country, including methodologies to perform: water quality control, visual inspection, sipping tests, corrosion monitoring by coupon assessment, and non-destructive burn-up test. In addition, a common database on inventories of fuel in the region was also developed, for the use of the participant countries and IAEA, and a burn-up exercise was performed in order to intercompare the codes and methodologies used by the reactor physicists who are responsible for the fuel management in their respective countries. On the “Nuclear Safety and Regulation” area the activities were developed toward the adoption of safety criteria for RRSF management activities, such as transport and interim storage, within the region. This activity pointed to the formalization of regulation documents, compatible with the regional realities, to ensure agreement to international standards, and guidelines, mainly from IAEA. Four documents, known as “recommendations”, were produced within this activity: 1. Recommendations for design of installations for interim storage of RRSNFs 2. Recommendations for the safety analysis of installations for interim storage of RRSNFs
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
19
3. Recommendations for the transport of RRSNFs 4. Recommendations for design of casks to transport RRSNFs The four documents were produced in Spanish and actually are being translated to Portuguese. The activities in the “Public Communication Strategy” area were structured in order to be a way of implementing institutional communication strategies associated with spent fuel management in the region, adapted to the diverse sites and possible scenarios, to improve the public’s perception, and to help gain public acceptance that will allow the safe and secure management of the spent fuel from RRs in the region. This activity was considered one of the most important in the project, because it is directly related to public acceptance of the proposed alternatives for storage and disposal of the spent fuel from RRs, and we recognize that public opinion is mostly negative regarding any possibility of disposal (particularly geological) of radioactive waste. In this regard, a continuous and effective public information programme in each country must be implemented in order to counter this perception. This modification of public attitudes is necessary prior to site selection and construction of any disposal facility. One of the activities in this area was the elaboration of a brochure to be distributed to politicians, media professionals, professors, and interested persons. The purpose of the brochure was to describe the benefits provided by RRs, with a description of all RRs in the region, and with an introduction of the spent fuel problem. The brochure included a questionnaire for the receivers to produce some feedback information to be used in future activities. A total of 10,000 copies were produced, 8,000 in Spanish and 2,000 in Portuguese. According to the strategy defined, the brochure, and the questionnaire, should be mailed with a letter signed by the highest possible national authority related to nuclear activities in the country. Due to some reorganization in some national nuclear institutions, the distribution process could not be completed in all countries before the end of the project. The area of Option for Spent Fuel Management was the core of the RLA/4/018 Project. Its purpose was to identify and assess all technically viable options for RRSF management, particularly for operational and interim storage (wet and dry); spent fuel transportation, including the development of a dual purpose cask); spent fuel conditioning, and final disposal. Moreover, it aimed to be the main source of information to design and implement related facilities. These options were presented in view of the economic realities, in the national, regional, and extra-regional contexts for the countries that participated in the project, and which currently operate nuclear RRs for radioisotope production, fundamental research in physics and biology, materials irradiation, and education and training. The final document, which is to be published by the IAEA as a TECDOC, describes the expert opinions on all technically feasible
20
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
options available, without excluding or failing to evaluate any option that might be impossible because of current national laws of the participating countries at the time of writing. The participants of the project understood that it may be approximately 50 years before there is any deep geological repository available in the region, and in the interim there will certainly be technological advances, political and legal changes, as well as transformation of public attitudes that may become favourable for regional options to be implemented. Therefore, the document produced is assumed to be a technical guide for decision makers of Latin American region and is expected that it could be revised and updated from time to time. The main outcomes of the “Option for Spent Fuel Management” area were: 1. Experts of countries participating in the project have defined the best option to be applied in the short and medium range to the reality of each country. Argentina will use a 16 m deep stainless steel pool, available in a postirradiation test facility, to store all spent fuel elements produced in the RRs of the country. Brazil, Chile, Mexico, and Peru are strongly considering the possibility of using interim dry storage facilities. The engineers and researchers of Brazil, Chile, and Mexico started analysing the feasibility of dry storage casks, and Peruvian engineers and researchers have started the conceptual design of a dry storage facility. 2. A dual purpose cask, for transport and storage, was designed as part of the project. The cask was dimensioned for 21 MTR type and 78 TRIGA type spent fuel elements, and Figure 7 shows its conceptual design. Actually a half-scale model of the cask is being built, and the plans are to test it by the end of 2005. 3. Development of activities related to processing and conditioning of spent fuel. This activity was assumed mainly as R&D in the areas of the HALOX processing method (a dry decladding process developed by Argentine researchers), in electrochemical separation (by Brazilian researchers) and in conditioning/immobilization. Regarding “conditioning/immobilization”, at the beginning of 2003 a joint programme was established, between researchers of Brazil and Argentina, to study the behaviour of some matrices proposed for immobilization of HLW. Within this programme, seven matrices were selected. The matrices were produced and characterized in both laboratories, and at Centro Atomico la Reina, in Chile, which has the necessary infrastructure to accelerate the corrosion process of glasses using a technique known as “Vapor Hydration Test” (VHT). The work developed in Argentina is mainly related to the utilization of iron phosphate glasses, and the work developed in Brazil is related to the utilization of niobium phosphate glasses. 4. The development of the spent fuel disposal model. This exercise was developed to quantify the amount of spent fuel, and its derivatives, that should be
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
21
Figure 7. Conceptual design of a dual purpose cask
managed assuming that all RRs in the region will continue operation for the next 30 years. The main assumptions taken into account for the modelling were: •
• • • • •
•
The regional countries share a common strategy for spent fuel processing and join in a partnership to install a conditioning plant in one of the partner’s countries. The plant will process the whole spent fuel inventory (turnover and backlog). The annual spent fuel turnover for the countries of the region corresponds to the estimated operation regimes of the RRs in a realistic scenario, as shown in Figure 8. The RRs of the region will operate for 30 years more. Burn-up of 50% in 235U and cooling time of 2 years were considered for the spent fuel, in order to get conservative results for radionuclides quantities, alpha activity, and thermal loading calculations. The structural components of the spent fuel assemblies—upper and bottom ends, side plates, etc.—are removed before conditioning. Thus, just the fuel elements (plates) are conditioned. A uranium isotopic dilution down to 1% enrichment of 235U is necessary in order to match the final enrichment of spent fuel from nuclear power plants intended for direct disposal. Such isotopic dilution is achieved with natural uranium. Immobilization is carried out with a phosphate-based glass and an assumed waste loading of 15 wt% reaching a density of 2.85 kg/dm3.
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
22
Extra-regional host country
Regional host country
53 SFA/yr 1 69.3 kgU/yr 8.71 kg235U/yr 394 gPu/yr 297 g239Pu/yr 255 kgAl/yr
Regional partners
Country 1 4 SFA/yr 4.07 kgU/yr 0.498 kg235U/yr 20.3 kgAl/yr
Country 2 Regional conditioning plant with operational storage Estimated capacity 100 SFA/yr (1)
22 SFA/yr 28.9 kgU/yr 3.96 kg235U/yr 103 kgAl/yr
Country 3 25 SFA/yr 33.8 kgU/yr 3.97 kg235U/yr 121 kgAl/yr
LILW-LL 20.9 ton/yr 7.33 m3/yr International deep geological repository
VLLW 60.4 kg Al/yr 0.0223
Country 4 2 SFA/yr 2.58 kgU/yr 0.281 kg235U/yr 10.6 kgAl/yr
Figure 8. Flow diagram for a regional RRSF conditioning plant
1
______ 1
The estimated capacity was established as 100 SFA/year to allow the processing of the spent fuel already stored in the reactors pools.
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
23
The spent fuel turnover is described by the quantities of total U, 235U, total Pu, 239Pu, and Al shown in Figure 8. The output of the regional conditioning plant, presented in Table 9, is described by the quantities of total immobilized spent fuel derivatives (LILW-LL waste form), total U, 235U, total Pu, 239Pu, Al, fission products and minor actinides (Np, Am, and Cm) as well as the volume of the LILW-LL waste form. Additionally, the activity of alpha emitters in the immobilized waste form and its thermal load, have been calculated. The quantity and the volume of the very low level waste (VLLW) Al stream, from the disassembled structural parts of the fuel assemblies, are also shown. According to the model and the assumptions made, after 30 years of operation we will have, in the region, about 1,600 spent fuel assemblies, which after dilution to 1% enrichment and subsequent immobilization will result in the output shown in Table 9. TABLE 9. Output of regional RRSF conditioning plant over 30 years Materials, elements, and nuclides
Mass
Volume
Immobilized SF derivatives (LILW-LL)
627 t
220 m3
P-based glass matrix
532.3 t
—
U
87.8 t
—
878 kg
—
11.8 kg
—
8.92 kg
—
Al
5.84 t
—
Fission products
156 kg
—
1.37 kg
—
1.81 t
0.670 m3
Composition:
235
U
Pu 239
Pu
Minor actinides (Np, Am, Cm) Structural Al from SF(VLLW) (2) (2)
If the structural Al parts of the spent fuel assembly are not removed in the partner countries of the region, immobilization in the main waste stream at the conditioning plant is the preferred approach
24
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
In Table 9, it is worth comparing the figures represented by fission products and minor actinides (about 157.4 kg), with the total 627 t of LILW-LL generated. 9 Conclusions Alternatives for the end point of RR fuel cycle must consider the possibility to use multinational deep geological repositories. It is assumed that any multinational facility will be subject to international conventions and internationally accepted standards involving safety, safeguards, physical security, and environmental protection Consequently, gathering together weapons-usable materials and radioactively hazardous wastes that are the inevitable products of the RR fuel cycle in preferably one (or a very small number) multinational, long-term interim storage facility, and eventually final repository, compared with the current situation where hundreds of locations are involved, has obvious benefits to all mankind. First and foremost among these will be nuclear threat reduction through reductions in proliferation risk and opportunities for theft of materials that could be used in radioactive dispersion devices. Real advantages can be identified also for the host country and participating countries. The only obvious drawback would be an increase in the transportation of fuel and HLW from the participating countries to the multinational facility. The IAEA should take the lead in initiating serious discussions of multinational solutions. The USA and Russian Federation, as the main suppliers of enrichment services to RRs, should also play important roles. There are some effective multinational spent fuel storage facilities for RRSNF existing at present: RBOF and L-Basin at Savannah River Site hold foreign RR fuel (originally enriched in the USA) from all over the world pending a decision on final disposition. Similar holding pools at reprocessing plants in France and Russia store RR fuel from foreign countries pending reprocessing, and eventually, Idaho National Engineering and Environmental Laboratory will be storing TRIGA fuel from 19 countries. Clearly, the technical problems of transportation and storage have been solved. Nevertheless, the most difficult obstacles—political willingness, legal issues of cost sharing, and liability and, of course, public acceptance—have yet to be addressed. Now is the time to start serious discussions.
WORLD STATUS OF RESEARCH REACTOR SPENT FUEL
25
References 1.
2.
3. 4.
Adelfang, P., Ritchie, I.G., “Back-End of the Research Reactor Nuclear Fuel Cycle: International Status and Perspectives” International Conference Research Reactor Fuel Management (RRFM’04), Munich, Germany, 21–24 March 2004. Adelfang, P., Ritchie, I.G., “Overview of the Status of Research Rectors Worldwide” International Meeting on Reduced Enrichment for Research and Test Reactors, Chicago, United States of America, 10 October 2003. Joint Convention on the Safety of Spent Fuel Management and on the Safety of Radioactive Waste Management. IAEA-INFCIRC/546, 24 December 1997. Options for Management of Spent Fuel from Research Reactors in Latin America Countries Participating in Regional Project RLA/4/018. To be published as IAEA TECDOC.
REGULATORY APPROACH FOR MANAGING RADIOACTIVE WASTE IN THE REPUBLIC OF KAZAKHSTAN
T. ZHANTIKIN∗ AND M. SHARIPOV Kazakhstan Atomic Energy Committee Almaty, Kazakhstan Abstract: This paper describes the regulatory approach that has been developed to manage radioactive waste in the Republic of Kazakhstan. Radioactive waste is a serious and dangerous problem for all human activity and specific management is necessary in order to reduce risk to facility personnel and to the general public from radioactive waste, and to reduce its impact on the environment. At present there are considerable quantities of radioactive waste in Kazakhstan and it is imperative to establish an adequate state-wide regulatory system to deal with it. Following a brief description of the distribution of radioactive waste in Kazakhstan, the legislative basis for state regulation and the role of various state bodies, including the Kazakhstan Atomic Energy Committee, are described. Keywords: uranium mining contamination, nuclear testing, radioactive waste (RW), legislation, regulation, supervision of RW
1 Introduction One of the necessary conditions for the evolution of civilization is increase in the use of energy. Nowadays, human requirements for thermal and electrical energy double every 12–15 years. New thermal and nuclear power plants are built, and mining operations are increased. As a consequence an intense contamination of the atmosphere takes place by chemical, thermal, and physical means, and different kinds of waste accumulate. This waste has a negative influence on both human beings and the environment.
______ ∗
To whom correspondence should be addressed: T. Zhantikin, Kazakhstan Atomic Energy Committee, Liza Chaikina 4, Almaty, Kazakhstan; e-mail:
[email protected] 27 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 27–36. © 2007 Springer.
28
MANAGING RADIOACTIVE WASTE IN KAZAKHSTAN
Radioactive waste is one of the most dangerous forms of waste for humans and living organisms because radiation can have detrimental effects on all life. Radioactive waste has a number of very specific characteristics and it demands special treatment, storage, and disposal. Systems for handling radioactive waste (RW) must be designed and operated with the aim of adhering to the principle of “as low as reasonably achievable” (ALARA) with regard to radiation. 2 Radioactive Waste in Kazakhstan Before outlining the state control system for RW let us point out the main sources of RW formation, its total amount of RW, and its main locations. RW in Kazakhstan includes: • • • • •
Waste from the uranium mining industry Waste from Soviet nuclear tests over 1950–1990 Waste from production of nuclear energy Waste from non-uranium branches of industry and Waste from enterprises that produce and use radioisotopes
Figure 1 shows the sources and distribution of RW in Kazakhstan.
Figure 1. Sources and distribution of radioactive waste in the Republic of Kazakhstan
MANAGING RADIOACTIVE WASTE IN KAZAKHSTAN
29
A rough estimate of the total quantity of RW is 220 million tons, including low and intermediate RW. The majority of RW—205 million tons—derives from uranium mining activities and is distributed as follows: • • •
About 25 million tons in the south part of Kazakhstan formed due to three sealed uranium deposits About 80 million tons in the northern part of Kazakhstan formed due to exploring and mining of uranium deposits and processing the uranium ores About 100 million tons in the western part of Kazakhstan formed due to exploitation of uranium deposits
There are about 67,450 m3 of low-level RW associated with the sites of Soviet nuclear testing. Bearing in mind the potentially dangerous nature of RW, its quantity and its wide distribution, it is clear that the corresponding regulation for handling RW should be carried out at the state level. 3 State Regulation of RW Handling One can say that the state regulation system for handling RW is now operative in the Republic of Kazakhstan. The state regulation system for RW handling— like any another activity bound up with the use of atomic energy—is based on the following elements: normative legal regulation, license regulation, and supervision regulation. Normative, legal regulation includes: • • • • •
Establishment of legal base of public relations regulation in the sphere of atomic energy use License regulation Supervision regulation Developing safety criteria Creation of the system of standards and safety rules Licensing includes:
• •
Examination and expertise of safety grounds Creation of license action conditions
The following kinds of activities are the components of state supervision on safe use of atomic energy: • •
Gathering and analysis of information about the safety status of atomic energy-related facilities Organization and performance of inspections, and analysis of their results
30
•
MANAGING RADIOACTIVE WASTE IN KAZAKHSTAN
Wherever necessary, the application of sanctions for violating safety requirements
3.1 NORMATIVE LEGAL REGULATION The following main laws of the Republic of Kazakhstan have been developed and used: • • • •
Law on Atomic Energy Use (№ 93-1) Law on Radiation Safety of the Population (№ 219) Law on Environment Protection (№ 160-1) Law on Licensing (№ 2200)
The law on Atomic Energy Use (1997) establishes the principle of state policy in the sphere of atomic energy use, rights, and responsibilities of state bodies, and also the mandatory conditions of activity bound up with the use of atomic energy. The law on Radiation Safety of the Population (1998) determines the principles for providing radiation safety, state management system, supervision, and control in the sphere of providing radiation safety, common demands for providing radiation safety. The law on Environment Protection (1997) determines the legal, economic, and social grounds of environmental protection of present and future generations and is aimed at providing ecological safety, prevention of the harmful influence of economic activity, preservation of biological diversity, and organization of a rational management of nature. Taking into account RW treatment Kazakhstan has signed “The joint convention on safety treatment with spent nuclear fuel and on safety treatment with RW”. Work is ongoing on a law to enforce this convention (Convention Ratification Law). 3.2 PROVISIONS FOR RADIOACTIVE WASTE DISPOSAL IN KAZAKHSTAN The Provisions for radioactive waste disposal in the Republic of Kazakhstan (№ 1283) were enforced by the Government Decree of 18 October 1996. The Provisions define the order for radioactive waste disposal in bowel, the procedure for obtaining permission from the regulatory bodies for its disposal in bowel and also establishes the list of necessary documents for this procedure. The Provisions also establish the authority of the various competent regulatory bodies as defined below.
MANAGING RADIOACTIVE WASTE IN KAZAKHSTAN
31
3.2.1 Kazakhstan Atomic Energy Committee The Kazakhstan Atomic Energy Committee (KAEC) carries out: •
•
Issuance of licenses to legal and physical entities on the right of fulfillment of work with RW including RW disposal (design of disposal sites, construction, commissioning and operation, special refinement, reprocessing, transportation, and disposal) Issuance of conclusions about possible future use of RW from disposal, except waste of radioactive balance minerals in tailings belonging to ore mining and exploration enterprises of the uranium industry
3.2.2 Ministry of Energy and Mineral Resources The Ministry of Energy and Mineral Resources of the Republic of Kazakhstan carries out: •
•
Issuance of licenses on the right for RW disposition for legal and physical entities and to construct near-surface and geological disposition facilities for storage/disposal of RW and also to issue licenses for parts of disposition works separated for recultivation of mine tailings, mining, and mining elaboration works Issuance of conclusions to interested legal entities regarding possible use of mine tailings for extraction of trace minerals
3.2.3 Ministry of Environmental Protection The Ministry of Environmental Protection of the Republic of Kazakhstan carries out: • • • •
State ecological oversight of RW storage/disposal sites projects Issuance of ecological conclusions on works, connected with RW storage/ disposal for getting license Issuance of permission on RW disposal Ecological control of RW storage/disposal sites
3.2.4 Ministry of Health The Ministry of Health of the Republic of Kazakhstan carries out: • •
State sanitary-hygienic oversight of construction projects for RW storage/ disposal sites State sanitary-hygienic control of RW storage/disposal sites
32
•
MANAGING RADIOACTIVE WASTE IN KAZAKHSTAN
Issuance of sanitary passports for RW storage/disposal sites
3.2.5 Ministry of Emergency Situations The Ministry of Emergency Situations of the Republic of Kazakhstan coordinates: • • •
Construction projects for RW storage/disposal sites, in which it is provided a heading of mounting framing, well-boring and other works RW technology reprocessing using blasting operations and virulently poisonous substances At the design stage, provision of supervision on technical safety of specialpurpose technological equipment (rolling-stock, vessels under pressure, tare, electrical equipment, and others) as control-measured apparatus supply
3.2.6 Ministry of Internal Affairs The Ministry of Internal Affairs of the Republic of Kazakhstan carries out: • •
Coordination of construction projects for RW storage/disposal sites Organization of defense and guarding of RW storage/disposal sites to interdict the theft of radioactive substances
3.2.7 Local executive authorities Local executive authorities (regional or district) of the Republic of Kazakhstan carry out: • •
Drawing up of an act on land place with district calculation of losses of agricultural and wood industry, protocols of agreement of interested sides Drawing up of a decision of withdrawal and grant of land place for radioactive waste storage/ disposal sites
3.3 OBLIGATIONS AND DOCUMENTS OF THE RW PROVISION The Provision defines the main obligations of physical and legal entities bound by its activity with RW. It defines the following list of main documents that regulate activities with RW: • • •
Statement of expending of radionuclide sources Statement of radiometric survey according to appendix of the acting regulations Balance sheet of minerals mining
MANAGING RADIOACTIVE WASTE IN KAZAKHSTAN
• • • • • • •
33
License to perform work with radioactive substances (waste) Statement of preliminary site selection for RW storage/disposal sites License for the right of disposition (only for near-surface and deep sites for RW storage/disposal) Design of construction of storage/disposal sites Decision on withdrawal and grant land for RW storage/disposal sites (only for near-surface and deep sites for RW storage/disposal) Emergency preparedness instruction for the population in the case of incidents or other extreme circumstances Logbook of radioactive waste
4 Role of the Kazakhstan Atomic Energy Committee The KAEC is part of a department of the Ministry of Energy and Mineral Resources of the Republic of Kazakhstan. The main task of the KAEC is to implement state supervision of the maintenance of nuclear and radiation safety on the territory of the Republic of Kazakhstan. One of the Committee’s functions is to implement the licensing of activities associated with atomic energy use and to control observance of normative legislation acts in the sphere of atomic energy use by physical and legal entities independently from their department subordination and ownership form. 4.1 APPROVAL OF NORMATIVE LEGAL DOCUMENTS The Committee approves the list of the main normative, legal documents in the sphere of atomic energy usage. In the list there are about 200 main documents which regulate safety for all kinds of activities in atomic energy use, including RW treatment. Below one can see short characteristics of the main documents establishing the common demands on accident prevention with RW treatment. Norms of radiation safety (NRB-99) put into force by resolution of the main state sanitary inspector of the Republic Kazakhstan by № 10 from 09.12.09. The present norms are developed with account of international main safety norms for protection from ionized radiation adopted jointly by Food and Agriculture Organization of the UNO, IAEA, Organization of Economic Collaboration and Development, Pan-American and World-wide Health Organizations (safety series N115, 1996). The present norms are applied for human safety guaranteed at any influence of natural or human-caused radiation. Norms provide the main dose limits, allowable levels of ionizing radiation and other demands on human irradiation constraints.
34
MANAGING RADIOACTIVE WASTE IN KAZAKHSTAN
By the order N97 from 31.03.2003 of the Minister of Health, it was approved the “Sanitary-hygienic demands to radiation safety guarantee”. The document establishes the requirements for: • • • • • •
Radiation objects site and design Organization management for radiation sources Delivery, accounting, storage, and transport of radiation sources Work with sealed radiation sources, technical systems, sanitary inspection room and lock (sluice) structure, and radiation control during the work with radiation sources RW treatment Radiation safety at radiation emergency, medical radiation
The classification of RW is determined in Sanitary Rules of RW Treatment (SPORO-97) taking into account their: • • •
Aggregative state (liquid, solid, gas) Specific activity (high, intermediate, low) Half-life (long, short)
The main demands for guaranteeing radiation safety at different stages of RW management are: • •
Accumulation, temporary storage, transport, receiving, processing, and final disposition of RW Radiation safety in disposal and equipment of RW disposition sites, radiation control, account and control of RW, emergency preparedness, occupational exposure measures, and personal hygiene
Nevertheless, one can note that in spite of a sufficient amount of corresponding normative documents some questions demand elaboration in additional documentation. In particular the present normative documents have been written at different times in a different political regime, by different teams of specialists, with different levels of scientific and technical knowledge. A significant part of normative documents on the list are approved and came into force under competent authorities of the former USSR. However new laws, normative legal acts of the President and Government of the Republic Kazakhstan have also come into force. There are in progress and reconsideration norms and rules revealing and determining laws demands. These circumstances make it necessary to elaborate and reconsider normative documents in the sphere of atomic energy use, effective regulation of activities related to atomic energy use in Kazakhstan.
MANAGING RADIOACTIVE WASTE IN KAZAKHSTAN
35
Today work is in progress on further reconsideration of the system of normative documents in the sphere of atomic energy use. Work is planned in two directions: • •
Elaboration of guide documents of the Committee on question of regulation requirements of safety for use of atomic energy Reconsideration of norms and rules currently in force in the sphere of atomic energy use and elaboration of new normative documents
The development of the system of the Committee’s guide documents is necessarily for work effectiveness increase of the Committee and practical implementation of its functions at carrying out normative-legislative, licensing, and supervision activities. 4.2 LICENSE REGULATION As mentioned earlier the Committee carries out licensing of activities that are associated with atomic energy use. RW treatment is related to this kind of activity. Laws of the Republic Kazakhstan “On Atomic Energy Use”, “On licensee” are the legal basis for the process of licensing. The licensing activity of the Committee was defined by the publication in 1998 of the Government Decree “On Approval of Provisions on Licensing of Activities Associated with Atomic Energy Use”. This document determines qualifying requirements for physical and legal entities, and the procedure and conditions of license issue License procedure includes the following stages: • • • • •
Examination of the statement on licensee issue and carrying out of preliminary checking of documents presented for license issue Assessment of documents confirming nuclear and radiation safety guarantee and if it is necessary the carrying out of scientific-technical expertise of the provided documents for safety basis If it is necessary to carry out inspection with the aim of examination of availability and adequacy grounds for safe performance of declared activity Formation of license activity conditions Issuance of license
With account of its organizational structure, the Committee has developed and approved licensee regulation which determines the procedure of consideration of license application statement and documents into the ground of safe carrying out of declared activities in structural departments of the Committee and the procedure of license issue.
36
MANAGING RADIOACTIVE WASTE IN KAZAKHSTAN
4.3 REGULATION OF SUPERVISION The Committee carries out inspections on the performance of organizations with regard to supervisory activities. Inspection is the form of state supervision that includes the examination of organization activity for assessment of definition of works, constructions, systems, equipment, documentation, and personnel qualification correspondence to establish safety requirements for atomic energy usage. Inspections are carryied out: • • •
At licensing At fulfillment of supervision activity At checking of appeals, statements, and other official statements except anonymous ones, and also with participation of the IAEA in inspections on safeguards implementation, examinations carried out by other state bodies
5 Summary On the whole one can say that a statewide system of regulating the handling of RW has been created and is fully functioning in the Republic of Kazakhstan. There is a normative-legislative basis for regulating management of RW. Compliance with its requirement and statements produces a significant level of radiation safety. Necessary legislative acts have been adopted (laws and regulation of the Government of the Republic of Kazakhstan). For example, dose limits were established in NRB-99 for any radiation source, including RW, both for personnel and the general population. These limits are in accordance with recommendations of the main International Basic Safety Standards for Protection from Ionizing Radiation (INSAG 1996). The conditions for and procedure of licensing activities associated with atomic energy use and established qualifying requirements to physical and legal entities claimed to license receiving for works bound with RW treatment. State bodies perform supervision and control of requirements of legislation, rules, norms, and license conditions for management of RW. Today the KAEC and other authorized state bodies continue their work to perfect the state regulatory system for the safety of atomic energy use, including management of RW. Toward this end there is considerable support from international organizations and competent bodies of other states in reviewing current documentation, and in developing new documentation in keeping with international requirements and taking into account the situation in Kazakhstan.
SPENT FUEL MANAGEMENT IN POLAND
S. CHWASZCZEWSKI∗ Institute of Atomic Energy, Swierk, Poland Abstract: Spent nuclear fuel in Poland is from the two research reactors EWA and MARIA. Information on these reactors and their spent fuel are presented. After discharge, storage has been under water in away-from-reactor (AFR) facilities for EWA spent fuel, and in the reactor pool for MARIA spent fuel. Design information and experience are given for these facilities. Because of lengthy wet storage, a broad programme of physical investigation of the spent fuel and storage facilities condition was begun. Visual inspection showed onset of corrosion after 20 years and strong corrosion after 30 years. Corrosion seems to depend on the presence of fuel under the cladding. Sipping tests were performed on WWR-SM assemblies for storage times of 4–31 years and a limit for wet storage established. Based on this work, the daily leakage rate of Cs-137 from WWR-SM and WWR-M2 spent fuel was estimated and compared with measurements in 1999 and 2000. The results of systematic sipping test of Ek-10 fuel and WWR-SM fuel assemblies discharged from EWA reactor before 1970 is also described. Additionally, an ultrasonic scanning technique was used to determine the extent of corrosion on cladding surfaces. The equipment for these techniques is described. The condition of the steel liner and aluminium tank of the storage pool was investigated with a wide range of techniques, e.g., dye penetrants, ultrasonic probes, and others. Improvements were made on the basis of results. Because of the limited time for safe wet storage, the concept of dry storage was proposed and is presented in this paper. The spent fuel for dry storage will be encapsulated in stainless steel capsules filled with inert helium. The technology for encapsulating MR spent fuel is described. A similar technique for encapsulating Ek-10 and WWR spent fuels is in the design stage. Keywords: Ek-10, MR and WWR-M fuels, EWA and MARIA research reactors, wet storage, pitting corrosion, sipping tests, dry storage
______ ∗
To whom correspondence should be addressed: Professor Stefan Chwaszczewski, Institute of Atomic Energy, 05-400 Otwock-Swierk, Poland; e-mail:
[email protected] 37 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 37–54. © 2007 Springer.
38
SPENT FUEL MANAGEMENT IN POLAND
1 Introduction In Poland, the problem with the management of spent nuclear fuel started at the time of first fuel discharge from the EWA reactor in 1959. The first spent fuel, after short storage in the EWA reactor tank, was transported to the Ek-10 storage facility. At the beginning, it was thought that the problem of spent fuel would be solved by the supplier of the fuel, namely by organizations elsewhere in the Soviet Union: after initial cooling in wet storage it would return to the supplier. The common belief in such a solution persisted till the end of 1992. But after the political, economic, and social transformation of the Soviet Union into the Russian Federation, and similar transformations in countries of Eastern and Central Europe, the principles of spent fuel management had to be established from scratch, under radically new conditions. There were two paths analysed for spent fuel management in the future: reprocessing in the Russian Federation, United Kingdom, or France; or long-term storage in a dry storage facility. In the latter option, CASTOR and NUHOMS containers, and the dry storage facility available at the EWA reactor after decommissioning, were all considered. Taking into account legal, technical, and economic aspects, the reprocessing solution was eliminated: it is an expensive, legally complicated, and technically difficult solution. In all cases, the high-level waste (HLW) originating in the reprocessing process would have to be stored in Poland, while the scale of the management problems with spent fuel and HLW are approximately similar. Therefore, the concept of long-term storage of spent fuel in a dry condition was chosen. Of several options, the design for dry storage in the decommissioned EWA facility was chosen as optimal.1 2 Spent Nuclear Fuel at the Research Centre, Świerk In the beginning, Ek-10 fuel rods were used in the EWA research reactor. The Ek-10 fuel rod had no certificate nor did it bear an identification mark. After reconstruction in 1966, the EWA reactor used fuel assemblies of the WWR-SM type, and from 1990, the WWR-M2 type. With these new fuel types EWA was operated at a thermal power of 6 MW, then 8 MW and finally, in the last period, 10 MW. The reactor was shut down in February 1995. In 1974, the second research reactor—MARIA—was put into operation. The reactor uses fuel assemblies of the MR-6 type. In the reactor core, the MR fuel assemblies are placed in special channels that supply cooling for the fuel. Spent fuel is discharged from the reactor core together with its channel and is stored in special pool at the MARIA reactor. Basic data for all spent fuel stored at the Institute of Atomic Energy are given in Table 1.
SPENT FUEL MANAGEMENT IN POLAND
39
TABLE 1. Basic parameters of spent fuel stored at the Institute of Atomic Energy Parameter/fuel type
Ek-10
Reactor Period of application Spent fuel storage facility Fuel mixture
EWA 1958– 1966 Ek-10 Rep. UO2 in Mg 81.4
Mass of uranium in SFA* (g) 8.05 Mass U-235 in SFA* Enrichment (%) 10 0.966 Average burn-up MWd/SFA. 1 Thickness of Al clad (mm) Number of elements/ 2,595 SFAs rods Calculated values after discharge Burn-up MWd. 1.00 Uranium mass (g) 80.2 U-235 mass (g) 6.85 Plutonium mass (g) ~0.147 Pu-239 mass (g) ~0.136
WWRSM EWA 1966– 1995 WWR Rep. U in Al 108
MR-6 (80%) WWRM2 EWA MARIA 1976– 1990– 1995 WWR MARIA Rep. Pool U in Al U in Al UO2 UO2 in Al in Al 124,5 429
MR-6 (36%) MARIA
MARIA Pool UO2 in Al 1500
38.9
44.8
345
540
36 14,36
36 16,02
80(36) 96
36 170 max
0.9
0.8
0.8
0.6
2,095 in SFA
445 in SFA
274 (MR514)
42
15 88.1 19.5 0.55 0.43
15 108 25.4 0.56 0.44
100 324 218 0.38 0.24
150 1251 356 20.6 18.6
*SFA–Single fuel assembly. WWR-SM and WWR-M2 types use single and triple assemblies.
Type Ek-10 fuel rods discharged in the first period of EWA operation are stored in the water pool of the first storage facility. This facility (No 19) is a concrete block with dimensions 735 × 585 cm and height 360 cm; a vertical section of the facility is shown in the Figure 1. In the central part of the concrete block are four chambers, 140 cm in diameter and 333 cm deep (Figure 2), situated on a square grid. Each chamber is lined with austenitic steel about 4 mm thick. The liner steel is an integral part of the storage chamber. At a height of 160 cm from the bottom of every chamber in the sheath side an austenitic tube was welded for ventilating purposes. The four ventilating tubes are joined to a collector on the axis of the building. All connections were checked for leak tightness using the dye penetrant method. Each chamber has an internal pool made from 6 mm aluminium plate. Spent fuel is placed in these internal pools in containers situated in a separator device.
40
SPENT FUEL MANAGEMENT IN POLAND
Figure 1. Storage facility for Ek-10 fuel
Figure 2. Storage chamber for Ek-10 fuel
Spent Ek-10 fuel elements are placed in chamber No 3, with the internal pool filled with demineralized water. The upper part of chamber is closed off by shielding. A vertical cross section of the storage chamber is shown in Figure 2. The spent Ek-10 fuel rods are stored in sets of 41–50 rods in a separate canister. Spent fuel assemblies of the WWR-SM and WWR-M2 types are stored in the second storage facility, which became operational in 1968. The design of the store is based on the assumption that spent fuel would have cooled in the atreactor store for a minimum of 3 months after discharge from the EWA reactor. The WWR spent fuel storage facility consist of two parts: an underground part made of reinforced concrete and an upper part made of bricks (Figure 3). There are two storage pools situated in the concrete block, each pool is 300 cm long, 270 cm wide, and 550 cm deep. The pools are double lined with stainless steel: inner wall 6 mm and outer wall 3 mm. The store has a 10 t crane, a ventilation system, and an activity-monitoring system. Spent fuel stored is stored in the pool in vertical channels forming a square lattice with 110 mm pitch in aluminium storage baskets. Each basket can hold 40 triple or 120 single fuel assemblies.
SPENT FUEL MANAGEMENT IN POLAND
41
Operation room Transport container
5,5 m
3m
Pool No. 2
Pool No. 1
(MR)
Spent
Auxiliary room Droga transportowa
fuel WWR
CONCRETE
Figure 3. Storage facility for WWR and (in future) MR spent fuel
MR assemblies from the MARIA reactor are stored in the “technological pool” of the reactor. After discharge, a spent fuel assembly is transferred in its channel and suspended from the pool wall. After removing channel hardware, the assembly is loaded into a basket located at the bottom of the pool. The fuel channel is removed with a cutting device located in the flowgate between the reactor and its pools. This handling sequence is shown in Figure 4. The water chemistry of the technological pool is controlled each week to a conductivity of <3 µS/cm and a pH value in the range of 5.7–6.2. The water is filtered when either limit is exceeded. Spent fuel in technological channels
Hot chamber
Flowgate
+.6.33
Technological pool
R 3250 +4.1
12250
Spent fuel +2.25
Reactor
R 2250
+1.1
-2.85 -2.85
Figure 4. Spent fuel handling at the MARIA reactor
42
SPENT FUEL MANAGEMENT IN POLAND
In general terms, in the spent fuel storage facilities at the Institute of Atomic Energy, as of 31 December 2004, there was a total of 5,410 spent fuel assemblies. These assemblies had produced 1.681 TWh (1.681 × 109 kWh) of thermal energy; they contained 625.6 kg HM. The total activity from this accumulated spent fuel is estimated to be 1.5 MCi (5.5 × 1016 Bq). 3 Physical Assessment of Spent Fuel Condition Systematic inspections of spent fuel stored in the research centre at Swierk were performed, starting in 1997. The Government Strategic Program “Management of Radioactive Waste and Spent Fuel in Poland” was established. In the framework of this programme, the following scope of activities with respect to the spent fuel from EWA and MARIA reactors was established: • • • • •
Estimation of safe storage time in wet condition Identification of failed fuel elements and their encapsulation Elaboration of proposals for improvement of storage conditions Modernization of the spent fuel storage facilities Elaborate the concept (with preliminary technical, safety, and economic analysis) of the final solution for spent fuel management in Poland
3.1 EXAMINATION TECHNIQUES AND EQUIPMENT For analysis of spent fuel condition after long-term storage underwater, the following techniques were used: • • •
Visual testing of the outside surface of spent fuel Sipping test of specified fission product from fuel element Ultrasonic scanning of clad surface
For visual investigation of spent fuel the VIDEO-TEST system2 was used; it consisted of the following parts—Figure 5: • • • • •
Video camera Panasonic GP-KS162CUDE TV monitor SVHS Panasonic WC-CM14150 Video recorder SVHS Panasonic AG-4700 EY PC with MUTEC video card and LUCIA programming system High-resolution colour printer MITSUBISHI CP-4700 D
Underwater pictures of spent fuel cladding were recorded on video tape, converted to bitmap files using a MUTEC video card, and stored on computer or CD-ROM. Pictures were then transformed with the LUCIA programming system and printed on a high-resolution Mitsubishi CP-4700D colour printer.
SPENT FUEL MANAGEMENT IN POLAND
43
Figure 5. VIDEO-TEST system equipment
A set of video testing experiments were carried out before measurements were made. The influence of gamma radiation on the video picture quality was investigated, among other factors. It was found that the video camera worked satisfactorily in a gamma field of up to 15 Gy/h. For the analysis of spent fuel, the STEND-1 system3 was constructed for precision manipulation of measuring equipment. It consisted of devices for the underwater precision rotation of fuel elements and precision vertical movement of monitoring devices like the video camera and ultrasonic head. The system is controlled by a PC. The configuration of the STEND-1 system is shown in Figures 6 and 7. 3.2 VISUAL INSPECTION Video inspection of Ek-10 spent fuel revealed advanced corrosion, especially by pitting. Views of the cladding on Ek-10 elements are shown in Figure 8. The condition of the WWR-SM fuel elements depends on storage time. No visual defects were observed in the fuel elements stored less then 15 years. For elements with longer storage, uniform and pit corrosion of the surface were observed. In the case of old (over 30 years) spent fuel, advanced corrosion was observed. The pictures of such corrosion are shown in Figure 9.
44
SPENT FUEL MANAGEMENT IN POLAND
Figure 6. The STEND-1 facility
Figure 7. Underwater part of STEND-1 (left); and its controls (right)
Figure 8. Ek-10 spent fuel cladding after 40 years wet storage
SPENT FUEL MANAGEMENT IN POLAND
45
Figure 9. Cladding surface of WWR-SM fuel discharged from EWA in 1970
Visual inspection of MR-6 spent fuel confirmed results of the WWR-SM investigation. In elements stored under wet conditions less than 15 years, good cladding condition was observed. Spent fuel wet stored since 1977–1978 exhibited intensive corrosion. Pictures of MR-6 fuel assembly are shown in Figure 10.
Figure 10. View of an MR-6 spent fuel assembly: lower (left), middle (centre), and upper (right) parts
46
SPENT FUEL MANAGEMENT IN POLAND
In Figure 10, intense corrosion (uniform and pitting) was observed. The difference of the corrosion processes in the region with nuclear fuel and without fuel can be observed. Such effects were observed in other MR-6 and WWR-SM assemblies stored for long (over 15 years) times in the wet condition. The nonuniform boundary of the fuel layer in MR-6 assembly was proved by the radiological investigation of fresh MR-6 fuel assembly. 3.3 SIPPING TESTS For sipping tests Cs-137 isotope was chosen as the marker of release of fission products from tested fuel assembly.4 The reasons for selecting Cs-137 are sufficient efficiency of this isotope production in the fission processes and its 30 years half-life. The fuel element was closed in the capsule with pure fresh water. After suitable time, the sample of water was released from the capsule, and activity of the sample was measured using calibrated scintillation gammaray spectrometer with a NaI(Tl) crystal. The investigated fuel elements were placed in a special container. After tightening, the container was filled with fresh water and after 24 h the activity of Cs-137 in a water sample from the container was measured by scintillation gamma spectrometry. Based on the result of the measurements, the total activity of the released Cs-137 was determined. Therefore all the data of the sipping technique were shown as activity of Cs-137 isotope released per day. The container was designed for sipping test of triple WWR-SM fuel assembly. If measurement was done for a single fuel assembly, the filler was applied to minimize the amount of water in the container. The results of release measurements versus assembly storage time are shown in Figure 11.
Figure 11. Release of Cs-137 from WWR-SM assemblies as a function of storage time
SPENT FUEL MANAGEMENT IN POLAND
47
As it can be seen, the release increases dramatically for storage time more than 30 years. Extrapolation of measured data for storage time more than 30 years shows that the release tends to infinity for a storage time equaling 41 years. Thus the estimated time limit for WWR-SM fuel storage in water is approximately 41 years. This method was used for sipping tests on Ek-10 and MR spent fuel. 3.4 ULTRASONIC TESTING Visual inspection does not give information about the depth of a corrosion hole. To determine the corrosion shape, the ultrasonic technique was used.5 The principle of the method is based on measuring the arrival time of the echo of an ultrasonic pulse in a water environment. The ultrasonic pulse is emitted by an ultrasonic probe located 15 mm from the cladding surface. The ultrasonic probe was designed in such a manner that the ultrasonic beam was focused at the cladding surface in a spot of 0.1 mm diameter. Thus the space resolution of surface scanning is 0.1 mm. The ultrasonic frequency was 5 MHz, so the surface scanning can be defined with accuracy limit of 0.02 mm (speed of ultrasonic waves in water is equal 0.148 cm µs–1). For ultrasonic measurements the ultrasonic defectoscope EPOCH III, model 2300, made by Parametrix and a 5-MHz UT probe designed for water environment with shape focusing of the beam at 15 mm of water were used; Figure 12 shows the setup. The results of surface scanning of WWR-SM spent fuel for crevice corrosion6 are shown in Figure 13.
Figure 12. Setup for UT scanning
Figure 13. UT scans of crevice corrosion
48
SPENT FUEL MANAGEMENT IN POLAND
4 Dry Storage of Spent Fuel The set of spent fuel elements were stored for a considerable time in the wet condition. Some are sources of significant leakage of fission products due to corrosion of the aluminium cladding. Therefore it was decided to store failed fuel elements or assemblies in the dry condition in capsules filled by with helium. After encapsulation the spent fuel capsule is placed in the storage pool. In the process of encapsulation of the MR-type spent fuel it was necessary to distinguish the two regions of operations to be accomplished with the fuel, i.e., the storage pool and the hot cell of the MARIA reactor. The most important operation of the MR-type spent fuel encapsulation process, namely, drying, leak-tight closure of capsules, and checking their tightness, are carried out in the hot cell of the MARIA reactor. 4.1 HOT CELL AT MARIA REACTOR The hot cell is located at the end of the storage pool adjoining the external shell of the building. The basic equipment for the cell comprises the following devices and systems: machine tool; transport device; overhead crane; lifting magnet; two manipulators; bolting machine; ventilation, waste, and decontamination system; radiological monitoring system; television camera and five culverts. For encapsulating MR-type spent fuel, the cell has been additionally equipped with a drying chamber and a vacuum chamber located in recesses of the cell floor; chucks fixed on the face milling cutter to be used for the welding head, helium probe, and milling cutter for cutting off the capsule. This peripheral equipment is arranged on a platform at the front of the cell; also both the crane and lifting magnet are operated from this platform. The spent fuel assemblies are located in the storage pool. Before removing the spent fuel element from the sheath the spent fuel element is cut off from the channel construction and the water contained inside is monitored. Fuel channel with leaking fuel will be separately cut off by the water lock. Then it will be closed in a leak-tight thimble in which water is filtered. To safely bring the fuel element into the hot cell, it is necessary to install on it a gripping head accommodated to the lifting magnet, which belongs to one of the cell accessories. This operation is accomplished in the storage pool below the water level. The fuel element modified in this way will be brought to the transport sleeve which is placed in a special rack on the floor of the storage pool, and then along with the transport sleeve on the carriage of the transport equipment. After approaching the spot right below the inlet hatch of the dismantling cell, the spent fuel element will be pulled into the transport sleeve, dried out,
SPENT FUEL MANAGEMENT IN POLAND
49
and prepared for placing inside the dry storage facility. To reach this goal it is necessary to apply the technology of the encapsulation process that is shortly described below. 4.2 ENCAPSULATION PROCESS Following the developed procedure, a spent fuel element is first transferred to the drying chamber. After placing the spent fuel element in the drying chamber it is closed and the drying system is activated. The electric air heater with temperature regulation within the range of 50–110°C at the inlet to the culvert is located at the outlet of this system. The drying process is carried on until the air relative humidity at the outlet of the system is around 5%. After reaching an appropriate relative humidity (RH), the dried fuel element is ready to be closed in the capsule. The most important moment of the encapsulation operation is the leak-tight closing of the dried spent fuel element in the capsule. The capsule and its cover have been earlier manufactured of certified materials and they are verified as regards weld tightness. The dried fuel element is transferred to the capsule thimble. The operation of loading the spent fuel element into the capsule is recorded on video to identify the numbers on the capsule and fuel element. In the next operation the welding head is positioned onto the welding spot and rotation of the chuck and the TIG welding machine are activated. Weld tightness on the capsule perimeter is checked by helium leak detection using a ASM-type detector manufactured by the ALCATEL company. If the capsule is leak tight, it is filled with helium to an overpressure of 0.2 MPa. Next, operation of mounting the gripping head on the capsule is started. It is the same as described below during the leak-tight closing of the capsule. Figure 14 shows the set up. After the gripping head has been mounted, the capsule is transported to the vacuum chamber. A suction head is installed on the chamber and the attached helium leak detector is activated. When the absolute value of the overall leak rate of the capsule has been measured and recorded the leak-tight capsule is removed from the vacuum chamber ready for transport to the storage pool. The review of the encapsulation technology of the spent fuel elements has confirmed its full compatibility with the required safety and strength of the elaborated procedure relevant to storage of spent fuel in the dry storage facility for at least 50 years. Till now 114 MR fuel assemblies have been encapsulated and are ready for transport to pool No 2 in the WWR storage facility. With this good experience with MR spent fuel assemblies a similar technology is being designed for use with Ek-10 fuel elements and WWR assemblies.
50
SPENT FUEL MANAGEMENT IN POLAND
Figure 14. Leak-tight capsule after mounting its gripping head in the hot cell
5 Dry Storage Facility at EWA Reactor The results of the assessment of spent fuel condition indicates the necessity of removing it from wet storage. As mentioned in the introduction, the choice is of dry storage of spent fuel for at least 50 years. From the analysis of possible solutions, the concept of using the decommissioned EWA reactor was chosen, as proposed by T. Matysiak.6,7 The concept is based on using the biological shielding of the reactor for storage. Inside the shaft six rack modules will be placed. In each module, the hexagonal grid of stainless steel tubes with diameter 114 mm, wall thickness 4.5 mm, and grid pith 130 mm in concrete will be installed. A vertical section of the dry storage facility is shown in Figure 15. The spent fuel elements or assemblies after drying will be encapsulated in stainless steel capsules and placed in vertical channels in the racks. The safety analysis of the proposed solution show that the neutron multiplication coefficient of such a storage facility is below 0.9 in normal condition and 0.95 in the fully flooded condition. The temperature of fuel cladding in normal ventilation will be below 80°C and in the case of ventilation failure—below 100°C.
SPENT FUEL MANAGEMENT IN POLAND
Biological shield of EWA reactor
51
Rack modules
Figure 15. Vertical section of the proposed EWA dry storage facility.
The safety analysis of outside situations included earthquake, flood, hurricane, and a direct hit by an aircraft. Analysis of inside situations included fire, drop of transport devices (casks), operator errors, and electric power failures. The proposed use of the EWA biological shield as a storage facility is considered safe for all these conditions. 6 Measurement of Fissile Material in MR Spent Fuel Measurement of fissile material content can be made by two techniques, the passive neutron technique and the active one.8,9 The first one is based on the measurement of inherent neutron emission rate. In case of the MR fuel, the dominant sources of inherent neutrons emission are the spontaneous fission source and the fission neutrons induced by the spontaneous fission neutrons through the fissile materials remaining in the spent fuel. The second technique is based on
52
SPENT FUEL MANAGEMENT IN POLAND
the measurement of neutrons emitted from spent fuel as a result of fission within the fissile materials. The fission is induced by external neutron source. The number of emitted neutrons is proportional to the amount of fissile materials present in the spent fuel. In the active neutron technique, neutrons escaping from the source (Pu-Be) have rather high energies and before directing to the fuel should undergo moderation. This is effected by positioning the neutron source in the centre of a moderating container having a cylindrical shape and filled with water playing a role of moderating medium. The moderating container is lined by a cadmium sheet of 1 mm thickness, which cuts all thermal neutrons reaching the container boundary. The purpose of absorbing the thermal neutrons going out from the moderating container is to reduce background and to minimize multiplication in the rest of the spent fuel. To make the measurement of the fissile materials content along the spent fuel assembly possible, a collimator of rectangular shape has been inserted to the moderating container. The distance between the neutron source and inlet window of the collimator has to be chosen to achieve the maximum outward thermal neutron current in the collimator. Neutrons escape the collimator through the outlet window and intersect the fuel assembly induce fission through the fissile nuclei. Fast neutrons emitted from fission undergo thermalization in water and then detection by the fission chamber. The fission chamber is not adjacent to the fuel surface. Certain distance from the fuel surface guarantees proper thermalization of fission neutrons in the water. The principle of the method is shown in Figure 16. The measuring installation is submerged underwater to provide shielding against intensive gamma radiation emitted from the spent fuel.
Figure 16. Schematic diagram of the measuring equipment
SPENT FUEL MANAGEMENT IN POLAND
53
The idea of passive neutron technique is to establish a functional relationship between the fuel burnup and the experimentally measured inherent neutron source. The signal of the inherent neutron source is proportional to the amount of actinide isotopes having a significant spontaneous neutron emission, and on the neutron emission rate of each isotope. The emitted neutrons undergo multiplication within the fissile materials remained in the spent fuel. The ORIGEN code has been used to calculate the spontaneous neutron source (fission rate) as a function of burnup for certain cooling time values. The spontaneous fission rate R s has been found to be correlated with the burnup above 60 MWd/fuel assembly by a power functional relationship:
R s = α ⋅ (Bu ) γ Thus, the burnup can be expressed as:
Bu = K ⋅ (CR in ) 1 / γ where K can be recognized as a calibration constant. The burnup and the amounts of fissile materials distribution along the axial length of the spent fuel can be evaluated using the detector count rate given the values of γ and K. The active measurement method turned out to have a poor axial resolution and rather can be used for measuring the absolute value (average) of the amount of fissile materials. The active method required numerical calculations to verify the linearity between the detector signal and the amount of fissile materials. The verification has been done by Monte Carlo simulation The passive method proved to have an excellent axial resolution, thus it can be used to verify the burnup and the amount of fissile materials in the spent fuel. The passive method makes the measurements simpler and faster. The passive method requires at least one calibration point for absolute burnup determination. This must be realized by the reference method. Also calculations should be used to complement the passive neutron measurements. These calculations relate the burnup with the inherent neutron emission. ORIGEN code has been used for these calculations. The sensitivity of the neutron emission rate to various parameters such as cooling time and reactor power was studied pointing to a very good stability in the relationship between burnup and spontaneous neutron emission rate on the fuel power at the same time showing significant dependence on the cooling time. To get good statistical results, the measurements have been repeated 60 times for every position of 1 min measuring time.10 The standard deviations were about 7%. Both techniques gave a good agreement compared with the reactor operation data. The active neutron technique gives a difference of about 2% while the passive one gives about 4%.
54
SPENT FUEL MANAGEMENT IN POLAND
References 1.
Chwaszczewski, S., Włodarski, J., Nuclear Safety and Radiological Protection, 1(41), Warszawa, Poland, 2000, p. 3. 2. Chwaszczewski, S., Czajkowski, W., Borek-Kruszewska, E., Matysiak, T., Madry, M., “Management physical assessment of the spent fuel from research reactors in Poland”, 5th International Topical Meeting on Research Reactor Fuel Management, Aachen, Germany, 1–3 April 2001, p. 78. 3. Czajkowski, W., Borek-Kruszewska, E., “Stand for visual and ultrasonic testing of spent fuel”, ibid., p. 160. 4. Chwaszczewski, S., WWER Fuel Management and Quality Investigation. CRP on “Corrosion of research reactor aluminium clad spent fuel in water”, First Regional Co-ordinating Meeting, Buenos Aires, Argentina, November 2002. 5. Borek-Kruszewska, E., Bykowski, W., Chwaszczewski, S., Czajkowski, W., Mądry, M., “Encapsulation technology of MR-6 spent fuel and quality analysis of the Ek-10 and WWR-SM spent fuel stored more 30 years in wet conditions”, 6th International Topical Meeting on Research Reactor Fuel Management, Ghent, Belgium, 17–20 March 2002, p. 170. 6. Matysiak, T., “The concept of dry storage facility in decommissioned EWA reactor”, International IAE Report (in Polish). 7. Mieleszczenko, W., Mołdysz, A., Hryczuk, A., Matysiak, T., “Nuclear spent fuel dry storage in the EWA reactor shaft”, International Topical Meeting on Research Reactor Fuel Management Aachen, Germany, April 2001, p. 83. 8. Chwaszczewski, S., Pytel, K., Abou-Zaid, Attya A., “Neutron multiplication method for measuring the amount of fissile isotopes in the spent fuel”, ibid., 2001. 9. Abou-Zaid, Attya A., “Neutron multiplication method for measurement of the amount of fissile isotopes in the spent fuel”, PhD Dissertation, Warsaw University of Technology, Poland, 2000. 10. Pytel, K., Prokopowicz, R. Strzałkowski, L., Mieleszczenko, W., Pytel, B., “Measurements of vertical burn-up distributions of MR-6/80% fuel”, Institute of Atomic Energy Internal Report B, 2004.
AN OVERVIEW OF SPENT FUEL STORAGE AT COMMERCIAL REACTORS IN THE UNITED STATES
J. D. B. LAMBERT1∗ AND R. LAMBERT2 Argonne National Laboratory, Chicago, Illinois 2 Electric Power Research Institute, Palo Alto, California 1
Abstract: A legacy of the growth in the use of nuclear power in the USA over 1960–2005 has been the accumulation of large quantities of spent nuclear fuel (SNF) at the reactor sites, a situation initially caused by an embargo on fuel reprocessing and later exacerbated by delays in the opening of a national repository. The nation’s inventory of SNF stands at ~54,000 metric tons of heavy metal (MTHM) in late 2005, and grows at an annual rate of ~1,750 MTHM. Storage pools are becoming full as a result, despite the use of high-density fuel racks. To alleviate the problem older SNF assemblies are being dry stored in casks placed on concrete pads in independent spent fuel storage installations (ISFSIs). There are now 30 ISFSIs in the USA and their number is growing. Experience with both wet and dry storage has been very reassuring. Wetstored SNF derives its stability from an adherent oxide layer on its Zircaloy cladding which resists further corrosion, even in water of poor quality. The destructive examination of SNF that had been vacuum-dried and stored in Hefilled casks for nearly 15 years revealed no increase in the internal fission-gas pressure or increase in the cladding creep of its fuel rods. The terrorist attacks of September 2001 focused attention on the security of reactors and their associated SNF storage. Analysis suggests that dry storage at an ISFSI may be more secure than wet storage at a reactor. However, offloading SNF to dry storage and a return to the use of open-frame racking are measures that could improve security at spent fuel pools. Keywords: PWR and BWR reactors, spent nuclear fuel (SNF), wet and dry storage, fuel storage pools, dry storage casks, independent spent fuel storage installations
______
∗ To whom correspondence should be addressed: J.D.B. Lambert, Argonne National Laboratory, 9700 S. Cass Avenue, Argonne, IL 60439, U.S.A.; e-mail:
[email protected]
55 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 55–70. © 2007 Springer.
56
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
1 History of US Nuclear Power The era of commercial nuclear power in the USA began in 1957 when the Westinghouse 63-MWe Shippingport pressurized water reactor (PWR) was started up on the Ohio River, 25 miles from Pittsburgh.1 Three years later, the first boiling water reactor (BWR)—the 210-MWe Dresden unit designed by the General Electric Company—began operation in Morris, Illinois. Despite the accidents at Three Mile Island in 1979 and Chernobyl in 1986, which shook public confidence in nuclear power, the next 25 years or so saw a sustained growth in the number of reactors of both types that were designed, built, and brought into operation2 in the USA (Figure 1). Although no orders were placed for new reactors after 1978 and the last reactor (Watts Bar 2) began operation in 1997, the number of units grew to a peak of 110 in 1990. In 2005 there were 104 operating power reactors in the USA—69 PWRs and 35 BWRs, which supply 19% of the nation’s needs for base-load electricity. A further 14 reactors are shut down permanently and are at various stages of being decommissioned. Over the last 15–20 years the combined effect of improved fuel designs, longer reactor run lengths, Figure 1. Number of operating U.S. reactors increase in burnup of fuel, and electricity generation, 1960-2000 and much greater attention Source: Nuclear Energy Institute to maintenance resulted in plant factors at the better-run reactors routinely approaching ~90%, more than compensating for the units that were taken out of service over the same period. Nowadays, there is increasing optimism about nuclear power. A sign of this current mood is the number of utilities seeking life extension for their reactors: by September 2005, the Nuclear Regulatory Commission (NRC) had approved license extension from 40 to 60 years for 33 reactors, with 16 more applications under review, and 28 further applications expected.3 A second even more promising sign is that NuStart—a consortium of major US utilities—is planning license applications to build two new reactors: at the Tennessee Valley Authority’s Bellafonte site in Alabama, and at Energy’s Grand Gulf site in Mississippi4; the applications are due in late 2007 and early
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
57
2008, respectively. The cost of these new reactors will be shared by NuStart and the US Department of Energy (US-DOE) in its 2010 Nuclear Power Initiative. 2 The Legacy of Spent Nuclear Fuel An unforeseen consequence of the success story of US nuclear power has been the inventory of spent nuclear fuel (SNF) which has accumulated in wet storage around the nation. It was always intended that SNF would be cooled in fuel storage pools (FSPs) at the reactor sites for 3–5 years only. The SNF would then be reprocessed to recover unused uranium and plutonium produced inreactor. In anticipation of this logical step commercial reprocessing facilities were built in the late 1960s and early 1970s at Morris, Illinois, West Valley, New York, and Barnwell, North Carolina; FSPs at reactor sites were sized accordingly. 2.1 BAN ON FUEL REPROCESSING In May 1974 India detonated a nuclear device made from plutonium separated at its reprocessing facility in Trombay. This event caused fear regarding nuclear proliferation and led President Ford to declare in 1976 that “the avoidance of proliferation must take precedence over economic interests”. This deferral of civilian reprocessing was made an embargo in 1977 by President Carter. Although the embargo was lifted in 1981, reprocessing was no longer believed to be economic and the facilities built earlier were never really operated. Since then the USA has followed a “once-through” nuclear fuel cycle strategy, thereby creating an ever-increasing inventory of SNF in what appears to be permanent wet storage at reactor sites. 2.2 NATIONAL WASTE POLICY ACT OF 1982 In recognition of the growing need to deal with SNF (and with the radioactive waste from some early reprocessing and production of medical radioisotopes), Congress passed the National Waste Policy Act of 1982. This Act addressed the SNF problem by recommending the characterization of a site on the Yucca Mountain in Nevada as a national repository for both SNF and high-level waste (HLW). Progress in this project has been painfully slow ever since for a myriad (mainly political) reasons with 2017 the latest projected date for opening the facility.5 Table 1 shows the SNF that accumulated meanwhile at US reactors up to December 2002 as a result of the once-through nuclear fuel cycle and delays at Yucca Mountain. By December 2002 there was a total SNF inventory of ~47,000
58
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
MTHM, including ~3,500 MTHM from 14 shut-down reactors; ∼3,500 MTHM of the inventory were in dry storage. By September 2005 the total had increased to 54,000 MTHM 6 and the amount in dry storage to 8,300 MTHM. The 100,000 MTHM level will be exceeded by about 2035, given that all existing reactors seek and receive NRC permission to operate for 60 years. TABLE 1. Historical and projected SNF discharges from U.S. commercial reactors Year 1970 1975 1980 1985 1990 1995 2000 2002 2005 2015 2025 2035
Cumulative Number of Fuel Assemblies 128 6,352 24,812 46,908 78,005 113,418 150,701 165,854 187,000 252,000 300,000 365,000
Cumulative U Content (MTHM) 45 1,557 6,542 12,675 21,442 31,992 42,605 47,023 54,000 72,000 86,000 104,000
Source: DOE Office of Civilian Radioactive Waste Management
2.3 NEED FOR DRY STORAGE OF SNF The annual increase in the national inventory of commercial SNF is ∼1,750 MTHM (Table 1). Given this rate, and essentially no relief until 2017 from the Yucca Mountain Repository, it is clear that FSPs are filling up fast and that recourse must be increasingly made to dry store SNF at reactor sites. Figure 2 shows that 65 reactors had full FSPs in 2005 and that most pools will be full by 2012–2013. By then every reactor must be able to off-load cooled SNF to dry storage in order to safely fuel handle and to continue operation.
Figure 2. Increase with time in the number of spent fuel pools at capacity in the U.S., based on ability to off-load the entire core following an emergency Source: Nuclear Regulatory Commission
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
59
3 Mechanics of SNF Handling and Storage 3.1 FUEL HANDLING A light-water reactor is hermetically closed during operation by a pressure head through which control-rod drives and instruments enter via special seals. After shutdown, the reactor cavity is flooded with water to a level considerably above this pressure head. To change out fuel, the control-rod drives and instruments are disconnected, the head unbolted and lifted aside to reveal the core and SNF replaced—one assembly at a time—with an overhead fuel handling crane. The crane removes the SNF assembly from the core under water, transfers it along a canal to the fuel storage pool (FSP), and brings back a fresh assembly to occupy the space vacated by the spent assembly. 3.2 FUEL STORAGE POOLS The FSP in a BWR is located in the reactor building well above ground level; in a PWR, it is typically in a separate building outside of main containment, at ground level or below. In both reactor types, pools are about 12 m deep and can be 12 m or more in each horizontal direction (Figure 3). The pool walls are constructed from 1.2 to 2.5 m thick reinforced concrete. A steel liner 6–13 mm thick is attached to the walls with studs embedded in the concrete. The pools have 4 m high storage racks with feet to provide space between the racks and the pool bottom to allow water circulation. Pools now contain “dense” racks with steel-clad panels filled with a boronated polymer (Boraflex), see Figure 4. These racks allow up to five times as many fuel assemblies to be stored as in the original open-frame racks. Such reracking of SNF has been a common practice at most reactors since the late Figure 3. Spent fuel pool at a PWR 1970s. Source: Nuclear Regulatory Commission
60
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
Figure 4. Dense SNF storage racks for a BWR spent fuel storage pool (From: Nuclear Regulatory Commission.)
3.3 DRY STORAGE CONTAINERS With pools filling up despite reracking, utilities have turned increasingly to dry storage to alleviate their SNF problem. The use of a dry storage cask was first approved by the NRC in 1986 for Virginia Power’s Surry-2 Nuclear Station PWR. Table 2 lists the US cask designs that have since been approved by the NRC. The approved storage period is 20 years so that early designs will soon need to be relicensed. Dry storage casks can be vertical or horizontal, although vertical casks tend to be more often used. The reason perhaps is that vertical casks are stand-alone units that come complete with their own shielding—a concrete pad is the only additional requirement at the site. Horizontal casks are less heavily shielded and must be placed in steel-lined concrete bunkers for safety. An example of each type of cask is shown in Figure 5. When dry storage begins at a reactor site, it is declared an independent spent fuel storage installation (ISFSI), which is regulated by the NRC; Figure 6 shows licensed ISFSIs. In 2005, there were 8,300 MTU of SNF in 690 casks at 30 ISFSIs. In 2010, the projected numbers are 13,000 MTU in 1,300 casks at 51 ISFSIs.6 Dry storage will continue to increase in ensuing years as plant life extension takes hold in the US reactor fleet.
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
TABLE 2. NRC-approved dry spent fuel storage designs7 Vendor
Storage design model
Compliance certificate date
General Nuclear Systems
CASTOR V/21
08/17/90
NAC S/T NAC-C28 S/T NAC-MPC NAC-UMS VSC-C24 Fuel Solutions
08/17/90 08/17/90 04/10/00 11/20/00 05/03/93 02/15/01
Holtec International
HI-STAR 100 HI-STORM 100
10/04/99 05/31/00
Transnuclear.
TN-24 TN-68 TN-32A/B NUHOMS-24P NUHOMS-61BT NUHOMS-52BT NUHOMS-24PT1
11/04/93 05/30/00 02/20/01 09/12/01 09/12/01 09/12/01 02/05/03
NAC International
BNFL Fuel Solutions
Figure 5. Examples of SNF dry storage casks Left: vertical stand-alone TN-68s; Right: horizontal NUHOMS-32s in bunkers (From: Areva (2005).)
61
62
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
Figure 6. Licensed independent spent fuel storage facilities (From: Nuclear Regulatory Commission.)
4 Spent Fuel Storage Experience in the USA 4.1 STATISTICS OF SNF STORAGE In 2005 there were approximately 107,000 discharged BWR fuel assemblies and 81,000 discharged PWR fuel assemblies in storage at US reactor sites.6 Of these totals—which together comprise 54,000 MTHMs—approximately 17,000 BWR and 13,000 PWR fuel assemblies have experienced storage under both wet and dry conditions. The longest wet-stored SNF has been underwater for 40 years. The average age of all currently wet-stored fuel is in the region of 12–18 years, a value that will decrease in years to come. Because dry storage was first permitted in 1986, the longest dry-stored SNF has to be from the Surry-2 PWR, which has remained in a CASTOR V/21 cask for almost 20 years. The average time in dry storage is currently 8–12 years. This value will increase as progressively more SNF is moved to dry storage at ISFSIs. 4.2 WET STORAGE EXPERIENCE Wet storage experience in the USA includes nearly 200,000 SNF assemblies that have been under water for an average of 12–18 years. In general terms, experience has been excellent, with no evidence to suggest that storage under water has led to any significant degradation in SNF condition. Indeed, no fuel cladding failure has been attributed to water storage per se.
