Accident-Tolerant Materials for Light Water Reactor Fuels 0128175036, 9780128175033

Table of contents :
Cover
Accident-Tolerant Materials for Light Water Reactor Fuels
Copyright
Contents
Preface
List of abbreviations and acronyms
1 Nuclear power is clean and safe
Overview
Introduction
Benefits of nuclear energy
The first steps of commercial nuclear power
2 Current materials in light water reactors. Why do we need a materials renewal?
Overview
The light water nuclear power reactor
Materials for light water reactors
Boiling water reactors
Pressurized water reactors
Reactor vessel for boiling water and pressurized water reactors
Fuel assemblies for boiling water and pressurized water reactors
Light water reactor fuels and the excellent performance of urania
How zirconium alloys became the material of choice for fuel cladding
In praise of zirconium alloys
Waterside corrosion of zirconium alloys
Nodular corrosion
Hydrogen pickup by zirconium alloys
Iodine stress corrosion cracking of zirconium alloys
Shadow corrosion of zirconium alloys
Crud deposition on zirconium alloys
Irradiation damage of zirconium alloys
3 Worldwide development of accident tolerant fuels, areas of study, claddings, and fuels
Overview
Accident tolerant fuels—from crisis to opportunity
The events at the Fukushima nuclear power stations of March 2011
International effort to develop safer materials for nuclear power plants
Timeline for the accident tolerant fuels development
Assessment on current accident tolerant fuels maturity concepts
The accident tolerant fuels program in the United States
Industrial civilian nuclear power participation in the accident tolerant fuels efforts in the United States
Nuclear Energy Institute
Electric Power Research Institute
4 Accident-tolerant fuels cladding concept: coatings for zirconium alloys
Overview
Introduction to the use of zirconium alloys as cladding for nuclear fuels in light water reactors
Why do we consider coatings for accident tolerant fuel zirconium alloys?
Oxidation protection of coatings for zirconium alloys
Family of candidate coatings for zirconium alloys
Ceramic coatings
Chromium coatings for zirconium alloys in the French ATF program
Aluminum-based and iron–chromium–aluminum coatings for zirconium alloys
Silicon-based coatings for zirconium alloys
Fabrication and implementation of zirconium-coated rods
Performance of coated zirconium under reactor normal operation conditions
Performance of coated zirconium under accident conditions
Coated zirconium irradiation behavior
Coated zirconium licensing for reactor use
5 FeCrAl—iron–chromium–aluminum monolithic alloys
Overview
What are FeCrAl alloys?
Metallurgy and microstructure of FeCrAl
Earlier considerations of FeCrAl alloys for nuclear applications
Why are FeCrAl considered for accident-tolerant fuel cladding? Benefits and challenges
Thermal properties of FeCrAl
Mechanical properties of FeCrAl
Oxidation resistance of FeCrAl under LWR’s normal operation conditions
Composition of the oxide films on FeCrAl coupons
Electrochemical behavior of FeCrAl alloys in high-temperature water
Shadow corrosion
Galvanic corrosion
Resistance to crud deposition under normal operation conditions
Resistance to EAC of ferritic alloys under LWR normal operation conditions
Resistance to fretting under normal operation conditions
Resistance of monolithic FeCrAl cladding to thermal shock
Interaction between the urania fuel and the FeCrAl cladding
Oxidation resistance of FeCrAl in high-temperature gas environments
Mechanism of protection at accident condition temperatures
The Roles of metal oxides on the surface of FeCrAl
Normal operation oxidation to accident oxidation scenario and vice versa
Scenario 1: Water-oxidized APMT tubes exposed to superheated steam
Scenario 2: Steam-oxidized APMT tubes exposed to high-temperature water
The versatile oxidation behavior of FeCrAl Alloys
Fabrication and implementation of cladding tubes
Welding of FeCrAl alloys
Mitigation measures to parasitic neutron absorption of FeCrAl
Mitigation measures to increased tritium release into the coolant
Irradiation behavior of FeCrAl
Corrosion behavior of used FeCrAl cladding in cooling pools
Licensing for reactor use
6 Silicon carbide and ceramics metal composite
Overview
Why do we consider silicon carbide composites for accident tolerant fuel? Benefits and challenges
Thermal properties and permeability of SiC/SiC fuel cladding
SiC/SiC fuel cladding, fabrication, and implementation
Environmental behavior of SiC/SiC under normal operation conditions
Environmental behavior of SiC/SiC under accident conditions
Irradiation behavior
Licensing for reactor use
7 Alternative fuels to urania
Overview
Introduction
The urania nuclear fuel
The urania excellent performance
Accident tolerant fuels under consideration
Improved urania fuels by doping
Modified urania performance under normal operation conditions
Modified urania performance under accident conditions
Higher density fuels: uranium silicide
Reactivity of uranium disilicide
Reactivity of U3Si2 with the cladding
Fabrication and implementation of U3Si2 fuels
Higher density fuels: uranium nitride
Reactivity of uranium mononitride fuel
Fabrication paths for uranium mononitride
Behavior of uranium mononitride under irradiation
8 Maturity of the accident-tolerant fuel concepts: the fuel cycle and used fuel disposition
Overview
Assessment on accident-tolerant fuel maturity concepts
NEA assessment on maturity of ATF cladding concepts
Assessment on maturity of ATF fuel concepts
The nuclear fuel cycle
9 Licensing and the increased safety of power reactors’ operation
Overview
Licensing process in the United States
Increased safety of nuclear power plant operation
Evolutionary trend of the nuclear fuel
Safety analysis and source term
10 Looking to the future
Overview
References
Index
Back Cover

Citation preview

Accident-Tolerant Materials for Light Water Reactor Fuels

Accident-Tolerant Materials for Light Water Reactor Fuels

Raul B. Rebak General Electric Research, Schenectady, NY, United States

Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States Copyright © 2020 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress ISBN: 978-0-12-817503-3 For Information on all Elsevier publications visit our website at https://www.elsevier.com/books-and-journals

Publisher: Matthew Deans Acquisition Editor: Christina Gifford Editorial Project Manager: Joshua Mearns Production Project Manager: R. Vijay Bharath Cover Designer: Matthew Limbert Typeset by MPS Limited, Chennai, India

Contents Preface List of abbreviations and acronyms

ix xi

1.

Nuclear power is clean and safe

1

Overview Introduction Benefits of nuclear energy The first steps of commercial nuclear power

1 1 5 8

2.

Current materials in light water reactors. Why do we need a materials renewal? Overview The light water nuclear power reactor Materials for light water reactors Boiling water reactors Pressurized water reactors Reactor vessel for boiling water and pressurized water reactors Fuel assemblies for boiling water and pressurized water reactors Light water reactor fuels and the excellent performance of urania How zirconium alloys became the material of choice for fuel cladding In praise of zirconium alloys Waterside corrosion of zirconium alloys Nodular corrosion Hydrogen pickup by zirconium alloys Iodine stress corrosion cracking of zirconium alloys Shadow corrosion of zirconium alloys Crud deposition on zirconium alloys Irradiation damage of zirconium alloys

3.

15 15 16 17 19 19 20 22 23 24 27 30 34 35 36 37 40 41

Worldwide development of accident tolerant fuels, areas of study, claddings, and fuels

43

Overview Accident tolerant fuels—from crisis to opportunity The events at the Fukushima nuclear power stations of March 2011

43 44 45

v

vi

4.

5.

Contents

International effort to develop safer materials for nuclear power plants Timeline for the accident tolerant fuels development Assessment on current accident tolerant fuels maturity concepts The accident tolerant fuels program in the United States Industrial civilian nuclear power participation in the accident tolerant fuels efforts in the United States Nuclear Energy Institute Electric Power Research Institute

56 57 60

Accident-tolerant fuels cladding concept: coatings for zirconium alloys

63

47 49 51 53

63

Overview Introduction to the use of zirconium alloys as cladding for nuclear fuels in light water reactors Why do we consider coatings for accident tolerant fuel zirconium alloys? Oxidation protection of coatings for zirconium alloys Family of candidate coatings for zirconium alloys Ceramic coatings Chromium coatings for zirconium alloys in the French ATF program Aluminum-based and iron chromium aluminum coatings for zirconium alloys Silicon-based coatings for zirconium alloys Fabrication and implementation of zirconium-coated rods Performance of coated zirconium under reactor normal operation conditions Performance of coated zirconium under accident conditions Coated zirconium irradiation behavior Coated zirconium licensing for reactor use

76 77 79 80

FeCrAl—iron chromium aluminum monolithic alloys

83

Overview What are FeCrAl alloys? Metallurgy and microstructure of FeCrAl Earlier considerations of FeCrAl alloys for nuclear applications Why are FeCrAl considered for accident-tolerant fuel cladding? Benefits and challenges Thermal properties of FeCrAl Mechanical properties of FeCrAl Oxidation resistance of FeCrAl under LWR’s normal operation conditions Composition of the oxide films on FeCrAl coupons Electrochemical behavior of FeCrAl alloys in high-temperature water

84 84 85 87

64 65 67 68 69 71 72 74 75

88 92 93 96 102 106

Contents

Shadow corrosion Galvanic corrosion Resistance to crud deposition under normal operation conditions Resistance to EAC of ferritic alloys under LWR normal operation conditions Resistance to fretting under normal operation conditions Resistance of monolithic FeCrAl cladding to thermal shock Interaction between the urania fuel and the FeCrAl cladding Oxidation resistance of FeCrAl in high-temperature gas environments Mechanism of protection at accident condition temperatures The Roles of metal oxides on the surface of FeCrAl Normal operation oxidation to accident oxidation scenario and vice versa Scenario 1: Water-oxidized APMT tubes exposed to superheated steam Scenario 2: Steam-oxidized APMT tubes exposed to high-temperature water The versatile oxidation behavior of FeCrAl Alloys Fabrication and implementation of cladding tubes Welding of FeCrAl alloys Mitigation measures to parasitic neutron absorption of FeCrAl Mitigation measures to increased tritium release into the coolant Irradiation behavior of FeCrAl Corrosion behavior of used FeCrAl cladding in cooling pools Licensing for reactor use

6.

7.

vii 108 111 112 113 115 116 118 119 121 124 124 125 126 127 129 131 134 134 137 139 140

Silicon carbide and ceramics metal composite

143

Overview Why do we consider silicon carbide composites for accident tolerant fuel? Benefits and challenges Thermal properties and permeability of SiC/SiC fuel cladding SiC/SiC fuel cladding, fabrication, and implementation Environmental behavior of SiC/SiC under normal operation conditions Environmental behavior of SiC/SiC under accident conditions Irradiation behavior Licensing for reactor use

143

149 153 155 155

Alternative fuels to urania

157

Overview Introduction The urania nuclear fuel The urania excellent performance Accident tolerant fuels under consideration Improved urania fuels by doping

157 158 159 159 160 160

144 148 148

viii

8.

9.

Contents

Modified urania performance under normal operation conditions Modified urania performance under accident conditions Higher density fuels: uranium silicide Reactivity of uranium disilicide Reactivity of U3Si2 with the cladding Fabrication and implementation of U3Si2 fuels Higher density fuels: uranium nitride Reactivity of uranium mononitride fuel Fabrication paths for uranium mononitride Behavior of uranium mononitride under irradiation

163 164 164 165 167 167 168 169 169 170

Maturity of the accident-tolerant fuel concepts: the fuel cycle and used fuel disposition

171

Overview Assessment on accident-tolerant fuel maturity concepts NEA assessment on maturity of ATF cladding concepts Assessment on maturity of ATF fuel concepts The nuclear fuel cycle

171 172 174 180 185

Licensing and the increased safety of power reactors’ operation

187

Overview Licensing process in the United States Increased safety of nuclear power plant operation Evolutionary trend of the nuclear fuel Safety analysis and source term

187 188 188 188 190

10. Looking to the future Overview References Index

197 197 199 213

Preface The aim of this book is to provide a snapshot on the state-of-the-art development of a family of materials called accident tolerant fuels (ATF) for commercial light water reactors. These materials include advanced cladding components and fuel forms. This book is not meant to be a comprehensive or exhaustive collection of data or information on materials for civilian nuclear power generation since its status is continuously changing, practically on a weekly basis. A description is provided on how the enthusiasm of the current development of newer innovative materials started after six decades of stagnation or complacency in the nuclear industry. The book is intended for graduate students or new professionals who are getting started in the field; so they can put things in perspective and understand how we reached 2019, the year of ATF. The concept of ATF was born after the 2011 unfortunate events at the Fukushima Daiichi nuclear power stations. Initially, there was a state of great concern that the destructive tsunami wave not only washed away the diesel generators at the affected plants, but it also swept away the resolve of using nuclear fission to generate civilian electricity, especially in the western world. However, only a few months after the disaster, the nuclear materials’ international community was able to recover from the negative reporting regarding the explosions shown live on television worldwide and offer materials solutions to ensure that the events of Fukushima would not repeat themselves. It was a beautiful thing to witness how governmental funding agencies, nuclear fuel vendors, regulatory agencies, trade organizations, university professors, reactor owner utilities, research institutes scientists, and plant operators came seamlessly together to offer solutions for the continuing use of light water reactors. The effort to look for solutions came to life at both national and international levels, and there was (is) a great continuous cooperation between all the involved parts. The developmental programs are evolving so quickly that all the initial schedules are being beaten. There is also such an agility in the execution of the programs that many initial ideas not deemed viable of fast implementation are being shed quickly and researchers are joining forces on the growth of the remaining more robust concepts. Therefore, the focus of this book is to describe and assess the technology readiness level of only the few strongest ideas rather than to meander trying to cover all early proposed ideas that may be far away from fruition.

ix

x

Preface

Initial chapters succinctly describe the history and evolution of nuclear energy as a source of civilian electricity, praising the materials that were able to make possible the existence of nuclear power for almost seven decades. The main goal of the more technical chapters was to review the literature data using the same set of guidelines, metrics, or parameters to evaluate the maturity of each concept and to assess the progress or important gaps that exist in each of the engineering answers to the newer materials challenge. This book does not cover efforts and results from the field of modeling and simulation. In each one of the proposed solutions there is a solid foundation of science justifying the viability of each concept, but for a utility to be able to implement the proposed accident tolerant fuel, it also needs to be simple, practical, economical, and safe. It is an exciting time to be working in the field of advanced technology materials for making light water reactors safer to operate. The use of electricity originating from nuclear sources represents a crucial contribution to a clean environment and to the reduction in the atmospheric release of climate-changing greenhouse gases. Finally, I would like to acknowledge the patience and understanding of my husband William W. Wickline, who makes me laugh and who prefers to ignore the travails of the real world by watching vintage Hollywood movies. Raul B. Rebak Schenectady, New York, May 15, 2019

List of abbreviations and acronyms ANL AOOs ATF ATR BDBAs BNL BWR CANDU CA-PVD CEA CHF CILC CMC CNL CRIEPI CTE CVD CVI DBA DNB DOE EATF EBSD ECCS ECR EDF EDS EGATFL EPR EPRI ETF HFIR HWC IAEA IMAGO INL

Argonne National Laboratory (United States) Anticipated operational occurrences Accident-tolerant fuel or advanced technology fuel Advanced test reactor Beyond-design-basis accidents Brookhaven National Laboratory (United States) Boiling water reactor Canada Deuterium Uranium Cathode arc physical vapor deposition French Alternative Energies and Atomic Energy Commission Critical heat flux Crud-induced localized corrosion Ceramic matrix composite (SiC/SiC) Canadian Nuclear Laboratories Central Research Institute of Electric Power Industry (Japan) Coefficient of thermal expansion Chemical vapor deposition Chemical vapor infiltration Design-basis accident Departure from nucleate boiling Department of Energy (United States) Enhanced accident-tolerant fuel Electron back scatter diffraction Emergency core cooling system Equivalent-cladding reacted Electricite´ de France Energy dispersive spectroscopy Expert Group on Accident-Tolerant Fuels for Light Water Reactors (NEA) European pressurized reactor Electric Power Research Institute Elongation to failure High-flux isotope reactor Hydrogen water chemistry International Atomic Energy Agency Irradiation of Materials for Accident-tolerant fuels in the Go¨sgen reactor Idaho National Laboratory (United States)

xi

xii

List of abbreviations and acronyms

IRSN JAEA KAERI KIT KTH LBM LFA LFR LHGR LOCA LTA LTR LWR METI NDE NEA NEAMS NFD NPP NPS NRC NWC OD ODS ORNL PCI PCMI PIE PNNL PRW PSI PVD PWR RBMK RCS RE RIA RPV RT SBO SCC SEM TD TIG TREAT TRISO TRL

Institut de radioprotection et de suˆrete´ nucle´aire (France) Japan Atomic Energy Agency Korea Atomic Energy Research Institute Karlsruhe Institute of Technology (Germany) Royal Institute of Technology (Sweden) Laser beam welding Lead fuel assembly Lead fuel rods Linear heat generation rate Loss-of-coolant accident Lead test assemblies Lead test rod Light water reactor Ministry of Economy, Trade and Industry (Japan) Nondestructive evaluation Nuclear Energy Agency Nuclear energy advanced modeling and simulation Nippon Nuclear Fuel Development Nuclear power plant Nuclear power station Nuclear Regulatory Commission (United States) Normal water chemistry Outer diameter Oxide dispersion strengthened Oak Ridge National Laboratory (United States) Pellet clad interaction Pellet clad mechanical interaction Postirradiation examination Pacific Northwest National Laboratory (United States) Pressure resistance welding Paul Scherrer Institute (Switzerland) Physical vapor deposition Pressurized water reactor High Power Channel-Type Reactor (Russia) Reactor coolant system Rare-earth elements Reactivity-initiated accident Reactor pressure vessel Room temperature Station blackout Stress corrosion cracking Scanning electron microscopy Theoretical density Tungsten inert gas Transient reactor test facility Tri-structural Isotropic Technology readiness level

List of abbreviations and acronyms UTS VHT VVER XRD YS

Ultimate tensile strength Very high temperature Water water energetic reactor (Russia) X-ray diffraction Yield stress

xiii

Chapter 1

Nuclear power is clean and safe Chapter Outline Overview Introduction Benefits of nuclear energy

1 1 5

The first steps of commercial nuclear power

8

Overview The electricity in the interconnected commercial grid comes from many sources including fossil fuels, renewable energies (e.g., solar and wind), and nuclear energy. The energy harvested from nuclear fission sources is clean, it does not release greenhouse gases to the environment, and it does not contribute to climate change. Moreover, nuclear sourced electricity is the safest type for civilian use. In the United States and around the globe, the number of human deaths per kilowatt hour of generated electricity from nuclear sources is five orders of magnitude lower than for coal and other hydrocarbon sources. Nuclear electricity is even safer than solar or wind electricity. Nuclear power was first used for military purposes, especially for the propulsion of submarines. However, in the early 1950s, the United Nations program of Atoms for Peace, both the United States and the Soviet Union developed their first civilian nuclear power plants which were connected to their respective grids in the late 1950s. At around this time, the trusted fuel rod pair of zirconium alloy cladding containing pellets of urania fuel was born and used for many decades since. Currently, there are approximately 30 nations which use 451 civilian reactors to generate electricity. The largest expansion of nuclear reactors was in the decades of 1970s and 1980s. In 2019 there were 54 new reactors under construction, mainly in Asia.

Introduction Clean energy is an energy that does not pollute the environment nor increases the amount of greenhouse gases that may contribute to climate change. Nuclear energy is clean energy since it does not change the carbon footprint in the planet. Nuclear energy is also the safest form of all energies Accident-Tolerant Materials for Light Water Reactor Fuels. DOI: https://doi.org/10.1016/B978-0-12-817503-3.00001-8 © 2020 Elsevier Inc. All rights reserved.

