Accelerator-driven Systems: Safety and Kinetics Marcus Eriksson Doctoral Thesis Department of Nuclear and Reactor Physics Royal Institute of Technology Stockholm 2005 Akademisk avhandling som med tillstånd av Kungliga Tekniska Högskolan framlägges till offentlig granskning för avläggande av teknologie doktorsexamen fredagen den 18 mars 2004 kl. 10.00 i sal FA32, Albanova universitetscentrum, Roslagstullsbacken 21, Stockholm. ISBN 91-7283-988-0 TRITA-FYS 2005:13 ISSN 0280-316X ISRN KTH/FYS/--05:13—SE Copyright Marcus Eriksson Tryckeri: Universitetsservice US-AB, Stockholm 2005 2 Abstract The accelerator-driven system (ADS) is recognized as a promising system for the purpose of nuclear waste transmutation and minimization of spent fuel radiotoxicity. The primary cause for this derives from its accelerator-driven, sub-critical operating state, which introduces beneficial safety-related features allowing for application of cores employing fuel systems containing pure transuranics or minor actinides, thereby offering increased incineration rate of waste products and minimal deployment of advanced (and expensive) partitioning and transmutation technologies. The main theme of the thesis is safety and kinetics performance of accelerator-driven nuclear reactors. The studies are confined to the examination of ADS design proposals employing fast neutron spectrum, uranium-free lattice fuels, and liquid- metal cooling, with emphasis on lead-bismuth coolant. The thesis consists of computational studies under normal operation and hypothetical accidents, and of evaluation and identification of safety design features. By itself, subcritical operation provides a distinct safety advantage over critical reactor operation, distinguished by high operational stability and additional margins for positive reactivity insertion. For a uranium-free minor actinide based fuel important safety parameters deteriorate. Specific analyses suggest that operation of such cores in a critical state would be very difficult. The studies of unprotected transients indicate that lead-bismuth cooled accelerator-driven reactors can be effective in addressing the low effective delayed neutron fraction and the high coolant void reactivity that comes with the minor actinide fuel, but some supportive prompt negative feedback mechanism might be considered necessary to compensate for a weak Doppler effect in case of a prompt critical transient. Although lead- bismuth features a high boiling point, the work underlines the importance of maintaining a low coolant void reactivity value. The transient design studies identified a molybdenum-based Ceramic-Metal (CerMet) fuel with favourable inherent safety features. A higher lattice pitch is suggested to avoid mechanical failure during unprotected loss-of-flow. Detailed coupled neutron kinetics and thermal hydraulic analyses demonstrated that the point kinetics approximation is capable of providing highly accurate transient calculations of subcritical systems. The results suggest better precision at lower keff levels, which is an effect of the reduced sensitivity to system reactivity perturbations in a subcritical state resulting in small spatial distortions. In the course of a beam reliability study, the accelerator was identified as responsible for frequent beam interruptions. It is clear that extensive improvement in the mean-time between beam failures is required. 3 List of Papers The thesis is based on Papers I through VI. These papers are appended in the thesis and are referred to in the text by their Roman numerals. I. M. Eriksson, J. Wallenius, M. Jolkkonen, and J. E. Cahalan Inherent Safety of Fuels for Accelerator-driven Systems Accepted Nuclear Technology (January 2005). II. J. Wallenius and M. Eriksson Neutronics of Minor Actinide Burning Accelerator Driven Systems with Ceramic Fuel Accepted Nuclear Technology (January 2005). III. M. Eriksson, J. E. Cahalan, and W. S. Yang On the Performance of Point Kinetics for the Analysis Accelerator-driven Systems Nuclear Science and Engineering, 149, 298-311 (2005) IV. M. Eriksson, J. Wallenius, J. E. Cahalan, K. Tucek, and W. Gudowski Safety Analysis of Na and Pb-Bi Coolants in Response to Beam Instabilities Proc. 3rd International Workshop on Utilisation and Reliability of High Power Proton Accelerators, Santa Fe, May 12-16, NEA/OECD (2002). V. M. Eriksson and J. E. Cahalan Inherent Shutdown Capabilities in Accelerator-driven Systems Annals of Nuclear Energy, vol. 29/14 pp 1689-1706, (2002). VI. M. Eriksson and C. Piaszczyk Reliability Assessment of the LANSCE Accelerator System Proc. Workshop on Utilisation and Reliability of High Power Proton Accelerators, Mito, Japan, 13-15 October (1998) Author’s contribution With the exception for paper II, the author is the principal writer and bears the main responsibility for the papers included in the thesis. The author took active part in the discussions and preparation of paper II, but his contribution is primarily related to the subject reported in Paper I. Paper III consists mainly of computational studies and data interpretation, which were performed by the author, with important comments and suggestions by the co-authors. The author performed the main parts of Papers IV and V including the transient and thermal hydraulics calculations, data analysis and the writing of the manuscripts, supported with neutron transport data for Paper IV. Paper VI is the result of Diploma work performed by the author at the department. 