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
63
The intense radiation field in an LWR core causes radiolysis of the coolant water. The hydrogen that is freed leads to hydriding of Zircaloy cladding, while the oxygen reacts with the Zircaloy to form an oxide surface layer typically 50–75 µm in thickness at discharge (Figure 7). These effects of radiation-induced hydrolysis are life-limiting for LWR fuel: hydriding embrittles the cladding, while the oxide surface layer degrades heat transfer from the fuel rods. Paradoxically, the oxide layer Figure 7. The oxide layer on Zircaloy that forms in-reactor is also the versus fuel burn-up and cladding type feature that protects SNF during Source: H.W. Wilson et al. 8 wet storage and allows prolonged immersion without problems, even in water of poor quality. In order to limit the possibility of degradation, the NRC suggests that the thickness of the oxide layers should be limited to <100 µm to avoid them flaking off (or being knocked off during fuel handling), and that the electrical conductivity of the pool water and its aggressive ion content (particularly Cl–) should be minimized to limit further corrosion in the event that damage to the layers cannot be prevented.9 4.3 DRY STORAGE EXPERIENCE Dry storage experience in the USA up to 2005 includes a total of about 30,000 SNF assemblies located in a variety of casks; the average storage time is 8–12 years. There have been no reported problems encountered with this mode of containing and storing SNF, although to observe any evidence of degradation requires opening sealed casks and inspecting the fuel. Tests and calculations were performed on casks in the early 1980s10–12 to identify appropriate drying procedures and heat loads to ensure SNF stability in dry storage. Vacuum drying to 0.27 kPa (2 mm Hg) was found to be sufficient with a final hold time of 6 h—if pressure increases at all during this period, the cask is filled with He to 1.3 kPa (10 mm Hg) and the evacuation repeated. The cask interior is finally filled with high-purity (99.995%) He. Figure 8 shows the maximum allowable dry storage temperature versus the age of the SNF and its internal gas pressure, the criterion being the allowable creep of the cladding.
64
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
To observe any degradation due to dry storage a Castor V/21 cask used to store 21 assemblies from the Surry-2 PWR was examined in 1999, 14 years after being loaded with SNF. Gas analysis of the cask showed no signs of air ingress or failure of the cladding and fission gas release. The cask lid Orings were in good condition and all materials inside the cask appeared the same as they had at cask loading.13 Twelve fuel rods were removed from the hottest SNF assembly (T11) in the cask and were destructively examined; the results were similarly 14 Figure 8. Allowable storage temperature reassuring : for SNF versus age and internal pressure Source: B.A. Chin & E.R. Gilbert.12
• • • •
The maximum cladding creep that may have occurred in storage was <0.6%; actual creep values were probably considerably less. There appeared to be no additional release of fission gas during storage. There was no evidence of hydrogen pickup or hydride reorientation during the storage period. No significant annealing of the cladding had occurred during the prestorage testing or the storage period itself.
4.4 DRY CASK TESTS The NRC, the Electric Power Research Institute (EPRI), and the DOE Office of Civilian Waste Management are participating in a cooperative research program to determine the long-term integrity of dry cask storage systems and the SNF they contain. The first experiment was with the Castor V/21 cask which is described above. A number of other commercial dry casks are presently being tested at the Idaho Nuclear Technology and Engineering Center (INTEC),15 as shown in Figure 9. The interiors of the casks are periodically sampled for fission gas to check for possible fuel failure during dry storage. As yet no fuel failures have been detected; based on the results of examining the Surry-2 SNF, none is anticipated in the future.
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
65
Figure 9. Commercial dry storage casks being tested at the INTEC Left to Right: REA 2023; VSC-17; TN 24P; V/21; 125B; MC-10 (From: Hill and Fillmore, this workshop.)
5 Safety Issues of Commercial SNF Storage 5.1 CRITICALITY CONTROL The fuel racks in FSPs were originally designed for use without recourse to any absorbers—the possibility of a criticality event during fuel handling being precluded by the generous pitch of the racks. With time, however, not only did the U-235 enrichment of LWR fuel rise significantly in order to achieve higher discharge burnups (typically from 2–3% U-235 to 4–5%), but the inability to off-load SNF caused FSPs to gradually fill up, as earlier described. The initial industry response was to “rerack” SNF into closer packed arrays (Figure 4) which use neutron absorbers for the necessary additional control. Although a variety of absorbers like Boral have been tried, Boraflexis the most commonly used material. The silicon polymer matrix of Boraflex breaks down in the high gamma field of SNF, releasing silica and boron to the FSP. Corrective action is frequently needed. This led to an EPRI R&D program to understand the degradation process,16 and to the formation of a Boraflex Users Group, where product information and experience are shared. One outcome of the users group has been validation of the RACKLIFE code for predicting how long Boraflex may be safely used. A controversy surrounds the “burnup credit” of SNF. As fission products accumulate with burnup, some of them act as neutron absorbers, reducing the possibility of criticality. But no one can agree on how much credit can be taken for this effect. In the absence of a consensus, the NRC allows no credit to be taken for fission-product absorption, either during SNF storage or transport.
66
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
Although limiting the fuel handling practices used by the industry, such conservatism errs on the side of an increased safety of reactor operations. 5.2 FAILED FUEL Throughout the 45 years of commercial nuclear power in the USA, LWR fuel has generally exhibited high reliability with overall failure rates below 0.1%. From time to time, however, for any number of possible reasons—ignorance of a materials phenomenon, or changes in the coolant water chemistry, assembly design or operating procedures, there has been what can only be referred to as a rash of fuel failures. Most causes of failure were rapidly identified and corrected. An example is fuel densification that was encountered in the 1970s when, because of the use of interrupted sintering to produce low fuel density, pellets shrank in-reactor and caused failure of the cladding between pellets under the influence of the coolant pressure.17 Once identified, the cause of failure was eliminated by adjusting the procedure used to sinter pellets. Failed fuel rods release fission gas (and volatiles like cesium) at power inreactor, during removal from the core, and in the early stages of wet storage. If mechanical interaction has split the cladding—as opposed to failure by creep rupture that causes pin-hole leaks, slight loss of fuel may also occur. However, contamination from fuel failures has not been a significant safety issue during reactor operations (nor, apparently, during decommissioning)—perhaps because UO2 oxidizes slowly in water, swells and tends to seal the failure sites. Whatever the cause of failure, almost all SNF with failures remains in wet storage today: only a small fraction has been put into dry storage in ISFSIs, and only then because the associated reactor has been decommissioned. The Maine Yankee PWR is a good example. The reactor was shut down permanently in 1997 after 25 years operation and was successfully decommissioned over 1997– 200418; all 1,436 of its SNF assemblies are contained in 60 NAC-UMS vertical casks at the on-site ISFSI (Figure 10). In common with other reactors over 1985–1995, Maine Yankee experienced fuel failures due, among other causes, to entrained debris in the coolant and almost 300 of the SNF assemblies were deemed to contain “nonstandard” fuel because of the presence of actual or potential fuel failures; these assemblies are included in the casks. Following removal of SNF to dry storage, the SFP was filtered, drained, demolished, and removed along with the rest of the reactor. Today, the site of the Maine Yankee PWR has been returned to a “green field” condition, with only the ISFSI left to remind people of its former use.
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
67
Figure 10. The Maine Yankee ISFSI pad and NAC-UMS dry storage casks containing the reactor’s total inventory of SNF (From: New Horizon Scientific LLC.)
5.3 PHYSICAL SECURITY OF WET AND DRY STORAGE In the wake of the terrorist attacks of September 2001, the response of nuclear reactors to sabotage became a controversial issue for the US government agencies and the public. Although some expert witness on the subject was for obvious reasons classified, in general terms it was agreed that the hardened structure of reactors makes them less vulnerable than many industrial installations, which may be considered “soft” by comparison. For example, an airplane crash on the reactor dome is one case now addressed in the safety analysis of containment. The same conclusion cannot be reached about the security of FSPs.19–21 They are located either outside containment in less hardened buildings at PWRs, or within primary containment but much above grade in BWRs. In either case, loss of pool water could lead to the ignition of the Zircaloy cladding and the release of large quantities of Cs-137 (a typical FSP in the USA contains ~400 t of SNF and a Cs-137 inventory of 30–40 MCi). 21 Reracking of SNF has heightened the effect, because in the original openframe racks, SNF that was cooled for >5 days after shut down before transfer to the FSP could survive a complete loss of pool water without cladding failure. Thus much improvement could be made if SNF was removed as soon as practical to dry storage and the original open-frame storage racks were reinstalled and used. Alvarez et al.21 suggest that these changes could reduce the Cs-137 inventory of a typical FSP by a factor of four, and eliminate cladding ignition in the case of a total loss of pool water.
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
68
In contrast, dry storage casks are massive structures that are subjected to extensive mechanical testing before NRC approval, so that they will withstand very significant physical abuse before failure and any possible release of their radioactive content. There are two additional points in favor of dry casks over FSPs: • •
They contain much smaller inventories of SNF and Cs-137 activity. They are not so easily subject to a loss-of-coolant accident.
Discussion of the security of dry tasks is therefore mainly concerned with their ability to withstand intentional missile attacks or, in the worst case, an intentionally crashed plane, which could lead to fire from spilled aviation fuel. There are no clear conclusions to be made about any of these scenarios except, perhaps as a last resort, to enclose ISFSIs in hardened buildings or placed behind earth berms. On balance it does seem that dry storage of SNF is inherently less vulnerable than wet storage. One might thus conclude that the current trend towards more and more dry storage is going in the right direction in terms of physical security. 6 Summary For historical reasons, by late 2005 approximately 190,000 SNF assemblies had accumulated in storage at the 104 operating commercial reactors in the USA This inventory of ~54,000 MTHM was due to a lack of reprocessing capability, which stemmed from a fear of nuclear proliferation in the mid-1970s, and to delays in the opening of the national repository at Yucca Mountain, Utah. In consequence, fuel storage pools today are fast filling up with SNF, despite highdensity racking of assemblies. Older SNF is increasingly being dry-stored in casks placed on concrete pads in ISFSIs. There are presently 30 ISFSIs in the USA and their number is rapidly growing. Experience with both wet and dry storage of SNF assemblies has been very reassuring with no indication of fuel failure. Wet-stored SNF derives its stability from an adherent oxide layer on its Zircaloy cladding, which resists corrosion in wet storage provided it is not damaged. The behavior of SNF from the Surry-2 PWR, which was vacuum dried and kept in He-filled casks for nearly 15 years, was similarly benign: no increase in internal fission-gas pressure, or increase in cladding creep, was measured on 11 fuel rods that were destructively examined. Analysis of security issues since September 2001 suggests that dry storage of SNF at ISFSIs may be more secure than wet storage at reactors. However, off-loading SNF to dry storage as soon as possible and a return to the use of
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
69
open-frame racking of assemblies are simple measures that could substantially improve security at a spent fuel pool by removing the risk of fire in the event that the pool is drained of water. References 1. 2. 3. 4. 5. 6. 7. 8. 9.
10. 11.
12.
13.
14.
15. 16. 17. 18.
US Atomic Energy Commission, The Shippingport Pressurized Water Reactor, Addison-Wesley, Reading, Massachusetts (1958). See Nuclear Energy Institute website: http://www.nei.org. “Maximising the Assets: A status report on license renewal and power uprates”, Nuclear News, September 2005. Kranhold, K., “Nuclear-power industry sees signs of a US revival”, Wall Street Journal, November 10, 2004. House Report 109–474—Energy and Water Development Appropriations Bill, 2007. See Nuclear Regulatory Commission website: http://www.nrc.gov/waste/spentfuel-storage.html. Nuclear Regulatory Commission 10CFR 72.214 (2003). Wilson, H.W., et al., “Westinghouse Fuel Performance in Today’s Aggressive Plant Operating Environment”, LWR Fuel Performance Conference, ANS (1997). “Effects of radiation and environmental factors on the durability of materials in spent fuel storage and disposal”, IAEA-TECDOC-1316, International Atomic Energy Agency, Vienna, Austria (2002). McKinnon, M.A., et al., “BWR Spent Fuel Storage Cask Performance Test”, Pacific Northwest Laboratory Report PNL-5777 Vol. 1 (1986). Cunningham, M.E., et al., “Control of Degradation of Spent LWR Fuel During Dry Storage in an Inert Atmosphere”, Pacific Northwest Laboratory Report PNL-6364 (1987). Chin, B.A., Gilbert, E.R., “Prediction of Maximum Allowable Stresses for Dry Storage of Zircaloy-clad Spent Fuel in Inert Atmosphere”, Nuclear Technology, 85 (1989): 57–65. Kenneally, R.M., Kessler, J.H., “Cooperative Research Program on Dry Cask Storage Characterization”, Proceedings 8th International Conference on Nuclear Energy, Paper 8180, ASME (2000). Einziger, R.E., et al., “Examination of Spent PWR Fuel Rods after 15 Years in Dry Storage”, Proceedings 10th International Conference on Nuclear Energy, Paper 22456, ASME (2002). Hill, T.J., Fillmore, D.L., “Managing Spent Nuclear Fuel at the Idaho National Laboratory”, this workshop. Michaels, A., Kessler, J.H., “Research and Development Overview”, Dry Storage Information Forum, Key Biscayne, Florida, May 10–11, 2005. Freshley, M.D., et al., “Irradiation-induced Densification of UO2 Pellet Fuel”, Journal of Nuclear Materials, 62 (1976): 138–66. See Maine Yankee website: http://www.maineyankee.com/storage/Default.html.
70
SPENT FUEL STORAGE AT COMMERCIAL U.S. REACTORS
19. US General Accounting Office, “Spent Nuclear Fuel—Options Exist to Further Enhance Security”, GAO-03-246 (July 2003). 20. National Academy of Science, “Safety and Security of Commercial Spent Nuclear Fuel Storage: Public Report (2005)—Board on Radioactive Waste Management (BRWM)—National Research Council” (2005). 21. Alvarez, R., et al., “Reducing the Hazards from Stored Spent Power-Reactor Fuel in the United States”, Science and Global Security, 11 (2003): 1–51.
MANAGING SPENT NUCLEAR FUEL AT THE IDAHO NATIONAL LABORATORY
T. J. HILL1∗ AND D. L. FILLMORE2 1 Idaho National Laboratory, Idaho 2 Idaho Closure Project, Idaho Abstract: The Idaho National Laboratory (INL) has a large inventory of diverse types of spent nuclear fuel (SNF). This legacy derives from the history of the INL as the National Reactor Testing Station, and from its mission to recover highly enriched uranium (HEU) from SNF and to test and examine SNF after irradiation. The INL also has a large diversity of SNF storage facilities, around 50 years old. SNF at INL has many forms—from intact assemblies down to metallurgical mounts, and some fuel has been wet-stored for over 40 years. SNF is stored bare or in metal cans under water, or dry in vaults, caissons or casks. Inspection shows varying corrosion and degradation of the SNF and its storage cans. SNF has been stored in 10 different facilities: 5 pools, one cask storage pad, one vault, two generations of caisson facilities, and one modular Independent Spent Fuel Storage Installation (ISFSI). The pools range in age from 40 years to the most modern in the US Department of Energy (DOE) complex. The nearterm objective is to move SNF from older pools to interim dry storage, allowing shutdown and decommissioning of the older facilities. This move involves drying methods that are dependent on fuel type. The long-term objective is to have INL SNF in safe dry storage and ready to be shipped to the National Repository. The unique features of the INL SNF requires special treatments and packaging to meet the proposed repository acceptance criteria and SNF will be repackaged in standardized canisters for shipment and disposal in the National Repository. Disposal will use the standardized canisters that can be codisposed with high-level waste (HLW) glass logs to limit the total fissile material in a repository waste package. The DOE standardized canister also simplifies the repository handling of the multitude of DOE SNF sizes and shapes. Keywords: spent nuclear fuel, dry storage, wet storage, drying spent nuclear fuel, dry storage test program
______ ∗
To whom correspondence should be addressed: Thomas J. Hill, Space Nuclear Systems & Technology, Idaho National Laboratory, PO Box 1625, Idaho Falls, Idaho 83415, USA; e-mail:
[email protected] 71 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 71–84. © 2007 Springer.
72
SPENT FUEL AT IDAHO NATIONAL LABORATORY
1 Introduction The Idaho National Laboratory (INL) was established in 1949 as the National Reactor Testing Station. As the name implies, the mission was reactor testing. Over the years, 52 reactors were constructed and operated by a variety of organizations for a variety of programs. Portions of the site operated under different Federal Agencies and field offices of the Atomic Energy Commission/ Department of Energy (DOE). Most of the spent nuclear fuel (SNF) generated by these reactors has remained on site. A chemical reprocessing facility was constructed in 1950 for the recovery of fissile materials from SNF. SNF was also sent to the INL from many other locations for testing, examination, and reprocessing. Sources of this SNF included naval reactors, DOE and university test reactors, and even some commercial reactors. SNF was stored in a number of facilities at several different locations, including: Test Area North (TAN), Materials and Fuel Complex (MFC) (formerly Argonne National Laboratory—West), Idaho Nuclear Technology and Engineering Center (INTEC), Reactor Technology Complex (RTC), and Power Burst Facility (PBF). Much of the fuel was reprocessed for the recovery of enriched uranium. At present, 257 metric tons of heavy metal (MTHM) with over 250 different types of SNF is being managed at four facilities at the INL. This paper addresses management of all SNF at the INL, except for naval fuel. 2 INL Current Status 2.1 SPENT NUCLEAR FUEL DESCRIPTION There are many characteristics of SNF important to its management. Chemical characteristics of its fabrication materials, such as the fuel meat and cladding, and other constituents, determine how it can be stored and what treatment may be required for it to meet geologic repository acceptance criteria. Physical characteristics such as length, weight, heat generation rate, radiation levels, fissile content and enrichment, and its physical condition, dictate storage and handling requirements. The INL has a large inventory of diverse types of SNF, possibly the most diverse and varied inventory in the world. To illustrate the point, Table 1 gives some of the materials of fabrication and Table 2 describes the variety in physical characteristics of the fuel. The fuel also consists of several different fissile materials of varying enrichments. Much of the SNF is highly enriched in 235U (>60%) and was sent to the INL for 235U recovery. There is also a significant quantity of typical commercial light-water reactor fuel that was brought to the INL for examination and testing. The INL also stores fuel that contains 233U and
SPENT FUEL AT IDAHO NATIONAL LABORATORY
73
thorium. Specialized test reactor fuel such as the sodium-bonded metal fuel used in the Experimental Breeder Reactor-II is also stored on-site. In order to understand the inventory and quantity of the SNF, the projected 2035 inventory was placed into groups based on fabrication materials. Table 3 describes the 16 groups. TABLE 1. Typical INL SNF fabrication materials Cladding material Aluminum Stainless steel Zirconium
Fuel meat composition Uranium metal Uranium oxide Uranium alloy Uranium carbide
Other materials Stainless steels Thorium carbide Beryllium oxide Metallic sodium Thorium oxide
TABLE 2. INL SNF physical characteristics Characteristic Weight Length Heat generation rate Radiation field Weight
Range 0.01–1,600 kg 10–410 cm 0–600 w 0–10,000 R/hr Scrap to excellent
The SNF is stored in a variety of configurations. Some is dry-stored in casks, caissons, underground silos, or vaults, and much is stored underwater in spent fuel pools. Some fuel is stored unencapsulated as bare units, either intact or having had structural material removed. Some fuel is stored in containers, sometimes referred to as cans, having been previously disassembled for destructive examination or security purposes. At one time the cans were dry on the inside; however, a number are now known to have leaked. Some cans have been replaced underwater to retain the fuel handling capability. Some fuel has been in storage for >40 years and physical degradation has occurred. 2.2 STORAGE FACILITY DESCRIPTION 2.2.1 Wet storage facilities Both wet and dry storage are used at the INL. The wet-stored fuel is in concrete pools, some lined with stainless steel, either at the INL reactor site or at INTEC, the nonreactor facility formerly the Idaho Chemical Processing Plant (ICPP).
SPENT FUEL AT IDAHO NATIONAL LABORATORY
74
TABLE 3. SNF Inventory Groups at INL Group 1
4
Group description Intact, low-enriched (<5%), UO2 clad with Zr or stainless steel Intact, medium-enriched (5–20%), UO2 clad with Zr or stainless steel Intact, highly enriched (>20%) UO2 clad with Zr or stainless steel Disrupted, low-enriched, UO2
5
Disrupted, highly enriched UO2
6
14
Intact medium-enriched U-zirconium hydride clad with Al, Zr, or stainless steel Some disrupted, highly enriched U-zirconium hydride clad with Zr or stainless steel Low-enriched, U metal or U alloy or Zr or Mo with various claddings Highly enriched, U metal or U alloy or Zr or Mo with various claddings Highly enriched UC in SiC coated particles in a graphite matrix Highly enriched UCin non-SiC coated particles in a graphite matrix Highly enriched UC in a nongraphite Matrix Highly enriched U-233 with thorium in Zr cladding Metallic sodium bonded
15
Aluminum clad U compounds
16
Others not included above
2 3
7
8 9 10 11 12 13
Quantity 31 types; 30 m3 76.8 MTHM 8 types; 1.4 m3 4 MTHM 21 types; 9.3 m3 8.7 MTHM 34 types; 145 m3 87.5 MTHM 44 types; 24 m3 6.2 MTHM 97 types; 6.6 m3 1.8 MTHM
Fuel example Typical commercial
24 types; 1.3 m3 0.2 MTHM
TRIGA Flip
14 types; 0.8 m3 2.0 MTHM 6 types; 2 m3 3.9 MTHM 1 type; 196 m3 23 MTHM 7 types; 35 m3 3 MTHM 2 types; 5 m3 0.06 MTHM 1 type; 52 m3 39 MTHM 33 types; 15 m3 60 MTHM 14 types; 43 m3 3.4 MTHM 5 types; 4.3 m3 0.2 MTHM
HWCTR
PBF Shippingport PWR TMI-2 TORY Standard TRIGA
Fermi Fort St. Vrain Peach Bottom graphite SRE Shippingport LWBR Fermi blanket EBR-II ATR
The CPP-666 pool is the most up-to-date storage pool in the DOE complex (Figure 1). Built to store SNF waiting for reprocessing, it is lined with stainless steel and has modern leak detection and water purification systems. Other pools—such as the unlined TAN concrete pool (TAN-608) built in 1955 for the Aircraft Nuclear Propulsion program and used to store TMI-2 fuel—have been emptied and the SNF put in dry storage casks. The Advanced Test Reactor pool supports reactor operations and stores SNF until transfer to the pool at INTEC.
SPENT FUEL AT IDAHO NATIONAL LABORATORY
75
Figure 1. The modern steel-lined CP-666 Storage Pool at the Idaho Nuclear Technology and Engineering Center (INTEC)
2.2.2 Dry storage facilities Dry storage facilities are now consolidated at INTEC, MFC, and Fort St. Vrain, Colorado. The dry storage casks at TAN SNF storage pad have been relocated to INTEC on a new pad. Commercial light-water reactor fuels, bare and consolidated assemblies, have been in various commercial storage casks for up to 15 years under a joint DOE and Nuclear Regulatory Commission (NRC) testing program; Figure 2 shows the casks at the INL.
Figure 2. Commercial dry storage casks used in SNF tests at the INL Left to Right: REA 2023; VSC-17; TN 24P; V21; 125B; MC-10
76
SPENT FUEL AT IDAHO NATIONAL LABORATORY
INTEC has four different types of dry storage, not counting the dry storage cask systems. The Irradiated Fuel Storage Facility (IFSF) is an air-cooled vault originally built to store the Fort St. Vrain SNF. It presently stores many different types of SNF. The first generation of storage caisson, CPP-749, has steel-lined below-grade individual vaults, built to store Peach Bottom graphite SNF (a general view of the CPP-749 facility is shown in Figure 3). The secondgeneration caissons were a significant improvement in design and were built to store the Shippingport light water breeder reactor SNF.
Figure 3. The below-grade storage facility CP-749 at the INL used to dry-store Peach Bottom Reactor SNF
MFC operates a below-grade silo storage facility (Radioactive Scrap and Waste Facility) which was built in 1965. This facility stores spent fuel and remote handled mixed and radioactive waste, primarily from the EBR-II. The newest dry storage facility is a modified NUHOMS® design built to hold TMI-2 reactor SNF removed from the damaged core after the accident. This was the second DOE facility licensed by the NRC, and the first one for which DOE actually prepared the license application. 2.3 RECEIPT AND TRANSFER The INL has an active SNF generation and receipt program. The ATR is one of the few operating reactors in the DOE complex, and it is scheduled to operate till 2025. The ATR produces over 30 spent fuel assemblies each year. The SNF
SPENT FUEL AT IDAHO NATIONAL LABORATORY
77
is cooled in the reactor pool for a short time and then transferred to INTEC for interim management. The DOE programmatic Environmental Impact Statement (EIS)1 and the Foreign Research Reactor EIS2 both identify the INL as the receiving location for nonaluminum-based SNF presently located at the DOE sites as well as SNF generated in foreign, domestic and university reactors. This includes all SNF containing metallic sodium. The INL will, according to the EIS record of decision, ship all existing INL aluminum-based SNF to the Savannah River Site. MFC will receive sodium-bonded fuel removed from the Fast Flux Test Facility at the Hanford site for treatment. The INL will receive nonaluminum-clad SNF from 16 universities, 8 domestic sites, 18 foreign sites, and 5 DOE sites. 3 INL Short-Term Objectives 3.1 SAFE INTERIM STORAGE Some of the INL SNF facilities have been in operation for over 40 years, although they had an operating lifetime of 20 years when constructed. These facilities do not meet present standards and needed to be shut down and decommissioned. Some of these older facilities contained SNF that has been in storage for over 40 years. The condition of these SNF types deteriorated and presented problems in safe management. Both of these situations were identified as vulnerabilities to the INL and corrective actions taken to eliminate the vulnerabilities. The DOE reached an agreement with the State of Idaho3 in 1995 to move the SNF to safe dry storage by 2023 and have it removed from the state by 2035. 3.1.1 Consolidate SNF and shut down older facilities The INL had several older, wet SNF storage facilities that did not meet present standards4. As the SNF was removed from these facilities, these facilities have been shut down and decommissioned. The first facility targeted for this activity was the CPP-603 pool. The CPP-603 pool consisted of three bare concrete pools that contained a large variety of SNF. The north and middle pools stored SNF that hung from a monorail on hangers that kept the fuel in the proper configuration. There had been considerable corrosion of the carbon steel hangers and the SNF containers. The south pool kept the SNF units in racks that sat on the basin floor. The north and middle pools were emptied of SNF by 1996. This was accomplished by transferring the SNF that was in relatively good condition to CPP-666 and consolidating the SNF that would require additional treatment into the south
78
SPENT FUEL AT IDAHO NATIONAL LABORATORY
pool. The remaining fuel in the CPP-603 pool was repackaged into new storage cans or buckets and transferred to CPP-666 or to IFSF. A total of 1,340 units of SNF were transferred between 1994 and April 2000. The project was completed 8 months ahead of schedule. Decontamination and decommissioning of the facilities is now underway. The SNF storage pool located at the TAN was also a bare concrete pool that stored SNF in metal racks. The primary SNF type that was stored in the TAN pool was the TMI-2 SNF debris contained in stainless steel containers. There was also some commercial and loss of fluid test SNF in the pool. The TAN facilities, located 20 miles from INTEC, were emptied of SNF in 2002 and the material moved to INTEC. The MTR canal was a pool that was constructed to support the operation of the MTR reactor and had been maintained to store SNF that was left over from the research and testing programs at the INL. It was a bare concrete pool that contained SNF in storage racks. This SNF was moved to INTEC in 2001. The PBF pool was a small stainless steel lined pool that was built to support the operation of the PBF reactor. The reactor has been shut down and defueled. The only fuel stored in the pool was the PBF fuel. The INL moved this SNF to INTEC in 2002. The INL SNF management plan called for moving all the SNF from these older, remote facilities into modern wet or dry storage and eventually all into dry storage. Significant projected cost savings have been realized from the shutdown of these older facilities. The IFSF plays an important part in the INL consolidation plans. The transfer of all of the INL SNF from wet storage to dry storage is an important step in reducing storage costs as well as meeting the State of Idaho agreement. The IFSF was constructed to store all of the SNF from the Fort St. Vrain reactor in Colorado. Because two-thirds of the fuel will remain in Colorado until it is time to prepare it for repository disposal5 there is excess room for the storage of other SNF types. SNF was moved out of the older facilities into IFSF and the older, wet storage facilities were shut down without having to wait for new facilities to be built. Much of the SNF that was moved out of the CPP-603 pools was moved into IFSF. The foreign research reactor fuel that is now being received is being put into IFSF, as well as all the SNF removed from the MTR and PBF facilities. The Radioactive Scrap and Waste Facility (RSWF) is used for the bulk of interim fuel storage at MFC. Spent fuel is dry-stored, below grade, in passively cooled cathodically protected carbon steel liners or silos. All of the sodiumbonded fuel destined for electrometallurgical treatment will pass through RSWF and the high-level waste (HLW) forms will be stored there awaiting geological disposal.
SPENT FUEL AT IDAHO NATIONAL LABORATORY
79
The INL constructed an NRC-licensed facility6 for the dry storage of the TMI-2 SNF. DOE decided to make this facility the first NRC licensed facility at a DOE site. This design is a modification of an existing NUHOMS® design that was adapted for the TMI-2 SNF. DOE authorized the construction of a privatized SNF storage facility. The project is called the Spent Nuclear Fuel Dry Storage Project (SNFDSP) and the contract was awarded to Foster Wheeler Environmental Corporation.7 This facility would initially accept three types of INL SNF but it can be expanded to accept the entire INL SNF inventory. It would have the capability to dry and characterize the SNF to meet the repository acceptance criteria. It is currently on hold as DOE decides its long-term SNF management approach for the INL. 3.1.2 Maintaining capabilities MFC has two large inert hot cells which will be maintained in their current operational status. Electrorefining of fuel takes place in the fuel-conditioning facility while final waste form production is carried out in the Hot Fuels Examination Facility (HFEF). In addition, HFEF is actively providing spent fuel characterization services to DOE and non-DOE customers alike. The TAN hot shop is the largest operational hot shop in the USA. It was constructed as part of the Aircraft Nuclear Propulsion program in the 1960s. It was used to service the dry storage casks located on the SNF dry storage pad at TAN. It was also used to dry the debris from the TMI-2 reactor which has been stored underwater at TAN for over 20 years. It is currently being used to support decontamination and decommissioning activities at TAN. This facility is scheduled for shutdown in the fall of 2006 and will eventually be dismantled. 3.1.3 Technology development The technology needed to characterize the INL SNF and prepare it to meet repository acceptance criteria did not exist when reprocessing was terminated. The INL began a technology development program to ensure that the technology would be available when it was needed. The program focused on three major areas: drying, conditioning, and nondestructive characterization. The first technology developed and implemented at the INL was an ultrasonic system to nondestructively look inside sealed stainless steel cans to determine if there was water inside and also to determine the condition of the fuel inside the cans. Many research projects investigated the effects of water on the SNF for long periods of time, and methods to ensure water removal from SNF. The technologies needed for long term dry storage and dispositioning include: (1) removing the free and bound water; (2) determining the effects of hydrogen from the radiolysis of water on the materials; (3) material interactions
80
SPENT FUEL AT IDAHO NATIONAL LABORATORY
between the SNF and the containers; and (4) material corrosion modeling for all of the SNF and container materials. The technologies needed for conditioning the fuel for repository dispositioning include: (1) removal and stabilization of metallic sodium; (2) removal of metallurgical mount epoxy; and (3) methods to ensure that the condition of the fuel after 40 years of dry storage will still meet the transportation and repository acceptance criteria. The technology needed for characterization included nondestructive assay to determine the radionuclide inventory of the container, including the fissile material inventory, and nondestructive examination of the SNF condition. 3.2 PREPARE FUEL TO MEET STORAGE CRITERIA Most of the INL SNF inventory has been stored underwater for many years. Some of the fuel stored as been bare assemblies and some of it has been in cans. For the fuel that has remained intact, the drying is fairly simple. For the fuel that is not intact or that is stored in cans that are not intact, the drying is more involved. One example of the dry process is that designed for fuel being moved into the IFSF from CPP-603 basins. The fuel needed to be hot vacuum-dried. The SNF had insufficient decay heat for drying so it had to be actively heated as well as vacuum-dried. However, the SNF included aluminum-clad SNF what could not be dried at high temperatures. In addition, a drying system had to be designed to fit with in the existing IFSF SNF receiving area. A second example of a unique drying station is the one designed for the TMI-2 SNF. The debris has been stored in stainless steel cans full of water for over 20 years. The stainless steel canisters also contained low-density concrete as well has the SNF debris that had become saturated with water. It was necessary to design a hot vacuum-dry system that would heat the debris to over 800°C to drive all of the free as well as the bound water. 4 INL Long-Term Objectives 4.1 REPOSITORY AND TRANSPORTATION REQUIREMENTS The SNF and its container will need to meet strict transportation and repository acceptance criteria in order to be transported and dispositioned in a repository. The existing transportation criteria are contained in 10 CFR 71. The INL assumes that these requirements will not change significantly over the next 30 years and plans to package the SNF now so that it will meet these requirements and will not have to be repackaged prior to shipment to the repository. This involves treating the fuel and ensuring the structural and criticality control features employed during packaging will meet the 10 CFR 71 criteria.
SPENT FUEL AT IDAHO NATIONAL LABORATORY
81
The repository acceptance criteria are still evolving as the design of the repository is being developed. There are many of the criteria that are firm enough that design solutions to address them can be developed and implemented now. 4.1.1 Treatment to meet repository requirements Nuclear criticality in the repository was an issue that surfaced early in the repository planning. The initial repository design called for large-diameter waste packages. For HEU fuel this design represented a difficult task to show that no criticality would occur over the lifetime of the repository. During discussion with DOE-Office of Civilian Radioactive Waste Management (RW) the idea of putting only small quantities of HEU in a waste package was determined to simplify the criticality analysis. However, the idea of having to handle a very large number of small diameter waste packages was not appealing to the repository operators. A concept was identified where the SNF and the HLW glass logs could be packaged together in the same waste package. This allowed the repository to use large-diameter waste packages, yet limited the quantity of fissile material in a waste package. This concept was named codisposal.8,9 A design of a waste package that contained a ring of five HLW glass logs around the outside and one SNF canister in the middle was prepared. The utilization of this “extra” space in the waste package would allow the disposition of small quantities of HEU yet not increase the number of waste packages or increase handling costs. The original design has been expanded and the concept has been adopted by RW. The codisposal concept also simplified the handling of the INL SNF in the surface facility. Because of the diversity of SNF sizes DOE-RW felt that it would be impossible to design an effective means to handle all of the SNF assemblies one at a time. The original RW surface facility design for SNF handling included moving one assembly at a time from the transportation cask to the waste package. This is still the plan for the commercial SNF, where there is some consistence in the design of the handling fixtures. The standard canister design simplified the handling process for the DOE SNF. Four standard canisters were designed. The dimensions of 18 inch and 24 inch diameter were chosen with lengths of 10 feet and 15 feet. Forty years of research, development, and operation of liquid metal cooled fast breeder reactors have generated ~60 MTHM of sodium-bonded SNF. The bulk of the fuel is from the EBR-II reactor (~25 MTHM) and the Fermi reactor (34 MTHM) and is stored at facilities on the INL. This fuel is distinguished from commercial uranium oxide fuel by use of uranium or plutonium metallic alloy fuel bonded to the cladding by metallic sodium. The presence of sodium potentially complicates disposal certification and licensing for geological disposal because the draft waste acceptance criteria does not allow spent fuel
82
SPENT FUEL AT IDAHO NATIONAL LABORATORY
which contains explosive, pyrophoric, or chemically reactive material.10 The Record of Decision for the Treatment and Management of Sodium-Bonded Spent Nuclear Fuel chose the electrometallurgical treatment at the MFC as the means to stabilize the chemically reactive sodium found in the EBR-II fuel type. An electrorefiner, using a molten salt electrolyte, is used to separate the sodium and fission products from the fuel meat, resulting in high-level radioactive ceramic and metallic waste forms which will be qualified for geological disposal. Treatment of the Fermi fuel has been postponed pending further study of the electrometallurgical and alternative treatment technologies. A number of INL SNF types consist of small quantities of unique fuel, sometimes referred to as “dogs and cats”. A few of these SNF types consisted only a few quart-size cans of severely disrupted SNF. In one of the INL SNF task team meetings an observation was made that “if we could just make this fuel as durable as other fuels it would be acceptable.” The idea of a highintegrity canister (HIC) was born.9 A HIC would be constructed out of extremely corrosion resistant materials, such as zirconium with diameters small enough that the container would be critically safe by geometry. The HIC as designed will not corrode over the lifetime of the repository, will not break open even under beyond design basis accidents, and will not go critical. Nuclear criticality analysis will still be necessary because several HICs may be placed into a standard canister. The presence of organic material in the repository is as yet an unanalyzed potential problem.10 Several potential scenarios have been hypothesized. Some of the INL SNF contains epoxy in the form of sample mounts for metallurgical examination. These may be considered an organic material. The INL is presently evaluating technologies that might be used to remove epoxy as part of a treatment program to meet repository requirements. The chemical reactivity of uranium spent fuel is another issue being addressed by researchers at the INL in order to meet repository safety criteria. Uranium hydride can form as a byproduct of aqueous corrosion and its pyrophoric nature is well known. Experiments sponsored by the National SNF Program (NSNFP) at MFC have resulted in a clearer understanding of the oxidation kinetics of uranium hydride under repository conditions.11 4.2 REPOSITORY DISPOSITIONING Soon after the decision was made by DOE to terminate reprocessing of SNF for the recovery of enriched uranium and plutonium, the INL began exploring other options for treatment and final disposition of SNF. Initial contacts were made with DOE-RW in 1996 and a task team was set up with INL and DOE-RW to evaluate possible options for dispositioning SNF. This task team prepared a report that outlined the direct disposal options for the INL SNF. The INL has
SPENT FUEL AT IDAHO NATIONAL LABORATORY
83
also been working very closely with the National SNF Program located at the INL to ensure that all Yucca Mountain planning, recommendations, and licensing activities include the INL SNF. The INL has also opened contacts with the NRC through DOE-RW and the NSNFP.8,9 Because of the diversity of the INL inventory it was concluded that each SNF type could not be individually addressed in the repository analysis. An evaluation was performed and it was determined that the SNF types could be grouped by characteristics important to the particular analysis. The group could then be represented by one fuel. This approach greatly simplified the analysis.9 The INL SNF was included in the Yucca Mountain Environmental Impact Statement as part of the SNF inventory, and the electrometalurgical products as HLW products. The data included chemical, physical, and radiological properties of the material. The INL inventory was included in the analysis of all of the potential impacts. There were no adverse impacts attributed to the INL SNF. INL SNF is also included in the Yucca Mountain Site Recommendation Report that is presently being prepared by DOE-RW and will be included in the repository license application that will be submitted to the NRC. The same grouping approach is being used in these two documents that were used in the EIS, with additional in enhancements in the characterization data. 4.3 KEY MILESTONES The following key milestones relate to the Consent Order and Settlement Agreement. These milestones, if not met, could result in halting all shipments of SNF into the State of Idaho as per the terms of the Settlement Agreement. • • • • • •
Commencement of TMI-2 fuel debris movement from TAN wet basins— completed 03/31/1999 Removal of all SNF from CPP-603 wet storage by 12/31/2000—completed 04/28/2000 Removal of TMI-2 fuel debris from TAN wet basins by 06/01/2001— completed 06/2001 Begin loading spent fuel into dry storage (excluding TMI-2 fuel debris covered separately) by 07/01/2003—completed 06/2003 Removal of all SNF from wet storage by 12/31/2023 Completion of shipment of SNF from the State of Idaho by 01/01/2035
5 Conclusions A legacy of the INL was a large, diverse inventory of SNF in a large number of aging facilities. With termination of reprocessing there was no path forward for disposition of SNF. Managing this legacy has included: characterization of the
84
SPENT FUEL AT IDAHO NATIONAL LABORATORY
SNF and facilities, resolving SNF storage vulnerabilities, consolidating SNF and closing facilities, managing domestic and foreign assigned fuel receipts, developing technologies needed, and ensuring that the SNF inventory or its treatment byproducts are acceptable in a federal repository. References 1.
Department of Energy’s Record of Decision for Programmatic Spent Nuclear Fuel Management and Idaho National Engineering Laboratory Environmental Restoration and Waste Management Programs, (60 FR 28680) May 30, 1995 and Amendment, March 1996. 2. Department of Energy’s Record of Decision for Nuclear Weapons Nonproliferation Policy Concerning Foreign Research Reactor Spent Nuclear Fuel, (61 FR 25092), May 13, 1996. 3. Settlement Agreement, State of Idaho and the Department of Energy, October 16, 1995 re: Public Service of Colorado v. Batt, No. CV 91-0035-S-EJL (D. Id.) and United States v. Batt, No. CV-91-0054-S-EJL (D. Id.) 4. Defense Nuclear Facilities Safety Board Recommendation 94-1 to the Secretary of Energy, May 1994 and DOE Implementation Plan, February 28, 1995. 5. Settlement Agreement, Contract No. DE-AC07-96ID13425, February 9, 1996, between the United States Department of Energy and Public Service Company of Colorado, Modifying Contract at (04-3)-633, as amended, and Agreement No. DESC07-79ID01370, as amended. 6. INEEL ISS Facility Project DE-AC07-99ID13727. 7. Spent Nuclear Fuel Dry Storage Project DE-AC07-00ID13729. 8. Technical Strategy for the Treatment, Packaging, and Disposal of Aluminum Plate Fuels, June 1996. 9. Technical Strategy for the Management of INEEL Spent Nuclear Fuel, Volume I, March 1997. 10. Civilian Radioactive Waste Management System, MGDS Waste Acceptance Criteria, B00000000-01717-4600-00095, Rev. 00, Draft A, September 1997. 11. Totemeier, T.C., Pahl, RG., Frank, S.M., “Oxidation Kinetics of Hydride-Bearing Uranium Metal Corrosion Products”, Journal of Nuclear Materials, 265 (1999): 308–320.
RADIOLOGICAL PROBLEMS OF SPENT FUEL STORAGE
ASSESSMENT OF ENVIRONMENTAL IMPACT OF REACTOR FACILITIES IN KAZAKHSTAN
K. K. KADYRZHANOV, S. N. LUKASHENKO,∗ AND V. N. LUSHCHENKO Institute of Nuclear Physics (INP), Almaty, Kazakhstan Abstract: The two main reactor facilities in Kazakhstan today are the WWR-K research reactor close to Almaty and the BN-350 fast reactor in Aktau on the Caspian Sea. The WWR-K reactor is operated as a source of neutrons for irradiation of materials, production of radioisotopes, and basic neutron physics. Until 1998, BN-350 was operated as a power reactor to produce electricity and to desalinate water for the region for over 25 years; with international help it is now being decommissioned. This paper describes radioecological surveys that were performed around both reactor facilities, the techniques used, and results obtained. It was found that measured values of radionuclides in the regions surrounding both reactors were no higher than the values measured elsewhere which were caused by radioactive fallout. It was concluded that the reactors have minimal additional impact on the environment. Keywords: WWR-K and BN-350 reactor operation, radioecological surveys, global radioactive fallout, 137Cs activity, 238,239 + 240Pu, 90Sr, tritium
1 Background About 30 km from Almaty—the largest city in Kazakhstan, and in the territory of Alatau—is situated the research nuclear reactor belonging to the Institute of Nuclear Physics of the National Nuclear Centre of the Republic of Kazakhstan (INP NNC RK). The research nuclear reactor WWR-K (Kazakhstani water-moderated reactor, 10 MWt power) has been in operation from the autumn of 1967. Since then the reactor worked continuously without variance from a normal operating mode of 20-day cycles, up until the Chernobyl accident in 1988, when all research reactors in the former Soviet Union were stopped. Repeated start-ups
______ ∗
To whom correspondence should be addressed: S. N. Lukashenko, Institute of Nuclear Physics, Ibragimov Str.1, 050032, Almaty, Kazakhstan; e-mail:
[email protected] 87 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 87–94. © 2007 Springer.
88
ENVIRONMENTAL IMPACT OF REACTORS IN KAZAKHSTAN
of the reactor occurred in 1997 after work to improve seismic stability and to modernize its control systems had been performed. By its design and physical features WWR-K is considered to be one of the safest types of reactor, and by its experimental capabilities and efficiency of use, one of the best research reactors in the whole of the Commonwealth of Independent States (CIS). WWR-K has operated for more than 25 years. The dual-purpose industrial fast neutron reactor facility BN-350 in Aktau was designed to produce steam and breed plutonium.1 Steam generated by the reactor was used to produce electricity (up to 150 MW) and to desalinate water from the Caspian Sea (up to 120,000 t/day). BN-350 began operation in July 1973 with a design life of 20 years. After this period was over in July 1993, subsequent operation was carried out based on annual decisions to prolong the term of operation for another year. From December 1998 BN-350 stood idle and on April 22, 1999, the RK Government Resolution N456 was adopted “On Withdrawal of BN-350 Reactor Facility in Aktau in the Mangistau Region”. Hence, the total operating period of the reactor was finally more than 25 years. 2 Experimental Techniques The basic mechanism for the potential impact of reactor installations on the radioecological conditions of their surroundings is the emission of radioactive gases and aerosols into a surface layer of the atmosphere.2,3 Accordingly, the greatest attention has been paid to analysis of the radionuclide content of the soil, which accumulates radioactive fallout from the atmosphere. To estimate radioecological conditions in the studied territories a complex set of measurements including areal gamma surveys, sampling of objects of the environment and their radionuclide analysis in the laboratory was carried out. In the Alatau settlement around WWR-K the level of radon concentration in the air of occupied houses and offices was also estimated. For processing and representing results modern computer methods based on geographic information system (GIS) technologies have been used. The first stage of the measurement at BN-350 is complete and consists of surface and layer-wise soil sampling, ground and surface water sampling, and vegetation and bottom sediment sampling within the reactor facility site, the buffer area, and the radiation-control zone of the BN-350 reactor. Figure 1 shows the scheme of surface soil sampling and bottom sediment sampling within the buffer area and the radiation-control zone. The INP NNC RK analytical laboratories have carried out gammaspectrometric analysis of the total number of the samples collected, while a part of them were analyzed for 238,239 + 240Pu and 90Sr content. In addition, the tritium content of water samples was measured.
ENVIRONMENTAL IMPACT OF REACTORS IN KAZAKHSTAN
89
Figure 1. The sampling scheme in the buffer area and radiation-control zone of the BN-350 reactor facility and 238Pu/239 + 240Pu isotope ratio for Karakol Sor samples
3 Results and Discussion 3.1 WWR-K REACTOR The gamma survey of streets, avenues, yards, human and cottage settlements, and industrial objects was carried out on a regular grid of 20 × 10 m. For points at the corners of human settlement quarters, exposure rates were metered and topographic positioning was made using satellite global positioning system (GPS) receivers. The results of the gamma surveys showed that the measured values for the equivalent dose of gamma radiation were in the range 0.08–0.21 µSv/h, which is within the average background values for the region of 0.25–0.27 µSv/h. According to analysis of the radionuclide content of selected ground surface samples, the predominant artificial radionuclide was 137Cs. The concentration of 137 Cs in surface layers (Figures 2 and 3) measured with a semiconductor gamma spectrometer ranged from 0.5 to 42 Bq/kg, with an average value of 8.3 Bq/kg, corresponding to the background level from global fallout. The concentration of natural radionuclides in the ground is typical for the given type of ground.
90
ENVIRONMENTAL IMPACT OF REACTORS IN KAZAKHSTAN
Figure 2. 137Cs activity values in soil around WWR-K reactor (size of circles corresponds to level of concentration)
Figure 3. Frequency of occurrence of 137Cs activity around the WWR-K reactor
The absence of a double hump in the distribution curve in Figure 3 suggests there is only one source of 137Cs. The distribution of 137Cs with depth averaged
ENVIRONMENTAL IMPACT OF REACTORS IN KAZAKHSTAN
91
on the basis of four profiles is shown in Figure 4. The depth profile is typical of global fallout. For estimating the body burden for the population of the Alatau settlement, selective measurements were made of the effective equilibrium volumetric activity of radon in the air of homes and offices. The measured values were 20–40 Bq/m3, compared with a maximum permissible value of 200 Bq/m3 for buildings in use.
Figure 4. Averaged 137Cs activity with depth
For an estimation of the level of environmental contamination by tritium, snow sampling and subsequent analysis for tritium were carried out using a liquid scintillation beta spectrometer. The concentration of tritium in samples varied from 8 to 119 Bq/l, with an average value of 14 Bq/l. For a number of sampling points located outside the settlement, an increased concentration of tritium was registered in comparison with the background values (Figure 5) and points with slightly increased tritium contents were concentrated in the direction of a wind rose from the WWR-K reactor. It should be noted that the radiation safety norms in Kazakhstan allow a maximum permissible tritium concentration of 7.7 × 103 Bq/l.
92
ENVIRONMENTAL IMPACT OF REACTORS IN KAZAKHSTAN
Figure 5. Tritium concentrations around the WWR-K reactor
3.2 BN-350 REACTOR The radiation environment around a nuclear power station forms from carryover of radioactive gas and aerosols to the near-surface layer of the atmosphere and from waste water discharge. For the BN-350 reactor radionuclide analysis of the soil and of water and bottom sediments from Karakol Sor into which waste water was discharged after decontamination is of particular interest. Quantitative data obtained made the error-free conclusion possible that technogenic radionuclide content of environmental samples collected within the buffer area and the radiation-control zone of the BN-350 reactor facility is no greater than background values, i.e., corresponding to the average level of global radionuclide fallout in the West Kazakhstan territory (see Table 1). Available retrospective material on radioactive release4 authenticates that the radioactive isotope content of the reactor waste discharge was never more than the annual acceptable level during the whole operating period of the reactor. Atmospheric gas–aerosol carryover amounted to tenth parts of a percent from maximum permissible emission.
93
ENVIRONMENTAL IMPACT OF REACTORS IN KAZAKHSTAN
TABLE 1. Radionuclide content of the soil in buffer area and radiation-control zone of the BN-350 reactor Activity range in samples studied (Bq/kg)
Global radionuclide fall out for Western Kazakhstan (Bq/kg)
137
Cs
<2.1–33.5
<0.5–50
90
Sr
<5–10.5
<5–30
<1.4–2.5
Not measured
<0.05–0.55
<0.05–5
Radionuclide
152
Eu
239 + 240
Pu
The precision and sensitivity of the analysis is confirmed by the measured Pu/239 + 240Pu isotope ratios for the samples. According to assessments of other authors5,6 this ratio for global radionuclide fallout in the northern hemisphere is 0.02–0.06. Analogous data obtained by INP specialists for a range of objects in Western Kazakhstan lie within these limits. 238
238
Pu Bk/kg
60 238
Pu= 0.74239+240Pu + 2.1 R2 = 0.96
40
20
239+ 240
0 0
10
20
30
40
50
60
70
Pu Bk/kg
80
Figure 6. 238Pu/239 + 240Pu isotope ratio for silt samples from the BN-350 waste water disposal plant
The measured 238Pu/239 + 240Pu isotope ratio for silt samples collected from the waste water disposal plant was 0.7–0.9 (Figure 6) and diminished from 0.45 to 0.15 for Karakol Sor bottom sediments with distance from the reactor (see Figure 1). This was clear evidence of plutonium technogenic genesis in the
94
ENVIRONMENTAL IMPACT OF REACTORS IN KAZAKHSTAN
samples. Moreover, some samples of bottom sediment showed trace contents of 152 Eu (2–3 Bq/kg). The latter testifies that the contribution of the BN-350 reactor to the radioecological situation within the territories under investigation can be identified but its value is quite negligible. 4 Conclusions Data obtained from a radioecological survey of the buffer area and radiation control zone of the BN-350 reactor show the level of contamination of objects is no more than the background values due to global radionuclide fallout. Radioecological conditions in the territory around the WWR-K research reactor and in the Alatau settlement are normal. Levels of radioactive pollution of the ground do not exceed background values typical for the region from global radioactive fallout. The impact of the reactor on the environment is fixed and consists of increased concentrations of tritium in snow at several locations in comparison with background values. However, the peak measured values are several orders of magnitude below maximum permissible values. Thus, the fixed influence has not led to deterioration of the environment, nor does it represent a hazard to the health and life of the population. In general we conclude that operation of reactors in Kazakhstan has not impacted the environment of adjoining territories. References 1. 2.
3.
4. 5. 6.
Nazarenko, P.I., “BN-350 as a Firstling of Kazakhstan’s Power Engineering”, Report thesis. Kadyrzhanov, K.K., Lukashenko, S.N., “Radioactivity in Kazakhstan. Cases and Consequences”, Environmental Protection Against Radioactive Pollution, Kluwer Academic, The Netherlands, 2003, pp. 11–17. Nazarenko, P.I., Chakrov, P.V., Lukashenko, S.N., et al., “On the Assessment of Environmental Effect of BN-350 Reactor Facility over Its Operating Period”, ibid., pp. 69–74. “Complex Technical and Radiation Survey of the BN-350 Reactor Facility. Preliminary Technical and Radiation Survey Report”, Aktau, 2001. Kuznetsov, Y.V., Legin, V.K., Pospelov, Y.N., Simonyak, Z.K., “The Baltic Sea Bottom Sediments Analysis for 239, 240Pu Content”, Radiochemistry, N2, 1988. Hanson, W.S., Transuranium Elements in the Environment, 1985.
DESIGN AND MANUFACTURE OF FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
V. V. ROZHIKOV, А. А. ENIN, А. B. ALEXANDROV, AND А. А. TKACHEV∗ Novosibirsk Chemical Concentrate Plant, Novosibirsk, Russia Abstract: The most widely distributed fuel assembly designs with tubular fuel elements (WWR-M2, WWR-M5, IRT-2M, IRT-3M, MR, MIR, WWR-TS, IVV-2M) with more than 20% U-235 enrichment nuclear fuel used in research reactors of Russian design are described. The results of activities to convert foreign research reactors of Russian design to low-enrichment (less than 20% U-235) fuel that are performed in accordance with the Russian Program on Reduced Enrichment of Research Reactors are presented. Finally, recent accreditation of quality assurance (QA) and safety practices at the Novosibirsk Chemical Concentrate Plant (NCCP) to international standards is described. Keywords: research reactors, tubular fuel pins, U-Al alloy, UO2-Al, U-9% Mo, lowenrichment uranium (LEU) fuels, ISO-9001-2000 and ISO-14001-98 standards
1 Introduction The Novosibirsk Chemical Concentrate Plant (NCCP) manufactures nuclear fuel—both fuel pins and fuel assemblies—for power plants, commercial, and research reactors.1 Since 1973, NCCP has manufactured about 30,000 assemblies of 56 varieties for 30 research reactors of pool and channel-type design developed in Russia. The geography of its nuclear fuel supply covers research centers in Russia, former USSR countries, Europe, Asia, and the Middle East. The research centers and the types of supplied assemblies are shown in Figure 1.
______ ∗
To whom correspondence should be addressed: A. A. Tkachev, ОАО Novosibirsk Chemical Concentrate Plant, 94, Khmelnizkiy St., Novosibirsk 30110, Russia; e-mail:
[email protected] 95 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 95–105. © 2007 Springer.
96
FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
Figure 1. Geography of NCCP nuclear fuel supply
From the start of manufacture of fuel for research reactors until the early 1980s, NCCP produced fuel pins containing U-Al alloys. In the early 1980s the fuel composition was switched to UO2 dispersed in an aluminum matrix. Today research reactors of Russian design are operated in Russia (all types), Czech Republic (LWR-15), Hungary (VVR-SМ), Kazakhstan (VVR-K), Poland (МARIA), Libya (IRT-1), Ukraine (VVR-M), and Uzbekistan (VVR-SМ). 2 Designs of Fuel Pin and Assembly Fuel pins of tubular and fuel-rod types are used in assemblies manufactured by NCCP. The most popular design is where tubular fuel pins are coaxially installed inside each other (Figure 2). Figure 3 shows the structure of an MP-type pin: the center (b) contains the U-containing fuel kernel, while the outer and inner claddings (a, c) are made from an aluminum alloy. Figure 2. Tubular fuel pins in IRT-type assemblies
FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
97
а b c
Figure 3. Tubular structure of the MP-type fuel pin (MARIA reactor, Poland): (а) outer cladding, (b) fuel kernel, and (c) inner cladding
Fuel pins manufactured by NCCP have different shapes with the main cross sections shown in Figure 4. Changing the shape, as well as the uranium mass in a fuel pin allows fabrication to the exact fuel specifications of the customer.
Figure 4. Cross-sectional shapes of tubular fuel pins
98
FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
A quite large number of different research reactors operate with NCCP fuel pins and the specific features of these reactors determine a similar number of modifications in fuel-pin and assembly manufacture, as shown by the examples described below. 2.1 IRT-TYPE ASSEMBLY The IRT fuel assembly contains 3, 4, 6, or 8 coaxial square-section fuel pins. Figure 5 shows the IRT-4M fuel assembly, which is used in the LVR-15 and VVR-SM reactors in the Czech Republic and Uzbekistan, respectively. IRT-2M Assembly Russia, the Czech and Uzbekistan Republics, Libya, Bulgaria, Iraq IRT-3M Assembly Russia, the Czech and Uzbekistan Republics IRT-4M Assembly The Czech and Uzbekistan Republics
Figure 5. The IRT-4M fuel assembly
2.2 IVV-2M ASSEMBLY The IVV-2M assembly has five coaxial hexagonal fuel pins (Figure 6). The IVV-10 (a minor design variant) has six fuel pins. Both types of assembly are used in Russian research reactors only.
Figure 6. The IVV-2M fuel assembly
FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
99
2.3 VVR-Z FUEL ASSEMBLY The VVR-Z fuel assembly has four coaxial hexagonal fuel pins and a central tubular fuel pin that is circular in shape (Figure 7). The VVR-Z assembly is used in the VVR-K reactor in Kazakhstan and several Russian research reactors.
Figure 7. The VVR-Z fuel assembly
2.4 MR FUEL ASSEMBLY The MR fuel assembly has four, five, or six coaxial circular fuel pins (Figure 8). The MR fuel assembly is used in the MARIA reactor in Poland and in several Russian research reactors.
Figure 8. The MR fuel assembly
2.5 FUEL ASSEMBLY VARIANTS Figure 9 shows the cross-sections of minor actinides (MA) fuel assemblies with six or four fuel pins. It shows how an empty inner cavity can be introduced in which controls, reactor safety controls, or different experimental devices can be installed.
100
FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
Figure 9. Modification of an MR assembly to produce a central cavity: (a) original six-tube assembly and (b) four-tube assembly with cavity
Figure 10 shows how an IRT-2M fuel assembly can be modified by the customer to produce an empty inner cavity by taking out a fuel pin.
Figure 10. IRT-2M assembly modified for a central experiment
FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
101
As well as fuel assemblies made of a number of fuel pins used as single units, some assembly designs allow a keying together of three assemblies to produce a larger fuel unit which speeds loading fresh fuel and unloading spent fuel from the reactor core. Both the VVR-M2 and VVR-M5 assemblies can be assembled into larger units this way (a single and a triple VVR-M2 fuel assembly are shown in Figure 11).
Figure 11. Single and triple VVR-M2 fuel assembly
2.6 TARGETS FOR ISOTOPE PRODUCTION The various mastered and applied technologies for fuel pin manufacture have allowed NCCP develop targets for production of radioisotopes (see Figure 12). The targets can be placed in the central cavity of modified fuel assemblies. 3 Current and Future Fuel Manufacture The major characteristics of the most popular fuel assembly designs are given in Table 1. It shows that NCCP currently manufactures research reactor fuel assemblies with U235 enrichment of 90%, 36%, and 19.7%. Mass concentration of uranium in the fuel column varies from 1.25 to 3 g/cm3, with a corresponding bulk concentration from 14 to 34.5 wt%. The future trend will be to reduce fuel enrichment in all assemblies to less than 20% U-235, in accordance with aims of the Russian Program on Decrease of Enriched Fuel,2 which began in 1993.
FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
102
Figure 12. Targets for radioisotope production Left: for С14, Co60, Ir192, Tm170, Tl204, Cs137, Sb204, Se75. Right: for Мо99
TABLE 1. Parameters of research reactor fuel assemblies manufactured by NCCP Assembly type
Number of fuel pins
U235 mass in assembly (g)
% U235
45 38 VVR-M2
3
36
42 50
19.7
VVR-M5
54
6
66
IRT-2М
4(3)
230(198)
IRT-3М
8(6, 4)
IRT-4М
8(6, 4)
МR
6(5, 4)
Fuel thickness (mm)
500 600
65 5
Fuel length (mm) 600
2.5
500 600
36 90 36
352 (309, 235)
36
300 (264, 201)
90
300 (264, 201)
19,7
550 (504, 439)
36
470 (432, 377)
90
1.3 500 600
2
600
1.4
600
1.6
1,000
2
FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
103
МIR
4
377
90
1,000
2
VVR-Z
5(3)
109(83)
36
600
2.3
IVV-2М
5
225
90
500
1.4
IVV-10
6
287
90
500
1.4(1.7)
TVR-С
1
8
36
100
4
IR-100
7
39
36
500
2.5
3.1 REDUCED-ENRICHMENT FUEL ASSEMBLIES Starting in 1998, and in cooperation with the Moscow Electrotechnics Research and Design Institute and with the Institute of Physics and Power Engineering (IPPE), Obninsk, NCCP has helped develop fuel assemblies with high density metal-ceramic fuel based on U-9% Mo alloy with a 19.75 U-235 enrichment. As mentioned before, this development has been carried out within the Russian Program on Decrease of Enriched Fuel; it aims at the conversion of research reactors to fuel with U-235 enrichment not higher than 20%, as well as a further improvement in fuel performance and in the provision of high neutron fluxes in reactors with fuel of low enrichment. High density fuel is going to be used for fuel pins of the traditional tubular and fuel-rod types. Figure 13 shows a multipurpose fuel-rod type fuel pin with the new fuel composition of U-9% Mo. Figure 14 shows how a multipurpose fuel-rod type fuel pin could be used to replace standard tubular fuel pins in the IRT-3M, MR, and VVR-M2 fuel assemblies.
Figure 13. Multipurpose fuel-rod type fuel pin based on U-9%Mo: (а) crosssection view of the multipurpose fuel pin and (b) general view
FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
104
Figure 14. Potential combination patterns for VVR-M2, IRT-3M and MR fuel assemblies on the basis of the same multipurpose fuel-rod type fuel pin (data from: VNIINM, Moscow)
By the end of 2006 the schedule is to manufacture and load for in-reactor testing four full-fledged assemblies of the IRT type with both tubular and fuelrod type fuel pins that contain U-9%Mo fuel. These fuel assemblies will be delivered for testing to reactors in the Czech Republic, Uzbekistan, and Libya. 4 Quality Assurance at NCCP In improving the technology and design of fuel assemblies, NCCP has focused recently on the quality and environmental safety of their operations. The quality assurance (QA) management at NCCP has been certified at the national and international levels (RF Gosstandart, TUV CERT) and meets the requirements3 of ISO-9001-2000. To develop a positive image to the public and to attract potential customers, NCCP has introduced and licensed its environment management system with TUV CERT, confirming that its system meets ISO-14001-98. QA management and environmental-friendly management assure the customer of the quality and safety of a product. NCCP provides customers an opportunity to view documentation for adherence to technical specifications at all stages of product manufacture and acceptance. References 1.
Enin, A.A., An Overview of Russian Research Reactor Types, their Fabrication and Quality Control, IAEA Regional Workshop on Characterization, Management and Storage of Spent Fuel from Research Reactors, Swierk, Poland, May 8–12, 2000.
FUEL ASSEMBLIES FOR RUSSIAN RESEARCH REACTORS
2.
3.
105
Aden, V.G., et al., “Russian Programme of Reduced Fuel Enrichment in Research Reactors”, Proceedings of the 16th International Conference on Reduced Enrichment for Research and Test Reactors, Orai, Japan, October 4–7, 1993, JAERI-M 94-042. International Organization for Standards: (http://www.iso.org/iso/en/ISOOnline/).
STRATEGY FOR HANDLING SPENT BN-350 CESIUM TRAPS IN THE REPUBLIC OF KAZAKHSTAN
O. G. ROMANENKO,1* I. L. TAZHIBAEVA,1 D. WELLS,2 A. HERRICK,2 J. A. MICHELBACHER,3 C. KNIGHT,3 V. I. POLYAKOV,4 U. PRIVALOV,4 M. SOBOLEV,4 U. SHTYNDA,4 A. GAINULLINA,4 I. L. YAKOVLEV,5 U. P. SHIROBOKOV,5 A. I. IVANOV,5 AND G. P. PUGACHEV5 1 Nuclear Technology Safety Center (RK) 2 NUKEM Limited (UK) 3 Idaho National Laboratory (USA) 4 Research Institute of Atomic Reactors (RF) 5 Mangyshlak Atomic Energy Complex-Kazatomprom (RK) Abstract: A major task during decommissioning of fast reactors is draining and processing the primary sodium, which—by end of operations—is usually significantly contaminated by fission products and fuel released from failures. The most hazardous fission product is Cs-137 because of its large contribution to total activity of the primary sodium and its 30-year half-life. In fact, direct processing of sodium without removal of Cs-137 is problematic. Cesium was removed from primary sodium in the BN-350 fast reactor in two ways: in the 1980s by means of graphite traps in special in-core assemblies, and, before draining in 2003, by means of traps connected to the primary circuit, which contained reticulated vitreous carbon (RVC) of very high surface area. The use of both types of very efficient trap concentrated radioactive cesium from 10 to 15 years of reactor operation in relatively small devices, and gave safe working conditions around the primary circuit. However, the traps are highly active solid waste, they are chemically active and prone to fire, and must be transferred in a secure state suitable for long-term storage. This paper describes the design and operation of BN-350 cesium traps and discusses the means being considered for their safe handling and disposal in Kazakhstan. Keywords: BN-350 fast reactor, primary sodium purification, cesium contamination, trap accumulators, reticulated vitreous carbon (RVC), small-size adsorber for removal of radioactivity (MAVR)
_________ ∗ To whom correspondence should be addressed: O.G. Romanenko, Nuclear Technology Safety Center, Liza Chaikina 4, Almaty 050020, Kazakhstan; e-mail:
[email protected] 107 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 107–142. © 2007 Springer.
108
HANDLING SPENT CESIUM TRAPS FROM BN-350
1 Introduction Over the past few decades, effective methods of cleaning radionuclides from sodium circuits, cesium1–4 in particular, have been developed in the sodium technology of fast reactors. The technology of cleaning cesium from sodium was realized in the form of cesium traps of two types: stationary devices connected to the circuit that is to be cleaned, and devices installed in the core of reactor at shutdown. The efficiency of removal of cesium by traps was achieved by using carbon-graphite materials under conditions of specially organized mode of trapping. During the different periods of BN-350 reactor operation both types of cesium traps were used: in-core devices designed at the Research Institute of Atomic Reactors (NIIAR), Russian Federation (RF)—the MAVR series using low-ash granulated graphite; and stationary devices designed at the Institute of Atomic Energy of the National Nuclear Center (IAE NNC), Republic of Kazakhstan (RK); Argonne National Laboratory (ANL), USA; and Mangyslak Atomic Energy Complex (MAEC)-Kazatom-prom, RK, using reticulated vitreous carbon (RVC) material. Cleaning sodium by stationary traps was carried out before draining in order to reduce the radiation load on personnel involved later in processing primary sodium. At the same time, cleaning cesium radionuclides from primary sodium is one of the stages of BN-350 reactor decommissioning. Purification of sodium of the reactor primary circuit has resulted in reduction of general radioactive background in the primary circuit premises and from equipment that further will facilitate its dismantling. Application of high-performance cesium carbon-graphite sorbents has allowed concentration of radioactive cesium, accumulated in the reactor circuits for the decades of their operation, in relatively small volumes of traps that practically provided radiation-safe conditions for personnel working with coolant and equipment of the primary circuit. Four spent cesium traps of the MAVR series designed by NIIAR, and seven spent cesium traps developed jointly by IAE NNC, ANL, and MAEC-Kazatomprom are stored at BN-350. Specific activity of Cs-137 in traps with maximum activity for NIIAR traps is equal to 2.6 × 1013 Bq/kg, and for stationary traps – 8.9 × 1013 Bq/kg. By specific activity of cesium the traps are comparable with the activity of spent fuel elements of the reactor facilities cooled within 10 years –– approximately 1014 Bq/kg.5 Cesium traps, as products unfit for further use, represent solid radioactive wastes (SRWs). According to IAEA safety standards, they can be classified as wastes formed at decommissioning of an NPP.6 According to standard practice, an SRW handling system includes their collection, sorting, packing, temporary storage, conditioning (concentrating, hardening, pressing, burning), transporting, long-term storage, and/or disposal. Self-ignitable and explosive radioactive
HANDLING SPENT CESIUM TRAPS FROM BN-350
109
wastes (RAWs) should be transferred into a safe state. Spent cesium traps contain some amount of chemically active sodium, which is able to react violently with water, as well as potassium and cesium––possibly in the form of potassium–carbon, cesium–carbon layer compounds. The presence of these alkali metals in the traps and a lack of a corresponding legislative and technical basis for handling highly active SRW in the RK results in a series of problems concerned with their handling. This is a problem of transfer of cesium traps into safe state, as well as a problem of their disposal, since they are classified as high-level RAWs by the level of specific activity. The objective of this paper is to describe the strategy for handling spent cesium traps from BN-350, which was jointly developed by specialists from Kazakhstan, UK, USA, and Russia, and finally accepted in the RK. 2 Characterization of BN-350 Cesium Traps 2.1 TRAP ACCUMULATORS (IAE, ANL, AND MAEC-KAZATOMPROM) The BN-350 primary sodium was purified from cesium radionuclides before its drainage during the year 2002–2003 using stationary trap accumulators (TAs). These TAs were developed by IAE NNC in collaboration with ANL-West, now the Idaho National Laboratory (INL), and MAEC-Kazatomprom. Equipment was fabricated by the Joint Stock Company “Belkamit” (RK). 2.1.1 TA design characteristics A detailed description of the cesium trapping system, as well as the results of its operation were reported in a recent ANS paper.7 Here, we repeat parts of that paper which are relevant to this discussion. An elevation view of the cesium trap system is shown in Figure 1. Each component of the system is contained inside radiological shielding for personnel protection during operation and for storage of the traps. Each cesium trap is contained inside an integral biological shield allowing the handling of the whole assembly as a replaceable module (3). After being used and cut out of the primary sodium circuit, the entire trap and shielding module was transported to an approved interim storage site. All components of the trap and shielding module, including instrumentation inside the shielding, were designed to remain operable and intact for up to 50 years of interim storage.