1

2

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 1.1 World energy consumption.

used to generate electricity. Nuclear energy is even safer than the renewable solar or wind energies. Electricity generated using fossil fuels may increase the carbon footprint by releasing methane gas to the atmosphere in the upstream (during exploration and production) and then releasing carbon dioxide to the atmosphere in the downstream when the fossil fuels or hydrocarbons are burnt to extract the heat. Steam turbines generate most of the electricity used in the United States and in the planet. Power plants that are fed by uranium or coal are similar in the sense that both use steam turbines to move the generators to produce electricity. Nuclear and coal plants differ in the way how they make their steam needed for the turbines. One type of plant splits the atom through a fission reaction which generates vast amounts of energy, and the other type of power plant burns hydrocarbons through a combustion reaction. Overall, the operation of both types of plants greatly differs in the fact that one plant produces steam in the safest and cleanest way possible while the other does not. Worldwide, the generation of electric power has several sources of energy that can be grouped as: (1) fossil fuels (coal, petroleum, and natural gas), (2) nuclear, and (3) renewable (wind, solar, hydroelectric, geothermal, biomass, etc.) sources. Fig. 1.1 shows that the world energy consumption in the next two decades will be still dominated (B80%) by the burning of fossil fuels (liquid, gas, and coal). Nuclear energy represents only 6% of the energy consumed worldwide. In the United States, 30% of the consumed natural gas is used to generate about 20% of the electrical power produced in the country. The other 70% of the consumed natural gas is used for purposes other than to generate electricity, for example, for heating. On the contrary, 100% of the nuclear energy generated is used to produce electricity. Nuclear or atomic energy is released when atoms are split in a reactor. The release of nuclear fission energy is transferred to water, which makes

Nuclear power is clean and safe Chapter | 1

3

FIGURE 1.2 Operational power reactors (IAEA, 2019).

steam that is used to spin the turbines which rotates the shaft or an electrical generator. Electricity generated using nuclear sources represents about 20% of all the electrical power consumed in the United States. In other countries, such as France, the electricity generated from nuclear fission represents over 70% of the total electricity consumed in the nation. There are currently around 30 countries in the globe that use nuclear sources to generate civilian electricity (IAEA, 2019). Fig. 1.2 shows that the worldwide number of operational reactors slightly increased between 2013 and 2019 from 430 to 446 (IAEA, 2019). The largest increase in newer power reactors was in Asia and Eastern Europe (China, Korea, India, Pakistan, and Russia). The notable increase was in China where the number of reactors more than doubled from 18 in 2013 to 46 in 2019. In the western world, the number of operational reactors mostly decreased, except for Argentina where it increased from two to three. Fig. 1.3 shows the nuclear power percentage of the total electricity consumed in the 30 current countries with nuclear power for 2013 and 2019 (IAEA, 2019). The countries with the largest share of nuclear power are France, Slovakia, Ukraine, Belgium, and Hungary. France is also exporting electricity to neighboring countries since their cost of power generation is low compared to other nonnuclear sources in those neighboring countries. There was little change between the share of nuclear power between 2013 and 2019 for the 30 countries listed. In the United States, for example, the share of nuclear power stayed the same at 20% between these 2 years, considering that the country had four fewer reactors in 2019 than in 2013. One

4

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 1.3 Nuclear power share (IAEA, 2019).

of the reasons is that the power reactors became more efficient and another reason is because the total consumption of electricity decreased due the environmental green measures. Fig. 1.2 shows that Japan had approximately 40 operational reactors between 2013 and 2019 but the share of nuclear power was only in the order of 3%. The reason for this is that most of the Japanese reactors were not producing electricity in this period because of their temporary shutdown following the Fukushima accident of March 2011. Before the accident at the Fukushima site, the nuclear power share in Japan was approximately 30% (IAEA, 2019). Fig. 1.4 shows the age distribution of the 451 power reactors operating worldwide in May 2019 (IAEA, 2019). Sixty-eight percent of the reactors are 30 or more years old and only 13% of the reactors have an age of 10 years or less. The worldwide fleet of power reactors is aging, but it also appears that there is a slight increase in the number of reactors with 5 years or less in age. It is important to prevent the shutdown of older reactors by license renewal. One of the measures to prevent the shutdown of older reactors is to retrofit them using accident tolerant fuels, which will make the reactors safer to operate. In the United States, there are currently 98 operating commercial light water power reactors, 34 boiling water reactors (BWRs), and 64 pressurized water reactors (PWRs). Approximately 90% of the US power reactors are at least 30 years old and 45% of the reactors are at least 40 years old. Only one

Nuclear power is clean and safe Chapter | 1

5

FIGURE 1.4 Age power reactors.

new reactor was connected to the grid in the United States, in the last 25 years and only two new reactors are currently under construction (2019). Fig. 1.5 shows that 19 countries have currently power reactors under construction as compared to 15 countries in 2013. Four countries (Bangladesh, Belarus, Turkey, and United Arab Emirates) that did not have nuclear power before are currently planning to enter the international community of generating electricity from nuclear energy (IAEA, 2019). China, India, and Russia have currently the largest amount of nuclear power plants under construction, even though the construction rate in China seemed to have slowed down in the last quinquennium. The United Arab Emirates has four reactors under construction which is more than double than it was in 2013. It is likely that Egypt will follow soon with a power reactor of their own in the Arab community. Egypt has had a nuclear program since 1954. There are also current advocacy groups to bring nuclear power to Australia, considering that Australia has more than 30% of the world deposits in uranium, similar to the deposits of Canada and Kazakhstan.

Benefits of nuclear energy Nuclear power plants work all the time, and they do not need specific climatic conditions to operate. Nuclear energy protects the environment since it does not release nitrogen oxide, sulfur oxide, or carbon dioxide. The

6

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 1.5 Reactors under construction.

FIGURE 1.6 Casualties per kilowatt hour (kWh).

production of nuclear electricity does not generate particulates or spread mercury and it does not contribute to the release of methane to the atmosphere. Nuclear energy is also the safest type of energy since it historically

Nuclear power is clean and safe Chapter | 1

7

produced the lowest number of casualties or fatalities per kilowatt hour of energy generated (Fig. 1.6). The largest number of human casualties around the world because of generating electricity is caused by air pollution, more specifically by the presence of fine particulate matter (,2.5 µm diameter) in the atmosphere. The fine suspended particulate matter, which originates mainly through the burning of coal, may cause more than 2 million deaths a year mainly in the form of cardiopulmonary diseases and lung cancer (Silva et al., 2013). Most of these annual premature deaths are in Asia (India, Southeast and East Asia) (Silva et al., 2013). Fig. 1.6 shows the mortality rate in the United States per kilowatt hour of electricity generated according to five sources of energy. The human fatalities caused by the burning of coal are five orders of magnitude higher than the fatalities caused by nuclear power. Major accidents causing immediate deaths and related to the production of electricity include the collapse of dams (hydroelectricity), coal mining accidents, and the rupture of pipelines followed by explosions. For example, the rupture of the Shimantan Dam in 1975 produced more the 170,000 human fatalities. Coal mining accidents produced almost 5000 immediate deaths in 2006 in China alone. Moreover, in China, more than 70,000 miners a year suffer from the black lung disease (pneumoconiosis). In the United States, the fatalities in coal mines were in the order of 1500 per year until the 1970s, which eventually declined to less than 100 per year between 1990 and 2012. In contrast, in the United States, there were zero fatalities related to the production of electricity from nuclear sources. Why does nuclear power still have a mixed support in the public? It is mainly because nuclear energy is complex and nontransparent. The generation of civilian power using nuclear energy has an unprecedented and unparalleled safety record. Its implementation follows strict licensing procedures and regulations and the nuclear power plants are operated and maintained by highly trained professionals. Only three major accidents are related to the production of commercial electricity from nuclear fission sources and are described in Table 1.1. The accident that caused the largest number of casualties was in Ukraine by the Chernobyl Reactor Number 4. Approximately 30 first responders’ firefighters died during or immediately after the Chernobyl accident and it is estimated that the human casualties may have reached several thousands in the three decades since the accident, mainly because of cancer. By contrast, the accident of Three Mile Island had zero immediate and long-term casualties. The Fukushima accident sits somewhere in the middle. The only casualties at the Fukushima site during the tsunami and the immediate events were due to the drowning of two technicians who were in the basement when the second wave hit the turbine buildings. Related premature casualties because of the Fukushima accident emanated from the evacuation of elderly and ailing residents from near the power

8

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 1.1 Commercial nuclear accidents. Commercial nuclear accident

What happened?

Direct human casualties (delayed, estimated)

Three Mile Island, 1979

Partial meltdown of the PWR Unit 2 reactor due to a loss of coolant in the primary circuit caused by a valve stuck open. Cause, human error, & lack of training.

0 (0)

Chernobyl, 1986

Steam explosion followed by open air graphite fire of the light water graphite moderated RBMK Unit 4 reactor during a test to simulate a station black out. The fire burned for 9 days releasing fission products to the atmosphere. Cause, operator error, negligence, & lack of training or knowledge.

30 (134 up to 1996, maybe a total of 4000, leukemia and cancer)

Fukushima Daiichi, 2011

Hydrogen gas explosions because of loss of coolant due to station black out caused by a tsunami wave. Release of radioactive products to the atmosphere from BWR Units 1, 2, and 3. Cause, failure of operator to meet basic safety requirements.

2 drowned (1600 elderly related to evacuation, not to radiation)

BWR, Boiling water reactor; PWR, pressurized water reactor; RBMK comes from “Reaktor Bolshoy Moshchnosti Kanalnyy” which means “High Power Channel-type Reactor”.

station. Casualties due to irradiation exposure in the Fukushima Prefecture may not be proven yet.

The first steps of commercial nuclear power Civilian nuclear power was developed almost simultaneously in the USSR and in the United States in the mid to late 1950s. Fig. 1.7 shows a few milestones of nuclear power from 1950, mostly centered around events in the United States. The utilization of nuclear energy to generate electricity originated from the discovery of the nuclear fission (and the release of heat) in December 1938 in Germany. The controlled release of nuclear energy by a self-sustaining chain reaction was realized by Enrico Fermi at the University of Chicago in December 1942. Criticality was reached in the Fermi built pile reactor with natural uranium. A reactor achieves criticality and becomes critical, when each fission event releases enough neutrons to sustain an ongoing

FIGURE 1.7 Nuclear power timeline.

10

Accident-Tolerant Materials for Light Water Reactor Fuels

series of reactions, producing a continuous release of heat. Approximately 99% of the natural uranium is the uranium-238 (U-238) isotope which is nonfissile. The natural uranium also contains approximately 0.7% of the fissile U-235 isotope. After the Fermi sustained chain reaction demonstration, efforts were undertaken at the Clinton Engineer Works at Oak Ridge to enrich the uranium fuel in the U-235 isotope from approximately 0.7% to levels in the order of 3% 5%. The early 1940s efforts and actions to harvest energy from nuclear fission were accelerated as a response to the World War II events in Europe. The use of nuclear fission energy to produce electricity in the United States was mainly the result of the labors of one person, Hyman George Rickover. Rickover was an officer in the US Navy and after spending time during World War II in the Philippines and Japan, he returned to the United States and decided to reinvent his own career by dedicating himself to learn how to use nuclear energy to propel a submarine. Between 1946 and 1947, Rickover spent a year at Oak Ridge to learn as much as he could about nuclear energy and then he returned to Washington, DC, to be the director of reactor development in the newly created (1946) Atomic Energy Commission. Rickover’s main objective was to build a reactor for a submarine but because of political backlash, the US political system decided to demonstrate in the early 1950s that nuclear energy could be used for peaceful purposes, such as generating electricity for the civilian distribution grid. The first pressurized light water cooled power reactor to be connected to the electrical grid and to operate commercially was the atomic power station 1 (APS-1) at Obninsk, USSR, on June 27, 1954 (Simnad, 1981). The Obninsk reactor also served as a research station for several decades until it was decommissioned in 2002. Research and development conducted at the Obninsk power station helped in the design and improvement of the RBMK soviet power reactor, a type of power reactor which is still in use today. After an RBMK type reactor suffered the accident in Chernobyl in Ukraine in April 1986, the existing operating RBMK reactors were modified and retrofitted to make them more resilient to accidents. The RBMK design is simple, it uses light water for coolant, graphite with boron carbide for control rods, and zirconium niobium cladding for 2% enriched uranium 235 in urania fuel. As mentioned earlier, the first western production of electricity by ways of harvesting nuclear fission energy was a political move by the United States to fulfill the goal of peaceful uses of atomic energy, as a follow-up of the December 8, 1953, “Atoms for Peace” speech by the US President Dwight D. Eisenhower delivered at the United Nations General Assembly. A PWR that was under construction in the early 1950s at the Westinghouseoperated Bettis Laboratories (near Pittsburgh) for an aircraft carrier was repurposed to be used for the civilian generation of electricity. The US Atomic Energy Commission partnered with the Westinghouse Co and the

Nuclear power is clean and safe Chapter | 1

11

Duquesne Light Company to take this first civilian power reactor to completion (Lustman, 1981; Van Duysen and Meric de Bellefon, 2017). The location for this first civilian reactor was Shippingport in Beaver County on the Ohio River approximately 40 km south of Pittsburgh. The responsibility of the construction of the reactor was given to Rear Admiral H. G. Rickover, Director of the Atomic Energy Commission Division of Naval Reactors. The construction company for the Shippingport reactor was Westinghouse and the utility that was going to participate in the construction and later distribute the electricity was the Duquesne Light Company. The groundbreaking ceremony for the civilian nuclear power station on Shippingport was on September 6, 1954 (Karoutas et al., 2018). The nuclear power station reached criticality 3 years later, on December 2, 1957 (Fig. 1.7). The first electricity from this plant arrived at the electric grid on December 18, 1957, and the plant generated electricity until it was permanently shut down in 1982. The objective by the US Atomic Energy Commission in the early 1950s was to have a public utility run a nuclear power reactor before the Second United Nations International Conference on the Peaceful Uses of Atomic Energy in Geneva September 1 13, 1958 (United Nations, 1958; Lustman, 1981). As in the case of the Obninsk power plant in the USSR, Shippingport was built to serve a dual purpose, not only as a power generator for the grid, but also as a test facility for the advancement of nuclear power technology. The reactor at Shippingport was a PWR based on previous designs made for the US Navy nuclear submarines such as the Mark 1 prototype built in Idaho and the second comparable reactor built for the Nautilus submarine (Lustman, 1981; Van Duysen and Meric de Bellefon, 2017). The first submarine type PWR called S1W was built on land and it achieved criticality on March 30, 1953. The second PWR was installed in the Nautilus submarine and commissioned on September 30, 1954. Two engineering and material issues had to be resolved at that time: (1) what to use for the fuel and (2) what to use for the cladding of the fuel in the reactors. The corrosion behavior of nuclear candidate materials in the United States in the earlier 1950s was characterized by urgency and secrecy (Wanklyn and Jones, 1962). Remarkably, in the mid to late 1950s, the basic phenomena of the oxidation of Zircaloy-2 in air, water and steam were already rather well tested and recognized. Then, the decision was made that Shippingport was going to use Zircaloy-2 as cladding for the fuel based on the positive results from the previous two built reactors, the land based in Idaho and the Nautilus one, even though Zircaloy was very expensive at that time since the Kroll production process was not optimized yet. Also, at the time of the fabrication of the Shippingport reactor at Bettis, the only known fuel was metallic uranium, sometimes minimally alloyed to control grain size (Lustman, 1981). Compatibility studies of metallic uranium with hot water that may come in contact via cladding defects were not optimistic. When 9% 12% of molybdenum was added to the uranium metal, the researchers achieved the gamma

12

Accident-Tolerant Materials for Light Water Reactor Fuels

or face-centered cubic-phase stabilization of the uranium metal and a decrease of several orders of magnitude in the corrosion rate of metallic uranium in hot water. It was also believed that Mo additions would limit swelling during irradiation (Lustman, 1981). At around the same time in the mid-1950s, Glatter, a Bettis ceramist, was pushing for the consideration of urania (UO2) for the fuel in the three reactors built or under construction in the United States (Lustman, 1981). One of the main issues that many resisted on the implementation of urania as fuel was its poor thermal conductivity. High thermal conductivity is needed to remove the heat quickly from the fuel, via the cladding and into the coolant water. Nevertheless, urania was considered attractive because of its resistance to interaction with water in the case of a cladding breach and because of its resistance to irradiation damage (Ang and Burkhammer, 1960). Eventually in the 1950s, after many fabrication and performance trials, the manufacturing of urania was deemed a reproducible, controllable, and predictable process (Lustman, 1981). Therefore in the late 1950s, the couple or pair of urania fuel and Zircaloy-2 cladding was born and adopted ever since for many decades to come all over the world (Fig. 1.7) (Lustman, 1981). At the same time, the US Navy used Zircaloy cladding for the uranium dioxide (UO2) or urania fuel, but in the early 1960s, the cost of zirconium was high enough that most of the commercial nuclear power stations built in the United States used stainless steels for the cladding of the urania pellets (Simnad, 1981). Eventually as the cost of zirconium started to decrease after the Kroll production process was optimized, the commercial plants shifted from austenitic stainless steels such as type 304 to Zircaloy-2 or Zircaloy-4. The design, construction, and testing of BWRs, without a second loop to produce steam, were explored both in the United States and the USSR on the mid-1950s to early 1960s (Simnad, 1981). In the United States, Argonne built an experimental BWR which operated between 1956 and 1967 demonstrating the feasibility of an integrated BWR plant. General Electric built the Vallecitos prototype BWR which operated between October 1957 and December 1963. The GE Vallecitos reactor started operation 4 months before the Westinghouse Shippingport in Pennsylvania (Fig. 1.7). Even though Vallecitos generated electricity for the Pacific Gas & Electric company grid, its power contribution was small enough that Vallecitos is not generally accepted as the first US civilian power reactor, the record that belongs to Shippingport (Van Duysen and Meric de Bellefon, 2017). The actual first US civilian BWR was the Dresden power station designed by GE, which started operations in 1960 in Illinois (Van Duysen and Meric de Bellefon, 2017). Dresden was the first reactor designed and built by private initiative and without government assistance. Meanwhile in the USSR, a core of a BWR started testing inside the Obninsk APS-1 reactor in 1954 (Simnad, 1981). The first fully commercial PWR in the United States was Yankee-Rowe power station in western Massachusetts, which started operation in January

Nuclear power is clean and safe Chapter | 1

13

FIGURE 1.8 US reactors first connected to grid.

1961 and ended in February 1992. Initially the fuel in the Yankee-Rowe PWR plant was urania pellets and the cladding was type 348 austenitic stainless steel, with some elements having Zircaloy-2 cladding (Simnad, 1981). Figs. 1.7 and 1.8 show that the highest number of power reactors connected to the US grid was 12 in 1974 (IAEA, 2019). Fig. 1.8 shows for a span of 60 years the number of reactors connected annually to the US electrical grid. Most of the reactors were connected in the first 30 years (1960 90). For the last 30 years (1990 2019), only four new reactors were connected to the grid. The largest number of reactors was connected in the mid-1970s and later again in the mid-1980s. The Vallecitos reactor in Sunol (California) had the US Atomic Energy Commission Power Reactor License N 1. Vallecitos was the first privately owned commercial prototype BWR owned by General Electric. This reactor initially used austenitic type 304 SS cladding for slightly enriched urania fuel. Later some of the cladding was replaced using Zircaloy-2 and Zircaloy4 (Simnad, 1981). Eventually, some of the type 304 SS cladding was found to suffer intergranular stress corrosion cracking from the coolant side, mainly because of the radiolytic oxygen-containing environment (Terrani, 2018). The type 304 SS used in the 1960s contained a considerable amount of carbon and during the welding of the end caps, the austenitic steel would sensitize, making the weld seam area vulnerable to stress corrosion cracking from the coolant side, mainly because of the high corrosion potential in the water

14

Accident-Tolerant Materials for Light Water Reactor Fuels

by the presence of radiolytic oxygen and hydrogen peroxide. Since at the time when environmental cracking was observed in the welds of fuel rods the process of fabrication of Zircaloy was optimized and of lower cost, the use of austenitic stainless steel was discontinued for fuel cladding use. Because of the current efforts to develop advanced accident tolerant fuel materials, and many decades after austenitic stainless were discontinued, the use of ferritic stainless type alloys has been proposed for fuel cladding. In contrast to austenitic stainless steels, ferritic stainless steels are resistant to stress corrosion cracking in high temperature water. Moreover, due to their low carbon content, modern ferritic stainless alloys can be welded without undergoing sensitization by using pressure resistance welding, a solid-state welding process.

Chapter 2

Current materials in light water reactors. Why do we need a materials renewal? Chapter Outline Overview 15 The light water nuclear power reactor 16 Materials for light water reactors 17 Boiling water reactors 19 Pressurized water reactors 19 Reactor vessel for boiling water and pressurized water reactors 20 Fuel assemblies for boiling water and pressurized water reactors 22 Light water reactor fuels and the excellent performance of urania 23 How zirconium alloys became the material of choice for fuel cladding 24

In praise of zirconium alloys Waterside corrosion of zirconium alloys Nodular corrosion Hydrogen pickup by zirconium alloys Iodine stress corrosion cracking of zirconium alloys Shadow corrosion of zirconium alloys Crud deposition on zirconium alloys Irradiation damage of zirconium alloys

27 30 34 35 36 37 40 41

Overview Light water reactors (LWRs) have been generating electricity for over five decades for the electrical grid of more than 20 countries. Most of these reactors are built using a series of alloys and materials that have changed very little over the many decades. The most common structural materials are based on the three elements, iron (Fe), chromium (Cr) and nickel (Ni) such as stainless steels and nickel-based alloys. Most of the Fe-Cr-Ni alloys contain enough Cr to render them passive with low general corrosion in typical LWR environments. Since the chemistry of the water in the reactor is highly controlled, without aggressive impurities such as chloride or sulfate ions, localized corrosion is not a problem for the common structural alloys. The most common failure mode of the Fe-Cr-Ni alloys is environmentally assisted cracking or stress corrosion cracking (SCC), which is currently well understood and successfully mitigated, mainly by dissolving hydrogen gas into the coolant. Accident-Tolerant Materials for Light Water Reactor Fuels. DOI: https://doi.org/10.1016/B978-0-12-817503-3.00002-X © 2020 Elsevier Inc. All rights reserved.

15

16

Accident-Tolerant Materials for Light Water Reactor Fuels

The other fundamental material in the reactor core is the zirconium-based alloy used for cladding of the fuel rods. The traditional fuel rod alloys were Zircaloy-2 and Zircaloy-4, which were used regularly in boiling water reactor and pressurized water reactor systems, respectively. Very early in the development of the alloys (early 1950s), it was understood that a small amount (,2%) of alloying elements greatly reduced the general corrosion rate of the zirconium alloys. The traditional corrosion problems of zirconium alloys, such as nodular corrosion, shadow corrosion from the waterside, and SCC from the fuel side, are now well understood and under control. Debris fretting from the coolant side remains the main failure mode of zirconium alloys fuel cladding. Over time the degradation processes of zirconium alloys were successfully identified and managed to allow for their use in nuclear power generation for more than six decades.