4 Acknowledgements I owe thanks to many people who supported me in various different ways during the completion of this thesis. I wish to express my appreciation to friends and colleagues (past and present) of the reactor physics department (Royal Institute of Technology) for, among many other things, their helpful interactions and advice, and for providing a rich social life at the department as well as many memorable events outside work, including stimulating discussions on various topics covered, conference trips, unforgettable “fikas”, and so on. My sincere appreciation goes to W. Gudowski, who originally stimulated my interest in reactor physics and accepted me as graduate student, and to J. Wallenius, from whom I received much guidance and inspiration over the years and also for his efforts in reading the manuscript. It is also a pleasure to acknowledge the continued guidance of J. Cahalan of Argonne National Laboratory, for devoting so much time to me and for his untiring assistance on the topics of reactor safety analysis, whose thinking on the subject in many ways influenced my own. Also, thanks are due to numerous people working at Argonne National Laboratory at the former Reactor Analysis Division who provided considerable assistance and hospitality during my stay as visiting graduate student, in the year 2000-2001, and for their courtesy in allowing me to use their computing facilities. Sincere thanks are due to Svensk Kärnbränslehantering AB for supporting my research in transmutation systems for more than half a decade. I gratefully acknowledge The Swedish Centre for Nuclear Technology, for generously providing the extra resources to carry out part of my studies in the U.S., which I had the fortune to experience. Thanks are also due to the students and staff of the nuclear physics department not mentioned thus far for providing an enjoyable, friendly, and inspiring working environment. Further, I wish to express my deep appreciation and love to my wife, Ellinor, for her continued support, encouragement, patience, and extraordinary understanding for my countless late returns and absent weekends throughout the many years of research, and for providing tasteful lunches (although I must confess I lost 9% of my initial body weight during the writing of this text!) and for her tending of the family during the trying period of writing. Without her support this task would have been impossible. I also extend my affectionate to my two daughters, Elvira and Lovisa, who truly provided the greatest inspiration of all, for their cheerful smile and loving company (understandably with varying degree of patience). Last but not least, special thanks must be given to my parents, Thomas and Elsy, who always followed my work with genuine interest, no matter what, and for their early stimulation on matters that finally brought me into academia. Marcus Eriksson Stockholm 2005 5 Content Chapter 1: Nuclear Waste 1 Introduction 1 Classification of radioactive wastes 1 Spent nuclear fuel 3 Options for spent fuel management 11 Chapter 2: Partitioning & Transmutation 15 Partitioning 15 Transmutation 20 Chapter 3: Transmutation Strategies 24 Introduction 24 Transmutation of plutonium 25 Transmutation of minor actinides 27 Transmutation of fission products 31 Chapter 4: Accelerator-driven Systems 33 General principles 33 Safety features of uranium-free cores 36 Motivation for subcritical operation 42 Responsiveness of the ADS 44 Accident performance 45 Safety performance in super-prompt critical transients 52 Computational performance of the point kinetics method 55 Accelerator reliability 60 Conclusions 62 Appendix A: Reactor Kinetics Equations 64 Appendix B: Short Summary of the SAS4A Code 67 Bibliography 68 Papers 6 “A grad student in procrastination tends to stay in procrastination unless an external force is applied to it” Also known as Newton’s first law of graduation or the “Law of Inertia” originally discovered experimentally by Galileo when he threatened to cut his grad student’s funding! (From http://www.phdcomics.com,