110
HANDLING SPENT CESIUM TRAPS FROM BN-350
Figure 1. Elevation view of the BN-350 cesium trap system 1: economizer; 2: air-sodium heat exchanger; 3: cesium trap in biological shielding; 4: heat exchanger block biological shielding; 5: piping biological shielding; 6: base plate; 7: drip pan; 8: support frame; 9: dose rate detector
The cesium trap, shown in Figure 2, is a pressure vessel designed to remove Cs-137 from the primary sodium and to store it. Sodium enters the trap through a central inlet pipe (7) at a pressure of ∼0.4 MPa. After reversing and decreasing flow, sodium passes through the RVC adsorbent bed (9). The RVC is installed as stacked discs inside the stainless steel shell (1) and is kept in place by perforated plates (8) on the top and bottom of the RVC bed. After passing through the RVC the cleaned sodium passes through a stainless steel mesh particulate filter (10) that traps any RVC particle, which may have been entrained in the sodium stream. From the filter plenum the sodium leaves the trap and flows back through the economizer before returning to the reactor primary system. Each trap is wrapped with electrical heaters and thermocouples (not shown) to allow the contained sodium to be melted, if necessary after a system shut down and cool down. Thermal insulation (15) is installed to support the electrical leads in the trap shielding.
HANDLING SPENT CESIUM TRAPS FROM BN-350
111
Figure 2. BN-350 cesium trap 1: shell; 2: bottom plate; 3: top plate; 4: inlet pipe; 5: outlet pipe; 6: inlet nozzle; 7: central inlet pipe; 8: perforated plate; 9: RVC; 10: mesh filter; 11: electrical support frame; 12: electrical conduit support; 13: label; 14: support leg; 15: thermal insulation
The cesium trap biological shielding, shown as item 3 in Figure 1, was designed to protect personnel from ionizing radiation, localize possible sodium and cesium leaks, and to assist with passive fire extinguishing of a sodium leak. The carbon steel biological shielding thickness was designed to reduce the ionizing radiation levels to less than 11 nSv/s at 1 m as required by BN-350 and regulatory documents of RK. The first three cesium traps were expected to accumulate up to 111,000 GBq (3,000 Ci) of cesium in each trap but the remaining four traps were only expected to accumulate up to 18,500 GBq (500 Ci). Consequently, the shielding thickness for the traps were sized to contain a 111,000 GBq (3,000 Ci) source and the remaining shielding thicknesses were decreased to contain only 18,500 GBq (500 Ci) sources. This decreased the total material required and the total shielding weights. Weight of biological shielding is 5,625 kg for the first three traps and 4,447 kg for the next four traps.
HANDLING SPENT CESIUM TRAPS FROM BN-350
112
2.1.2 TA trapping results Seven cesium traps were fabricated and used to remove cesium from BN-350 primary sodium. Each trap remained in operation until its RVC became saturated with cesium, after which it was removed and replaced by a clean trap. Table 1 summarizes trap information for each of the seven operating runs, including the RVC masses, cesium activity changes in the primary system sodium, cesium accumulation on each trap, and the cumulative amount of cesium removed as a percentage of the estimated initial total cesium in the system. TABLE 1. Cesium trapping results Trap no.
RVC mass (kg)
Activity before trapping (MBq/kg)(1,2)
Activity after trapping (MBq/kg)(1)
Cesium on trap (GBq)/(Ci)
1(4) 622 2(4) 621 3(4) 623 4(5) 624 5(5) 627 6(5) 625 7(5,6) 626 Total
1.58 1.20 2.25 2.25 2.95 3.04 3.07 16.34
267 88.1 25.0 10.4 5.7 10.2 6.6
87.3 16.9 9.7 2.9 6.9 4.8 0.3
140,600/3,800 56,980/1,540 18,500/500 25,900/700 8,584/232 3,922/106 851/23 255,337/6,901
Run no.
Cumulative Cs-137 removed from primary system (%)(3) 54.3 76.3 83.4 93.4 96.7 98.3 98.6 98.6
Notes: (1) Activity values were obtained with the “CENA” cesium monitoring system. (2) For runs 2–7, activity before trapping was higher than the preceding activity after trapping because the cesium on internal surfaces returned to solution in the primary sodium during the time that one trap was removed and the next trap was installed. (3) Cs-137 inventory was estimated as 233,100 GBq (6,300 Ci) in the primary circuit and 25,900 GBq (700 Ci) in the storage tanks for a total of 259,000 GBq (7,000 Ci). (4) Value includes only the amount from the primary circuit. (5) Value includes amounts from both the primary circuit and the sodium storage tanks. (6) The activity before trapping was obtained from a sample of the primary sodium.
2.2 SMALL-SIZE MAVR ADSORBERS DESIGNED BY NIIAR, RUSSIA 2.2.1 BN-350 operating history with MAVRs MAVR (small-size adsorber for removal of radioactivity) devices were used at the BN-350 reactor from 1978 to 1989 for reducing the activity of cesium radionuclides in primary sodium. Spherical granules of reactor graphite were
HANDLING SPENT CESIUM TRAPS FROM BN-350
113
used as the sorbent for cesium and these were placed in a device whose overall and bore dimensions corresponded with the dimensions of a standard BN-350 fuel assembly. The existing fuel assembly reloading and transportation system was used for loading and unloading MAVRs. Spent traps were drained and then unloaded into a leakproof shielding package, filled with inert gas, and placed into dry storage. The weight of a MAVR including the shielding package was about 9,000 kg. Cleaning cesium radionuclides from primary sodium using MAVRs was carried out four times at the BN-350 reactor: November 1979, April–May 1980, June–July 1984, and May 1989. 2.2.2 Design of small-size adsorbers Adsorber “MAVR” is designed in the form of a fuel assembly (sorbent is placed in cartridges inside a package-simulator); the assembly can be placed in any core location. Sodium flows in the axial direction through four cartridges containing sorbent. Sodium is supplied (and drained) to and from the sorbent cartridges by a system of independent tubes. The system of cartridges is fitted with a chip filter of stainless steel, through which all sodium runs from the cartridges. The design of a MAVR is sorbent cartridge shown in Figure 3. 2.2.3 Results of trapping with MAVRs Table 2 presents the basic characteristics of four MAVRs used in BN-350 over 1979–1989 (note that the MARV-3 unit was never used). The “reduced” value of activity of Cs-137 in the table corresponds to the sum of activities of Cs-137 and Cs-134 at the moment of unloading from the reactor. These results are to be used in the calculation of the activity of cesium radionuclides and gammaradiation dose rate for the time of handling the MAVR adsorbers. Parameters of primary coolant on the unloading of MAVRs from the reactor core are the following: • • •
Temperature of sodium of the primary circuit Specific activity of cesium-137 of the primary coolant being purified Ratios of specific activities of cesium-134 to cesium-137.
HANDLING SPENT CESIUM TRAPS FROM BN-350
114
Figure 3. Design of MARV sorbent cartridge used in BN-350 TABLE 2. Characteristics of BN-350 MARVs MAVR no.
Date (mm/yy)
Mass of graphite (kg)
1
11.79
∼10
2
05.80
∼10
5
06.84
10.01
4
05.89
10.04
Total:
∼40
Cumulative activity Ci
TBq
7,000 ± 1,000 2,000 ± 500 2,000 ± 750 2,500 ± 850 13,500
259 ± 15% 74 ± 25% 74 ± 35% 92,5 ± 35% 500
MARV parameters Na temp.
ACs-137
°С
MBq/kg //Ci/kg 107//2.9 × 10–3 207//5.6 × 10–3 59//1.6 × 10–3 63//1.7 × 10–3
225 190 180 185
ACs-134 ACs-137 % 19 18 6.65 11.7
HANDLING SPENT CESIUM TRAPS FROM BN-350
115
2.3 ESTIMATING QUANTITY OF LONG-LIVED RADIONUCLIDES IN TRAPS The BN-350 primary sodium is contaminated with fission products and alphaemitting nuclides because of fuel element failures. Purification of sodium in the period 1979–1989 using MAVRs allowed reduction in the primary circuit contamination by cesium radionuclides and possibly by transuranic elements. However, from 1991, before cleaning cesium from sodium using traps with RVC adsorber, fuel failures up to the degree of fuel erosion took place and fuel could again enter into the primary sodium. The presence of alpha-emitting nuclides was never detected in the primary coolant throughout the whole period of BN-350 power operation. Therefore, in estimating radionuclide composition and its activity in primary sodium, analysis of the previous results concerning calculations of fuel and fission fragments released from fuel pins into primary sodium of BN-350 reactor was performed. The results of conservative estimation of the quantity and composition of fuel released in primary sodium from fuel failures showed that plutonium isotopes made the main contribution to alpha activity of fuel release to primary sodium— the alpha activity of uranium isotopes is lower by several orders of magnitude. Approximate estimation of the quantity of alpha nuclides accumulated in each trap could be made, if analysis of the content of fuel and transuranic elements in the primary sodium for periods before and after purification of coolant from cesium by each MAVR and trap accumulator (TA) were known, but, unfortunately, such data are not available. Based on the analysis performed there was made an assumption that 1.06 × 106 MBq of fuel entered in primary circuit in total for the whole period of operation of the reactor BN-350. However, it would be incorrect to consider that all this fuel could be accumulated in the traps, since a major quantity of alpha-active nuclides could not reach traps being adsorbed on the internal surfaces of equipment of the primary circuit, but not in sodium solution. Indeed, literature analysis—such as the work of Feuerstein1—suggests that the concentration of alpha-active nuclides in the wall layer of sodium and on the surface of components removed from fast reactors is higher by 100–1,000 times than in the coolant. It was also experimentally demonstrated that specific alpha activity of sodium is decreased after reactor shutdown, and this fact is to be interpreted as precipitation of fuel microparticles on the walls of the primary circuit. According to results from radiometric analysis, alpha-active nuclides occur in sodium mainly in the form of fine particles of fuel and complex oxygen-containing compounds of variable composition involving uranates and plutonates. According to known experimental data, the distribution coefficient (ratio of surface activity of nuclide on steel to bulk activity in sodium) makes up 550 ± 50 cm for Pu-239 at a temperature of 350°С. It means that 90–95% of the Pu-239 is in precipitates.
HANDLING SPENT CESIUM TRAPS FROM BN-350
116
It may be assumed that 90% of the total quantity of fuel (1.06 × 106 MBq), which entered into the primary sodium during the whole of BN-350 operation is in the form of precipitates on internal surfaces of equipment in the primary circuit. Furthermore, the 10% of the fuel contained in the sodium is assumed to reach the cesium traps and be distributed between four traps of MAVR type and seven TAs. In addition, the value of accumulation of plutonium isotopes in the traps is assumed to be proportional to the accumulation of Cs-137 radionuclides in them. We will not take into account the fact that a considerable portion of fuel could settle in the cold traps during the regular purifications of sodium of the primary circuit to remove oxides before the commissioning of cesium traps. For MAVR traps and TAs with RVC sorbent, Table 3 gives the experimental and estimated values of specific activities of Cs-137 and Pu radionuclides accumulated in sorbent. TABLE 3. Estimates of specific activities of Cs-137 and Pu isotopes accumulated in BN-350 traps Trap no.
MAVR-1 MAVR-2 MAVR-5 MAVR-4 TA-622 TA-621 TA-623 TA-624 TA-627 TA-625 TA-626
Specific activity of Cs-137 in sorbent Ci/kg MBq/kg 2.6 × 107 700 7.4 × 106 200 7.4 × 106 200 9.3 × 106 250 8.9 × 107 2,410 4.8 × 107 1,280 8.2 × 106 222 1.2 × 107 312 2.9 × 106 78.3 1.3 × 106 34.7 2.8 × 105 7.5
Specific activity of Pu isotopes in sorbent Ci/kg MBq/kg 3.6 × 103 0.1 1.03 × 103 0.03 1.03 × 103 0.03 1.3 × 103 0.04 1.3 × 104 0.34 6.6 × 103 0.18 1.1 × 103 0.03 1.6 × 103 0.04 4.0 × 102 0.01 1.8 × 102 0.005 3.9 × 101 0.001
2.4 ESTIMATED CONTENT OF CHEMICAL SUBSTANCES IN CESIUM TRAPS 2.4.1 Graphite According to certificate data the quantities of RVC in the TAs were: TA-621: 1.2 kg, TA-622: 1.58 kg, TA-623: 2.25 kg, TA-624: 2.25 kg, T-625: 3.04 kg, TA-626: 3.07 kg, and TA-627: 2.95 kg. Total RVC sorbent in TAs: 16.36 kg.
HANDLING SPENT CESIUM TRAPS FROM BN-350
117
Spherical granules of reactor low-ash graphite were used as cesium sorbent in the adsorber MAVRs. Each adsorber contained ∼10 kg of graphite; thus, the total quantity of graphite in the four MAVRs was ∼40 kg. 2.4.2 Sodium The TAs were filled with 16.6–17.4 kg of sodium, depending on the quantity of sorbent in the trap. The MAVRs were drained of sodium after operation. Experiments made at NIIAR on drainage from an MAVR mockup8 showed that nondrainable sodium remaining on the surface of the graphite granules was about 0.5 kg or 550 cm3, while about 1,300 g (1,400 cm3) of sodium remained in the pores of the graphite granules. 2.4.3 Cesium The mass (m) of radionuclide (g), of A curies in activity, is given by: m = 8.9 × 10−14 × A0 × T1⁄2 × A, where A0 is atomic weight, and T1⁄2 is the radionuclide half-life (sec). For example, the mass (m) of Cs-137 (g) with activity 3,800 Ci in TA-622 is given by: m = 8.9 × 10−14 × 137 × 30.3 × 364 × 24 × 3,600 × 3,800 = 44 g, or 0.028 g of Cs per gram of carbon. The amount of stable cesium can be estimated from this quantity and is about 90 g. The mass (m) of Cs-137 (in g) with activity 7,000 Ci in the MAVR-1 adsorber is given by: m = 8.9 × 10−14 × 137 × 30.3 × 364 × 24 × 3,600 × 7,000 = 81 g, or 0.008 g of cesium per gram of carbon. Again, the quantity of stable cesium nuclide can be estimated from the same quantity and is about 160 g. 2.4.4 Potassium Potassium is a natural impurity of sodium, which derives from the raw material used in sodium production. The potassium content in sodium supplied from the plant is in the range 500–700 ppm. There are no sources of contamination of sodium by potassium in the reactor circuits during power operation.
HANDLING SPENT CESIUM TRAPS FROM BN-350
118
The content of potassium in the primary coolant is then given by: m=M×c where m is mass of potassium in coolant (kg), M is mass of coolant (kg), and c is the fractional concentration of potassium in coolant. m = 550,000 kg × 0.0005 = 275 kg. Potassium, as well as cesium may accumulate in the traps. It is known that potassium forms irregular solid solutions with carbon, where potassium atoms are located in the interlayer interstices of the carbon lattice. At temperatures above 200°С, graphite absorbs potassium in amounts up to 40 wt%.9 If the total content of carbon in the TAs is 16.36 kg, then 16.36 × 0.4 = 6.5 kg of potassium may accumulate in the traps. This value represents 2.4% of the total quantity of potassium in the circuit. Assuming that the sorption coefficient does not depend significantly on the quantity of potassium in sodium, the content of potassium in the TAs will be: 0.48 kg (TA-621), 0.632 kg (TA-622), 0.9 kg (TA-623 and TA-624), 1.216 kg (TA-625), 1.228 kg (TA-626), and 1.188 kg (TA-627). One adsorber of the MAVR type may absorb 10 × 0.4 kg = 4 kg of potassium. These authors do not have literature data characterizing the sorption of potassium by graphite in a sodium medium. However, detailed study of layer compounds being produced in the vapor phase10 demonstrates that the stability of layer compounds of graphite with alkali metals (composition С8М) is increased going from K to Rb and to Cs. For example, the following data are presented regarding heat of reaction with alkali metals within interval of temperatures 66–95°С: –∆Н = 7.8 kcal/g-atom for С8K, 11.1 kcal/g-atom for С8Rb and 15.3 kcal/g-atom for С8Cs. Proceeding from this reaction heat value for graphite interaction with alkali metals, one can assess that the coefficient of sorption for potassium in graphite should be lower by two to three orders in comparison with the coefficient of cesium sorption in graphite for reactions in sodium medium. Then, if we assume that the coefficient of sorption for potassium in graphite is lower by two orders compared to the cesium sorption coefficient in sorbent and concentration of potassium in sodium was maintained at the level 500 ppm, it is to be expected that up to 2,200 g of potassium could be accumulated in TA-622 for 44 g of Cs137. As for a MAVR with an activity of 7,000 Ci of Cs137, the accumulated quantity of potassium could be up to 4,050 g. Thus, by estimating the quantity of potassium accumulated in the traps by two independent methods, we obtain comparable values, hence they may be accepted as reasonable estimates.
HANDLING SPENT CESIUM TRAPS FROM BN-350
119
One should note that usually the amount of potassium that can be accumulated in graphite is taken into account in the design of cesium traps to prevent destruction of the graphite by sodium during operation, because its destructive influence on the material could be noticeable. We believe that for our case, an amount of potassium accumulated in cesium traps does not influence significantly on the process of choosing technologies for conditioning cesium traps. 3 Classification of Spent Cesium Traps as Radioactive Waste 3.1 REGULATORY LEGISLATION IN THE RK At present, in the RK, there is no concept of handling high level RAW, although a strategy for handling low and medium-level RAWs is under consideration. In this section, we consider some basic provisions of two normative documents that were developed for regulation of activities on handling low and medium-level RAWs in the RK and that we will refer to in the future, in the process of selection of optimal variants for handling cesium traps. 3.1.1 “Safety requirements for collecting, processing and storing radioactive wastes” (SRCPS-2003) The document “safety requirements for collection, processing and storing of radioactive wastes” (SRCPS-2003) was put into force in Kazakhstan on August 1, 2003, to ensure safety when handling RAWs. This document sets safety assurance requirements for the systems of collection, processing, conditioning, and storing of low and medium-level RAWs of the atomic energy utilizing objects, where these wastes are formed and/or processed, including specialized RAW processing plants and RAW disposal plants. SRCPS-2003 sets requirements for conditioned package of RAW, namely: 1. RAW package must prevent unacceptable spread of radionuclides into the environment and must not contain: • Strong oxidants and chemically unstable substances • Poisonous, pathogenic and infectious substances • Biologically active substances • Highly inflammable, explosive, or fire-risk substances • Substances that can detonate or decompose with explosion • Substances that enter in exothermic reaction with water accompanied with explosion • Substances that contain or are capable of generating toxic gases, vapors or sublimates
120
HANDLING SPENT CESIUM TRAPS FROM BN-350
2. Content of free liquid in RAW package must not exceed 0.5% of the package volume. The classification of liquid and SRWs is given in Table 4. TABLE 4. Classification of liquid and solid radioactive wastes by specific activity Waste category
Low activity Medium activity High activity
Radionuclide specific activity (kBq/kg) Beta-emitting Alpha-emitting Transuranics (except transuranics) <103 <102 <101 3 7 2 6 10 –10 101–105 10 –10 7 6 >10 >105 >10
3.1.2 “Safety rules for near-surface disposal of radioactive wastes” (SRNDRW) “Safety rules for near-surface disposal of radioactive wastes” (SRNDRW) were developed in the framework of elaborating a strategy for handling RAWs in the RK. The rules set requirements for SRW near-surface disposal objects and are aimed at excluding unacceptable risk of causing damage to health of the population and environment now and in the future. These rules are already developed, but not in force and are at the approval stage between ministries of the RK. According to SRNDRW, the basic principles of assurance for long-term safety of near-surface RAW disposals are as follows: •
• •
Principle of application of several levels of shielding, including: form and stability of wastes, reliability of their packing, natural barriers of the RAW near-surface disposal site (RAWNDS), engineering barriers (structures) of RAWNDS, and RAWNDS operating procedure and methods of control. Not exceeding the dose limit quota established by the regulatory documents of RK for population from RAW disposal. The burden of long-term monitoring and control over safety of disposal on future generations shall not be excessive.
SRNDRW formulates criteria for acceptance of RAWs for near-surface disposal. RAWs acceptable for near-surface disposal shall have the following properties: Aggregative state: Wastes accepted for disposal shall be only in the solid form. If wastes initially were liquid, they shall be conditioned to solid low leachability state (less than 10–3 g/cm2 × days by Cs-137 and Sr-90). Density: Wastes shall be compacted in order to minimize their volume and to reduce the surface area over which leaching can occur.
HANDLING SPENT CESIUM TRAPS FROM BN-350
121
Radionuclide composition and waste activity: Radionuclide composition of wastes, specific and total activity of radionuclides in the packages (maximum and average values for RAWNDS) shall meet requirements established by the design of the disposal site. Limits of specific activity of radionuclides are established on the basis of estimations of radiological effect of the disposal site on the whole population and taking into account possible scenarios of radiation impact on individuals from the population during operation of the site, its decommissioning, and after its closure. In the estimations it is necessary to take into account that accidental or deliberate entrance of persons from population to the disposal site during established period of control may happen. Specific activity shall be averaged over the whole volume of the waste package. At the same time, specific activity of the package for long-lived alpha sources shall be limited to 4,000 kBq/kg, and averaged activity for all waste packages shall not exceed 400 kBq/kg. Thermal stability: Wastes shall be stable against degradation conditioned by residual heat after disposal and effect of external heat sources. Package design: Design of all packages by weight, volume, form, and size shall correspond to design of the storage, transportation conditions and be easyto-operate. Mechanical strength: Wastes and containers shall have sufficient mechanical strength to hold form and withstand rough usage. Compression strength limit for cemented wastes shall not be less than 50 atm, including after exposure to radiation dose 106 Gy. A thermal destruction test shall consist of 30 cycles of heating to +60°С and cooling to –40°С. Packages shall withstand vertical loads corresponding to the design of RAWNDS. Stability: Stability of wastes may be provided by form of wastes or by package. Medium level RAWs shall maintain stability during the period not less than 300 years. Toxicity: Content of chemically toxic, poisonous, pathogenic and infectious substances in wastes shall be determined with adequate accuracy and limited as much as it is possible. Low dispersion ability: In order to avoid surface contamination at handling with packages, they shall not contain dispersing wastes (dust, powder), and light contaminated materials (paper, polyethylene film) shall be briquetted. Surface radioactive contamination: Level of radioactive contamination on the external surface of package shall not exceed dose limits established by the regulatory documents of RK for service personnel. Chemical stability: Wastes shall not contain strong oxidants, chemically and corrosion-active and unstable substances. Wastes shall not decompose and
122
HANDLING SPENT CESIUM TRAPS FROM BN-350
release gases and fumes. In case of presence of organic or self-ignitable materials in wastes, they shall be processed into explosion- and flameproof form. Amount of free liquid shall not exceed 0.5% of the waste package volume. Liquid shall have рН from 4 to 11. For liquid wastes solidified in cement, free liquid shall have minimal рН = 9. Chemical compatibility: Content of stable complexing substances used, for example, at decontamination, as well as possible chemical transformations in wastes that may increase their migration ability in future, shall be taken into account at the stage of preparation of wastes for disposal. Medium-active wastes containing long-lived radionuclides (with half-lives of more than 30 years), except transuranics, with total specific alpha activity more than 104 kBq/kg, and/or transuranic radionuclides with total specific activity more than 102 kBq/kg, shall be disposed in such a way that their upper level is located at a depth of not less than 5 m from the daylight surface, or shall be provided with shielding barriers that have a service life not less than 500 years. SRNDRW classifies RAWs in the same way as SRCPS-2003 (see Table 4). 3.2 CATEGORIZATION OF BN-350 CESIUM TRAPS Cesium traps designed by IAE NNC RK, ANL, and MAEC Kazatomprom were developed for primary sodium purification that is a stage in the decommissioning of BN-350. In accordance with IAEA safety standards number 69,6 all spent cesium traps can be rated among the wastes being formed at decommissioning of NPP, which may require “special processing and conditioning”. Standards recommend such wastes to be classified “in accordance with the content of radionuclides in them, physical form and sizes, as well as with nature of materials”. By the content of radionuclides (β-emitting Cs-137 radionuclide) in cesium traps they are wastes of high level of activity, therefore spent cesium traps can be classified as high-active wastes, since activity of cesium in trap is “comparable with activity of spent fuel” and “formed by highly radioactive substance containing fission products”. By physical form they are rated among solid wastes and by nature of materials––among chemically and fire risky materials, since they contain sodium and carbon. IAEA safety standards #69 set technical criteria on assurance of technical basis of safety principles for handling with RAWs to be disposed underground. There is a reason to believe that fulfillment of these criteria will also provide safety at handling with spent cesium traps after their conditioning. Technical basis for assurance of safety at placing wastes in storage (disposal) is the following: firstly, criterion of acceptability of wastes by the content of supposed or actual radionuclides on basis of which the storage, where traps are planned to be placed, was designed; secondly, form of waste, it is necessary that highly
HANDLING SPENT CESIUM TRAPS FROM BN-350
123
active wastes were in solid form, their chemical and physical properties shall provide containment of radionuclides, and they shall correspond to accepted system of disposal. It is obvious that spent cesium traps do not meet both requirements of SRCPS-2003 and SRNDRW, since cesium traps belong to the category of highly active wastes, they contain chemically active substances: sodium and cesium–carbon compounds. 4. Existing Technologies for Handling Spent BN-350 Cesium Traps 4.1 AUTOCLAVE REACTOR (FRANCE) The autoclave process was developed in France for neutralization of limited amounts of radioactive NaK;11,12 however, if recalculated, the thermodynamic parameters of reaction can also be applied for sodium processing. If the volume of waste is less than 1 kg, the reaction is conducted with a plentiful volume of water. Reaction runs in the autoclave reactor (Figure 4), developed specially for assurance of safety at fast pressure buildup caused by the reaction of NaK with water. The NaK alloy was placed in sealed vessel that was placed into the reactor. The reactor was closed and filled with inert gas. Then water was supplied to the reactor, the vessel with NaK was opened using a special mechanism and moved into the water using a hydraulic jack. Pressure in the reactor is increased by the reaction of NaK with water. When pressure is stabilized, pressure in the autoclave reactor is released and all gases (mixture of hydrogen, inert gas, and vapor) are diluted by an inert gas flow. Monitoring of various parameters (temperature, pressure, hydrogen yield, and others) allow control of the completeness of the reaction. The system operates as a batch process and is adapted for limited quantities of NaK. At present, this process is used for processing the radioactive NaK wastes of the French experimental reactor Siloé located at Grenoble. Main advantages of the process are safety and efficiency. The amount of liquid wastes formed is also minimized. Safety is guaranteed, since there are no physical reasons for extra uncontrolled liberation of energy beyond the designed one that the reactor can absorb. Besides, the system includes automatic interlocks that prevent the possibility of introducing into the reactor an amount of NaK larger than it is designed for. This principle can be applied for sodium wastes of various sources: leaks and various sodium-containing elements of equipment. Prototype was tested, and industrial application is under study for future sodium waste handling unit ATENA. Developers of the unit assume that this autoclave reactor will be capable of processing various wastes, especially highly active graphite cartridges of cesium traps. French specialists selected the
124
HANDLING SPENT CESIUM TRAPS FROM BN-350
To the ventilation duct Pressure sensors
Entrance of water
Vessel in gas
Max. level of water Water level detectors
Vessel in water
Water drainage
Hydraulic jack
Figure 4. Autoclave reactor being used at SILOE reactor
following technology of processing of their cesium traps: trap cutting into cartridges, removal of sodium from them in autoclave reactor and separate processing of highly active alkaline solutions and solid residues of graphite and metal. Cesium traps have not yet been conditioned by this technology. 4.2 REACTION WITH AMMONIA (GREAT BRITAIN) For safe removal of sodium, potassium, and cesium from cesium traps with graphite sorbent, specialists of Great Britain considered a technology based on interaction of alkaline metals with ammonia.13 It was proposed to fill graphite sorbent containing cesium, potassium, and sodium residues with liquid ammonia at temperature –33°С. Then cooling is shutdown, and ammonia transforms at room temperature into gas phase, all sodium, potassium, and cesium are transformed into amides according to reaction: M + NH3 (Cs, Na, K)
↔
MNH2 + 1/2H2 amide
hydrogen
HANDLING SPENT CESIUM TRAPS FROM BN-350
125
At the next stage, a damp atmosphere is supplied into the trap and metal amides transform without vigorous reaction into corresponding hydroxides: MNH2 + H2O ↔ amide
MOH + NH3 ammonia
The trap can then be filled with water, hydroxides can be transferred finally into solution and highly active alkaline solution and solid residues of graphite/metal can be processed separately in the same way as it is assumed to be done in the French variant. 4.3 LEAD FILLING (RUSSIA) In Russia (NIIAR, Dimitrovgrad), where studies on development of cesium traps and corresponding technologies had been performed for a long period of time and as a result, dozens of their types were practically used at the reactors BOR-60, BN-350, BN-600, the technology of washing in water was eliminated at the first stage. Main disadvantages of the method are the complexity of units for assurance of explosion-proofness, cesium carryover, large volume of RAW to be processed and exposure of personnel to radiation during the long-term process. According to the opinion of Russian specialists, the most simple, reliable, and safe method is the method of “washing and storing in lead”.14,15 In experiments to validate the “lead fillup” method, cesium traps of the reactor BOR-60 were heated to a temperature of about 350°C and then were slowly lowered into a metal container filled with molten lead. Lead penetrated easily into the space between the graphite granules in the trap. To check this fact, the container with trap was cut after cooling the lead. It turned out that the free space between granules was filled completely with solidified lead. Activities of cesium isotopes in graphite granules were measured after their saturation in sodium, and then after staying in molten lead for 90 min at a temperature from 350°C to 500°C. The quantity of cesium lost from graphite into the molten lead was found to be from 8% to 12%. It was also determined that exit of cesium from surface of molten lead in the form of aerosols accounted for 3–6 × 10–3% of the total activity in granules. Hence, losses of cesium from trap at application of this technology are minor, and cesium that entered into lead remains, as it were, within the safety barrier. The method of filling cesium adsorbers with lead was repeatedly applied at the BOR-60 and BN-600 reactors. After sealing the traps with lead, the container was closed with a plug and placed in a dry channel of the BOR-60 reactor, and at the BN-600 reactor––into spent fuel cooling pond. At present, all cesium traps of the reactor BOR-60 are held in dry storage of RAW in a special
126
HANDLING SPENT CESIUM TRAPS FROM BN-350
building in NIIAR, and traps of the MAVR type from the BN-600 reactor are stored in the spent fuel storage pond. 5 Selection of Best Option for Handling Spent BN-350 Cesium Traps This section contains materials of the Options Workshop, which was held at the Nuclear Technology Safety Center, Almaty, Kazakhstan on November 8–10, 2004. 16 5.1 SELECTION PROCESS USED An approach accepted in Great Britain was used for selection of optimal options for handling spent cesium traps. It consists in passing through the following six stages: 1. Determine goals
2. Identify all options
3. Rationalize the options
4. Identify and set criteria
5. Develop the options
6. Assess the options
HANDLING SPENT CESIUM TRAPS FROM BN-350
127
5.1.1 Determination of goals Goals were determined in the following way: Handling spent traps shall meet requirements of both Kazakhstan and the international community: There should exist required procedures for reaching this goal in the form of:
Regulatory management Technical management Physical security and safety management
Long-term goal: To carry out conditioning of spent cesium traps in a form suitable for disposal in Kazakhstan as soon as radioactive waste storage is constructed. 5.1.2 Selection of decision-making criteria Decisions regarding selection of best options for handling traps were made using the following criteria:
Cost Environmental impact Assessment of risks (technical, political, and financial) Safety Physical security/opinions of organizations involved Regulatory requirements Technical factors Time factor - Could the proposed option be realized in the framework of existing schedule of work for transfer of the BN-350 reactor into safe-store state without coming into serious conflicts with other projects? Application to MAVRs - Could the proposed option be applied easily to TAs and to MAVRs?
Proposed options of handling spent traps were assessed according to these criteria.
128
HANDLING SPENT CESIUM TRAPS FROM BN-350
5.1.3 Options for handling spent traps Above considered options of handling traps were taken for further analysis: 1. To leave all as it is. 2. To pack in canisters identical to canisters with packaged fuel and send for disposal (long-term storage) to “Baikal”. 3. To drain sodium, to fill traps with some inert material (lead, plastic, etc.) without removal of sodium residues, to place it for long-term storage (disposal). 4. To drain sodium, to remove sodium residues, to fill traps with some inert material (lead, plastic, etc.), to place it for long-term storage (disposal). 5. To drain sodium, to remove sodium residues, to withdraw sorbent, and to use it as raw material for production of ionizing radiation sources. 6. To drain sodium, to remove sodium residues, to withdraw sorbent, to pack it into sealed package, and to place it for long-term storage (disposal). 5.2 ASSESSMENT OF OPTIONS A proprietary software package “HIVIEW” was used to record the scoring process to assist the panel in the assessment and analysis of results. This package is produced by the London School of Economics, UK, and is designed to facilitate Multi-Attribute Decision Analysis for choosing between options. 5.2.1 Procedure for options assessment Assessment of options is started from the fact that all options are ranked from one to another by each criterion. If such data as, for example, volumes of RAWs or cost are known, they are entered right away as points. For more subjective criteria, the options are ranked from 0 to 10. The option that is considered as the best by some criterion (safety, technical factors, etc.) receives 10 points, and the worst option receives zero. All other options are ranked within this interval depending from the fact how they are assessed in relation to the best and the worst options. Then all points are normalized in the interval from 0 to 100. The normalization process has two main functions. Firstly, it provides differentiation of options, i.e., the possibility to rank them by amount of points. If an option is the best or worst one for given criterion, it is very important to realize it. Secondly, normalization provides equal limits of points for all criteria, in spite of the fact that initially points may vary in arbitrary interval.
HANDLING SPENT CESIUM TRAPS FROM BN-350
129
In some cases it may turn out that options do not differ noticeably by some criterion. This problem is solved by the criterion-ranking procedure. If options are differentiated weakly by some criterion, this criterion receives low weight. 5.2.2 Procedure for weight determination It often happens that for decision-making, among identified criteria some criteria are more important than others. In order to take into account this fact at assessment of options, a weight is assigned to each criterion. Usually it is made after assignment of points by each criterion to options. Criterion weight reflects the level of its importance and its ability to differentiate options between each other for making specific decision. For example, safety is an important criterion, but if all options are considered as similar with respect to safety, this criterion shall be assigned with relatively low weight so that minor difference in points of options by this criterion does not dominate in making a decision. If criteria are grouped under some criterion of upper level, such as, for example, environmental impact, all criteria inside of this group shall be weighted relative to each other. Then criteria of upper level are weighted relative to each other in the same manner. In total, each individual weight is expressed as percentage from total weight. 5.3 RESULTS OF ASSESSMENT OF TRAP HANDLING OPTIONS 5.3.1 Assignment of weights to criteria Weights assigned to criteria are presented below:
Application to MAVRs Safety Physical security/involved parties Regulatory requirements Technical factors Cost Environmental impact Risks Time factor
2 3 3 2 1 3 3 2 1
Such criteria as “safety”, “physical security/involved parties”, “cost” and “environmental impact” were determined as crucial ones for making decision on the project of disposal of spent cesium traps. Each of these criteria was determined as “approving” the work under the project.
HANDLING SPENT CESIUM TRAPS FROM BN-350
130
“Application to MAVRs”, “regulatory requirements”, and “risks” were considered in order to perform differentiation of options and, besides, it was very important to consider how easily each of the options meets these criteria. “Technical factors” and “time factor” were determined as important criteria for decision-making, but they have less effect on differentiation of options. Amounts of points for each option in relation to weighted criteria are presented in Table 5. TABLE 5. Results of trap options assessment Criterion
Application to MAVRs Safety Physical security/involved parties Regulatory requirements Technical factors Cost Environmental impact Risks Time factor
Total
Weight
1
2
3
4
5
6
Leave as is
Send to “Baikal”
Drain, fill, and dispose
Drain, remove residues, fill, and dispose
Drain, remove residues, and withdraw sorbent for sale as source
Drain, remove residues, withdraw and immobilize sorbent for disposal
2 3
100 100
90 90
0 70
0 50
70 0
70 20
3
0
10
70
70
20
100
2
100
40
80
70
0
30
1 3
100 100
80 50
60 90
40 60
0 0
20 40
3 2 1
100 56 0 76
100 44 100 64
50 60 80 62
40 53 60 50
0 18 20 13
20 56 40 46
5.4 FURTHER ANALYSIS OF OPTIONS FOR BN-350 CESIUM TRAPS After fulfillment of assessment, it is necessary to try to analyze each option and give comments on them: 1. Leave as is (76 points)
This option does not meet the requirements for transfer of nuclear reactors into a “safe storage” (SAFESTOR) state that implies either removal of all mobile active sources from BN-350 or their immobilization for long-term storage before transfer of the reactor into SAFESTOR state.
HANDLING SPENT CESIUM TRAPS FROM BN-350
131
This option does not provide conditioning of traps and their further disposal, so this task is shifted to shoulders of the next generations. At the same time, there is a high probability that it will be more difficult to perform conditioning of traps in the future because of problems with knowledge retention and the departure of qualified personnel from BN-350. It is unlikely that this option would be acceptable for the involved parties.
2. Pack in canisters identical to canisters with packaged fuel and send for disposal (long-term storage) to “Baikal” (64 points)
Feasibility of this option strongly depends on successful realization of the project for transportation of BN-350 spent fuel from Aktau to the Baikal site, Kurchatov city. Feasibility of this option depends on whether regulatory authority would allow loading of traps into licensed packages that are to be designed for transportation and storing of spent fuel of BN-350. Probably, it will require additional radiation, strength, and heat calculations to confirm the safety of transporting and storing traps at the Baikal site. Safety analysis shall also take into account the presence of sodium and cesium in traps, and hence, possibility of their interaction with water in case of a design accident during transportation – i.e., the fall of package from the height of 9 m and its filling with water. As well as Option 1, conditioning of traps and their further disposal are not considered here—this task falls on the shoulders of the next generations.
3. Drain sodium, fill traps with some inert material (lead, plastic, etc.) without removal of sodium residues, and place in long-term storage (disposal) (62 points)
Advantage of this option is that it contains a minimal number of steps. The option implies disposal of traps. Issues remaining indefinite are: to what extent the traps can be filled and what level of leachability of sodium and cesium will remain in the final form of waste. In comparison with Options 1 and 2, but not to the extent as Options 4 and 6, Option 3 considerably improves the degree of physical protection for the radioactive material contained within the conditioned traps. Option 3 can also be used for MAVRs, but with extra difficulties. Option 3 does not provide withdrawal of sorbent from the traps and, hence, the possibility of dilution of activity in the RAW package is limited. It is necessary to assess whether the RAW package would be classified as medium-active waste that can be disposed according to existing concept of disposal in the Republic of Kazakhstan or whether alternative options of handling traps after their conditioning by this option would need to be proposed.
132
HANDLING SPENT CESIUM TRAPS FROM BN-350
4. Drain sodium, remove sodium residues, fill traps with inert material (lead, plastic, etc.), and place in long-term storage (disposal) (50 points)
Considerable difficulty may arise for this option with MAVRs, because handling MAVRs implies removal from their biological shielding. The option does not provide withdrawal of sorbent from the traps and, hence, the possibility of dilution of activity is limited. It is necessary to assess whether the RAW package would be classified as medium-active waste that can be disposed according to existing concept of disposal in the RK or whether, as in the previous option, alternative options of handling traps after their conditioning by this option would be needed. The option is less flexible in comparison with Option 6.
5. Drain sodium, remove sodium residues, withdraw sorbent and use as raw material for production of ionizing radiation sources (13 points)
This option contains the maximum number of steps and requires conducting a considerable amount of research. The option is strongly subject to political and commercial risks. There is a high probability that the cost of producing ionizing radiation sources by this route will turn out to be higher than their market cost.
6. Drain sodium, remove sodium residues, withdraw sorbent, pack it into sealed package and place it in long-term storage (disposal) (46 points)
This option is the most flexible in allowing selection of the final form of waste (cement, geocement, asphalt, vitreous form, etc.). The option allows concentration or dilution of radionuclides in the RAW package. The option requires handling highly active sorbent during the conditioning of the traps. A problem may result from the fact that complex processes of handling highly active RAW whose amount is not large will be required.
Proceeding from the results of assessment of options, from their analysis and taking into account the above-mentioned comments, only three of six considered options could be selected for further consideration:
Option 2: Pack in canisters identical to canisters with packaged fuel and send for disposal (long-term storage) to “Baikal”. Option 3: Drain sodium, fill traps with some inert material (lead, plastic, etc.) without removal of sodium residues, and place in long-term storage (disposal). Option 6: Drain sodium, remove sodium residues, withdraw sorbent, pack it into sealed package and place in long-term storage (disposal).
HANDLING SPENT CESIUM TRAPS FROM BN-350
133
It is clear that additional research and development work should be implemented to substantiate Options 3 and 6. The issues related to this fact are discussed in the next section. 6 Experiments to Support Cesium Trap Handling Options 6.1 DETERMINING PRIORITIES In the previous section six quite generalized options for handling spent BN-350 traps were considered; three options were chosen for further consideration. However, there are still a lot of issues to be considered for each of the chosen options in order to identify the best handling method. For example, it is not clear yet which of the technologies is best for draining sodium from the traps (depressurization or overpressurization with noble/inert gas). A preferred material (such as lead, bitumen, or other) for immobilization of sodium and radionuclides has not yet been chosen. Neither is it clear which technology will work better for removal of sodium residues from the traps (vacuum distillation, chemical reactions, nitrogen/steam treatment, alcohol, etc.). As a result, in order to obtain answers to all these issues, it is necessary to perform a considerable amount of research, which correspondingly increases project costs significantly. That is why it would be reasonable at this stage to narrow the directions for further investigation and try to determine optimal technological approaches for chosen trap handling procedures. It is proposed to run experiments to verify the possibilities for realization of chosen optimal technological approaches. But the intention is that the program for experimental investigations should be completed within a reasonable time at a comparatively reasonable cost. Let us analyze the problem of developing the program for future experiments from the point of view of its optimization. Issues related to realization of the option “Pack the casks identical to casks with packed fuel and deliver them to safe-store at ‘Baikal’ facilities” do not require any experiments, so this option is not further considered here. Let us consider Option 3 “Draining of sodium; filling-up the traps with some inert material such as lead, plastic or other without removal of sodium resides; safe-store (disposal)” and Option 6 “Draining of sodium; removal of its residues; extraction of the sorbent; packing it into sealed container and delivery to safe-store (disposal)”. Both options imply draining of sodium from the TAs and, therefore, it is necessary to evaluate the possibilities for draining of sodium from the TAs and to verify the assessments in experiments. In considering Option 6, removal of sodium residues from the traps is of particular concern, and the choice of optimal technological approaches for residue removal would definitely require considerable research. Choice of final
134
HANDLING SPENT CESIUM TRAPS FROM BN-350
waste form and filler material for the immobilization of extracted sorbent with cesium radionuclides in its matrix would also require many experiments to be performed. In considering Option 3, the choice of immobilizing filling material for the traps and the technology for filling the traps with this material is important. Obviously, carrying out experiments with all hypothetically appropriate filler materials would require huge efforts; therefore, it would be reasonable, at first, to analyze the possibility of using lead as filling material, since this technology has also been used and approved in NIIAR, Dimitrovgrad, Russian Federation for stabilization of cesium traps supplied from reactors BOR-60 and BN-600. Additionally, one should take into account the fact that when scoring various options, Option 3 gained 62 points, while Option 6 gained only 46 points; that means Option 3 is considered more favorable and we would consider it as a first priority in our further analysis. One of the crucial shortcomings of Option 3 is the inability to dilute the activity in the package of SRW since this option does not imply extraction of sorbent from the traps. As was mentioned above, currently in the RK there is no concept developed for disposal of high-level RAW and there are only rules for handling of low- and intermediate-level RAWs. The upper level for intermediatelevel waste acceptable for surface disposal in the RK is 104 MBq/kg for betaactive radionuclides and 4 MBq/kg for alpha-active radionuclides for one package of SRW. Supposing that upon sodium draining from TAs (sodium have already been drained from MAVRs) the TAs and MAVRs are filled in with lead, let us assess the specific activity in the traps in order to evaluate their category of RAW. Results of calculated specific beta- and alpha activities performed for the spent traps upon their filling in with lead considered together with the casing and biological shieldings as a SRW packages are presented in Table 6. This is allowable because according to SRNDRW “specific activity can be averaged over the whole volume of the waste package”. One can see from Table 6 that according to the specific activity of longliving alpha-active radionuclides all the traps, except MAVR-1, can be considered as intermediate level waste packages acceptable for near surface disposal. Assessment of specific activities for alpha-active radionuclides in MAVR-1 provides the value 4.2 MBq/kg, which slightly exceeds the allowable 4 MBq/kg, but taking into account the conservatism of the assessment performed in paragraph 2.3, one can accept that, upon additional substantiations, permission for its near surface disposal would be obtained. Still, regarding beta activities the packages for the traps MAVR-1, MAVR-4, and TA-622 would presently be considered as high-level SRW. Other traps can be disposed as intermediatelevel SRW packages in near surface RAW disposal. The traps MAVR-1, MAVR-4, and TA-622 can be disposed only after 40–50 years following Cs137 decay in the traps down to acceptable levels. Taking into account that currently
HANDLING SPENT CESIUM TRAPS FROM BN-350
135
there is no surface disposal for RAW constructed in the RK, this postponed solution with a safe-store period until final disposal can be quite acceptable if one succeeds to prove that the sorbent containing cesium and residual sodium after its immobilization in lead matrix would meet the requirements for the form of RAW for their near surface disposal. A site for temporary safe-store of conditioned traps could then be allocated in the building of the BN-350 reactor, or at the store for radiation sources, or in a room at the RAW handling site. TABLE 6. Specific activity in the traps on lead filling (as SRW packages) Trap number MAVR-1 MAVR-2 MAVR-5 MAVR-4 TA-622 TA -621 TA -623 TA -624 TA -627 TA -625 TA -626
Calculated Cs-137 activity on filling trap with lead (with casing/shielding) (МBq/kg) 3.0 × 104 8.6 × 103 8.6 × 103 1.1 × 104 2.4 × 104 9.9 × 103 3.2 × 103 5.7 × 103 1.9 × 103 8.5 × 102 1.9 × 102
Calculated Pu activity on filling trap with lead (with casing/shielding) (МBq/kg) 4.2 1.2 1.2 1.5 3.4 1.4 0.5 0.8 0.3 0.1 0.03
It is obvious that Option 6 does not have such problems due to its flexibility and a priori makes it possible to transfer the traps into RAW packages acceptable for near surface disposal in the RK. The main shortcoming of this approach is the high cost stipulated by its complexity and the considerable amount of research and development required. Besides that, the amount of secondary RAW generated in Option 6 would be much higher than that envisaged for Option 3. So, in order to optimize the program for further scientific investigations, it might be possible to focus at first on the problems related to possibilities of Option 3 realization for the technology of trap filling-in with lead, and then, in case of difficulties with realization of this option, one should consider substantiation experiments for realization of Option 6. Let us make a brief comparative analysis of the chosen options in order to make sure that the correct priorities have been assigned; so, we would consider not generalized, but specific technological approaches that accompany realizetion of these options.
136
HANDLING SPENT CESIUM TRAPS FROM BN-350
6.2 COMPARISON OF TECHNOLOGIES FOR TRAP CONDITIONING The safety of handling cesium traps is conventionally associated with the problem of residual sodium. The design of MAVRs envisages sodium draining during removal from reactor and thus in the MAVRs sodium may remain in graphite pores only. However, in TAs, all space inside the trap shell including inlet and outlet elbows is filled with sodium. The priority task then becomes to remove sodium from the traps completely, though we just should keep in mind that a trap with drained sodium containing a high content of cesium does not become less fire- or explosion-hazardous. Let us consider a method of carbonization17 or the process “SANDS”––gas oxidation––developed in Belgium and patented in the USA and EU. Analysis of a technology to perform cesium trap treatment with carbonization shows the need to include the following main steps: • • • • • •
Sodium extrusion from the TAs with gas Passivation of sodium residues in MAVRs and TAs by expulsion with dry or wet carbon dioxide gas for carbonization Cutting the traps in the hot cells Sorbent removal, its grinding and obtaining a homogeneous mixture Mixing with cement for solidification and decreasing its specific activity to intermediate-level RAW requirements Tempering of cement mortar with water, mixing and packing into barrels for compound hardening and transfer to safe-store.
Application of carbonization technique requires establishing a remotely controlled line for mechanized processing of graphite sorbent, considerable amounts of auxiliary equipment, and results in the generation of considerable amounts of RAW, which would not be recommended by IAEA. Besides the technological and economical challenges with application of carbonization, this method reveals major shortcomings: decomposition of graphite material structure on formation of sodium carbonates. Decomposition of graphite material structure may result in considerable release of cesium into the gas environment and radioactive contamination of the equipment, accompanied with the necessity to have additional equipment for its entrapment. Even if one succeeds in designing complex technological equipment for carbonization of sodium residues in a trap, problems would still persist with sorbent extraction from the traps and its further handling. We face the same problems when analyzing the method of hydrocarbonate formation. The model for a technological process implies the following: • •
Draining of sodium from the traps Expulsion with wet atmospheric air and carbon dioxide gas
HANDLING SPENT CESIUM TRAPS FROM BN-350
• •
137
Sorbent removal Its immobilization and further disposal
Experiments performed in NIIAR showed that processing of sodium pre-treated graphite samples (LAG and RVC) with air resulted in practically complete destruction of the graphite or carbon material. As a result of such processes hydrocarbonate and sodium hydroxide are formed. Formation of hydroxide and its further interaction with carbon dioxide in air results in formation of friable reaction products, together with formation of a fragile carbonate cake.18 At expulsion of wet nitrogen with carbon dioxide over a pan with a 6–45 mm thick sodium layer a carbonate cake was formed; the amount of material formed was 10 times greater than the volume of initial sodium. Upon expulsion with wet air a friable mixture was formed in the trap casing; the mixture resembled “sugaryalbuminous kiss” of carbon, cesium compounds, sodium carbonates, and hydrocarbonates. After such procedure, cutting off and sorbent removal from the trap became a complicated task due to sorbent destruction. Later, it would be quite difficult to transform such a mixture of RAW and immobilize it for safe disposal. Hence, the process of hydrocarbonate formation (hydrocarbonization) may, in principle, be used for removal of sodium residues from nondraining volumes, but this process seems to be of little use for cesium traps, particularly for those with high accumulated activities, as the traps from BN-350. A few words should be said about steam-gas washing technique; the method is based on reaction of sodium with water and formation of a sodium hydroxide water solution. Steam-gas method has been used for washing out of various equipment at almost all reactor facilities.19–21 In the case of trap handling this method was never used due to its main shortcoming revealed at washing out of equipment highly contaminated with radionuclides. When using this method a large amount that requires further processing. As well, there are other shortcomings such as the considerable time required for the processing, difficulties with the process control and possibilities of achieving high temperatures in local places. Substantiation for application of this method for conditioning of cesium traps from BN-350 would require considerable research to be performed. Methods of sodium residues removal with alcohol22,23 were also never practiced on equipment that contains, besides sodium, high concentrations of radionuclides; therefore, these methods can only be recommended for conditioning of BN-350 traps upon extensive scientific investigations. Considering the sodium distillation technique,24,25 when applied to removal of sodium residues from the traps, the main problem is in the considerable cesium release from sorbent and, correspondingly, contamination with cesium isotopes of the distillation equipment. In order to reduce cesium release from the sorbent, one needs to reduce the temperature of the process and, probably, operate within a temperature range where sodium sublimation takes a long time
138
HANDLING SPENT CESIUM TRAPS FROM BN-350
and is not effective. As an alternative for removal of sodium residues one can consider sodium distillation followed by distillation of cesium on heating of the traps with cesium trapping in capsules with a cold sorbent agent and consequent immobilization of the capsules with concrete in barrels. Technologically, such heating of a trap up to required temperature is quite difficult. According to experimental data obtained in NIIAR for low-ash graphite treated in sodium and saturated with cesium, heating up to 600°C removed about 15% of Cs-137, at 1,000°C––35% , and at 1,200°C there remained in graphite about 40% of cesium. The shortcoming of the technology is in the large amount of concrete blocks with cesium capsules which would result. For instance, processing of the trap TA-622 would result in the generation of 141 concrete blocks of 0.7 m3 each: 100 times decrease in gamma-radiation dose rate for Cs-137 equal to corresponding decrease of activity without shielding is achieved at concrete thickness of 44 cm. Another shortcoming of this high-temperature distillation method (temperatures within the range 800–1,200°C) is in potential hazard from accidental failure of the vacuum system at such temperatures and the long time the process takes (potential release of cesium to the environment). If the choice is to use a processing technology connected with removal of sodium residues from the traps, the steam-vacuum method26 might be effective since it assures removal of sodium from fine pores. On the other hand, when applying this method one can envisage considerable release of cesium isotopes to the vacuum system. The method of washing the equipment out with water in vacuum may assure process safety for removal of sodium residues and assures sodium removal from fine slits in the equipment but accidents may occur with loss of control over the process. A large amount of generated liquid RAW with a tenfold increase only at the first stage of processing is the main shortcoming of the method. Expected growth in the total amount of generated RAW only at the first stage of processing is 15 times. In France there was proposed and tested another technology: cutting off traps in hot cells; taking fragments with specific amount of sodium; placing them into an autoclave reactor; dissolution of sodium and, partially, of cesium; preparation of RAW portions with specified activity and their immobilization in concrete matrixes. But, there are concerns that cesium may be present in the BN-350 traps in the form of lamellar compounds that demonstrate pyrophoric properties; this fact leads to concern about the safety of the entire process of BN-350 trap cutting and their consequent washing out for sodium removal. While the ratio “cesium atoms to carbon atoms” in the traps is one order of magnitude less than the same of typical pyrophoric lamellar compounds, these considerations are still of particular concern when choosing a method for the trap TA-622 due to its much higher cesium concentration compared to the cesium concentration in traps from the reactor RAPSODIE. Hence, safety requirements would call for over-all experimental tests and verification of the
HANDLING SPENT CESIUM TRAPS FROM BN-350
139
method chosen for the TAs from BN-350. Therefore, safety considerations as well as large amounts of liquid RAW subjected to further processing raise serious doubts about the applicability of the method in the RK, particularly, taking into account the much higher activity of cesium in BN-350 traps compared to that in traps of the RAPSODIE reactor. Filling in the traps with lead (alloy Pb-Bi) outperforms other technologies in terms of minimization of amounts of generated RAW, in the comparative simplicity of the method and in the comparatively lower cost of the technology. Investigations performed in NIIAR proved experimentally the homogeneity of pore filling-in with lead (Pb-Bi) between the grains of low ash graphite. That means the adsorber MAVR in a cask filled-in with lead (Pb-Bi) may be considered as a casing with multibarrier shielding: • •
•
Radioactive cesium atoms are chemically bonded in carbon compounds and, therefore, are kept to a certain degree in the graphite matrix both on leaching with water and high-temperature distillation. Lead (Pb-Bi) that fills up all volume between grains and cask walls is another barrier for migration of the radionuclides: it offers a high melting temperature, low corrosion rate in water and air, and a low diffusion coefficient. Steel casing of MAVR and sealed steel cask are the barriers (in analogy: trap walls and steel shielding are barriers for the TA traps).
A priori one can suppose that such packing of highly active cesium traps should assure environmental safety during storage for hundreds of years. Even emergency situations such as explosion with cask disintegration would result in release of only a small fraction of the stored activity. 7 Conclusion This paper has discussed the possible options for handling spent BN-350 cesium traps designed by IAE NNC RK, ANL, and MAEC Kazatomprom and for MAVRs designed by NIIAR, together with an overview of optimal options. Following analysis, three options were chosen as most favorable for further consideration:
Pack the traps into canisters identical to the packed fuel canisters and send them to disposal (safe-store) at “Baikal” facilities. Drain sodium, fill-in the traps with some inert material (such as lead, plastic, or other) without removal of sodium residues and send them to safestore (disposal). Drain sodium, process residues, extract the sorbent, pack it in sealed containers, and send to safe-store (disposal).
HANDLING SPENT CESIUM TRAPS FROM BN-350
140
More detailed consideration of these options would make it possible to choose the priority directions for further scientific investigations. As a priority technology that requires additional experimental verification we propose to consider conditioning of all cesium traps from reactor facilities BN-350 by filling-in with lead (Pb-Bi alloy) in accordance with the option “Drain sodium, fill-in the traps with some inert material without removal of sodium residues and send them to safe-store (disposal)”. Advantages of this technology over other ones are as follows: the technology has already been used previously; relative safety of technological procedures, minimum volumes of RAW; the technology is simple and economically effective; prepared packages are environmentally safe for long-term storage or disposal. It has been shown that upon filling-in the traps with lead there is a possibility to dispose of eight of the eleven traps as intermediate-level SRW packages at near surface disposal site for RAW. The traps MAVR-1, MAVR-4, and TA-622 can be disposed within about 40–50 years on decrease in Cs-137 activity to acceptable levels. In that the RK currently has no surface disposal facility for RAW, such a postponed solution—safe-store until final disposal—is considered appropriate. 8 Acknowledgments This work was financially supported jointly by the US Department of State and by DTI, UK through the ISTC K-512 project. References 1.
2.
3. 4.
5. 6.
Feuerstein, H. et al., “Mechanism of release of radioactive products into liquidmetal coolants, their transport within the circuits and removal from LMFBRs”, Atomic Energy Review, 17(3), 1979, рр. 697–761. Sobolev, A.M. et al., “Systems of purification of sodium coolant of fast reactors from cesium”, Report NIIAR Workshop “Units and systems of fast reactors”, USSR, 1988. Sobolev, A.M. et al., “Systems of purification of sodium coolant of the fast reactors from cesium”, VANT, Issue #1, 1990, pp. 66–72. Sobolev, A.M. et al., “Graphite cesium traps of the reactor BOR-60. Results of tests”, NIIAR Workshop “Monitoring and purification of impurities, mixing hydrodynamics at low coolant flows, development and operation of sodium equipment”, USSR, 1989. Gusev, N.G., Belyaev V.A., “Radioactive releases in biosphere”. Reference book, Energoatomizdat, M., 1986. Handling radioactive wastes of nuclear power plants. Code of provisions. IAEA safety standards, Series of publications on safety #69, IAEA, Vienna, 1987, p. 11.
HANDLING SPENT CESIUM TRAPS FROM BN-350
7. 8.
9. 10. 11.
12.
13. 14.
15. 16. 17. 18.
19.
20. 21.
22.
23. 24.
141
Romanenko, О. et al., “Cleaning cesium radionuclides from the BN-350 primary sodium”, Nuclear Technology, 150, 2005. Kizin, V.D., Polyakov, V.I., Sobolev, A.M., Preparation for disposal of radionuclide traps of the fast reactors––VANT, series “Nuclear engineering and technology”, ISSN 0321-3099, Vol. 6, 1991, pp. 41–47. Subbotin, V.I., Ivanovsky, M.N., Arnoldov, M.N., Physical and chemical fundamentals of application of liquid metal coolants. Atomizdat, Moscow (1970). Novikov, Y.N., Volpin, M.E., Layer compounds of graphite with alkali metals–– achievements of chemistry. Т.XL. Issue 9, 1971, pp. 1568–1592. Petitfour, B. et al., Oxidation and conditioning of contaminated metallic sodium, Meeting: Radioactive Sodium Waste Treatment and Conditioning, Lyon, France (2002). Rodriguez, G., Saroul, J., Arnauld des Lions, J.P., “Methods for sodium waste treatment coming from liquid metal fast reactors”, ENC’98, Nice, France, October 1998. Green, T.H., Mackinnon, D., UK Patent No.14056, 1986. Kizin V.D., Polyakov V.I., Sobolev A.V. “Preparation of radionuclide traps of sodium cooled reactors for disposal”, Report for Russian–French seminar on treatment of radioactive sodium wastes, Dimitrovgrad, Russia, 15–16 October 1998. Kizin V.D., Shtynda Y.E., Cesium Trap Conditioning for Disposal. IAEA Meeting, “Radioactive Sodium Waste Treatment and Conditioning”, Lyon, France (2002). Options Workshop Report 85765/T9/9.1/OPT/005. Radioactive Sodium Waste Treatment and Conditioning. Review of main aspects, document for final consultants’ meeting, IAEA-TECDOC, Vienna. Rodrigues, G., Gastaldi, O., “Sodium carbonation process development in a view of treatment of the primary circuit of LMFR in decommission phases”, International Conference on Radioactive Waste Management and Enviromental Remediation, Bruges (2001). Bohnel, K., Hanke, D., Stade, K.C., “Washing sodium contaminated components at KNK II”, Proceedings of the Technical Meeting at Phenix, Marcoule, France, 1995. Whitlow, W.H., “Sodium cleaning and decontamination of PFR fuel charge machine”, Proceedings of the LIMET-84, London, Vol. 2, 1984, pp. 125–132. Mesnage, B., Marieteau, P., “Creys-Malville nuclear plant. Handling line operating experience”, Proceedings of the International WG on Fast Reactors, IAEA, Vienna, 1996, pp. 79–89. Welch, F.H., Steele, O.P., “Non-aqueous removal of sodium from reactor components”, Proceedings of the Specialists Meeting on Sodium Removal and Decontamination, Richland, Washington, 1978, pp. 173–177. Ruther, W.E. et al., EBR-II experience on sodium cleaning, ibid., 1978, pp. 182– 193. Abrams, C.S. et al., “Development of disposal method and burial criteria for radioactive sodium wastes”, Proceedings of the LIMET-84, Vol. 2, London, 1984, pp. 165–170.
142
HANDLING SPENT CESIUM TRAPS FROM BN-350
25. Smith, C.C., “TNO experiments on sodium cleaning of large plant components by vacuum distillation”, Procееdings of the Specialists Meeting on Sodium Removal and Decontamination, Richland, Washington, 1978, pp. 8–12. 26. Kirisava, T. et al., “Evaluation tests on several cleaning methods for crevices of sodium components”, Proceedings of the Third International Conference on Liquid Metal Engineering and Technology––LIMET-84, Vol. 2, London, 1984, p. 139.
ACCOUNT AND CONTROL OF NUCLEAR MATERIALS AT THE WWR-SM REACTOR IN THE INSTITUTE OF NUCLEAR PHYSICS, TASHKENT B. S. YULDASHEV, U. S. SALIKBAEV, S. A. BAYTELESOV,∗ A. A. DOSIMBAEV, AND U. A. KHALIKOV Institute of Nuclear Physics Tashkent, Uzbekistan Abstract: The WWR-SM research reactor has been in operation since 1959, with a variety of fuels—10%-enrichment Ek-10, 90%-enrichment IRT-3M (1972), 36%-enrichment IRT-3M (1998). The move today is toward use of IRT-4M fuel with <20% enrichment. Tight control of nuclear materials has been maintained over the years by physical and administrative means: fresh fuel is stored in a dry locked room, while spent fuel is kept in racks in three fuel storage pools (FSPs) containing high-purity water that minimizes cladding corrosion. Between 1991 and 1994 the accounting and control of nuclear materials was according to the method used by Ministry of Atomic Energy and Industry (MINATOM) in Russia. Since 1994, nuclear materials data are forwarded to the International Atomic Energy Agency (IAEA) and are checked by quarterly inspections. An agreement was signed with Russia in 1997 for supply of fresh fuel and take back of spent fuel for reprocessing (with return of fission-product waste to Uzbekistan). Keywords: WWR-SM research reactor, Ek-10, IRT-3M and IRT-4M fuels, spent nuclear fuel (SNF), fuel storage pools, inventory change report, physical inventory list, materials balance report
1 Storage of Reactor Fuel 1.1 FRESH FUEL The storage of fresh (unirradiated) assemblies at the WWR-SM research reactor at the Institute of Nuclear Physics (INP) is in airtight containers in a dry locked room, where the presence of water is excluded. The location of ______ ∗
To whom correspondence should be addressed: S.A. Baytelesov, Institute of Nuclear Physics, Ulugbek, Tashkent, Uzbekistan 702132; e-mail:
[email protected] 143 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 143–146. © 2007 Springer.
144
NUCLEAR MATERIAL CONTROL AT WWR-SM REACTOR
fuel assemblies on shelves ensures a subcritical state of not less than 0.05 for all conditions, even total flooding of the room with water. The following surveillance devices are installed in the storage room: • • • •
Two fission detectors with automatic alarm signal warning Fire detector Video camera Intrusion sensor
1.2 SPENT FUEL There are three fuel storage pools (FSPs) at the WWR-SM reactor made from stainless steel and covered by deck plates. Spent nuclear fuel (SNF) assemblies unloaded from the core are installed in them, where the SNF is aged for not less than 3 years before shipping for reprocessing. The first FSP has 60 cells and the second FSP (Figure 1) has 192 cells on two levels, on a pitch of 150 mm, which provides a subcritical limit of not less than 0.05 when filled with fresh fuel. The third FSP was added by INP in 2000 and has four tanks with a capacity of 162 cells: 44 in the first and fourth tanks, and 37 in the second and third tanks.