The light water nuclear power reactor In a nuclear power reactor, the heat released during the fission of uranium in the fuel is captured by the water surrounding the fuel rods to produce steam, which is used to spin the turbines to generate electricity. The water that extracts the heat generated by the nuclear fission reaction is called the coolant. The fuel inside the rods is urania (or uranium dioxide UO2), which is used in the form of approximately 10-mm tall and 10-mm diameter pellets that are piled about 4-m high inside vertical metallic tubes called the cladding. A metallic tube filled with the ceramic fuel pellets is called a fuel rod. Currently the alloy for the cladding is generally a zirconium alloy such as Zircaloy-2, Zircaloy-4, M5, or Zirlo (Rebak et al., 2009; Bragg-Sitton et al., 2014; Motta et al., 2015; Tang et al., 2017). In the United States and around the globe, there are two main types of light water power reactors: (1) boiling water reactors (BWRs) and (2) pressurized water reactors (PWR) (IAEA, 2019). Currently, there are 451 operable civilian power reactors in the globe, of which 299 are PWR and 73 are BWR. The other types of power reactors are 49 pressurized heavy water moderated and cooled reactors, 14 gas cooled graphite moderated reactors, 13 light water cooled graphite moderated reactors, and three fast breeder reactors (IAEA, 2019). The main difference between the two common light water reactors (LWRs) is that in the BWR, the coolant that makes the steam is in direct contact with the fuel rods. In the PWR, the water coolant that is in direct contact with the rods (primary side) transfers its heat though a steam generator to water in the secondary side, which makes the needed steam for the turbines. In both the BWR and PWR systems, the water in contact with the external wall of the fuel rods is kept liquid at about 275 C 288 C with 7.5-MPa pressure in BWRs and at 290 C 330 C with 15.5-MPa pressure in PWRs (Zinkle and Was, 2013). The chemistry (composition) of the cooling water is highly controlled. The water may contain dissolved hydrogen gas to lower the corrosion potential of

Current materials in light water reactors Chapter | 2

17

the metallic components in contact with the water. The presence of hydrogen in the water gas depresses the corrosion potential of most of the metallic materials to electrochemical potential regions where the general corrosion and stress corrosion cracking (SCC) of the alloys is minimized (Ford et al., 2006; Scott and Combrade, 2006). That is, the purpose of the reducing hydrogen gas dissolved in the water is to bring the open circuit potential or corrosion potential of the engineering components alloys to the vicinity of the “a” line in a Pourbaix diagram where SCC is essentially controlled (Jones and Nelson, 1990). In PWRs the water may also contain dissolved boric acid (as a neutron moderator) and lithium hydroxide (to control the value of the pH to minimize corrosion) (Scott and Combrade, 2006). Other additions to the water may include low concentrations of zinc injections to increase the protectiveness of the oxide films formed on metallic surfaces in contact with the coolant (Betova et al., 2011). Zirconium alloys have been used for over six decades to clad the urania fuel in light water power reactors. Zirconium alloys have adequate resistance to corrosion in high-temperature water under the standard operation conditions of PWRs and BWRs at near 300 C (Lemaignan, 2006; Motta et al., 2015). However, the corrosion resistance in water and steam of zirconium alloys decreases rapidly as the temperature increases above 400 C. The oxidation of zirconium with water is highly exothermic. During the plant blackout events of March 2011 at the Fukushima Daiichi nuclear power stations, the water temperature inside the reactor increased to above the normal operation condition values for the reactors and the zirconium alloys reacted rapidly with water and steam forming hydrogen gas and releasing large amounts of oxidation reaction heat to the environment. After almost seven decades of successfully using zirconium alloys in the reactors, the international nuclear materials community is now searching for safer and more resistant alternative materials for the fuel components.

Materials for light water reactors There are two well-established types of LWRs: (1) boiling water reactors of BWR and (2) pressurized water reactors of PWR (Figs. 2.1 and 2.2). The circuit in the BWR is simpler than in the PWR configuration, which requires two water circuits. Except for the two smaller red boxes in Figs. 2.1 and 2.2 showing the fuel components in both reactors, almost all other components are made with alloys containing the three common elements: iron (Fe), chromium (Cr), and nickel (Ni) (Fig. 2.3). Iron is used for structural low-cost components, and chromium and nickel are used to provide resistance to environmental effects. Around the world, BWRs represent approximately 20% of the installed nuclear power capacity and PWRs represent 67% of the nuclear power (NNL, 2018). The concept of a light water power reactor was initially conceived at Oak Ridge National Laboratory in 1946 (Van Duysen and

18

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 2.1 Boiling water reactor materials. Courtesy from R.W. Staehle.

FIGURE 2.2 Pressurized water reactor materials. Courtesy from R.W. Staehle.

Meric de Bellefon, 2017). Also in 1946, while working at Oak Ridge, H.G. Rickover made the historical decision to use LWRs over gas cooling reactors to power the submarines for the US Navy he was sponsoring and designing at that time.

Current materials in light water reactors Chapter | 2

19

FIGURE 2.3 Ternary diagram. Most materials in light water reactors are based on Fe, Cr, and Ni. Courtesy from R.W. Staehle.

Boiling water reactors After its original conception at Oak Ridge, the technical development of BWRs was conducted at Argonne National Laboratory, and prototypes were built in the mid-1950s in Idaho and Illinois (Simnad, 1981; NNL, 2018). General Electric (GE) established an atomic power business unit in 1955 and picked up the BWR design to build the first ever US-licensed reactor at the Vallecitos site near Sunol, California. The Vallecitos BWR was the basis for the future construction of commercial power BWRs all around the globe. This GE series of worldwide BWR constructions began in 1959 with the Dresden Unit 1 reactor in Morris, Illinois, which started to generate civilian electricity in 1961. The design of the BWRs evolved through the decades and currently several types and forms of BWR nuclear power stations are found in the United States, Japan, Sweden, Finland, Spain, Switzerland, Germany, Mexico, and India (NNL, 2018).

Pressurized water reactors As the BWRs were first engineered at Argonne National Laboratory, the first PWRs were designed and technically developed at Westinghouse Bettis Atomic Power Laboratory near Pittsburgh, with the initial purpose of using the PWR system for the propulsion of nuclear submarines (Cummins and Matzie, 2018). Currently in the United States, PWR power is used for

20

Accident-Tolerant Materials for Light Water Reactor Fuels

propulsion of submarines, aircraft carriers, and ice breakers. Besides Westinghouse, two other companies in the United States (Combustion Engineering and Babcock & Wilcox) also supplied designs and components for PWRs, both for the Navy and civilian power (Cummins and Matzie, 2018). Eventually, Combustion Engineering became a part of Westinghouse and Babcock & Wilcox became a part of Framatome. The first civilian USproduced electricity using PWR atomic power was the Shippingport reactor for the Duquesne Light Company, south of Pittsburgh, a utility that was founded by George Westinghouse in 1912 (Cummins and Matzie, 2018). The materials of construction for LWRs have changed very little in the six decades of civilian nuclear power (Allen et al., 2010; Rippon, 1984). Figs. 2.1 and 2.2 show the most relevant materials of construction for LWRs. Most of the alloys are based on the three elements iron, chromium, and nickel (Fig. 2.3). Table 2.1 shows a condensed list of alloys and their chemical composition used for the construction of LWRs. Iron-based alloys containing enough chromium for passivation are stainless steels. Fig. 2.3 shows that most stainless steels contain approximately 20% of chromium. If the stainless steels contain some nickel, they would be austenitic because they adopt the FCC or gamma phase structure. When the steel does not contain nickel, its phase would be ferritic, BCC, or alpha. Fig. 2.3 (right) shows the nickel-based alloys, which also contain approximately 20% chromium for passivation. The lowest chromium content in the nickel alloys are in X-750 and alloy 600 and the highest chromium content is 29% for alloy 690 (Table 2.1). The main failure mode of austenitic stainless steels and chromium containing nickel alloys during normal operation in the reactor core is SCC or environmentally assisted cracking (EAC). The cracking susceptibility of structural stainless steels and nickel alloys is generally minimized by lowering the redox potential in the coolant via the injection of hydrogen gas (Jones and Nelson, 1990). All the alloys containing at least 12% chromium would have a low general corrosion rate because of the formation of a protective chromium oxide film on the environment exposed surface, which kinetically passivates the metallic structures and the dissolution (or corrosion) of the component becomes negligible. Most of the materials in Table 2.1 have been used for many decades. The newest material listed in this table is Alloy 690 (N06900), which was introduced for steam generators tubing in PWRs in 1989. That is, the newest material used in LWRs is currently 30 years old. Materials not listed in Table 2.1 are the zirconium alloys for fuel rod cladding, which are discussed separately.

Reactor vessel for boiling water and pressurized water reactors In Figs. 2.1 and 2.2 there is a vertical gray line. To the right of the vertical gray line, both reactor systems are similar since the common feature is the

Current materials in light water reactors Chapter | 2

21

TABLE 2.1 Nominal composition of light water reactor materials in mass percent. Alloy—UNS

Fe

Cr

Ni

Others

Carbon steel A508

B97

# 0.25

0.4 1.0

# 0.25C, 1.2 1.5Mn, 0.45 0.60Mo, # 0.40Si

308 SS—welds

B70

19.5 22.0

9.0 11.0

1 2.5Mn, 0.08C max, 0.75Mo max

309 SS—clad

B64

23.0 25.0

12.0 14.0

1 2.5Mn, 0.3 0.65Si, 0.12C max

Alloy 600 N06600

6 10

14 17

72 min

0.15C max, 1Mn max, 0.5Si max

Alloy 182—welds

10 max

13 17

59 min

0.1C max, 5 9.5Mn, 1Si max, 1 2.5 (Nb 1 Ta)

Alloy 690 N06900

7 11

27 31

58 min

0.05C max, 0.5Mn max, 0.5Si max

Alloy 800 N08800

39.5 min

19 23

30 35

0.1C max, 0.15 0.6Al, 0.15 0.6Ti

304 SS S30403

B74

18 20

8 12

0.08C max, 2Mn max, 0.75Si max

316 SS S31603

B74

16 18

10 14

0.08C max, 2 3Mo, 2Mn max, 0.75Si max

X-750 N07750

5 9

14 17

70 min

0.08C max, 2.25 2.75Ti, 0.4 1Al, 0.7 1.2 (Nb 1 Ta), 1Mn max

A286 S66286

B62

13.5 16

24 27

0.08C max, 2Mn max, 1Si max, 1 1.5Mo, 1.9 2.35Ti, 0.1 0.5 V

405 ferritic steel S40500

B88

11.5 14.5

0.5 max

0.08C max, 0.1 0.3Al, 1Mn max, 1Si max

duct carrying the steam across this vertical gray line to propel the turbines. Even fossil fuel plants have a similar configuration as the components shown on the right of the gray line. The components to the left of the gray line [including the reactor pressure vessel (RPV) and the heat exchangers] are enclosed in the reinforced concrete containment building of the nuclear power station. The concrete containment building is the last barrier for radionuclides and other contaminants before they can reach the external

22

Accident-Tolerant Materials for Light Water Reactor Fuels

environment, which may affect humans, plants, and animals. The vessel of each reactor is the part of the power plant that gives the power station its identity. The pressure vessel is the most important component and its integrity must always be guaranteed (Van Duysen and Meric de Bellefon, 2017). The nuclear fission reactions produce toxic radioactive derivatives and large amounts of heat happen inside of the pressure vessel. The so-called reactor includes the vessel itself and all the needed internal ancillary components. Most of the internal components can be repaired or replaced during the maybe 80-year life of a nuclear plant. However, the vessel cannot be replaced. The reactor vessel is constructed using fine-grained low alloy steel (e.g., A508 in Table 2.1) to provide structural integrity to the container. For PWR reactors, the inner diameter (ID) of the pressure vessel is 4 5 m with a wall thickness of up to 250 mm. In general, the BWR vessels are larger than the PWR vessels. The carbon steel structure inside of the vessel is clad using austenitic stainless steel weld overlay (e.g., type 308 SS in Table 2.1), generally 0.125 in. (3 mm) thick, to provide protection against corrosion in the 300 C pressurized water environment. In the BWR vessels the stainless steel cladding is only applied to the portion of the vessel that is in contact with condensed water. Since the head of the vessel in a BWR is in contact with saturated steam, it is not clad with stainless steel. In the PWR vessel, the entire ID surface of the vessel is clad with stainless steel since the entire container is flooded with condensed water.

Fuel assemblies for boiling water and pressurized water reactors Both in BWR and PWR reactors, the fuel systems are similar. They consist of urania pellets piled inside long vertical narrow tubes made of zirconium alloys. In the BWR the zirconium alloy is mostly Zircaloy-2 and in the PWR the zirconium alloy may be Zircaloy-4, M5, or Zirlo in most Western countries (Motta et al., 2015). The zirconium alloy tubes have an outside diameter (OD) of 9.55 mm for PWRs and 10.26 mm for BWRs. The wall thickness of the zirconium alloy tubes may vary between 0.6 and 0.8 mm. The combination of a zirconium alloy tube filled with stacked urania pellets is called a fuel rod. The rods in both types of reactors are approximately 4-m long, and they are assembled in square bundles of, for example, 10 3 10 (B100 rods) in BWRs and 14 3 14 or 17 3 17 (200 300 rods) in PWRs. The fuel rod bundles are stacked vertically side-by-side in a specially designed grid. There could be 800 bundles per core in BWRs and 150 200 fuel assemblies in PWRs (Terrani, 2018). The BWR bundles may contain shorter rods and the entire bundle or assembly is encased in a slender box called channel, which are also made of Zircaloy-2 alloy. The channels are used to reduce water crossflow between

Current materials in light water reactors Chapter | 2

23

fuel assemblies. Controls rods in the reactors are used to maintain the desired rate of fission reactions. The BWR control rods or blades are in the form of a cross and they run from the bottom of the reactor in the space between four channeled bundles or assemblies. In PWRs the control rods driving mechanisms are mounted on the head of the pressure vessels and are inserted from above into the reactor. Control rods contain neutron absorbing elements (such as boron) encased in stainless steel cladding and their objective is to stop the thermal neutron chain reaction.

Light water reactor fuels and the excellent performance of urania The driving force in the development of nuclear power fuel was motivated not only by the technical performance of the fuel itself but also by targeting a lower fuel cycle cost and to reduce plant operation costs. Lower fuel costs were essential to compete with electricity generated using fossil fuels in the decades before clean energy for a clean environment and climate change were some of the value forces behind civilian power generation (Van Duysen and Meric de Bellefon, 2017). Besides the economical requirements, the fuel elements were designed considering the following essential technical capabilities: (1) nuclear reactivity, (2) suitable heat transfer capability, (3) retention of fission products, (4) maintaining structural and mechanical stability, and (5) intrinsic safety under accident conditions (Simnad, 1981). As regards the economical and management characteristics, the manufacturability of the fuel and the proper associated quality assurance procedures were also extremely important. Of all the fuels that were investigated in the early years of civilian nuclear power, UO2 or urania often showed at the top of the needed requirements for the fuel. The urania fuel was prepared in the form of cylindrical cold-pressed and sintered pellets with densities of the order of 95% of the theoretical density. Since the early 1960s the urania pellets were mostly clad in zirconium alloys but some austenitic stainless steel tubing was also used for early cladding (Simnad, 1981; Lustman, 1981; Terrani, 2018). The excellent performance of urania pellets clad in Zircaloy-2 tubes in the Shippingport plant in Pennsylvania in the mid-1960s cemented the acceptance of this fuel cladding pair combination for hundreds of reactors worldwide in the decades to come. A gradual decrease in fuel bundle failures and an overall performance increase of fuel rods (Zircaloy cladding with urania pellets) has been taking place over the years and decades. For example, the early creep collapse of zirconium clad rods has been minimized by pressurizing the fuel cavity with helium. This also helped with heat transfer to the cladding and coolant, and therefore reduced the overall temperature of the fuel.

24

Accident-Tolerant Materials for Light Water Reactor Fuels

How zirconium alloys became the material of choice for fuel cladding In the United States, civilian commercial nuclear power started mostly as an expansion of nuclear power developed for nuclear submarines for the Navy. The motivation was similar for the former USSR, now the Russian Federation. As of today, a group of civilian reactors in the United States is called a fleet, like a group of submarines. Naval reactors were mostly built using the materials and alloys commercially available in the late 1940s and early 1950s, except for the zirconium alloys, which were purposely developed on an industrial scale for the submarine nuclear reactors’ fuel cladding. The role of cladding is to separate the water from the fuel and to avoid the migration of fission products into the coolant and possibly into the environment (Kass, 1964). The cladding also acts as a structural and spatial support for the fuel and as a heat exchanging surface in which the water removes the heat of fission that diffuses from the fuel pellet across the clad wall. Therefore besides the requirement of low thermal neutron absorption cross section, the tube clad material needs to be resistant to attack by hot water from the OD, it needs to be sufficiently strong to stand its own weight at reactor temperatures, it must maintain a coolable geometric grid, and possess high thermal conductivity (Kass, 1964). In 1946 47, there were a few candidates for the nuclear fuel cladding for the submarines pressurized water propulsion reactors, including zirconium, aluminum, beryllium, and stainless steels (Rickover et al., 1975). Initially, zirconium was not highly promising as it was not produced on an industrial scale necessary to supply tubing for the planned reactors and it absorbed more neutrons than was originally anticipated (Hillner, 1977). The decision point to use zirconium for cladding materialized in December 1947 when scientists from Oak Ridge National Laboratory were able to remove from the produced zirconium metal the 2% hafnium impurity. Hafnium is a companion of zirconium in natural ores, and while the neutron absorption cross section for zirconium is 0.184 barns, for hafnium it is 104 barns, almost three orders of magnitude higher. Rickover acknowledged that the presence of hafnium in the manufactured zirconium metal in the 1940s was responsible for the higher absorption of the neutrons needed for the fission reaction to proceed (Rickover et al., 1975). It is interesting to note that the decision by Rickover to use zirconium for the fuel cladding was taken before any corrosion data was available for zirconium in high-temperature water (Kass, 1964). In the decade following the decision to use zirconium for fuel cladding, the major problem encountered by the developers of the first nuclear submarine was the procurement of sufficient amounts of zirconium. Zirconium metal was first produced on a commercial scale at the US Bureau of Mines in Albany Oregon using the Kroll Process (Stephens, 1984; Adamson, 2010). Although zirconium did not have a wide commercial interest until the

Current materials in light water reactors Chapter | 2

25

mid-20th century for nuclear power, its existence was known since 1824 because of the early studies by Berzelius. Since the initial developments on the production of Zirconium in Albany, Oregon, the next production step for the naval reactors was carried out at the Carborundum Metals in Akron, New York, in August 1953 (Stephens, 1984). By 1957 the annual production of zirconium was 147,400 kg. As the demand for zirconium continued, four more plants were opened in the United States and one in Japan. In 1956 the Wah Chang Company was given the contract by the US Atomic Energy Commission to run the zirconium plant in Albany, Oregon. This Oregon plant is still in operation in 2019, continuing to make zirconium ingots by the Kroll process. A completely integrated Western Zirconium plant started producing zirconium at more than a million kilogram a year in the early 1980s in Ogden, Weber County, Utah (Stephens, 1984). The reduction process to zirconium metal in Western Zirconium is basically the same as that used by Wah Chang (Stephens, 1984). The first time zirconium was used in a reactor was in the experimental Mark I at Idaho National Laboratory which reached criticality on March 30, 1953. Meanwhile the behavior of zirconium was investigated to make it more predictable from the point of view of corrosion response, by adding alloying elements. The first corrosion tests of zirconium metal in hightemperature water showed that below 316 C, the oxide on the surface of the zirconium coupons was adherent and protective; however, at temperatures of 360 C and higher, the corrosion rate was fast (Kass, 1964). Impurities in the tested material were the main reason for uneven corrosion test results in hot water. One of the first corrective measures in the production of zirconium alloys was to reduce the level of nitrogen in the ingots, and to add smaller (less than 2.5%) amounts of tin (Sn), which was said to countereffect the action of nitrogen in the metal. Efforts to make zirconium alloys corrosion resistant were pursued for the Rickover’s Nautilus submarine and tin was found to be the most promising alloying element. When tin was added to a quantity of 2.5% to zirconium, the alloy Zircaloy-1 was created, but this alloy did not show consistent corrosion behavior as a function of time (Rickover et al., 1975). The Bettis Atomic Power Laboratory near Pittsburgh is considered to be the birthplace of zirconium alloys. At Bettis, a piece of stainless steel was added apparently by mistake to a batch of zirconium and the result was an alloy with great corrosion resistance properties. The resulting alloy between zirconium and 1.5% tin and small amounts of stainless steel was called Zircaloy-2, which had highly improved corrosion resistance in hot water and steam compared with unalloyed zirconium. Zircaloy-2 was selected by Rickover in August 1952 to be used in the Nautilus submarine before even an ingot with the right composition of Zircaloy-2 was produced and tested. This first nuclear reactor in a submarine generated electricity on December 30, 1954 and the submarine started navigation using nuclear propulsion on January 17, 1955.

26

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 2.2 Commercial zirconium alloys for cladding. Alloy

Nominal composition in mass percent

Zircaloy-2 or R60802

Zr 1 1.2/1.7Sn 1 0.07/0.20Fe 1 0.05/0.15Cr 1 0.03/0.08Ni (Fe 1 Cr 1 Ni 5 0.18 0.38)

Zircaloy-4 or R60804

Zr 1 1.2/1.7Sn 1 0.18/0.24Fe 1 0.07/0.13Cr (Fe 1 Cr 5 0.28 0.37)

ZIRLO

Zr 1 1Sn 1 1Nb 1 0.1Fe (optimized Zirlo has 0.67Sn)

M5

Zr 1 1Nb 1 0.14O

E110

Zr 1 1Nb

E635

Zr 1 1.2Sn 1 1Nb 1 0.35Fe

Zr-2.5Nb or R60904

Zr 1 2.4/2.8Nb

Eventually the nickel in Zircaloy-2 was replaced by iron to minimize hydrogen absorption by the cladding, which was attributed to the presence of nickel. The iron containing alloy was called Zircaloy-4 and was adopted in all US submarines since the mid-1960s (Rickover et al., 1975) (Table 2.2). The early wisdom of adding less than 0.5% of transition metals (Ni, Cr, and Fe) to zirconium to obtain the enduring classic Zircaloy-2 (R60802) and Zircaloy-4 (R60804) alloys was recognized half a century later (Lemaignan, 2002; Motta et al., 2015). In the past 30 years, two newer zirconium alloys (Zirlo and M5) were developed in the Western world to replace Zircaloy-4 in civilian PWRs (Mardon et al., 2000; Sabol, 2005; Motta et al., 2015) (Table 2.2). Zircaloy-2, internally lined with softer zirconium metal, is still in use in BWRs 70 years after its discovery. Heavy-water CANDU power reactors initially used Zircaloy-2 and now they use alloy Zr 1 2.5% Nb (R60904), originally developed in the former Soviet Union (Table 2.2) (Cheadle, 2010). The main reason that Canada shifted from Zircaloy-2 to Zr 1 2.5Nb was because of the higher strength of the latter. It was later demonstrated that Zr 1 2.5Nb would pick up less hydrogen generated by corrosion than Zircaloy-2. Moreover, the CANDU tubes are fabricated in such a manner that if platelets of hydride precipitate they will be in the circumferential orientation, that is, in the least deleterious alignment (Cheadle, 2010). Zirconium alloys had a myriad of minor technical problems, which were understood, corrected, or managed over the decades. One positive attribute of zirconium alloy was its resistance to SCC from the coolant (water) side, which has been a major concern for other early cladding materials such as austenitic stainless steels, at least in their initial application in BWRs (Lustman, 1979). It was eventually demonstrated that zirconium can suffer

Current materials in light water reactors Chapter | 2

27

SCC from the fuel side when the fission product iodine is formed in the ID of the tube. The problem of the pellet-cladding mechanical interaction (PCMI) and iodine-induced cracking was solved at GE by the introduction of a ,100-µm layer of softer zirconium metal inside the Zircaloy-2 cladding. This softer layer of zirconium reduces the stresses on the Zircaloy-2 tube wall due to fuel pellet swelling mainly during ramping and therefore the stress-related failures from the fuel cavity decreased (Pickman, 1994). The United States was the leader in developing the alloys Zircaloy-2 and Zircaloy-4, mainly at the Bettis Atomic Power Laboratory. These two alloys have been tested and used in many countries in the Americas, Europe, and Asia for heavy and LWR cladding applications. Over the decades the two Zircaloy alloys may have had small changes or optimization, and newer improved zirconium alloys came into the market; however, in general, zirconium alloys have been used reliably for power generation for many decades now. That is, the smart decision by Rickover in December 1947 is valid 70 years later, not only in the United States but also around the globe. For example, the suitability of Zircaloy was never questioned by the German utilities when the construction of the first pressurized heavy water reactor started in Karlsruhe in 1961 (Garzarolli et al., 1996). It was deemed that zirconium alloys were reliable not only under normal operation conditions but also under postulated accidental conditions and for long-term intermediate storage after used fuel discharge from the reactor core.

In praise of zirconium alloys In the first few years of commercial nuclear power the cladding for the fuel and other internal components was made of stainless steel, because zirconium was not much available and information about zirconium alloys was scarce. Zirconium alloys were expensive for civilian power generation purposes and were used mainly for submarine reactors. But eventually by the mid-1960s, most commercial reactors switched from austenitic stainless steels to zirconium alloys for the fuel cladding materials. The main driving force was the need to improve neutron efficiency and reduce the fuel cycle cost (Lemaignan, 2006). The structural metallic components inside a water power reactor need to withstand triaxial dynamic stresses while under neutron irradiation and environmental attack, such as from radiolytic ionized water at 300 C. In the case of the fuel cladding made with Zircaloy, it also needs to withstand attack from stresses and fission products from the inside diameter of the tube in the two most typical reactors, the BWR and the PWR. From the point of view of the fuel interaction with the internal diameter of the Zircaloy cladding, tubes in both BWR and PWR reactors show similar behavior. However, from the OD of the tubing, in contact with the water coolant, the environmental resistance may be different between BWR and PWR. For example, BWR water may have more oxidants in the water

28

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 2.3 Typical operational conditions for fuel rods in light water reactors. BWR

PWR

Coolant chemistry and high-purity water

Pure water, 0.3 ppb O2, 3 ppb H2

, 1 ppb O2, 3 ppm H2, 2 ppm Li, 600 ppm B

System pressure

7 MPa

15 MPa





Cladding temperature

280 C 320 C

290 C 400 C

Time in the reactor

1800 days

1500 days

Mass of Zr per core

B33 tons

B25 tons

Burnup at discharge

55 MWd/kgU

60 MWd/kgU

Source: After Garzarolli, F., Stehle, H., Steinberg, E., 1996. Behavior and properties of Zircaloys in power reactors: a short review of pertinent aspects in LWR fuel. In: Bradley, E.R., Sabol, G.P. (Eds.), Zirconium in the Nuclear Industry: Eleventh International Symposium, ASTM STP 1295. ASTM International, pp. 12 32; Lemaignan, C., 2006. Corrosion or zirconium alloy components in light water reactors. In: ASM Metals Handbook, Vol. 13C Corrosion: Environments and Industries. ASM International, Metals Park, OH, p. 415; and Terrani, K.A., 2018. Accident tolerant fuel cladding development: promise, status, and challenges. J. Nuclear Mater. 501, 13 30 https:// doi.org/10.1016/j.jnucmat.2017.12.043.

than PWR water but the temperature is lower (Garzarolli et al., 1996). Table 2.3 shows comparatively the main operating conditions in both reactors’ cores. To avoid the water to boil at 325 C, the pressure inside the PWR primary circuit must be at least 150 times the atmospheric pressure (Table 2.3). The pressure in the PWR secondary circuit is lower and the water is allowed to boil inside the steam generators. Since each fuel assembly or bundle in a PWR may contain 200 250 rods and there are about 150 250 fuel assemblies per PWR core, the total mass of zirconium inside a PWR could be of the order of 25 tons (Terrani, 2018). In the simpler BWR configuration, there is only one circuit and the water is allowed to boil in the reactor core at a pressure that is approximately 75 times higher than the atmospheric pressure to keep a boiling temperature of approximately 285 C. The fuel rods in a BWR are arranged in vertical bundles of approximately 100 rods per bundle, and there are approximately 700 800 bundle assemblies per core. Each assembly is enclosed in Zircaloy-2 boxes, called channels. Therefore the total mass of zirconium (including cladding and channels) in a BWR core could be 33 tons (Terrani, 2018). That is, the amount of zirconium inside a BWR reactor core is approximately 30% higher than in a PWR core. Besides the tubes for the cladding of the fuel pellets, zirconium alloys have also been used for the frame of the fuel bundles such as the guide tubes in the PWR and the channels or channel box in the BWR and the pressure tubes for the CANDU reactors (Lemaignan, 2006). As mentioned before,

Current materials in light water reactors Chapter | 2

29

FIGURE 2.4 Degradation of zirconium alloys.

zirconium alloys performed well in water reactors for many decades, despite several issues that were eventually understood and fixed or controlled over the years. The degradation processes of zirconium alloys fuel tubes can be separated into two large groups: (1) from the water or coolant side and (2) from the fuel cavity side (Fig. 2.4). Over time, most of the environmental degradation issues of zirconium alloys have been observed, studied, and mitigated or controlled. The pervasive degradation that is still occurring in functioning reactors is the debris fretting against the fuel rods from the coolant side. The debris fretting may occasionally perforate the cladding tube. Fretting degradation happens when a foreign piece of debris such as a wire may be carried by the water flow and may get trapped in the separator grit.

30

Accident-Tolerant Materials for Light Water Reactor Fuels

The repetitive touching of the debris against the tube wall would make a scar and may infrequently lead to the perforation of the cladding. It is accepted that the use of accident-tolerant concepts for cladding such as hard coatings for zirconium alloys or iron chromium aluminum cladding would greatly mitigate this failure mode.

Waterside corrosion of zirconium alloys In the temperature range of operation of LWR, zirconium alloys develop an adherent protective zirconium oxide (ZrO2) on the surface according to the reaction Zr 1 2H2O-ZrO2 1 2H2. During reactor irradiation service, the water oxidation of zirconium is accelerated compared with out-of-pile laboratory autoclave immersion tests. The effect of irradiation on oxidation acceleration is less important as the oxide thickness on the alloy increases. The effect of the temperature on the oxidation of zirconium is a classical Arrhenius dependency. It has been known for decades that in the formation of surface oxides, two kinetic periods are generally identified (Wanklyn and Jones, 1962; Lemaignan, 2006). The first period is cubic or parabolic (or subparabolic) up to an oxide thickness of less than 5 µm and the second or linear period is for oxide thicknesses higher than 5 µm (Garzarolli et al., 1996; Lemaignan, 2006; Motta et al., 2015). The second linear period of faster oxide growth period on Zircaloy is sometimes called breakaway corrosion (Wanklyn and Jones, 1962; Motta et al., 2015). For zirconium niobium alloys (Table 2.2) the oxidation kinetics transition between the parabolic/ cubic and linear periods is less pronounced than for the Zircaloy-type alloys (Lemaignan, 2006). Black zirconium oxide is protective, while white zirconium oxide is non-protective, and it flakes off the surface (Hillner, 1977; Franklin, 2011). The oxide on a zirconium alloy exposed to water reactor temperatures grows at the interface of the oxide with the metal (Lemaignan, 2006; Motta et al., 2015). The entire oxide formation or corrosion process is controlled by the diffusion of species through the oxide film (Cox, 2005). During the oxidation process, the oxygen anion dissociated from the water at the oxide surface (in contact with the coolant) migrates through the existing oxide to react with a zirconium cation at the oxide metal border. It was claimed that the water dissociation at the oxide surface may be the rate limiting step for the oxidation of zirconium (Lemaignan, 2006). But others have argued against this assertion, saying that the rate limiting step is the diffusion of oxygen anions through the oxide film to meet the zirconium cation at the metal oxide interface (Cox, 2005). Cox stated that the zirconium atoms in the oxide lattice are immobile. The resulting oxide on the surface of the zirconium alloy cladding is an n-type semiconductor. The electrons liberated by the zirconium atom at the metal oxide interface travel to the surface of the oxide and reduce the hydrogen ions from the dissociated water to form either atomic hydrogen or hydrogen gas (Motta et al., 2015; Rebak et al., 2009).

Current materials in light water reactors Chapter | 2

31

Others argue that the positive hydrogen ions may enter the oxide before being reduced by the movable electrons. Therefore if the proton diffusion is faster than the electron (liberated by the zirconium cation), more hydrogen entrance into the oxide may occur. The activation energy for the oxidation of a standard Zircaloy in an LWR is high, meaning that a 20K temperature increase will double the oxidation rate of zirconium (Lemaignan, 2006). Both Zircaloy-2 and Zircaloy-4 contain transition metals Cr, Ni, or Fe which are highly insoluble into the alpha zirconium matrix, therefore these transition elements precipitate in the form of intermetallic compounds such as Zr2(Fe,Ni) or Zr(Fe,Cr)2 called secondphase precipitates (SPPs). The size and distribution of the SPP control the oxidation behavior of the zirconium alloys, and it has been reported that the optimum corrosion resistance in PWR environments requires larger size precipitates ( . 150 nm) than the BWR environments, which may need an optimum size of ,80 nm (Lemaignan, 2006). The size and distribution of the precipitates in the zirconium alloys can be controlled by thermal treatments and cooling rates of the final product (e.g., tubes or channels). If the alloy is heated to the beta or body-centered cube temperature range, all the alloying elements would dissolve into the zirconium matrix. By controlling the cooling rates into the alpha or hexagonal temperature range, the precipitates will develop, and the SPP sizes can be controlled by controlling nucleation rates and/or growth rates. Initially, as it first forms, the surface zirconium oxide consists of small equiaxed grains with a mixture of mostly dense and protective tetragonal and some monoclinic zirconia, but as the oxide grows it becomes columnar monoclinic and less protective (Lemaignan, 2006; Motta et al., 2015). The type of oxide that develops on the surface of zirconium alloys during reactor service will depend on the temperature, time, and environment of service. Fig. 2.5 shows the appearance of Zircaloy-2 tube specimens after 9 months’ immersion under the three tested conditions (Table 2.4). After 9 months the tube specimens tested in the three conditions look practically undistinguishable from each other, with surface oxides that are lustrous and adherent. The appearance of the surface oxides suggests that they are thinner than 5 µm in thickness or in the pre-transitional parabolic stage of oxidation. Zirconium

FIGURE2 .5

Zircaloy-2 tube specimens.

32

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 2.4 Boiling water reactor and pressurized water reactor out-of-pile testing conditions. Autoclave

Test conditions

S-2

Simulated PWR, high-purity water, 3.75 ppm H2, 330 C

S-5

Simulated BWR, hydrogen water chemistry—HWC (0.3 ppm H2), 288 C

S-6

Simulated BWR, normal water chemistry—NWC (1 ppm O2), 288 C

FIGURE 2.6 Zircaloy mass gain. Immersion tests of Zircaloy-2 tube specimens in three different conditions.

alloys always gain mass (due to the incorporation of oxygen) during hightemperature water immersion testing, both in reducing (containing hydrogen gas) and oxidizing conditions (containing oxygen or hydrogen peroxide). However, the total mass gain (i.e., oxide thickness) can be different depending on the temperature or redox potential in the system. In reactor service, for the standing 4-m long fuel rod, the zirconium alloy tubing generally has a thicker oxide at the top of the rod than at the bottom, since the water becomes hotter as it rises in the reactor (Lemaignan, 2006). Fig. 2.6 shows the mass gain for Zircaloy-2 tube specimens as a function of the immersion time in simulated pure water BWR and PWR out-of-pile autoclave tests. In the hydrogen environments (PWR 330 C and BWR HWC 288 C) the mass

Current materials in light water reactors Chapter | 2

33

FIGURE 2.7 Zirconium oxide formed on tube. After 6-month immersion in hydrogenated water at 330 C the ZrO2 thickness was 1.5 µm. SPP rich in Ni and Fe are observed in the base metal and in the oxide.

gain was lower than in the oxidizing environment (BWR NWC 288 C). In the hydrogen environments the mass gain was higher for the PWR-simulated environment because the temperature was 42 C higher (330 C vs 288 C for BWR HWC). The general rule in the industry is that a mass gain by oxidation of zirconium alloy of 0.15 mg/cm2 corresponds to a ZrO2 oxide thickness of 1 µm. For example, in Fig. 2.6 at 26 weeks (6 months) immersion in PWR-type water, the mass gain is B0.22 mg/cm2, which may correspond to an oxide thickness of 1.5 µm. Fig. 2.7 shows the characteristics of the oxide formed on the OD of a Zircaloy-2 tube specimen exposed for 6 months (26 weeks) to the simulated PWR environment (Table 2.4). After the 6-month exposure at 330 C in pure hydrogenated water the oxide consisted of

34

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 2.8 Nodular corrosion. Nodular corrosion formed on a susceptible Zircaloy-2 tube specimen after the ASTM G2 test in steam at 400 C (Rebak et al., 2009).

columnar grains with a thickness of 1.5 µm, which correlates to the mass gain shown in Fig. 2.6. The oxide and the substrate in Fig. 2.7 also show the presence of SPP particles, which are rich in nickel and iron. As the zirconium oxide (ZrO2) grows at the oxide metal interface, the SPPs stay in place while the zirconium oxidizes around them. It has been reported in the literature that oxides grown in the reactor in reducing PWR-type environments are generally black and uniform in thickness, probably because they are pre-transition oxides or hypo-stoichiometric (Lemaignan, 2006). After the transition period to linear growth kinetics, the oxides may become lighter and be of nonuniform thickness with internal cracks parallel to the surface. Under more oxidizing conditions or at higher temperatures, the oxide may grow thicker in discrete regions and then the oxide may grow sidewise. If the thickness of the zirconium oxide reaches 50 100 µm, it may become non-protective and start to spall off the tube (Lemaignan, 2006).

Nodular corrosion Nodular corrosion is an anomalous type of corrosion oxidation of zirconium alloys from the waterside (Fig. 2.4) (Lemaignan, 2006). This type of localized oxidation happens mostly in BWR reactors (more oxidizing conditions than in PWR) at the higher end of temperatures 400 C 500 C. Fig. 2.8 shows that nodular corrosion appears as white patches where the oxide

Current materials in light water reactors Chapter | 2

35

thickness is much higher (e.g., 250-µm thick) than in the nearby metal which would have a more uniform and thinner oxide (e.g., 10 µm). Eventually the nodular patches may grow laterally together on the surface of the tube. It has been shown repeatedly that the size and distribution of intermetallic secondphase particles (SPP) in the cladding may be responsible for nodular corrosion by a galvanic like corrosion process. The SPPs are rich in Fe, Cr, and Ni and more noble than the surrounding zirconium material. The cathodic reaction of hydrogen reduction on the particles is higher than in areas where there are no particles, or where the particles are very small. The enhancement of the cathodic reduction of hydrogen on the SPPs promotes the anodic oxidation of nearby zirconium increasing significantly the oxide thickness where the SPPs are present. More recent thermomechanical processing of zirconium alloys made the SPP finer and uniformly distributed and the nodular corrosion phenomenon was controlled and mitigated since the driving force for having a preferential area for hydrogen reduction would vanish. Typical reactor irradiation flux during the service of the fuel rods in the reactor may also alter the size and distribution of the particles in the cladding, and therefore change the corrosion resistance properties of the cladding over time.

Hydrogen pickup by zirconium alloys In the natural oxidation of zirconium with water, atomic hydrogen (H ) or molecular hydrogen (H2) is generated according to Zr 1 2H2O-ZrO2 1 4H or -ZrO2 1 2H2. Some of the hydrogen generated diffuses into the cladding wall and may remain in solid solution in the metal (probably to total soluble bulk concentrations of up to 150 ppm hydrogen). Other diffused hydrogen accounting for more than the terminal solid solubility (TSS) in the metal may react with metallic zirconium to form stable hydride platelets according to Zr 1 4H -ZrH4. The solubility of atomic hydrogen in Zircaloy increases with the temperature, and since the temperature of the cladding wall is the highest at the ID and the coldest at the OD, the dissolved hydrogen saturation level will be reached first at near the OD of the cladding, where the hydride platelets precipitate due to supersaturation. It was reported that most of the hydride precipitation happens in 10% 40% of the external wall of the cladding (Garzarolli et al., 1996). After irradiation in the reactor, the oxidized zirconium alloy tubes may show the presence of hydride platelets, which may have different morphologies and orientations depending on the cooling rate of the tubes, on the level of stresses in the matrix of the tube (which may be controlled by the expansion produced by the precipitation itself), and on the texture of the alloy in the region of precipitation (Lemaignan, 2006). The presence of hydrides in the matrix of zirconium alloy tubes may have deleterious ductility effects (mostly at ambient temperature), and the degree of ductility damage will depend highly on the

36

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 2.9 Pellet-cladding mechanical interaction (PCMI) of zirconium alloys.

orientation of the hydrides with respect to the applied tensile stress (Sawatzky and Ells, 2000; Lemaignan, 2006). Because of its small content of nickel, Zircaloy-2 may absorb a higher amount of hydrogen in service than Zircaloy-4, since nickel can enhance catalytic reactions with hydrogen reduction (Lemaignan, 2006). The formation of hydrides in the cladding material may contribute to the phenomenon of delayed hydride cracking (DHC), first reported in 1964. To avoid DHC the fabrication process of CANDU Zr 1 2.5Nb pressure tubes was modified to avoid hydrogen absorption during their reactor life (Cheadle, 2010).

Iodine stress corrosion cracking of zirconium alloys A phenomenon of SCC or EAC was reported for zirconium alloys in presence of iodine, both under reactor performance conditions and in out-of-pile laboratory experiments. For EAC to occur, three conditions are needed simultaneously: (1) a susceptible material (in this case the zirconium alloy), (2) the presence of tensile stressed in the material (provided here by the hooped stretched cladding in the ID due to swelling of the fuel pellet), and (3) an aggressive environment (the iodine gas I2 formed by nuclear fission in the fuel) (Fig. 2.9). In the reactor, under a temperature increase during irradiation ramping, the size of the fuel pellets inside the Zircaloy cladding may increase, and this change in size would impart mechanical stresses into the ID of the cladding wall (red block arrows in Fig. 2.9). If the swelling process

Current materials in light water reactors Chapter | 2

37

of the fuel is slow, the cladding may creep around the fuel and relieve the mechanical stresses, which would minimize the occurrence of EAC. But during fast power ramps, there is no time for the metal of the cladding wall to creep and the ID of the wall will develop tensile stresses. This phenomenon is called PCMI. During fast reactor power ramping, the fission process may generate iodine in the fuel, which with the assistance of tensile loads may promote cracks in the ID of the tube, and the crack propagation may even perforate the cladding wall (Garzarolli et al., 1996). Laboratory studies showed that for the same level of stresses, the crack length increased with the amount of iodine generated. Other factors that influence the cracking of zirconium alloys are the yield stress, texture, and grain size of the cladding material. The higher the yield stress of the tube material the higher the susceptibility to iodine-induced cracking. Irradiation increases the yield stress by a mechanism of irradiation hardening. Tubes with smaller grain sizes are more resistant to cracking. The most active paths for cracking seem to be at approximately 30 degrees from the hexagonal crystallographic basal planes (hcp) of the zirconium structure. That is, it is possible to design a tube with texture that is resistant to EAC. To avoid PCMI failure, it is necessary to remove at least one of the fundamental conditions (tensile stress, sensitive material, aggressive environment), which are responsible for EAC, that is, remove at least one of the rings in Fig. 2.9. The principal types of remedies may include (1) modification of the cladding tube, (2) operation restrictions, and (3) fuel design improvements, all of which would minimize the level of tensile stresses on the cladding ID. GE had the best solution to control iodine-induced cracking of Zircaloy-2 using a barrier layer in the ID of the tube made with a softer purer zirconium liner, which would accommodate better the expansion behavior of the fuel pellets avoiding the development of tensile stresses on the substrate Zircaloy cladding. Zirconium alloys are not susceptible to EAC from the waterside of the fuel rod (Fig. 2.4).