Figure 1. Entrance to fuel storage pool No. 2 of WWR-SM reactor through the reactor deck plate
NUCLEAR MATERIAL CONTROL AT WWR-SM REACTOR
145
The parameters for the water in the FSPs are listed below: they are the same as for the water used in the reactor primary circuit. рН 5.5–6.5
Electroconductivity <4
СI mcg/kg <50
AI+3 mcg/kg <50
2 Accounting and Control of Nuclear Materials The accounting and control of nuclear materials has been carried out for over 40 years at the WWR-SM reactor. Since 1959 nuclear fuel was of the EK-10 type with 10% U-235 enrichment, used when the reactor operated at 2 MWt. After reconstruction in 1972 IRT-3M fuel with 90% U-235 enrichment was used. After installation of a gas-cleaning system in 1979, reactor power was raised to 10 MWt. IRT-3M fuel with 36% U-235 enrichment has been used since 1998. Conversion to IRT-4M fuel (<20% U-235) will occur in the near future. Accounting starts on receipt of nuclear fuel at the WWR-SM reactor and consists of: • •
Passport data for the nuclear material Registration in “The journal for account of fuel assemblies”, where the following parameters are entered: - No. of the assembly - Amount of isotopes U235 + U238 (g) - Mass U235 (g) - Date of fabrication - Date of entry into INP - Date of loading - Date of unloading - % Burn-up - Remaining mass of U235 (g) - Mass of isotopes U235 +U238 (g) after cycle - Energy output of the assembly (MWt days) - Journal No. and page “History of fuel assembly loading” - Date of shipment and reprocessing
Since 1991 quarterly accounting and control of nuclear materials has been carried out according to the method of the Ministry of Atomic Energy and Industry (MINATOM) of the Russian Federation. Since 1994 all information about account of nuclear materials has been sent to the State Atomic Inspectorate of the Republic of Uzbekistan and on further through official channels to the International Atomic Energy Agency (IAEA).
146
NUCLEAR MATERIAL CONTROL AT WWR-SM REACTOR
This nuclear information is presented in the following form: • • •
Inventory change report – ICR Physical inventory list − PIL Materials balance report – MBR
Every 3 months IAEA inspectors visit the INP to check this information. The storage of fuel assemblies conforms to standard procedures in as much that the number of free cells in the FSPs can accommodate a complete unloading of the core following an emergency. 3 Status of Spent Fuel at WWR-SM Reactor Until 1992, SNF assemblies were transported regularly to MAYAK, Russia, for reprocessing after 3 years aging (from 1974 to 1992, 440 assemblies were sent: 128 Ek-10s, 90 IRT-2Ms, and 222 IRT-3Ms). In August 2004, 11 unclaimed SNF assemblies were sent to Russia for reprocessing (1 with 90%, 4 with 36%, and 6 with 10% enrichment) and 12 assemblies (3 with 90%, 4 with 36%, and 5 with 10% enrichment). At present there are 370 fuel assemblies at the reactor: 28 in fresh storage, 18 in the core, and 321 in the three FSPs: -
33 in FSP No. 1 (29 36% and 4 19.8% enrichment) 126 in FSP No. 2 (48 with 90%, 67 with 36%, and 11 with 10% enrichment) 162 in FSP No. 3 (all with 90% enrichment).
The average burn-up is 60% and 252 assemblies have been prepared for shipment to MAYAK for reprocessing. On December 22, 1997 the agreement was signed between Uzbekistan and Russian Federation Governments about scientific technical collaboration in the field of peaceful use of atomic energy. Article 8 deals with the supply of fresh fuel to the Uzbekistan Republic and the back shipment of SNF to the Russian Federation for reprocessing with the subsequent return of radioactive waste to Uzbekistan; these activities are legislated by the State in which collaboration is carried out and also contracts between enterprises and organizations of Uzbekistan Republic and the Russian Federation. Shipments of items of nuclear export from the Russian Federation to Uzbekistan will be carried out on completing the placement of all nuclear activities in Uzbekistan under IAEA guaranty.
ACTIVITIES AT THE KHARKOV INSTITUTE RELATED TO THE PROBLEM OF SPENT NUCLEAR FUEL MANAGEMENT
V. M. AZHAZHA, I. M. NEKLYUDOV, S. Y. SAYENKO,∗ AND V. N. VOYEVODIN National Science Center Kharkov Institute of Physics and Technology of the National Academy of Sciences of Ukraine, Kharkov, Ukraine Abstract: The National Science Centre Kharkov Institute of Physics and Technology (NSC KIPT) is one of the oldest and largest centers for science in Ukraine. For many years experts at NSC KIPT have carried out scientific and technological research in various areas of the nuclear industry. During the last 10 years special attention has been given to resolution of problems of radioactive waste (RAW) isolation. As a prospective method for RAW and spent nuclear fuel (SNF) management, encapsulation in a protective glass–ceramic form with a composition similar to natural minerals has been proposed. Research and development in the synthesis of crystal materials—titanates, zirconates, and alumino-silicates—is being performed. Actinides, and also strontium and cesium due to isomorphic replacements, can be located in the structure of these crystal materials. Technologies of obtaining these materials are based on methods of compacting, and sintering these powder materials under pressure. As initial components, industrial powders, the powders manufactured by grinding from natural rocks––the granite and clay, and also the powders obtained by chemical sedimentation of corresponding solutions of salts are used. Use of hot isostatic pressing (HIP) at the stage of sintering creates a material with a low coefficient for radionuclide diffusion and a low rate of leaching by ground water. The results of scientific and technological researches carried out at the NSC KIPT in the field of SNF isolation are presented and briefly discussed. Keywords: radioactive waste (RAW), spent nuclear fuel (SNF), interim storage, deep geologic disposal, encapsulation, granite and clay monoliths, hot isostatic pressing (HIP), radionuclide diffusion, groundwater leach tests
______
∗ To whom correspondence should be addressed: S.Yu. Sayenko, NSC Kharkov Institute of Technology, 61108 Akademicheskaya st.1, Kharkov, Ukraine; e-mail: sayenko@ kipt.kharkov.ua
147 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 147–164. © 2007 Springer.
148
SPENT FUEL MANAGEMENT IN UKRAINE
1 Introduction The National Science Center Kharkov Institute of Physics and Technology (NSC KIPT) is located in the city of Kharkov. Kharkov is one of the biggest industrial and scientific centers of Ukraine. The NSC KIPT has deep historical roots. In 2003, we celebrated the 75th anniversary of its foundation. The Institute is mainly known for the fact that on October 10, 1932, the famous scientists A. Valter, K. Sinelnikov, A. Leipunsky, and I. Latyshev for the first time in the USSR carried out an outstanding physics experiment––they split the nucleus of lithium. This achievement can be truly considered the starting point for a subsequent dynamic development of nuclear physics, physics of accelerators, and nuclear materials science. The main activity of the NSC KIPT is to carry out research in science and technology and to provide for the development of nuclear power engineering in Ukraine. 2 Spent Nuclear Fuel in Ukraine Spent nuclear fuel (SNF) management is one of the key problems that determine prospects for nuclear power development around the world. About 17% of global electricity needs are satisfied by nuclear power plants (NPPs), and 65% of these are 1,000-MWt pressurized-water reactors (PWRs). When operated at 1,000 MWt, a PWR will produce about 21 t of SNF per year. According to independent evaluations, by the year 2010 about 400,000 t of SNF will have been accumulated around the world. Dealing with the problem of such a large quantity of radioactive material presents the biggest immediate challenge to nuclear power for the present and the future. It is well known that three variants of SNF management exist; they are: reprocessing, intermediate storage, and disposal as radioactive waste (RAW). Today not one country has completely resolved the problem of SNF management. Each country with NPPs defines, by itself, its national approach to SNF management, depending on its own specific technical, economic, and political conditions. The majority of countries have accepted the variant of “delayed decision” with regard to SNF; namely, a lengthy (30–100 years) controlled storage in special storage of the “dry” type. The variant of a delayed decision on SNF is also the national strategy in Ukraine. The first dry storage facility has been constructed at the Zaporoghskaya NPP for storage of its SNF. Recently the decision to construct a large central dry storage facility was accepted by all Ukrainian NPPs. The construction of this facility will be performed by the American company “Holtec”. It is necessary to note that in earlier times part of Ukrainian SNF was sent to Russia for chemical reprocessing. According to the agreement with Russia,
SPENT FUEL MANAGEMENT IN UKRAINE
149
and according to international practice, wastes from reprocessing Ukrainian SNF will be returned to Ukraine in a vitrified form (after 2010). Today in Ukraine more than 250,000 t of SNF are unloaded from NPPs; 90,000 t have been sent for reprocessing, and the remainder is in the state of temporary storage. In Ukraine, therefore, the tasks are well set for realizing the scientific– technological feasibility of temporary storage conditions for SNF in a central storage facility and of vitrified wastes in specialized storage. After temporary storage of SNF and wastes the problem of disposal of RAW in geologic structures will arise. In this regard the technology of radioactive materials encapsulation into protective forms suitable for geological disposal must be studied. Many institutions and scientific institutes including NSC KIPT are involved in solving these problems. Below are described the NSC KIPT activities in this field. 3 Development of Protective Materials for SNF Encapsulation 3.1 BACKGROUND The concept of permanent disposal of un-reprocessed SNF is under intensive investigation in many countries. A decision in favor of disposal will be based on conditioning techniques, in particular encapsulation of the SNF in a form suitable for deep geological emplacement. Radioactive high-level waste immobilization in natural mineral-like matrices is known to be an alternative to vitrification, i.e., incorporation into crystalline matrices, which are more stable in view of water resistance and radiation stability. A lot of leading research centers working in the field of RAW immobilization synthesize similar matrices with the obligatory use of hot pressing, or hot isostatic pressing (HIP). 1,2 The NSC KIPT is one of the biggest research centers in Ukraine, possessing qualified scientific and engineering personnel, a research base for tests of materials complying with IAEA standards, and up-to-date technological equipment, including HIP equipment and vacuum pressing facilities. All this allows fundamental and applied investigations into immobilization of RAW. The research activities are concerned with the selection of immobilizing substances, and comprehensive studies of their properties and compatibility. As immobilizing matrices the materials similar in composition to natural minerals are being studied. The research of solid phase synthesis of crystalline matrices-immobilizers of RAW with the aid of the HIP process has been conducted. The results of testing the obtained materials allows their characterization as being comparable with, or even exceeding, the well-known ceramic Synroc.3–5 These investigations have been carried out together with Russian specialists using HIP
150
SPENT FUEL MANAGEMENT IN UKRAINE
equipment developed in NSC KIPT and placed in a hot cell at the Radium Institute in St. Petersburg, Russia. For SNF isolation with the purpose of following interim storage with final disposal in geological formations, enclosing in an artificial mineral-like glass– ceramic monolith has been proposed. To obtain a material for the protective monolith we propose to use a powder mixture, as initial material for further treatment by compacting (cold pressing at room temperature), preliminary sintering and final sintering under pressure (HIP). The main components for the powder mixture in this case are natural granite together with the following additions: natural clays (to increase plasticity during compaction of the mixture of powders), stable mineral phases (which can include radionuclides in their structure, in the first place volatile Cs and Sr), and some chemicals (to increase radiation and chemical durability of the compositions). It is noteworthy that the proposed approach is expedient for isolating failed SNF elements whose chemical reprocessing is hardly possible. 3.2 WASTE FORM CHARACTERIZATION AND TECHNOLOGICAL APPROACH The treatment of SNF presents the most interest for reactor RBMK because of the large quantity of this kind of waste in the Chernobyl region. Also, encapsulating into stable and chemically durable monoliths of SNF in the shape of intact bundles, or cut into pieces, may be used for other kinds of reactors besides RBMKs. The scheme of the proposed technological approach and protective multibarrier system for SNF storage and final deep geological disposal is shown in Figure 1. The SNF assembly is divided into two fuel bundles and a connection rod by cutting the connection rod. The fuel bundles are placed in stainless steel capsules (∼4–5 mm thin-walled tubes with welded lower lids), filled with the protective material (powder mixture or sintered from powder mixture), dried and evacuated by pumping gas out through the metal capillary tube welded into the upper lid; the capillary tube is squeezed and finally sealed by welding the tube end. During further treatment by sintering under pressure by HIP, due to the triaxial action of the high pressure argon gas at the selected temperature, the powder in the hermetically sealed capsule is compacted to a completely porefree body. The resulting compacted glass–ceramic block prevents any release of radioactive materials outside the metal capsule, thus solving the important problem of immobilizing volatile fission products.
SPENT FUEL MANAGEMENT IN UKRAINE
151
Figure 1. Scheme for treatment of RBMK SNF
After HIP the capsules are loaded into stainless steel cans and sealed by capping, vacuuming, and welding. Two further steps are performed: (1) the cans are placed in standard canisters, which are sealed by capping, welding and inerting with argon, and the canisters housed in concrete vaults or modules for radiological protection for interim storage, and (2) the cans are placed in standard transport containers, which are transferred to the repository facility for final disposal.
152
SPENT FUEL MANAGEMENT IN UKRAINE
With this concept, based on the multiple barrier principle, an absolute isolation of the radioactive materials may be obtained for a very long time. 3.3 PHYSICAL AND CHEMICAL ASPECTS To manufacture the engineering protective barrier for SNF encapsulation, we propose to modify natural granite with the help of grinding; adding selected other components, compacting by cold pressing, and heat treating the green pellets by sintering under pressure. What does this result in? We have to decide two problems: (1) dense and strong protective ceramic, and (2) immobilizing volatile fission products which can pass from SNF during HIP-ing. It is known that the most attractive matrices for immobilizing radionuclides are synthetic materials, produced according to the principle of modeling complex mineral phases (hollandite, zirconalite, and perovskite)—the so-called Synroc composition—and natural minerals and rocks (granite, basalt, etc.), which have the ability to be isomorphously substituted and dissolved in the crystalline structure of some radionuclides. According to some recommendations, alumino-silicates, namely nepheline5 (a sodium alumino-silicate Na3K[AlSiO4]4), are proposed as the host for Cs. Attention should also be paid to other natural minerals, such as sphene (CaTiAlSiO8), and its cesium analogue leucite CsAlSi2O6, which can also incorporate Cs. Granite deposits of the Ukrainian crystalline shield are considered at the moment to be the most suitable for RAW repository. Thus it seems expedient to take granite rocks as the geological barrier and for producing a protective stable monolith to contain SNF. Here, the principle of phase compatibility keeping the natural balance is observed: the material of the protective monolith is similar in composition to that of the bedrock at the place of disposal. Natural granite, possessing chemical stability and lasting durability, is always characterized by a wide spectrum of grain size, some porosity and jointing. It can be regarded as its disadvantage if it is used as a geological barrier for disposal of RAWs, because the principle mechanism for the transport of radionuclides is migration along cracks and diffusion in water-filled pores. The modification of natural granite which we propose will allow, on the one hand, preserving the chemical composition of the mineral components (the most radiation and corrosionresistant phases are chosen, e.g., feldspar and quartz), and on the other hand, to obtain material with a fine grain structure that will completely exclude cracks and pores. As initial natural minerals, the granites of the Ukrainian crystalline shield were chosen with the composition of the main mineral phases: feldspars K(Na,Ca)AlSi3O8 up to (65–70) wt%, quartz (SiO2) up to 30 wt%, mica at a minimum. The use of additions of clay rocks ensures high plasticity of the
SPENT FUEL MANAGEMENT IN UKRAINE
153
synthesized composition and will encourage an increase in the corresponding material mechanical strength characteristics. Following published work, analysis of the physicochemical properties of sintering the mixture of natural granite rock and clay (in particular, kaolinite clay) taken in various weight proportions was conducted. Despite the wide range of components in granite–clay mixtures (5–70 wt% granite and clay for the rest), all of them according to the triple phase diagram CaO + K2O + Na2O)/ Al2O3/SiO2 move into the compositional area where the feldspars, mullite (3Al2O3⋅2SiO2) and alkaline-aluminosilicate glasses occur. So, in glass-ceramics, the crystalline phases, which existed before or are newly formed secondary ones, will provide a wide isomorphism for the alkaline ion Cs+ in the cation sublattice (e.g., replacing Na for Cs in K(Na,Ca)AlSi3O8). At the same time, the glassy multicomponent phase will be capable of absorbing part of the alkaline ions and creating a reliable immobilizing medium for volatile radionuclides, which can be released from UO2 during the HIP process at temperatures of about 1,000°C. 3.4 EXPERIMENTAL PROCEDURES In laboratory-scale researches, the possibility of manufacturing samples of the monolith simulator made of natural mineral-like compositions has been studied at the NSC KIPT. The main tasks were: • • • •
Manufacture of the glass–ceramic materials on the basis of natural granites and clays Optimization of parameters for sintering and sintering under pressure Determination of the main characteristics of the glass–ceramic materials proposed as protective engineering barriers for SNF encapsulation Grounding of terms of safe long-term storage and disposal of SNF contained in a glass–ceramic monolith
The technological parameters of obtaining compact glass–ceramic materials from the powder compositions of natural components (granite + kaolin clay) were optimized. The process of manufacturing glass–ceramic compositions included the following operations: • • • • • • •
Breaking granite into 1–5 mm size pieces Crushing the granite pieces in a ball mill to particles <200 µm in size Mixing with the clay powder and with glass-stabilizing additions Grinding the powder mixture to particles less than 100 µm in size Compacting by cold pressing at room temperature Preliminary sintering in air at ∼970°С Final sintering by HIP at 950–1,050°C and a pressure of 100–150 MPa
154
SPENT FUEL MANAGEMENT IN UKRAINE
Coming from physicochemical characteristics of SNF with Zr cladding on UO2 pellets, the maximum operating temperature is above 1,100°C. Research has proved that sintering the granite powders to dense material by solid phase synthesis with the help of HIP at temperature up to 1,000°C is not possible. Full sintering is possible only through the liquid phase at a temperature of ∼1,200°C. To solve this problem, glass-forming additions on the basis of natural clays were used to reduce the melting point, and sintering at 950–1,000°C during HIP was conducted. After the HIP treatment the metal capsules were cut with the help of a diamond-impregnated saw and samples were cut, which were used for further measurements and tests. As initial materials, the following natural components were used: granite from Ukrainian crystalline shield (Korostenian region), and kaolin clay. Glass–ceramic materials in the form of a glass-like matrix (40–45%), containing crystals of feldspar, mullite, and α-quartz are obtained. Crystalline grains are tied by matrix from glassy phase (Figure 2). As initial materials powder compositions (70% granite + 30% kaolin) are chosen as optimum.
Glassy matrix Crystalline phases
Crystalline phases: feldspar (Na(K,Ca)AlSi3O8), quartz (SiO2), and mullite (3Al2O3⋅2SiO2 ); Glassy matrix: (Al-Si-O) system Figure 2. Scheme of glass–ceramic composition
3.5 ANALYTICAL EQUIPMENT AND TESTS Methods of radiation materials science are based for analytical investigation and include: • •
Simulated irradiation of various materials Investigation of physical, mechanical, and chemical properties of the materials before and after irradiation
SPENT FUEL MANAGEMENT IN UKRAINE
155
The analysis of the structure of the material obtained and its phase composition is conducted with the help of scanning electronic microscopy (SEM), X-ray phase analysis, infra-red spectroscopy, and crystal optics. NSC KIPT experts carry out investigations of new perspective protective ceramic materials in cooperation with Kiev’s Institute of geochemistry of environment of NASU. The manufactured samples as simulators of materials for monolithic protective waste block, were characterized by density, compression strength, and corrosion rates. Characterization techniques include density measurements using method of direct hydrostatic weighing in liquid medium, defining maximum mechanical strength at axial compressing and accelerated 1-, 7-, 14-, and 28-day static leaching tests with standard methodology in deionized water by defining the total decrease of the sample mass. To study the effects of external γ-irradiation effect on the experimental samples, the bremsstrahlung of the electron linear accelerator was used. The scheme of irradiation is shown in Figure 3.
1
2 e-
3 γ,n
4 γ
eFigure 3. Scheme of γ-irradiation in electron accelerator: e–––electrons, n––neutrons, 1––electron accelerator, 2––Ta-converter, 3––filter for electrons and neutrons, 4––experimental sample
The beam of electrons with energy of 20 MeV while going through the tantalum converter generates a stream of γ-rays, i.e., bremsstrahlung. Neutrons, which appear in the converter (γ + 181Ta → 180Ta + 1n0) have an isotropic distribution. The further placed system of filters (aluminium and paraffin) reduces the electron and neutron components of the radiation beam, producing a practically pure stream of γ-rays with an average energy (by spectrum) of 2.0– 2.3 MeV. The maximum absorbed dose rate was ∼104 Gy/h.
SPENT FUEL MANAGEMENT IN UKRAINE
156
3.5 RESULTS OF ANALYTICAL INVESTIGATIONS AND TESTS It was established, that during sintering under pressure the dense glass–ceramic material is produced which consists of the crystalline phases (feldspar, quartz, mullite, α-tridymite (SiO2)) and the glassy phase. The quartz was found both in the form of primary phases and secondary ones in the nanometer size range (20–400 nm). Feldspar, mullite, and α-tridymite were found only as secondary phases. These phases were created as a result of processes of crystallizing in the glass phase during sintering under pressure. It was established that the glassy phase has a eutectic composition and is formed because of the availability of amorphous SiO2 and the dissolving of part of the crystalline phases. This results in a decrease in the content of quartz from 21 to 12 wt% and in complete dissolution of feldspar, the sizes of quartz grains being decreased from 40–50 to 15–25 µm. Results of tests on mechanical durability and leaching, and physical characteristics of modification of natural granite by sintering the powder mixtures under pressure are given in Table 1. TABLE 1. Physical characteristics of modified granite samples Powder mixture
Natural granite Granite + 30 wt% clay Granite + 30 wt% clay
Density (g/cm3)
Properties Compression strength (MPa)
–
2.6–2.7
100–160.0
Corrosion rate·(10–6 g/cm2 day) 1.3–1.8
950
2.45
100.0–120.0
1.2–1.5
1,050
2.65
160.0
1.0–1.5
Temperature of HIP (°C)
Investigations showed that the fabricated glass–ceramic material was characterized by high stability of chemical and phase compositions, high resistance to water, and a low porosity compared with the value in natural basalt. The corrosion rates for glass–ceramic compositions obtained during leaching tests were: (1.0–1.5) × 10–6 g/cm2d. The average value of the velocity of linear corrosion in water of the protective material made of the glass–ceramic composition determined experimentally was ∼15 µm per year. This allows use of glass–ceramic compositions effectively as an engineering barrier in the system of SNF deep geological disposal and increase the lifetime of the glass– ceramic monolith, in particular, up to 3,000 years with the layer thickness of
SPENT FUEL MANAGEMENT IN UKRAINE
157
∼40 mm. Our short-scale studies demonstrate the full potential of modified granite as additional defense barriers for using in the concept of SNF deep geological disposal. From the irradiation tests it was found that glass–ceramic compositions are characterized by appropriate mechanical stability (Figure 4).
Figure 4. SEM image: cracking is limited by glassy matrix, ×20,000; glass–ceramic composition: 30 wt% granite +70 wt% kaolin; γ-irradiation: absorbed dose of 1.2 × 107 Gy
With calculation it was shown that the absorbed dose of the glass–ceramic material by SNF irradiation over 1,000 years storage (interim controlled storage or deep geological disposal) following 10 years of preliminary cooling will be ∼3 × 108 Gy, which is 2–3 orders of magnitude less than the values corresponding to preserving radiation resistance and functional parameters of glasses and ceramics. Therefore, during the first 1,000 years, migration of radionuclides beyond the waste container will be limited to a large extent by the glass– ceramic barrier, whose structure and composition will not change much in the course of irradiation. The possible release of radionuclides by diffusion through glass–ceramic layer from the fuel elements, while in interim storage in air is estimated. Solving the task about spreading the radionuclides in the glass–ceramic medium can be brought together to the solution of the system of equations as: Ri
dCi d 2 Ci = Di − λ Ri C i , dt dx 2
(1)
158
SPENT FUEL MANAGEMENT IN UKRAINE
where i: index, corresponding to the number of the thin film of the protective material, whose boundaries are perpendicular to the stream of radionuclides; Ri: coefficient of radionuclide delay in the film i; Ci(x,t): concentration of the radionuclide at a distance x from the fuel element at time t; axis x is along the stream; Di: coefficient of radionuclide diffusion in the film I; λ: decay constant. The calculations are made taking into account the possible increase in the coefficients for radionuclide diffusion up to 10–14 m2/s as a result of SNF irradiation affecting the protective layer.6 Solving Eq. (1) showed that the protective barrier (layer about 40 mm, internal diameter of fuel bundle––80 mm) on the basis of the glass–ceramic composition ensures reliable isolation of the environment from release of radionuclides at the controlled near-surface interim storage. The summary relative release of radionuclides will be about 1% for the period of more than 400 years, and 10% over 1,000 years (Figure 5). For this period of time the radionuclides 90Sr and 137Cs will completely transform into stable 90Zr and 137Ba and decay of many transuranic elements will occur.
Figure 5. Relative release of radionuclides
Thus, the use of the additional glass–ceramic protective barrier (in a layer not less than 40 mm) in long-term interim storage will allow solution of the problem of “delayed decision” of managing SNF for a minimum period of 400 years. During this period the variant of the system to be used for final isolation of SNF can be chosen completely.
SPENT FUEL MANAGEMENT IN UKRAINE
159
4 Investigations of Deep Geological Disposal of SNF 4.1 INTRODUCTION It is known that geological disposal of RAW or SNF eventually will result in contact between groundwater and the RAWs upon failure of the disposal metal container (e.g., mechanical breach or metal corrosion). Therefore, studying the mechanisms of radionuclides release from SNF and the migration of radionuclides through granite (as the geological disposal medium) upon contact with aqueous solutions is now becoming important. The study of granites of Korostenian deposit from the Ukrainian crystalline shield is of most interest now because it is located near the “Shelter” unit and is being considered as a perspective place for disposal of long-lived RAWs in Ukraine.8 The results of the first experiments on investigation of the transport of 169 Yb in pristine and γ-irradiated (up to doses about 107 Gy) granite are presented in this article. 4.2 EXPERIMENTAL PROCEDURES A piece of Korostenian granite was cut into test specimens in the form of blocks 10 × 10 mm in cross section and 30 mm in length. Each block was covered with shellac adhesive on one surface. The following steps were used to study the migration of radionuclides as actinide simulators into granite matrix. Pellets of Yb2O3 were irradiated by bremsstrahlung obtained in an electron accelerator (Figure 3). In the course of irradiation the 169Yb tracer isotope was produced in accordance with reaction: 168 Yb (n,γ) 169Yb (half-life T1/2 = 30.7 days). The mass of ytterbium oxide pellet after irradiation was 0.1 g. Then the pellet was dissolved in concentrated HCl acid with the volume of 0.2 ml and finally a solution of pH = 1.8 was prepared. This solution (about 40 ml) was transferred into thermo-resistance flask supplied with reverse motion refrigerator. Experimental granite blocks were placed inside the flask. The flask was heated by water steam within 32 h. Then each granite block was washed in distilled water for 24 h and dried at 60°C. Layers were removed by precision grinding of the uncovered surface of the block (to avoid tracer loss). The thickness of removed layers varied from 2 to 50 µm. Material of the removed layer was used for γ-spectrometry with a Ge(Li)-detector. The 169Yb gamma intensity was measured for 20 min. Studying radionuclide migration was made with granite specimens in both natural pristine and γ-irradiated states. Irradiation was carried out up to absorbed doses of 0.3 × 107, 1.0 × 107, and 3.0 × 107 Gy; exposure times were
SPENT FUEL MANAGEMENT IN UKRAINE
160
respectively 7, 21, and 60 days.7 Realization of these severe conditions may be of greater interest to estimating chemical and radiation stability of the granite matrix under conditions of HLW disposal. The selected range of absorbed doses corresponds to international standards on tests of the radiation strength of the protective materials in geological repositories (in particular, materials to be exposed to absorbed doses ranging from 105 to 108 Gy). 4.3 RESULTS OF EXPERIMENTS AND DISCUSSION Characteristic γ-spectrums of tracer isotope 169Yb in removed layers were obtained. Thickness of removed layer was 2–50 µm. The absolute magnitudes of ytterbium concentrations on each removed layer of pristine and irradiated specimens were determined by calibrated measurements of γ-spectrums with the use of 137Cs as standard γ-emitted source. The maximum magnitude of ytterbium concentration was determined as 3.0 × 1020 at/cm3 that corresponded to the first layer (i.e., thickness 50 µm) of block irradiated up to external dose of 3.0 × 107 Gy (Figure 6).
20
concentration, *10 at/cm
3
10.00
a b c
1.00
0.10
0.01 0
200
400
depth, µ m
Figure 6. 169Yb concentration in granite specimens; a: after dose of 3.0 × 107 Gy, b: pristine state with pegmatite structure; c: pristine state with uniform grain structure
Curves of ytterbium concentration in various specimens of granite are shown in Figure 6 as function of matrix depth. To explain the character of the curves, one can argue as follows. The total diffusant mass in a matrix is known to be determined by both grain-volume diffusion and grain-boundary diffusion. The grain boundary diffusion is usually of importance for natural minerals.9 Besides this, structural defects (pores, cracks) may also enhance the penetration of the diffusant in the matrix. This means that the diffusant enters the rock matrix far from the surface mainly along grain boundaries (in the absence of
SPENT FUEL MANAGEMENT IN UKRAINE
161
obvious micro or macro defects). Both processes of grain-volume and grainboundary diffusion may be described by the following two Eqs. (2) and (3):9 C ( x, t ) =
2 ⎞ ⎛ C0 ⋅ exp⎜⎜ − x ⎟⎟ , 2πDt ⎝ 4 Dt ⎠
(2)
where C––tracer concentration; C0––initial tracer concentration; D––diffusion coefficient; and x––distance from specimen surface. C ( x, t ) = C ⋅ erfc 0
⎛ ⎜2 ⎝
⎞, ⎟ Dt ⎠
x
(3)
where
()
erfc x = 1 −
2
π u=
x
( )
⋅ ∫ exp −u 2 du , and 0
x . 2 Dt
Based on the above explanation, analysis of concentration profiles shown in Figure 6 was carried out. Two characteristic regions of 0–60 and 60–400 µm were revealed. The first one is in a good agreement with classic grain volume diffusion (penetration inside the bulk of grains). It was determined by crystaloptical analysis that the grains of minerals, forming the granite rock, i.e., plagioclase and feldspars, were exposed to interaction with tracer solution to a greater extent than grains of quartz. This may be explained by taking into consideration that in natural granite the quantity of microcracks in plagioclase and feldspars is higher than in quartz. Curves of profiles for natural unirradiated granite blocks (b, c in Figure 6) are satisfactorily described by Eq. (2). The diffusion coefficients were 1.4 × 10–15 and 1.09 × 10–15 m2/s, respectively. On the other hand, the curve of the ytterbium concentration profile in a granite block irradiated to a dose of 3.0 × 107 Gy was different from profiles in pristine blocks. In this case, the characteristic profile view of irradiated block (a in Figure 6) cannot be ascribed to grain-volume diffusion because Eq. (2) does not correspond to obtained experimental results. This can be explained by microstructure changes and accumulation of micro defects (pores, cracks) occurring during irradiation. Therefore, diffusion of ytterbium into deep granite block with depth up to 60 µm occurs mainly on these micro defects. To achieve mathematical accordance, the correct coefficients in Eq. (2) for estimating diffusion coefficients were introduced. Average magnitude of ytterbium diffusion coefficient on depth from surface 0–60 µm for specimen of irradiated block was 3.2 × 10–15 m2/s.
SPENT FUEL MANAGEMENT IN UKRAINE
162
In the range 60-400 µm the concentration profiles for all specimens (both pristine and irradiated) have similar nature, connected with grain boundary diffusion, which is described quite precisely by Eq. (3). It is noteworthy that in irradiated granite specimens up to doses more than 1.0 × 107 Gy, the diffusion coefficient was 10–12 m2/s, which is higher than for classic grain-boundary diffusion. Based on the results of crystal-optical analysis, one may suppose that it is connected with the modification of grains and change of condition of grain boundaries during irradiation. 4.4 FINAL COMMENTS In the study of radionuclide (actinide simulators) migration under conditions simulating the destruction of the metal disposal container in a geological repository for SNF, further interaction of SNF with groundwater and transport of leached actinides into environmental geological medium are proposed. To create the simulated γ-irradiation from SNF, bremsstrahlung from an electron accelerator was used. Penetration profiles of 169Yb (as actinide simulator) in specimens of natural granite rocks were obtained with the use of nuclear-physics methods. Diffusion characteristics for Yb-isotopes were estimated from the concentration profiles, taking into consideration the microstructure of granite rock. The observed diffusion coefficients for ytterbium were: for the depth 0 to 60 µm––about 1.09–1.4 × 10-15 m2/s concerning unirradiated specimens and 3.2 × 10–12 m2/s as to irradiated specimens; for the depth 60 to 400 µm––about 1.0–1.2 × 10–8 m2/s, which was practically the same for both kinds of specimen. Radiation beams were shown to influence the process of radionuclide migration in granite on account of both the changed internal structure of the rock matrix during irradiation and enhanced grain-volume diffusion. The magnitude of the characteristic dose of γ-irradiation which is required to increase radionuclide transport was determined to be 3.0 × 107 Gy. Because such high doses are hardly feasible from SNF, the release of actinides on account of their transport in granite rocks will depend primarily on the structure of natural granite itself. 5 Conclusions •
Glass–ceramic materials with an alumino-silicate structure have been made using natural components of granite and clay. These materials are suggested for SNF encapsulation in protective monolithic blocks for the purpose of long safe storage and subsequent geological isolation.
SPENT FUEL MANAGEMENT IN UKRAINE
•
• • • •
163
The parameters of a two-stage process for obtaining compact glass–ceramic materials from the powder compositions of the natural components of granite and clay have been optimized: preliminary sintering in air at 970°С followed by sintering under pressure by the method of HIP treatment at 920°С and 100 МPa. Irradiation of glass–ceramic samples with γ-rays up to doses ∼107 Gy causes little change in their density, structure or durability. The diffusion characteristics of granite in pristine condition and after γ-ray irradiation up to a dose 3 × 109 Gy were determined. The bremsstrahlung from electron accelerators are a useful way to generate γ-ray irradiation. To determine radionuclide concentrations in granite and other protective materials nuclear-physical methods of element analysis were developed in the NSC KIPT.
Acknowledgments The authors are grateful to Professor A.N. Dovbnya (NSC KIPT) for his advice and continuous interest in every aspect of this research. Our thanks are due also to Dr. V.L. Uvarov, Dr. N.P. Dikiy, Dr. E.P. Medvedeva, Dr. B.A. Shilyaev, Dr. S.V. Gabelkov, Dr. E.P. Shevyakova, and Mr. R.V. Tarasov for their assistance in diffusion studies, and Mr. Y.V. Lyazhko for his help in computer calculations. The basic ideas and technological decisions developed in NSC KIPT were recognized and approved by foreign experts. In 2000–2005 investigations were supported by two international grants, STCU #1580 and STCU #1761. References 1.
2.
3.
Hoenig, C.L., Larker, H.T., “Large Scale Densification of a Nuclear Waste Ceramic by Hot Isostatic Pressing”, Ceramic Bulletin, 62(12), 1983, pp. 1389– 1390. Anderson, E.B., Burakov, B.E., Vasilev, V.G. et al., “Methods of Synthesis of Crystalline Composite Materials for Immobilization of Long-Lived Radionuclides with the Purpose of Their Environmentally Safe Disposal”, Presented at International Symposium, Antwerp, Belgium, October 19–23, 1992. Sayenko, S.Y., Kantsedal, V.P., Tarasov, R.V., “Radioactive Waste Immobilization in Protective Ceramic Forms by the HIP Method at High Pressures”, Proceedings of the International Conference on Nuclear Waste Management and Environmental Remediation, Vol. 1, Prague, September 5–11, 1993, pp. 149–153.
164
4.
5.
6.
7.
8.
9.
SPENT FUEL MANAGEMENT IN UKRAINE
Sayenko, S.Y. et al., “Encapsulation of Spent Fuel Assemblies Using Defense Mineral-Like Barriers”, Proceedings of the International Conference on Radioactive Waste Disposal, Hamburg, Germany, September 1998, pp. 425–429. Oversby, V.M., “Ceramic Waste Forms for Fuel-Containing Masses at Chernobyl”, Proceedings of the International Topical Meeting “Spectrum-94”, Vol. 3, 1994, pp. 2065–2068. Sayenko, S.Y., Dikij, N.P., Uvarov, V.L., Shevyakova, E.P., “Studying the Influence of Simulated Gamma Irradiation on Radionuclide Migration in Granite Matrices”, submitted at the International Conference “Ecological Problems of Radioactive Wastes Disposal”, Kiev, Ukraine March 9–11, 2000. Dikiy, N.P., Sayenko, S.Y., Uvarov, V.L., Shevyakova, E.P., “Application of Nuclear-Physics Methods for Studying Radioactive Transport in Granite Rocks”, Problems of atomic science and technology, “Nuclear-Physics Investigations”, 2, 2000, pp. 54–57. Deconik, Z.M., Sabotovich, E.V., Shestopalov, V.M., Skvortcov, V.V., “Investigation of Possibility of Radioactive Waste Disposal in Deep Geological Formations (in Ukrainian)”, Bulletin of Ecological State of Chernobyl Zone, 13, 1999, pp. 64–66. Kaur, I., Gust, W., Fundamentals of Grain and Interphase Boundary Diffusion, Ziegler Press, Stuttgart, Germany, 1989.
TECHNICAL ISSUES OF WET AND DRY STORAGE
UNDERSTANDING AND MANAGING THE AGING OF SPENT FUEL AND FACILITY COMPONENTS IN WET STORAGE
A. B. JOHNSON, Jr.∗ Pacific Northwest National Laboratory Richland, Washington, USA Abstract: Storage has become the leading management option for spent reactor fuel and many storage facilities are operating longer than originally anticipated. Aging is a term that focuses attention on the consequences of such extended operation on the systems, structures, and components (SSCs) of storage facilities. The key to mitigating age-related degradation in storage facilities is to understand and manage aging of the facility materials. A systematic approach to preclude serious effects of age-related degradation is addressed in this paper, which mainly deals with test and research reactors (RRs). The first need is to assess the facility materials and their environments. Access to historical data on facility design, fabrication, and operation can facilitate assessment of expected materials performance. Methods to assess the current condition of facility materials are summarized in the paper. Each facility needs an aging management plan to define the scope of the management program, involving identification of the materials that need specific actions to manage age-related degradation. For each material, one or more aging management programs (AMPs) are developed and become part of the plan. Several national and international organizations have developed comprehensive approaches to aging management. A method developed by the US Nuclear Regulatory Commission (NRC) is recommended as a template for organizing measures to effectively manage age-related degradation of storage facility materials, including the scope of inspection, surveillance, and maintenance needed to ensure successful operation of the facility over its required life. Important to effective aging management is a staff alert to evidence of materials degradation and committed to carrying out the AMPs. Keywords: fuel storage pools, pitting corrosion, microbially influenced corrosion, intergranular stress corrosion cracking, aging management programs
______
∗ To whom correspondence should be addressed: A. B. Johnson, Jr., Pacific Northwest National Laboratory, Richland, Washington 99352, USA; e-mail:
[email protected]
167 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 167–180. © 2007 Springer.
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
168
1 Introduction Numerous nuclear fuel storage pools (FSPs) have been operated longer than originally intended. Circumstances that contribute to extended FSP operations include: delays in availability of permanent repositories, delays or cancellation of fuel reprocessing capacities, delays in fuel take-back programs, and, often, decisions to operate the associated reactors longer than initially intended. Aging is a term that has emerged to focus attention on the potential consequences of extended operation on the systems, structures, and components (SSCs) that comprise the facilities. Concepts related to aging are more readily understood with access to definitions of aging terminologies.1 Time-related degradation of materials that may occur in extended operation needs systematic consideration and mitigation. When facility staffs have not effectively managed effects of aging, materials failures have occurred that—in some cases—have caused extended and expensive recovery operations. In other FSPs, aging was detected in its early stages and was effectively mitigated. Storage facilities at nuclear power plants or large independent spent fuel storage installations are generally subject to comprehensive regulatory oversight and are staffed and funded to conduct comprehensive aging management actions. Smaller facilities have smaller staffs and smaller budgets to apply to aging management. However, by application of a systematic approach, outlined in this paper, aging management can be effective, providing a sound basis for extended operation, including relicensing when required. 2 Examples of Age-Related Degradation in Nuclear Wet Storage Facilities Predominantly, nuclear FSPs operate with minimal impacts of age-related degradation. Temperatures are low, conditions are generally quiescent, and high water purities are maintained in many facilities. However, case histories are referenced here that involved significant degradation of FSP materials, including fuel cladding. Materials that are prominent in FSPs are identified here, with references to cases illustrating degradation of materials during storage: • • • • • • • •
Pitting corrosion of aluminum (Al) alloys2–5 IGSCC of sensitized stainless steels4, 6 Corrosion of carbon steels (CSs)3, 4 Corrosion of copper (Cu) alloys3,4 Uranium metal corrosion3,7 Deterioration of some neutron absorbers8 Degradation of coatings9 Biological impacts on facility operations4,10,11
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
169
The key to effective aging management is to understand how to operate the facility to mitigate degradation of the materials over its intended life. The paper includes a concise summary of methods to mitigate age-related degradation of FSP materials. 3.1 UNDERSTANDING AGING Understanding aging of SSC materials involves systematic assessment of the materials and the environments that they are exposed to in a given facility. Essential to understanding aging issues in a given facility is to develop and implement an aging management plan, which is addressed in the section 3.2.1. Ideally, aging issues should be addressed during design and fabrication of the facility. Records of materials compositions and conditions should be archived for ready access. During operations, records should be maintained involving key data, such as water chemistries, operational problems and their solutions, maintenance methods, and schedules. 3.1.1 Considerations for understanding metal durability The envelope of successful operation for metals needs to be understood on the basis of the following parameters: • • • •
Metal composition Metallurgical condition, including effects of thermal treatments and irradiation where relevant Stress levels Range of environments, including thermal, chemical, galvanic, hydraulic, and radiation, as appropriate. Sludge deposits can accelerate corrosion where they are in contact with materials. Biological species, notably algae, have developed in some FSPs
3.1.2 Condition assessment for FSP materials Methods to assess the condition of materials during FSP operation include: • • • •
Visual observations, including walk-downs of accessible equipment; inspections of pool materials with binoculars, boroscopes, video, etc. Nondestructive examinations Sipping of fuel elements to detect cladding breaches Destructive metallurgical examinations (limited due to expense for irradiated materials)
170
• • • • • • • •
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
Monitoring the condition of coupons of typical materials in actual facility environments Electrochemical monitors for corrosion and water chemistry assessment Monitoring pool water for impurities and radioisotopes Targets-of-opportunity, e.g., inspection/sampling of components removed from service or during maintenance operations Consideration of components that include galvanic couples or high radiation fields that involve potentially vulnerable materials Awareness of pool water levels to monitor for leakage Consideration of hard-to-access locations that may have vulnerabilities, such as buried piping Awareness of materials behavior in similar facilities and from the technical literature
The scope of condition assessment should be adapted to the complexity and operational history of a given facility. 3.1.3 Case histories Examples of problems that developed in wet storage facilities are given here. The value of coupon monitoring is illustrated in Figures 1 and 2. The wet storage facility included Al alloys, Cu alloys, CSs, stainless steels (SSs), and Zircaloy (Zry).3 Coupons of each material were inserted into the pool water. After 4 years of operation, the pool water was dosed with a biocide that was not appropriate for the mix of materials. The pool water turned blue, signaling an abnormal corrosion condition. The fact that the coupons were in the water before, during, and after the chemical excursion provided a basis to monitor the extent of accelerated corrosion on each material (illustrated in Figures 1 and 2 for an Al alloy and CS). The Al alloy specimens reflect return to normal corrosion rates about 2 years after onset of the chemical excursion in the 105-KW FSP. The CS specimens were exposed in the 105-KE FSP, where they were subject to an additional chemical excursion related to a major fuel handling campaign. Sufficient coupons were available to follow the course of the corrosion. The International Atomic Energy Agency (IAEA) sponsored a broad international study of the value of coupons to monitor Al alloy behavior in FSPs.5 In another wet storage case, changes in water level were interpreted to indicate leakage of pool water in the Omega West Reactor (OWR).10 By selective valving, the location of the leak was identified. The leakage was in an underground line, requiring excavation. The pipe failure was attributed to microbial attack on the outside of the SS pipe. Failure of the pipe was the principal motivation for shutdown of the reactor. Assessing the condition of underground piping and concrete presents formidable challenges, particularly in locations
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
171
with moist, aggressive soil conditions. Occasional exploratory excavations may be justified. Behavior of concrete in FSPs has been satisfactory in a wide range of environments. The leading consideration for concrete structures is that the integrity, including effects of aging, is consistent with seismic conditions where they are sited.
Figure 1. Uniform corrosion rates for 5086 aluminum3
Figure 2. Uniform corrosion rates for carbon steel3
172
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
3.1.4 Durability characteristics of FSP materials
The following sections address durability characteristics of FSP materials. Aluminum alloys: Several Al alloys have functioned well in many FSPs, in some cases for more than four decades. In other FSPs, Al alloys have degraded in a few months in aggressive water chemistries.2,4 Figure 3 illustrates an advanced pitting attack on Al-alloy fuel storage canisters. Wet storage facilities that have maintained water conductivities below ∼10 µS/cm have reported favorable performance of Al alloys for up to 25 years,5 although 1 µS/cm is a preferable target. Keeping impurities such as chloride, mercury, copper, and other heavy metals at low levels facilitates Al alloy longevity. However, pitting corrosion has occurred, even in high purity waters, due to deposition of iron oxide particles.5 Pitting corrosion is the leading degradation mechanism for Al alloys in FSP waters. Crevice corrosion and galvanic corrosion also have been observed, particularly in aggressive water conditions. Compact oxide films promote durability of Al alloys in FSP environments. Figure 4 involves Al-clad fuel elements after 25 years in FSP water that caused pitting on unfilmed Al alloy components (Figure 3). Surface films of boehmite formed during reactor service offer an explanation for the fuel element durability. However, damage to these films caused, for example, by scratching (Figure 5) can initiate local corrosion, as shown in work performed at the Savannah River Site.2
Figure 3. Pitting corrosion of Al-alloy fuel storage canisters after 31.5 years in filtered Columbia River water and 10 years in deionized water3,4
Figure 4. Durable Al-clad U fuel elements stored in sometimes aggressive water (as in Figure 3); stability attributed to a protective oxide film formed in the reactor3
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
173
Figure 5. Pit initiation (arrowed) on Al alloy cladding where oxide layer was damaged by scratching2
Stainless steels: SSs are prominent in FSP components. In general, they have had excellent durability for up to nearly 50 years in some facilities. Degradation of SS components by intergranular stress corrosion cracking (IGSCC) has occurred, however, even at low temperatures (30–50ºC) in water, due to a combination of sensitized microstructures, stresses, and degraded water quality,4 as illustrated in Figure 6.
Figure 6. Through-wall penetration of IGSCC crack in the heat affected zone of a weld which was sensitized in the weld procedure
Carbon steels: CSs are susceptible to rust formation in oxygenated water under FSP conditions. Corrosion is mitigated in high purity waters, but over extended periods, reddish brown corrosion products contribute to sludge inventtories and deposition on other components.3,4 Some facilities have successfully involved CSs in piping and even FSP liners by use of organic coatings that are effectively maintained. Copper alloys: Cu alloys are sometimes installed in heat exchangers and in peripheral applications in FSPs. There are concerns that Cu alloys and Al alloys should not exist in the same systems, although there are cases where they have coexisted when high water purity was maintained.3,4 In degraded water, corrosion was accelerated on both materials.
174
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
Uranium metal: Uranium metal has been widely used in defense, test, and research reactors (RRs). Because it is susceptible to aqueous corrosion, even at low temperatures, the integrity of the cladding is vital to durability of the metal fuel. Where damage to the cladding occurred during discharge from the reactor, exposing irradiated uranium,3 corrosion has been monitored by measuring 137Cs activities released to the FSP water, as shown in Figure 7. The leach rate of 137 Cs increased with water temperature and conductivity.
Figure 7. Leaching of 137Cs from uranium fuel in the Hanford K-Basin during three temperature cycles, showing effects of FSP water temperature and conductivity on corrosion of uranium exposed at cladding defects3
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
175
Concrete: Concrete FSP structures have a successful history of durability. Rarely, problems have arisen when the concrete was not formulated properly or pipe penetrations were not engineered appropriately. The characteristics of concrete for fuel storage applications have been addressed.9,12 Neutron absorbers: Organic materials are used in some neutron absorbers in fuel storage racks. As the radiation stability limit of the materials is approached, the materials become degraded and the neutron absorption properties are diminished. The GALL report13 includes guidance for an aging management program to assure that satisfactory Boroflex criticality control properties are maintained. Organic coatings: Organic coatings are applied to protect corrosion-prone materials, such as CS. They are also applied to the concrete walls in some FSPs, instead of installing a metal liner. With proper maintenance, organic coatings have performed well in FSPs for periods of more than 40 years. 3.1.5 Biological impacts in wet storage facilities Effects of biological species in FSPs include infestations of algae and microbially influenced corrosion (MIC). Algae can interfere with visibility in FSP waters and can promote corrosion if growths develop on metal surfaces, for example, due to differential oxygen cells. Operators of several US test reactors have reported on methods that were successful to suppress algae and other biological species.11 Measurements at the Savannah River Site indicate that the radiation threshold to suppress the growth of algae is about 25 R/h.12 Evidence for MIC attack on metals in FSPs has been very limited,11,14 although a case was cited earlier involving failure of an underground pipe by this mechanism.10 3.1.6 Decommissioning as an opportunity to assess aging of FSP materials A potentially valuable opportunity to assess aging of materials is available when a FSP is decommissioned. Parts of the facility not accessible during operations can be investigated, for example, concrete, including reinforcing bars. However, the value of the opportunity accrues when expert observers are provided with resources to accomplish specified examinations, which are then documented and become part of the technical literature. 3.2 MANAGEMENT OF AGING Effective aging management begins in design and construction phases of a FSP, though details are beyond the scope of this paper. In each facility, an aging management plan provides the basis for effective aging management. Two of
176
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
the major efforts to develop systematic approaches to aging management were mounted by the IAEA and by the US Nuclear Regulatory Commission (USNRC). 3.2.1 Aging management plan It is recommended that each storage facility have an aging management plan. The plan involves identification of the SSCs that need specific actions to mitigate age-related degradation. Also, the plan identifies aging management programs (AMPs) that are to be applied for each SSC. Development of the plan is facilitated by access to the following information and considerations: • •
• •
• • • •
•
Facility design and construction information relevant to aging of SSC materials. Identification of all materials in the facility, including alloy compositions, metallurgical conditions, and fabrication histories to the extent available. If archive materials are available, their location and inventory should be identified. Storage conditions should be reviewed to assure minimal deterioration during storage. Recognition of materials interactions, involving galvanic effects where significant. Assessment of the water chemistry history and current conditions to understand whether there have been periods that may have adversely affected any materials. Are specifications consistent with maintaining durability of all exposed materials? Is the sampling frequency consistent with effective water chemistry control? Are all needed parameters addressed in the sampling program? Are water chemistry measurements consistent with specifications? If not, what actions are needed to restore satisfactory control? Consideration of radiation sources and doses. Are they incident on any materials that are susceptible to radiation-induced degradation? Consideration of sludge building up on the bottom of the pool? If significant, what is the potential effect on exposed materials? Is action needed to mitigate sludge buildup, e.g., vacuuming. Is the pool water level stable, consistent with evaporation losses? If not, what action is needed to investigate the prospect of a leak? Identification of locations where aging of materials may be anticipated but difficult to access, such as buried piping and concrete rebar. Are AMPs needed, for example, excavation to assess pipe condition if water loss suggests leakage of pool water? Surveillance for evidence of biological impacts, including algae and MIC.
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
177
3.2.2 Alternative templates for aging management in FSPs The IAEA initiatives focus on major NPP components, such as reactor pressure vessels and PWR steam generators. The merit of the IAEA aging analysis method is that it integrates the essentials of understanding and managing aging.15 The drawback for the fuel storage application is that the IAEA initiative does not specifically address aging management issues for FSPs. Major guidance for the USNRC method is provided in the Generic Aging Lessons Learned (GALL) report.13 The report describes a template that is applied to all safety related reactor components, including those in fuel storage facilities. The template integrates the essentials of understanding and managing aging for each component on one page, including identifying the AMPs for the component. The template is given in Table 1 for components in BWR FSPs with high-purity water chemistry. TABLE 1. Spent fuel pool cooling and cleanup (boiling water reactor) Structure
Pipe, pipe fittings, and flanges
Filter housing
Filter elastomer lining
Material
Stainless steel
Stainless steel, carbon steel with elastomer lining, or SS cladding Same
Elastomers
Environment
Chemically treated oxygenated water up to 50°C Aging effect/ Material loss, mechanism pitting/crevice corrosion
Same
Valves (check hand valves), body, and bonnet Stainless steel, carbon steel with elastomer lining, or SS cladding Same
Hardening, Material loss, cracking/elastomer pitting/crevice degradation corrosion (only for CS after lining/cladding degradation) A plant-specific Aging Chapter Chapter AMP to assess the XI.M2 Water Management XI.M2 Water qualified life of Program * Chemistry for chemistry for linings in the BWR water BWR water in environment is to In BWRVIPBWRVIP-29 be evaluated 29 [EPRI TR[EPRI TR103515] 103515] * To be adapted to materials/environments in each FSP Material loss, pitting/crevice corrosion (only for CS after lining/cladding degradation) Same augment AMP if water chemistry is control effective
178
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
Facility operators must adapt the template to their FSP materials and conditions. For each component/material, one or more AMPs are identified in the column with that heading; section 3.2.3 describes the basis for the AMPs. 3.2.3 Basis to develop Aging Management Programs for FSP materials The GALL report identifies ten attributes of an effective aging management program: Scope, Preventive Actions, Parameters Monitored or Inspected, Detection of Aging Effects, Monitoring and Trending, Acceptance Criteria, Corrective Actions, Confirmation Process, Administrative Controls, Operating Experience. The development of an AMP for Al alloys is available.9 For components exposed in FSP water, Water Chemistry is an obvious program. Facility operators will use judgment to decide the extent that each of the ten attributes applies to their facility. The application should be no more complex than is required to assure that aging degradation of facility materials is effectively managed. Detection of Aging Effects could involve periodic visual inspections of fuel elements and/or components or it could involve insertion of coupons that are examined periodically. If dosing of pool water with biocides is to be conducted, a decision is needed regarding the scope of monitoring that is needed to assure that materials in the FSP are not damaged. A case of accelerated corrosion of FSP materials due to an aggressive biocide was referenced in section 3.1.3. Several US RRs have successfully conducted campaigns to control algae and other biological species.11 For equipment outside the pool water, exposed to air, a Walkdown Program offers a method to monitor for possible evidence of aging. The program should involve a protocol that identifies the equipment to be observed and the specific observations to be made. Depending on the level of concern, possibly prompted by operating experience in this or other similar facilities, nondestructive or other programs may be applied. 3.2.4 The role of Maintenance in Effective Aging Management Effective maintenance is essential to mitigate degradation of materials. Included under the maintenance umbrella are: preventive, periodic, planned, predictive, and corrective maintenance and refurbishment and repair.1 In a benign FSP, the maintenance program requirements are less stringent than for safety-related equipment operating under more demanding conditions. Even so, maintenance is a key element of effective aging management.
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
179
3.2.5 Targets-of-opportunity—value in aging management The concept of targets-of-opportunity was identified in section 3.1, involving observations of component/material condition during maintenance or removal. Examples are removing an aluminum storage rack, allowing inspection for signs of degradation; and maintenance on a SS valve and its weld, offering the prospect of observing if IGSCC is progressing. 3.2.6 Importance of attentive staff Important to effective aging management in any facility is a staff which is alert to evidence of materials degradation and which is committed to conducting AMPs in a disciplined manner. 4 Summary • • • • • • •
Nuclear fuel storage facilities are often required to operate longer than expected, requiring increased attention to understanding and managing the age-related degradation of the materials of their SSCs. Each FSP facility should have an aging management plan. The plan identifies SSC materials that need aging management. It is necessary to understand what is needed for effective environmental control to mitigate aging of SCC materials. The plan should involve development of AMPs for each SSC material that is identified. Management cooperation and necessary resources are very important. Facility staff should be alert for evidence of age-related degradation and be committed to effective aging management.
References 1. 2.
3. 4.
Glossary of Nuclear Power Plant Aging, ISBN 92-64-05842-7. Nuclear Energy Agency, Paris, France (1999). Howell, J.P., “Corrosion of Aluminum-Clad Alloys in Wet Spent Fuel Storage”, WSRC-TR-0343, Westinghouse Savannah River Technology Center, Aiken, South Carolina, USA (1995). Johnson, A.B., Jr., Burke, S.P., “K Basin Corrosion Program Report”, WHC-EP0877, Westinghouse Hanford Company, Richland, Washington (1995). “Durability of Spent Nuclear Fuels and Facility Components in Wet Storage”, IAEA-TECDOC-1012, International Atomic Energy Agency, Vienna, Austria (1998).
180
AGING OF FUEL AND POOL COMPONENTS IN WET STORAGE
5.
“Corrosion of Research Reactor Aluminium Clad Spent Fuel in Water”, Technical Reports Series No. 418, International Atomic Energy Agency, Vienna, Austria (2003). “Effects of radiation and environmental factors on the durability of materials in spent fuel storage and disposal”, IAEA-TECDOC-1316, International Atomic Energy Agency, Vienna, Austria (2002). Nelson, D.Z., Howell, J.P., “Metallography of Pitted Aluminum-Clad Depleted Uranium Fuel.” CORROSION95 Paper No. 430, NACE International, Houston, Texas (1995). “Spent Fuel Performance Assessment and Research”, IAEA-TECDOC-1343, International Atomic Energy Agency, Vienna, Austria (2003). “Understanding and Managing Aging of Materials in Spent Fuel Storage Facilities”, Technical Report Series 443, International Atomic Energy Agency, Vienna, Austria (in press). Andrade, A., “Stress-assisted Microbial-induced Corrosion of Stainless Steel Piping and other Aging Issues at the OMEGA West Reactor.” GKSS 95/E/51, Proceedings of the International Topical Seminar on Management of Aging of Research Reactors, Geesthacht, Hamburg, May 1995. Published by the International Atomic Energy Agency, Vienna, Austria (1995). Johnson, A.B., Jr., Burke, S.P., “General and Specific Perspectives on Biological Impacts in Wet Storage Facilities for Irradiated Fuel”, J.H. Wolfram, et al. (eds.), Microbiological Degradation in Radioactive Waste Repository and Nuclear Fuel Storage Areas, pp. 103–112. Kluwer Academic, Dordrecht, The Netherlands (1997). “Assessment and management of aging of major nuclear power plant components important to safety: concrete containment buildings”, IAEA-TECDOC-1025, International Atomic Energy Agency, Vienna, Austria (1998). “Generic Aging Lessons Learned (GALL)”, NUREG-1801, Vols 1 and 2, US Nuclear Regulatory Commission, Washington, DC (2001). Louthan, M.R., Jr., “The Potential for Microbiologically Influenced Corrosion in the Savannah River Spent Fuel Storage Pools.” J.H. Wolfram, et al. (eds.), Microbiological Degradation in Radioactive Waste Repository and Nuclear Fuel Storage Areas, Kluwer Academic, Dordrecht, The Netherlands (1997). “Methodology for the Management of Aging of Nuclear Power Plant Components Important to Safety”, Technical Report Series No. 338. International Atomic Energy Agency, Vienna, Austria (1992).
6.
7.
8. 9.
10.
11.
12.
13. 14.
15.
LONG-TERM (100–300 YEARS) INTERIM DRY STORAGE FOR SPENT FUEL: PACKAGE AND FACILITIES DEVELOPMENT INCLUDING SAFETY ASPECTS AND DURABILITY ASSESSMENT PROGRAM
J.-P. SILVY,* N. MOULIN, AND F. LAURENT CEA DEN/Valrhô, BP 17171 30207 Bagnols sur Cèze Cedex, France Abstract: In the framework of the French radioactive waste management act, the Commissariat à L’énergie Atomique (CEA) has carried out a major research program on long-term interim storage of intermediate-level long-lived waste, high-level waste, and spent fuel [1]. This paper describes work on the long-term interim storage of spent fuel: container development, design of surface and subsurface storage facilities, durability and functional performance demonstrations. Section 1 describes the long-term design and safety principles, section 2 covers long-term interim storage concepts, and section 3 discusses the development and demonstration program. Keywords: spent fuel, dry storage, container, cast iron, durability, underground facility, thermohydraulics
1 Safety and Design Principles The notion of “long-term” interim storage contrasts with that of existing Industrial facilities designed for several decades of operation. The results obtained to date with these facilities demonstrate that safe operation can be extended to a hundred years. Long-term operation concerns several generations, and the design must take into account the consequences of this duration, as discussed below. The CEA has selected a period ranging from 100 to 300 years for these studies. It must be emphasized that an interim storage facility is intended only for temporary use, not as a final disposal site.
______ * To whom correspondence should be addressed: J. P. Silvy, CEA DEN/Valrhô, BP 17171, 30207 Bagnols sur Cèze Cedex, France; e-mail:
[email protected]
181 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 181–188. © 2007 Springer.
182
R&D ON LONG-TERM SPENT FUEL STORAGE IN FRANCE
The perspective of this duration requires a technical and scientific validation not only of the constituents of the facility but also of the societal context, which is liable to change during this period. The CEA has therefore defined a novel “loss of technical control” scenario under which the safety of the facility must be demonstrated. This scenario assumes that society continues to conserve responsibility for the facility, but no longer ensures maintenance and limits its surveillance to the presence of security guards for a period of up to 10 years. The following safety and design principles have been defined to comply with these requirements: • • • •
Allowance for aging from the design stage by specifying components with known degradation behavior over time. Allowance for surveillance and maintenance requirements from the initial design stage. Limitation of the burden on future generations. Robustness with respect to contingencies: – Design and construction using components as simple and passive as technically feasible; “passive” refers to components operating naturally without moving parts. – Maximum inertia in the facility, allowing an extended time interval between the occurrence of a malfunction and the implementation of remedial action to restore nominal operating conditions.
In compliance with these principles, the CEA organized its research and development (R&D) program in three areas: • •
•
Control of component aging in a long-term interim storage facility. The degradation behavior of the constituents over time must be determined and modeled; this is the research aspect of the CEA studies. Qualification and performance demonstration of the processes that will be used for the construction of the long-term interim storage facility; this is the development work, consisting mainly in the construction of demonstrators and the performance of integrated tests. Designing the facilities and equipment based on the results of the research and development work to quantify indicators related to safety, cost, and durability.
The design results and demonstration program are discussed below for the container and for the interim storage facility.
R&D ON LONG-TERM SPENT FUEL STORAGE IN FRANCE
183
2 Long-Term Interim Storage Concepts As mentioned above, the selected design involves passive operation. This concerns the radioactive containment and the removal of thermal power from the spent fuel. Passive containment is obtained using a double-wall metal container that provides the two barriers necessary for safety purposes. Concerning heat transfer, a dry interim storage facility was selected with natural convection cooling. Two types of layout were examined: a surface facility, or bunker, in which the canisters are stored vertically beneath a slab, and a subsurface facility located on a hillside, in which tunnels and dry interim storage shafts are excavated. The studies were carried out for generic sites for which parameters were defined with reasonable and plausible physical values with regard to the risk of external aggression. The generic sites assume hard rock (granite or limestone), a fault network, and a position above the water table. Figure 1 shows a limestone generic site.
Figure 1. Limestone generic site considered in design
2.1 CANISTER CONCEPT As the leak tightness of the fuel cladding cannot be demonstrated over secular time periods, each spent fuel assembly is placed in an individual leak tight canister that constitutes the primary containment barrier. The canister is made of stainless steel sealed by TIG welding. Helium is used for the atmosphere inside the canister for effective transfer of the thermal power released from the fuel. The second containment barrier is formed by a metal container designed to accommodate several canisters. The design studies addressed the materials, the mechanical and thermal properties, corrosion resistance and subcriticality of the package. The container will be made of carbon steel or cast iron, sealed by full-thickness welding (several variants have been tested). It will contain 7 UOX canisters or
184
R&D ON LONG-TERM SPENT FUEL STORAGE IN FRANCE
4 MOX canisters (thermally equivalent). Cast iron is the most suitable metal: it ensures a homogeneous container with an integral bottom plate, and the manufacturing cost is 3–4 times lower than for rolled or welded carbon steel. A drawback is that it is impossible to weld cast iron thicknesses of a few centimeters. The CEA has overcome this difficulty by developing a process in which a carbon steel rim insert is metallurgical bonded to the top of the container. The cover is welded directly to the rim, as shown in Figure 2.
Figure 2. Metal container for secondary containment of canisters
2.2 INTERIM STORAGE CONCEPTS 2.2.1 Surface facility Following an examination of the preliminary concepts, the selected surface interim storage concept consists of dry interim storage in a semi-underground bunker. The facility comprises two separate areas: an interface zone for spent fuel entry and loading into containers, and the actual interim storage zone comprising a series of independent bunkers. A modular solution was preferred to allow flexibility with respect to inventory and capital costs. Figure 3 shows a cross-sectional view of a bunker. Heat removal from the packages is provided by natural ventilation that maintains the humidity at a level guaranteeing a very low container corrosion rate. In this type of cooling, transverse airflow predominates over convective updrafts. The airflow remains stable in time and space. There are a few recirculation or stagnant zones; the airflow can Figure 3. Surface interim storage be considered equivalent to forced convection and can therefore be modeled.
R&D ON LONG-TERM SPENT FUEL STORAGE IN FRANCE
185
Remote operation was selected for normal container handling, although the possibility of nonremote operation was considered in the event of unexpected or degraded conditions. 2.2.2 Subsurface facility The underground portion of the facility is excavated into a hillside to allow horizontal access and permit water runoff by gravity flow. The overall architecture calls for vertical storage of the packages in naturally ventilated shafts. Figure 4 shows a cross section of the shafts together with the lower cool air inlet and upper hot air exhaust tunnels. Two packages are stored in each shaft. As in the surface concept, a modular design is used. The cooling airflow rate necessary in each shaft determines the number of shafts per tunnel and number of tunnels per module, so as to limit the airflow velocity and tunnel diameters. Figure 5 shows an overall view of the facility. Handling is again carried out remotely under normal conditions in order to minimize personnel exposure to radiation.
Figure 4. Subsurface storage facility
Module interface (accueil des châteaux, conditionnement et surveillance)
Modules d’entreposage
Modules d’entreposage regroupant 6 galeries de 120 puits
1100 m
1500 m
Figure 5. Overview of subsurface facility
3 Development Programs The objective of the development programs is to specify qualification procedures for the processes that will be used to produce the components of a long-term interim storage facility. Qualification concerns not only the manufacturing operations but also aspects related to durability under interim storage conditions. The CEA has therefore produced representative demonstrators of the containers and of the storage facility subassemblies. The demonstration is supported by a
186
R&D ON LONG-TERM SPENT FUEL STORAGE IN FRANCE
container development program including the fabrication of full-scale prototypes, qualification and durability testing, and programs covering the facilities: • •
Thermo-hydraulic test facility representative of a storage bunker (VALIDA experiment), borehole (PROMETHEE), and tunnel network (SIGAL) Full-scale concrete infrastructure subjected to mechanical and thermal loading (GALATEE experiments now being set up)
3.1 DEVELOPMENT OF A SPENT FUEL INTERIM STORAGE CONTAINER As noted above, the fuel assemblies are inserted into individual stainless steel canisters that are placed in carbon steel or cast iron containers (materials subject to uniform corrosion). The fabrication and closure processes considered for the industrial implementation and a full-scale demonstrator were manufactured according to the refeFigure 6. Full-scale demonstrator of interim rence solution (cast iron with storage container an annular steel insert and electron-beam welding). The full-scale demonstrator is shown in Figure 6. The performance demonstration consists in characterizing these objects, carrying out drop tests on mockups (November 2005), and confirming their durability. The main factors liable to affect container aging are corrosion and heat transfer. With regard to corrosion, a simplified approach in compliance with applicable standards yielded an estimated total metal thickness of less than 350 µm consumed after 300 years on a container 45 mm thick. Predictive corrosion modeling and a systematic study of historical analog materials are now in progress. Thermal aging can modify the structural properties of the metal components and welds. Studies are now in progress to confirm the effect of aging on the container airtightness and integrity. This basically involves an aging test program lasting one, three, five and ten years on representative samples and mockups. The initial condition was examined and characterization examinations are repeated over time to check for any significant variations. The tests began in 2002 and the initial results confirm the predictions.