Shadow corrosion of zirconium alloys Shadow corrosion in zirconium alloys is an enhanced local oxide growth similar to the phenomenon of nodular corrosion. The increased local oxide growth in shadow corrosion can be associated with the enhancement of the local oxide electronic conductivity (Cox, 2005). Shadow corrosion mostly happened in BWR reactors when imprints (or shadows) of structural components made of stainless steels or nickel alloys appeared on zirconium alloy components, such as boxes. Fig. 2.10 (left) shows an image of the shadow corrosion of a control rod handle on the Zircaloy channel encircled in red. On the right, there is a schematic representation of how the control blades fit in between four channels in a BWR. The oxide thickness in the imprint area

38

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 2.10 Channel shadow corrosion.

on the Zircaloy component was much larger than the oxide in the surrounding surface. The shape of the shadow attack corresponds to the shape of the handle. For example, the zirconium oxide thickness on the channel could be of the order of 4 µm; however, in the “shadow” area the oxide thickness of the zirconium oxide could be more than ten times higher. The first observations of shadow corrosion may have been reported as nodular corrosion under spacer grits made of a more corrosion-resistant alloy in the reactor. Shadow corrosion is not observed in PWR with enough hydrogen gas additions (Cox, 2005). Shadow corrosion was not observed either in out-of-pile autoclave laboratory tests, suggesting that irradiation was a necessary condition for shadow corrosion to develop (Lysell et al., 2005). Traditionally, two hypotheses have been used to explain the phenomenon of shadow corrosion: (1) galvanic corrosion and (2) local radiolysis. However, there are evidences that these two hypotheses may not explain by themselves all the occurrences of shadow corrosion. Lysell et al. (2005) argued that the most likely cause of shadow corrosion is the theory of galvanic corrosion. They also state that the reason shadow corrosion is not reproducible in autoclave (out-of-pile) tests is because of the enhanced photoconductivity effect caused by reactor irradiation on the zirconium alloy oxide. Adamson (2007) summarized the variables influencing shadow corrosion, mainly from reactor field observations. Adamson confirms that shadow corrosion occurs only under irradiation in a reactor and it was never reproduced in out-of-pile laboratory autoclaves. He lists three mechanisms by which irradiation assists the shadow corrosion process: (1) increase in the conductivity of the oxide film on the Zr alloy, (2) increase in the conductivity of the water by ionizing it, and (3) it accelerates the reaction of the reduction of molecular oxygen to form hydroxyl anions. Adamson (2007) also acknowledged (like others before him) that shadow corrosion is observed in BWR plants but not in PWR plants or in plants with high hydrogen content (under highly reducing conditions). This important observation suggests that shadow corrosion may be caused by a separation in

Current materials in light water reactors Chapter | 2

39

the corrosion potential of the two types of materials involved in the shadow corrosion process in the reactor environment (Kim et al., 2011). That is, when hydrogen is present in the system, it forces all metallic materials in the reactor to have a low (and very similar) corrosion potential and therefore there is no gap of potential between a stainless steel handle and a Zircaloy-2 nearby channel. Based on experiences from BWR, CANDU, and PWR reactors, Cox (2005) pointed out that a dissolved hydrogen content higher than 10 cc/kg may be necessary to eliminate the occurrence of shadow corrosion since this amount of hydrogen would guarantee that all the materials would be operating at the hydrogen reversible reaction potential level (near the “a” line in a Pourbaix diagram). Adamson (2007) noted that the fact that anomalous enhanced corrosion (oxidation) that may be observed at the location of shadow corrosion did not necessarily imply that also a higher hydrogen pickup would occur in that area. A similar observation was reported by Cox (2005) and he implied that it could be because the hydrogen reduction may be occurring on the neighboring stainless component producing the “shadow” or imprint. This fact may differentiate shadow corrosion from other types of accelerated corrosion (like nodular corrosion), since in the latter the enhanced thickening of the oxide film is generally accompanied by a higher hydrogen pickup. Adamson (2007) presented a case from the field in which shadow corrosion was “cast” over a Zr alloy channel from a stainless steel handle of a control blade, even though the handle was more than 5-mm away from the channel. This observation may suggest that the formation of a special environment (i.e., concentration of impurities or radiolytic species and such) between the two metallic components may not be needed for shadow corrosion to occur. In addition, Adamson (2007) pointed out that in the classical type of galvanic corrosion, the cathode must be of similar area or larger than the anode that is affecting, but in the case of shadow corrosion the cathode (e.g., stainless steel handle) is generally of a much smaller area than the affected anode (i.e., the Zr alloy channel). This is another argument on why shadow corrosion may not be a pure galvanic corrosion phenomenon. Kim et al. (2011) performed electrochemical tests in laboratory autoclaves under ultraviolet (UV) illumination to study the electronic properties of the oxides formed on Zircaloy and on the “shadowing” corrosion-resistant alloys such as stainless steels and nickel-based alloys. They reported that in a BWR type of water at 300 C containing 1 ppm dissolved oxygen the corrosion potentials of shadowing metals become more positive under UV illumination, while the corrosion potential of the shadowed Zircaloy-2 becomes more negative under the same illumination. That is, the corrosion potential gap between stainless steel nickel alloy and zirconium alloy becomes larger under UV illumination or irradiation (Fig. 2.11). Kim et al. attributed the phenomenon of shadow corrosion to the different types of oxide electronic properties that develop on these two types of alloys. On stainless steels the oxides are rich in p-type chromium while on

40

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 2.11 Shadow corrosion mechanism.

zirconium the oxide is n-type zirconia (Kim et al., 2011). Kim et al. (2011) also performed in-situ electrochemical impedance spectroscopy studies and showed via Nyquist plots that under UV illumination the corrosion resistance of Zircaloy-2 decreased compared with non-illuminated Zircaloy-2. It was also revealed that UV illumination increased the galvanic current of Zircaloy-2 coupled with dissimilar alloys such as Alloy X-750, 304 SS, or Pt compared with the galvanic current without illumination (Kim et al., 2011). When tests were conducted in hydrogenated high-temperature water, there was no increase in galvanic current when UV illumination was applied. Their results support the field observation that shadow corrosion may be a phenomenon only associated when oxidizing species (oxygen or hydrogen peroxide) may be present in the reactor water, such as the radiolysis products in BWR water. The actual mechanism of shadow corrosion is not completely understood, but it could be a direct response of irradiation on the electronic properties of the protective oxide films on the shadowing and shadowed components in the reactor.

Crud deposition on zirconium alloys Crud is a term used to designate the buildup of deposits or fouling on the OD of fuel rods in LWRs. The mass built up on the tubes is troubling since this deposit can become activated and the deposit can also be an impediment for the transfer of fission heat to the coolant (Deshon et al., 2011). In PWR the deposits are generally porous and rich in nickel iron chromium spinel that could be of the order of 75-µm thickness. In BWR reactors the crud may be mainly composed of iron, nickel, and copper oxides (Buongiorno, 2014). The OD tube deposits are mostly a consequence of corrosion in the reactor cooling system that release ionic species and particulates that are dissolved in the water and later deposit on the fuel rods (Deshon et al., 2011; Buongiorno, 2014). The crud scale is porous and may contain oriented chimneys for the steam to escape. The presence or crud may increase the waterside corrosion of the fuel cladding by a mechanism of crud-induced localized corrosion (CILC). The issue of the crud deposition in civilian

Current materials in light water reactors Chapter | 2

41

power plants has not been eliminated yet, mainly because of the lack of understanding on the buildup mechanism (Short, 2018).

Irradiation damage of zirconium alloys Reactor irradiation changes the properties of alloys and materials in the core, including the fuel and the zirconium alloy rod components. Adamson (2000) reported that the swelling in austenitic stainless steel under irradiation may be associated with the formation of voids in the metal; however, zirconium alloys do not form voids under irradiation. The irradiation damage to zirconium seems to be by the formation of dislocation loops in the base metal. Irradiation also may turn the intermetallic SPP amorphous, and therefore changing over time (and dosage) the corrosion behavior of the cladding alloy in the reactor environment. It was also reported that reactor irradiation could completely dissolve the SPP into the alpha Zircaloy matrix (Adamson, 2000; Cox, 2005). However, Cox (2005) argued that there is no evidence that either BWR or PWR reactor irradiation increases in some way the oxidation rate of zirconium. Cox argued that there is only an indirect effect of irradiation by the redistribution of alloying elements inside the alloy, for example, by dissolving the intermetallic particles (SPP) into the matrix of the alloy.

Chapter 3

Worldwide development of accident tolerant fuels, areas of study, claddings, and fuels Chapter outline Overview Accident tolerant fuels—from crisis to opportunity The events at the Fukushima nuclear power stations of March 2011 International effort to develop safer materials for nuclear power plants Timeline for the accident tolerant fuels development

43 44 45 47 49

Assessment on current accident tolerant fuels maturity concepts The accident tolerant fuels program in the United States Industrial civilian nuclear power participation in the accident tolerant fuels efforts in the United States Nuclear Energy Institute Electric Power Research Institute

51 53

56 57 60

Overview Until a few years ago, very few considered the possibility of using fuel systems in light water reactor that were not based on a zirconium alloy and uranium dioxide (UO2). All standards and guidelines developed over the last 60 years in the commercial nuclear power business are based on Zr/UO2. Any improvement over the decades was to understand and solve issues related to the performance of Zr/UO2. Now, there is a possibility for newer materials in power reactors, and regulatory bodies and trade organizations may be accepting this change as an unavoidable reality. The newer accident tolerant fuels ATF can be classified broadly into (1) newer claddings and (2) newer fuels. Each one of the cladding and fuels main groups has subgroups. In the cladding, subgroups are (1) coated zirconium alloys, (2) monolithic FeCrAl alloys, and (3) ceramic composites or SiC/SiC made of SiC fibers embedded in an SiC matrix. The subgroups in the fuels are (1) modified or doped UO2 and (2) high-density fuels including uranium nitride and uranium silicide. The most near-term concepts for implementation may include coated zircaloy cladding and modified urania fuel. The second closest implementation could be FeCrAl cladding for urania fuel. The ceramic composite Accident-Tolerant Materials for Light Water Reactor Fuels. DOI: https://doi.org/10.1016/B978-0-12-817503-3.00003-1 © 2020 Elsevier Inc. All rights reserved.

43

44

Accident-Tolerant Materials for Light Water Reactor Fuels

SiC/SiC for cladding and the higher density fuels UN and U3Si2 are concepts that may need longer term developments. In the United States (and in many other countries), the entire nuclear power industry is prepared for the changes. In the first few years of development of the ATF concepts, the regulatory agency and the commercial side of nuclear power were observant bystanders. However, in the last few years, they became fully engaged. Currently, utilities are working closely with the fuel producers to have one or more concepts installed in their commercial reactors within the next quinquennium. The supportive organizations for the utilities such as the Electric Power Research Institute and the Nuclear Energy Institute are dedicating technologists to evaluate the benefits offered by the newer ATF concepts. The regulatory body (Nuclear Regulatory Commission) is investing in personnel to work in parallel with the US Department of Energy and the fuel vendors to allow a full reload into a reactor as soon as technically possible. The funding at federal level in the United States is robust for universities, national laboratories, fuel vendors, commercial utilities, and trade organizations to be working collectively with the main goal of making nuclear power plants safer to operate and to continue having the option of using nuclear power to supplement other clean energy sources and to minimize the indiscriminate burning of fossil fuels in order to generate electricity. It is always rewarding to have variety of technological and engineering organizations working with the same goal of making nuclear power stations more robust and safer to operate. Different technologists with diverse backgrounds would look at and evaluate the same issue of increased safety in reactor operation from a different perspective. The Fukushima accident was a critical unfortunate event for nuclear power generation; however, it eventually galvanized the industry to offer solutions for the continuing use of clean nuclear energy

Accident tolerant fuels—from crisis to opportunity The concept of accident tolerant fuels (ATF) was born in 2011 in response to the unfortunate events in north east Japan at the Fukushima Daiichi nuclear power stations. During the Fukushima Daiichi accident, the lack of cooling in the reactors triggered a temperature rise above the safe design margins for the operation of zirconium alloys components in the reactors. As the temperature raised, the oxidation rate of zirconium accelerated by an autocatalytic exothermic oxidation reaction between the zirconium and the superheated water or steam in the cores, which produced large amount of ignitable hydrogen gas. Zr ðsÞ 1 2 H2 O ðl; vÞ 5555 ZrO2 ðsÞ 1 2H2 ðgÞ 1 Heat of reaction

Worldwide development of accident tolerant fuels, areas of study Chapter | 3

45

The heat of this reaction is listed as 586 kJ/mol (Electric Power Research Institute, EPRI, 2019). Initially, just after the accident, the nuclear power community around the world was in state of shock, and worldwide pessimism invaded the nuclear power business because engineers and scientists knew this unfortunate event was going to be another setback for the validation on the continuing safe use of commercial nuclear power. However, after only a short period, the scientists and engineers came to the realization that this nuclear power crisis could be turned into an opportunity for the renewal of materials used in power reactors (Chapter 2: Current materials in light water reactors. Why do we need a materials renewal?). Most of the countries with civilian nuclear power embarked on the quest of accident tolerant materials, a needed upgrade that was delayed for several decades. The nuclear materials community may have been complacent and comfortable with the materials they have been using for decades and did not feel the need to change, evolve, or innovate. The Fukushima plants disaster was a wakeup call to try something different if nuclear energy was going to be considered viable and competitive again in the world electricity market. In the 1970s and 1980s, nuclear power was flourishing due to the many international oil market crises. However, in the last 20 years, with the aggressive exploration of shale natural gas to make nations more independent from the old oil cartels, nuclear energy was struggling to be price competitive against the burning of natural gas. A major factor against the competitiveness of nuclear power was the heavy burden of regulations and safety-imposed rules when compared to the more relaxed set of guidelines applied to fossil fuel (e.g., natural gas) power stations. For example, the global raise in the threat of terrorism in the last 20 years has forced the nuclear power stations to maintain an expensive shield of security that is not obligatory to natural gas or coal plants. Moreover, nuclear power continued to have an unparalleled bad rapport in the uneducated public opinion. The public at large seem to have an irrational fear of nuclear power, probably because its use and application are not obvious and therefore uneasily grasped. Nuclear power is the safest form of energy per kilowatt-hour generated among all forms of civilian electricity (including wind and solar), but this statistic does not seem to permeate into the nations’ mainstream information systems. It is almost tragically comic that communities which are close to a nuclear power station have a much larger support for nuclear power generation than communities that do not have nuclear-based electricity produced nearby.

The events at the Fukushima nuclear power stations of March 2011 The Fukushima Daiichi nuclear power stations in north east Japan are situated in a beautiful geographical valley surrounded by hills on the north, west, and south and by the ocean to the east. The six boiling water reactors

46

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 3.1 Reactors in Fukushima preaccident.

(BWRs) of the Fukushima Daiichi site sit on a north to south line parallel to the shore at approximately 100 m from the ocean. The newest reactors Units 5 and 6 are located on the northern part of the site and reactors Units 14 are in the southern part of the site (Fig. 3.1). Seawalls were built in front of the reactors to reduce the impact of ocean waves. The six reactors of the Fukushima Daiichi power station were designed by General Electric for Tokyo Electric Power Co. Three reactors were supplied by General Electric, two reactors by Toshiba, and one reactor by Hitachi. The site was first commissioned to produce electricity in 1971 with the operation of Unit 1. Units 2, 3, 45, and 6 started operation in 1974, 1976, 1978, and 1979, respectively. In 2011 at the time of the accident, the construction of Units 7 and 8 was under consideration. Following the strong magnitude 9 Tohoku earthquake on March 11, 2011, the nuclear reactors Units 13 automatically “scrammed,” using the control rods to stop the fission reactions and to practically shut down the production of steam and electricity. At the time of the earthquake, Units 46 were in shutdown mode waiting for refueling. The earthquake generated two tsunami waves that hit the coast of northeastern Japan. The first wave was stopped by the seawalls in front of the reactor line. The second larger wave arrived approximately 1 h after the earthquake, and it was between 13 and 15 m tall, which overcame the 6 m tall seawalls, hitting the turbine buildings and disabling the emergency diesel power generators that were in the basement of the turbine buildings

Worldwide development of accident tolerant fuels, areas of study Chapter | 3

47

(Fig. 3.1). Since the six units were already disconnected from the grid due to effects of the earthquake damage, and because the diesel generators were incapacitated by the tsunami, there was a complete plant black out at the site, that is, there was no electricity to power the pumps that recirculated the water to extract the residual fission heat from the fuel bundles in the reactors and in the cooling pools. At least four of the light water reactors (LWRs) lost water recirculation in the reactor cores. The diesel generator for Unit 6 was not affected by the tsunami, since it sat on higher ground. The diesel generator of Unit 6 was able to supply cooling power to both the reactor cores of Units 5 and 6, which did not ultimately suffer hydrogen gas explosion. Even after scramming, the fuel bundles in the reactors have a residual fission decay heat that eventually would heat the water above the safe low corrosion rate range for the zircaloy elements inside the reactor (Hofmann, 1999; Powers et al., 2016). The oxidation reaction of zircaloy with water and steam is highly exothermic, and therefore the reaction progressed quickly autocatalytically generating in the order of 1000 kg of hydrogen gas inside each of the BWR pressure vessels. Due to the gas pressure build up inside the reactor pressure vessels, the hydrogen gas eventually escaped the pressure vessels and accumulated in the containment building housing the pressure vessels. The ignition of hydrogen accumulated in the containment buildings was witnessed live around the globe in the form of building explosions. The upper part of pressure vessel was not affected by the explosions, which happened above the pressure vessels. The largest explosions were in buildings housing reactor Unit 1 on March 12, 2011, and Units 3 and 4 on March 15, 2011. The building housing reactor Unit 2 had only a section or window blown off on the building wall facing the ocean. Fig. 3.2 shows the appearance of the Fukushima Daiichi site in 2018. The forest behind the line of reactors was cleared to house rows of tanks holding contaminated water recovered from underground leaks from the units affected on the site. The contaminated water in the tanks accumulates of the order of one tank per week or up to 80,000 tons per year. The water recovered under the reactors is treated to remove radioactive elements, but the cleaning is not exhaustive enough to allow yet for the release of the treated water into the ocean. Despite the severity of the loss of coolant accident at the Fukushima Daiichi power stations, only two people died on the entire site by drowning in the basement of the turbine buildings when the second tsunami wave arrived. No humans perished by exposure to radiation at the Fukushima Daiichi site or in the surrounding areas.

International effort to develop safer materials for nuclear power plants Since the natural disaster events of March 2011, the international nuclear materials community has been engaged in developing safer fuel alternatives

48

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 3.2 Reactors in Fukushima postaccident.

that can maintain their structural integrity for longer times without being water cooled inside the reactor pressure vessel (McCullum, 2019). The safer alternatives are called ATFs or advanced technology fuels (Kim et al., 2016a,b; Rebak, 2017; Sakamoto et al., 2017; Savchenko et al., 2015; Terrani, 2018). The newly proposed ATF concepts can be classified in two large groups considering the two components of the fuel rod (Fig. 3.3): 1. cladding and 2. fuel The desirable attributes of the ATF cladding are mainly high strength at high temperature (to maintain the geometry of the fueled coolable lattice) and resistance to oxidation (to avoid generation of hydrogen gas). The main desirable attributes of the ATF fuel pellets are enhanced thermal conductivity to remove faster the heat from the fuel and retention of volatile fission products to avoid polluting the environment in case of a cladding breach (Kim et al., 2016a,b). In the development of ATF, the current issues under consideration are not only the physical, chemical, and mechanical properties of the materials proposed but also the economics and safety for the continuing operation of nuclear power stations. The newer ATF systems need to be evaluated not only regarding their performance in the reactor but also in view of the entire fuel cycle, from the fabrication of the fuel assemblies (of both fuel and cladding) to the final disposition of the used fuel. In the family of ATF systems cladding concepts, three main groups exist (Chapters 46), and in the ATF fuel family concepts, there are two main concepts (Chapter 7: Alternative fuels to urania).

Worldwide development of accident tolerant fuels, areas of study Chapter | 3

Coated zirconium Alloys Cladding

49

FIGURE 3.3 Proposed accident tolerant fuels systems.

Monolithic FeCrAl alloys

SiC/SiC ceramic

Current ATF concepts

Modified or doped UO2 Fuel High density fuels (UN, U3Si2

1. Development of improved ATF cladding for the fuel a. Coated zirconium alloys b. Monolithic ironchromiumaluminum (FeCrAl) c. Silicon carbidebased composites 2. Development of improved ATF fuels a. Modified or doped UO2 fuels b. Higher density fuels, such as nitride (UN) and silicide (U3Si2) It is likely that in the near future, some of the ATF fuel concepts could be a combination of any of the three claddings with any of the two fuels listed above. In the following chapters, only the five ATF concepts above will be discussed, but the research community are also studying many other versions as well, for example, molybdenum-based cladding (Cheng et al., 2016) or metallic fuels (Hartmann et al., 2018). The common denominators of the proposed ATF fuel concepts are: (1) high thermal conductivity and (2) lower energy storing per unit mass than the current fuel system (EPRI, 2019). According to the EPRI (2019), the criteria initially developed by the civilian nuclear power industry to enable the acceptance of ATF cladding and fuel concepts included attributes such as (1) adequate neutron absorption cross sections to guarantee adequate operational and economic performance; (2) able to be manufactured to specifications and configurations appropriate for the current LWR fleet; (3) ample and inexpensive access to raw materials to satisfy the operational needs of the US and international power reactors; (4) compatibility with current LWR coolants under normal operating conditions; and (5) satisfy the existing design, operational, reliability, and licensing requirements for fuels.

Timeline for the accident tolerant fuels development Fig. 3.4 shows the most likely short-term (before 10 years) and midterm (1020 years) areas of development of ATF concepts (Carmack et al.,

50

Accident-Tolerant Materials for Light Water Reactor Fuels FIGURE 3.4 Timeline for accident tolerant fuels development.

2013). Several countries such as the United States, France, Japan, Korea, Russian Federation, and China are working in more than one of the ATF concepts outlined in Fig. 3.4. In the United States, the development of the ATF concepts is primarily driven by the fuel producers such as General Electric (GE), Framatome, and Westinghouse in cost-shared projects with the US Department of Energy (Hayes, 2019). In other countries and regions such as in Canada, Europe, and Asia, the efforts are primarily driven by government agencies or national research centers. The nearest term and evolutionary concept for ATF cladding is to use a coating (probably containing chromium) to protect the current zirconium alloy tubing to expand its temperature tolerance range to resist attack by water and steam from the current 350 C to a temperature closer to 1000 C. For example, General Electric is proposing to use their ARMOR proprietary coating concept that is aimed at providing not only resistance to attack by steam at higher temperature but also at providing wear resistance to minimize fuel rod perforation due to debris fretting under normal operation conditions (Lin et al., 2018). Framatome and Westinghouse are also pursuing coating concepts for their zirconium alloys, M5 and Zirlo, respectively (Chapter 4: ATF cladding concept: coatings for zirconium alloys). The next nearest term solution for ATF could be to use coating for the BWR Zircaloy channels to minimize the volume of hydrogen production in case of a loss of coolant accident. The channels offer a simple geometry and do not have the requirement of fretting resistance or sealed hermeticity like for the fuel rod tubes. The coating for the channels could be a simpler and more costeffective solution than for the fuel rods. Eventually, in a follow-up development, the channels in BWRs could be fabricated using nuclear grade silicon carbide composite materials. These ceramic composites for the BWR channels would not contain either boron or cobalt, and they would be resistant to dissolution in water at near 300 C. General Electric in the United States and countries such as Japan, Korea, and China are also considering using monolithic cladding of FeCrAl alloys

Worldwide development of accident tolerant fuels, areas of study Chapter | 3

51

to replace the current zirconium-based alloys. The international materials community is currently assessing the economics of FeCrAl, since it would be less transparent to neutrons than zirconium-based alloys. The apparent cladding material goal (in 10 years or more; Fig. 3.4) would be to use a nuclear grade silicon carbide composite that (1) can be sealed hermetically to avoid the release of radionuclides to the coolant, (2) to resist thermal cycles without developing microcracks that could be a path for releasing of contaminants into the coolant, and (3) to be compatible with 300 C cooling water. On the fuel side of ATF materials (Fig. 3.3), many consider that doping urania fuel would make it more effective at transferring the fission heat to the cladding and subsequently to the coolant. A faster diffusion of heat to the coolant will decrease the highest centerline temperature in the fuel, which is important for avoiding the crumbling of the fuel, which would further increase resistance to heat flow. The use of dopants may help maintaining the geometry of the fuel pellet for higher burnup conditions. The second ATF concept on the fuel family is to use higher density fuels such as uranium silicide or uranium nitride. These fuels are currently in the laboratory stage of development and not proven to be industrially viable. For example, a higher density fuel could be a good solution for a monolithic FeCrAl cladding due to the higher thermal neutron absorption of the FeCrAl cladding compared to the current zirconium alloys.