R&D ON LONG-TERM SPENT FUEL STORAGE IN FRANCE
187
3.2 THERMOHYDRAULIC TESTING 3.2.1 VALIDA This test facility determines the experimental convection coefficients in a container room for the design of long term-surface interim storage bunkers. It comprises a strong turbulence airflow test section at representative scale. The models have been validated by characterization data on the flow velocity and temperature profiles in the specific range of interaction between natural vertical convection along the packages and the transverse airflow, and between the impact of the hot exhaust air and the ceiling. 3.2.2 PROMETHEE This program was initiated because of the lack of published data in this area. It consisted in a full-scale investigation of the thermal aspects of natural convection in a subsurface dry interim storage shaft. The test facility included a heating element representing a container, an enclosure corresponding to the shaft, and a system of measuring instruments (200 thermocouples and two flowmeters). This experiment allowed us to assess convective heat transfer between the container and the air inside the shaft. The results were incorporated in models simulating the behavior of packages in interim storage shafts. 3.2.3 SIGAL Very high singular pressure losses occur in the tunnels of a subsurface interim storage facility, and no published data are available concerning the geometric characteristics of such tunnels. SIGAL is a 1/20 scale mockup ensuring functional Reynolds similarity, used to determine the correlations necessary to model singular pressure losses in the air supply and distribution lines of the interim storage facilities, as well as in distribution bends and tee junctions. 3.3 INFRASTRUCTURE BEHAVIOR 3.3.1 GALATEE experiment GALATEE is a tunnel segment of a subsurface long-term interim storage facility. It was built not only as an illustration of the system, but also to demonstrate the long-term performance of concrete structures subjected to thermal and mechanical stresses representative of an interim storage facility. The initial experiment is now being set up (Figure 7). The arch temperature will be about 80°C, and corresponds to the maximum beginning-of-life temperature of the facility.
R&D ON LONG-TERM SPENT FUEL STORAGE IN FRANCE
188
Figure 7. The GALATEE mockup of an interim storage facility
Temperature and humidity sensors have been installed in the concrete structure. Sensor readings during the experiments and some dimensional measurements before and after the experiment will be used to monitor variations in the infrastructure. 4 Conclusion CEA studies of long-term interim storage of spent fuel are supported by an ambitious program on the behavior of the components of an interim storage facility, as well as on representative technological demonstrations. Some of this work is still in progress but the initial results have confirmed the predictive models. The data available today are sufficient to conclude that a long-term interim storage facility is technically feasible. References 1. CEA Scientific Report “Les Déchets Radioactifs à Haute Activité et à Vie Longue” — Loi du 30 Décembre 1991—Axe 3 Conditionnement et entreposage de longue durée CEA/DEN/DDIN/2004-643, February 2005.
TECHNICAL ISSUES OF WET AND DRY STORAGE FACILITIES FOR SPENT NUCLEAR FUEL E. D. FEDEROVICH∗ I. I. Poluzunov Boiler-Turbine Research and Design Institute St. Petersburg, Russia Abstract: The satisfactory removal of residual heat from spent fuel assemblies (SFAs) in wet and dry storage facilities is an all-important safety provision. To this end, heat–mass transfer processes should be studied in detail for both normal and accident conditions. Examples of modeling and experiments carried out to this end at the State Polytechnic University and the Central BoilerTurbine Research and Design Institute in St. Petersburg, Russia, are presented. They include: mathematical modeling of nonstationary accident processes (like pool drying in a wet storage facility); experimental studies of mixed convection in a small-scale pool model, and natural convection in a simulated bundle of spent fuel rods; experiments on heat transfer and pressure drops in vertical SFA casks, and on natural convection in horizontal casks (simulating SFA transport conditions); and, finally, experiments and modeling of air flows and heat–mass transfer in various dry-storage configurations. The utility of such work is in the possibility of defining the temperature regimes of components in storage facilities, including critical ones like fuelrod cladding and cask seals, for normal and upset conditions. Also described are ways to improve heat–mass transfer modeling, thus augmenting heat transfer from SFAs and allowing improved storage capacity. Keywords: spent fuel assembly (SFA), spent nuclear fuel (SNF), wet and dry SNF storage, modeling heat–mass transfer, natural and mixed convection, casks, cask accidents
1 Introduction In practically all industrially developed countries, the problems of safe spent nuclear fuel (SNF) storage and its technology development are critical ones. Most spent fuel assemblies (SFAs) are stored now in water pools at nuclear
______ ∗
To whom correspondence should be addressed: E. D. Federovich, I. I. Poluzunov BoilerTurbine Research and Design Institute, 24, Politechnicheskaya str., St. Petersburg, 194021, Russia; e-mail:
[email protected] 189 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 189–208. © 2007 Springer.
190
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
power plants (NPPs). Several countries, including the USA, plan to transfer SNF to regional or national “dry” storage facilities, explaining this decision by the necessity to increase levels of radiation and nuclear safety. For example, the American National Academy of Science recently concluded that the level of safety of wet storage at NPPs is unsatisfactory; US official bodies plan the removal of SNF from such storage. This decision looks not to be very realistic, bearing in mind the huge technical and financial resources required for its realization. This situation, which one can say is also characteristic of the Russian Federation, demonstrates that several types of SNF storage will coexist on our planet in the nearest decades: •
•
•
•
•
Water pools located close to nuclear reactors, which are intended for initial storing and cooling of SNF for approximately 1 year after its extraction from the reactor. The fuel in such storage has relatively high thermal power (up to several tens of kilowatts per SFA) and is cooled by intense flow of water. Water pools also located at the NPP territory in separate buildings, but intended for interim storage of SNF during several years or several decades before its reprocessing or transportation to “dry” storage. In the Russian Federation, where the fuel from the RBMK-type reactors is not reprocessed for time being, and fuel from VVER-type reactors is reprocessed only partly, the major part of SNF is kept in such storage. Water pools located at nuclear fuel recycle plants. These plants reprocess SNF or condition it for further storage, or use it as a raw material for secondary fuel manufacturing. Significant amount of SNF can be stored at this kind of facility, including imported SNF. Typical examples are the Sellafield plant in the UK, La Hague plant in France, and the Krasnoyarsk Mining-Chemical Complex in Russia. Metal or metal/concrete casks [9,10,17], usually filled with inert gas (helium) and intended for interim “dry” storage (up to 50 years) of a limited number of SFAs (6 to 30). These casks often play the double role of a facility for SNF storage and SFA transportation between an NPP and other nuclear fuel recycle facilities. “Dry” storage of the cask type is usually located at NPP sites. Their number remarkably increased in some countries (Germany, USA, Lithuania), and one can expect significant growth of this tendency, because in the short-term the cask-type storage alternative looks attractive in providing a step-by-step introduction of “microstorage” units, without a large one-time expenditure. In the long term, however, this option is not so wise, because it still involves many expensive casks. “Dry” storage in the form of large-scale structures—made of concrete and located either on the ground or beneath—require huge capital investment,
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
191
but their capacity can meet national demands (or even demands of several nations) for long-term SNF storage. For example, the dry storage facility which is now under construction at the Krasnoyarsk Complex is intended for SNF containing 30,000 t of heavy metal (Figure 1). The Yucca Mountain project is a similarly large-scale enterprise.
Figure 1. Ground-level large-capacity dry storage facility (Russian Federation)
2 Role of Reliable Heat Removal in Safe SNF Storage Specific features characterize the safety of each type of spent fuel storage. The main safety requirements are defined by the following factors (acting over a period characteristic of each particular type of storage): • • •
Minimizing the danger of hermetic sealing damage (leading to release of radioactivity by the fuel) Minimizing the release of radioactivity and its migration to the environment in case of fuel-element cladding damage Minimizing the mechanical damage of SNF elements as a result of external impacts (earthquakes, fires, explosions, projectiles, fall of flying objects).
Each factor of safety is supported by design, construction, operational, and administrative measures. We will consider in more detail the measures of the design and construction character, directly providing the desired (and allowed) temperature levels in all the SNF elements, and (indirectly) the long-term stability of the SNF materials under the effect of the processes defining their life periods (corrosion, radionuclide migration, stresses). In Table 1, the heat–mass
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
192
transfer processes that define temperatures of SNF elements under normal conditions are indicated. Table 2 does the same for accidents. TABLE 1. Heat–mass transfer for reliable heat removal under normal conditions Storage type
Processes Heat release in fuel matrix due to radioactive decay Heat conduction in the fuel and through the cladding Convective heat transfer from fuel rods to pool water (or in-canister water, if SFAs are placed in waterfilled canisters) due to natural or mixed convection 4. Evaporation of the water from the pool and from the canisters into the ventilated air flow above water levels; heat release from the pool by evaporation 5. Condensation of the steam from the steam-air mixture on the relatively cold surfaces of storage elements (metal floor, walls) 6. Heat conduction through the pool construction to environment 1 and 2. As above (typical for all kinds of storages) 3. Convection heat transfer from fuel rods due to natural convection of gaseous coolant inside vertically or horizontally oriented casks 4. Thermal radiation inside casks, radiation heat transfer between the rows of fuel rods and between the fuel and basket-surrounding elements 5. Heat conduction through internal elements of the cask and through its thick body wall 6. Natural convection and thermal radiation from the cask’s outer surface to the environment 1. Natural circulation and mixed co-current convection (circulation up-flow plus natural convection at the heat-releasing surfaces of the SNF canisters) 2. Thermal radiation heat transfer from the fuel rods, SNF to the surrounding elements 1. Mixed co-current convection (forced flow from air ventilation plus natural convection at heated surfaces). 2. Thermal radiation 3. Evaporation/condensation in air should be included 1. 2. 3.
Water pools
Casks
Surface storage (cask, vault, module, and chamber type). Underground storage
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
193
TABLE 2. Heat–mass transfer processes for SNF under accident conditions Storage type
Water pools
Casks
Processes and accident situations The same, as under normal conditions plus: 1. “Pure” natural convection (not forced) as a result of an accidental stop of organized water circulation in the pool. 2. Temperature rise in fuel and all storage components, leading to the evaporation of water at increasing rate and a decreasing water level in the pool and canisters (for canister-type storage). 3. Above-mentioned situations may happen in an electrical power failure. The ventilation system will then also stop. It means the deterioration of the heat removal from SNF with evaporating water and steam-air mixture evacuation via the ventilation system. The worst situation is when water loss is at a high rate. It may occur when the ventilation channels are blocked (by damaged structures, by human factor negative effect, etc.), and heat removal is possible only by heat conduction and thermal radiation through storage elements. 4. When water levels fall below the top of SFA, natural convection in the air–steam mixture, and later in dry air, provides heat removal, plus thermal radiation, the role of which increases with rise in temperatures. 5. The main feature of all processes accompanying an accident is their nonstationary character. Eventually, thermal equilibrium will be reached in the storage environment. 6. The most important parameters that should be defined in the thermal analysis of a hypothetical accident situation are: • Maximum temperatures at critical points, where such overheating leads to loss of hermetic tightness, or strength (fuel rods claddings, supporting structures). • Time intervals in which dangerous temperatures and eventual thermal equilibrium can be reached, because personnel will have only this time to locate and terminate the accident and make necessary repairs, including efforts to treat exhausted air in the ventilation system which may be contaminated. Accidents during transportation can lead to dangerous impacts and the potential for fire. IAEA rules specify drop tests for casks to simulate accident shock and impact, and corresponding design criteria for cask component behavior are well developed. On a positive note, international experience reveals an absence of significant cask damage in several thousands of transportation events. The objective factor helping here is the (Continued)
194
Surface dry storage
Underground dry storage
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
large thickness of the cask body, bottom, and lids (usually double lids with a control volume between them), necessary for the radiation protection. IAEA rules assume a cask is in a fire for 30 min. Thermal cask design for fire situations should perhaps consider longer periods—at least several hours, bearing in mind the time taken for a cask to heat up and cool down. Fires can have different sources, including transport accidents involving flammable substances, leaks and ruptures of gas and oil pipelines, terrorist attacks, etc. Nonstationary heating of system elements may arise as a result of air-cooling channels blocking, when only small-scale vortextype natural convection of air is possible (general air circulation is absent) together with thermal radiation. Situation is similar to the previous one, but the difference lies in two additional possible scenarios: • Accidental termination of the forced air circulation along the branched system of shafts, horizontal and vertical tunnels located deep underground, intended for technological aims and for storage containers with conditioned (compacted) SNF (see example in Figure 4). In this case, natural circulation replaces forced convection, and thus the rate of heat removal falls. • Blocking air input and/or output in storage channels (one or many). Small-scale vortex natural convection plus thermal radiation and the heat conductivity of the mine-surrounding rock are the remaining heat transfer mechanisms over the period of time when measures to mitigate consequences of the accident must be taken.
3 Design and Modeling of SNF Storage Systems A legitimate question to ask is how developed and how correct are modern design and modern methods of physical and mathematical modeling in predicting values of important safety parameters for all probable stages of storage operation (normal, including the long-term aspect, and accident ones). The answer is not an easy one, taking into account the huge number of influencing factors. 3.1 WORLD EXPERIENCE World experience of operating wet storage facilities in terms of their nuclear radiation and ecological safety is basically positive. For example, the serious accident of stopping water circulation in the large-scale facility at Sellafield in the UK did not lead to a dangerous situation, because the rate of rise in water
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
195
temperature was small and personnel had enough time to repair the circulation system. In addition, many pools at NPPs have operated safely for decades and NPPs are planning to prolong their life well beyond design limits. A typical example is the Leningradskaya NPP (Russian Federation), which has safely operated its wet storage system for 31 years and is undertaking intensive research and development (R&D) to prove the possibility of prolonging its life to 50 or even 100 years [19]. Sweden has experience of underground water pool type storage operation [18]. Swedish specialists consider as advantageous the features of wet storage (compared to dry storage), because modern water chemistry allows control of the water composition—and maintenance of it during operation—with great precision. This means it is possible to keep the corrosion rate at a minimum level and predict it precisely in the relevant time period. This is different from dry storage; there the atmosphere can be considered only conditionally as dry, because air is basically a wet medium that contains impurities and components which are difficult to control. At the same time, one can observe an overloading of NPP pools—the result of growth in nuclear power combined with a stalling of SNF reprocessing. The growing risks associated with large amounts and concentrations of SNF lead to the tendency to reload SNF into dry storage, but the lack of a large centralized storage capacity makes the idea of the cask-type storage development attractive. Such storage is usually located at NPP sites. This path of least resistance is characterized by relatively modest expenses by today’s standards, but in the long term one can expect the reloading of SNF into large-scale dry storage facilities. This tendency correlates with the shift in several countries—USA and Germany, in particular—from the reprocessing of SNF to its final disposal [20]. The requirements for the safety of SNF storage constantly grow and one can expect further increase as a consequence of SNF accumulation and growth in technogenic and anthropogenic risks. 3.2 STUDIES AT ST. PETERSBURG POLYTECHNIC UNIVERSITY Modern physical sciences and engineering have accumulated considerable experience in design of SNF storage systems and in predicting the processes that control their operational parameters. Nevertheless, gaps of understanding still exist in this field, which should draw the attention of specialists. We would like to consider these gaps in more detail using the heat–mass transfer processes in Tables 1 and 2 as examples.
196
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
3.2.1 Accident progression in wet storage after loss-of-flow The scheme of a thermal twodimensional computation model pertaining to this problem was developed and is shown in Figure 1. In this scheme, a spent fuel pool is, in fact, reduced to one SNF “microstorage” unit, which has all elements of an actual storage facility that participate in heat– mass transfer, e.g., fuel rods, canisters of SNF, pool water, and pool structures (walls, bottom, metal deck-plates above the water level). So, one can expect that after the 1: support rod; 2,3: rows of fuel rods; determination of specific para4: canister water; 5: canister wall; meters of the nonstationary pro6: pool water; 7: pool wall cess, namely the amounts of heat removed and water evaporated Figure 2. Elementary pool cell used to from one cell, and after the submodel loss-of-flow in SNF storage sequent multiplication of these parameters by the number of SFAs in storage, the values for the entire facility can be determined. In this model, all sizes of the canisters are assumed to be real ones. However, the intercanister spaces are modeled in the form of annular channels, which is simpler compared to real geometry, but similar with regard to heat– mass transfer processes development. All sizes of the volumes external in relation to a single canister, are chosen in a way that all the intercanister space cross-section dimensions and concrete walls restrictions of the model are the same as the real ones. The width of two heat-generating rings, modeling two rows of fuel rods inside a SFA (such is the geometry of the RBMK-type reactor SFA), is chosen in a way that the evaporation mirror inside the canister of the model is equal to that in a real-life storage. The conditions of thermal modeling are satisfied by the equality of the meanings of the Bio criterion of the model and that of the real object and in recalculation of the Fourier criterion, taking into account the difference between concrete walls restrictions of the model and ones for the real pool. In the above model the water–air boundary is mobile and moves downwards in the course of evaporation from the pool and canisters.
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
197
System of heat–mass transfer equations including equations of movement in cylindrical coordinates taking into account the temperature dependence of the media physical properties (x—radial coordinate, y—axial coordinate) is presented as follows: сρ
⎞ ∂T 1 ∂ ⎛ ∂T ⎞ ∂ ⎛ ∂T ⎜⎜ λ = − сρUxT ⎟ + − сρVT ⎟⎟ + Q v ; ⎜ xλ ∂τ x∂x⎝ ∂x ∂ y ∂ y ⎠ ⎝ ⎠ ⎞ ∂C 1 ∂ ⎛ ∂C ⎞ ∂ ⎛ ∂C ⎜⎜ D = − UxC ⎟ + − VC ⎟⎟ ; ⎜ xD y y ∂τ x ∂x⎝ ∂x ∂ ∂ ⎠ ⎝ ⎠
(1)
⎛∂ U 1 ⎛∂ U U⎞ ∂U ∂ U⎞ ∂P ∂ ⎛ ∂ U ⎞ ∂ ⎡ ⎛ ∂ U ∂ V ⎞⎤ ⎟⎟ = − ⎟⎥ + ρFx ; ρ⎜⎜ +U +V + 2µ ⎜ − ⎟+2 + ⎜µ ⎟+ ⎢µ⎜ τ x y x x x x x ⎝ ∂ x ⎠ ∂ y ⎣ ⎜⎝ ∂ y ∂ x ⎟⎠⎦ ∂ ∂ ∂ ∂ ∂ ∂ ⎝ ⎠ ⎝ ⎠ ⎛∂ V 1 ⎛∂ V ∂ U⎞ ∂V ∂ V⎞ ∂P ∂ ⎛ ∂ V ⎞ ∂ ⎡ ⎛ ∂ U ∂ V ⎞⎤ ⎟⎟ = − ⎟⎟ + 2 ⎜µ ⎟+ ⎟⎥ + ρFy ; ρ⎜⎜ +U +V + 2µ ⎜⎜ + + ⎢µ⎜ x⎝∂x ∂y⎠ ∂x ∂y⎠ ∂y ∂ y ⎜⎝ ∂ y ⎟⎠ ∂ x ⎣ ⎜⎝ ∂ y ∂ x ⎟⎠⎦ ⎝∂τ 1 ∂ (xU ) ∂ V + =0; x ∂x ∂y
G G F = gβ(Ta − T ) .
System (1) is solved by the finite difference numerical method under the following boundary conditions: • •
•
•
At solid surfaces “sticking” conditions apply: U = 0, V = 0. At the boundary “water–moist air” the normal velocity components are equal to zero (V = 0) and tangential stresses are absent (∂U/∂y = 0); this corresponds to no ventilation in the space above the water (conservative evaluation); if ventilation is included, the flow rate and velocity of ventilating air should be specified. At the boundary line “water–steam–air mixture” and near the upper metal floor, which is cooled by the natural air convection inside the storage building, the vapor is saturated, but at the final stages of accident process the steam pressure near the floor may be lower than saturation pressure. At the outer boundary of a concrete wall of the pool the boundary condition of the third type is specified: −λ
•
∂Т = α env (Т − Т env ) ∂x
Temperature at the lower boundary of a computational region is constant and is equal to the annual average soil temperature under the bottom (foundation) of the pool.
198
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
The solution of the system (1) is reduced by integer process to the integral heat balance equation for storage. So, for each moment in time it is: ΣQSNFA = Qh + Qev + Qtr – Qcond.
(2)
where ΣQSNFA: sum of all SNF heat generation; Qh: value of all media and structure heating; Qev: heat removal from SNF storage by evaporation; Qtr: heat transmitted to the environment through the boundaries (side walls, steel deck-plates, bottom, and pool foundation) due to heat conductivity, thermal radiation, and with ventilating air; Qcond: returned heat due to condensation of steam on solid surfaces colder than the water. In expression (2) the value of ΣQSNFA is specified and Qh, Qev, Qtr, and Qcond are calculated. To solve the system (1) and determine values for equation (2) components it is necessary to have closing empirical correlations for the heat–mass transfer coefficients. Sometimes it is possible to use literature data for this purpose, but otherwise it is necessary to perform special experiments modeling particular process conditions. The combination of natural and forced convection (mixed convection) in normal operation and at the initial moment of an accident is just such a case of water flow that requires experiments for its proper modeling. 3.2.2 Study of mixed convection using a small-scale experimental model A series of tests with a wet storage cell model (scale 1:10, with electrical heating) was performed in a set-up (Figure 3) consisting of 56 tubes submerged in water and heated by joule heat.
Figure 3. Experimental set-up used to study mixed convection
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
199
These tubes model either the fuel rods (when SFAs are stored without canisters), or the canisters submerged in the pool water. Two directions of mixed convection were studied: parallel flow, when the forced flow has the same direction as the natural one (upward direction), and counterflow, when the forced flow has a downward direction (this type of mixed convection is realized in the SNF storage at the Leningrad NPP). Under the conditions of counterflow convection, and assuming the relative distance between the tubes axes being S/d = 1.4, where S is the intertube distance, d is tube diameter (the tube grid is square), the following heat transfer correlation was obtained: Nu/Nu0 = 0.325 (Raq/Re)0.34,
(3)
where Nu and Nu0 are the Nusselt numbers for mixed convection and for pure forced convection provided by the turbulent flow in the circular tube of the same equivalent diameter as that of the intertube spacing (Nu0 value is taken from literature1). Using this equivalent diameter the Rayleigh number Raq and Reynolds number Re can be defined. The walls of the experimental model were transparent, allowing the observation of vortices—indicating turbulization—in the intertube space, which was accompanied by water temperature oscillations. The intensity of oscillation was similar to that of temperature differences in “liquid–tube walls”. Flows between tubes in smallscaled model: “cold” mode
test mode
Water temperature pulsation
Figure 4. Results of mixed-convection tests in water
200
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
Under the conditions of the co-current flow, the laminarization of the flow between the tubes was observed. The heat transfer intensity was lower than by the counter flow convection. This circumstance should be taken into account by choosing the scheme of water circulation in the SNF pools. On the one hand the parallel flow scheme, when the water is taken from the pool for cooling and for purification purposes from the upper layer is preferable, because in this layer the water is warmer and more “dirty” (in terms of concentration of hydrosols). It means more effective cooling in the outer system (larger temperature differences between the pool water and cooling water) and more effective water purification. On the other hand the lower heat transfer coefficients in parallel flow convection will give slightly higher cladding temperatures, which will be unfavorable from the point of view of corrosion. 3.3 STUDIES AT THE POLZUNOV BOILER-TURBINE INSTITUTE (CSKTI) 3.3.1 Natural convection in a fuel-rod bundle (or SFA canisters) This kind of small-scale, cell-type convection takes place in a pool when forced circulation across it has stopped (accident situation), or when the vessel with water has “deaf bottom”, as can be observed in SFA canisters or in waterfilled casks. Figures 5.1 and 5.2 present experimental data from the CSKTI5 for natural convection in the absence of circulation along a rod bundle with geometry close to that of an RBMK FA. The data are given in the form of a correlation between relative heat conductivity of water or air (λeff—effective heat conductivity of convecting medium); λ0—molecular effective heat conductivity of that medium) and the Rayleigh number Ra, for experiments (1,2) with different shell geometries.
Figure 5.1 Results from experiments on convection in a fuel-rod bundle
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
201
Figure 5.2 Results from experiments on convection in a fuel-rod bundle
3.3.2 Vertically oriented casks with SFAs Usually in this case the natural circulation circuit is formed inside the cask interspace with up-flow in its central part and down-flow in the annular channel between the basket, containing SFAs and the cask body. Experimental data obtained from CSKTI6 provides a possibility to perform the calculation of heat mass transfer coefficients (by stabilized parts of channels) by upflow in tubes, rod bundles, and annular channels): Nu/Nulam = [1 + (0,065 (Gr/Re)3/2]1/3,
(4)
where Nulam—Nusselt number corresponding to the laminar flow regime without natural convection influence (this theoretical value can be found in literature).This type of flow is usually called “viscous-gravitational flow”. It is characterized by the flow turbulization (similar to the one observed in our above experiments with small scale model of the pool cell in section 3.2.1) leading to the increase of both the heat transfer intensity and the hydraulic resistance of the channel. Another correlation is presented in the same work6 for down-flow in the annular channel around the basket:
202
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
_
Nu/ Nu = [1 + (0,125 ⋅ Gr ⋅ Re ) 4 ]1/ 4
(5)
Both correlations (4) and (5) are valid for the ratio Gr/Re in the range from 10 to 104. The experimental data for hydraulic resistance coefficients in viscous – gravitation regime, obtained in the same work,6 allows the possibility to calculate the natural circulation gas flow rate in a closed system of the circuit “SNF assemblies—180° turn—annular channel—180° turn”. With this flow rate value and the corresponding average flow velocity the Reynolds number in correlations (4) and (5) could be defined. 3.3.3 Horizontally oriented casks with SFAs This orientation is characteristic for cask transportation. The experimental data obtained in CsKTI7 for a wide range of similarity criteria, gas pressure and gas compositions, give an opportunity to define natural convection heat transfer coefficients in SNF rod horizontal bundles (Figure 5). The complex of the design recommendations which has been developed in CsKTI in the period of creation of all the national cask models (starting from the end of the 1960s) was included in the Design Code.14 The same was also for the design of the more modern cask generations, for example the ones used at the Ignalina NPP, Lithuania (CONSTOR, a metal– concrete cask developed jointly by GNS-GNB and CsKTI.10 3.3.4 Dry large-capacity SNF storage of the chamber type One of the largest dry SNF storage facilities, intended for storage of up to 30,000 t of SNF, is now under construction at Krasnoyarsk in the Russian Federation. Its general layout was shown in Figure 1. It consists of a row of chambers with sockets for positioning SNF canisters. The sockets are placed on a square grid and form complicated air channels inside the chambers, with air entering from the bottom end. Air moves under the effect of thermal and gravitational forces. Computations of the local air velocity fields, its temperature, local heat fluxes and Nusselt numbers demonstrate the complicated character of the cooling air space flow and thermal parameters distribution.12 Computational studies of the aerodynamics and thermal regimes of the chamber-type SNFS were complemented by experiments [15,16]. Local parameters of the natural-size channel of an air tube, relevant for the natural convection, were determined for the case of the fissile material storage (FMS), the one constructed in the Russian Federation in cooperation with American authorities, national laboratories, and companies. The averaged heat transfer coefficients, determined in the result of these experiments, can be calculated using following correlation:
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
Nu = 0,0816 Raq0,2815,
203
(6)
where Nu and Raq numbers are calculated using the tube radius and the heat flux at the tube wall. Thermophysical parameters (temperature, air velocity, heat flux, oscillations of temperature) of the chamber-type SNF storage facilities and of the fissile materials storage were determined.15 The averaged heat transfer coefficient by natural convection in SNFS model is described by the equation: Nu = 0,2635 Raq0,223,
(7)
Now it is planned to stage experiments using a large-scale model of the chamber-type SNF storage facility, which includes not only the sockets, but also the inlet and outlet collectors, turns, etc. 3.3.5 Cask in the fire accident situation The combined nonstationary heat transfer problem should be solved in this case, i.e., the temperature fields in the cask body, in its internals and in the outer-cask space are to be defined. The role of thermal radiation becomes significant at the stage of the cask heating and at the stage of its cooling, when the fire is terminated and the cask is removed from the heart of the fire. Nowadays, the problem is complicated because of technogenic and anthropogenic factors are numerical ones and feature a great variety of the boundary conditions. One can mention liquid-fuel fires at transport lines, consequences of natural gas pipes ruptures, explosions, terrorist attacks, etc. In these cases, it is necessary to model not only the heat transfer processes, but also the combustion process outside the cask. Universal and powerful computational methods and applied programs were developed for the analysis of the thermal radiation inside and outside the cask. These methods are based on the radiant fluxes algebra implementation.1 For the gas coolant inside the cask the radiation component (the heat exchange between the fuel rods and the basket) should be added to the convective one. In the Russian Federation and abroad, numerical fire tests of the casks and the cask models are performed to check their sealing tightness. However, no special measurements of the heat transfer parameters (local heat fluxes, air and combustion product velocity, emissivity coefficients of the surfaces, and media and their change during the history of the fire) are performed, as a rule. So, the computational models verification is difficult.
204
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
3.3.6 Heat removal from a finned outer-cask surface The outer fins on a cask body play an important role in augmenting heat removal. Under the conditions of natural convection and thermal radiation, this role lies mainly in enlargement of the heat release surface, because the convective heat exchange intensity decreases in the narrow spaces between the fins due to the friction of air and the fin plates. Opposite fin surfaces have equal temperatures, which lead to a decrease in the effectiveness of radiation. Nevertheless, fins increase the effective heat transfer coefficient (relation of the fin heat exchange to that on the fin-carrying smooth surface) by 1.5–2 times, and even more. Nowadays, while choosing a type of cask fin (axial, radial, or others), a cask designer thinks primarily about the manufacturing cost. Nevertheless, the detailed optimization of fin geometry is a necessary step, because the required cask capacity (released heat power) increasing constantly goes on, and the problem of the cask material consuming minimization becomes more important under the conditions of large-scales production. French casks of the TNtype have very effective fins comprised of copper tenons coated with stainless steel. The thermal power of these TN-casks reaches approximately 50 KW, which is far superior to other casks (30–40 KW at a maximum). If a cask moves in atmospheric air (e.g., during the evacuation from the heart of the fire), the rate of the finned surface cooling will be significantly higher than that of the smooth surface, because the fins intensify forced convection of heat due to turbulent air flow. It should be mentioned, however, that the publications in this field are rather scarce, and most of them relate to the finned surfaces of industrial thermotechnics and electronic devices having much smaller sizes compared to SNF casks. The factor of scale is important in convective processes. 3.3.7 Mass transfer processes in wet storage The phenomena of evaporation into still air and into a ventilating air flow, as well as steam condensation, have not been studied thoroughly enough. However, analysis demonstrates that about 20% of overall heat removal from the water storage to the environment can be realized by evaporating under normal operating conditions, and the role of evaporation becomes decisive (up to 90% and more) in the hypothetical, however probable situation of stopping the forced pool water circulation. The total heat flux from the water surface is determined by the following expression: _
qtot = r PM/R T ·βmass·(Ysurf – Yair) + αsurf (Tsurf – Tair),
(8)
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
205
_
where T is the averaged temperature of the boundary layer, and Y = Pp/P is the ratio of the partial pressure to the full (close to atmospheric) pressure. The values of the mass transfer (βmass) and heat transfer (αsurf) coefficients can be taken from literature or from special experiments. In case a coefficient is unknown, the analogy between the heat and mass transfer can be used8 to make tentative assumptions. At elevated water temperatures of up to 100°C, which can occur in an accident (as in the stoppage of water circulation), the influence of steam crossflow in the direction parallel with the water surface on the flow of heat and mass transfer in the boundary layer increases. This evaluation can be made based on theoretical work13 in the study of heat–mass transfer and flow friction in the boundary layer with cross-flow of a blown agent (evaporation in this instance. 3.3.8 Underground dry storage (UDS) UDS facilities are built for the long-term storage of massive amounts of SNF. For time being, their number is small and they are of relatively small capacity (Sweden, Germany, Finland, USA). Launching the larger ones is planned in the USA (Yucca Mountain project) and in several other countries. It is evident that SNF accumulation will lead to growth in these storages intended for long (several hundred years) or/and nearly infinite periods of life. For example, the Russian Federation has territories with geological conditions proper for UDS organization, e.g., northern Siberia, under conditions of international cooperation. Thermal design of UDS is based on following methods: • • •
Mathematical modeling of temperature fields in the system “conditioned (compacted) SNF—storage container—air-ventilated channel in the rock– metal liner of the shaft with a container, shaft surrounding rock” Aerodynamical analysis of the shaft ventilation system, operating in the forced ventilation mode (normal operation conditions), or in the natural air circulation mode (emergency conditions) Mathematical modeling of UDS as a whole, including the channels for air flow input and output, vertical shafts and horizontal galleries for container loading, in-storage transportation and unloading (if necessary), collectors, etc.
Modern applied sciences, such as the aerodynamics of ventilation systems for the mining industry, heat transfer theory, geophysics, and geochemistry, have collected enough material for this type of design. However, a large number of boundary conditions (underground water migration, big variety of mine
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
206
substances, compositions, and properties) leads to a large amount of preparatory investigations and calculations including on-site studies and tests. Taking into account the hazard of serious accidents (earthquakes, rock movement due to natural and anthropogenic effects, air channel blockage, etc.), we must perform the heat transfer analysis of these accidents by computation methods. The main goal of such analysis is to define the maximum temperatures of the most critical parts of the system (fuel matrix, fuel cladding, walls, and container sealing∗). Usually, in accordance with IAEA rules, the maximum temperature of a metal container (cask) body is 80–100°C, and the temperatures of zirconium alloy claddings at a level of 350°C (normal conditions). Evidently, in the accident situation, when heat removal conditions may be much harder, these temperatures will be higher. For providing enough safe operation, it is necessary to determine a priori the time interval in which a storage component will retain its integrity, and the personnel and the repair services will have a possibility to perform the necessary antiaccident and postaccident measures. 4 Conclusions 1.
Modern design methods provide prediction of the reliability of developed types of SNF storages at a high enough level, namely of: • • •
Wet storages of the pool type and water-filled casks Gas-filled casks of vertical and horizontal orientation Dry storage of different types (cask type, modular type, chamber type on the ground surface)
These characteristics in particular include: • • • • • 2.
Temperature regimes for normal operation, with deviations from normal and emergency conditions Air and water coolants flow rates and mass velocities Mean heat fluxes and heat balance components Mean heat and mass transfer coefficients Rates of evaporation
Nevertheless, the way to determine the optimum design and operational regimes of SNF storage has not yet been finalized. There is a lack of
______
∗ The copper is considered as a perspective material for a container body (3,18). The intensive studies of ceramics and other materials are conducted in different countries (Sweden, USA, Australia, Russian Federation, Germany, etc.)
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
207
experimental data used by designers, of data concerning local stationary and nonstationary characteristics, namely local components of heat flux and medium velocities, and local heat and mass transfer coefficients. These data are necessary for verification of computational programs and for accurate prognosis of following parameters: • •
Local temperatures, especially in critical zones and points of a system Local air (gas) humidities (steam concentration)
3. The search for the ways to augment heat removal from SNF should be continued, particularly in the following directions: • • •
Study of the materials for the containers internals (baskets, inserts, etc.) and bodies, which have high heat conductivity and high resistance to corrosion impact Optimization of fin geometry, that is size, form, and number of fins Organization of coolant natural circulation circuits, providing the required velocity values in normal and emergency situations and effective interaction with the forced circulation (if the forced convection option is dictated by technology requirements)
References 1. 2.
3.
4.
5.
6.
7.
Kutateladze, S.S., Fundamentals of Heat Transfer. M. Atomizdat, 1979 (in Russian); 19 pp. (in English). Zubkov, A.A., Petrenya, Y.K., Fromzel, V.N., “About New Generation of National Casks for Spent Nuclear Fuel and Radioactive Wastes”, Proceedings of CsKTI, 2002, Issue 282, pp. 45–51 (in Russian). Tveiten, B., Kersting, W., Karpov, A., Russian Weapons Plutonium and the Western Option; Nuclear Disarmament Forum AG, Zug-Switzerland, 2002, 199 pp. (in English). Schmakov, L.V., Fiodorov, M.P., Fedorovich, E.D., Karjakin, Yu.E., “Provision of the Long-Term Safe Compacted Storage of Spent Nuclear Fuel in Water Pools”, Proceedings of CsKTI, Issue 282, 2002, pp. 135–141 (in Russian). Fromzel, V.N., Schleifer, V.A., Fedorovich, E.D., “Heat Transfer by Free Convection in a Bundle of Vertical Heat-Generating Rods in Absence of Coolant Circulation”, Journal of Engineering Physics, 1984, Vol XVI(4), (in Russian). Lusin, I.P., “The Development of Thermal Design Methodics of Vertical Casks for Spent Nuclear Fuel Transportation on Base of the Viscous-Gravitational Coolant’s Flow Study”, Autoreferat of Candidate Degree Thesis, Leningrad, NPO CsKTI named after I.I. Polzunov, 1984, 22 pp. (in Russian). Andreev, P.A., Fromzel, V.N., Pervitskaya, T.A., Heat Transfer and Hydrodynamics of One-phase Flow in a Bundle of Rods, Academy of Sciences, Leningrad, 1979, pp. 121–129 (in Russian).
208
8.
9.
10. 11.
12.
13. 14.
15.
16.
ISSUES IN WET AND DRY STORAGE OF SPENT FUEL
Arefiev, K.M., Fedorovich, E.D., “Problems of Heat–Mass Transfer Study in Water Pools for The Nuclear Power Plants Spent Fuel Storage”, Proceedings of the XIII School-Seminar of Young Scientists and Specialists Chaired by Academician A.I. Leontjev, “Physical Fundamentals of the Experimental and Mathematical Modeling of Gas Dynamics and Heat and Mass Transfer Processes in Power Generating Facilities”, M.: Moscow Power Institute, Vol. 2, 2001, pp. 412–419 (in Russian and English). Zubkov, A.A., Fromzel, V.N., Fedorovich, E.D., Diersch, R. et al., “Heat Transfer Analysis in Steel/Concrete Cask CONSTOR”, Proceedings of 12th International Conference, “PATRAM-98” (in English). Yanberg, K., “Experience of Joint Development of Metal-Concrete Cask for Spent Fuel”, Proceedings of CsKTI, 2002, Issue 282, pp. 281–284 (in Russian). Fedorovich, E.D., Snegirev, A.Yu., Talalov, V.A., Stepanov, V.V., “The Heat Transfer Process Modeling in the Water-Filled Casks for NPP Spent Fuel Storage”, Material science problems, SPb.: the Prometey State Unitary Enterprise, 6(12), 1997, pp. 38–47 (in Russian). Smirnov, E.M., Zaitsev, D.K., “Finite Elements Method in Application to Problems of Hydrogas Dynamics and Heat Transfer in Complicated Geometry Zones”, Scientific-Technical Issues of St. Petersburg State Polytechnical University, Issue 2, 2004, pp.75–79. (in Russian). Kutateladze, S.S., Leontjev, A.I., Heat, “Mass Transfer and Friction in the Boundary Turbulent Zone”, Energia, Moscow, 1972, 342 pp. Thermal and hydraulic design of heat exchange equipment of NPP, Methodical directions (RD 24.035.05.89), Leningrad, Ministry of Heavy and Power Equipment—NPO CsKTI named after I.I. Polzunov, 1991, 210 pp (in Russian). Blinov, M.A., Lebedev, M.E., Muhina, I.S. et al., “Heat Transfer by Free and Mixed Convection of Cooling Air in storages of Nuclear Spent Fuel”, Proceedings of the V Minsk International HeatMassTransfer Forum. Lyikov HeatMassTransfer Institute, Minsk, May 24–28, 2004. Petrenya, Yu.K., Sudakov, A.V., Zubkov, A.A. et al., “The Works of NPO CsKTI Stock Holding Company on Developing Heat Regimes of Storages for Fissile Materials and Nuclear Spent Fuel”, Proceedings of the VIII International Conference on Safety of Nuclear Technologies: Radioactive Wastes Management, St. Petersburg, Russian Federation, May 27–30, 2004.
PROBLEMS OF NUCLEAR AND RADIATION SAFETY OF CASKS WITH SPENT FUEL DURING LONG-TERM DRY STORAGE
A. Z. AISABEKOV, S. A. MUKENEVA, E. S. TUR,∗ AND V. M. TSYNGAYEV
National Nuclear Center, Kurchatov, Republic of Kazakhstan Abstract: This paper considers problems of nuclear and radiation safety for long-term (~50 years) dry storage of spent nuclear fuel (SNF) from the BN-350 reactor in metal-concrete casks. This work is part of decommissioning BN-350 and—as the preliminary safety evaluation of a long-term storage site—will be helpful during site design, construction, and eventual operation. During the validation of the safety of long-term SNF storage particular attention should be given to preserving tight seals, the stability of the strength characteristics of materials used in cask design, and the regimes of heat removal under various environmental conditions. The basic structure and design configurations—fuel assemblies, canisters, metal–concrete casks, etc.—are shown in Figure 1. The major factors influencing nuclear and radiation safety are described: fuel burn-up, enrichment, fabrication tolerance, fuel assembly types, assembly configuration in the canisters, and canisters in the containers, background of assemblies placed in the reactor and cooling pool, configuration of containers on the storage site, and degree of radiation embrittlement of assembly materials. The conditions under which the SNF casks will be stored and probable accident situations are described in the paper. The results of neutron-physical and thermal calculations and calculations of SNF cask radiation fields are described for both normal and emergency conditions. The geometry to analyze criticality events is shown in Figure 2. For normal conditions, the calculated keff for this configuration was found to be 0.534; when fully flooded, the keff value was 0.901, i.e., the cask was still safe. Keywords: BN-350 reactor, spent nuclear fuel (SNF), SNF canisters, metal–concrete casks, heat transfer conditions, criticality calculations
______
∗ To whom correspondence should be addressed: E.S. Tur, National Nuclear Center of the Republic of Kazakhstan, 071100, Kurchatov, Kazakhstan; e-mail:
[email protected] .
209 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 209–210. © 2007 Springer.
210
SAFETY OF SPENT FUEL CASKS FOR DRY STORAGE
Figure 1. Dry storage cask for BN-350 SNF
Figure 2. Cask geometry used for criticality analysis (the cask contains 48 BN-350 SNF assemblies in 8 canisters)
TRIAL OF STORAGE CONTAINER TECHNOLOGY FOR RESEARCH REACTOR SPENT NUCLEAR FUEL
Z. I. CHECHETKINA,∗ Yu. V. CHECHETKIN, A. E. NOVOSELOV, V. G. BORDACHEV, V. V. MAKLAKOV, AND I. Yu. ZHEMKOV Russian State Scientific Center Research Institute of Atomic Reactors Dimitrovgrad, Russia Abstract: The Russian State Scientific Center Research Institute of Atomic Reactors (SSC RF RIAR) has implemented a program of experimental testing of container storage technology for research reactor spent nuclear fuel (RR SNF) using the TUK-108/1 container. The program is aimed as a trial of the “dry” storage technology for spent fuel assemblies (SFAs) as well as to determine the maximum permissible time and conditions of dry and wet storage. Standard SFAs from the MIR reactor after operation and wet storage for 4–40 years are subject to examination. The temperature in the container does not exceed 180°C from SFA decay heat, and the storage medium is air. An inspection bench, a facility to dry SFA ducts, and basics of transport technology were developed to estimate the external state of the assembly and perform a sipping test after long-term storage in the cooling pool. The calculated justification of radiation and nuclear safety was performed, design documents were developed, and main parts of the bench were fabricated. Keywords: research reactor (RR), spent fuel assembly (SFA), dry and wet storage, MIR reactor, TUK-108/1 container, SFA inspection stand
1 Introduction The majority of research reactors are operated using highly enriched nuclear fuel (10–90% U-235) and the reprocessing of such fuel is economically reasonable. However, accumulation of large amount of SNF in spent fuel pools has been caused by a lack of available reprocessing plants on the one hand, and by the absence up-to-date technologies on the other.
______ ∗
To whom correspondence should be addressed: Z. I. Chechetkina, State Scientific Center Research Institute of Atomic Reactors, Dimitrovgrad-10, 433510, Ulyanovsk region, Russia; e-mail:
[email protected] 211 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 211–223. © 2007 Springer.
212
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
The lack of technologies is related to the fact that the design of fuel rods and fuel assemblies (FAs), as well as fuel, absorbing and structural materials have changed whereas SNF reprocessing technologies were developed for the fuel alone. That is why at present there are no reprocessing technologies for many types of RR SNF stored in spent fuel pools. This situation causes problems for safe long-term storage of SNF and is the reason behind the current study. 2 Scope of the SNF Problem As a rule, RRs are operated using a wide range of fuel elements (fuel rods, FA) that differ in geometrical dimensions, type of fuel composition, cladding, and enrichment in U-235. The RR SNF fuel rods and fuel assemblies can be conveniently divided into the following main groups: • • • •
Dispersed fuel compositions UAlx–Al, UО2-Al in aluminum alloy claddings (MIR, ARBUS, VVER-C, IVV-2M, IRT-M, MR, IR-8 reactors) UО2 in zirconium alloy E-110 cladding (VK-50 reactor) Dispersed fuel comprising UО2 in Cu-Be matrix, cladding is made of EI847 steel, duct is made of Х18Н9Т steel (SM, RBT-6, RBT-10/1,2 reactors) UО2 in claddings made of different steels: EI-172, EI-847, ChS-68 (BOR60 and BR-10 reactors)
Some fuel rods and fuel assemblies were loaded with fuel in the form of carbides, nitrides, uranium beryllide, and uranium-based alloys. Different steels and alloys were used as claddings. Evidently, taking into account the application of different fuel and structural (cladding) materials, separate technologies should be developed for each SNF types as well as separate SNF reprocessing technologies and conditions for its safe storage within a set period of time. At present, in Russia about 30,000 spent fuel assemblies (SFA) of different types are stored which contain about 17 t of U-235 in total and of these only 15,500 SFAs are going to be reprocessed (they contain ~2.8 t of U-235, i.e., 16%). As for the rest of the SFAs, there are no reprocessing technologies at all, or their reprocessing is not economically reasonable. At the same time, the storage period of some SFAs exceeds 25 or even 35 years. There is also a serious problem related to the storage of leaking SFAs. Besides, 17 countries have 37 reactors built on the basis of Russian design and about 11,000 SFAs and fuel rods are stored there. Again, in most cases, they contain highly enriched uranium (HEU) (30–90% U-235). At present, the
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
213
countries are determined, from which SNF is subjected to shipment first of all. They are Ukraine, Bulgaria, Uzbekistan, and Yugoslavia. These countries have 995 SFAs and 8,030 fuel rods. The cooling and storage pools in Russian research centers are filled to more than 90% capacity, which makes reactor operations difficult. SSC RF RIAR itself is faced with a very serious problem having seven operating reactors. The RIAR storage facilities comprise about 7,000 pieces of FA that make up 99% of the design capacity. The storage time of some fuel now exceeds 30 years. At the same time, the pace of SNF shipment to the Mayak PA facility does not allow a decrease in the SNF inventory in the interim storage facility of SSC RF RIAR due to the lack of finances. 3 Fuel Storage Technology At present, fuel storage technology is of great importance. The International Standard is developed for the first fuel group fuel based on the water-chemical conditions of the “wet” storage as well as criteria and mechanisms to evaluate the fuel degradation during storage and technology for “dry” storage. The “dry” storage of spent fuel is an alternative to “wet” storage but it does not exclude preliminary cooling of SNF in water to allow decrease in its radioactivity and concomitant heat release. Dry storage has economical and operating advantages. It has not been implemented on a large scale yet but there are some positive results on dry storage of CANDU reactor fuel in Canada, Magnox fuel in Great Britain, and LWR and LMFBR fuel in the USA. Several options for SNF dry storage are being considered, which can be divided into three main categories: containers, storage tanks, and dry wells. 3.1 DRY STORAGE PROGRAM AT SSC RF RIAR SSC RF RIAR has initiated a program of experimental tests of dry container storage of RR SNF using the TUK-108/1 container. The TUK is a cylindrical metal structure with a bottom, consisting of three welded shells that form a body having a welded collar in the upper part. The collar is the supporting element for protective sealing of the inner and outer lids. The space between the shells is filled with reinforced, extra-strong, extraheavy concrete. Parameters of the TUK-108/1 container are given in Table 1.
214
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
TABLE 1. Parameters of the TUK-108/1 container Parameter Total height Outer diameter Inner diameter Cavity height Radial body thickness Lid thickness Bottom thickness
Value (mm) 4,490 1,645 775 3,474 435 190 410
In the inner part of the container there is a spacer grid that is a removable element and represents a welded structure made of diaphragms. The holes in the diaphragms make channels that provide the location and spacing of 7 ducts with FA. Seven ducts were developed and fabricated to locate the MIR SFA in the TUK. The duct represent a welded structure 242 mm in diameter, a bottom and four tubes 89.3 mm in diameter to locate one MRI FA per duct; the duct is covered with a lid. Figure 1 shows the MIR fuel assembly.
Figure 1. Fuel assembly for MIR reactor
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
215
3.1.1 Scope of dry storage program The program at SSC RF RIAR is aimed at being a trial of SFA dry storage technology under an advanced surveillance mode that uses the TUK-108/1 container, as well as determining the permissible time and conditions for dry storage of U-Al fuel. The following tasks should be solved during the course of the program: • • • • • • • •
•
Justification, selection and certification of SFA intended for testing under the dry container storage conditions Material science investigations and evaluation of the conditions of the selected SFAs before their location into the dry storage container stand Trial of the loading–unloading and transport technologies using separate SFAs, ducts and container Trial of the SFA drying technology before dry storage Trial of the technology of visual control, sipping control, and physical inspection of SFAs before dry storage Trial of the monitoring technology of SFA condition during dry storage Experimental examination of different modes (temperature and inner medium) of the RR SFA dry storage Periodic evaluation of the SFA conditions after the stage-by-stage testing at the surveillance stand using destructive and non-destructive control means and justification of the permissible time period and modes of RR SFA dry storage Evaluation of the conditions of the container structural materials, including the sealing unit, spacer grid, SFA wrappers, container body, as well as concrete that fills the space between the body coatings
The objects of investigation were standard MIR SFAs after operation and wet storage from 4 to 40 years. The reference and other SFAs and fuel rods are supposed for removal and further storage every 3, 7, 15, and 30 years. 3.1.2 Experimental support stand The experimental support stand comprises: • • • • •
Experimental TUK-108/1 Special removable elements to locate ducts with SFA Selected SFA Open platform with devices to fix the container Room with a measurement area and devices to control the stand parameters
216
•
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
A set of special equipment to perform SFA loading/unloading and operations with the TUK-108/1 container
The stand provides control over the following parameters: cladding temperature of RR SFAs; the gas medium in the inner space; the inner wall and outer surface of the metal–concrete container (MCC); fission product release during the long-term storage, FP activity measurement, and detection of leaking fuel rod claddings; air humidity in the MCC inner space; gas medium pressure in the MCC inner space; dose rate on the surface of the MCC external lid; leak tightness of the main unit of the MCC connector; and hydrogen release from the concrete. At present, a large scope of preparatory work has been performed. All the required documentation has been assembled. Examination of calculations was performed to verify nuclear and radiation safety and temperature conditions of dry storage for different SNF types loaded in the container stand. Seven ducts were fabricated to locate SFAs as well as coordinating plate to load the ducts into MCC, components and units of the transport technology, and process systems of the stand, electric furnace and protective reloading container. These elements are the components of SFA duct drying system. Some lacking units and systems for the stand are being fabricated now. 3.1.3 SFA inspection stand To evaluate the appearance of the SFAs and perform sipping control after the long-term storage in the spent fuel pool, an inspection stand was fabricated1; its major features are shown in Figure 2. KEY 1: upper sealing lid; 2: TV camera; 3: biological shielding; 4: sipping control case; 5: SFA stoppers; 6: body; 7: insulated electric heater; 8: drainage valve; 9: drainage valve drive; 10: unit base; 11: discharge manifold; 12: inlet manifold (for washing solution and air); 13: lid drive Figure 2. Unit for sipping control and process preparation of SFAs
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
217
The inspection stand comprises several units intended for measuring, mechanical purification of SFAs, sipping control of ducts, and preparation for SFA processes. The measurement unit is a protective metal cylinder consisting of a semiconductor detector and a collimator to measure the intensity of gamma-ray emission from SFAs, a device for the dose rate measurement, and a TV system for SFA inspection and surveillance of the inspection process. The unit of mechanical purification is used to grip an SFA and prevent it from rotating during purification, which is performed by a metal brush on a shaft driven by a worm-gearing electric motor. The brush can be moved perpendicular to the SFA axis depending on the SFA diameter. The sipping control unit consists of a cylindrical protective body that is covered by a sealed lid. Inside there is a case to locate the examined SFA. The case is equipped with a mechanism to fix the SFA in the vertical position. The TV camera controls the SFA positioning. In case of SFA damage, the sipping control unit is removed together with the assembly, but at the same time it provides a reliable isolation of the SFA from the environment. The sipping control of the fuel-rod cladding is performed by the 85Kr registration. The gas fission product release is stimulated by heating the SFA with the electric heater. The maximum temperature that can be achieved on the SFA surface is 230°C. Washing and drying of the SFA can be done without removal from the case. 3.1.4 Reloading container The reloading container is intended for the SFA transportation to separate inspection units of the stand,1 it is shown in Figure 3. The inspection stand can be used for checking RR SFA 68–200 mm in diameter and 910–2,550 mm in length. KEY 1: hoist; 2: reloading container; 3: biological shielding; 4: SFA; 5: grips to hold SFA during purification; 6: hole to install a detector and collimator; 7: mechanical brush.
Figure 3. Reloading container and measurement unit with the mechanical purification device
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
218
3.1.5 SNF characteristics
The MIR SFAs intended for loading into the dry storage container stand are subjected to certification. Below, there are given some data on characteristics of MIR SFAs removed from spent fuel pools, and the parameters of their operation and wet storage. Tables 2 and 3 present the characteristics and parameters of U-Al SFAs stored in the SSC RIAR central storage facility. Tables 4 and 5 present data on the water-chemical conditions of the MIR reactor coolant and pools. Figures 4–6 present the MIR fuel rod conditions after their operation and long-term storage in water. Table 6 gives the radiation characteristics of a 50 atom% burn-up MIR SFA after unloading from the reactor and cooling in the spent fuel pool. TABLE 2. Parameters of U-Al type SFAs in the SSC RIAR central storage facility2 Research reactor
Fuel type
% U235
No. fuel rods in SFA
U02
MIR
ARBUS
Cladding material
4
UAl Alloy U02 U02 А1 UA13 UA13 UA14
90 90 90 36
Al alloy SAV-6
4 6
Al alloy SAV-6
3–5
90
Storage time (years)
Mean atom% burn-up
16.1
37.5
23.4 28.3 16.2 19 33.6
40.6 38.4 4.4 2.1 5.9
22.6
3.9
TABLE 3. First loading of MIR SFA for dry storage trial in TUK-108/1
SFA type Fuel/Cladding
No. 1 6-tube UA1x + A1 SAV-6
Number of duct in container stand No. 2 No. 3 No. 4 No. 5 6-tube 4-tube 4-tube 4-tube UA1x + UA1x UO2 + UO2 + A1 + A1 A1 A1 SAV-6 SAV-6 SAV-6 SAV-6
Years of wet storage
36–37
22–23
22–23
4–5
4–5
4–5
4–5
Place of wet storage
Central storage pool
Central storage pool
Central storage pool
MIR pool
MIR pool
MIR pool
MIR pool
Parameter
No. 6 No. 7 4-tube 4-tube UO2 + UO2 + A1 A1 SAV-6 SAV
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
219
TABLE 4. Characteristics and quality of MIR coolant3 Parameters and indices water quality Pressure: at the channel inlet at the channel outlet Temperature: at the channel inlet at the channel outlet Supply рН Specific electric conductivity Dry residual Concentration of: chlorides ferrum ions aluminum copper Gas content (nitrogen) Oxygen content
Units MPa
Value 1.1–1.3 0.6–0.8
ºС
kg/s µSm/cm mg/kg mg/kg
40–60 50–95 8.3–22.2 5.6–6.0 <1.5 <1.5
ncm3/kg ncm3/kg
<0.02 <005 <0.04 <0.01 <50 <0.02
TABLE 5. Indices of water quality in the SNF cooling and storage pools2 Indices
MIR cooling pool 5.5–6.0 0.8–1.5
рН Specific electric conductivity (µSm/cm) Mass concentration (µg/kg): chlorides — iron 6–20 copper 5–12 Ag, Hg and other heavy metalsb Aluminum 5–20 — Total rigidity (µg -equiv/kg) Specific activity of γ-emitting radionuclides (3.7–37)10 4 (Bq/kg)c Water temperature (°С) 25–35 Time of SNF cooling before sending to the 1 year central storage facility a рН up to 8 is allowed b Heavy metal ions are very aggressive in pitting corrosion of Al c Activity increase shows the cladding corrosion
Central storage facility 6,4–6,5a 3,7–4,4 90–100 760–980 — — 40–60 (1.8–2.2)106 15–25 —
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
220
Figure 4. Appearance of a MIR SFA after operation and wet storage for 31 years
×2 ×5 Figure 5. Surface of the MIR fuel rod with rash-type defects at different magnifications after operation and wet storage for 31 years (a)
(b)
Figure 6. Macrostructure of MIR fuel-rod cladding (a), and microstructure of oxide film on fuel rod surfaces (b) after operation and wet storage for 31 years
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
221
TABLE 6. Radiation characteristics of a 50 atom% burn-up MIR SFA Nuclide
Cooling time (years) 3
5
10
20
30
50
Kr-85
1.33E + 12
1.17E + 12
8.47E + 11
4.44E + 11
2.32E + 11
6.37E + 10
100
Sr-90
9.25E + 12
8.81E + 12
7.80E + 12
6.12E + 12
4.80E + 12
2.95E + 12
8.73E + 11
Y-90
9.25E + 12
8.81E + 12
7.80E + 12
6.12E + 12
4,80E + 12
2.95E + 12
8.73E + 11
2.51E + 09
Y-91
1.80E + 09
3.16E + 05
1.29E–04
—
—
—
—
Zr-95
5.60E + 09
1.97E + 06
4.56E–03
—
—
—
—
Nb-95
1.24E + 10
4.35E + 06
1.01E–02
—
—
—
—
Tc-99
1.40E + 09
1.40E + 09
1.40E + 09
1.40E + 09
1.40E + 09
1.39E + 09
1.39E + 09
Ru-106
3.19E + 12
8.16E + 11
2.70E + 10
2.96E + 07
3.24E + 04
—
—
Rh-106
3.19E + 12
8.16E + 11
2.70E + 10
2.96E + 07
3.24E + 04
—
—
Ag-110m 1.67E + 09
2.22E + 08
1.42E + 06
—
—
—
—
Cd-113m 1.74E + 09
1.57E + 09
1.22E + 09
7.29E + 08
4.37E + 08
1.57E + 08
1.22E + 07
Sn-119m 3.08E + 08
3.91E + 07
2.24E + 05
—
—
—
—
Sn-123
4.13E + 08
8.20E + 06
4.55E + 02
—
—
—
—
Sb-125
6.38E + 11
3.87E +11
1.11E + 11
9.06E + 09
7.42E + 08
4.96E + 06
1.83E + 01
Te-125m 1.48E + 11
8.95E + 10
2.56E + 10
2.10E + 09
1.72E + 08
1.55E + 06
4.22E+00
3.65E + 09
3.51E + 07
3.17E + 02
—
—
—
—
Te-127m 3.73E + 09
3,58E + 07
3,24E + 02
—
—
—
—
Te-127
Cs-134
1.54E + 12
7.82E + 11
1.45E +11
4.94E + 09
1.69E + 08
1.97E + 05
Cs-135
1.49E + 08
1.49E + 08
1.49E + 08
1.49E + 08
1.49E + 08
1.49E + 08
1.49E + 08
Cs-137
9.96E + 12
9.51E + 12
8.48E + 12
6.74E + 12
5.36E + 12
3.38E + 12
1.07E + 12
5.07E + 12
3.20E + 12
1.02E + 12
1.26E + 10
6.38E + 07
1.17E + 02
Ba-137m 9.43E + 12
9.01E + 12
8.03E + 12
6.38E + 12
2.01E + 13
3.39E + 12
3.96E + 10
5.40E + 06
Pr-144
2.01E + 13
3.39E + 12
3.96E + 10
5.40E + 06
Pm-147
1.57E + 13
9.28E + 12
2.48E + 12
1.77E + 11
Sm-151
1.51E + 10
1.49E + 10
1.44E + 10
1.33E + 10
1.23E + 10
1.06E + 10
7.17E + 09
Eu-154
1,12E + 11
9.55E+10
6.39E + 10
2.86E + 10
1.28E + 10
2.55E + 09
4.52E + 07
Eu-155
7.39E + 10
5.59E + 10
2.78E + 10
6.87E + 09
1.70E+09
1.04E + 08
9.53E + 04
Total
1.04E + 14
5.64E + 13
3.60E + 13
2.60E + 13
2.03E + 13
1.26E + 13
3.85E + 12
Ce-144
Figure 7 presents data on the SFA heat release change with time calculated with the help of different programs. The residual heat release of the assembly does not exceed 2W. Moreover, actual operating temperatures will not exceed 180°C for an ambient temperature of 50°C under dry SFA conditions.
222
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
Figure 7. Residual heat release in an irradiated MIR SFA
Calculations show that nuclear safety requirements for container operation are met for all accidents. In particular, for the design-basis accident, when the container is filled with water and fragments of damaged SFA fall on the duct bottoms, the effective coefficient of system multiplication from an infinite number of containers with 28 MIR SFA does not exceed 0.95. Regulatory requirements on nuclear safety are therefore satisfied by the design decisions. Safety calculations for the container stand used for examination of SFAs showed that under normal operation the release of radioactive products from the container is several orders of magnitude lower than the admissible level. Analysis of emergencies showed that for possible accidents, including collapse of the container, no damage will occur to the SFA, while the design of ducts and inner devices excludes a criticality event. The radiation consequences of design-basis and beyond-the-design-basis accidents are slight: expected releases are several orders of magnitude lower than the admissible level. 4 Conclusion The developed container stand allows a full-scale trial of the technology for dry storage of RR SNF while satisfying all safety requirements.
STORAGE TRIAL FOR RESEARCH REACTOR SPENT FUEL
223
References 1.
2.
3.
Grachev, A.F., Ladzin, A.S., Konyashov, V.V. et al., “The Test Rig for Inspection of SFAs of Research Reactors of the SSC RF RIAR”, 7th International Topical Meeting on Research Reactor Fuel Management, March 9–12, 2003, Aix-enProvence, France. Grachev, A.F., Kalygin, V.V., Ladzin, A.S. et al., “Experience of Handling SNF at SSC RF RIAR”, Regional Workshop on Chracterization, Management and Storage of Spent Fuel from Research and Test Reactors, May 8–12, 2000, Swierk, Poland. Alexandrov, V.V., Grachev, A.F., Izhutov, A.L. et al., “Operating Experience of MIR FAs”, 6th International Topical Meeting on Research Reactor Fuel Management, March 17–20, 2002, Ghent, Belgium.
INTERIM STORAGE AND LONG-TERM DISPOSAL OF RESEARCH REACTOR SPENT FUEL IN THE UNITED STATES
N. C. IYER, ∗ D. W. VINSON, R. L. SINDELAR J. E. THOMAS, AND T. M. ADAMS Savannah River National Laboratory, USA Abstract: Aluminum clad research reactor spent nuclear fuel (RR SNF) is currently being consolidated in wet storage basins (pools). Approximately 20 metric tons (heavy metal) of aluminum-based spent nuclear fuel (Al-SNF) is being consolidated for treatment, packaging, interim storage, and preparation for ultimate disposal in a geologic repository. The storage and disposal of AlSNF are subject to requirements that provide for safety and acceptable radionuclide release. The options studied for interim storage of SNF include wet storage and dry storage. Two options have also been studied to develop the technical basis for the qualification and repository disposal of aluminum spent fuel. The two options studied include direct disposal and melt-dilute treatment. The implementation of these options present relative benefits and challenges. Both the direct disposal and the melt-dilute treatment options have been developed and their technical viability assessed. Adaptation of the melt-dilute technology for the treatment of spent fuel offers the benefits of converting the spent fuel into a proliferation resistant form and/or significantly reducing the volume of the spent fuel. A mobile melt-dilute system concept has emerged to realize these benefits and a prototype system developed. The application of the melt-dilute technology for the treatment of legacy nuclear materials has been evaluated and also offers the promise for the safe disposal of these materials. Keywords: Al-based spent fuels, wet storage, basin storage, dry storage, repository disposal, melt-dilute, mobile system, high-enriched spent fuel
______ *To whom correspondence should be addressed: Natraj C. Iyer, Savannah River National Laboratory,Aiken, SC 29808, USA; e-mail:
[email protected] 225 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 225–242. © 2007 Springer.
226
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
1 Introduction Aluminum-clad research reactor spent nuclear fuel (RR SNF) is currently being consolidated in wet storage basins (pools). Approximately 20 metric tons (heavy metal) of aluminum-based spent nuclear fuel (Al-SNF) is being consolidated for treatment, packaging, interim storage, and preparation for ultimate disposal in a geologic repository. The sources of the Al-SNF are domestic research reactors (DRR), and foreign research reactors (FRR). This spent fuel contains uranium and highly enriched uranium (HEU) that originated in the USA. The storage and disposal of Al-SNF are subject to requirements that provide for safety and acceptable radionuclide release. A number of alternative technologies have been developed to constitute the Al-SNF into a waste form acceptable for ultimate disposal in a geologic repository. These technologies include direct disposal using road ready packages and melt-dilute treatment of the highly enriched Al-SNF. The direct disposal technology offers the capability of treating and packaging the spent fuel in its existing form without altering the composition of the spent fuel. The melt-dilute treatment technology, on the other hand, offers the potential for diluting the HEU into a proliferation-resistant form prior to packaging and disposal. The melt-dilute technology can also be readily adapted to treat many Department of Energy (DOE) legacy waste streams with significant process versatility and modularity. The legacy material waste streams include the following: depleted uranium, HEU, streams containing organic materials including metallurgical mounts, and streams containing small amounts of plutonium, etc. This paper will provide an overview of the technologies developed in the USA for the geologic disposal of Al-SNF. It will also detail the development of alternative platforms for the emerging melt-dilute technology and its application to both SNF and legacy nuclear materials treatment. 2 Aluminum-clad Research Reactor Fuel The majority of the research reactor SNF assemblies consist of uraniumaluminum fuel cores encased in aluminum clad. The fuel consists of uranium aluminide particles in an aluminum matrix. Some fraction of the Al-SNF assemblies in the inventory also have cores of uranium silicide, uranium oxide or uranium carbide particles in an aluminum matrix. The uranium enrichment ranges from 20% to 90% and the burn-up of the spent fuel ranges from 30% to 70%. Figure 1 shows typical geometries and configurations of aluminum clad fuel assemblies.
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
227
Figure 1. Geometry of aluminum-clad SNF assemblies
3 Interim Storage of Aluminum-clad Spent Fuel Interim storage of Al-SNF will likely be needed for several decades since the licensing, construction and start-up of the geological repository is a long process. Dry storage in a monitored retrievable storage (MRS) facility and extended basin (wet) storage are the two alternatives for interim storage. 3.1 WET BASIN STORAGE The Al-SNF may be stored directly in wet storage basins provided that the basin water chemistry controls are in place. The typical basin standards for wet storage of Al-SNF are listed in Table 1. The wet basin chemistry limits are procedure limits and in order to ensure procedural compliance periodic water chemistry analyses are performed. Initial laboratory testing has shown this basin chemistry to be nonaggressive1 and basin surveillance confirms that present storage conditions are indeed nonaggressive.2 Radionuclide release from exposed fuel meat in water storage is another issue that must be managed for wet basin storage. For example, a typical basin limits to release from Al-SNF in basin storage is 20.7 µCi/h from an assembly for a large, well managed basin. Fuel can be cropped to expose the fuel meat if the expected release is below 20.7 µCi/h. Maintenance of a wet basin at or
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
228
TABLE 1. Table of basin chemistry standards and present conditions Parameter Conductivity pH Chloride Cs-137 Alpha Tritium Al Cu Fe Hg
Limit 10 µmho/cm 5.5–8.5 0.1 ppm 1000 dpm/ml 10 dpm/ml 1.0 µCi/ml 1.0 ppm 0.1 ppm 1.0 ppm 0.1 ppm
Typical Value 1 µmho/cm 6.2 <0.1 ppm (laboratory detection limit) 13 dpm/ml <1 dpm/ml (laboratory detection limit) 0.035 µCi/ml <0.05 ppm (laboratory detection limit) <0.05 ppm(laboratory detection limit) <0.05 ppm(laboratory detection limit) <0.002 ppm (laboratory detection limit)
below these level aids in ensuring that not only is water chemistry controlled to prevent fuel degradation but also limits the potential for personnel exposure/ contamination events. 3.2 DRY STORAGE Two distinct types of dry storage systems are envisioned: a “sealed system” which would store fuel assemblies in fully-sealed containers and a “nonsealed” system which would store assemblies in nonsealed containers or holders open to the environment of the facility. If a sealed system is used, the seal must be of adequate design to last for the duration of the storage period (e.g., 50 years) or allowance (e.g., cost) must be made for resealing. The need for a sealed storage system, in a dry storage facility will be driven by the requirements of confinement barriers including the acceptable level of radionuclide release from a confinement barrier and the number of barrier layers. The cladding itself provides one such confinement barrier with the release of radionuclides limited by the criteria for acceptable degradation. A sealed storage system is one in which a fully sealed container enclosing one or more fuel assemblies is placed within a dry storage facility. The approach to avoid excessive degradation in a sealed system is to dry the contents to a level of free water (remaining in the container to be sealed) such that if all water is fully consumed by corrosion of the fuel, that: (1) the conditions of acceptable degradation are not exceeded; (2) the production of hydrogen does not pose a threat to post-storage retrieval due to build-up of hydrogen to levels that could result in a deflagration event; and (3) the production of hydrogen does not pose a threat to post-storage retrieval due to production of pyrophoric substances. The temperature (upper) limit is based on the most limiting noncorrosion degradation mechanism.
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
229
A nonsealed system is one in which fuel would be in contact with the ambient air in a dry storage facility (e.g., outside air temperature and relative humidity conditions). Since the corrosion of aluminum exposed to ambient air progresses at a slow rate, even at high humidities (up to 100%) at near room temperatures, storage in a nonsealed system is feasible in “nondirty” atmospheres (those containing low chloride and sulphate compounds). It is not possible to produce high humidities (up to 100%) at high temperatures (up to 200°C) in a nonsealed system and therefore cladding materials would not experience the corrosion rates observed in corrosion testing program.3 The requirements for interim dry storage are based on providing for safe, retrievable storage. Retrievability is directly related to limiting degradation in a storage system. In developing the requirements for interim dry storage, limited fuel degradation was acceptable, consistent with the requirement of no gross rupture of fuel cladding during storage and post-storage handling since the fuel was assumed to be eventually removed from interim storage. 3.2.1 Fuel drying for a dry storage system The drying requirement for interim dry storage is based on avoiding hydrogen gas, H2, build-up during storage in sealed canister. H2 is generated through corrosion reactions. In a sealed storage system, H2 build-up gives a more stringent limit for free water in the canister than does corrosion consumption of the material. The total amount of water available for corrosion arises from three basic sources. These sources are: (1) free water (which includes water in pits in the cladding, crevices, etc.); (2) waters of hydration on existing oxide; and (3) adsorbed waters on oxide and on aluminum. The amount of adsorbed water on the surface of a metal varies with the relative humidity and the temperature. Volpe4 determined that at 20°C, the amount of absorbed water is approximately 20 monolayers at 100% relative humidity. This is much less than the water of hydration in a 50 µm Boehmite film, and is therefore negligible. Complete dehydration of the hydrated oxide layers is neither readily achievable nor necessary in drying. The waters of hydration from Boehmite (assumed to be released in storage due to radiolysis) for a film of 50 µm are assumed to be released and to corrode the aluminum cladding to form Al2O3. It does not recombine with the original Al2O3 left from dehydration of Boehmite. This results in 0.0002 inches of aluminum corrosion. Therefore, the maximum possible uniform consumption of aluminum dried to 1 ml/0.1 m2 of cladding surface in a sealed system is approximately 0.0003 inches. The limiting drying criterion is designed to avoid hydrogen build-up. At temperatures above approximately 80°C, hydrogen build-up occurs in a closed system containing aluminum and water according to5:
230
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
2Al + 4H2O = Al2O3•H2O + 3H2 Assuming the reaction goes to completion, 1 ml of H2O yields 0.042 moles of H2. The generation of hydrogen in the above mentioned reaction to produce boehmite bounds that for the reaction to produce gibbsite at temperatures below approximately 80°C. A general formula can be derived to relate the volume fraction of free water to the hydrogen pressure for the reaction at completion5: FW/V = 292505.PH2/(273.15 + T),
(1)
where FW is the free water volume in ml V is the volume of the container in m3 PH2 is the pressure due to H2 in atmospheres T is the temperature in °C One impact of hydrogen buildup is the potential for an explosion hazard. The lower concentration limit of flammability of hydrogen is 4% by volume in air at room temperature. The lower concentration level for a sustained burn of hydrogen in air is approximately 9% by volume in air. The concentration level at which a hydrogen/air mixture is explosive is 18%. Therefore, the partial pressure of hydrogen must be below 0.59 psia or the ratio of free water (ml) to the volume of the container (m3) must be kept below 39 to ensure that hydrogen at 4% by volume in air is not produced in a container. Another impact of hydrogen buildup is the potential for production of UH3, a compound that is pyrophoric under certain conditions. Dispersoids such as UAlx would not be reduced by expected hydrogen pressures to produce UH3; however, oxides of uranium could be if the partial pressure of H2O is low enough and the partial pressure of H2 is high enough in an H2O/H2 system. However, because aluminum surrounds the oxides in the dispersoid fuel and does not allow direct contact with the fuel particles, this is not expected to result in significant production of UH3. Considering the potential for existing UH3 on uranium metal fuels retrieved from basin storage, only uranium metal fuels may need to be stabilized. The INEL has developed a stabilization treatment with the technical bases6 to convert UH3 to uranium oxide. No significant amount of UH3 is expected to be present in Al-based fuels. In addition, the stringent drying requirements ensure that the expected H2 build-up is extremely low so that UH3 production due to H2 gas contacting exposed fuel meat is negligible.
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
231
3.2.2 Dryness specification To limit H2 build-up in a sealed canister to 4% by volume from free water, the maximum allowable free water (Wwater, in grams) is expressed as a function of the free volume (V, cm3) of the container. Wwater = 3.873x10–5 V.
(2)
3.2.3 Vacuum drying specification The Al-SNF must be vacuum-dried to at least 5 torr at an internal chamber temperature of 25°C or higher. The vapor pressure of water at 25°C is 24 torr, and any free liquid water on the fuel will be vaporized before 5 torr is reached.7 To ensure dryness, the vacuum pump should be isolated following drying and the chamber pressure must remain at ≤5 torr for 15 min to confirm dryness. Drying tests using unheated vacuuming to dry an instrumented canister containing residual free water and a mock fuel assembly were reported.7 These tests were successful at drying the assembly and canister, and showed that temperature, pressure, or relative humidity could be used as measures of free water removal. Applying a warm air purge during the vacuum drying improved the drying method. Recent field experience with two instrumented, shielded SNF test canisters indicates a continuous warm air purge (<74°C) under vacuum provides satisfactory dryness in a reasonable time interval. A canister with one assembly and ~0.6 pints of water was dried to ~0.25 torr in 2.5 h. The vacuum pump should have a pumping speed of at least 100 cfm and a rated ultimate vacuum of at least 0.5 torr. An air-cooled vacuum pump is recommended to eliminate the need for cooling water disposal with a watercooled pump. A water-sealed vacuum pump is recommended to eliminate the need for waste oil disposal with an oil-sealed pump. Redundant vacuum sensors should be placed close to the canister vacuum nozzle connection. The warm air purge system should have a thermostatically controlled heater and a flow of at least 25 cfm. Temperature monitoring must be provided for air entering and exiting the canister. Direct temperature monitoring of the canister bottom is desired. The drying temperature should not exceed 250°C to avoid the potential for hydrogen blistering and gross cladding failure. The drying specification calls for a low-temperature heated-air vacuum drying. The lower temperature during the drying process helps avoid the potential for blistering of the Al-SNF materials caused by H2 via vapor corrosion at high temperature (at and above 250°C).
232
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
4 Geologic Disposal for Al-clad Research Reactor Spent Fuel The path envisioned for ultimate disposition of the Al-SNF assemblies involves transfer and treatment of wet-stored assemblies into an Al-SNF form suitable for the geologic repository. Two options include packaging the fuel in either a “direct” or “melt-diluted” form in a sealed canister. The canisters would be in a “road ready” package, i.e., it would be suitable for interim storage for up to 40+ years prior to repository disposal. The canisters would be transported to the repository and placed into waste packages for ultimate disposition. The qualification of the road ready package for either of these two options include documentation, analyses, and/or validation of the spent fuel characteristics, criticality controls, corrosion performance and thermal analysis in the context of the repository performance assessment. The technical basis for the qualification and repository disposal of aluminum spent fuel has been developed for both the propose options. 4.1 DIRECT DISPOSAL TECHNOLOGY Direct disposal technology involves drying the SNF to remove the adsorbed and hydrated water, packaging and sealing the SNF in a canister, which has a diameter of approximately 17 inches and a length of approximately 120 inches. The storage criteria for Al-SNF in a road-ready package support the basis for both interim storage and repository requirements. The canister of SNF will be vacuum-dried and back-filled with helium. The SNF will be separated in the canister with a basket containing neutron absorber materials. Three to four baskets would be stacked within each canister. After the canister is back-filled and sealed, it will be temporarily stored in horizontal concrete storage modules. Ultimately, the canisters will be shipped to a federal Mined Geologic Disposal System (MGDS) repository for final disposal. There each of the SNF canisters will be placed inside a larger waste package containing five Defense Waste Processing Facility (DWPF) high-level waste (HLW) canisters before being emplaced in the repository. Exposure of Al-SNF forms to environments that may be present in the waste package will cause changes in the forms from their initial condition. AlSNF degradation may result in release of radionuclides from the SNF matrix and reconfiguration of fissile species within the engineered barrier system (EBS). This may directly affect the performance of the proposed repository. Degradation of the Al-SNF and reconfiguration of fissile materials controlled by the thermochemical stability and solubility of the many possible uranium compounds and rates of the many competing reactions has been extensively studied. Based on the natural occurrence of uranium bearing minerals within ore deposits in the western USA, thermochemical data, and the products
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
233
formed during laboratory corrosion experiments, the hydrated oxides and silicates of uranium and hydrated aluminum oxides or alumino-silicates are the most likely final degradation products. The reconfiguration and redistribution of materials within the waste package have been analyzed to support the criticality analysis. Loading neutron poison materials in Al-SNF canisters has been assessed as a method to avoid criticality with HEU SNF. Candidate materials are borated stainless steel, dispersions of europium oxide, gadolinium oxide, or samarium oxide in stainless steel, and cadmium. Mechanical properties, corrosion resistance, neutron absorption properties, cost and availability are the major factors evaluated for selection of a poison material. A preliminary basis has been developed for direct co-disposal of Al SNF with neutron absorbers. 4.2 MELT-DILUTE TECHNOLOGY The melt-dilute treatment process option has been developed for ultimate disposal of SNF from FRR and DRR in the monitored geologic repository (MGR). Most of these fuels contain HEU (>20% 235U).8 The melt-dilute treatment melts the SNF in a furnace and dilutes with depleted uranium. Figure 2 shows a schematic of the process. Dilution of the SNF to reduce the U235 content of HEU to LEU levels lowers the potential for criticality. The product is isotopically diluted SNF that can be tailored to optimize degradation characteristics by addition of aluminum or other elements. Significant benefits also accrue from the ~70% volume reduction, resulting in fewer canisters to be stored and shipped for repository disposal compared to direct/co-disposal. The melt-dilute treatment also minimizes characterization requirements through erasure of the SNF’s history and acquisition of in-process characterization data. Depleted U
Spent Fuel Assembly Induction Furnace Melting
Aluminum
Offgas
Diluted Ingot <20% 235U
Storage Canister Geologic Repository (Dry Storage)
Figure 2. Process flow schematic for melt-dilute process
234
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
Advantages of diluting the uranium to below 20% U235 and the eutectic composition include: (1) lower process operating temperatures; (2) minimum gravity segregation in the casting; (3) lower volume of off gas products; and (4) lower process and materials costs. Compared to other dilution methods, the 20% dilution offers the greatest versatility because waste forms containing 5–67 wt% uranium can be produced to give package volume reduction of up to 70%. The melt-dilute treatment technology has been advanced through the construction and start-up testing phases for a pilot-scale remote facility. The scale of the pilot facility is ideally sized for the treatment of the aforementioned SNF and legacy materials streams. Optimization of the design has lead to two off-gas system options: (1) traditional SRS off-gas design with a combination of dry zeolite beds and HEPA filters; or (2) a closed evacuated self-contained melting system. The pilot-scale facility design offers the opportunity to optimize design for SNF and legacy materials and installation of such units at multiple locations within the DOE complex. Alternatively, a transportable mobile unit is also envisioned for the treatment of spent fuel and/or legacy waste materials. Such a system will be capable of being readily adapted and modified to meet requirements in different parts of the world. 5 Pilot-Scale Melt-Dilute Facility A pilot-scale facility was constructed in a Hazard Category 2 structure (Figure 3).9 The facility has a control room where operators observe process operations that are performed inside a hardened structure. The furnace is inside a stainless steel box, which acts as primary containment.
Figure 3. A pilot-scale Melt-Dilute Facility
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
235
The floor plan for the facility is shown in Figure 4. Shown in the Figure are the universal power supply, the crane aisle access corridor, the trailer well, furnace and associated equipment, and the control room. The induction furnace is located behind a shield wall to protect operators from radiation during entry into the trailer well. SNF is brought to the facility in an 8 t cask on a flatbed trailer. The roll-up door is opened and the trailer is backed into the trailer well. After removing the cask lid bolts, the trailer is removed and the door closed. Further operations are done remotely using the crane. After the cask lid is removed, the cask is moved to the unloading station where the fuel basket containing the SNF assembly is removed and moved to the furnace. The fuel basket tool is placed on guide pins located on top of a datum plate that is attached to the furnace; the fuel basket is then lowered into the graphite crucible containing the carbon steel liner. The unloading tool is then returned to the tool stand shown in Figure 5.
UP S
Crane Aisle A Cont l Bld
Trailer
Figure 4. Layout of pilot-scale Melt-Dilute Facility
Next the equipment plate containing two melt samplers, the primary zeolite bed and the crucible camera is lowered on top of the furnace crucible as shown in Figure 6. The guide pins on the datum plate help guide the equipment plate to align the zeolite bed and crucible. The equipment plate engages a robot end-ofarm tool changer on the datum plate to connect pneumatic and electrical lines for operation of equipment on the equipment plate. The containment box lid is put onto the enclosure and clamped in place using pneumatic clamps.
236
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
Figure 5. Unloading tool returning to the tool stand
Figure 6. Lowering the equipment plate onto the furnace using the overhead crane
The melt-dilute facility (MDF) is operated from the control room. All operations are observed on video screens located on the console and operating parameters are displayed on the center video screen. The operator controls the induction furnace start-up and power levels, the airflow into the furnace containment box and through the zeolite bed. The pressure drop across the containment boundary is monitored to maintain a negative pressure so all airflow is from the room into the box. Dilution air from inside the box is used to cool hot offgas so the exit gas temperature is less than 50ºC. The console is shown in Figure 7.
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
237
Figure 7. Control room for melt-dilute pilot-scale process
The facility was fully operational and start-up testing begun using aluminum assemblies. Several successful tests were made where data on the furnace, ventilation system, cooling water system, and uninterruptible power system were obtained. Furnace operations and all remote tooling functioned according to design. 6 Melt-Dilute Treatment Using a Mobile Platform The melt-dilute technology was converted into a mobile platform to develop the mobile MMD for treatment of SNF and/or legacy materials at storage locations around the world, thereby avoiding the cost of building treatment facilities at each site and shipment of HEU assemblies.10, 11 The mobile melter concept is based on SRNL tests and modular pilot-scale facilities at SRS for treatment of US SNF; it was developed in conjunction with Argonne National LaboratoryWest, 12 now Idaho National Laboratory (INL). Laboratory tests at SRNL showed the feasibility of operating both a closed and a filtered off-gas system. One concept for the MMD would be to build a facility that would utilize the closed system approach. A summary flow sheet schematic is shown in Figure 8. The MMD process simply involves: (1) loading spent fuel assemblies in a canister with depleted uranium; (2) welding a lid on the canister; (3) drying and evacuating the canister; and (4) melting the HEU fuel assemblies and diluting the U235/U–aluminum alloy to less than 20% enrichment in U235. After treatment, the sealed canister containing the solidified aluminumuranium ingot can be placed in interim storage pending reprocessing or emplacement into long-term storage. Thus, HEU material can be treated using the
238
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
MMD process to generate a safe and secure LEU ingot. As envisioned, the MMD system will be compact and staged on a transportable vehicle, with the capability to treat and encapsulate research reactor spent fuel at either the reactor or storage site.
Figure 8. Simplified flow sheet for MMD process
The furnace would be enclosed with an outer container or dome, similar to a reactor dome, to contain any volatile gases in the unlikely event the closed container leaked during melting. The dome may be designed to provide sufficient shielding so none is needed around the furnace area. Ideally, only one element would be melted at a time, but the system could be designed to melt 4–6 elements per batch. With one element the recipe for dilution and alloy composition control would be easier. It is expected that the furnace and controls would be located in separate international shipping containers that could be easily loaded onto trailers and unloaded at the work site for assembly. Loading and unloading of the furnace with spent fuel would be done remotely. Once the fuel assembly is brought to the MMD facility in a cask, it would be unloaded using a forklift or crane and placed onto a remote system to move it to the furnace. It is expected that the cask would provide radiological shielding while furnace loading, unloading, and transporting the spent fuel. The mobile melter can be transported in two over-the-road trailer assemblies, one trailer containing the melter and associated equipment, and the other, the control and forklift systems, allowing free movement between locations where SNF is stored, eliminating the need to transport SNF, and minimizing its handling. Only the melter and control system will move along roads, it is not necessary to move shielding or structures. If additional shielding for process
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
239
activities is needed, it can be easily erected at individual sites. In addition, the control and monitoring equipment are housed separately, keeping the more costly equipment in a clean environment at all times. Should contamination of the melter become a problem, the furnace can be replaced at minimal cost. Figure 9 shows how the equipment could be installed in shipping containers and staged at the storage facility. Inexpensive shielding, if needed, can be erected quickly and inexpensively with corrugated steel panels like those used to protect aircraft from strafing or ground attack. These panels can be erected to form cavities of predetermined thickness and filled with dirt. Control and power cables which must pass from the control unit to the melter can be installed through these cavities before they are filled, preventing radiation “shine”. This shortens setup time to make the system operational, leaving only positioning the trailers, connecting the cables, and testing the system. The melter was conceived as a closed system with a condenser to trap volatiles, but an off-gas system was added to avoid a pressurized system that contained radionuclides and to reduce waste volume. In addition, a system similar to the one developed at SRS could be designed for mobility. 7 Melt-Dilute Treatment of Legacy Materials Currently various classes of nuclear materials have no defined disposition pathway. They include off-spec HEU metal and oxides, surplus depleted and natural uranium, uranium-233, Pu-238 scrap, miscellaneous Pu-239 scrap, metal, and powder, and various SNF “cats and dogs.” Development of disposition pathways— which may include various stabilization, treatment, or disposition technologies— is paramount for an effective nuclear materials stewardship. In order to achieve this goal, the most appropriate technology must be identified to ensure effective management of the legacy materials inventory. One approach is to explore/ evaluate currently existing stabilization, treatment, and disposal technologies with respect to their ability to handle diverse new/alternate feed streams. The melt-dilute technology offers a way to treat and dispose of materials in the DOE-NMS program 13 and it could treat a variety of materials. The pilotscale design or the modular mobile design can be adapted to a “mini-batch” system which could be turnkey fabricated and placed at any site needing SNF treatment and disposition,15 or installed as a mobile treatment system.
240
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
Figure 9. Modular transportable melt-dilute system for SNF
Summary About 20 metric tons (heavy metal) of aluminum-clad research reactor SNF (Al-SNF) is currently being consolidated in wet storage basins, with the goal of treatment, packaging, interim storage, and preparation for ultimate disposal in a geologic repository. The handling of this Al-SNF is subject to requirements that provide for safety and acceptable radionuclide release. Both wet and dry interim storage of SNF is being considered. Two options have also been studied to develop the technical basis for the qualification and repository disposal of Al-SNF, namely direct disposal and melt-dilute treatment. The implementation of these options offers benefits and challenges. Both options have been developed and their technical viability assessed. Adaptation of the melt-dilute technology allows conversion of SNF into a proliferation-resistant form and of significantly reducing its volume. A mobile melt-dilute system concept has been developed and a prototype system designed. The application of the melt-dilute technology for the treatment of legacy nuclear materials has been evaluated and also offers the promise for safe disposal of these materials.
MANAGING RESEARCH REACTOR SPENT FUEL IN THE U.S.
241
References 1.
2.
3.
4 5 6 7
8.
9.
10.
11.
12.
13.
14.
Chandler, G.T., Sindelar, R.L., Lam, P.S., “Evaluation of Water Chemistry on the Pitting Susceptibility of Aluminum”, National Association of Corrosion Engineers International, Paper No. 104 at CORROSION/97, Houston, TX, March 1997. “Corrosion Surveillance in Spent Fuel Storage Pools”, National Association of Corrosion Engineers International, NACE’97, Paper 107, Houston, TX, March 1997. Peacock, H.B., Jr., Sindelar, R.L., Lam, P.S., Murphy, T.H., “Evaluation of Corrosion of Aluminum-Base Reactor Fuel Cladding Materials During Dry Storage (U)”, Savannah River Technology Center, Aiken, South Carolina, NAC CORROSION 96, Paper 96134. Volpe, L., Proceedings of the 10th International Congress on Metallic Corrosion, 1987. Lam, P.S., Sindelar, R.L., Corrosion of Aluminum—Uranium Alloys in Water Vapor at 200C, Scientific basis for Nuclear Waste Management XXII, 1998. Ebner, M.A., “The Potential Pyrophoricity of BMI-SPEC and Aluminum Plate Spent Fuels Retrieved From Underwater Storage”, INEL-96/0235. Large, W.S., Sindelar, R.L., “Vacuum Drying Methods for Spent Nuclear Fuel”, Proceedings of 1998 High-Level Radioactive Waste Management Conference, 1998. Adams, T.M., Peacock, H.B., Jr., Sindelar, R.L., Iyer, N.C., Swift, W.F., Rhode, F.C., Brook, H.M., “Melt-Dilute Treatment Technology for Aluminum-Based Research Reactor Fuel”, Proceedings of the ANS-DOE Spent Nuclear Fuel and Fissile Management, 2000, pp. 41–45. Peacock, H.B., Jr., Fisher, D.L., Adams, T.M. et al., “Development of a Pilot-Scale Facility for Melt-Dilute Treatment of Spent Nuclear Fuel”, ANS DOE Spent Nuclear Fuel Conference, Charleston, SC, September 2002. Adams, T.M., Peacock, H.B., Jr., Fisher, D.L. et al., “Development of the Mobile Melt Dilute Technology for the Treatment of Russian Research and Test Reactor Fuel”, GLOBAL 2003, Atoms for Peace, New Orleans, LA, September 2003. Adams, T.M., Peacock, H.B., Jr., Iyer, N.C., “Mobile Melt-Dilute Treatment for Russian Spent Nuclear Fuel”, ANS DOE Spent Nuclear Fuel Conference, Charleston, SC, September 2002. Peacock, H., Fisher, D., Adams, T., Sindelar, R., Iyer, N., Sell, D., Allen, K., Howden, E., Westphal, B., “A Mobile Melt-Dilute Module for the Treatment of Aluminum Research Reactor Spent Fuel”, RERTR 2004. Adams, T.M., Peacock, H.B., “Treatment of Legacy Materials Using the MeltDilute Treatment Technology”, ANS DOE Spent Nuclear Fuel Conference, Charleston, SC, September 2002. Adams, T.M., Peacock, H.B., Jr., Williams, W. et al., ORNL-DES-03, Engineering Study of the Disposition of Uranium-233: Uranium-Aluminum Melt-Dilute Alternatives, September 2000.
MATERIALS STABILITY ISSUES OF SPENT FUEL STORAGE
MANAGING SPENT FUEL IN WET STORAGE AT THE SAVANNAH RIVER SITE
R. L. SINDELAR,∗ P. R. VORMELKER, R. W. DEIBLE, AND J. E. THOMAS Savannah River National Laboratory, USA Abstract: Spent nuclear fuels (SNFs) received from reactor sites around the world are being stored in the L-Basin at the Savannah River site (SRS) in Aiken, South Carolina. The predominant fuel types are research reactor fuel with aluminum-alloy cladding and aluminum-based fuel. Other fuel materials include stainless steel and zircaloy cladding with uranium oxide fuel. Standing chemistry control and corrosion surveillance programs have been established and upgraded since the early 1990s to minimize corrosion degradation of the aluminum cladding materials, so as to maintain fuel integrity and minimize personnel exposure from radioactivity in the basin water. Recent activities that have been initiated to support additional decades of wet storage include fuel inspection and corrosion testing to evaluate the effects of specific water impurity species on corrosion attack. An overview is given of the activities that have been successful at maintaining spent fuel integrity in wet storage at SRS. Additional topics, including the role of the surface oxide layer in stabilizing aluminum against corrosion during wet storage, and recent research program findings of exfoliation of aluminum during high-heat flux exposure to water that may occur in high power reactor operation, are discussed. Keywords: al-based fuels, spent nuclear fuel (SNF), pitting, galvanic and general corrosion, corrosion coupons, control of water pH and conductivity, cyclic polarization tests, high heat-flux corrosion
______ *To whom correspondence should be addressed: R. L. Sindelar, Savannah River National Laboratory, Aiken, SC 29808, USA; e-mail:
[email protected] 245 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 245–266. © 2007 Springer.
246
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
1 Introduction 1.1 BASIN STORAGE OF SPENT FUEL AT THE SAVANNAH RIVER SITE Spent nuclear fuels (SNFs) from nuclear materials production operations and fuel material provided by the US Department of Energy and used in foreign and domestic research and test reactors are being accepted and stored at the Savannah River site (SRS), now exclusively in the L-Basin. Production reactor fuels were previously stored in K-, L-, P-Reactor basins, and the research and test reactor fuels were stored in the receiving basin for offsite fuel (RBOF) facility. The L-Basin is a 3.5 million gallon volume structure constructed of reinforced concrete and coated with vinyl paint. The L-Basin fuel storage configuration consists of vertical and horizontal tube storage, and bucket storage. Oversize cans are used to store damaged and failed fuel and fuel pieces. Damaged or degraded cladding or structural features of a fuel assembly, if significant, can result in radiological, criticality safety, waste, and accountability issues.1 The recent L-Basin inventory includes approximately 10,000 materials test reactor equivalent fuel assemblies with aluminum cladding and 700 assemblies with stainless steel or zircaloy cladding. The storage experience at SRS, and experience from basin storage at many sites worldwide, shows that storage of aluminum-based fuel materials in water basins is most challenging because of the need to avoid conditions aggressive to corrosion of the aluminum. Corrosion control and surveillance activities were developed at SRS to minimize aluminum cladding corrosion degradation through water chemistry operational limits and storage configurations to avoid galvanic couple incompatibilities. 1.2 MECHANISMS OF ALUMINUM CORROSION IN WET STORAGE Identification of the mechanisms that cause corrosion provides a basis for the actions required to minimize corrosion in wet storage. The types of corrosion that can be expected for aluminum in wet storage are localized corrosion attack in the form of pitting or crevice corrosion, galvanic corrosion, and general corrosion. The aluminum cladding typically has an initial film of an aluminum oxyhydroxide due to reactor irradiation or pretreatment. Aluminum is thermodynamically active in water, leading to the formation of oxyhydroxide films with various structures and degrees of hydration.2–5 The formation of oxyhydroxide films is dependent on the temperature and pH of the water. Aluminum is amphoteric and its oxyhydroxides will dissolve at pH levels below ∼4 and above ∼10.
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
247
The chemical precursor of the crystalline oxyhydroxide film structures is a gelatinous pseudo-boehmite. The gelatinous film ages to form a tri-hydroxide with the structure of Gibbsite (hydrargillite) [γ-Al(OH)3] if the pH is lower than 5.8 or higher than 9, and Bayerite [α-Al(OH)3] if the pH is between 5.8 and 9 at temperatures below about 80°C. Boehmite, a monohydroxide [γ-AlOOH] forms at temperatures above ∼80°C. These passive films are stable (limited oxide dissolution) for waters with pH levels between about 4 and 10 and boehmite provides high resistance to continued corrosion. Autoclave treatment to form a passivating, adherent boehmite film on the surfaces of the clad fuel assemblies has been used prior to reactor irradiation at several sites. Boehmite films on the cladding over the high-temperature fuel regions during reactor operation can also form if no pretreatment is applied. General corrosion results in a consumption of aluminum cladding. At pH conditions in which the oxyhydroxide layer is stable, the corresponding growth of the hydrated oxide layer with time in general corrosion can be expressed as:6 Al consumption or oxyhydroxide film growth = A + B ln(time),
(1)
where the coefficients A and B are dependent on the temperature and water flow rate. Using distilled, deionized water, with a conductivity of 0.71 microSiemens per centimeter (µS/cm) and low flow conditions, Draley et al.6 determined the values of A and B for temperatures of 40°C, 70°C, and 90°C for the weight loss of aluminum alloy 1,100 in both helium-saturated and oxygen-saturated water. As an estimate using the helium-saturated coefficients at 70°C, and neglecting the dissolution of the hydrated oxide, consumption of only 2 µm of aluminum metal would be expected for a 10-year exposure. Since the initial cladding thicknesses on aluminum fuels are typically at least 250 µm thick, cladding thinning by general corrosion is not significant to impact any desired basin storage period up to many decades. The body of experimental work on pitting of aluminum with an oxyhydroxide film has been reviewed within a mechanistic framework for localized corrosion by Foley7 and Szklarska-Smialowska.8 Fundamental steps in the process of pitting attack are adsorption of reactive anions on the film, formation of soluble compounds, dissolution of the film, and direct attack of the exposed aluminum by the anion. There is substantial experimental evidence that adsorption of halides, particularly chloride, a common impurity, is the preliminary step to pitting. The aggressive anion (e.g., chloride ion) is adsorbed “competitively” with hydroxyl ions or water molecules that would promote passivity. A category of pitting corrosion, which leads to rapid attack, is nodular pitting in which an occluded cell is developed at the pit. Porter and Hadden9 reported that this form of pitting develops when calcium bicarbonate, chlorides, copper salts, and dissolved oxygen are present together in the water. Davies10
248
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
and Rowe and Walker11 have also reported that solutions containing soluble salts of copper, chloride, and bicarbonate have an accelerating effect on the corrosion of aluminum. The distribution of pits in aluminum exposed to aggressive water has been observed to follow an extreme value statistical distribution in which, after initiation and development of the pit distribution, only a small percentage (∼1%) continue to propagate.12 The growth rate (pitting rate) of these deepest pits has been reported by the authors to follow a power law:13 Maximum Pit Depth = A (time)1/3,
(2)
where A, in general, depends on the alloy and water quality. Site experience has shown that aggressive water quality conditions can cause pitting and penetration of the claddings especially if the protective boehmite film is ruptured;14–16 fuel stored in high-quality (high-purity) water has not shown significant corrosion even after many years of storage. Conductivity of the basin water has been observed to be a primary factor in the incidence of pitting in the storage of aluminum-clad fuels. Galvanic couples between the cladding and stainless steel equipment used in storage have enhanced this corrosion. Fuel returned to SRS from several fuel storage sites with aggressive chemistry and storage conditions have exhibited advanced corrosion attack. Figure 1 has still photographs from inspections of aluminum fuel assemblies from the RA-3 reactor in Argentina that had been stored under aggressive conditions at the Ezeiza Central Storage Facility. The photos show advanced crevice corrosion degradation between the side plates and the end fuel plate. Andes et al. provide additional details on the storage conditions that caused the corrosion attack on the RA-3 fuel.17 Rapid, significant degradation of aluminum has not been observed and would not be anticipated in high-quality (high-purity) water. Chemistry monitoring and controls, in conjunction with a surveillance program, have been used at the SRS to minimize corrosion degradation during fuel storage. The specific levels of impurities and their combination in water that create conditions “aggressive” to cause pitting in aluminum over extended exposure periods have not been rigorously determined. Rather, separate effects of aggressive species from literature information, and service experience, have formed the basis for water chemistry control. The International Atomic Energy Agency (IAEA) provides a guide describing water chemistry conditions to avoid corrosion damage to aluminum-clad fuel.18
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
249
Figure 1. RA-3 Assembly #243. Left: In Ezezia CSF (1999), with crevice corrosion in poor water. Right: In SRS (2001); corrosion nodules dislodged in transport or handling
2 Chemistry Monitoring and Control Program A Basin Water Chemistry Control Program was established in the early 1990s and upgraded in 1995. The program is designed to minimize personnel exposure from radioactivity in the basin water and corrosion degradation of SNF in the basin. The chemistry control program establishes and maintains chemistry limits to ensure that the basin water quality is controlled to minimize initiation and progression of corrosion. The chemistry control program activities include sampling and control of basin water quality parameters such as pH, temperature, and conductivity, and species such as chloride, mercury, iron, aluminum, microbes, and Cs-137 activity (see Table 1). TABLE 1. L-Basin water quality operating limits Water quality parameters Ph Conductivity Activity
Cu concentration Hg concentration Cl concentration Fe Al Temperature
Operating limit 5.5–8.5 10 µS/cm Cs-137: 500 dpm/ml Alpha: 3 dpm/ml Tritium: 0.4 Ci/ml (8.88 × 105 dpm/ml) 0.1 ppm 0.014 ppm 0.1 ppm 1.0 ppm 1.0 ppm 45°C
Monitoring frequency Weekly Weekly Weekly Monthly Biannual Biannual Biannual Biannual Biannual Biannual Weekly
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
250
The overall water quality in L-Basin has been maintained since the early 1990s within chemistry program limits. Previous to July 1995, a portable deionizer was rotated among C-, K-, L-, and P-Reactor basins to control basin chemistry. A temporary deionizer was installed in L-Basin in mid-1995 when the basin had an initial conductivity ∼100 µS/cm. By the end of October 1995, conductivity values decreased to 3 µS/cm and below. A dedicated deionizer system was placed in service in June 1996 to maintain this high water quality. Example trend data for conductivity and pH levels is shown in Figure 2. Basin activity levels have peaked during the recent storage period due to new additions of fuel and high activity water from materials added to L-Basin. Figure 3 shows that increases in Cs-137 activity levels did not affect pH and conductivity levels over the time period from late 2002 to early 2004 when the samples were received from RBOF and placed into L-Basin, and this suggests that corrosion processes are not strongly affected by increases in Cs-137 activity levels. Water chemistry shows that the L-Basin water is mostly being maintained within the operating limits since 1996. 9
12 11
8
pH
7
9 6
8 7
5
pH
Conductivity, uS.cm
10
6 4 5 3
4 3
2 2
Conductivity 1
1 0
02
0
2 r-0
er ar be ob m t nu c e a J O ec D
03 y-
3 3 3 03 04 03 03 04 04 t-0 r-0 r- 0 ilyyhher s r e e a c c n a b M gu ar ar Ap ob Ju nu M M ct em Au Ja O ov N
Figure 2. L-Basin conductivity and pH from January 2003 until March 2004
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
251
Figure 3. Cs-137 activity levels in L-Basin versus pH and conductivity during October 2002 to the end of February 2004
3 Corrosion Surveillance The corrosion surveillance program, originally designed to monitor the condition of fuels stored in K-Basin at SRS, was initiated in 1992 and expanded to include L- and P-Basins and the RBOF facility in 1993. The P-, K-, and RBOF-Basins were de-inventoried and their surveillance programs were discontinued in 1996, 2002, and 2003, respectively. The L-Basin Corrosion Surveillance Program is continuing to demonstrate the resistance of the fuel claddings to corrosion in the basin water and to provide assurance of continued safe, interim storage. The corrosion surveillance program monitors susceptibility to pitting, crevice, galvanic, and general corrosion. The corrosion surveillance specimens are of coupons and other test specimens in the basin to provide early detection of corrosion susceptibility in the basin water environment. The corrosion specimens include sets of standard corrosion coupons (70 and 32 mm diameter disks), that include materials representative of the aluminum cladding materials and storage rack materials used at SRS and in the US origin research and test reactors fuel. Standard sets of coupons were developed through a cooperative, Department of Energy complex-wide assessment of basin corrosion and surveillance programs initiated in 1995.16
252
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
Recent surveillance program results show that corrosion can occur even under good water quality conditions if specimen configurations are present that may facilitate the development of localized volumes of water that may be aggressive to metals, e.g., intimate contact of dissimilar metals (galvanic corrosion) or in crevices and pits. This corrosion of the surveillance coupons are expected, and are the result of the coupon design that promotes the evolution of aggressive localized chemistry conditions. 3.1 CORROSION SURVEILLANCE COUPONS Sets of standard corrosion coupons, including coupons representative of the aluminum cladding materials and storage rack materials used at SRS and in the research and test reactors were placed in the SRS basins. The surveillance coupons had a finish of 120 grits and were not pretreated to increase the thickness of the air-formed oxide before being placed in the basins. The use of an air-formed film provides a surface that is more susceptible to corrosion than the irradiated fuel cladding because high-temperature reactor operations enhanced the protective quality of the oxide film.15 The use of these nonirradiated coupons allows early detection of corrosion in advance of potential effects on fuel cladding and provides the opportunity to associate that corrosion with changes in the basin water chemistry. Aluminum alloy coupons of 1100, 5086, 6061, and 6063, listed in Table 2, were used to represent various fuel cladding and furniture rack alloys. TABLE 2. Composition of Al surveillance coupon alloys
Al Alloy designation 1100
Si
Maximum elemental composition (%) Cu Mn Mg
Zn
Other 1.0 Max Si + Fe 0.50 Fe/0.25 Cr 0.35 Cr 0.35 Fe/0.10 Cr 0.8 Fe
–
0.20
0.05
0.05
0.10
5086
0.40
0.5
4.0
0.4
0.25
6061 6063
0.8 0.6
0.40 0.10
0.15 0.10
1.2 0.9
0.25 0.10
*R4043 6.0 0.30 0.05 *Welding rod for furniture rack samples
0.05
0.10
A “Ray Gun” coupon assembly that has 70 coupons is shown in Figure 4. A “Jr. Ray Gun” coupon assembly has 36 coupons, and does not contain the 70 mm disk stainless steel coupons on the Ray Gun assembly, but does contain the 1:1 stainless steel to aluminum coupon galvanic couple specimens. A sketch
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
253
showing the configuration of the galvanic coupons on the Ray Gun is shown in Figure 5. Additional specimen types in the surveillance program include “furniture rack” specimens that are flat plates and U-bend specimens of 6061-T5 and 6063-T5 materials, respectively with R4043 weld beads.
Figure 4. “Ray Gun” assembly: 70 surveillance disk coupons on SS rod, as received from L-Basin
Figure 5. Sketch of Ray Gun galvanic coupon arrangement. The arrangement creates a galvanic cell between the Al and a larger SS coupon, with potential crevices between each coupon and between the coupon and each washer
3.2 EVALUATION OF CORROSION SPECIMENS Coupon assemblies have been withdrawn and the corrosion reported on an annual basis. The corrosion evaluation of the coupons consists of: • • • •
Photographs of both sides Weight gains of all coupons Cleaning of selected coupons to remove oxides for detailed pit analysis Pit characterization in accordance with ASTM G-46.19
The reported information includes coupon photographs including pitting at a high magnification. Comparisons with past coupon surveillance photographs may be reported. The weight gain data of each specimen is reported. The pitting
254
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
density, and average (of the 10 deepest pits) and maximum pit depths are reported for the disk face regions and crevice regions. Pit depth growth rates are evaluated. Comparison of the information with the chemistry data from the storage period is reported, and the trend of data from previous coupon surveillance evaluations is reported. A recent literature paper provides details of the 2003 corrosion surveillance results.20 Several of the results from the surveillance coupons from 2004 are summarized below. Selected images of individual surveillance coupons retrieved from the basin in 2004 after 8 years immersion are shown in Figure 6. Percentage weight gains of ∼0.85–1.05% for the 2004 aluminum samples were observed. These coupons appear visually comparable to coupons pulled in 2001, 2002, and 2003. The individual coupons from all 4 years of evaluation are undergoing uniform corrosion with a buildup of an oxyhydroxide layer.
Figure 6. Selected Surveillance Coupons removed from L-Basin in 2004. The aluminum specimens exhibited a uniform oxide scale and pitting attack near the crevice region created by the PTFE washer. The aluminum specimens with an autogenous weld bead exihibited pitting attack near the weld region
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
255
The most appreciable oxide buildup observed is in galvanically coupled specimens21 marked by heavy oxidation between the samples of dissimilar metals (see Figure 7). Similar behavior was noted in the previous examinations in 2001 and 2003, respectively. Pitting was apparent when the oxides were removed by the cleaning of these specimens in 16 M nitric acid solution.
Figure 7. Aluminum 1100, 6061, and 6063 (small) coupons galvanically coupled with the 304 stainless steel (large) coupons shown with mating surfaces in view
Pit depths were measured on all galvanically coupled aluminum alloy coupons and summarized in Figure 8 with a display of average and maximum pit depths. Within each two alloy groupings, the diameter of the specimen on the left is 32 mm and that of the specimen on the right is 70 mm. The size difference is intentional in order to accelerate pitting rates from galvanic corrosion in the smaller specimen. The highest average (6 mils) and maximum (8.5 mils) pit depths were revealed on the 6063 coupon, which was coupled with 304. The pit formation is attributed to a combination of galvanic corrosion and end grain effects. Similar results from the 6063/304 coupled coupons withdrawn in 2003 were observed. Maximum pit depths on the aluminum alloy couples were located near the PTFE washer circumference and on the ID and OD of the samples.
256
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
Figure 8. Average and maximum pit depths of 2004 galvanically coupled coupons. From left, the first coupon (group of two) is the smallest (32 mm OD), while the second mating coupon is the largest (70 mm OD). The first three bar sets are coupons mated to 304 SS while the remainder mate two Al alloys. Maximum pit depths were measured near the edge of the PTFE washer area or at the coupon ID. No pitting was observed on all 304 and on one of the 6063 coupons mated to 6061
Average and maximum growth rates for the galvanically coupled coupons are shown in Figure 9. In all but one coupon set, the average growth rate is less than 0.4 mils per year (m/y). The 6063 coupon mated to 304 show the highest average growth rate (∼0.7 m/y), which is about double that of other coupons. Figure 10 shows the pronounced pitting on the weld-filler material on the U-bend specimens removed in 2003 and 2004 and cleaned to remove the oxide deposits. In the autogenous-weld disk specimens, the base material suffered pitting attack in crevice regions whereas the weld deposit did not. The following are general conclusions from the surveillance program from specimens removed in 2001–2004 that had been exposed to high-quality water, within the limits of Table 1: •
Uniform surface corrosion was observed on the aluminum surveillance coupons. No significant difference in general corrosion performance was
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
•
257
observed among the aluminum alloy corrosion surveillance disk coupons (i.e., 1100, 6061, and 6063). It appears that oxide growth was initially high and has grown at a much slower rate or has stabilized. Pitting occurred in several regions, including the inner diameter and outer diameter circumferences (due to end grain attack), on and near autogenous and filler-metal welds, in crevices beneath the PTFE insulating washers, and in crevices between galvanically coupled stainless steel and aluminum, and between galvanically coupled aluminum coupons.
These results show that basin storage in low-impurity, high-quality water provides for only limited, slow corrosion of aluminum cladding and structural materials. It is emphasized that the corrosion surveillance coupons were deliberately designed to be susceptible to corrosion attack, and that the aluminum cladding materials are less susceptible to corrosion attack due to protective oxyhydroxides and limited galvanic couples in the basin storage configurations.
Figure 9. Average and maximum pit growth rates for galvanically coupled coupons. Coupon order and sizes are identical to those in Figure 8. The same pit depths were used with depth divided by time in basin to calculate growth rates
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
258
2003
Weld toe
2004
Figure 10. Cleaned R4043 weld beads on 2003 (upper) and 2004 (lower) 6063 T5 U-Channel samples after immersion in L-Basin for 85 and 97 months, respectively. Width of each weld bead is approximately 6.4 mm. Numerous pits are visible in the weld bead and weld toe on each side of bead. Few pits were found in the base metal
4 Basin Storage Life Extension A strategy to extend the storage life of fuels in the SRS basins is based on the management of corrosion degradation of aluminum-clad fuels through several program elements. These are: •
Development of water quality standards to minimize corrosion of fuel and fuel storage structures.
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
• • •
•
259
Validation of the water quality standards through detailed examination of surveillance specimens exposed to basin conditions. Monitoring of the basin conditions (water quality, radioactivity, physical condition), and the fuel condition in an inspection program to verify acceptable conditions are maintained. Specification of an acceptable fuel condition (i.e., maximum extent of loss in net section and maximum release rate of radionuclides) to remain in direct basin storage. Fuels exceeding the acceptance criteria would be place in isolation canisters. Development of water-quality-dependent corrosion models to allow prediction of the general condition of the fuel and basin storage structures as a function of time.
Several elements of this strategy have been previously implemented including ongoing corrosion specimen surveillances and monitoring of basin conditions. Additional tasks are in progress to provide the technical bases to enable additional decades of safe basin storage. These tasks are outlined below. 4.1 CHEMISTRY ENVELOPE DEFINITION FOR CORROSION MINIMIZATION Specifications and operating standards are in place to govern water temperature, basin level, chemistry, conductivity, and water radioactivity for the L-Basin. These standards are primarily based on empirical observations. As part of the activities for extended basin storage, an experimental program is being conducted to provide detailed technical bases for impurity levels for water quality. Water quality standards would be set within the region of water qualities which are non-aggressive to pitting or, at least, lead to a low pitting rate that would not adversely impact the fuel for a desired storage period. Cyclic polarization testing is being performed to identify the region of “aggressive” water qualities where corrosion attack would occur if an existing oxide layer were scratched, exposing the bare metal, or the aluminum components were without an oxide layer. The testing is based on cyclic polarization methods reported in references 22 and 23. Figures 11A–D show initial results from the testing using Al 1100 prepared to a 1200 grit finish. Figures 11A and 11B show that the 1 ppm chloride solution is not likely to cause pitting, whereas the 10 ppm and 100 ppm chloride solutions clearly showed pitting attack, as verified by post-test metallography. Figure 11D shows that nitrate at low levels may serve as an inhibitor to aggressive solutions containing chloride ions. Additional testing will support definition of a “chemistry envelope” for water qualities non-aggressive to pitting attack of aluminum.
260
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
Figure 11. Cyclic Polarization Result for (A) Low-Chloride non-aggressive (low current density) solution and (B) High-Chloride aggressive (high current density) solution.
Figure 11C. Cyclic Polarization Results for Chloride and Sulfate Solution (10 ppm Cl and 10 ppm SO4). The results show an aggressive solution (high current density).
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
261
Figure 11D. Cyclic Polarization Results for Chloride with Nitrate Solution (10 ppm Cl and 100 ppm NO 3). The results show a non-aggressive solution when nitrate is added to the chloride solution.
4.2 FUEL INSPECTION PROGRAM The L-Basin storage conditions are expected to provide for minimal degradation of the fuel. Nevertheless, periodic inspection of the wet-stored fuel provides a direct measure of the condition of the fuel with storage time. Direct visual examination of selected fuels in L-Basin over the past several years has shown that the fuel is typically in good condition, and is not degrading noticeably with time in L-Basin storage.24 The L-Basin has received many different fuel types with various cladding conditions, including through-clad damage that occurred prior to storage in the SRS basins. Aluminum-based fuel with through-clad damage with even moderate leakage is shown to be acceptable for continued direct basin storage.1,25 Other fuel types (e.g., oxide fuel) with damage are stored in isolation canisters with J-tube vents that are effective at eliminating radioactivity release into the basin water. An inspection program to characterize, trend, and disposition fuel condition is being planned. Specifications for acceptable fuel conditions to remain in direct basin storage will be included in the plan.
262
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
5 Corrosion Under High Heat Flux Conditions Accelerated corrosion on aluminum cladding has been observed at several research reactor sites during power upgrades. The corrosion mode appears to be of a different nature than typical boehmite formation, growth, and spallation. The effects of radiation and heat flux on the corrosion of aluminum are discussed.26,27 The rate of corrosion is expected to increase under radiation fields and under heat flux conditions. For materials experiencing large thermal gradients, the expected rate of general corrosion and oxyhydroxide film growth is given by: W = Ktn,
(3)
where W is the weight gain, t the time of exposure, and n is an empirical constant. The factor K is expressed as K = A exp[Q/R(Ti + 10φ)],
(4)
where Ti is the temperature of the metal/oxide interface and φ the heat flux, and A is an empirical constant. This suggests that the effective activation energy decreases with increasing heat flux. This decreased activation energy causes the anticipated corrosion rate for materials exposed under heat transfer conditions to be higher than the corrosion rate of similar materials exposed to isothermal conditions. Corrosion under high heat flux conditions is being investigated using an apparatus to provide a high heat flux across a specimen shown in Figure 12. Figure 13A and 13B shows the optical macro- and micrographs of a 1/8″ thick disk of aluminum 1100 with an initial 800 grit finish exposed to a back side temperature of 130°C in water at 96 ± 2°C for 6 weeks. Limited formation of boehmite was observed. Rather, the aluminum surface appeared to undergo exfoliation due to this exposure. It is speculated that hydrogen ingress may be causing local embrittlement of the aluminum surface. Additional investigation is in progress.
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
Figure 12. Heat flux apparatus
Figure 13A. Post-test aluminum 1100 specimen surface region
263
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
264
Figure 13B. Cross-section of specimen at 6 weeks exposure, 96 ± 2°C front temperature, 130°C back temperature. Note incipient exfoliation of the aluminum at the near-surface region
References 1.
2. 3.
4. 5.
6. 7.
Carlsen, B., Fillmore, D., McCormack, R.L., Sindelar, R.L., Spieker, T.J., Woolstenhulme, E., “Experience with Damaged Spent Nuclear Fuel at U.S. DOE Facilities”, ICONE14-89314 in Proceedings of 14th International Conference on Nuclear Engineering, July 17–20, 2006, Miami, Florida, USA, to be published. Alwitt, R.S., “The Aluminum-Water System”, Chapter 3 in Oxides and Oxide Films, Vol. 4, ed. J.W. Diggle. New York, Marcel Dekker, 1976, pp. 169–254 Troutner, V.H., “Observations on the Mechanisms and Kinetics of Aqueous Aluminum Corrosion. Part 1––Role of the Corrosion Product Film in the Uniform Aqueous Corrosion of Aluminum”, Corrosion, 15(1959): 9–12. Wefers, K., Misra, C., “Oxides and Hydroxides of Aluminum”, AlCOA Laboratories Technical Paper No. 19, Revised, 1987. Thomas, J.K., Ondrejcin, R.S., “An Evaluation of the Thickness and Emittance of Aluminum Oxide Films Formed in Low-Temperature Water”, Journal of Nuclear Materials, 199(1993): 192–213. Draley, J.E., Mori, S., Loess, R.E., Journal of Electrochemical Society, 114(1967): 353. Foley, R.T., “Localized Corrosion of Aluminum Alloys––A Review”, CORROSION––NACE, 42, 5(1986): 277–288.
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
8. 9. 10. 11. 12.
13. 14. 15.
16. 17.
18. 19. 20.
21. 22.
23.
24.
265
Szklarska-Smialowska, Z., “Pitting Corrosion of Aluminum”, Corrosion Science, 41(1999): 1743–1767. Porter, F.C., Hadden, S.E., “Corrosion of Aluminum Alloys in Supply Water”, Journal of Applied Chemistry, 3(1953): 385–409. Davies, D.E., “Pitting of Aluminum in Synthetic Waters”, Journal of Applied Chemistry, 9(1959): 651–660. Rowe, L.C., Walker, M.S., “Effect of Mineral Impurities in Water on the Corrosion of Aluminum and Steel”, CORROSION––NACE, 17(July 1961): 353t–356t. Aziz, P.M., “Application of the Statistical Theory of Extreme Values to the Analysis of Maximum Pit Depth Data for Aluminum”, CORROSION––NACE, 12 (October 1956). Aziz, P.M., Godard, H.P., “Pitting Corrosion Characteristics of Aluminum”, Journal of Industrial and Engineering Chemistry, 44, 8(1952). Howell, J.P., “Durability of Aluminum-Clad Spent Nuclear Fuels in Wet Basin Storage”, Corrosion/96, Paper no. 128, Houston, TX, NACE International, 1996. Louthan, M.R., Jr., Iyer, N.C., Sindelar, R.L., Peacock, H.B., Jr., “Corrosion of Aluminum-Clad Fuel and Target Elements: The Importance of Oxide Films and Irradiation History”, Proceedings of the Embedded Topical Meeting on DOE Spent Nuclear Fuel and Fissile Material Management, American Nuclear Society: La Grange Park, Illinois, 1996, pp. 57–61. Howell, J.P., “Corrosion Surveillance in Spent Fuel Storage Pools”, Corrosion/97, Paper No. 107, Houston, TX, NACE International, 1997. Andes, T.C., Large, W.S., Castle, R.B., Louthan, M.R., Valdes, V.S., Sindelar, R.L., “Characterization of Corrosion Damage on Aluminum Fuel Assemblies in Basin Storage”, Proceedings of the 5th Topical Meeting on DOE Spent Nuclear Fuel and Fissile Material Management, American Nuclear Society, La Grange Park, Illinois, 2002, ISBN:0-89448-668-3, ANS Order No. 700294. IAEA, “Corrosion of Research Reactor Aluminum Clad Spent Fuel in Water”, Technical Reports Series 418, 2003. ASTM G-46, “Standard Guide for Examination and Evaluation of Pitting Corrosion”, American Society for Testing and Materials, 1999. Vormelker, P.R., Duncan, A.J., Mercado, D.C., “Corrosion Evaluation of Aluminum Alloys in a Spent Fuel Basin”, Corrosion/2005 Paper No. 5595, Houston, TX, NACE International, 2005. ASM Handbook Ninth Edition, Volume 13 Corrosion, “Evaluation of Galvanic Corrosion”, ed. J.E. Davis, ASM International, September 1987, p. 235. Chandler, G.T., Sindelar, R.L., Lam, P.S., “Evaluation of Water Chemistry on the Pitting Susceptibility of Aluminum”, Corrosion/97, Paper No. 104, Houston, TX, NACE International, 1997. ASTM G-5, “Standard Reference Test Method for Making Potentiostatic and Potentiodynamic Anodic Polarization Measurements”, American Society for Testing and Materials, 1994. Vormelker, P.R., Vinson, D.W., “FY04 Inspection Results for Wet Uruguay Fuel in L-Basin (U)”, WSRC-TR-2005-00216, Westinghouse Savannah River Co., Aiken, SC, 29808, September 2005.
266
WET STORAGE OF SPENT FUEL AT SAVANNAH RIVER SITE
25. Vinson, D.W., Deible, R.W., Sindelar, R.L., “Impact of Degraded Al-SNF on Shipping and Basin Storage”, Proceedings of the 5th Topical Meeting on DOE Spent Nuclear Fuel and Fissile Material Management, American Nuclear Society: La Grange Park, Illinois, 2002, ISBN:0-89448-668-3, ANS Order No. 700294. 26. Sindelar, R.L., Lam, P.-S., Louthan, M.R., Iyer, N.C., “Corrosion of Metals and Alloys in High Radiation Fields”, Materials Characterization, 43(1999): 147–157. 27. Griess, J.C. et al., “Effect of Heat Flux on the Corrosion of Aluminum by Water, Part IV. Tests Relative to the Advanced Test Reactor and Correlation with Previous Results”, Oak Ridge National Laboratory, ORNL-3541, February 1964.
CORROSION OF ALUMINIUM ALLOY SAV-1 AND AUSTENITIC STAINLESS STEELS 12Cr18Ni10Ti AND 08Cr16Ni11Mo3—CORE STRUCTURAL MATERIALS FOR WWR-K AND BN-350 REACTORS O. P. MAKSIMKIN∗ Institute of Nuclear Physics, Almaty Abstract: Pitting corrosion was studied in aluminum alloy SAV-1 and two austenitic steels subjected to various neutron doses and heat treatments. Based on mechanical testing data and optical and electron microscopy it was shown that irradiation changes both the microstructures and properties of these alloys. SAV-1 is used in the WWR-K reactor as material for the automatic control rods. Irradiation was found to change the aging kinetics of this aluminum alloy, deteriorating its corrosion resistance. Data on the influence of temperature on its structure and corrosion resistance are presented. The relationship between SAV-1 alloy structural changes and its corrosion resistance after irradiation up to various doses has been established. Microstructure and composition in corroded surface and subsurface layers have been studied for the steels 12Cr18Ni10Тi (analogue of AISI 321) and 08Cr16Ni11Мo3 (analogue of AISI 316)––the duct materials of spent BN-350 fuel assemblies. Two techniques for investigating corrosion in irradiated stainless steels were developed and used: study of surface element composition and accelerated corrosion tests. On the basis of EM studies, as well as data from local analysis of the material chemical composition near pitting areas, the crucial role of fine structure in pitting corrosion was determined. It has been shown that the processes of corrosion and aging in neutron-irradiated austenitic stainless steels depends considerably on such parameters as the latent energy, the concentration of transmutation helium, and the formation of the martensitic α′-phase due to deformation and irradiation. Keywords: BN-350 reactor, WWR-K reactor, austenitic stainless steels, SAV-1 aluminum alloy, pitting corrosion in water storage, anode polarization curve
______
∗ To whom correspondence should be addressed: O. P. Maksimkin, Institute of Nuclear Physics, National Nuclear of Republic of Kazakhstan, Ibragimov Str.1, 050032, Almaty, Kazakhstan; e-mail:
[email protected]
267 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 267–279. © 2007 Springer.
268
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
1 Introduction According to the International Atomic Energy Agency (IAEA), an alarming situation connected with the problems of safe storage of spent nuclear fuel exists in the world at large. In states having research or industrial nuclear reactors but no plants for nuclear fuel processing, there are a number of fuel assemblies collected in storage pools. The monitoring of their condition shows that despite thorough abidance with parameters for safe storage (such as correct water chemistry and temperature), after a long period of time (∼30 years) signs of corrosion damage are observed in the spent fuel. For the purpose of preventing fuel assembly destruction under those conditions and environmental contamination with radioactive decay products, in several cases it was recommended to remove spent fuel assemblies from storage pools and prepare them for “dry” storage. For this the complex checking of the condition of the assembly structural materials (such as assembly ducts) are necessary. The task of checking is both to investigate changes in mechanical properties and structure of metallic materials after irradiation and to evaluate the scale of corrosion damage and ability to withstand long-time storage under water. The above factors concern directly the situation in Kazakhstan, where nowadays three research reactors are in operation and plans are developed for decommissioning of the first industrial sodium-cooled fast reactor, the BN-350 reactor, which was successfully exploited for more than 25 years. One of the main questions connected with safe storage of spent fuel is estimation of corrosion level in materials stored for a long time in water, such as stainless steels—duct material for BN-350 assemblies, and an aluminum alloy (SAV-1) used in assemblies for the WWR-K light-water research reactor. The present paper describes investigations on the corrosion degradation of these materials. 2 Sample Preparation The materials investigated were industrial steels of austenitic (12Cr18Ni10Ti, analog AISI 321) and ferritic-martensitic (12Cr13Mo2, analog HT-9) types, as structural material of ducts of spent fuel assemblies from the BN-350 reactor. The irradiation temperature range was 280–410°C, and maximum damage dose was 80 dpa. Plat samples of stainless steel with sizes 50 × 10 × 2 mm were cut in the hot cells at the BN-350 reactor from standard fuel assembly ducts at the level of the core center and at other levels (Figure 1) These plates were further cut in hot cells of the reactor complex of the INP NNC RK using specially developed remote devices to prepare samples of smaller size, which were investigated by optical metallography (Figure 2), scanning electron microscopy, and microhardness measurements.
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
Figure 1 (a). BN-350 sample cutting layout for assembly ducts (the studied surfaces are shown by arrows)
269
Figure 1(b). INP sample cutting layout for miniature duct samples (dimensions in mm)
Figure 2. Difference in thickness of corrosion layers on external (right) and internal surface (left) of the wall of a BN-350 duct of 12Cr13Mo2
In order to produce samples from the alloy of aluminum with magnesium and silicon (SAV-1) the casing of a standard rod for automatic control (AC-rod) of the WWR-K reactor was used (i.e., a tube with 27 mm diameter, 850 mm length, and wall thickness of 1 mm). The composition of materials investigated is given in Table 1. Samples of the SAV-1 alloy were cut from the bottom (in the reactor core and so irradiated to a high neutron dose) and the top (unirradiated control) of the tubular casing of the AC-rod. The AC-rod was in the WWR-K reactor core for ∼30 years, 20 years of which consisted of cyclic reactor operation to a maximum power of 10 MW. During reactor operation the rod casing was bathed with a reactor cooling water
270
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
TABLE 1. Elemental compositions of investigated alloys Alloy Fe
Element composition (wt%) Al C Cr Ni Mo
12Cr18Ni10Ti (AISI 321)
Base
–
0.12 17.0 10.66
12Cr13Mo2 (HT-9)
−
–
0.12 12.3
Al-Mg-Si (SAV-1) 0.2 Base – Alloy Ti 12Cr18Ni10Ti (AISI 321)
–
–
–
0.5 0.34 1.6 0.032 0.013
–
–
12Cr13Mo2 (HT-9)
–
0.25 0.34 0.021 0.07 0.19
Al-Mg-Si (SAV-1)
–
1.36
–
–
0.17 1.53 0.45
Element composition (wt%) Si Mn P S V
–
Nb
–
–
– Mg – – 0.87
with the following parameters, pH of 5.5 and specific electrical conductivity of 1.5–2.5 µS/cm. Ion concentrations in the water varied as follows: Fe3+, 0.1–0.8; Al3+, 0.01–0.1; Cl1–, 0.01–0.1 (mg/l). The temperature of the water was 35–50°C during operation and at room temperature during shutdown periods. The fast fluence value at discharge was 1.3 × 1022 n/cm2 (E > 0.1 MeV). After removal the rod was placed for 2 years in the storage pool. 3 Experimental Results 3.1 STEEL CORROSION FROM IRRADIATION AND WATER STORAGE The direct measurement of the depth of the corroded layer on spent fuel elements and assembly ducts from the BN-350 reactor was not carried out until 1999. But based on the results of visual inspections during the course of stripping of these assemblies, the suggestion was made that corrosion was not extensive and would not impact mechanical stability of the structures.1 Examination of samples from 12Cr13Mo2 assembly ducts revealed that the internal surface of every face had a dark-gray color and was dense (probably it was Fe3O4), whereas the color of an external cladding surface was yellow-brown; the corrosion layer was not dense and crumbled in some places (probably it was Fe2O3). The depths of the corrosion layer on the inner and outer surfaces of assembly ducts were not the same. For the inner surface, which was intensively bathed with sodium in-reactor, the depth of the corrosion layer (2–3 µm) was substantially lower than on the outer surface, where it was 50 µm (see Figure 2).
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
271
Measurements of the duct wall showed that after irradiation and subsequent storage in water it was thinned by no more than ∼2%. Metallography of a “thick” layer (Fe2O3) indicated that it includes as observable fine particles of material oxides as separate coarse fractions of the material itself (with size of 50–70 µm). Detailed investigation of the material within the boundary of the corrosion layer carried out after easy chemical treatment of a section with 1%-spirit solution of nitric acid revealed the non-compact places observed in between separate ferritic grains that represented “loose” Sorbite. In other words the most corrosion active component in the surface layer of the 12Cr13Mo2 steel was the Sorbite phase, whose damage, however, was not deeper than 20 µm. Figure 3a shows that corrosion proceeding around a ferritic grain led to its crumbling.