Assessment on current accident tolerant fuels maturity concepts The timeline for the development of the ATF concepts is directly tied up to the maturity in the progress of each one of the concepts (Fig. 3.3). A condensed evaluation on the maturity of the five ATF concepts, for three claddings and two fuel concepts, is presented in Chapter 8, Maturity of the accident tolerant fuel concepts: the fuel cycle and used fuel disposition, after each concept is discussed separately in Chapters 47. Most of the data cited in Chapter 8, Maturity of the accident tolerant fuel concepts: the fuel cycle and used fuel disposition, regarding the technology readiness or maturity evaluation can be referenced to the document prepared by the Expert Group on Accident Tolerant Fuels for Light Water Reactors (EGATFL) of the Nuclear Energy Agency (NEA, 2018). The NEA is part of the Organisation for Economic Co-operation and Development (OECD), which was founded in 1961 to stimulate economic progress and world trade. The OECD currently has 36 country members from around the globe (mostly developed countries), including Europe, Asia, and the United States. The NEA is part of the OECD and has 33 members including Argentina, Germany, Japan, Russia, and United States. Country members of the NEA are not necessarily

52

Accident-Tolerant Materials for Light Water Reactor Fuels

members of OECD. The mission statement of the NEA is listed as “To assist its member countries in maintaining and further developing, through international cooperation, the scientific, technological and legal bases required for a safe, environmentally sound and economical use of nuclear energy for peaceful purposes. It strives to provide authoritative assessments and to forge common understandings on key issues as input to government decisions on nuclear energy policy and to broader OECD analyses in areas such as energy and the sustainable development of low-carbon economies.” The NEA has several work areas including the one on nuclear science, which at the same time has groups of expert scientists and technologists and working parties. The groups are dedicated on collecting information and opinion on multiple areas of common concerns within NEA. One of the expert groups is the EGATFL, which was established in 2014 following two workshop meetings in 2012 and 2013. Scientists from 35 institutions in 14 member countries— Belgium, the Czech Republic, France, Germany, Japan, Korea, the Netherlands, Norway, Russia, Spain, Sweden, Switzerland, the United Kingdom, and the United States—as well as invited technical experts from China participated. The scientists and technologists of EGATFL formed three subgroups and selected leaders for each of the three subgroups. The three subgroups were (1) definitions and metrics, (2) cladding materials, and (3) fuel materials. The members of the three subgroups met approximately two times a year for approximately 5 years following the progress on ATF materials development around the globe until finally releasing their consensus statement document in September 2018 (NEA, 2018). One of the outcomes of the EGATFL document was the ranking and rating of the several ATF concepts based on their technical and industrial maturity and the technical readiness level of the concepts (NEA, 2018). The need to evaluate and rank the ATF concepts using the same set of rules or criteria was presented before by Bragg-Sitton et al. (2014) under the designation of metrics or attributes. The NEA (2018) prepared attribute qualitative charts to compare the maturity level of several ATF concepts including cladding and fuel using a set of colors describing knowledge categories (Table 3.1). The ranking of the ATF concepts was performed for several cladding and fuel concepts regarding their (1) fabrication and manufacturability including economics, (2) performance under normal operation conditions and AOOs, (3) performance under accident conditions such as design basis accidents and beyond design basis accidents, and finally (4) likely properties and performance at the end of the fuel cycle such as the possibility of repurposing the used fuel. The assessment by the NEA EGATFL was comprehensive, addressing as many combinations and permutations brought up by the country members of NEA. Chapter 8, Maturity of the accident tolerant fuel concepts: the fuel cycle and used fuel disposition, presents a simplified version of the NEA rankings. Many concepts could not be ranked by color due to the lack of information (Table 3.1).

Worldwide development of accident tolerant fuels, areas of study Chapter | 3

53

TABLE 3.1 Attributes for cladding and fuels (NEA, 2018). White color

Lack of knowledge does not allow for color identification, property not addressed

Green color

The concept is mature, data are available, and the results are positive

Yellow color

Data are available, but the results are not fully positive, and optimization may be needed

Blue color

There are no data, but the lack of data does not appear challenging

Purple color

There are no data, and this lack of data may be troubling

Gray color

A potential show stopper has been identified

The accident tolerant fuels program in the United States The first nationwide ATF program started in the United States in 2012 when the US Department of Energy (DOE) implemented a 10-year program to develop ATF concepts. The plan milestone was to have a prototype called a lead test assembly (LTA) installed in a commercial reactor by the year 2022 (Fig. 3.5; Hayes, 2019). The DOE engaged three fuel vendor companies (Westinghouse, Framatome, and Global Nuclear Fuels) through cost share contracts to develop alternative fuel rods to the traditional urania/zirconium alloy pair. The three fuel producers submitted proposals for which the research and developing cost was covered at 80% by the government (US DOE), and the industry would provide the remaining of the funds. This plan initially had three phases: (1) 4 years of Phase 1, also called a period of Feasibility studies, (2) 6 years of Phase 2 also labeled as Development and Qualification of ATF concepts, and (3) a Phase 3 of Commercialization to start after the year 2022 (Fig. 3.5). The initial goal of the DOE was to have articles of ATF concepts from the three fuel vendors inserted into commercial power plants by the year 2022 (Hayes, 2019). Each of the US fuel producers is working in several ATF concepts (for both cladding and fuel) simultaneously and advancing fast to be able to make the ATF concepts attractive to the current aging fleet of commercial LWRs. As of 2019, Framatome was working on (1) Cr-coated M5 cladding, (2) Doped UO2 for improved thermal conductivity and performance, and (3) SiC/SiC cladding (Hayes, 2019). General Electric was

54

Accident-Tolerant Materials for Light Water Reactor Fuels

Phase 2 Development/Qualificaon

Phase 1 Feasibility

Workshops

Phase 3 Commercializaon

Concept priorizaon Lead fuel rod ready

Feasibility studies on advanced fuel and clad concepts – Bench-scale fabricaon – Small-scale irradiaon tests – Steam reacons – Mechanical and chemical properes – Furnace tests – Fuel performance modeling

Steady state loop and capsule tests Transient irradiaon tests Loss of coolant accident (Loca)/Furnace tests

Assessment of new concepts – Impact on economics – Impact on fuel cycle – Impact on operaons – Impact on safety envelope – environmental impact

Industry-led projects (Phase 1a)

2012

2013

2014

Fuel performance code Fuel safety basis

Industry-led projects (Phase 1b)

2015

2016

Industry-led projects (Phase 2)

2017

2018

2019

2020

2021

2022

FIGURE 3.5 Ten-year plan.

working on (1) Coatings for Zircaloy-2 cladding, (2) Coatings for Zircaloy-2 channels, (3) Monolithic FeCrAl alloys for cladding, (4) Nuclear grade SiC/SiC for channels, and (e) Nuclear grade SiC/SiC for claddings. Westinghouse was working on (1) Coated Zirlo cladding, (2) SiC/SiC composite cladding, and (3) Higher density uranium silicide fuel with improved thermal conductivity (Hayes, 2019). The initial milestone of the 10-year plan by the US DOE was to have an ATF article inserted into a commercial power plant by the year 2022 (Fig. 3.5). General Electric was the first fuel vendor who succeeded in this goal by inserting ATF articles into the Hatch Unit 1 reactor in Georgia in February 2018, that is, 4 years ahead of original schedule from DOE (Nuclear News, 2018). General Electric and Global Nuclear Fuels installed two concepts of ATF into the Hatch-1 reactor, namely, ARMOR or coated zirconium alloy cladding and IronClad or FeCrAl thin wall monolithic cladding. Initial results from this installation should be available during poolside inspection and harvesting of exposed material during the following plant outage or fuel cycle of 24 months in February 2020 (Nuclear News, 2018). In 2019 all three fuel vendors (Framatome, General Electric, and Westinghouse) have plans to insert articles into commercial power reactors, which will also be ahead of schedule of the initial DOE plan for 2022. In 2019 Framatome is planning to have chromia-doped UO2 fuel in chromiumcoated M5 cladding in the pressurized water reactors (PWR) Southern Nuclear Vogtle plant in Georgia; General Electric is planning to have IronClad and ARMOR components in the BWR Exelon Clinton plant in Illinois; and Westinghouse is planning to have chromium-coated Zirlo and U3Si2 fuel in the PWR Exelon Byron plant in Illinois (Hayes, 2019).

Worldwide development of accident tolerant fuels, areas of study Chapter | 3

55

In the first few years of the DOE 10-year long plan, only DOE and the three fuel companies were engaged in research and reporting progress at workshops and integration technical meetings. However, in approximately 2016, the US Nuclear Regulatory Commission (NRC), the Nuclear Energy Institute (NEI), the EPRI, and more than one utility (e.g., Exelon Generation and Southern Nuclear) also became fully involved in the development of ATF, creating an engaged community that meets, interacts, and exchanges information several times a year. Fig. 3.5 shows that the initial goal of the DOE Office of Nuclear Energy was to have ATF concept articles inserted into a commercial reactor by 2022. However, the three fuel producers in the United States are planning to have ATF articles in commercial reactors by 2019 or 3 years ahead of the original schedule (Hayes, 2019). It has currently been reassessed that Phase 2 of Development and Qualification, which started in 2017 to continue for 6 years, may last until the year 2026 for a total of 10 years (Hayes, 2019). During the extended Phase 2 decade, the ATF concepts will be evaluated under normal and off-normal conditions in test reactors such as Advanced Test Reactor (ATR) in Idaho National Laboratory (INL) and in the commercial plants (Hatch, Clinton, Byron, and Vogtle). During Phase 2, the fuel vendors will collect and analyze performance data, develop fuel performance codes, build reload manufacturing capabilities, and design and fabricate batch reloads. In parallel, during Phase 2, the fuel vendors will periodically interact with the NRC to obtain licenses for partial core reloads, qualify fuel performance attributes, qualify fuel fabrication processes, and facilities and license transportation issues to the commercial plants. Similarly during Phase 2, the fuel vendors will also engage trade associations such as the NEI and the EPRI to strengthen the business case for ATF, to establish and reinforce the additional safety benefits, and to assess the commercial viability in the prevailing electricity distribution market (Hayes, 2019). One of the most important features of Phase 2 is the acquisition of data on the behavior of the ATF concept under representative irradiation conditions. There are two irradiation studies being conducted at the ATR of the INL called ATF-1 and ATF-2. In the ATF-1 study, short rods of approximately 15 cm in length (named rodlets) and comprising of a cladding and fuel are being neutron irradiated in presence of dry helium inside capsules made of type 316 L SS. The results from these tests are going to establish if the rodlets maintain their geometrical stability in a neutron irradiation field, and to determine whether there is a chemical or mechanical interaction between the fuel pellets and the fission products with the internal diameter (ID) surfaces of the claddings. In the ATF-2 study, the rodlets will also be irradiated at the ATR but exposed directly to a water loop that mimics the water chemistry condition of a PWR operation condition. That is, in addition to the conditions of the ATF-1 test, the ATF-2 tests will include the environmental effect of the coolant on the outer diameter (OD) of the rodlets under a neutron irradiation field (Hayes, 2019). DOE and fuel vendors are also engaged in the testing of ATF

56

Accident-Tolerant Materials for Light Water Reactor Fuels

concepts in the TREAT reactor, also located at INL, which was designed to perform fast transient tests called reactivity initiated accidents (RIAs). One of the outcomes of the RIA TREAT tests will be to measure the total enthalpy absorption by the fuel rodlets before their bursting; therefore these tests will determine the extended margin of safety of the ATF concepts when compared to the traditional zirconium/urania fuel concepts (Hayes, 2019). During Phase 2, the fuel vendors and their utility partners will also be able to remove from commercial power reactors their ATF articles and poolside examine them after one or two cycles after insertion to these commercial reactors. After poolside inspection, the removed articles may be shipped to hot cells across the country for further evaluation. The first removal of articles from a commercial utility is expected to be performed by General Electric and Southern Nuclear from the Hatch-1 plant in the Spring 2020. On January 31, 2019, the US DOE Office of Nuclear Energy (DOE-NE) announced the distribution of 111 million dollars to the three fuel producers to continue the fuel development work in Phase 2. The mandate from the DOE-NE was that the fuel vendors would (1) confirm that an ATF article concept or an LTA is installed in a US commercial power plant; (2) confirm that prototypic pin segments or rodlets have been installed in the INL ATR’s water loop (ATF-2 Test Plan); (3) engage the US NRC on licensing efforts for their ATF concepts that include the involvement of at least one nuclear power plant owner/operator per ATF concept; and (4) have licensing plans from the US NRC for partial core reloading by the mid-2020s (DOE-NE, 2019). The 5-year term of the Commercialization Phase 3 would probably start in 2027 and extend until 2031 (Hayes, 2019), in which the industrial manufacturing capabilities may be expanded through the development of a qualified supply chain, and a competitive price marketing. In parallel, the fuel vendors will be working with the NRC to apply for licensing of full core reloads utilizing the proposed ATF concepts.

Industrial civilian nuclear power participation in the accident tolerant fuels efforts in the United States Although the main financial effort to develop ATF concepts and materials is from the federal government through the DOE, the entire nuclear community is engaged through the missions of their own organizations. There are two US nuclear energy trade organizations that are fully engaged in the development of ATFs: (1) the NEI and (2) the EPRI. Other advocacy groups include (1) Nuclear Matters, which has the mission to inform the public and legislators on the benefits of nuclear produced electricity and (2) the American Nuclear Society, which has mainly the objective of education and dissemination of information pertaining nuclear science and technology.

Worldwide development of accident tolerant fuels, areas of study Chapter | 3

57

Nuclear Energy Institute The NEI represents the nuclear power industry interests before the US Congress and regulatory entities such as the NRC and other national agencies as well as international organizations. The NEI also provides a forum for the nuclear power industry to resolve technical and business issues and works to deliver a unified voice to impact public policy concerning nuclear power. In their webpage, the NEI advocates that nuclear energy is carbon free, it is available 24 h a day 7 days a week, and it is vital to a future of clean energy. Currently, there are 98 nuclear power reactors in the United States, which deliver approximately 20% of the electricity to the US grid, becoming the largest sector of clean or carbon-free source of electricity in the country (Fig. 3.6). More than 50% of the clean electricity consumed by the United States has nuclear sources, which is a complement to renewable energy sources that are not always available. The NEI is engaged in a campaign to avoid the premature shutdown and decommissioning of nuclear power stations. In the last 10 years, the number of nuclear reactors in the United States decreased from 105 to 98, mainly due to economic issues. The cost of operation for nuclear power plants is mainly due to the burden imposed regarding security and safety, which is much higher than the burdens imposed to natural gas or coal plants. To avoid the risk of early closure of more nuclear stations, which may result in an increase in the emission of greenhouse gases, in 2016, the states of New York and Illinois reclassified the status nuclear energy as carbon-free or zero emission to offer protection to nuclear power generating stations. This protection by policy makers was followed by Connecticut and New Jersey in 2017 and 2018, respectively. Efforts to recognize the value of nuclear power are currently being discussed in Ohio and Pennsylvania. Another reason to avoid nuclear reactor shutdown is the decrease in fuel diversity, which may derive into price volatility and 2017 Clean Energy Share in the United States

Nuclear

Hydro

Wind

Solar

Geothermal

FIGURE 3.6 Clean energy.

58

Accident-Tolerant Materials for Light Water Reactor Fuels

harm local communities that depend on taxes paid by the nuclear utilities and on jobs for the highly trained and skilled work force who operate the reactors. One of the core critical issues behind the survival of nuclear energy as a civilian source of electricity in the United States is the overcoming of the prevalent fear associated with it, to overcome the public perception that nuclear technology is not safe, added to the disbelief that nuclear is a needed source or energy. With the stepwise evolution of nuclear fuel technology using the ATF concepts under consideration by the DOE-NE-supported fuel vendors, it may be possible to instill into the technical community and the public the idea that nuclear energy is not only clean but it is also the safest type of energy. There are two issues that will make nuclear energy accepted: (1) the public perception about its safe technology needs to change and (2) it must be cost competitive compared to the burning of natural gas. Both issues can be addressed by the currently discussed ATF concepts. The reduction in costs may come, for example, (1) by allowing the nuclear fuel to run for longer periods between refueling (increased burnup), (2) by increasing fuel enrichment, (3) by reducing the amount of used or spent fuel generated by the power plants, and (4) by reducing overall maintenance and operational costs of the nuclear power stations. The NEI estimates that the plants operational costs can be reduced by 20 % by 2030 using the ATF concepts. The plant operation savings may encourage the reactor owners to request plant licenses extension to 80 years, which will take the current US nuclear power stations to operate beyond 2050. The NEI assessment is that the ATFrealized benefits will be plant specific; however, in general, the following ATF benefits may be anticipated: (1) fuel cycle optimization, (2) improved operational flexibility, (3) improved coping times, (4) enhanced fuel performance, and (5) enhanced fuel reliability (Holtzman, 2019). The NEI has published a timeline on the ATF milestones implementation of the ATF concepts by the DOE-NE-supported fuel vendors (Table 3.2; Holtzman, 2019). The most critical initial milestones are the insertion of fuel rod articles or LTAs into operating reactors to obtain undisputable data directly from operating power reactors in actual reactor environments. These articles are going to be of partial or full-length rods as parts or components of standard bundles in the reactors of power generating plants. It is planned that by the end of 2019, the three US fuel vendors will have ATF components inserted in probably four active operation nuclear reactor cores. Some of the bundles may be removed in the following fueling cycle for poolside inspection and/or for post irradiation examination in hot cells. The NEI perceives that the coated zirconium alloy concept may be the less problematic for NRC regulatory approval, since it implies the addition of a 10 µm corrosion or wear protection layer on top of the current monolithic zirconium alloy tube. If the zirconium alloy tube wall is currently 0.8 mm thick, the added coating protection coating would represent only

Worldwide development of accident tolerant fuels, areas of study Chapter | 3

59

TABLE 3.2 The Nuclear Energy Institute milestones for accident tolerant fuels (Holtzman, 2019). Year

Fuel vendor

Achievement

2018

GE/GNF

Insertion of articles of IronClad and ARMOR into Southern Nuclear Hatch-1 Station in Georgia

2019

Framatome

Insertion of articles of chromia-doped urania fuel, and chromium-coated M5 cladding into Southern Nuclear Vogtle Station in Georgia

GE/GNF

Insertion of articles of IronClad cladding and ARMOR-coated Zircaloy-2 cladding for urania into Exelon Clinton Station in Illinois

Westinghouse

Insertion of articles of chromia- and alumina-doped urania fuel and chromium-coated Zirlo cladding into Exelon Byron Station in Illinois

Framatome

Insertion of longer term developmental concepts such as SiC cladding and chromia-doped urania

Westinghouse

Insertion of longer term developmental concepts such as chromia- and alumina-doped urania and U3Si2 high uranium density fuel

Westinghouse 1 General Atomics

Insertion of longer term developmental concepts such as SiC cladding

2023

All fuel vendors

First batch reloads

2026

All fuel vendors

Full core loads

2022

GE, general electric; GNF, global nuclear fuels

approximately 1% of the total cladding thickness (Holtzman, 2019). All the required functions of the cladding such as structural supporting its own weight and be able to withstand the internal pressure in the fuel cavity will be performed by the underlying zirconium alloy (like in the traditional noncoated system). The 10 µm coating will not substantially add to any increase in mechanical properties, since the coating will offer only an external fretting and/or corrosion resistance protection. Many may consider this coating not different from a typical crud build up on the monolithic zirconium alloy after some years of service in the reactor. The NEI envisions the interaction with NRC for regulatory approval to be performed in parallel with the LTA testing in the commercial reactors. That is, the NRC will be informed by the fuel vendors as the results from LTA insertions become available; therefore when the vendors License Amendment Request becomes available to the engineers and scientists of the NRC for review, the data would not be entirely new or unexpected.

60

Accident-Tolerant Materials for Light Water Reactor Fuels

Electric Power Research Institute EPRI is a nonprofit organization dedicated to research of generation, delivery, and use of all sources of electricity. Therefore EPRI has an office dedicated to exploring the behavior of materials used in the generation of nuclear power, which includes the Fuels Reliability Program (FRP). The FRP deals with the performance, safety, and sustainability of fuels in the reactors (EPRI, 2019). Approximately 50 billion US dollars’ worth fuel are currently installed in reactors worldwide; therefore maintaining the integrity and performance of the fuels is important for the utilities since the cost of fuel accounts for approximately 25% of the total plant cost, which includes capital, fuel, maintenance, and operations. One of the tasks for the EPRI FRP is to monitor the failure modes of fuels in both BWR and PWR plants around the world and to help the industry (utilities) address the issues leading to reported failures. Figs. 3.7 and 3.8 show the type of failures experienced by LWR fuels in the 10-year period between 2009 and 2018 (Mervin, 2019). For BWR reactors, the number of cycles shutdown was practically the same in the first 5-year period (22 shutdowns) than in the last 5-year period (21 shutdowns). In both periods, the overwhelming reason for shutdown ( . 90%) was debrisrelated failure, with 20 shutdowns in each period of 5 years. Overall, in the last 10 years (200918), there was no change in the failure rate of BWR fuel rods. For PWR reactors, the total number of cycles shutdown was 58 in the 200913 period, and it decreased by more than 50% for the period 201418 at 21 total failures. In the 200913 period, the dominant mode of failures for PWR fuels were grid to rod fretting, followed by debris and fabrication related, which accounted for 86% of all failures. In the most recent period of 201418, the dominant mode of failure for PWR fuel was debris related, followed by grid to rod fretting. For both BWR- and PWR-type reactors, the current prevalent failure mode of fuel is connected to the presence of solid debris in the coolant. Therefore in the realm of ATF research and development, it may seem imperative that a reliable ATF fuel should have debris failure resistance under normal operation conditions. EPRI is currently fully engaged in the development of ATF concepts across countries (Csontos, 2019). EPRI produced several risk-informed documents in support of the newer materials, mainly in the areas of economic and safety benefits for anticipated operational occurrences (AOO), design based accidents (DBA), and beyond design base accidents (BDBA) scenarios. They outlined the business case for the utilities by balancing the costs and the benefits upon adopting the ATF concepts into the utilities (Csontos, 2019). On the ATF cost side, EPRI envisions the cost for research and development, the cost for licensing the newer ATF concept, and the cost by utility to implement newer ATF concepts. On the

Worldwide development of accident tolerant fuels, areas of study Chapter | 3 2009–13 Global BWR cycles shutdown with failures Debris

Grid to rod fretting

Baffle related

Fabrication

Duty related

Unknown

61

FIGURE 3.7 Boiling water reactor fuel failures.