Figure 3. Cross-section of a surface layer for samples from the 12Cr13Mo2 steel (a) and 12Cr18Ni10Ti steel (b). (The arrow indicates a separate ferritic grain) ×900
Concerning the structure of the “thin” corrosion layers, they consist mainly of a mixture of oxide particles (Fe3O4). Metallographic examination of corrosion in the 12Cr18Ni10Ti austenitic steel indicated that the thickness of the corroded layer was two times lower than in the ferritic-martensitic steel and the corrosion is basically intercrystalline in nature (Figure 3b). Microhardness through the duct thickness of 12Cr13Mo2 steel was a constant 360 kg/mm2 ± 3%, compared to 450 kg/mm2 for 12Cr18Ni10Ti steel. A subject of study was a specimen of 12Cr18Ni10Тi steel, taken from the duct of the BN-350 fuel assembly H-214(I) at the +375 mm location. Using an Amrey 1200B scanning electron microscope and a Quantum Systems 4000 X-ray microanalyzer, the metallographic specimen surfaces were examined and the elemental composition of the steel determined. The elements Cr, Ni, Ti, Fe, responsible for the steel physico-mechanical properties, have been analyzed. The specimen dimensions were 3 × 6 × 2 mm; the contact dose rate was 450 µR/s. Visually, the plane surfaces of the specimen were different: the duct outer side was light with the corrosion layer removed whereas the duct inner surface was brown. The averaged results of analysis are given in Table 2. The table
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
272
demonstrates that the contents of the main doping elements associated with the inner and outer surfaces are quite different. For example, the content of nickel on the dark surface is lower, on an average, by 1.9 wt% than on the outer one, whereas the chromium content, on the contrary, is higher on the dark surface by 0.5%. The titanium content also is higher on the shroud inner surface. Microstructure and element contents in the corroded layer on the outer and inner surfaces of the duct have been studied the same way at the +75 mm and +375 mm locations. Results are given in Table 2 and Figure 4. TABLE 2. Surface compositions of an assembly duct (N-214(I), 12Cr18Ni10Ti steel) Location (mm) −375 +75 +375
Cr 17.1 17.7–17.9 18–21 18–20
Element (wt%) Ni Fe 12.4 69.8 10.4 71.2 9.2 69.2 7-8 71.4
Duct surface Ti 0.5 0.7–0.8 1.1–1.3 0.9–1.0
Outer (free of oxide layer) Inner (brown) Inner (brown) Inner (brown)
Figure 4. (a) Optical image of the oxide film on 12Cr18Ni10Тi steel at the +75 mm location; (b) SEM image of the same (×800)
3.2 MEASUREMENT OF CORROSION STABILITY 3.2.1 Stainless steels A great deal of work is devoted to investigation of the corrosion stability of stainless steels. The overwhelming majority of investigators share the opinion that the reason for the high corrosion stability of stainless steels is its inclination for anode passivation at the expense of formation of chromium oxide film on a metal surface. Stability of any stainless steel against corrosion depends on its
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
273
structural state, reaching a maximum for a single-phase and homogeneous structure. The presence of a secondary phase, the peculiarities of its distribution within the matrix, as well as the presence of defects, may significantly affect corrosion stability. As the material energy state and structure can vary under irradiation, corrosion stability is not constant. In scientific publications data may be found to support both radiation-induced increase and decrease in the corrosion stability of stainless steels; a decrease is typical of aqueous halogen containing solutions. Due to ambiguity of such conclusions, some items of the corrosion stability investigation of constructional material applied in nuclear reactors and suffered from in-pile irradiation are quite urgent. This work is aimed to investigate the effect of neutron irradiation on the corrosion stability of industrial stainless steels in chlorine-containing media at room temperature. The subjects of enquiry were specimens of Cr18Ni9Ti and 12Cr18Ni10Ti stainless steels in two states: unirradiated and after irradiation in the WWR-K reactor core. Irradiated specimens of 12Cr18Ni9Ti steel were cut from the automate control rod (AR) operated for a long time in the WWR-C reactor core and acquired the fluence comprising 1.3⋅1022 n/cm2. The pitting corrosion rate has been assessed by the specimen weigh loss per a surface unit versus time of staying in aggressive solutions 0.5N FeCl3 and 1N FeCl3 at room temperature and higher. Weighing was done with the CERN-770 grade electron analytical balance with a measurement accuracy of 0.0001 g. Also the corrosion behavior of steels was studied by the electrochemical technique of evaluating the stationary potential with derivation of the anode polarization curve at a room temperature in a 0.1N FeCl3 solution. For the experiments the laboratory installation consisted of a potentiostat, a cathode voltmeter, a milliammeter (the F30 grade), an electrolytic cell, an auxiliary platinum electrode, and a reference calomel electrode. Before each experiment, the specimen working surface area was determined, and the non-working section was covered with waterproof insulating material. The potential was measured discretely in 0.1-V steps at the rate of 2-V/hour. In Figure 5 the weight loss curves are presented for the original and the irradiated specimens of 12Cr18Ni10Ti steel after immersion for different periods of time in a 1N solution of FeCl3 at room temperature. The corrosion rate was found to be higher for the unirradiated material. The weight loss of the irradiated specimen after staying immersion for 212 h was less than the loss of the unirradiated specimen by a factor of 2.4. The pattern of anode polarization curves obtained for both unirradiated and irradiated stainless steels is typical of materials that tend to passivate (Figure 6). Irradiation has shifted toward the positive side the breakdown potential and the polarization curve section related to the re-passivation region, without changing
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
Mass loss, g/m2
274
Time, hr
Potential, V
Figure 5. Weight loss in 12Cr18Ni10Тi samples versus immersion time in a 1N FeCl3 solution (1: prior to irradiation; 2: after neutron irradiation)
Current density, mA/cm2 (10-2)
Figure 6. Anode polarization curves for (a) unirradiated and (b) irradiated 12Cr18Ni10Тi stainless steel
the anode process kinetics. As a whole, this phenomenon can be characterized as a delay in the anode dilution of metal material. The re-passivation potential level for 12Cr18Ni10Ti and Cr18Ni9Ni steels is the function of the steel
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
275
state. In the unirradiated specimens of 12Cr18Ni10Ni and Cr18Ni9Ti steels the re-passivation potential values, characterizing the material tendency to point corrosion, were 1.0 and 0.9 V, respectively. After irradiation the re-passivation potential values increased to 1.9 and 1.8 V for 12Cr18Ni10Ti and Cr18Ni9Ti steels, respectively. Regular examinations of the stainless steel specimen exposed to aggressive solutions made it possible to establish peculiarities in the occurrence of pitting areas. In Figure 7 photos of specimens subject to staying in the 1N FeCl3 solution for different periods of time are given. The picture demonstrates that tiny pitting particulates are formed at first on the metal surface; they grow in size under the long-term effect of the aggressive medium, then merge and form deep pits, converting into through pores. The metallographic studies have shown that near to pitting areas the surface has contrast coloration of light and dark relief. Determination of the elemental content in irradiated 12Cr18Ni10Ti steel within these areas by means of X-ray analysis made it possible to establish its origin. It was found that the number of basic elements in the matrix had not changed and corresponded to its typical contents in the steels under studies (Ti: 0.79%, Cr: 17.6%, Ni: 9.9%), whereas near the pitting formations observed in the optical microscope as dark areas the titanium content has increased up to 3.12%. In the light-color areas the titanium and chromium contents were 1.26 and 20.4%, respectively. In all light and dark areas the presence of chlorine was detected. The data obtained show that the pitting corrosion in stainless steel in chlorine-containing media is accompanied by selective dissolution of some elements, the centers for which are located near carbide inclusions on a base of titanium and chromium.
Figure 7. Surface of the unirradiated sample of 12Cr18Ni10Тi steel (a) before and (b) after trials for pitting corrosion; pitting formation on a surface of irradiated steel after (c): 1 h; (d): 7 h; (e): 11 h
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
276
3.2.2 SAV-1 alloy control rod Visual examination of the rod assembly showed that the original light-green color of the tube surface was only over the upper end (low-dose region) of the tube. At the bottom end (high-dose region) the tube appeared yellow-brown. Most of the surface was dull. Significant destruction of the protective coating was observed over the bottom end of the tube, but much less over the upper end (Figure 8). At the upper end the coating was smooth and adhered well to the matrix, whereas at the bottom end it was more porous, cracked and came off easily from the matrix. In fact, this degraded porous coating consisted of separate grains with sizes coinciding with those of the matrix (20–100 µm). Surface areas with lost coating ranged in size from several microns to hundreds of microns. The alloy surface of alloy was observed to have corroded grain boundaries in these areas. Metallographic examination of cross-sections also revealed grain boundary corrosion of the alloy under the protective coating. The elemental composition of the SAV-1 grain surface and its protective coating was determined by X−ray microanalysis in the scanning electron microscope; data are given in Table 3. The elemental composition of the coating is seen to differ substantially from the alloy composition, which is likely related to the way the coating formed and impurity accumulation in the porous structure of the protective layer during operation and storage. Treatment of a polished section of the specimen, irradiated with high dose of neutrons, in the solution 50%HNO3 + 47%HCl +3%HF revealed features with regular rectangular shapes, identified in the study as mainly etching holes (Figure 9a) and, partially, precipitation of intermetallic phases. The enhanced content of silicon was found in some of these holes, several times exceeding the mean content in the alloy matrix. One can surmise the holes contain a siliconrich secondary phase (probably Mg2Si), because formation of Mg2Si hardening phase particles has been observed by others2 in SAV-1 alloy after irradiation. An increase in the iron content in the etching holes was also noted, a feature warranting further investigation. TABLE 3. Composition of the anodized coating and the SAV–1 alloy (bottom of AC rod) Mg Coating – SAV-1 0.78
Al 93.43 97.8
Si 0.14 1.3
P 0.57 –
S 0.90 –
Ca 0.28 –
Ti 0.66 0.01
Cr 0.80 –
Fe Cu Zn 0.90 1.84 0.47 0.05 – –
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
Figure 8. Corrosion damage of aluminum alloy (AC rod tube, bottom end): (a) a piece of the AC rod tube; (b) protective coating, SEM image; (c) lost protective coating (residual coating is in upper left corner), SEM image; (d) residual coating at higher magnification, optical image; (e) grain boundary corrosion under the coating (cross section of tube), optical image
277
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
278
Figure 9. Etched sections of SAV-1 alloy (a) irradiated and (b) unirradiated
Apparently, the regular geometric etch pits are formed only in high purity (99.99%) aluminum.3 Thus, it was proposed that irradiation and aging caused “purification” (depletion) of the aluminum matrix in SAV-1 from doping elements, as the result of solid solution decomposition and precipitation of secondary phases. Such depletion of the matrix has been reported elsewhere.4 To characterize the degradation in mechanical properties of the protective coating, comparative measurements of microhardness (Нµ) of the coating and alloy at bottom and upper ends of the control rod shroud were carried out. It was found that the initial hardness of the anodized coating is characterized by a value of 510 kg/mm2, whereas the value for the unirradiated alloy was only 56 kg/mm2 under a 50 g load. For the upper end of the shroud tube (low dose) these values were 460 and 83 kg/mm2, respectively, and for the bottom end of the tube (high dose) – 190 kg/mm2 and 122 kg/mm2. Supplementary data were obtained in sclerometric tests. These experiments were performed by scratching a specimen with the diamond pyramid of a PMT3 instrument under various loads. The scratch width on the low-irradiated surface was 2 µm, for a 100 g load, whereas for similar conditions at the highirradiated end of the tube it was wider by a factor of 100. In the first case the scratch was neat, and triangular in profile, but in the second case the indenter simply removed a surface layer. This indicated that the strength/anticorrosion properties of the coating were getting drastically worse as a result of long-term irradiation in the water environment. 4 Summary The microstructure and composition of the corroded surface layer and subsurface layers have been studied in stainless steels 12Cr18Ni10Тi (analogue of AISI 321) and 12Cr13Мo2 (analogue of HT9), the structural materials of spent fuel assemblies of the BN-350 reactor.
CORROSION OF SAV-1 ALLOY AND AUSTENITIC STEELS
279
On the basis of electron microscopy studies, as well as data from local analysis of the material composition near pitting areas, the processes of corrosion and aging in neutron-irradiated austenitic stainless steels were found to depend on such parameters as latent energy, the concentration of helium caused by transmutation, and the amount of martensitic α′ phase caused by deformation or by irradiation. A protective anodized coating substantially prevents SAV-1 alloy from corrosion in water, although its condition strongly depends on neutron dose. Partial destruction of the protective layer and substantial loss of protection are observed at neutron doses of ∼1 × 1022 n/cm2 (E > 0.1 MeV). Evidence of grainboundary corrosion was found on the alloy surface that lost its coating. This result is important because fuel-element cladding for the WWR-K reactor is made of the SAV-1 alloy, but without the protective coating. The neutron dose for fuel elements is substantially less than for examined AC-rod tube, but the time of water storage is comparable. Besides, low coolant temperature (40– 50°C) favors formation of Bayerite on the aluminum surface, which has low protective properties. Further clarification of corrosion in irradiated SAV-1 alloy will allow us to predict the behavior of the main structural material of the reactor, and to develop appropriate anti-corrosion measures. Acknowledgments The author wishes to thank L.G. Turubarova, N.S. Sylnyagina, T.A. Doronina, and A.V. Yarovchuk for their great help in implementing this work. References 1.
2.
3.
4.
Karaulov, V.N., Blynsky, A.P., Yakovlev, N.L., Golovin, S.V., Lambert, S.V., “Investigation Changes of Short-Time Mechanical Properties for Materials of the BN-350 Reactor Spent Jackets Depending on Vacancy Swelling”, Proceedings of 2nd International Conference on Nuclear and Radiation Physics, Almaty, October 8–11, 1997, pp. 44–54. Karasev, V.S., Zaritsky, N.S. et al., “Structure and properties of the SAV-1 alloy after pong-time reactor irradiation”, Radiation Material Science (The Proceedings of the International Conference, Alusha, May 22–25, 1990, Charkov, Ukraine, 1991, pp. 112–117. Votinov, S.N., Sharov, B.V. et al., “Acceleration of Structural Changes in Aluminum Alloy Under Reactor Irradiation”, preprint NIIAR P-88, Melekess, 1970, 15 p. Krast, H.B., Liveensh, M.E. et al., “Investigation of aluminum casing of spent fuel elements of the IRT-22 reactor”, Atomic Energy, 1969, 27(4), pp. 286–289.
CORROSION OF FAST-REACTOR CLADDINGS BY PHYSICAL AND CHEMICAL INTERACTION WITH FUEL AND FISSION PRODUCTS
V. A. TZYKANOV, V. N. GOLOVANOV, V. K. SHAMARDIN, F. N. KRYUKOV,∗ AND A. V. POVSTYANKO Research Institute of Atomic Reactors, 433510, Dimitrovgrad, Russia Abstract: Fuel-cladding chemical interaction in fast breeder reactor (FBR) fuel pins can cause both matrix and intergranular corrosion of the inner surface of the cladding. Matrix corrosion is uniform nonselective interaction with fuel and fission products, causing the cladding to thin. Intergranular corrosion occurs on grain boundaries, weakening both them and the grains. Interaction with fission products may be the cause of microcrack formation at the cladding inner surface and subsequent propagation through the cladding under the tensile stress from gas and fuel pressure, and by mechanical interaction of fuel pins with assembly ducts. A way of considering corrosion in fast-reactor cladding is to divide data into two groups. The first group contains information from study of the physical and chemical interaction of separate fission products with cladding in laboratory experiments. Because experiments in a reactor take significant time, and are complicated, out-of-pile experiments may be used to model certain stages of the corrosion process. Such an approach allows systematic study of the effect of various factors on corrosion of cladding materials, and promotes the discovery of the fission-product behavior causing corrosion. The second data group obtained from FBR fuel pins is of greatest practical interest. The large statistical base obtained from examining fuel pins, combined with purposeful modeling experiments, has allowed links to be made between interaction of the basic fission products with cladding materials and the character of corrosion, and to define the influence of doping with rare earth elements and cladding pretreatment on corrosion stability. In the real environment of FBR fuel pins, the experimental dependencies of depth of corrosion on operational parameters, and on cladding mechanical properties are obtained. Features of the distribution of fission products and the corrosion environment are described in terms of the mode of operation of fuel pins.
______ ∗
To whom correspondence should be addressed: F. N. Kryukov, Research Institute of Atomic Reactors, 433510, Dimitrovgrad, Russia; e-mail:
[email protected] 281 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 281–293. © 2007 Springer.
282
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
Keywords: BOR-60 fast reactor, fuel-pin claddings, austenitic and ferritic-martensitic steels, oxide fuel and fission products, matrix and intergranular corrosion of steels.
1 Introduction UO2 and (U, Pu)O2 oxides are compatible with stainless steel at least to 1,000°С, so there has been no concern about corrosion of cladding caused by physical and chemical interaction with fuel in the initial stage of operation of fast neutron reactors. The issue of compatibility of oxide fuel with cladding arose in the early 1970s during studies of irradiated fuel pins when corrosive fission product materials accumulated in the fuel column attracted attention. Corrosion decreases short- and long-term strength and plasticity of cladding material. Corrosion along grain boundaries may cause microcracks and their further propagation into cladding due to pressure of gas and fuel, as well as mechanical interaction of fuel pins with assembly ducts. Loss of plasticity of the cladding material due to neutron irradiation leads to development of microcracks at the points of physical and chemical interaction of the cladding with fission products and becomes one of the major reasons for fuel pin depressurization during operation. It can limit the life-time of a fuel pin, which mainly depends on impermeability of cladding; it hinders reaching design burn-up and, eventually, affects the economic characteristics of fast-neutron reactors. Therefore the issues of physical and chemical interaction of fuel pin with cladding were studied in those countries where fast-neutron reactors were developed and operated. Researchers focused their attention on obtaining comprehensive data on physical and chemical interaction of fuel and fission products with cladding, on thermodynamic computations of potential chemical reactions inside fuel pins, on searching for analytical dependency of the corrosion depth on various technical and operational parameters, on methods to account for corrosion damage of cladding in the design of fuel pins, and on measures to limit or prevent corrosion of cladding. The other aspect of fission product accumulation is linked with storage and reprocessing of irradiated nuclear fuel. To justify safe storage and transportation of fuel and to develop technologies for reprocessing and to optimize reprocessing conditions, it is necessary to have, in addition to computed data on the quantities and composition of accumulated fission products, information about their distribution, local concentrations, and chemical status, and the changes in compositions of fuel and cladding caused by them.
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
283
2 Characteristics of Fuel Pins and Operational Conditions in BOR-60 Before 1989 a standard BOR-60 fuel pin had a 400 mm fuel column made of 90%-dense sintered UO2 pellets in a sealed cladding of 0Cr16Ni15Mo3Nb austenite stainless steel. The outer diameter was 6 mm, and the cladding wall thickness was 0.3 mm. On both ends of the column there were 100 mm long insulator regions in the form of sintered pellets of depleted UO2. From the mid1970s to the end of the 1980s a broad scope project on radiation testing of potential structural materials was implemented in BOR-60. Under this project a number of assemblies with lead test fuel pins were irradiated; a specific feature of fuel pins was that claddings were either made of prospective materials or/and had undergone different preprocessing treatments. Major irradiation parameters for fuel pins are shown in Table 1. In the table the maximum temperatures given for the inner surface of fuel pin cladding are for the central channel of an assembly without hot-spot factors. TABLE 1. Irradiation conditions of test fuel pins1 Assembly type Standard BOR-60 Lead test BOR-60
Max. linear power (kW/m)
Max. fuel burnup (%hа)
Max. cladding temperature (°С)
Fluence, 1022 n/сm2 (Е > 0.1)
39.1–49.2
10.8–15.3
650–700
8.4–12.2
36.2–51.2
8.0–9.7
610–715
6.3–15.6
Lead test fuel pins for BOR-60 had claddings made of chromium-nickel 0Cr16Ni15Mo3Nb and 0Cr16Ni15Mo3NbB stainless steels in the austenitic state (А), cold work (CD), and mechanical-and-thermal treatment (МТT); some under-went isothermal annealing at 700°С, and there was chromium stainless steel (1Cr13Mo2NbVB) after hardening and tempering. For two assemblies the 0Cr16Ni15Mo3NbB steel tubes were used with added rare-earth metals and yttrium (REМ and Y); the weight percentages were 0.05% and 0.01%, respectively. Physical and chemical interactions were studied as a part of postirradiation examination of the fuel pins. 3 Impact of Cesium on Corrosion of Cr–Ni Stainless Steel Cladding 3.1 MATRIX CORROSION Metallographic studies of fuel pin cross-sections1–3 show that the physical and chemical interaction with oxide fuel containing fission products can lead to general (matrix) and intergranular corrosion in cladding at temperatures above
284
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
500°С. Matrix corrosion is an even, unselective interaction of cladding material with the fuel column that thins the cladding wall (Figure 1). At burn-ups of up to >15 atom%, matrix corrosion thins the cladding wall by 30–40 µm in the upper cross-sections of fuel pins, and causes interaction products to form a layer up to 150 µm thick to fill in the gap between the fuel and cladding. The Figure 1. Matrix corrosion in products of this matrix corrosion are non0Cr16Ni15Mo3NbB steel at 650°С metallic compounds with metallic parAnd 15 atom % burn-up ticles that are in the gap between the fuel and the major metal of the cladding. X-ray spectroscopy microanalysis shows that matrix corrosion products contain iron, chromium, nickel, cesium, and small amounts of iodine. At cladding temperatures >600°С the segregation of the following chemical elements are observed in the corrosion products: Cr–Cs, Fe–Cr, and in some cases Cs–I. Metallic particles are made of Fe and Ni. In the particles the Ni concentration is higher than in the original stainless steel and reaches 36%. Chromium depletion of the inner surface is observed. The deepest point of the depleted Cr layer is about 20 µm and is observed in the upper cross-sections of fuel pins. Thus, the major process causing matrix corrosion of Cr–Ni stainless steel cladding is interaction of Cs with Cr that leads to destruction of the protective oxide film on the cladding surface, chromium depletion of stainless steel and partial oxidation of its components with generation of Cs–Cr–O, Cr–Fe–O chemical compounds and the nonoxidized metallic component on a basis of Fe with the increased content of Ni. 3.2 INTERGRANULAR CORROSION Experimental data accumulated during studies of the composition of substances in the corroded boundaries of grains, distribution of stainless steel components near the grain boundaries and distribution of chemical elements in the peripheral part of fuel in the area of corroded cladding allows one to conclude there are two kinds of intergranular corrosion in austenitic Cr–Ni steel. Each kind has a specific behavior of fission products and chemical elements making the cladding material. Physical and chemical processes causing either kind of intergranular corrosion are specified on the basis of the set of features and with regard of the simulated corrosion tests2,3and thermodynamic characteristics.3,4 Evidence of both kinds of corrosion can be seen in the same region of the
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
285
Figure 2. (a) Microstructure of 0Cr16Ni15Mo3NbB steel cladding, and (b) distribution of elements in the area of intergranular corrosion caused by interaction with Cs at 650°С and burn-up 15 atom %
cladding but, as mentioned above, due to difference in conditions of physical and chemical processes in microareas it is even possible to see different kinds of the discussed corrosion in different places of the fuel pin cladding of the same cross-section of a fuel pin. The difference in the two kinds of intergranular corrosion (Figures 2 and 3) is observed even in microstructural features: in the first case the grain boundaries are filled in by the products of interaction, in the second case they are porous. For the corrosion shown in Figure 2 it is specific to have Cs in the corroded grain boundaries where in addition to Cs there is, as a rule, slightly increased content of Cr compared to the matrix. Some chromium depletion of the area close to the boundaries of grains is registered; it is seen at grains larger than 30 µm. Such corrosion is noticed in the upper part of all examined austenite stainless Cr–Ni steel cladding at burn-ups of 8.0–19.7 atom %. Greater irradiation time increases the corroded area down to the middle of the active part of fuel pins. However, maximum depth of such corrosion is not increased with burn-up in the range of 8.0–19.7 atom % and does not exceed 60–70 µm. Thus, the test data show that under certain conditions local areas of selective interaction with Cs appear in Cr–Ni stainless steel claddings in the austenitic state; these areas are at grain boundaries. Interaction with Cs causes not only matrix corrosion but also intergranular corrosion. The available test and computation data show that one of the reasons for intergranular corrosion of claddings made from austenitic 0Cr16Ni15Mo3Nb and 0Cr16Ni15Mo3NbB steels is interaction of Cs with the Cr carbides segregated at the grain boundaries. The maximum depth of this type of corrosion in fuel pins with 8% burn-up is 60 µm and it does not go deeper with increase of irradiation time to 15% ha burn-up of fuel. It is explained by coagulation of segregated carbide in
286
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
(a)
(b)
Figure 3. (a) Microstructure of 0Cr16Ni15Mo3Nb (А) steel cladding, and (b) distribution of elements in the intergranular corrosion area due to interaction with iodine at 690°С and 15.3% h.а.burn-up. the grain boundaries of such steels due to radiation thermal aging during irradiation of fuel pins up to 7–8% ha burn-up at 600°C. It results in a decrease in the filling in of grain boundaries by carbide segregation, and intergranular corrosion is stopped because of the impact of Cs. 4 Iodine Impact on Corrosion of Stainless Cr-Ni Steel Claddings X-ray spectral microanalysis of fuel pin samples reveals a second kind of intergranular corrosion in Cr–Ni stainless steel. It differs from the one discussed in the way in which chemical elements are distributed in the corrosion area. The specific feature of this corrosion is the presence of iodine in the corroded grain boundaries and a change in the ratio of major components of the steel, with a characteristic Fe and Cr decrease and Ni increase compared to the initial content (Figure 3b). The components of the steel are also observed in the nearest peripheral areas of the fuel column. Fe is noticed in metallic inclusions at a distance up to 900 µm from the fuel surface (Figure 4) or in inclusions containing fission products Ru and Mo in addition to Fe at a distance of 0.5–1.2 mm from the cladding. Cr in the composition of oxide phase or in combination with Cs is found in the fuel-cladding gap and in the outer part of the fuel column to a depth of 100 µm from the surface. Sometimes grains with corroded boundaries would tear off from cladding; probably when reactor power was decreased (Figure 5). X-ray spectroscopy microanalysis of the ground crosssection of a fuel pin (see Figure 5) shows that the amount of Ni in grain fragments separated from the cladding is 2.5–3 times higher than in the initial composition of steel.
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
Figure 4. Intergranular corrosion area in 0Cr16Ni15Mo3Nb steel due to interaction with iodine at 690°С and 15.3% hа burn-up
287
Figure 5. Grains in 0Cr16Ni15Mo3Nb steel, with weaker links because of corrosion, get separated from cladding
Thus, the major feature of the second kind of intergranular corrosion of cladding of irradiated fuel pins is the transfer of the steel components Fe and Cr to the fuel column from the cladding. The same phenomenon is observed in simulated corrosion tests of stainless steel samples under a thermal gradient and the presence of iodine or CsI in the closed area.5 During the fission process the yield of iodine is ten times less than that of Cs. These elements can produce a thermodynamically stable chemical compound CsI that is inert to the cladding at low partial pressure of oxygen, as outof-pile experiments under a thermal gradient show. In these experiments the mass transition phenomenon in presence of CsI correlates with the partial pressure of oxygen in the system. The chemical activity of iodine toward steel in this case depends on the chemical potential of oxygen. At rather high partial pressure of oxygen, Cs participates in competing reactions generating Cs chromates that are more stable than CsI, the latter is confirmed by X-ray spectroscopy microanalysis of sample surfaces. Thus, transition of iodine into status chemically active with steel takes place only in the case of production of cesium chromates on the surface of samples during interaction with steel at rather high partial pressure of oxygen. Presence of phases containing cesium at the inner surface of cladding and in the peripheral part of the fuel column, similar to the phase on the surface of samples after simulated corrosion tests, proves the same role of oxygen in the nonisothermal transfer of Fe and Cr both in fuel pins and simulated tests. CsI migrates under the influence of the radial temperature gradient from the fuel column where it enters into chemical interaction with Cr and O under the condition of rather high oxygen activity. A specific feature of this phenomenon in fuel pins is (i) in addition to Cs chromates it is possible to produce Cs uranates or Cs uranium-plutonates, and (ii) radiation decomposition of CsI and transition of iodine to the chemically active form. Figure 6 shows the temperature dependence of the free energy of formation of metal iodides that make the basis of stainless steels, and of Cs as of the study1.
288
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
Свободная энергия, кДж/моль
20 1
-30
2
3
-80 -130
4
-180 -230 -280 -330 200
400
600
800
1000
1200
1400
Температура, оС
Figure 6. Free energy of metal iodide formation: 1: 1/2Ni + I ═1/2NiI2; 2: 1/2Fe + I ═1/2FeI2; 4: 1/2Cr + I ═ 1/2CrI2
Using the data and equation:
∆G = − RT ln D
(a MI 2 )1 / 2 ( a M )1 / 2 p I
where ∆G D is the free energy, a MI 2 , a M , p I are, respectively, the activity of metal iodide, the activity of metal itself in stainless steel, the pressure of iodine vapor, it is possible to calculate the minimal pressure of iodine needed to form iodide of a specific metal. For example, pΙ, values needed to form iodides of Cr, Fe, and Ni in the cladding of steel with the basis of Cr16Ni15 at 700°С would be 20, 70, and 3.1·103 Pa, respectively. Calculations of stability of iodides of some metals depend on their activity (αm) and partial iodine pressure (pΙ) at 730°С.8 For example, the values of pΙ, needed to form iodides of Cr, Fe, and Ni in the 316 steel cladding at the said temperature are 40, 100, and 5·103 Pa, respectively. In case only 0. 1% of iodine, formed during fission of fuel, is in the atomic state then it is enough to have about 0.1 atom % burn-up to make pressure needed to form iodides of Cr and Fe on the surface of cladding at 700°С; for Ni iodide formation it is enough to have about 10 atom % burn-up. 5 Effect of Composition and Pre-Processing on Corrosion 5.1 CORROSION IN 0Cr16Ni15Mo3Nb AND 0Cr16Ni15Mo3NbB AUSTENITIC STEELS Experimental data obtained during examination of fuel pins irradiated to 15% burn-up show the principal difference in corrosion behavior of claddings made of 0Cr16Ni15Mo3Nb and 0Cr16Ni15Mo3NbB steels in austenitic state and confirm the results obtained by simulated corrosion tests of these materials.
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
289
Глубина коррозии, Corrosion depth , µm мкм
120 The results of simulations and irradiated fuel pin tests show that transfer of Fe and Cr from 0Cr16Ni 80 15Mo3Nb steel to the high-temperature area is of a selective nature. 40 Mainly the elements from the grain boundaries of 0Cr16Ni15Mo3Nb steel are transferred and this is the 0 reason for its intergranular corrosion. 500 550 600 650 700 Fe and Cr are transferred from Температура, o °С Temperature, C 0Cr16Ni15Mo3NbB frontally, the corFigure 7. Depth of intergranular rosion does not go deep and then it corrosion with temperature in fades because the upper layer gets 0Cr16Ni15Mo3Nb (А) (○) and enriched with Ni. So, for claddings of 0Cr16Ni15Mo3NbB ( А) (●) stainless 0Cr16Ni15Mo3Nb steel there are two steels at 15 atom burnup types of intergranular corrosion: one caused by interaction with Cs carbide segregated in the grain boundaries, the other by selective iodide mass transfer. In 0Cr16Ni15Mo3NbB steel cladding only the first type of intergranular corrosion is observed. The depth of corrosion in cladding of these steels is different regardless of similar operational conditions (Figure 7).
5.2 EFFECT OF PRETREATMENT ON CORROSION 100
Corrosion depth, µm
Fuel pins were irradiated to the maximum burn-up of 10.8 atom % hа to examine the impact of pretreatment on the corrosion of 0Cr16Ni15Mo3Nb and 0Cr16Ni15Mo3NbB claddings caused by physical and chemical interaction with fuel and its fission products. The data obtained confirm the positive effect of preceding CD and МТT on the resistance of cladding against intergranular corrosion caused by impact of Cs, and lack of such impact on intergranular corrosion because of interaction with iodine. Thus no intergranular corrosion is observed in claddings of 0Cr16Ni15Mo3NbB (CD and
80 60 40 20 0 490
530
570
610 650 Температура, °С
Temperature, oC
Figure 8. Depth of intergranular corrosion with temperature at 10.8% hа burn-up: ∆, □––0Cr16Ni15Mo3Nb (CD/ МТT) ●, ○, +––0Cr16Ni15Mo3NbB (А/CD/МТT)
290
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
МТT) steel. However, for 0Cr16Ni15Mo3Nb, the type and depth of corrosion are the same as in the austenitic state after the above treatment (Figure 8). 5.3 EFFECT OF ALLOYING WITH RARE-EARTH ELEMENTS AND YTTRIUM Corrosion depth, µm Глубина коррозии, мкм
Fuel pins of two assemblies 160 irradiated to 18.1 and 19.7 atom % burn-up were used for examining 120 the specific physical and chemical interaction of 0Cr16Ni15Mo3NbB 80 austenitic cladding. The maximum depth of intergranular corrosion 40 was 140 µm at 670°C. The maximum depth of corrosion of 0 the 0Cr16Ni15Mo3NbB claddings 480 530 580 630 680 that are additionally alloyed with o Температура, °С Temperature, C REM and Y, on fuel pins after 18.1 atom % burn-up was below Figure 9. Depth of intergranular corrosion 70 µm (Figure 9). X-ray spectrowith temperature at 18 atom % burn-up in: scopy microanalysis data showed •––0Cr16Ni15Mo3Nb, о–– that up to >18 atom % burn-up, 0Cr16Ni15Mo3NbB with REM and Y 0Cr16Ni15Mo3NbB cladding tended to corrode intergranularly due to mass transfer of iodine. Additional alloying of steel with REM and Y prevented the intergranular corrosion caused by the described processes but did not affect the corrosion caused by Cs interaction. The data obtained during post-irradiation examination of fuel pins with a wide range of burn-ups and by corrosion simulated tests showed specific corrosive resistance of 0Cr16Ni15Mo3Nb and 0Cr16Ni15Mo3NbB steels when they interact with iodine under the temperature gradient. The specific features can be due to the difference in the state of boundaries of grains of these materials. The grain boundaries in polycrystalline materials are regions with the highest free energy and chemical activity;9,10 diffusion and corrosion take place rapidly there, as well as transmutations and phase segregation, where the increased solubility of inserted elements is observed. This effect of grain boundaries can be neutralized by adding chemically active elements, e.g., boron. Boron enriches the grain boundaries, produces refractory and thermo-dynamically resistant chemical compounds with insertion elements, and slows down diffusion at grain boundaries and decreases their free energy. So, it could be concluded that B in 0Cr16Ni15Mo3NbB steel, in decreasing the free energy of grains, decreases chemical activity, and in this way provides the steel with better corrosion resistance than 0Cr16Ni15Mo3Nb. This feature of the
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
291
physical–chemical interaction of 0Cr16Ni15Mo3NbB with oxide fuel and its fission products ensures its better resistance compared to 0Cr16Ni15Mo3Nb in fuel pins after 15 atom % burn-up in BOR-60. Further irradiation decreases resistance to intergranular corrosion due to iodide chemical transport reactions. It could be due to re-distribution of B on boundaries and in the matrix of grains because of production of B-contained compounds, e.g., carbonitroborides [Nb(C,N,B)], or partial burn-up of B in-reactor. It changes grain-boundary energy, and initiates selective interaction of the steel with iodine at grain boundaries to cause intergranular corrosion. Additional alloying with REM and Y (similar to B in terms of impact on the state of grain boundaries10) ensures high resistance to intergranular corrosion due to iodide transport reactions. 6 Interaction of Fission Products with Ferritic-Martensitic Steel Cladding The strong resistance of 1Cr13Mo2NbV to radiation swelling and radiation creep makes researchers interested in its application as the material for fuel pin cladding for fast-neutron reactors. In this connection it becomes a pressing issue to examine compatibility of 1Cr13Mo2NbV steel claddings. Fuel pins of two
Figure 10. Matrix corrosion (а), pitting corrosion (b), and corrosion depth (c) in 1Cr13Mo2NbVB cladding versus temperature at a burnup of 10.6 atom %
292
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
assemblies at 10.6 atom % burn-up were used for such examination.1 The results show that interaction with Cs causes matrix corrosion of 1Cr13Mo2NbV steel while iodide transfer of chemical elements to fuel causes pitting corrosion. At the same time the resistance of cladding to corrosion is satisfactory at 650– 670°C but drops significantly at 670–710°C (Figure 10).
7 Conclusions 1. Matrix corrosion of Cr–Ni 0Cr16Ni15Mo3Nb and 0Cr16Ni15Mo3NbB steel claddings comes from interaction with Cs at the oxygen potential of oxide fuel with an initial stoichiometry of 2 ± 0.01. Such corrosion increases with cladding temperature and is 30 µm at 680°С and 18.1 atom % burn-up. 2. Carbide segregation at grain boundaries due to radiation-thermal aging of 0Cr16Ni15Mo3Nb and 0Cr16Ni15Mo3NbB steels leads to intergranular corrosion of cladding by Cs. It is 60 µm deep at 8 atom % burn-up and does not increase with further burn-up. CD and MTT efficiently increase resistance to this intergranular corrosion. 3. In fuel pins of fast-neutron reactors some chemical elements, mainly Fe and Cr, may be selectively transferred from claddings to fuel due to iodide chemical transport reactions under the radial thermal gradient, causing intergranular corrosion of 0Cr16Ni15Mo3Nb steel claddings. Such corrosion is 120 µm at 15 atom % burn-up and 700°С. CD and MTT have no effect on this kind of corrosion. 4. The presence of boron in 0Cr16Ni15Mo3NbB steel prevents intergranular corrosion caused by selective mass transfer up to 15 atom % burn-up. Additional alloying of 0Cr16Ni15Mo3NbB steel with REM and Y prevents this corrosion to burn-ups above 18 atom % during irradiation in the BOR60 reactor. 5. Interaction of 1Cr13Mo2NbVB steel claddings with Cs and I results in matrix and pitting corrosion. Temperature has a strong effect on the depth of 1Cr13Mo2NbVB cladding corrosion. At temperatures <550°С the corrosion is negligibly small (≤10 µm), it increases slowly over 550–650°С (10–60 µm), but abruptly increases with temperature to 710°С (60–200 µm).
INTERNAL CORROSION OF FAST-REACTOR CLADDINGS
293
References 1.
Shamardin, V.K., Golovanov, V.N., Kryukov, F.N., Vsaimodeystvie materialov obolochek s oskolkami deleniya oksidnogo topliva reaktorov na bystrykh neitronakh. Voprosy atomnoi nauki I tekhniki . Ser. Fisika radiatsionykh povrezhdeniy I radiatsionnoe materialovedenie, 1989, 2(49), 25–38. 2. Kryukov, F.N., Golovanov, V.N., Shamardin, V.K., Rol’ produktov deleniya topliva v korrosii obolochek tvelov reaktora BOR-60: Preprint NIIAR-21 (536). Dimitrovgrad: NIIAR, 1982. 3. Tsykanov, V.A., Davydov, Y.F., Klochkov, Y.P., Shamardin, V.K., Golovanov, V.N., Kryukov, F.N., Issledovanie fisiko-khimicheskogo vsaimodeistviya oksidnogo topliva s obolochkami tvelov bystrogo reaktora. Atomnaya energiya, 1984, 56(4), 195–199. 4. Kryukov, F.N., Golovanov, V.N., Shamardin, V.K., Rol’ produktov deleniya topliva v korrosii obolochek tvelov reaktora BOR-60: Preprint NIIAR-21 (536). Dimitrovgrad: NIIAR, 1982. 5. Shamardin, V.K., Kryukov, F.N., Metody issledovaniya I resultaty eksperimentov modeliruyushchikh vsaimodeystvie nerzhaveyushchikh staley s produktami deleniya yadernogo topliva. M.: TsNIIAtominform, 1989, p. 31.. 6. Fee, D.C., Kim, K.Y., Johnson, C.E., “Phase Equilibria in the Cs-Cr-O System”, Journal of Nuclear Materials, 84 (1979): 286–294. 7. Fee, D.C., Johnson, C.E., “Cesium-Uranium-Oxygen Chemistry in UraniumPlutonium Oxide Fast Reactor Fuel Pins”, Journal of Nuclear Materials (1981): 99, 107–116. 8. Aubert, M., Calais, D., Le Beuze, R., “Role de l’iode sur le développement de la réaction entre le combustible et sa gaine en acier inoxydable (type 316) dans les réacteurs”, Journal of Nuclear Materials, 58 (1975): 257–277. 9. Kolotyrkin, Y.M., Kasparova, O.V., Segregatsiya primesei na granitsakh nerzhaveyushchikh staley. Korrosiya. Vol. 6. Under the editorship of Y.M. Kolotyrkin, M. VINITI, 1978, pp. 180–217. 10. Savitskiy, Y.M., Popov, V.F., Keis, N.V., Lyubimov, V.N., Vliyanie redkosemelnykh metalov I ikh okislov na plastichnost I antikorrosionnye svoistva nerzhaveyushchikh staley. Voprosy teorii I primeneniya redkosemelnykh metalov. Under the editorship of Ye.M. Savitskiy,. M. Nauka, 1964, pp. 214–217.
CORROSION OF RESEARCH REACTOR ALUMINUM CLAD SPENT FUEL IN WET STORAGE
L. V. RAMANATHAN, * S. M. C. FERNANDES, AND O. V. CORREA Instituto de Pesquisas Energéticas e Nucleares, São Paulo, Brazil Abstract: This paper gives an overview of activities carried out at IPEN, Brazil, in the IAEA-sponsored Coordinated Research Project (CRP) “Corrosion of Research Reactor Al-clad Spent Fuel in Water” and in the Latin American regional project on “Management of Spent Research Reactor Fuel”. In both projects Al alloy coupons in various configurations were exposed to the spent fuel storage pool of the IEA-R1 reactor for various times and then examined. Pitting was the main form of corrosion and pit distribution on coupons revealed the marked influence of settled solids. The AA6061 and AA1050 alloy coupons withdrawn after 2 years were stained with a layer of Bayerite caused by an increase in the average pool water temperature. This observation—part of a normal sequence of activities within the project––enabled reactor operations to explain the perplexing change in color of most in-reactor Al alloy components. It exemplified advantages of conducting a corrosion surveillance program. Analysis of sediments obtained from the pool revealed that the main constituents were oxides of aluminum, iron, silicon, and calcium. Coupons of the Russian alloy SZAV, pitted more inside crevices compared to the alloy AA6061. The former pitted more when coupled with stainless steel. Alloy grain orientation also influenced pitting. Coupons of rolled AA6061 alloy pitted more than coupons made from extruded AA6061 alloy. Details about the spent fuel storage facilities in IPEN and the water purification systems of the IEA-R1 research reactor (RR) are also presented. Keywords: corrosion, aluminum alloys, spent fuel cladding, pitting, settled solids, corrosion surveillance
______ ∗
To whom correspondence should be addressed: L.V. Ramanathan, Instituto de Pesquisas Energéticas e Nucleares (IPEN- CNEN/SP), Av. Prof. Lineu Prestes 2242, Cidade Universitária, 05508-000 São Paulo, Brazil; e-mail:
[email protected]. 295 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 295–306. © 2007 Springer.
296
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
1 Introduction Concerns related to corrosion of Al-clad spent research reactor (RR) fuel stored for long periods in light-water filled pools around the world led to the establishment, by the International Atomic Energy Agency (IAEA), of a Coordinated Research Project (CRP) on “Corrosion of Research Reactor AluminumClad Spent Fuel in Water.”1,2 This CRP was conducted in two phases. Brazil was one of fourteen countries that were invited to participate in this CRP. The other participating countries were Argentina, China, Czech Republic, Hungary, India, Kazakhstan, Pakistan, Poland, Romania, Russia, Serbia and Montenegro, Thailand, and the USA. The IAEA provided a detailed work package and standard corrosion test coupons to each participant. The materials selected for testing were representative of typical aluminum cladding alloys used world wide for RR fuel, handling tools, and storage racks. Aluminum alloys AA5086, AA1100, AA6061, AA6063, and SZAV-1 were used. In addition to the IAEA rack of corrosion coupons, many participants immersed site-specific alloy coupons in their spent fuel pools.3 The participants of the IAEA sponsored Regional Technical Co-operation Project for Latin America (RLA) included Argentina, Brazil, Chile, Mexico, and Peru.4 The objectives of this project were to determine the basic conditions for managing RR spent fuel during operation and interim storage as well as final disposal, and to establish forms of regional co-operation for spent fuel characterization, safety, regulation, and public communication. The Energy and Nuclear Research Institute (IPEN) in São Paulo, Brazil took part in both projects; this paper summarizes the main activities related to this participation. 2 The IEA-R1 Research Reactor IEA-R1 is a pool type, light water moderated, and graphite reflected RR at the “Instituto de Pesquisas Energéticas e Nucleares” (IPEN), which is part of the Brazilian Nuclear Energy Commission. Although designed to operate at 5 MW, IEA-R1 has operated at 2 MW during most of its life. It presently operates at 4 MW in a continuous cycle of 64 h/week. The reactor is used to perform research in nuclear and solid state physics, radiochemistry and radiobiology, production of radioisotopes and to provide irradiation services. Since start-up, about 160 fuel assemblies have been used and about 20 are burned annually. 2.1 SPENT FUEL STORAGE The wet storage facility, located at one end of the IEA-R1 reactor pool contains stainless steel spent fuel storage racks. The racks have been used since 1977. In 2003 the stainless steel racks were lined with an aluminum alloy to minimize
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
297
bimetallic contact between the fuel cladding and the rack and thus reduce galvanic corrosion of the fuel. Figure 1 illustrates the storage section in the IEA-R1 reactor pool.
Figure 1. Storage section of the IEA-R1 reactor pool
The dry storage facility consists of horizontal silos in a concrete wall. Each silo, made of carbon steel, is 20 cm in diameter and 3.5 m long. This facility is presently not in use. However it was used for over 40 years to store spent fuel. 2.2 WATER PURIFICATION SYSTEMS Table 1 gives the typical IEA-R1 reactor basin water parameters. There are two deionization systems to control the water parameters in the reactor basin and in the spent fuel storage section (SFSS).
2.2.1 Radioactive water deionization system This system carries out water purification continuously and removes dissolved solids from the primary circuit water. The water is pumped through one of two parallel deionization loops. In both loops a filter is used to remove all the suspended particles. Removal of dissolved impurities is carried out in mixed bed ion exchangers. Softened water from the non-radioactive deionization system is used for regeneration of the ion exchange resins. Conductivity of the
298
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
water, both entering and leaving the ion exchangers is measured. The system is operated continuously when the reactor is operating.
2.2.2 Non-radioactive water deionization system This system provides deionized make-up water to the reactor pool, to the horizontal irradiation tubes (beam holes), to the radiochemistry laboratory, and for regenerating the ion exchange resins. The system consists of a filter to remove solid particles over 25 µm; a 400 l tank of R–H resin to remove mainly the Ca and Mg ions; a coal filter to retain chlorine and organic impurities and a mixed bed ion exchanger, similar to that in the radioactive deionization system. The ion exchangers and filters are connected in parallel, to permit the use of the flow circuit while the resins in the second loop are being regenerated. There are two conductivity indicators in this circuit as well. TABLE 1. Typical pool water parameters for IEA-R1 research reactor Parameters pH Conductivity Cl ions Fe ions Na ions Temperature Dissolved solids 99 Mo 131 I 133 I 132 Te 239 Np
Units – µS/cm Ppm ppm ppm o C ppm Bq/l Bq/l Bq/l Bq/l Bq/l
Typical values 5.5–6.5 <2.0 <0.02 <0.001 <0.4 25–40 <2 <310 <90 <430 <95 <750
3 Materials and Methods used in Coordinated Research Project Coupons of aluminum alloys AA1100, AA6061, and SZAV-1 (chemical composition shown in Table 2) received from the IAEA, within the context of the two projects were assembled in stainless steel racks with alumina separators. The separators were used to avoid contact between coupons and between the coupons and the rack. A pre-oxidized and scratched coupon of the different alloys was included in the study to simulate the surface of a damaged fuel plate. Site-specific alloy AA6061 coupons were also added to the racks. Surface preparation of the site-specific alloy coupons was carried out as per orientations
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
299
received. The coupons and the racks were photographed and the racks were immersed in the SFSS of IEA-R1. In the RLA project, six racks were immersed in the SFSB. Three racks were positioned horizontally with its coupons oriented vertically. The other three racks were positioned vertically with its coupons oriented horizontally. One rack from each set was withdrawn after 1, 2, and 3 years of exposure.5 Upon withdrawing from the SFSS, the racks were photographed and standard procedures outlined in the CRP “Test Protocol” were used to handle the rack and coupons.6 The pH in the crevices of the crevice and bimetallic couples was measured. The coupon surfaces were examined in an optical microscope coupled to an image analysis system. The pH, conductivity, chloride content, and temperature of the water in the reactor pool and the SFSS were monitored periodically. Prior experience had shown that the amount of dissolved impurities in the water was well below detection limits of standard measurements and was reflected in the conductivity values. The water in the main pool and the SFSS circulates constantly and conductivity was measured at the inlet to the deionization system. TABLE 2. Chemical composition of the aluminum alloys Alloy AA1100 AA6061 SZAV-1
Cu 0.16 0.25 <0.01
Mg <0.1 0.94 0.53
Mn 0.05 0.12 <0.05
Si 0.16 0.65 0.71
Fe 0.48 0.24 0.09
Ti 0.005 0.04 <0.005
Zr 0.03 0.03 0.03
Cr 0.005 0.04 <0.005
3 Program Results 3.1 OBSERVATIONS ON COUPON REMOVAL During disassembly of the racks the coupled coupons were difficult to separate due to formation of oxides within the crevice. Also during disassembly of the rack, the pH of the water in the crevice between the various couples was measured. In all cases, independent of coupon orientation, the pH of the water in the crevice was about 5.5, one point below that of the bulk water pH. 3.2 OPTICAL MICROSCOPY OF COUPONS The AA6061 alloy coupons were almost dark grey whereas the pre-oxidized and scratched coupons of this alloy were quite bright and unattacked. The oxide on the coupon surface was removed to enable examination in an optical microscope. The exposed surfaces of the alloys revealed pits, independent of the orientation of the coupon. However, many features were specific to the alloy,
300
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
the position of the coupon in the rack, and the orientation of the coupon. The SZAV alloy coupons pitted where as the AA6061 coupons did not pit to the same extent. A number of pits were observed on the contact surface of the SZAV alloy compared to that on the AA6061 alloy coupon, and even more in the SZAV alloy contact surface of the galvanic couple SZAV-stainless steel. The distribution of pits on the coupon surfaces was determined and histograms of the number of pits (counts) as a function of pit diameter were plotted. 3.2.1 Horizontally oriented coupons (Rack-47) (1 year exposure) Figure 2 shows the histograms of pit count versus pit size on the AA1050 alloy upward and downward facing surfaces. The surface facing upward revealed a large number of pits, ∼90, in the size range 40–50 µm while the surface of the same coupon facing downwards, revealed only 6–8 pits in the same size range. The shape of the pits on this coupon varied from irregular to round (Figure 3).
Figure 2. Histograms of pit count versus pit size on the surface facing upwards (47506Up) and the surface facing downwards (47506Down) on a horizontally oriented AA1050 coupon exposed for 1 year to the IEA-R1 SFSS
Most pits revealed a bright region around them, characteristic of a cathode region around a localized anode region. The shape of this region varied from circular to elliptical (Figure 4). On the AA1050 alloy coupon surface in contact with the AISI 304 stainless steel coupon, larger pits were observed, revealing the deleterious effect of a bimetallic contact. The two surfaces of the pre-oxidized and scratched AA1050 alloy coupon revealed few small pits and no bright regions around the pits. Also, no pits were observed along the scratch.
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
301
Figure 3. Micrographs revealing pits and bright regions on horizontal AA1050 alloy coupon surfaces
3.2.2 Vertically oriented coupons (Rack 44) (1 year exposure) On the surfaces of individual AA1050 alloy coupons, round and irregularly shaped pits were observed. Many pit clusters were also observed. The bright cathode areas associated with the pits were shaped like a comet with a tail, giving a clear indication of the top and bottom of the vertically oriented coupons (Figure 4) Almost all the pit features observed on AA1050 alloy coupon surfaces were observed on the exposed surfaces of AA6061 alloy coupons. Bright regions or comet tails were not observed around irregular shaped pits indicating that the mechanism associated with the formation of round and irregular shaped pits are different. The contact surfaces of the crevice couple coupons, AA1050– AA1050, AA1050–AA6061, and AA6061–AA6061 couples were stained and did not reveal any pits. The stains on the surfaces of the two alloys were distinct and characteristic of the alloy. The surfaces of the pre-oxidized and scratched AA1050 alloy coupon revealed a few small pits and no bright regions around the pits.
302
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
Figure 4. Optical micrograph of vertically oriented AA1050 alloy coupon surface revealing the comet shaped bright region around a pit
Histograms of number of pits versus pit sizes on coupon surfaces exposed horizontally and vertically for 2 and 3 years were also obtained. Similar observations were made to those on coupons exposed for 1 year. 3.2.3 Effect of coupon orientation on pitting corrosion Comparison of pit histograms for horizontal and vertical surfaces of AA1050 coupon exposed for 1 year (Figure 5) revealed that twice as many pits (size range 40–50 µm) form on the horizontal coupon as on the vertical coupon.
Figure 5. Histograms of pit count versus pit size on the AA1050 alloy coupon surface facing upwards (47504Up) and a vertically oriented surface of the same alloy (44504A)
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
303
Comparison of pit histograms of vertically and horizontally oriented AA6061 alloy as well on coupons of both alloys exposed for 2 and 3 years also revealed similar behavior. This indicated that among the many parameters that control pit formation—such as alloy composition, metallurgical state, and water parameters—settled solids contribute to pit initiation and formation. 3.3 EFFECT OF POOL WATER TEMPERATURE ON COUPON CORROSION This part of the study was initiated when IEA-R1 reactor operations reported greying of all Al surfaces inside the reactor pool. Consequently some racks were withdrawn after 22 months of exposure instead of the planned 24 months.The surfaces of coupons exposed for 22 months (nominal 2 years) were darker and stained compared to those exposed for 12 months. These stains obscured the pits from being observed in an optical microscope. The stained regions were analyzed by SEM and XRD and found to be the aluminum oxide Bayerite, as shown in the micrographs in Figures 6 and 7. This oxide forms at temperatures below 70ºC. The pool water temperature was reported to be 10ºC higher than normal during a 6-month period. The rack with coupons exposed for 12 months were withdrawn prior to the increase in reactor water temperature. This 6-month period also coincided with reactor operation at 5MW. The increase in pool water temperature was attributed to insufficient cooling capacity of the heat exchanger. The reactor power has subsequently been reduced to 4MW awaiting installation of a new heat exchanger. This exemplifies the added advantage of conducting a corrosion surveillance program.
Figure 6. Scanning electron micrograph of AA1050 alloy coupon surface exposed for 12 months in the IEA-R1 SFSS
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
304
Figure 7. Scanning electron micrograph of AA1050 alloy coupon surface exposed for 22 months in the IEA-R1 SFSS
3.4 EVALUATION OF SETTLED PARTICLES ON COUPON SURFACES A sediment collector was installed in the SFSS close to the corrosion test racks for a period of 4 months. After this period the collector was removed, the water and settled solids retained in it were stirred, filtered, and the sediments collected on a filter paper. Subsequently the sediments were dried in an oven at 100ºC for 24 hours, weighed, mixed, and representative specimens examined in the SEM and analyzed by X-ray fluorescence spectroscopy. Table 3 gives the quantitative analysis of the sediments; they consisted primarily of the oxides of Al, Si, Fe, and Ca. The rate of sedimentation during the 4-month period was found to be 0.178 mg/cm2/month. TABLE 3. Quantitative X-ray fluorescence spectroscopic analysis of sediments from the dust collector Oxide Percent Oxide Percent
Al2O3 56.785 P2O5 0.432
SiO2 21.042 PbO 0.336
Fe2O3 14.93 MnO 0.186
CaO 2.352 ZnO 0.136
Cr2O3 1.594 ZrO2 0.102
TiO2 0.757 Ag2O 0.102
NiO 0.580 CuO 0.063
K2O 0.563 MoO3 0.041
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
305
4 General Discussion Comparison of pit histograms of horizontally oriented top surfaces of AA1050 alloy coupons with that of vertically oriented surfaces of the same alloy for one, 2 or 3 years revealed that twice as many pits (size range 40–50 µm) form on the former as compared to that on the vertically oriented coupon. Comparison of pit histograms of vertically and horizontally oriented AA6061 alloy coupon surfaces also revealed a similar behavior. This indicated that among the many parameters that control pit formation, such as alloy composition, metallurgical state and water parameters, settled solids contribute to pit initiation and formation. Microscopic examination of the crevice surface of the SZAV–SZAV crevice couple revealed a number of small pits besides the aluminum oxide. The contact surface of the SZAV coupon with stainless steel in the galvanic couple revealed many more pits compared to the contact surface with the same alloy. The contact surface of AA6061, either in contact with AA6061 or stainless steel, on the other hand did not reveal a similar distribution of pits. No pits were observed on the pre-oxidized AA1050 and AA6061 surfaces. No pits were observed along the scratch on the pre-oxidized and scratched coupons. The facing surfaces of the crevice couple coupons, AA1050–AA1050, AA1050–AA6061 and AA6061–AA6061 were stained and did not reveal any pits. The stains on the surfaces of the two alloys were distinct and characteristic of the alloy. 5 Conclusions 1. Pitting was the main form of corrosion of aluminum alloy coupon surfaces exposed to the spent fuel storage section of the IAE-R1 reactor. 2. The top surfaces of horizontally oriented coupons pitted more than the surfaces facing downwards. The extent of pitting on the top surface of horizontal coupons decreased with the position of the coupon from top to bottom in the rack. 3. The two sides of vertically oriented coupons of both alloys pitted to the same extent. 4. The extent of pitting of vertically oriented coupons was considerably less than that of the horizontally oriented coupons. This indicated that pit formation is influenced by, among other factors, settled solid particles on the coupon surface. 5. The horizontally and vertically oriented pre-oxidized coupon surfaces pitted to a lesser extent than the corresponding un-oxidized alloy coupon surfaces. 6. Coupon orientation had no noticeable effect on crevice or galvanic corrosion.
CORROSION OF AL-CLAD SPENT FUEL IN WET STORAGE
306
7. The contact surface of SZAV coupons, from a crevice or a galvanic couple, revealed a number of small pits. The surface in contact with stainless steel revealed more pits. 8. Rolled AA6061 coupons pitted more than extruded AA6061 coupons. 9. The surfaces of coupons exposed to water at 45ºC were stained with a layer of the Bayerite. 10. The main constituents of the SFSS sediments were oxides of aluminum, iron, silicon, and calcium. 11. Overall, coupon orientation has a marked effect on the corrosion behavior of aluminum alloy coupons. References 1.
2.
3.
4.
5.
6.
Howell, J.P., “Corrosion Surveillance for Research Reactor Spent Nuclear Fuel in Wet Basin Storage”, Corrosion-99, National Association of Corrosion Engineers, Houston, USA, paper 462, 1999. Howell, J.P., “Criteria for Corrosion Protection of Aluminum Clad Spent Nuclear Fuel in Wet Storage”, Corrosion-2000, National Association Of Corrosion Engineers, Houston, USA, paper 200, 2000. Ritchie, I.G., Ramanathan, L.V., Howell, J.P., Haddad, R., Luo, S., Bendreskaya, O.S., Yakovlev, V., Laoharojanaphand, S., Hussain, N., De, P.K., Johnson, A.B., Jr., Vidowsky, I., “Corrosion of Research Reactor Al-Clad Spent Fuel in Water”, Proceedings of 24th Reduced Enrichment for Research and Test Reactors, RERTR2002, Bariloche, Argentina, 2002. Ramanathan, L.V., Haddad, R., Ritchie, I., “Corrosion Surveillance Programme for Latin American Research Reactors Al-Clad Spent Fuel in Water”, Proceedings of 24th Reduced Enrichment for Research and Test Reactors, RERTR-2002, Bariloche, Argentina, 2002. Correa, O.V., Lobo, R.M., Fernandes, S.M.C., Marcondes, G., Ramanathan, L.V., “Effect of Coupon Orientation on Corrosion Behavior of Aluminium Alloy Coupons in the Spent Fuel Storage Section of the IEA-R1 Research Reactor”, Proceedings of the International Conference on Research Reactor Utilization, Safety, Decommissioning, Fuel and Waste Management, Santiago, Chile, 2003, p. 637. Corrosion of research reactor aluminium clad spent fuel in water, Technical Report Series No. 418, International Atomic Energy Agency, Vienna, 2003, ISBN 92-0113703-6.
INFLUENCE OF NEUTRON IRRADIATION ON MECHANICAL AND DIMENSIONAL STABILITY OF IRRADIATED STAINLESS STEELS AND ITS POSSIBLE IMPACT ON SPENT FUEL STORAGE F. A. GARNER∗ Pacific Northwest National Laboratory Richland, Washington, USA Abstract: Stainless steels used as cladding and structural materials in nuclear reactors undergo very pronounced changes in physical and mechanical properties during irradiation at elevated temperatures, often quickly leading to an increased tendency toward embrittlement. On a somewhat longer time scale there arise very significant changes in component volume and relative dimensions due to void swelling and irradiation creep. Irradiation creep is an inherently undamaging process but once swelling exceeds the 5–10% range austenitic steels become exceptionally brittle. Other processes also contribute to embrittlement and thereby contribute to difficulty in storing and handling of spent fuel assemblies removed from decommissioned fast reactors. In light water reactors other forms of embrittlement develop prior to reaching significant levels of void swelling. A review is presented of our current understanding of the radiationinduced changes in physical and mechanical properties that contribute to embrittlement. Keywords: stainless steels, fast-neutron flux, displacements per atom (dpa), transmutation, embrittlement, void swelling, radiation-induced segregation, phase instabilities
1 Introduction Stainless steels are used as structural materials in a variety of nuclear environments, most particularly in water-cooled test reactors, water-cooled power reactors, and in sodium-cooled fast reactors. In these various reactors there are significant differences in both neutron flux-spectra and neutron fluence accumulated in various structural components constructed from stainless steels. However, when considering fuel storage it is important to note that stainless steels are not used to directly enclose fuel in modern light-water power reactors, but serve
______ * To whom correspondence should be addressed: F. A. Garner, Pacific Northwest National Laboratory, Richland, Washington 99354, USA; e-mail:
[email protected] 307 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 307–327. © 2007 Springer.
308
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
only as structural components that support or surround fuel assemblies where the cladding is made from various zirconium alloys. In some water-cooled test reactors and most sodium-cooled fast reactors, however, stainless steel is used not only for most structural components but also for direct containment of fuel in cladding tubes or plates. It is in this context that safety issues relevant to spent fuel storage are most important. Safety issues associated with storage of stainless steel-clad fuel fall into two major categories, radiation-induced embrittlement of the steel and cracking/ corrosion issues. This paper will concentrate primarily on the embrittlement issue but will indicate where the second issue is influenced by the first. 2 Neutron Damage When assessing the pronounced changes in physical, dimensional, and mechanical properties of irradiated stainless steels it is important to note that damage arises from two primary processes. These are the neutron-induced displacement of atoms from their matrix lattice positions and the chemical alteration of the steel via transmutation. Both of these processes are sensitive to the details of the neutron flux-spectra and both can strongly impact the mechanical properties of the steel during irradiation. In addition to the brief summary presented below on flux-spectra issues the reader is referred to a recent review paper by the author.1 There are significant differences in neutron flux-spectra for water-cooled and sodium-cooled reactors. There are even larger differences in various fusion or spallation neutron environments.2 These differences lead to strong variations between various reactors in their ability to displace atoms from lattice positions. 2.1 DISPLACEMENT DAMAGE In order to evaluate the differing potential for displacement-induced damage, as reflected in Figure 1a for radiation-induced strengthening in different neutron spectra, it was necessary for the radiation damage community to develop an exposure parameter that is proportional to the total deposited nuclear energy capable of inducing atomic displacement. This exposure parameter is designated as “displaced atoms per atom”, usually referred to as “dpa” for shorthand.3,4 When the accumulated neutron flux-spectrum induces 10 dpa for instance, it means that every atom has been displaced from its lattice site an average of 10 times over its in-core residence time. As seen in Figure 1b, the use of dpa as an exposure correlation parameter successfully collapses the data into one response function.5 This allows data to be compiled and compared between different reactors and neutron spectra, providing that transmutation differences do not strongly influence the property of interest.
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
309
Figure 1. Radiation-induced yield stress changes of 316 stainless steel versus (a) neutron fluence, and (b) displacements per atom (dpa).5
The dpa produced per unit neutron exposure is referred to as “displacement effectiveness” and varies across a given reactor. These variations reflect primarily the role of neutron leakage in smaller cores such as EBR-II and changes in composition of fuel, coolant and structural materials in various regions in or out of core, such as shown for the larger FFTF reactor in Figure 2. The mid-core displacement effectiveness for EBR-II is significantly higher than that for FFTF, reflecting the difference between metal fuel and mixed oxide fuel, respectively, on shaping the neutron spectra.
310
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
Figure 2. Variation of “displacement effectiveness” per fast neutron with core position in EBR-II (top) and FFTF (bottom) fast test reactors1
Depending on the reactor type and particular component the lifetime accumulated dose can range from a maximum of ∼1–3 dpa in a BWR shroud, to 70–100 dpa in the reentrant corners of PWR baffle-former assemblies, and to above 100 dpa in fast reactor fuel cladding and fuel assembly ducts.
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
311
The majority of the displacement dose occurs for atomic collisions taking place at the higher neutron energies (>20 keV). Most transmutation events, however, occur at energies well below those producing displacements. Thermal neutrons especially are very important in transmutation, exhibiting a Maxwellian distribution centered at the coolant temperature, which is ∼0.023 eV at room temperature. Therefore an important figure of merit to determine the relative influence of transmutation per dpa is the thermal-to-fast neutron ratio. This quantity varies strongly between various types of reactor but also can vary with position and often with time in a given reactor.1 2.2 TRANSMUTATION EFFECTS In stainless steels the most significant transmutation changes that arise in various fission reactor spectra involve primarily the loss of manganese to form iron, loss of chromium to form vanadium, conversion of boron to lithium and helium, and formation of helium and hydrogen gas, the latter gases arising predominantly from the various nickel isotopes but also to a lesser extent from other elemental components.6–8 Each of these changes can impact the evolution of mechanical properties. Vanadium forms carbide precipitates that change the activity of carbon in the alloy matrix. Carbon plays a number of important roles in the evolution of microstructure and especially grain boundary composition. The latter consideration is very important in determining the grain boundary cracking behavior, designated irradiation-assisted stress corrosion cracking, or IASCC, especially with respect to the sensitization process.9,10 The strong loss of manganese in highly thermalized neutron spectra has been suggested to destroy the stability of insoluble MnS precipitates that tie up S, Cl, and F—all active elements in grain boundary cracking.11,12 Late-term release of these impurities to grain boundaries may participate in cracking but has not yet been conclusively demonstrated. Helium also affects the development of matrix microstructure and the integrity of grain boundaries.13 The potential role of hydrogen in IASCC is especially uncertain at present. To the first order, however, it is the displacement-induced evolution of microstructure that drives the changes in mechanical stability, with transmutation in most cases initially playing only a secondary role. In the following sections the displacement-induced damage will be covered, with transmutationinduced contributions introduced where appropriate.
312
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
3 Radiation-Induced Microstructural and Microchemical Evolution When metals are subjected to displacive irradiation, especially at elevated temperatures, an intricate and coordinated coevolution of microstructure and microchemistry commences that is dependent primarily on dpa rate and temperature, and secondarily dependent on applied or internally generated stresses. The driving force for this evolution is the simultaneous displacement-induced production of both vacancies and interstitial atoms, each of which has different attractions to various microstructural components, different modes of aggregation and especially different diffusional characteristics. When combined with the different diffusional characteristics of the various elemental components, the steel is now subject to new and powerful non-equilibrium driving forces that alter all facets of microstructure and microchemistry. The original dislocation microstructure quickly responds to mobile displacement-generated defects, leading to reductions in dislocation density and distribution, especially in cold-worked steels most frequently used for fuel cladding and structural components. These dislocations are quickly replaced by new microstructural components at very high densities, with two-dimensional interstitial Frank loops first dominating the microstructure, then generating new line dislocations via unfaulting and interaction of loops. At lower temperatures found in water-cooled test reactors especially, the microstructural features appear to be three-dimensional vacancy clusters and two-dimensional vacancy or interstitial platelets which are probably also small dislocation loops. These “defect clusters” at temperatures below ∼200°C are usually too small to be easily resolved via microscopy and are often characterized as “black dots”. The cluster and dislocation loop evolution is frequently concurrent with or followed by the loss or redistribution of preexisting precipitates. Most importantly, new radiation-stabilized precipitates at high density often appear with crystal structure and composition that are not found on an equilibrium phase diagram for austenitic steels. Subsequently the microstructure at higher doses often develops very high densities of crystallographically faceted, vacuum-filled holes called voids, thought to nucleate on helium clusters formed by transmutation. The latter phenomenon is not a volume-conservative process and the metal begins to “swell”, sometimes increasing in volume to levels of many tens of percent. Concurrently the dislocation microstructure responds to the local stress state, moving mass via a volume-conservative process designated irradiation creep. In general irradiation creep is not a directly damaging process but can lead to failures resulting from distortion that leads to local blockage of coolant flow or strong post-irradiation withdrawal forces. Both swelling and creep are interrelated processes that produce significant distortions in component dimensions. Figure 3 shows some examples of such distortion.
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
313
Figure 3. Swelling and irradiation creep of cold-worked 316 stainless steel in FFTF: (upper) extension and distortion in fuel pin length across the top of an assembly due to temperature, dpa rate and production lot variations, and (lower) spiral deformation of fuel pins via swelling-creep interaction with the spiral wire wrap.
Each of the evolving microstructural components interacts with the nonequilibrium driving forces mentioned earlier and alters its local chemical environment and that of other components, often accelerating the overall development of the microstructural ensemble. Some elements, especially nickel, silicon, titanium, and phosphorus, are strongly concentrated in new or preexisting precipitates but also become concentrated at the large and growing internal surface area associated with the voids. This elemental segregation and removal process strongly changes the matrix composition, pushing the matrix toward ferrite and martensite states. The author apologizes for the preceding discussion of radiation-driven evolution being rather descriptive, glossing over the many intricate details of the microstructural–microchemical evolution. A comprehensive description of the evolution and its dimensional and mechanical consequences is also available.14,15
314
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
Eventually the microstructural/microchemical ensemble approaches a quasiequilibrium condition or “saturation” state, usually at less than 10 dpa, with the transient duration in general decreasing with decreasing irradiation temperature. As a consequence the mechanical properties tend to stabilize at levels depending primarily on temperature and to a lesser extent on dpa rate. Interestingly, the saturation state is independent of the starting thermal-mechanical state of the material. If irradiation continues long enough the memory of the starting microstructural state and the associated mechanical properties is almost completely lost. The only microstructural component that succeeds in resisting this erasure process is that of preexisting, deformation-induced twin boundaries. If this quasi-equilibrium is maintained to higher neutron exposure no further change occurs in the steel’s mechanical properties. However, some slowly developing processes are non-saturable and continue to build-up, eventually forcing the system to jump toward a new instability. These new instabilities are usually associated with the microchemical evolution, involving continued segregation, continued transmutation, or a combination of the two factors.16 Three such late-stage instabilities of importance to the subject of this workshop are discussed later in this paper. 4 Radiation-Induced Changes in Mechanical Properties The first manifestation of the radiation-induced microstructural/microchemical evolution arises in changes of the yield properties,17 as shown in Figures 4 and 5. Note that radiation-induced changes in strength are roughly independent of composition within the 300 series stainless steels.
Figure 4. Strengthening of various annealed 300 series stainless steels versus dpa in various water-cooled reactors17
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
315
Figure 5. Strengthening versus temperature and the microstructural components causing it.17 Note peak at ∼300 C, theo approximate PWR outlet temperature
Movement of dislocations is impeded by the microstructural components and the strength of the steel therefore increases. Figure 6 shows that the yield and ultimate strength at most reactor-relevant temperatures not only increase but converge at a relatively low dose of <10 dpa, indicating the loss of workhardening ability.18 The convergence level for yield strength is independent of thermal-mechanical starting condition, but is sensitive to irradiation temperature.19 Similar behavior is observed in the evolution of hardness.20
Figure 6. Convergence of ultimate and yield strengths of annealed 304 stainless steel irradiated in EBR-II and tested at 370°C.18
316
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
Such convergence behavior has been observed many times, but there are exceptions; for example, cold-worked steels converge in their notch tensile strength, but not to the level reached by annealed steels.21 Such behavior is usually observed in steels which twin heavily during deformation and were irradiated at low temperatures that resist recrystallization. Twin boundaries are not easily erased by displacements, so their hardening contribution persists. Concurrent with an increase in radiation-induced hardening is a loss of ductility,22,23 as shown in Figure 7, which also demonstrates the effect of dpa level and irradiation temperature on yield strength and ductility of annealed steel.
Figure 7. Neutron-induced changes in tensile properties of annealed 1.4988 stainless steel irradiated in DFR. Ductility declines as strength increases.22
Figures 6 and 7 present the features of the first stage of embrittlement that saturates at low dpa levels long before the onset of any late-term instabilities. The most pronounced late-term instability arises from the slowly increasing level of void swelling. Figure 8 shows an example of voids and precipitates formed in AISI 304 stainless steel at dpa rates characteristic of PWR austenitic internals.24 Swelling is a non-saturable process in stainless steels, eventually proceeding at ∼1% per dpa at all fast reactor-relevant temperatures and can approach 100% volume change,25 as shown in Figure 9.
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
317
Figure 8. Voids and carbide precipitates in annealed 304 steel irradiated to 22 dpa in EBR-II at 0.84 × 10−7 dpa/s, similar to the corner of a PWR baffle-former assembly.24 The swelling is ∼1%. Carbides do not form at such densities at higher dpa rates.
Figure 9. Swelling measured by density change in 20% cold-worked AISI 316 stainless steel after irradiation in EBR-II.25
318
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
While voids in themselves are not strong hardening components, they have other indirect consequences. Swelling decreases the various elastic moduli at 2% per each percent of volume change,26 but this softening effect is only secondary in importance to determine the evolution of mechanical properties. At >5% swelling levels voids begin to inhabit a larger portion of any planar cross-section and thereby concentrate and locally elevate any applied stress.27 Most importantly, however, the continued intensive segregation of nickel to void surfaces and the concurrent rejection of chromium leads to strong changes in composition in the matrix during irradiation, pushing the matrix toward ferrite at higher temperatures, especially for steels with nickel content of <10%.28,29 In some observations voids encased in austenite shells exist in a pure ferrite matrix. To date no significant component failure has been observed to result from this particular late-term instability. Ferrite does not easily form in most irradiated stainless steels with nickel levels >12%, however, but the matrix approaches martensitic instability, especially when deformed at temperatures characteristic of storage conditions. This instability is a consequence of stress concentrations between voids, and changes in matrix composition that lead to lowered stacking fault energy and higher martensite start temperature.30 The first stage of the instability arises from stress-induced formation of very high densities of thin platelets of the epsilon martensite phase, which nearly doubles the yield stress during the first micro-moments of deformation. Epsilon martensite is basically a stacking fault in austenite that is more easily induced at lower temperature, lower stacking fault energy, and high stress concentration.
Figure 10. Flow localization in a 304 steel tensile specimen manifest as shearing of voids. At 50 dpa and ∼400°C there is 100–200% strain in the deformation band23
The early stages of this evolution involves increasingly intense localization of flow as swelling approaches 3–5%,23 as demonstrated by Figure 10. By ∼10% swelling, however, the steel has zero tolerance for any physical insult, especially
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
319
when stressed or shocked at lower temperatures. Any crack initiation event now involves intense stress concentration at the crack tip, which converts the epsilon martensite ahead of the crack to the very brittle alpha martensite. At this point the tearing modulus falls essentially to zero and any crack propagates quickly and completely through the embrittled component. Examples24,31–34 are illustrated by Figures 11–13. Whether slow straining or shock loading occurs the result is the same,35 as shown in Figures 14 and 15.
Figure 11. Void-induced embrittlement of an annealed 304 steel EBR-II assembly duct after ∼54 dpa at ∼400°C. The duct broke during routine handling in the hot cell. The numbers indicate flats on the hexagonal duct; the circular features are load pads.
Figure 12. Embrittlement failure in BOR-60 assembly ducts. The ducts were made of annealed X18H10T, Russian equivalent of 321 steel. The dpa/% swelling values were: BY-97: 53/ 27.8; BY-92:52/29.8; U-796: 34/14.24,32
320
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
Figure 13. Failure of 20% cold-worked D9 (Ti-modified 316) cladding during routine handling.33 Failure occurred where 90 dpa was reached at ∼460°C in EBR-II, producing ∼32% swelling. Fuel was lost from the open section.
Work by Porollo34 showed severe void-induced embrittlement can occur at temperatures of 335°C and 365°C, suggesting that PWR components which locally operate at these temperatures might also be subject to such instability. The severe, second-stage, void-induced embrittlement is the most serious consideration for long-term storage and safety of fuel clad with stainless steel. Note that in Figure 13 fuel has fallen out of the broken fuel pin.
Figure 14. Correlation between ductility loss and swelling in several heats of irradiated steels in PHENIX.35 By ∼5% total and uniform elongations converge and by ∼10% no ductility remains.
Figure 15. Correlation of swelling and embrittlement in Charpy impact tests of cold-worked Ti-modified 316 steel irradiated in PHENIX35.
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
321
5 Other Forms of Second-Stage Embrittlement Associated with LWRs In general the levels of transmutants formed in fast reactor spectra have been assessed to be of relatively small consequence to mechanical properties. Compared to light water reactors, transmutation proceeds in fast reactors at rates that are secondary in consequence compared to the influence of void swelling. Generation rates of helium and hydrogen by high energy neutrons are relatively low, and hydrogen is usually sucked out the steel into the sodium.36 Both helium and hydrogen are known to assist in void nucleation and at high levels to form bubbles, although conventional wisdom asserts that hydrogen, unlike helium, is sufficiently mobile in stainless steels at reactorrelevant temperatures that it can not build up to significant levels where it can influence the microstructural evolution or most importantly, the cracking behavior. Hydrogen has been implicated in various types of cracking in various structural alloys under some conditions, however.37 This assumption of hydrogen non-storage becomes even more important when nickel-bearing alloys are irradiated in water-cooled reactors with high fluxes of thermal neutrons. Under such conditions the helium and hydrogen generation rates are increased 1–2 orders of magnitude,38–40 although conventional wisdom has held that hydrogen still will not be retained. It is important to recognize, however, that under irradiation in water there are other strong nontransmutant sources of hydrogen that is injected or diffuses into the steel, sometimes overwhelming hydrogen produced by transmutation. Recently, however, this view of hydrogen non-storage has been challenged by a series of observations by Garner and coworkers’ showing that hydrogen storage indeed occurs under conditions where helium-nucleated voids or bubbles are formed.36,40,41 In one case previously identified “helium bubbles” containing approximately 3,000 appm helium were found to also contain equally large levels of hydrogen. Another new insight has arisen in the last decade that stainless steels irradiated in PWRs can develop not only gas-filled bubbles but also voids at significant levels.24,38 It appears that the lower dpa rates characteristic of PWRs actually accelerate the onset of swelling, occurring at lower dpa levels than observed in fast reactors. Previously thought to be relevant only to fast reactors, void swelling is now considered to be an issue that must be addressed before license extension from 40 to 60 years can be granted.42,43 Even if swelling and its attendant distortion develop into a significant issue for PWRs, until recently there has been no evidence that voids, bubbles and stored hydrogen can cause or accelerate cracking.44 Based on some recent studies, however, the possibility of cavity-stored hydrogen affecting cracking must now be examined further.
322
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
Connerman and coworkers45 with support by Edwards and coworkers46 have recently examined intergranular stress corrosion cracking (IGSCC) of cold-worked 316 stainless steel thimble tubes irradiated to very high dpa levels in a PWR. As the %IGSCC measured in post-irradiation slow strain rate tests climbed to ∼100% with increasing dpa level, retained hydrogen was measured to climb at a correspondingly increasing rate. At 70 dpa and 330°C for example, helium was measured to be ∼600 appm and hydrogen to be ∼2,500 appm. Most significantly, electron microscopy revealed a very high density (1.6 × 1023 m−3) of exceptionally small (<3 nm) cavities (probably bubbles) that not only populate the matrix but strongly coat the grain boundaries. To image these very hard-to-see bubbles Edwards found it necessary to strongly underfocus the electron beam. This suggests the possibility that such cavities may have been overlooked in earlier studies on lower exposure specimens. There is a strong suggestion that the hydrogen and helium filled bubbles on the grain boundaries are accelerating the development of boundary cracking. Connerman and coworkers compared their results with comparable specimens irradiated in the BOR-60 fast reactor.45 There was very little IGSCC in these specimens, very little helium and essentially no hydrogen in these specimens, again suggesting a role of hydrogen in cracking. 6 Other Considerations Concerning Safety of Spent Fuel To this point emphasis has been on embrittlement of steels that develops during irradiation. It should be remembered that radiation-embrittled stored fuel must be stable in all stages of its journey through time. This includes short-term storage in water or sodium either in or out of the core, subsequent storage in air or inert atmosphere in hot cells or interim storage units, and long-term “permanent” storage. Corrosion and cracking in these environments may push a marginally stable, embrittled fuel component beyond its residual endurance limit, leading to cladding failure and release of fuel. In addition to stresses or shocks that a fuel pin may receive during prestorage activities, temperature gradients may exist through the cladding from the fuel residual heat. These gradients will induce through-wall stress gradients that will be maintained for many years and may thereby promote cracking. The intense gamma radiation from fuel and cladding can induce radiolytic decomposition of moist air to produce nitric acid, requiring that moisture levels must be maintained at very low levels throughout the storage lifetime. Even in nominally inert atmospheres small amounts of moisture and air will exist and possibly increase due to leakage during extended storage. The resulting acidic vapor may contribute to corrosion and cracking.
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
323
Fast reactor fuel pins and assemblies can experience a problem arising from incomplete sodium removal. No matter how well washed, sodium will remain at low levels in cavities, cracks, and grain boundaries. Subsequent diffusion to the surface allows reaction with air and water to produce a caustic “white fuzz” often observed on long-residing components in hot cells. This caustic coating can induce slow-developing cracking failures often referred to as “hot cell rot”. 7 Conclusions Stainless steels irradiated at reactor-relevant temperatures will in general suffer two types of embrittlement. The first type is induced primarily by atomic displacements and is not strongly influenced by transmutation processes. This behavior is therefore relatively independent of neutron spectra as long as the exposure in expressed in dpa. In this first case both the microstructure and the mechanical properties approach a saturation or quasi-equilibrium state at exposures of 10 dpa or less. The second type of embrittlement arises from late-term instabilities developing at higher dpa exposures, most of which are associated with continued segregation, transmutation, or combinations of the two. In general this second category of embrittlement is more severe, leading to more abrupt and even catastrophic failure. Transmutation contributions to microstructural evolution and embrittlement are more pronounced in light water reactors than in fast reactors. The influence of dissolution of MnS precipitates, vanadium production and interaction with carbon, helium and hydrogen production, and finally hydrogen storage and bubble coating of grain boundaries have yet to be fully studied and understood. All of these processes contribute to potential fragility of stored spent fuel and must be considered as part of the decision process required to design, build and maintain safe long-term fuel storage. References 1.
2.
3.
Garner, F.A., Greenwood, L.R., “Survey of Recent Developments Concerning the Understanding of Radiation Effects on Stainless Steels Used in the LWR Power Industry”, 11th International Conference on Environmental Degradation of Materials in Nuclear Power Systems––Water Reactors, 2003, pp. 887–910. Garner, F.A., Greenwood, L.R., “Neutron Irradiation Effects on Fusion or Spallation Structural Materials: Some Recent Insights Related to Neutron Spectra”, Radiation Effects and Defects in Solids, 144 (1998): 251–283. Greenwood, L.R., “Neutron Interactions and Atomic Recoil Spectra”, Journal of Nuclear Materials, 216 (1994): 29–44.
324
4. 5.
6.
7.
8.
9. 10.
11.
12.
13. 14.
15. 16. 17.
18.
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
Norgett, J.J., Robinson, J.T., Torrens, I.M., Nuclear Engineering and Design, 33 (1975): 50. Heinisch, H.L., Hamilton, M.L., Sommer, W.F., Ferguson, P. “Tensile Property Changes of Metals Irradiated to Low Doses with Fission, Fusion and Spallation Neutrons”, Journal of Nuclear Materials, 191–194 (1992): 1177. Bates, J.F., Garner, F.A., Mann, F.M., “Effects of Solid Transmutation Products on Swelling in AISI 316 Stainless Steel”, Journal of Nuclear Materials, 103–104 (1981): 999. Greenwood, L.R., Garner, F.A., “Hydrogen Generation Arising from the 59Ni (n, p) Reaction and its Impact on Fission-Fusion Correlations”, Journal of Nuclear Materials, 233–237 (1996): 1530–1534. Garner, F.A., Greenwood, L.R., Oliver, B.M. “A Reevaluation of Helium/dpa and Hydrogen/dpa Ratios for Fast Reactor and Thermal Reactor Data Used in FissionFusion Correlations”, Effects of Radiation on Materials: 18th International Symposium, ASTM STP 1325, 1999, pp. 794–807. P. Scott, “A Review of Irradiation Assisted Stress Corrosion Cracking”, Journal of Nuclear Materials, 211 (1994) 101. Was, G.S., “Recent Developments in Understanding Irradiation Assisted Stress Corrosion Cracking”, 11th International Conference on Environmental Degradation of Materials in Nuclear Power Systems—Water Reactors, 2003, pp. 965–985. Chung, H., Garner, F.A., “Radiation-Induced Instability of MnS Precipitates and Its Possible Consequences on Irradiation-Assisted Stress Corrosion Cracking of Austenitic Stainless Steels”, Effects of Radiation on Materials: 18th International Symposium, ASTM STP 1325, 1999, pp. 647–658. Garner, F.A., Greenwood, L.R., Chung, H.M., “Irradiation-Induced Instability of MnS Precipitates and Its Possible Contribution to IASCC in Light Water Reactors”, Proceedings of the Eighth International Symposium On Environmental Degradation of Materials in Nuclear Power Systems––Water Reactors, 1997, pp. 857–860. Maziasz, P.J., “Overview of Microstructural Evolution in Neutron-Irradiated Austenitic Steels”, Journal of Nuclear Materials, 205 (1993): 118–145. Garner, F.A., Chapter 6: “Irradiation Performance of Cladding and Structural Steels in Liquid Metal Reactors”, Vol. 10A of Materials Science and Technology: A Comprehensive Treatment, VCH, 1994, pp. 419–543. Garner, F.A. “Evolution of Microstructure in Face Centered Cubic Metals during Irradiation: A Review”, Journal of Nuclear Materials, 205 (1993): 98–117. Garner, F.A., Toloczko, M.B., “High Dose Effects in Irradiated Face Centered Cubic Metals”, Journal of Nuclear Materials, 206 (1993): 230–248. Pawel, J.P., Ioka, I., Rowcliffe, A.F., Grossbeck, M.L., Jitsukawa, S., “Temperature Dependence of the Deformation Behavior of Type 316 Stainless Steel after Low Temperature Neutron Irradiation”, Effects of Radiation on Materials: 18th International Symposium, ASTM STP 1325, 1999, pp. 671–688. Holmes, J.J., Straalsund, J.L., “Effects of Fast Reactor Exposure on the Mechanical Properties of Stainless Steels”, International Conference: Radiation Effects in Breeder Reactor Structural Materials, 1977, pp. 53–63.
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
325
19. Garner, F.A., Hamilton, M.L., Panayotou, N.F., Johnson, G.D., “The Microstructural Origins of Yield Strength Changes in AISI 316 during Fission or Fusion Irradiation”, Journal of Nuclear Materials, 103, 104 (1981): 803. 20. Irvin, J.E., Bement, A.L., “Nature of Radiation Damage to Engineering Properties of Various Stainless Steel Alloys”, The Effects of Radiation on Structural Materials, ASTM STP 426, 1967, pp. 278–327. 21. Irvin, J.E., Bement, A.L., Hoagland, R.G., “The Combined Effects of Temperature and Irradiation on the Mechanical Properties of Austenitic Stainless Steels”, Flow and Fracture of Metals and Alloys in Nuclear Environments, ASTM STP 380, 1965, pp. 236–250. 22. Ehrlich, K., “Deformation Behavior of Austenitic Stainless Steels after and during Neutron Irradiation”, Journal of Nuclear Materials, 133–134 (1985): 119–126. 23. Fish, R.L., Straalsund, J.L., Hunter, C.W., Holmes, J.J. “Swelling and Tensile Property Evaluations of High Fluence EBR-II Thimbles”, Effects of Radiation on Substructure and Mechanical Properties of Metals and Alloys, ASTM STP 529, 1973, pp. 149–164. 24. Garner, F.A., Edwards, D.J., Bruemmer, S.M., Porollo, S.I., Konobeev, Y.V., Neustroev, V.S., Shamardin, V.K., Kozlov, A.V., “Recent Developments Concerning Potential Void Swelling of PWR Internals Constructed from Austenitic Stainless Steels”, Proceedings, Fontevraud 5, Contribution of Materials Investigation to the Resolution of Problems Encountered in PWRs, 2002, paper #22. 25. Garner, F.A., Gelles, D.S., “Neutron-Induced Swelling of Commercial Alloys at Very High Exposures”, Proceedings of Symposium on Effects of Radiation on Materials:14th International Symposium, ASTM STP 1046, Vol. II, 1990, pp. 673– 683. 26. Balachov, I.I., Shcherbakov, E.N., Kozlov, A.V., Portnykh, I.A., Garner, F.A., “Influence of Irradiation-Induced Voids and Bubbles on Physical Properties of Austenitic Structural Alloys”, Journal of Nuclear Materials, 329–333 (2004): 617– 620. 27. Neustroev, V.S., Golovanov, V.N., Shamardin, V.K., Atomnaya Energiya, 69 (1990): 345–348 (in Russian). 28. Porter, D.L., “Ferrite Formation in Neutron-Irradiated Type 304L Stainless Steel”, Journal of Nuclear Materials, 79 (1979): 406–411. 29. Porter, D.L., Garner, F.A., Bond, G.M., “Interaction of Void-Induced Phase Instability and Subsequent Void Growth in AISI 304 Stainless Steel”, Effects of Radiation on Materials: 19th International, ASTM STP 1366, 2000, pp. 884–893. 30. Hamilton, M.L., Huang, F.H., Yang, W.J.S., Garner, F.A., “Mechanical Properties and Fracture Behavior of 20% Cold-Worked 316 Stainless Steel Irradiated to Very High Exposures”, Effects of Radiation on Materials: Thirteenth International Symposium (Part II) Influence of Radiation on Material Properties, ASTM STP 956, 1987, pp. 245–270. 31. Porter, D.L., Garner, F.A., “Irradiation Creep and Embrittlement of AISI 316 at Very High Neutron Fluences”, Journal of Nuclear Materials, 159 (1988): 114–121. 32. Neustroev, V.S., Ostrovsky, Z.E., Teykovtsev, A.A., Shamardin, V.K., Yakolev, V.V., “Experimental Studies of the Failure of Irradiated Ducts in the BOR-60
326
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43. 44.
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
Reactor”, Proceedings of 6th Russian Conference on Reactor Materials Science, 11–15 September 2000, Dimitrovgrad, Russia (in Russian). Makenas, B.J., Chastain, S.A., Gneiting, B.C., “Dimensional Changes in FFTF Austenitic Cladding and Ducts”, Westinghouse Hanford Company Report WHCSA-0933VA, Richland, WA, 1990. Porollo, S.I., Vorobjev, A.N., Konobeev, Y.V., Dvoriashin, A.M., Krigan, V.M., Budylkin, N.I., Mironova, E.G., Garner, F.A., “Void-Induced Embrittlement of Austenitic Stainless Steel Irradiated to 73–82 dpa at 335–360°C”, Journal of Nuclear Materials, 258–263 (1998): 1613–1617. Fissolo, A., Cauvin, R., Hugot, J.P., Levy, V., “Influence of Swelling on Irradiated CW Titanium Modified 316 Embrittlement”, Proceedings of Symposium on Effects of Radiation on Materials: 14th International Symposium, ASTM STP 1046, 1990, Vol. II, pp. 700–713. Garner, F.A., Oliver, B.M., Greenwood, L.R., Edwards, D.J., Bruemmer, S.M., Grossbeck, M.L., “Generation and Retention of Helium and Hydrogen in Austenitic Steels Irradiated in a Variety of LWR and Test Reactor Spectral Environments”, 10th International Conference on Environmental Degradation of Materials in Nuclear Power Systems––Water Reactors, 2001 (on CD format). Louthan, M.R., Jr., Iyer, N.C., Morgan, M.J., “Helium/Hydrogen Effects on the Properties of Materials for the APT Target/Blanket Region”, Materials Characterization, 43 (1999): 179–186. Garner, F.A., Greenwood, L.R., Harrod, D.L., “Potential High Fluence Response of Pressure Vessel Internals Constructed from Austenitic Stainless Steels”, Proceedings of the Sixth International Symposium on Environmental Degradation of Materials in Nuclear Power Systems––Water Reactors, 1993, pp. 783–790. Garner, F.A., “Materials Issues Involving Austenitic Pressure Vessel Internals Arising From Void Swelling and Irradiation Creep”, Transactions of the American Nuclear Society, 71 (1994): 190. Garner, F.A., Simonen, E.P., Oliver, B.M., Greenwood, L.R., Hamilton, M.L., Wolfer, W.G., Scott, P.M., “Retention of Hydrogen in FCC Metals Irradiated at Temperatures Leading to High Densities of Bubbles or Voids”, Journal Nuclear Materials 356 (2006): 122–135. Tolstolutskaya, G.D., Ruzhitskij, V.V., Kopanets, I.E., Karpov, S.A., Bryk, V.V., Voyevodin, V.N., Garner, F.A., “Displacement and Helium-Induced Enhancement of Hydrogen and Deuterium Retention in Ion-Irradiated 18Cr10NiTi Stainless Steel”, Journal of Nuclear Materials, 356 (2006): 136–147. Tang, H.T., Gilreath, J.D., “Aging Research and Management of PWR Vessel Internals”, Proceedings, Fontevraud 5, Contribution of Materials Investigation to the Resolution of Problems Encountered in Pressurized Water Reactors, 2002, paper #19 (on CD format). Chung, H.M., “Assessment of Void Swelling in Austenitic Stainless Steel Core Internals”, NUREG/CR-6897, January 2006. Chung, H.M., Shack, W.J. “Irradiation-Assisted Stress Corrosion Cracking Behavior of Austenitic Stainless Steels Applicable to LWR core Internals”, NUREG/CR-6892, January 2006.
STEEL EMBRITTLEMENT EFFECTS IN SPENT FUEL STORAGE
327
45. Conermann, J., Shogan, R., Fujimoto, K., Yonezawa, T., Yamaguchi, Y., “Irradiation Effects in a Highly Irradiated Cold Worked Stainless Steel Removed from a Commercial PWR”, Proceedings of 12th International Conference on Environmental Degradation of Materials in Nuclear Power System––Water Reactors, 2005, pp. 277–287. 46. Edwards, D.J., Simonen, E.P., Bruemmer, S.M., Efsing, P., “Microstructural Evolution in Neutron-irradiated Stainless Steels: Comparison of LWR and FastReactor Irradiations”, Proceedings of the 12th International Conference on Environmental Degradation of Materials in Nuclear Power System––Water Reactors, 2005, pp. 419–428.
DEGRADATION IN MECHANICAL PROPERTIES OF STAINLESS STEELS C0.12Cr18Ni10Ti AND C0.08Cr16Ni11Mo3––MATERIALS FOR HEXAGONAL DUCTS OF SPENT FUEL ASSEMBLIES FROM THE BN-350 FAST NEUTRON REACTOR
K. K. KADYRZHANOV,1 S. B. KISLITSIN,1∗ O. P. MAKSIMKIN,1 O. G. ROMANENKO,2 AND T. E. TURKEBAEV1 1 Institute of Nuclear Physics, Almaty 2 Nuclear Technology Safety Center, Almaty Abstract: This paper discusses results of studying the structural behavior of the hexagonal assembly ducts of the BN-350 fast neutron reactor after irradiation in sodium and subsequent water storage at the reactor. Maximum irradiation dose for different ducts varied from ∼11 dpa for an assembly on the core periphery to ∼55 dpa for one at core center. The irradiation dose rate for various assemblies and along each assembly varied from ∼10−8 dpa/s to 10−6 dpa/s, and temperature of irradiation ranged from 280°C to 400°C. The changes in microstructure and physico-mechanical properties of 12Cr18Ni10Ti and 08Cr16Ni11Mo3 steels were studied from 50 × 10 × 2 mm samples cut from the assembly ducts at various locations. Changes in microstructure and phase composition were studied by optical microscopy, transmission electron microscopy (TEM) and scanning electron microscopy (SEM), as well as X-ray structure analysis. Mechanical tensile tests and microhardness measurements were performed. For each location on the ducts values of the hydrostatic density (HD) and the transmutation He content, which are dependent on damage dose and temperature, were also determined. The paper describes the dependence of swelling on dose, dose rate, and temperature for both steels, as well as the dependence of yield stress, ultimate strength, and uniform and total elongation, on these parameters. Keywords: BN-350 fast reactor, stainless steels, assembly ducts, damage dose, dose rate, irradiation temperature, mechanical property changes
______ *To whom correspondence should be addressed: S. B. Kislitsin, Institute of Nuclear Physics, National Nuclear Center of Republic of Kazakhstan, Ibragimov Str.1, 050032, Almaty, Kazakh-stan; e-mail:
[email protected] 329 J. D. B. Lambert and K. K. Kadyrzhanov (eds.), Safety Related Issues of Spent Nuclear Fuel Storage, 329–349. © 2007 Springer.
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
330
1 Introduction BN-350 was the first fast-neutron power nuclear reactor with liquid sodium coolant. It started operation in 1972 and in 1999 it was shut down. Besides producing electricity the reactor was used to desalinate sea water for the needs of the local population. Spent fuel assemblies unloaded from BN-350 during its operation were subject to “wet” storage in a spent fuel pool at the reactor. Interest in the research of structural materials of the active zone of the fast neutron nuclear reactor BN-350 arose for several reasons: •
• •
The rather limited amount of experimental data on high-dose irradiation of the internals of fast neutron nuclear reactors and the possibility to use these data to improve operating performance and lifetime of reactors of similar design; for example, the BN-600 fast neutron reactor. The possibility of applying data on radiation damage of structural materials in fast neutron reactors to help make decisions on prolonging the lifetime of light-water reactors. The maintenance of safe transport of SNF assemblies from the fast neutron nuclear reactor BN-350 to a place of permanent storage and safe long-term storage in nuclear waste product storage.
In recent years a large body of data on radiation damage of structural steels has become available for materials in the active zone of the fast neutron reactors BOR-60 (Russia) and EBR––II (USA).1,2 For operating parameters such as the maximum dose, dose rate, and coolant temperature rise, these two reactors are similar to BN-350 (see Table 1). Characteristics and properties of core structural materials are also similar; for example, the AISI 316 and AISI 304 austenitic stainless steels are analogs of the 08Cr16Ni11Mo3 and 12Cr18Ni10Ti austenitic steels used in BOR-60 and BN-350. Hence, data obtained on BN-350 stainless steels will allow a better, general understanding of the effects of fast neutron irradiation on materials structure and properties. TABLE 1. Operating parameters of LWRs and fast reactors Parameters Na temp. (°С) Inlet Outlet Dose rate (dpa/s)
VVER PWR
BN-350
BOR-60
EBR-II
250 350
280 450 − 10 8 to − 10 6
380 600 − 10 8 to 0.6 × − 10 6
370 590 − 1.5 × 10 8 to 8.3 × − 10 7
<10
−7
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
331
For a long time the BN-350 reactor was the facility in which new regimes of operation and new structural materials for use in the BN-600 reactor were studied. Thus data on the properties of its core structural materials can be useful to decisions on prolonging the service life of BN-600 reactor, which otherwise ends in 2010. The rather low temperature of the BN-350 sodium compared to EBR-II and BOR-60 raises the possibility of applying its data on radiation damage to the internal structures of light-water reactors. Again, such data might be the basis for developing decisions on life extension for these reactors also. Another important issue is transport of spent fuel from BN-350 in Western Kazakhstan to its place of permanent storage in Eastern Kazakhstan, and its safe long-term storage in a nuclear waste facility. The nuclear and radiation safety of transportation and storage of spent fuel appreciably depends on the mechanical properties of the material of the fuel assembly shrouds. Therefore design of containers for transportation and storage of spent fuel assemblies should be based on knowledge of mechanical properties of the shroud material. Interest in the study of the mechanical properties and structure of materials of the fuel assembly shrouds is caused by the fact that material of some hexagonal shrouds had cracks. Figures 1 and 2 show specimens of material of some hexagonal tube of fuel assembly shrouds cut out from different levels relative to the reactor core midplane.
Figure 1. Fuel assembly V-300 had a longitudinal crack on one side of its hexagonal duct. When cutting a sample from 500 mm above core midplane, it broke into six pieces
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
332
Figure 2. Specimen from 1,205 mm above core midplane of fuel assembly C-197 (two pieces, crack)
2 Results and Discussion 2.1 STUDIED MATERIALS The structural steels 08Cr16Ni11Mo3 and 12Cr18Ni10Тi are analogs of the AISI Type 304 and 321 steels, respectively; their composition is given in Table 2. Research into the microstructure and mechanical properties of these steels was carried out on samples prepared from the ducts of BN-350 fuel assemblies operated under various conditions: blanket assemblies N-214/1, N-214/2, N-42, N-110, core assembly CC-19 and internal blanket assemblies V-300 and V-337 (see Table 3). The ducts were cut in the hot cells at BN-350. Plates with the dimensions 50 × 10 × 2 mm (see Figure 1) were cut out from the hexagonal ducts from different levels relative to the core midplane (label “0”). Irradiation temperature, damage dose, and dose rate of the investigated specimens were calculated taking into the position of fuel assembly in the reactor core and the axial position on the fuel assembly.3,4 TABLE 2. Composition of BN-350 structural steels
С Ni Cr 12Cr18Ni10Тi Base 0.12 ∼10 ∼18 08Cr16Ni11Mo3 0.08– 11– 15–17.5 Base 0.12 12 Fe
Element (wt%) P Mn Mo 0.03
1.6
0.01 1–2
Si
B
S
N
–
0.34
–
0.02 0.01
2.5– 3.5
0.5–1
0.005
0.02 0.01
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
333
TABLE 3. Material, treatment, and damage dose of BN-350 assembly ducts examined after wet storage Assembly, material
Steel treatment
N-214 (II), C0.08Cr16Ni11Mo3 V-337, C0.08Cr16Ni11Mo3 V-300, C0.08Cr16Ni11Mo3 N-110, C0.08Cr16Ni11Mo3 Н-214 (I), C0.12Cr18Ni10Тi N-42, C0.12Cr18Ni10Тi CC-19, C0.12Cr18Ni10Тi
Mechanical and thermal (15% CW + 1 h anneal at 800°С) Mechanical and thermal (15% CW + 1 h anneal at 800°С) Mechanical and thermal (15% CW + 1 h anneal at 800°С) Mechanical and thermal (15% CW + 1 h anneal at 800°С)
Maximum dose (dpa) 15.6 10.8 12.1 12.8
Stabilized condition
15.6
Stabilized condition
17.0
Stabilized condition
55.4
2.2 SPECIMENS AND EXPERIMENTAL TECHNIQUES From the plates obtained from various locations on the assembly ducts samples were cut in the hot cells at INP for mechanical testing and study of microstructure. The cutting scheme is shown in Figure 3. Plates were labeled with the fuel assembly number and axial location relative to the core midplane. For example, N-214/1 “–900” means the plate was cut from assembly N-214/1 at 900 mm below the core midplane.
Figure 3. Scheme for cutting samples from plate specimens of BN-350 ducts
334
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
Each plate was cut into four parts (1, 2, 3, and 4), as shown in Figure 3. From part 1 was prepared a sample measuring 5 × 5 × 2 mm for study of microstructure by optical metallography and for microhardness measurement. From part 2 samples were prepared for investigation by transmission electron microscopy (TEM samples were disks of 3 mm diameter and 300 µm thick), scanning electron microscopy (SEM), and X-ray diffractometry. Samples prepared from this part were also used for measurements of hydrostatic density (HD) of the material. TEM was used to reveal of voids in the materials and to determine their swelling rate for comparison with the HD measurements. Mechanical test samples measuring 20 mm × 2 mm × 300 µm were prepared from part 3. Mechanical tests were carried out under uniform tensile conditions at ∼20°C at a strain rate of 0.5 mm/min. Yield stress σ0.2, ultimate strength σult, uniform elongation δunif, and total elongation δtot of the material were determined from the tensile tests. 2.3 SWELLING Void swelling was observed in both BN-350 structural steels, as shown in Figure 4. The size distribution of voids, their average size, and density were determined from a minimum of five pictures of a microstructure, and on this basis the amount of swelling was determined.
Figure 4. Voids in irradiated BN-350 steels: (a) C0.08Cr16Ni11Мo3, N-214/2 − “+500” Tirr = 365 °C, 6 dpa at 2 × 10 8 dpa/s, ∆V/V (TEM) = 0.16%, ∆V/V (HD) = 0.3; − (b) C0.12Cr18Ni10Тi, CC-19 “+300”, Tirr = 400°C, 50 dpa at 100 × 10 8 dpa/s, ∆V/V (TEM) = 4.8%, ∆V/V (HD) = 6.5%.
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
335
HD measurements of void swelling usually give higher values than TEM. For swelling <1%, there is a 4–5 times discrepancy. For 2–5% swelling, the discrepancy between measured TEM and HD values decreases. One reason is that the minimum void diameter revealed by TEM is 8–10 nm on the JEOL 100 CX instrument. At swelling of <1% after doses of 10–15 dpa, a sizeable number of voids are below this value and not accounted for in calculating swelling. When swelling reaches ∼5% at ∼50 dpa, however, the distribution of voids shifts to a larger size and thus a greater number of voids are revealed by TEM and accounted for in calculating swelling. Voids were not observed in all samples, or in some samples of both kinds of steel. Thus it could be argued that void formation did not depend on a single irradiation parameter—temperature, damage dose, or damage dose rate, but on a combination of these parameters. To these parameters should be added the key role played by impurities, particularly helium. For example, according to TEM data, swelling and voids were observed in a C0.12Cr18Ni10Тi sample (N214/1 “0”, Tirr of 337°C, dose of 15 dpa at 4 × 10−8 dpa/s, ∆V/V (TEM) = 0.23%) and at the same time were not observed in the same steel at much higher doses (CC-19 “–300”, Tirr of 330°C, dose of 50 dpa at 114 × 10−8 dpa/s, ∆V/V (TEM) of 0). The first stage of study was to determine swelling in samples cut from different axial locations of the N-214/2 assembly duct (C0.08Cr16Ni11Мo3) to determine axial variation in swelling rate. Further research was then undertaken to reveal dependence of swelling on dose rate and irradiation temperature. 2.3.1 Swelling along height of assembly duct Figure 5 shows the variation in swelling along the height of the N-214/2 C0.08Cr16Ni11Мo3 duct. From TEM data obtained from samples from levels “−900”, the “−500”, “0”, and “+500” mm one can see that swelling was rather small and did not exceed 0.5%. Attention should be paid to the non-uniformity of swelling along the height of a duct as it can lead to the appearance of mechanical stress in the duct material and to a bending of the duct. This fact may be important from the point view of long term storage of spent fuel.
336
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
Figure 5. Axial variation in the swelling of the duct on BN-350 assembly N-214/2; 1: HD data; 2: TEM data
2.3.2 Dependence of swelling on dose rate The dependence of swelling on dose rate for the two BN-350 structural steels are shown in Figure 6. In drawing these figures similar sample data were chosen for both steels with regard to irradiation temperature and damage dose. Note that swelling for these samples was not significant, the maximum swelling not exceeding 0.12% from the TEM data. The figures suggest that swelling decreases with increase in dose rate. This fact has been noted by a number of researchers.1,5,6 The conventional explanation is the lower point defect generation rate at low dose rates, which decreases the efficiency of recombination of vacancies and interstitials. Interstitials, being more mobile than vacancies, are preferentially absorbed by structural sinks like dislocations or grain boundaries, so that the irradiated material becomes oversaturated with vacancies. This favors the formation of vacancy agglomerates, i.e., voids, and results in an increase in swelling rate with decrease in dose rate. 2.3.3 Dependence of swelling on irradiation temperature The dependence of swelling on irradiation temperature for the two BN-350 structural steels is shown on Figures 7a and 7b. It is well known7 that with increase in temperature at the same doses and dose rate swelling increases and
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
337
Figure 6. Dependence of void swelling on dose rate for BN-350 structural steels
reaches a maximum at temperature of Т ∼0.5 Tmelt, and subsequently declines with further temperature rise. The data in Figures 7a and 7b show the initial growth stage in swelling rate with increase in temperature.
338
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
Figure 7a. Swelling rate with increase in irradiation temperature for BN-350 structural steel C0.08Cr16Ni11Мo3
Figure 7b. Swelling rate with increase in irradiation temperature for BN-350 structural steel C0.12Cr18Ni10Тi
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
339
2.3.4 Swelling at low irradiation temperature In our opinion one of the most significant results obtained from study of the BN-350 assembly ducts was the observation of void nucleation and swelling at very low damage doses and temperatures. As can be seen from Figure 8a voids were sporadically found in the C0.08Cr16Ni11Мo3 steel irradiated at 281°C to a dose of ∼1.3 dpa at a dose rate of ∼0.4 × 10−8 dpa/s. In the C0.12Cr18Ni10Тi steel voids and swelling were observed after irradiation at the same temperature at a dose rate of ∼0.4 × 10−8 dpa/s to a dose of only ∼0.7 dpa (Figure 8b).
Figure 8. TEM photos of voids in BN-350 steel ducts at 900 mm below core midplane (a) C0.08Cr16Ni11Мo3 (N-214/2); (b) C0.12Cr18Ni10Тi (N-214/1)
The possibility of investigating the low-temperature swelling of austenitic stainless steels comes from the design peculiarities of BN-350 reactor, which operated at a relatively low inlet temperature for its primary sodium. The sodium inlet temperatures in the EBR-II and BOR-60 test reactors were significantly higher (Table 1). The fact that nucleation of voids and swelling occur in stainless steels at low temperature and very low damage doses and dose rates is likely to be important to extending the service lifetime of light-water reactors. 2.4 MECHANICAL PROPERTIES Mechanical tests were performed to study the strength and plasticity properties (σ0.2, σult, δunif, and δtot) of both BN-350 steels as a function of damage dose, dose rate, and irradiation temperature. These data can be used to reach decisions
340
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
on prolonging the service life of fast neutron reactors and light-water reactors. Tests were carried out on specially prepared samples (section 2.2); values of σ0.2, σult, δunif, and δtot were determined from the results of tests of a minimum of two samples but as a rule from tests on three samples from each location of an assembly duct. Tests were carried out at room temperature (∼20oС). 2.4.1 Effect of damage dose on mechanical property changes The dependence on damage dose of σ0.2, σult, δunif, and δtot of 0.08Cr16Ni11Мo3 steel from the duct of assembly N214/2 duct are shown in Figure 9. One can see that at doses of about 7 dpa the rate of hardening is significantly decreased. Embrittlement of the steel, i.e., significant reduction in its plasticity, takes place even after doses as low as 1–2 dpa.
Figure 9. The yield stress, ultimate strength, total and uniform elongation of BN-350 steel 0.08Cr16Ni11Мo3 as function of dose for BN-350
The curves shown on Figure 9 are characteristic of the dose rate dependency of σ0.2, σult, δunif, and δtot for both types of BN-350 steels. Up to damage doses of ∼7–10 dpa a sharp increase takes place in σ0.2. With further increase in dose up to 56 dpa the strain–temperature curve shows saturation where insignificant increase is observed in σ0.2. The plasticity of irradiated steels displays the opposite behavior. Up to damage doses of 1–2 dpa a sharp decline in plasticity is observed: δunif and δtot decrease from unirradiated material values of ∼35–40% to ∼1–5%; thereafter, to damage doses up to 56 dpa, there is practically no further change. One should note that in the mechanical tests of some samples, for example, from the V-337 duct at 500 mm above core midplane, where the damage dose was 12 dpa and Tirr was 355°C, values close to those for unirradiated material (δunif = 28.5% and δtot = 31%) were observed. Results of investigation of strength and plasticity of
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
341
the BN-350 steels C0.08Cr16Ni11Мo3 and C0.12Cr18Ni10Тi with damage dose are similar to results obtained by other authors for AISI 316 stainless steel irradiated in the EBR-II fast neutron reactor.1,7 2.4.2 Dose rate effect on mechanical property changes The dose rate dependencies of the yield stress σ0.2, ultimate strength σult, uniform elongation δunif, and total elongation δtot obtained for samples of the two BN-350 steels C0.08Cr16Ni11Mo3 and C0.12Cr18Ni10Ti irradiated at the same temperature and damage dose, are shown in Figures 10 and 11. 10
1150 9 8
1050
σ
1000
σ
950
T=305 0 C, D =11.9 dpa
0.2 ult
T=311 0 C, D =12.8 dpa T=302 0 C, D =10.8 dpa
900
T=311 0 C, D =12.8 dpa
7
E lo n g a t io n , %
S tre n g th , M P a
1100
T=305 0 C, D =11.9 dpa
6
δ
5
δ
4 3 2
T=309 0 C, D =7.08 dpa 0
5
tot T=302 0 C, D =10.8 dpa
850 800
unif
T=309 0 C, D =7.08 dpa
1
10
15
Doze rate, dpa/s
20
0
25
5
10
15
20
25
Doze rate, dpa/s x10 -8
x10 -8
Figure 10. Dose rate dependencies of σ0.2, σult, δun, and δtot for C0.08Cr16Ni11Mo3 samples cut off from V337 “–500”, V300 “−500”, N110 “0”, N214/2 “−500” assembly ducts (parameters of irradiation are represented in the figure). 11 50
15 14
11 00
13 12
σ σ
10 00
11 0.2 ult
9 50 T = 2 90 0 C , D = 13 .2 d pa 9 00
T = 3 10 0 C, D = 12 .3 d pa
Elongation, %
Strengh, MPa
10 50
10 9
δ
8 7
δ
6
u nif to t
5 4 3
8 50
T = 2 90 0 C , D = 13 .2 d pa
T = 3 10 0 C, D = 12 .3 d pa
2 4
6
8
Doze rate, dpa/s x10 -8
10
4
6
8
10
Doze rate, dpa/s x10 -8
Figure 11. Dose rate dependencies of σ0.2, σult, δun, and δtot obtained for C0.12Cr18Ni10Ti steel samples cut from N214/1 (0 mm), N42(−300 mm) assembly ducts (parameters of irradiation are represented in the figure).
342
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
The main results, which are important for application to extending the service life of light-water reactors, were as follows: •
•
An abrupt increase in hardening with dose rate was observed in C0.08Cr16Ni11Mo3 steel irradiated at 302–311ºC at values <5 × 10−8 dpa/s (Figure 10). At dose rates >5 × 10−8 dpa/s, σ0.2 and σult did not change so strongly with dose rate, indicating—in our opinion—that there are several stages in the hardening process. At dose rates <5 × 10−8 dpa/s, hardening occurs much more intensively than at higher dose rates and probably saturates at a high damage dose rate. It is reasonable to suppose that the rate of hardening is determined not only by dose rate but also by irradiation temperature, and that an abrupt increase in hardening will be observed at higher irradiation temperatures at higher dose rate values. This process is connected with the accumulation of radiation defects causing hardening and, apparently, depends on irradiation temperature. The dose rate dependencies of σ0.2 and σult for C0.12Cr18Ni10Ti steel samples (N-214/1 “0”, N-42 “−300”) in Figure 11 at first glance seem to contradict the data for C0.08Cr16Ni11Mo3 steel shown in Figure 4, because σ0.2 and σult values increase with decrease in dose rate. But additional investigation revealed that tensile test specimen for N-214/1 “0” was cut from the area closer to a corner of the duct while specimen for N42 “−300” was taken from the center of a duct face, which means that there was initially a difference in microstructure between these samples caused by the duct fabrication process. TEM showed that specimens cut from the area located closer to a duct corner had a dislocation density that was an order of magnitude higher than for a sample cut from the center of a duct face. Hexagonal ducts were fabricated from tubing by means of roll knobbing and the region close to the duct corners was more deformed compared to the center of the duct faces. Figure 5 reflects this fact. Parameters σ0.2 and σult are very sensitive to the microstructure of the samples and adequately responded on this difference in microstructure state for the samples prepared for tensile tests from N214/1 and N42 fuel assembly ducts. Here, we would like to note that the results obtained are not in contradiction with the few data published by other authors,8,9 who conclude there is no strong dependence of σ0.2, σult on dose rate. The dose rate dependence of total elongation δtot demonstrates that both C0.08Cr16Ni11Mo3 and C0.12Cr18Ni10Ti steels are embrittled when damage dose rate decreases (Figures 10 and 11). Such behavior reflects the fact that voids were nucleated in the samples irradiated at low dose rates, while no voids were observed in samples with higher dose rates.
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
343
Based on these investigations and the few data published by other authors the shape of the dose rate dependencies of σ0.2, σult, δunif, and δtot are shown in Figure 12. 10
11 50 11 00
8
Elongation, %
Stress, MPa
10 50 10 00 9 50 9 00
σ
8 50
σ
0.2
6
δ 4
δ
u nif to t
2
ult
8 00 0
5
10
15
Doze ra te, dp a/s x1 0 -8
20
25
0
5
10
15
20
25
D oze ra te, dp a/s x1 0 -8
Figure 12. Dose rate dependencies of σ0.2, σult, δunif, and δtot on samples cut from C0.08Cr16Ni11Mo3 assembly ducts; suggested approximations are dotted lines.
For the dose rate dependencies of σ0.2 and σult a two-stage process is typical: in the low dose rate range, the rate of hardening increases significantly as dose rate increases up to some threshold; thereafter the rate of hardening changes insignificantly and achieves some saturation level. The range of strong dose rate dependence of hardening depends on the irradiation temperature and shifts to higher dose rates with increase in irradiation temperature. Within the range of dose rates and temperatures investigated the dose rate dependency of ductility parameters such as δunif and δtot increased monotonically with dose rate. 2.4.3 Effect of irradiation temperature on mechanical property changes Irradiation temperature dependencies of σ0.2, σult, δunif, and δtot obtained for the samples made of stainless steels C0.12Cr18Ni10Ti irradiated at the same temperature up to the same damage dose are shown in Figure 13. From this figure it was obvious that σ0.2, σult, δunif, and δtot essentially do not change in the temperature range 330–400°C—the range is too small to show any temperature dependence in mechanical property changes. The same lack of temperature dependence applied to data from samples of the C0.08Cr16Ni11Mo3 steel.
344
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
1200
120
σ
0.2
1100
110
σ
B
Stress, MPa
1000
100
"0", D = 50 dpa "-160", D = 56 dpa
900
900 "+300", D = 49 dpa
800
800
"+160", D = 56 dpa 700
700
600
600 500
500 340
360
380
400
Irradiation temperature, 0C
20
"-160", D = 56 dpa
δ δ
20 unif tot
Elongation, %
15
15
"0", D = 50 dpa 10
"+300", D = 49 dpa
10
"+160", D = 56 dpa 5
5
0
0 340
360
Irradiation temperature,
380
400
0C
Figure 13. Temperature dependencies of σ0.2, σult, and δunif, δtot for samples cut from C0.12Cr18Ni10Ti assembly duct CC-19 (level and damage dose are represented in the − curves, dose rate for all samples ∼100 × 10 8 dpa/s).
2.5 STRUCTURE OF C0.08Cr16Ni11Mo3 STEEL 500 mm ABOVE CORE CENTER
A study was made of the structure and mechanical properties of a sample of steel C0.08Cr16Ni11Mo3 from assembly duct V300 at 500 mm above core midplane. Samples for optical metallography and mechanical tests were cut from one of the pieces of plate. Unfortunately the material broke during sample preparation (Figure 14). Optical metallography showed numerous intergranular cracks along which the sample was destroyed. The material had completely lost plasticity and its structure corresponded to a material with high-temperature radiation embrittlement. Microhardness measurements showed that steel within grains had a hardness corresponding to irradiated metal and embrittlement had occurred due to intergranular cracks. It is known that high-temperature radiation embrittlement in steels like C0.08Cr16Ni11Mo3 can occur at temperatures of 650°C and above. The calculated operating temperature did not exceed 346°C.
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
345
Figure 14. Broken sample for mechanical testing (upper) and microstructure of the sample (lower)
In order to clarify the true irradiation temperature, X-ray analysis of the crystal of parts of the broken specimen was performed after one hour isochronous anneals in the temperature range from 100–1000°C at intervals of 50°C. At temperatures lower than working temperature of the duct material the dependencies of the lattice parameter (a) and electrical resistivity (ρ) on the annealing temperature should remain a constant but at temperatures above the operating value the lattice parameter will change and recovery of electrical resistivity will be observed; results are shown on Figure 15. From Figure 15 it is clear that the temperature of the steel in-reactor did not exceed 400°C, which is close to the calculated value. Fracture of the duct sample was probably due to the combined effect of internal stress and migration of impurities (particularly gas) to grain boundaries at temperatures of ∼400°C, and not to high-temperature radiation embrittlement. An interesting fact was the partial recovery of plasticity as a result of the isochronal anneals (see Figure 16).
346
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
Figure 15. Electrical resistivity and lattice parameter of duct samples versus annealing temperature
Figure 16. Parts of fractured sample before annealing (upper picture) and bent parts after isochronal annealing (lower picture)
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
347
3 Conclusions This study showed that the austenitic stainless steels C0.08Cr16Ni11Mo3 and C0.12Cr18Ni10Ti—the structural materials of the hexagonal ducts of fuel assemblies in the BN-350 fast neutron reactor—exhibited significant change in microstructure and mechanical properties after long-term irradiation in BN-350 and subsequent “wet” storage. Swelling was observed in both steels. For damage doses of ∼0.5 dpa up to 56 dpa swelling did not exceed 5% according to TEM data. Swelling was nonuniform along the height of fuel assembly duct, which led to the appearance of mechanical stresses in the material and to bending of the ducts. Void nucleation and swelling at very low damage doses of ∼1 dpa and irradiation temperature of ∼280°C were fixed in both steels. Earlier noted by other workers a decrease in swelling with increase in radiation dose rate was observed in our investigations. Swelling and voids could be observed in samples of steel irradiated to rather small damage doses and at low irradiation temperatures, while at the same time voids were not observed in samples of the same steel at considerably higher damage dose and irradiation temperature. Consequently it was concluded that void formation is not determined by any one factor of irradiation—damage dose, dose rate, or temperature—but depends on a combination of these factors. Impurity content (particularly helium) and internal and external stress in the materials also play important roles in void nucleation and swelling. Mechanical properties of the C0.08Cr16Ni11Mo3 and C0.12Cr18Ni10Ti steels fundamentally change under irradiation and radiation induced hardening and embrittlement of the materials take place. For damage doses up to ∼7–10 dpa, the yield stress of the steels rapidly increases to values about twice those for unirradiated steels; thereafter, they increase slowly up to doses of 55 dpa. The plasticity (elongation) of the steels abruptly decreased at damage doses of 1–2 dpa from unirradiated values of 30–40% to 2–5% for the irradiated materials. In some cases a complete loss of plasticity was observed in the assembly ducts. It resulted in the occurrence of cracks in the material in the top part of the ducts, where irradiation temperature was ∼400°C. The steel structures in these regions of crack propagation resembled those observed in high-temperature radiation embrittlement. Probably such a condition in the assembly ducts resulted from the joint action of internal stress in the duct metal and the migration of impurities (particularly gas) to grains boundaries at temperatures of ∼400°C.
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
348
4 Acknowledgments This work was funded by the European Union and implemented in the framework of ISTC Project K-437 in collaboration with Electricite de France (EdF) and Argonne National Laboratory (ANL) in the USA, whose assistance is highly appreciated. The authors are also grateful to A.I. Ivanov and I.L. Yakovlev of the Magylshak Atomic Energy Complex (MAEC), Kazakhstan, and the staffs of the Laboratory of Investigation of Physical-Mechanical Properties of Materials and the Laboratory of Reactor Materials at the Institute of Nuclear Physics (INP) in Almaty for assistance in the implementation of this work.
References 1. 2.
3.
4.
5.
6.
7.
Garner, F.A., “Irradiation Performance of Cladding and Structural Steels in Liquid Metal Reactors”, Material Science and Technology, 1999, pp. 420–534. Agapova, N.P., Ageeva, N.P. et al., “Investigation of the Structure Spent Fuel Element Shrouds Made from 15%CW Stainless Steel C0.08Cr16Ni15Mo3B, after Exploitation in Fast Neutron Nuclear Reactor BOR-60 after Burnout to 12,5%”, Voprosy Atomnoi Nauki i Tehniki. Atomnoe Materialovedenie, 5, 1982, p. 4 (in Russian). Kadyrzhanov, K.K., Romanenko, O.G., Turkebaev, T.E., Maksimkin, O.P., Kislitsin, S.B., Chumakov, Y.V., “Structure and Mechanical Properties of Stainless Steel C0.08Cr16Ni11Mo3 Irradiated in Fast Neutron Nuclear Reactor BN-350”, Proceedings of the Fifth International Conference on Interactions of Irradiation with Solids, Minsk, Belarus State University, 2003, pp. 157–159. Kadyrzhanov, K.K., Kislitsin, S.B., Maksimkin, O.P., Romanenko, O.G., Turkebaev, T.E., Tsai, K.V., “Swelling of Stainless Steels C0.08Cr16Ni11Mo3 and C0.12Cr18Ni10Ti Irradiated in Breeder Reactor BN-350”, Izvestiya VUSOV of Russia “FIZIKA” N 11, 2004, pp. 141–145 (in Russian). Brager, H.R., Blacburn, L.D., Greenslade, D.L., “Dependence of Displacement Rate on Radiation Induced Changes in Microstructure and Tensile Properties of AISI 304 and 316”, Journal of Nuclear Materials, 122–123 (1984): 332–327. Porter, D.L., Garner, F.A. “Swelling of AISI 304L Stainless Steel in Response to Simultaneous Variation in Stress and Displacement Rate”, ASTM STP V.870, 1985, pp. 212–220. Ibragimov, S.S., Kirsanov, V.V., Pjatiletov, U.S., “Radiation Damageability of Metals and Alloys”, Moscow, Energoatomizdat, 1985, p. 240 (in Russian).
MECHANICAL PROPERTIES OF BN-350 ASSEMBLY DUCTS
8.
349
Allen, T.R., Yoshitake, T. et al., “Tensile Properties of 12% Cold-Work Type 316 SS Irradiated in EBR-II Under Low Dose Rate Conditions to High Fluence”, Effect of Irradiation on Materials, 20th International Symposium on ASTM STP, Eds. American Society for Testing and Materials, West Conshohocken, PA, 2002, pp. 469–486. 9 . Cole, J.J., Allen, T.R. et al., “Swelling and Microstructural Evaluation in 316 SS Hexagonal Ducts following long term Irradiation in EBR-II”, Effect of Irradiation on Materials, 20th International Symposium on ASTM STP, Eds. American Society for Testing and Materials, West Conshohocken, PA, 2002, pp. 413–426.
LIST OF AUTHORS
ADAMS, T. M. Savannah River National Laboratory Aiken, SC 29808, USA ADELFANG, P. Division of Nuclear Fuel Cycle and Waste Technology International Atomic Energy Agency Wagramer Strasse 5, P.O. Box 100 A-1400 Vienna, Austria AISABEKOV, A. Z. National Nuclear Center Institute of Nuclear Physics Ibragimov Str.1, 050032 Almaty, Kazakhstan ALEXANDROV, A. B. ОАО Novosibirsk Chemical Concentrate Plant 94, Khmelnizkiy Str Novosibirsk, 30110, Russia AZHAZHA, V. M. National Science Center Kharkov Institute of Physics and Technology of the National Academy of Sciences of Ukraine Kharkov, Ukraine BAYTELESOV, S. A. Institute of Nuclear Physics Ulugbek, 702132 Tashkent, Uzbekistan
BORDACHEV, V. G. Russian State Scientific Center Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia CHECHETKIN, Yu. V. Russian State Scientific Center Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia CHECHETKINA, Z. I. Russian State Scientific Center Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia CHWASZCZEWSKI, S. Institute of Atomic Energy 05-0400 Otwock, Swierk, Poland CORREA, O. V. Instituto de Pesquisas Energéticas e Nucleares São Paulo, Brazil DEIBLE, R. W. Savannah River National Laboratory Aiken, SC 29808, USA DOSIMBAEV, A. A. Institute of Nuclear Physics Tashkent, Uzbekistan
352
LIST OF AUTHORS
ENIN, A. A. ОАО Novosibirsk Chemical Concentrate Plant 94, Khmelnizkiy Str. Novosibirsk 30110, Russia FEDEROVICH, E. D. I. I. Poluzunov Boiler–Turbine Research and Design Institute 24, Politechnicheskaya Str. St. Petersburg 194021, Russia FERNANDES, S. M. C. Materials Science and Technology Center Instituto de Pesquisas Energéticas e Nucleares São Paulo, Brazil FILLMORE, D. L. Idaho National Laboratory P.O. Box 1625, Idaho Falls ID 83415, USA GAINULLINA, A. Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia GARNER, F. A. Pacific Northwest National Laboratory P.O. Box 999, Richland WA 99352, USA GLUSHCHENKO, V. N. Institute of Nuclear Physics Ibragimov Str.1, 050032 Almaty, Kazakhstan
GOLDMAN, I. N. Division of Nuclear Fuel Cycle and Waste Technology International Atomic Energy Agency Wagramer Strasse 5, P.O. Box 100, A-1400 Vienna, Austria GOLOVANOV, V. N. Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia HERRICK, A. NUKEM Limited Kelburn Court, Daten Park Risley, Warrington Cheshire WA3 6TW, UK HILL, T. J. Idaho National Laboratory P.O. Box 1625, Idaho Falls ID 83415, USA IVANOV, A. I. Mangyshlak Atomic Energy Complex—Kazatomprom 466200 Aktau, Kazakhstan IYER, N. C. Savannah River National Laboratory Aiken, SC 29808, USA JOHNSON Jr., A. B. Pacific Northwest National Laboratory P.O. Box 999, Richland WA 99352, USA KADYRZHANOV, K. K. Institute of Nuclear Physics Ibragimov Str.1, 050032 Almaty, Kazakhstan
LIST OF AUTHORS
KHALIKOV, U. A. Institute of Nuclear PhysicsUlugbek 702132 Tashkent, Uzbekistan KISLITSIN, S. B. Institute of Nuclear Physics Ibragimov Str.1 050032Almaty, Kazakhstan KNIGHT, C. Idaho National Laboratory P.O. Box 1625, Idaho Falls ID 83415, USA KRYUKOV, F. N. Research Institute of Atomic Reactors 433510, Dimitrovgrad-10, Ulyanovsk Region, Russia LAMBERT, J. D. B. Argonne National Laboratory 9700S, Cass Ave, Argonne Chicago, IL 60439, USA LAMBERT, R. Electric Power Research Institute 3420 Hillview Ave, Palo Alto CA 94304, USA LAURENT, F. Commissariat à l’Énergie Atomique Nuclear Energy Division Marcoule BP 17171 F-30207 Bagnols-sur-Ceze CEDEX France LUKASHENKO, S. N. Institute of Nuclear Physics Ibragimov Str.1 050032 Almaty, Kazakhstan
353
MAKLAKOV, V. V. Russian State Scientific Center Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia MAKSIMKIN, O. P. Institute of Nuclear Physics Ibragimov Str.1 050032 Almaty, Kazakhstan MICHELBACHER, J. A. Idaho National Laboratory P.O. Box 1625, Idaho Falls ID 83415, USA MOULIN, N. Commissariat à l’Énergie Atomique Nuclear Energy Division Marcoule BP 17171 F-30207 Bagnols-sur-Ceze CEDEX France MUKENEVA, S. A. National Nuclear Center of the Republic of Kazakhstan 071100 Kurchatov, Kazakhstan NEKLYUDOV, I. M. National Science Center Kharkov Institute of Physics and Technology of the National Academy of Sciences of Ukraine 1, Akademicheskaya Str. 61108 Kharkov, Ukraine NOVOSELOV, A. E. Russian State Scientific Center Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia
354
LIST OF AUTHORS
POLYAKOV, V. I. Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia POVSTYANKO, A. V. Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia PRIVALOV, U. Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia PUGACHEV, G. P. Mangyshlak Atomic Energy Complex—Kazatomprom 466200 Aktau, Kazakhstan RAMANATHAN, L. V. Materials Science and Technology Center Instituto de Pesquisas Energéticas e Nucleares São Paulo, Brazil ROMANENKO, O. G. Nuclear Technology Safety Center Liza Chaikina 4 Almaty 050020, Kazakhstan ROZHIKOV, V. V. ОАО Novosibirsk Chemical Concentrate Plant 94, Khmelnizkiy Str. Novosibirsk 30110, Russia
SALIKHBAEV, V. S. Institute of Nuclear Physics Ulugbek, 702132 Tashkent, Uzbekistan SAYENKO, S. Yu. National Science Center Kharkov Institute of Physics and Technology of the National Academy of Sciences of Ukraine 1, Akademicheskaya Str. 61108 Kharkov, Ukraine SHAMARDIN, V. K. Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia SHARIPOV, M. Kazakhstan Atomic Energy Committee Liza Chaikina 4 Almaty, Kazakhstan SHIROBOKOV, V. P. Mangyshlak Atomic Energy Complex—Kazatomprom 466200 Aktau, Kazakhstan SHTYNDA, V. Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia SILVY, J.-P. Commissariat à l’Énergie Atomique Nuclear Energy Division Marcoule BP 17171 F-30207 Bagnols-sur-Ceze CEDEX France
LIST OF AUTHORS
SINDELAR, R. L. Savannah River National Laboratory Aiken, SC 29808, USA
TUR, E. S. National Nuclear Center of the Republic of Kazakhstan 071100 Kurchatov, Kazakhstan
SOARES, A. J. Division of Nuclear Fuel Cycle and Waste Technology International Atomic Energy Agency Wagramer Strasse 5, P.O. Box 100 A-1400 Vienna, Austria
TURKEBAEV, T. E. Institute of Nuclear Physics Ibragimov Str.1 050032 Almaty, Kazakhstan
SOBOLEV, M. Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia TAZHIBAEVA, I. L. Nuclear Technology Safety Center Liza Chaikina 4 Almaty 050020, Kazakhstan THOMAS, J. E. Savannah River National Laboratory Aiken, SC 29808, USA TKACHEV, A. A. ОАО Novosibirsk Chemical Concentrate Plant 94 Khmelnizkiy Str. Novosibirsk 30110, Russia TSYNGAYEV, V. M. National Nuclear Center of the Republic of Kazakhstan 071100 Kurchatov, Kazakhstan
TZYKANOV, V. A. Research Institute of Atomic Reactors 433510 Dimitrovgrad, Russia VINSON, D. W. Savannah River National Laboratory Aiken, SC 29808, USA VORMELKER, P. R. Savannah River National Laboratory Aiken, SC 29808, USA VOYEVODIN, V. N. National Science Center Kharkov Institute of Physics and Technology of the National Academy of Sciences of Ukraine 1, Akademicheskaya Str. 61108 Kharkov, Ukraine WELLS, D. NUKEM Limited Kelburn Court, Daten Park Risley, Warrington Cheshire WA3 6TW, UK
355
356
YAKOVLEV, I. L. Mangyshlak Atomic Energy Complex—Kazatomprom 466200 Aktau, Kazakhstan YULDASHEV, B. S. Institute of Nuclear Physics Ulugbek, 702132 Tashkent, Uzbekistan
LIST OF AUTHORS
ZHANTIKIN, T. Kazakhstan Atomic Energy Committee Liza Chaikina 4 Almaty, Kazakhstan ZHEMKOV, I. Yu. Russian State Scientific Center Research Institute of Atomic Reactors 433510, Dimitrovgrad-10 Ulyanovsk Region, Russia
9 14 7
10
5
Chwaszczewski (IAE) explains the dry storage design Chechetkina (RIAR) shows microstrucures of SNF from to be used at Swierck, Poland the reasearch reactor MIR
Johnson (PNL) describes occurrence of cracking in aluminum alloys in wet storage
Kadyrzhanov (INP) and Voyevodin (KIPT) discuss reasearch reactor operations
Maksimkin (INP) discusses corrosion of reactor materials; Chairman Johnson (PNL) takes notes
Garner (PNL) explains notch toughness; Chairman Silvy (CEA) looks for other questioners
Second day of workshop in full swing
Group photo of workshop participants* 1: 2: 3: 4:
F. N. Kryukov, Russia S. A. Baytelesov, Uzbekistan Z. J. Chechetkina, Russia S. Chwaszczewski, Poland
5: 6: 7: 8:
S. B. Kislitsin, Kazakhstan V. N. Voyevodin, Ukraine J. D. B. Lambert, USA R. B. Sindelar, USA
9: T. J. Hill, USA 10: L. V. Ramanathan, Brazil. 11: J. E. Thomas, USA 12: A. J. Soares, IAEA
13: 14: 15: 16:
N. J. Iyer, USA E. D. Fedorovich, Russia J.-P. Silvy, France A. B. Johnson, USA
*Missing: Garner (USA) ; Maksimkin, Kadyrzhanov, Sharipov, Tur and Yakovlev (Kazakhstan); Tkachev (Russia).