2

20

2014–18 Global BWR cycles shutdown with failures Debris

Grid to rod fretting

Baffle related

Fabrication

Duty related

Unknown

1

20

benefits side, engineers at EPRI see an improvement in the fuel cycle economics, an increase in the operational flexibility of the plant, and an increase in the fuel reliability. The incorporation of the ATF concepts into current LWRs may result in an increase in coping time from 1 to 3 h in the case of a severe accident (Csontos, 2019). The use of ATF concepts would provide a safety margin improvement by delaying core damage and may also provide a path for core restoration following a severe accident, since the fuel could retain their coolable geometry for longer periods allowing for quenching measures. EPRI is envisioning that the use of ATF will empower the overall

62

Accident-Tolerant Materials for Light Water Reactor Fuels 2009–13 Global PWR cycles shutdown with failures

Debris

Grid to rod fretting

Baffle related

Fabrication

Duty related

Unknown

FIGURE 3.8 PWR fuel failures.

5

17 14

3 19

2014–16 Global PWR cycles shutdown with failures

Debris

Grid to rod fretting

Baffle related

Fabrication

Duty related

Unknown

1 3

4 13

capability of nuclear power plants by a fuel cycle optimization process (Csontos, 2019). The improved fuel cycle efficiencies include (1) improved cladding performance, (2) likely increased burnups, (3) may enable increased enrichments, (4) will potentially lead to longer cycle lengths, and (5) will reduce nuclear waste generation. It is envisioned that with ATF concepts, the fuel burnup may be increased from 62 to 75 GWD/MTU.

Chapter 4

Accident-tolerant fuels cladding concept: coatings for zirconium alloys Chapter Outline Overview Introduction to the use of zirconium alloys as cladding for nuclear fuels in light water reactors Why do we consider coatings for accident tolerant fuel zirconium alloys? Oxidation protection of coatings for zirconium alloys Family of candidate coatings for zirconium alloys Ceramic coatings Chromium coatings for zirconium alloys in the French ATF program

63

64

65 67 68 69 71

Aluminum-based and iron chromium aluminum coatings for zirconium alloys Silicon-based coatings for zirconium alloys Fabrication and implementation of zirconium-coated rods Performance of coated zirconium under reactor normal operation conditions Performance of coated zirconium under accident conditions Coated zirconium irradiation behavior Coated zirconium licensing for reactor use

72 74 75

76 77 79 80

Overview The use of coatings to enhance the performance of zirconium alloys in nuclear reactors is considered the most logical evolutionary and near-term solution in the realm of accident-tolerant fuels (ATFs). Because the coating will provide protection by adding a layer in the order of 10-µm thickness to the current zirconium alloy wall, it is deemed that the coating solution may have the fewest licensing barriers for reactor use. It has been demonstrated that chromium coating by a physical vapor deposition process may be the most likely alternative. Chromium will not be a new element in the reactor core because traditional legacy alloys such as austenitic stainless steel type 304 and nickel-based alloy 600 rely on chromium for their corrosion resistance in the reactor environment. Chromium coating will not negatively impact the mechanical behavior of the zirconium alloy cladding and the Accident-Tolerant Materials for Light Water Reactor Fuels. DOI: https://doi.org/10.1016/B978-0-12-817503-3.00004-3 © 2020 Elsevier Inc. All rights reserved.

63

64

Accident-Tolerant Materials for Light Water Reactor Fuels

neutronic balance impact will be minimal. As for the other ATF concepts, the coating of zirconium alloy tubes needs to perform well under normal operation condition and under accident conditions. Many types of coatings were rated for performance in the normal operation and accident conditions and the highest ranking was for chromium-based coatings. The use of chromium coating is meant to provide short-term additional protection to the zirconium alloy cladding in the temperature range up to 1100 C. The top temperature for stability of Cr2O3 in superheated steam is approximately 1200 C. The presence of Cr2O3 on the surface of the cladding coating prevents not only oxygen access but also hydrogen diffusion into the cladding zirconium substrate. Full-length zirconium alloy cladding tubing can be coated in an efficient manner in an industrial production setting. Chromiumcoated zirconium alloy fuel rods are currently being evaluated in BWR and PWR commercial nuclear power plant environments. Supporting comprehensive reports for the chromium-coated concepts have been delivered to the Nuclear Regulatory Commission to increase their readiness to receive licensing topical reports from fuel vendors. It is expected that chromium-coated zirconium alloy cladding will be the first ATF concept to receive regulatory approval.

Introduction to the use of zirconium alloys as cladding for nuclear fuels in light water reactors Zirconium alloys were selected in the late 1940s to house the urania fuel in light water nuclear reactors for the US Navy submarines basically because of its high transparency to thermal neutrons and its relatively lower density compared to the also considered austenitic stainless steels (Rickover et al., 1975; Terrani, 2018). The use of zirconium alloys has been transplanted from submarine to civilian power generation in the late 1950s and have been used successfully in light water power reactors for over six decades, mainly because their corrosion and environmental degradation issues have been progressively understood, managed, and controlled over the years by the nuclear materials community (Motta et al., 2015). The fuel rods in light water reactors consist of zirconium alloy tubes of approximately 4-m length and 10-mm outer diameter with 0.6-mm wall thickness filled with 10-mm tall urania pellets as the fuel (Tang et al., 2017). The combination of the zirconium alloy tubing and the urania fuel pellets filling the ID of the tubing is called a fuel rod. Rods are assembled in bundles of approximately 100 rods each. The outer surface of the zirconium alloy rods is in contact with cooling water at approximately 300 C, which extracts the nuclear fission heat from the rods. Small additions of alloying elements (sometimes less than 0.5%) to zirconium made the alloys withstand the carefully monitored water chemistry under normal operation conditions of the light water reactors. The most common alloying elements for zirconium are tin (Sn) in

Accident-tolerant fuels cladding concept Chapter | 4

65

the United States and niobium (Nb) in Canada and Russia. The oldest continuously used zirconium alloys are Zircaloy-2 and Zircaloy-4. In the last 30 years, the newer alloys M5 and Zirlo were developed for the higher temperature of operation of the PWR cores (Motta et al., 2015; Tang et al., 2017). The most common zirconium alloys for nuclear fuel in Russia are E110 and E635 (Table 2.2). The reaction of zirconium with water in a nuclear reactor proceeds according to this equation (IAEA, 1998; Motta et al., 2015; Erbacher and Leistikow, 1987): Zr 1 2H2 O-ZrO2 1 2H2 1 Heat of Reaction The exothermic heat of reaction was reported 965 kJ/mol at 360 C (Motta et al., 2015), 2586 kJ/mol for 600 C 1500 C (Erbacher and Leistikow, 1987) and 584.5 kJ/mol at 1200 C (Tang et al., 2017). The reaction of zirconium with water produces an n-type oxide, and the growth of the oxide happens at the oxide/metal interface by diffusion of oxygen atoms via a vacancy mechanism through the oxide and reacting with the metal. This reaction liberates electrons that travel to the oxide surface, which may reduce protons from the water into atomic hydrogen and eventually produce molecular hydrogen gas. At the same time some hydrogen atoms from the surface may diffuse into the alloy reacting with Zr metal and producing hydrides (Motta et al., 2015; Tang et al., 2017). Zirconium and its alloys are not very stable in air and steam environments at temperatures higher than 800 C because they oxidize rapidly in an exothermic reaction producing zirconium oxide and combustible hydrogen gas (Motta et al., 2015; Erbacher and Leistikow, 1987). Regulators have established that the maximum oxidation allowed to the zirconium alloy is 15% (in Japan) to 17% (in the United States) of the tube wall thickness (Erbacher and Leistikow, 1987; Hache and Chung, 2001). This is generally known as the permissible equivalent cladding reacted (ECR). It was reported that to consume 17% of the zirconium alloy tube wall in steam, it will take 20 min at 1100 C, 8 min at 1200 C, and 4 min at 1300 C (Erbacher and Leistikow, 1987).

Why do we consider coatings for accident tolerant fuel zirconium alloys? As described in previous sections, zirconium alloys have performed rather well for over 60 years in light water reactors and the failures of the rods in the reactors in the last couple of decades were controlled to near zero (EPRI, 2019) (Chapter 3: Worldwide development of accident-tolerant fuels, areas of study, claddings, and fuels). However, it has been shown repeatedly that the temperature range for good oxidation and mechanical resistance of zirconium alloys is limited. As the temperature increases to approximately

66

Accident-Tolerant Materials for Light Water Reactor Fuels

700 C 1000 C, in conditions called design-basis accident (DBA) and beyond DBA (BDBA) conditions, the zirconium alloys initially suffer a rapid physical damage (such as permanent plastic deformation or ballooning) and eventually a chemical degradation (accelerated oxidation and generation of hydrogen gas), which may lead to a loss of the coolable geometry of the zirconium clad fuel bundles (Ocken, 1980; Erbacher and Leistikow, 1987; Hache and Chung, 2001; Terrani, 2018). The peak cladding temperature or the maximum allowed temperature for zirconium alloys in the reactor is 1204 C. The question here is how to have Zr alloys perform acceptably in DBA and BDBA conditions for longer periods. The simplest and most elegant evolutionary concept is to use a thin coating (e.g., ,10 µm thickness) or surface modification to protect the surface of the zirconium from the environment and to increase the useful time of the fuel rod (cladding) before it loses its coolable geometry (NNL, 2018; NEA, 2018). In the temperature range 1000 C 1200 C, the rate of oxidation of Zircaloy-4 in steam is almost two orders of magnitude higher than that of the oxidation of type 310SS, which relies on chromia (Cr2O3) for protection against attack by steam (Terrani, 2018). Therefore it may seem obvious that the first line of defense against severe accidents would be to protect the zirconium alloy of the fuel cladding using high-temperature oxidation-resistant coatings such a chromium, which will create a surface film of chromia for protection of the zirconium cladding. Chromium is an element well known to nuclear materials engineers because it has been used for decades in the protection of austenitic stainless steels (e.g., type 304 SS) and nickel alloys such as Alloy 600 and X-750. In the case of using a 10-µm-thick coating for the zirconium alloys, the mechanical integrity of the rod and the access to neutrons will be provided by the underlying zirconium alloy tubes, like in the current use, while the external coating will protect against oxidation and hydrogen uptake by the substrate. The use of coatings for current zirconium alloys seem the most near-term evolutionary solution to increase the safety and reliability of power reactors’ operation, at least under normal operation and in the design-based accident temperature range. Coatings can also provide protection of zirconium alloy tubes from fretting, the most common current failures of claddings (Chapter 3: Worldwide development of accident-tolerant fuels, areas of study, claddings, and fuels) (EPRI, 2019; Schneider et al., 2018). The most common waterside coatings will form a protective chromium oxide on the surface as a defense for further oxidation. This has been proved by 60 years of excellent performance of austenitic stainless steels and nickel-based alloys, which resist general corrosion in water and steam via the formation of a chromia film on the surface. Other coatings for zirconium alloys that may rely on corrosion protection by the formation of alumina or silica in superheated steam have also been proposed (Tang et al., 2017). However, silica and alumina are not stable in hightemperature water when compared with chromia.

Accident-tolerant fuels cladding concept Chapter | 4

67

Oxidation protection of coatings for zirconium alloys When metals such as zirconium, chromium, aluminum, and silicon react with oxygen or water, they develop oxides on the surface, which tend to slow down the degradation of the metals because the oxide on the surface acts as a barrier to prevent further attack from the environment. The protectiveness of the oxides on the surface of the metals depends on many variables such as the environment (e.g., temperature, presence of water vapor, oxygen) and the intrinsic electronic properties of the oxides. Terrani (2018) discussed comparatively the parabolic rate constants as a function of the temperature for the oxidation in steam of material candidates for accidenttolerant fuel (ATF) claddings. Table 4.1 shows the parabolic rate constants for oxidation of selected materials in steam at 1200 C (Terrani, 2018). It is evident that if the zirconium alloy is coated with a compound that will form a chromia protective film on the surface, the overall oxidation rate at 1200 C in steam would be approximately 100 times lower than for noncoated zirconium alloy. In the case of a compound that will form an alumina film on the surface, the decrease in the oxidation rate would be more than 300 times lower. Table 4.1 shows the benefits of the use of coatings on the current zirconium alloys to slow down degradation, mainly for design-basis accident conditions. The use of coating will also increase the coping time for reactor operators to allow for the flooding or quenching by the with fresh water in case of a beyond design-basis or severe accident situation. The attractiveness of thin coatings to protect zirconium alloys in nuclear reactors is that the neutronic impact in the reactor will be practically nil (Younker and Fratoni, 2016) and the coating may provide additional benefits such as resistance to fretting (Gray et al., 2007) and improved heat removal rate from the cladding wall by the coolant (Terrani, 2018). However, it is critical that the coating would be adherent during thermal cycles and be nonreactive with the underlying zirconium substrate. Several factors control the

TABLE 4.1 Parabolic oxidation rate. Material

Kp (g/cm2 s0.5)

Oxide for protection on the surface

Ratio of parabolic rate constant metal oxide/zirconia

Zircaloy-4

6 3 1024

Zirconia

1

25

Chromia

0.01666

26

310SS

1 3 10

APMT

2 3 10

Alumina

0.00333

SiC

4 3 1027

Silica

0.00066

68

Accident-Tolerant Materials for Light Water Reactor Fuels

oxidation resistance of the coating, including the coating composition and microstructure, the coating thickness, and the adhesion of the coating to the substrate (Meng et al., 2019). It was reported in the United States that three fuel vendors are currently working on the development of coatings for zirconium alloys (Geelhood and Luscher, 2019). For example, Westinghouse is working on applying a 20- to 30-µm-thick chromium coating on Zirlo by cold spray, Framatome is developing an 8- to 22-µm chromium coating on M5 using a physical vapor deposition (PVD) process, and Global Nuclear Fuels has a proprietary coating named ARMOR, which is being applied on Zircaloy-2 (Geelhood and Luscher, 2019).

Family of candidate coatings for zirconium alloys The oxidation performance of coatings for zirconium alloys must be evaluated not only under accident conditions but also under normal operation conditions of light water power reactors (Tang et al., 2017). Even though the coatings are designed to withstand accident conditions, they also need to provide fair protection in near-pure hydrogenated water at approximately 300 C (also called subcritical water). Several materials have been suggested for coatings of zirconium alloys, which can be divided into two large groups, 1. Ceramic coatings, and 2. Metallic coatings. The metallic coatings can be grouped by the type of oxide they would develop at high temperature for oxidation protection of the zirconium substrate (Table 4.1). Surface modification by ion implantation (such as carbon, nitrogen or oxygen) of zirconium alloys has been considered as well. The depth of surface modification is in general less than 1 µm. This implantation concept may provide some improvement to wear resistance and to lower the corrosion rate of zirconium in subcritical water, but they perform poorly in high-temperature steam (Tang et al., 2017). Tang et al. (2017) outlined clearly the conditions that a coating need to meet to be relevant for nuclear reactor applications. Some of the conditions include (1) affordable cost for a full-length tube (near 5-m long), (2) coating needs to be manufactured at temperatures low enough so that the microstructure and mechanical properties of the zirconium alloy substrate are not affected, (3) the coating does not contribute to the neutron penalty, and (4) the coating needs to be protective under normal operation condensed water conditions and in high-temperature steam typical of loss of coolant environments. General Electric reported that it is working on the development of a proprietary coating called ARMOR, which would reduce failures by fretting and provide corrosion resistance not only under normal operation conditions but also under loss of coolant accident (LOCA) DBA scenarios (Lin et al.,

Accident-tolerant fuels cladding concept Chapter | 4

69

2018). The fully dense, homogeneous and adherent ARMOR coating is applied in the entire external wall of the standard GE Zircaloy-2 tube cladding. The ARMOR coating is thermally compatible with the underlying Zircaloy-2 cladding because temperature cycles between 350 C and ambienttemperature quenching did not show delamination of the coating (Lin et al., 2018). Sliding and impinging fretting tests for up to a million cycles conducted at an ambient temperature between stainless steels wires and ARMOR-coated Zircaloy-2 and noncoated Zircaloy-2 tubes showed that the ARMOR-coated suffered smaller wear depth compared with the noncoated Zircaloy-2 tubes by a factor of three.

Ceramic coatings The currently proposed ceramic coatings for zirconium alloys include (1) MAX phases and (2) nitrides. One of the coatings considered to protect zirconium alloys are MAX phases, which are generally carbides or nitrides of a transition metal M (such as Ti, Cr, Zr), an A-group element (such as Al or Si), and carbon or nitrogen elements X with the general formula of Mn11AXn. One example of MAX phase compound is Ti3AlC2, which can be applied using thermal spray (Maier et al., 2015). Other investigated MAX phases include compounds such as Ti2AlC, Cr2AlC, Zr2AlC, and Zr2SiC (NEA, 2018). Ti2AlC offers low density (4.11 g/cm3), high thermal shock resistance, and does not spall during thermal cycling (Basu et al., 2012). The most attractive MAX phases for zirconium ATF coating are those containing aluminum. MAX phases possess properties of metals and properties of ceramics and are excellent electrical and thermal conductors (Basu et al., 2012; Tallman et al., 2013, 2015). The MAX phases coating thickness may be in the order of 20 µm. The desirable properties of MAX phases are their resistance to environmental attack, creep resistance, and low thermal expansion coefficient (Younker and Fratoni, 2016). MAX phases were found attractive for zirconium coating application in demanding nuclear applications owing to their versatile manufacturability and machinability (Tallman et al., 2015). One of the challenges in the use of MAX phases as coatings is to maintain a strict stoichiometry to deliver the desired properties because owing to the ternary nature of the compounds it is possible that unwanted compounds may form under other mass ratios of the elements (e.g., Ti, Al, C) (NEA, 2018). Another ceramic coating concept for zirconium alloys are nitride compounds. In the industry at large (not only in the nuclear sector), nitrides are generally used to modify the surface of metals and alloys, making them more resistant to wear and erosion. One of the most popular nitride coating is titanium nitride (TiN). In some industries, direct nitridation of the metallic substrate (such as a stainless steel) is performed, in a process that may be called surface modification. The nitridation of the external layer of a

70

Accident-Tolerant Materials for Light Water Reactor Fuels

component makes it more resistant to wear, but it may decrease the corrosion resistance of the component. Meng et al. (2019) used multiarc ion plating to deposit a 13-µm-thick CrN layer on zirconium alloy Zircadyne R60702 and studied the behavior of the coating for 1 h in air from 760 C to 1160 C and comparing the oxidation behavior with noncoated zirconium alloy for the temperature range 360 C 1160 C. For the noncoated zirconium alloy, they reported a compact oxide on the surface between 360 C and 760 C; however, for temperatures of 860 C and higher, the oxide in the noncoated zirconium alloy developed cracks owing to excessive growth. In the higher temperature range (860 C 2 1160 C), the surface compounds consisted of an external cracked layer of ZrO2 and a loose and porous internal layer of ZrN. This behavior is also reflected in the mass increase of the tested coupons. The mass gain was gradual and below 5 mg/cm2 up to 760 C, but it increased rapidly above 860 C nearing 85 mg/cm2 at 1160 C. One-hour exposure of the noncoated Zr alloy to air at 1160 C practically consumed the tested specimen (Meng et al., 2019). For the coated Zr specimens, the oxides developed on the surface remained intact and free of cracks, spalling, or deformation in the entire tested temperature range of 760 C 1160 C. The oxidation resistance of the coated Zr coupons was also reflected in the mass change, which was always below 2 mg/cm2 in the entire tested temperature range of 760 C 1160 C (Meng et al., 2019). At 860 C the Cr2O3 film on the surface of the coating was 0.4-µm thick and it became thicker at higher temperatures by diffusion of Cr from the CrN coating to the surface and reacting with oxygen in the environment. At 960 C, 1060 C, and 1160 C, the oxide thickness was 0.6, 2, and 4 µm, respectively (Meng et al., 2019). No interdiffusion between the CrN and the zirconium substrate was observed even at the highest tested temperature of 1160 C. Similarly, no spallation of the CrN coating occurred at the metal/coating interface. Daub et al. (2015) studied the suitability of coating Zircaloy-4 using CrN, TiAlN, and AlCrN. The coatings were produced using an arc evaporation or PVD, and the resulting thicknesses varied between 2 and 4 µm of nano-sized columnar grains. They tested the behavior of the coated zirconium alloy in water at 300 C and 350 C and in steam at 1100 C. The 300 C water exposures for 30 days showed a beneficial effect of the coatings on corrosion and hydrogen ingress into the alloy. The best overall performance was with the CrN coating. For the steam tests at 1000 C 2 1100 C for 15 min, the coated coupons had a better corrosion resistance than did the noncoated Zircaloy-4 (Daub et al., 2015). Again, in the steam tests, the best overall performance was with the CrN coatings; however, in areas where the coating was damaged, the underlying Zircaloy substrate suffered accelerated attack. In a follow-up study, Daub et al. (2017) performed testing of CrN PVD-coated Zircaloy-4 coupons in water at 300 C for up to 120 days and in steam at 500 C, 750 C, and 1000 C for 1 h. They found again a reasonable

Accident-tolerant fuels cladding concept Chapter | 4

71

temporary protection of Zircaloy by CrN coating both in water and in steam. In water, the coating was a protective barrier for hydrogen ingress into the Zircaloy substrate due to corrosion. In steam the attack of the coated Zircaloy happened only on the areas where the coating was purposely scratched (Daub et al., 2017).

Chromium coatings for zirconium alloys in the French ATF program French-led consortium of organizations (including CEA, EDF, AREVA (now Framatome) and the US Department of Energy) have been working in the development of a chromium coating for M5 zirconium alloys (Bischoff et al., 2018a; Brachet et al., 2019). The French program classified this chromium-coated cladding a short-term evolutionary concept in the realm of ATFs. The coating consists on an 8- to 22-µm-thick dense layer of Cr deposited using PVD on M5 cladding tubes (Bischoff et al., 2018a,b). They also published earlier (or pre-Fukushima accident) or first-generation studies of up to 10-µm-thick PVD process chromium-coated Zircaloy-4 tubes (Brachet et al., 2019). The resulting product offers an adherent homogenous (no cracks) layer of Cr, a porosity-free interface between the cladding and the coating, and an unmodified metallurgical condition of the zirconium alloy substrate. Bischoff et al. (2016) also demonstrated that the mechanical properties, including yield stress and elongation to failure, at an ambient temperature and at 400 C were similar between the noncoated M5 and Cr-coated M5. They suggested that this fact may ease the licensing for reactor application because the properties of noncoated M5 can be used to represent the mechanical properties of the Cr-coated tubes (Bischoff et al., 2016). Cr-coated and noncoated M5 coupons were exposed to a harsh “degraded” water containing 70-ppm lithium at 360 C for 168 days. While the noncoated M5 offered an oxidation mass gain of 1000 mg/dm2, the 15-µm-thick Cr-coated M5 coupons had a negligible mass gain of less than 50 mg/dm2 (Bischoff et al., 2018a). On the basis of the obtained results, it was suggested that using Cr-coated zirconium alloys, the normal operation conditions’ water chemistry constraints may be relaxed during the operation of the light water reactors. By the own nature of the higher hardness of the Cr-coated cladding, a reduction in the grid-to-rod and debris fretting failure of fuel rods under normal operation conditions of the reactors is expected. Bischoff et al. (2018a) performed an accelerated 100 h wear coupling test between both Cr-coated and noncoated M5 tubes cladding and an Inconel spring in 300 C PWR water containing 1000-ppm B plus 2-ppm Li. They reported that the cladding wear volume of the coated tube was reduced by two orders of magnitude when compared with the noncoated cladding under the same tested conditions (Bischoff et al., 2018a).

72

Accident-Tolerant Materials for Light Water Reactor Fuels

PVD Cr-coated M5 tubes were tested under LOCA conditions using creep tests and oxidation resistance properties (Brachet et al., 2017). Samples of 50-cm long and 10- to 15-µm-thick Cr-coated and noncoated M5 tubes were internally pressurized at 750 C and the in situ circumferential elongation was measured using laser techniques. It was observed that the coated zirconium alloy tubes offered a higher resistance to isothermal creep by a mechanism of strengthening of the underlying tube substrate. It was also reported that the time to rupture of Cr-coated M5 and Zircaloy-4 tubes via creep in the temperature range 600 C 1000 C was longer by a factor of two than the time to rupture of the noncoated zirconium alloy tubes. Less ballooning was observed in the coated tubes than in the noncoated ones. Moreover, when the rupture of the coated tubes happened, the coated Cr was still adhering to the zirconium alloy substrate next to the rupture area opening and the size of the mouth of the failure was much smaller for the coated tube than the opening for noncoated zirconium alloy tube (Brachet et al., 2017). Chromium-coated and noncoated M5 tubes, caped at both ends were tested in steam at 1200 C for up to 12,000 s and then water quenched to room temperature. For a 1200 C exposure time of 6000 s, the coated tubes gained approximately ten times less oxidation mass than the noncoated tubes. It was explained that the 10- to 15-µm-thick coating of Cr on the surface of the M5 tubes reduced the amount of oxygen diffusion into the zirconium matrix and therefore improved the ambient-temperature residual impact properties following quenching after the 1200 C steam exposure (Brachet et al., 2017). CEA started studying the coating of zirconium alloys in the early 1960s for applications in a CO2 atmosphere in a graphite-gas reactor for a maximum temperature of 650 C (Brachet et al., 2019). In the last decade (before the Fukushima accident), Brachet et al. (2019) performed exposure tests of coated Zircaloy-4 to steam at 1100 C for 850 s and quenched in water at an ambient temperature. Several coatings were investigated, including TiN, CrN, TiN/AlTiN, CrN/AlTiN, NbCrTi, Cr/NbCrTi, and Cr (single layer and multipass). They reported that Cr was the most effective PVD coating against the steam test exposure with the lowest mass gain, and the Cr multipass was more protective than the single layer chromium. The noncoated Zircaloy-4 material gained between 11 and 13 mg/cm2 while the Zircaloy-4 coated with Cr/Cr multipass (with 14 sublayers for a total thickness of 7 µm) gained between 1 and 2 mg/cm2 (Brachet et al., 2019).

Aluminum-based and iron chromium aluminum coatings for zirconium alloys The use of monolithic iron chromium aluminum (FeCrAl) alloys are suggested as structural material for replacing zirconium alloys for cladding

Accident-tolerant fuels cladding concept Chapter | 4

73

because of their outstanding oxidation resistance in steam at temperatures higher than 1000 C (see Chapter 5: FeCrAl—iron chromium aluminum monolithic alloys). FeCrAl material performs well under normal operation condition water owing to the formation of a compact protective layer of Cr2O3, and FeCrAl offers protection against high-temperature accident conditions steam attack by the formation of a thin film of alumina. One of the less favorable properties of FeCrAl is their parasitic absorption of thermal neutrons, which is approximately ten times higher than for the currently used zirconium alloys (Terrani et al., 2013). Therefore one alternative is to use current zirconium alloys as the structural material to contain the fuel and the fission products and coat the external wall with a thin layer of a FeCrAl alloy (Terrani et al., 2013; Tang et al., 2017). As mentioned before, a thin coating of FeCrAl would not affect greatly the neutronic balance in the reactor (Younker and Fratoni, 2016). Zhong et al. (2016) prepared approximately 1-µm-thick FeCrAl coatings with a wide range of 0%Al to 34%Al on the surface of Zircaloy-2 using magnetron sputtering. They tested the coatings in the not highly aggressive conditions of steam at 700 C for 15 h and reported a decrease in mass gain as the amount of aluminum increased in the coating. Tests performed in condensed water with 2 ppm of oxygen at 288 C for 20 days showed higher mass gain for the FeCrAl coated coupons than for the noncoated Zircaloy-2. They also reported a reaction between iron in the coating with zirconium in the Zircaloy-2 substrate forming a eutectic near 900 C (Zhong et al., 2016). It was argued that the 1-µm coating thickness was not enough to provide protection to the zirconium alloy substrate and that a 10-µm thickness could be more relevant. Terrani et al. (2013) used FeCrAl (74Fe 1 20Cr 1 5Al) and type 310SS (52Fe 1 25Cr 1 20Ni) 2-mm-thick metal sheets or cans to wrap or encapsulate slugs of hot isostatic-pressed zirconium powder and then determine their environmental resistance behavior in steam and interdiffusion behavior between the cans and slugs for up to 48 h at 1200 C and for 8 h at 1300 C. They reported that in the 48 h test at 1200 C, both the 310SS and the FeCrAl cans resisted the attack from the environment but in the test for 8 h at 1300 C only the FeCrAl can survived the test because of the additional protection offered by the aluminum in FeCrAl (Terrani et al., 2013). In the boundary between the Fe-alloy can wall and the zirconium slug, there was also more interdiffusion between zirconium and 310SS than between zirconium and FeCrAl. This test demonstrated that at temperatures higher than 1200 C, the presence of alumina is important to protect the iron-based alloy, and the presence of 25%Cr in type 310SS without aluminum was not enough. The use of FeCrAl and Cr/Al alloy coatings for Zircaloy-4 alloy has also been evaluated in the Korean and Chinese nuclear materials programs with dissimilar results (Kim et al., 2016a,b; Wang et al., 2018; Han et al., 2019).

74

Accident-Tolerant Materials for Light Water Reactor Fuels

When a film of FeCrAl was applied directly on a zirconium alloy tube to protect it from steam attack, the applied layer may react with the zirconium substrate if the tube is exposed to temperatures in the order of 1000 C.

Silicon-based coatings for zirconium alloys On the basis of the successful performance of silicide compounds for hightemperature applications in the industry, Hwasung et al. (2016) explored the possibility of using zirconium silicide (ZrSi2) to protect zirconium alloys against high-temperature attack in case of an accident. They sputter deposited ZrSi2 on flat coupons of Zircaloy-4 at room temperature to a total thickness of 850 nm. The coated coupons were exposed to air at 700 C for up to 5 h. They reported that the ZrSi2-coated Zircaloy-4 coupon had a 33% reduction in mass gain after 5 h at 700 C in dry air when compared with the noncoated Zircaloy-4 coupon (Hwasung et al., 2016). At 700 C the coated coupon developed an oxide of 1-µm thickness, while the noncoated Zircaloy4 coupon had an oxide that was 10-µm thick. They also conducted exposure tests of bulk samples of ZrSi2 and bare Zircaloy-4 coupons at 1000 C and 1200 C but not of Zircaloy-4 coated with ZrSi2. Kim et al. (2015a) explored the possibility of using silicon-based compounds to protect zirconium alloys from high-temperature oxidation. They based their research on the fact that SiC is stable in steam up to 1500 C. They coated Zircaloy-4 coupons using 90-µm silicon powder by plasma spray and plasma spray plus laser beam scanning (Kim et al., 2015a). The plasma spray process was repeated on the Zircaloy-4 coupon up to six times, resulting in a coated layer between 70 and 130 µm containing abundant porosity but no reaction between silicon and zirconium at the coating/ Zircaloy-4 interface is observed. The second type of coupon had a three-pass plasma spray with a thickness of 70 µm and then a laser beam scanning treatment, which reduced the porosity and cracks of the deposited silicon layer and the silicon reacted with zirconium from the substrate. They tested the two types of coated coupons in steam at 1200 C for 2000 s and measured the in situ mass change as a function of exposure time. They reported a modest decrease in the oxidation rate of the Zircaloy-4 coupons coated via plasma spray (with six passes) and plasma spray (three passes) 1 laser beam scanning when compared with the noncoated Zircaloy-4 coupon. For the noncoated Zircaloy-4 coupon the mass gain was in the vicinity of 2300 mg/ dm2, while for the 6-pass plasma spray it was approximately 1800 mg/dm2 and for the 3-pass plasma spray 1 laser beam scanning it was approximately 1500 mg/dm2 (Kim et al., 2015a). They also reported oxidation of the Zircaloy-4 substrate under the silicon-coated layers. The poor adhesion of the plasma spray layer and the Zircaloy-4 substrate could be related to the large difference in the thermal expansion coefficient between silicon and zirconium. There was no comment on the behavior of a silicon-coated layer

Accident-tolerant fuels cladding concept Chapter | 4

75

under reactor normal operation conditions in contact with hydrogenated water at near 300 C.

Fabrication and implementation of zirconium-coated rods There are several methods regarding how coatings can be applied over zirconium alloys cladding tubes, including spray coating, chemical vapor deposition (CVD), PVD, sol gel, laser, electrodeposition, and hot dipping. Vapor-deposited coatings are generally compact and defect free and they provide an excellent barrier to further attack by the environment. In the PVD process, the vaporized atoms in a near vacuum atmosphere condense directly on the zirconium tube while in the CVD process the deposition on the zirconium tube happens via a chemical reaction at temperatures in the order of 500 C (NEA, 2018). It has been argued that even though a PVD processes provide a uniform defect-free coating, it may be too slow or non-economical for an industrial-scale production of fuel cladding (NNL, 2018). In a thermal spray process the coating material is accelerated towards the zirconium tube target and deposit on the target by the transfer of kinetic energy from of the accelerated particles. The acceleration could be done at a higher temperature using oxygen fuel or at a lower temperature (cold spray) using a carrier gas. A spray process for Cr coating could be faster than the PVD process, but the end coating product may not have the same quality (NNL, 2018). AREVA (now Framatome) has announced that they have fabricated a PVD prototype machine to coat full-length M5 cladding tubes, which will be used for insertion into a commercial power reactor in 2019 (Bischoff et al., 2018a). It was also reported that Cr-coated M5 tubes can be welded using the standard “upset-shaped welding” process and without modification of the welding parameters. Basically, the Cr-coated rods can be industrially manufactured using the current production set up. The resulting joint was free of welding flaws, having survived corrosion and burst tests at 360 C. In the burst tests, the failure of the coated tubes always occurred in the tube area away from the weld joint (Bischoff et al., 2018a). One of the requirements of applying coatings to zirconium alloys for environmental protection is that the coating process did not change the mechanical and microstructural properties of the zirconium substrate tubing. These criteria may limit the type of coatings to be applied and the mode of application processes. The most desirable solutions for zirconium coatings seem to be a simple chromium or chromium nitride deposition using probably a PVD process. This outcome may make the coating route of full-length tubes quick and cost effective to implement. However, there is always a need for more research and development to make the coating reproducible and cost effective.

76

Accident-Tolerant Materials for Light Water Reactor Fuels

Performance of coated zirconium under reactor normal operation conditions Under normal operation conditions, the coating needs to be compatible with hydrogenated water in the order of 300 C 350 C. The PWR water may contain 2.2 ppm of dissolved lithium (as lithium hydroxide) and 1000 ppm of dissolved boron (as boric acid). It is known that current light water reactors have fuel rods of zirconium-based alloys that perform acceptably well under the normal operation conditions by the development of an adherent zirconium oxide on the surface. Other reactor internal components are made of austenitic stainless steels such as type 304 and 316, and nickel-based alloys such as alloy 600. The austenitic alloys resist corrosion in the normal operation condition environments by the development of a thin protective chromium oxide film on the surface (Rebak et al., 2017a,b). If the coating for zirconium alloy is chromium based, it would also resist corrosion by the development of a chromium oxide on the surface and therefore the impact of inserting these chromium-coated zirconium tubes into the reactor will not affect anything regarding material/environment compatibility. The presence of the chromium oxide on the surface will provide an additional benefit because it will be a barrier to reduce or eliminate the hydrogen uptake by the substrate zirconium tube. The reduction of hydrogen uptake will be in the form of a lower amount of atomic hydrogen produced on the surface owing to the lower corrosion rates of chromium and because the chromium oxide is an effective barrier for hydrogen diffusion to the underlying metal. The fact that less hydrogen may diffuse into the zirconium tube may allow for higher burn ups of the fuel (NEA, 2018). Bischoff et al. (2016) performed immersion corrosion tests of zirconium alloys Zircaloy-4 and M5 and Cr-coated M5 and Zircaloy-4 flat coupons in recirculating PWR-type water with less than 10 ppb of dissolved oxygen (no hydrogen gas added) at 360 C for up to 180 days. They reported that for noncoated M5 alloy the mass gain after 180 days was in the order of 50 mg/ dm2, while the mass gain of the Cr-coated coupons was less than 1 mg/dm2 (Bischoff et al., 2016). They noted that the zirconium material under the Cr coating did not show oxidation, demonstrating the protectiveness of the Cr coating. Little or no data exist on the behavior of thin FeCrAl coatings on zirconium alloys under LWR normal operation conditions. Because bulk monolithic FeCrAl have experienced light weight loss in the presence of 300 C hydrogenated water (Rebak, 2018), there is a concern that if a thin coating is applied to a zirconium alloy tube, it may lose mass and become thinner and therefore less protective to the underlying zirconium alloy tubing as a function of time (NEA, 2018). Little is known on the behavior of MAX phases coatings on zirconium alloys in hydrogenated water in the 300 C temperature range because limited

Accident-tolerant fuels cladding concept Chapter | 4

77

to no information has been published in the open literature (Tang et al., 2017; NEA, 2018). Roberts (2016) performed immersion tests at 360 C of noncoated and Ti-Al-C and Cr-Al-C magnetron sputtered coated Zirlo coupons for up to 14 days. The intention during the coating was to form MAX phases such as Ti2AlC and Cr2AlC. Immersion corrosion test results showed that the TiAlxC1-x coated coupons gained about twice the mass as the control noncoated Zirlo coupon. However, after 10-days immersion in 360 C water, the Cr-Al-C Zirlo coated coupons gained less mass or lost mass compared with the noncoated Zirlo coupons (Roberts, 2016). The reason for the small mass gain for the coated coupon was because the coating flaked off during the short duration testing; therefore the results became irrelevant. Both TiN and CrN coatings on zirconium alloys performed well when exposed to water at approximately 300 C, typical of reactor normal operation conditions (Alat et al., 2015; Tang et al., 2017). TiAlN was deposited up to 12 µm in thickness on Zirlo zirconium alloy using a cathodic arc PVD process. Exposure tests in water at 360 C for 3 days showed that the mass gain was in the order of 1 5 mg/dm2 for the TiAlN coated coupons and 1.2 mg/ dm2 for the TiN coated coupons when compared with the noncoated Zirlo coupon, which gained 14.4 mg/dm2 (Alat et al., 2015). This was a very short time testing, and not many conclusions can be drawn from it. In general, it is known that the aluminum in the coating reacts with water, forming a boehmite phase that is not highly protective under normal operation conditions (Tang et al., 2017). Multilayer coatings were also explored to avoid the boehmite phase in water at near 300 C and still have the aluminum available in the coating for protection under accident conditions (Tang et al., 2017). This is a complicated process that may need major development. General Electric performed the standard ASTM G2 immersion testing for their ARMOR-coated and noncoated Zircaloy-2 tubing at 400 C in 10.3 MPa steam for 72 h (Lin et al., 2018). They also performed an industry standard nodular corrosion test, which has two stages: (1) 410 C for 4 h in steam at 12 MPa followed by (2) 520 C for 16 h in steam at 12 MPa. In both tests, the ARMOR-coated Zircaloy-2 outperformed the noncoated Zircaloy-2 tubing. The zirconium was not oxidized under the ARMOR coating, proving the protectiveness of the coating towards the underlying tube. They also performed at least 4 cycles of the ASTM G2 tests for ARMOR-coated material and reported that the mass gain was approximately 2 mg/dm2 after the four cycles, while the mass gain of noncoated Zircaloy-2 was nearly 20 mg/dm2 for one cycle (Lin et al., 2018).

Performance of coated zirconium under accident conditions As discussed earlier, a chromium coating is considered to offer a higher resistance to attack by steam than noncoated zirconium alloys in conditions of DBA (Terrani, 2018). Bischoff et al. (2016) performed comparative

78

Accident-Tolerant Materials for Light Water Reactor Fuels

exposure tests of noncoated M5 and Cr-coated M5 in steam at 415 C for more than 120 h. They reported that while the noncoated M5 had a mass gain of near 160 mg/dm2, the Cr-coated coupon gained approximately 1 2 mg/dm2 (Bischoff et al., 2016). Brachet et al. (2016) performed LOCA testing of noncoated Zircaloy-4 and 15-µm chromium-coated Zircaloy-4 tubes by pressurizing the tubes at 120 bar and at temperatures ranging from 600 C to 1000 C and measured the time to rupture as a function of the initial hoop stress. They reported that the time to rupture at each temperature of the chromium-coated tubes was approximately two to five times higher than the time to rupture the noncoated Zircaloy-4 tubes, suggesting a strengthening effect of the coating at higher temperature on the zirconium alloy substrate (Brachet et al., 2016). Oelrich et al. (2018) performed oxidation tests of cold-sprayed chromium-coated Zirlo in steam at 1300 C and 1500 C for up to 25 min and they reported a nonsignificant weight gain at 1300 C but they noted surface melting at 1500 C (Geelhood and Luscher, 2019). Basu et al. (2012) performed oxidation tests for up to 120 h of Ti2AlC in air and steam in the temperature range 1000 C 1300 C. They reported that the resistance to oxidation in air and steam is given by the formation of a protective surface layer of alumina following a cubic rate law. A rutile (TiO2) layer may also form, but it is eventually evaporated owing to the reaction with steam, forming a gas phase of TiO(OH)2 (Basu et al., 2012). The thickness of the alumina layer never exceeded 20 µm, even after 120-h exposure to 1300 C. Basu et al. (2012) did not observe titanium or aluminum depletion in the coating matrix underneath the protective oxide. Tallman et al. (2013) reported that the oxidation resistance of Ti2AlC, Ti3AlC2, and Cr2AlC at 900 C 1400 C in air is due to the formation of a protective adherent alumina (Al2O3) scale on the surface. The oxidation kinetics of Ti2AlC in steam in the temperature range 1000 C 1300 C was only slightly faster than in air, but the activation energies were practically the same (Basu et al., 2012; Tallman et al., 2013). A lower activation energy was reported for Ti2AlC than for Cr2AlC (Tallman et al., 2013). Maier et al. (2015) explored the high-temperature oxidation of Ti2AlC coatings on flat Zircaloy-4 coupons deposited using the cold spray process, which is a solid-state deposition at supersonic velocities. The coating was homogeneous and adherent with a thickness of 90 µm. The environmental exposure tests were conducted for 20 min at 1005 C in a 100-standard cm3/min of argon, which was bubbled through boiling water before entering the reactor. After the high-temperature exposure tests the coupons were quenched in water at an ambient temperature. There was no oxidation of the Zircaloy-4 substrate under the coating and there was no spallation upon quenching (Maier et al., 2015). The environmental behavior or nitride coatings (TiN and CrN) in high-temperature steam was not sufficiently explored (Tang et al., 2017).

Accident-tolerant fuels cladding concept Chapter | 4

79

ARMOR coating by General Electric was tested for oxidation resistance in flowing steam under design-basis LOCA conditions at up to 1200 C (Lin et al., 2018). One test conducted for 5000 s at 1000 C, which is equivalent to 23% ECR, showed considerable attack of the noncoated inner diameter of the Zircaloy-2 tubing (oxide plus oxygen stabilized alpha layer). Meanwhile, the external surface of the tube coated with ARMOR only showed a thin layer of the oxygen stabilized alpha layer and no zirconium oxide under the coating (Lin et al., 2018).

Coated zirconium irradiation behavior AREVA (now Framatome) has announced that their chromium-coated M5 zirconium cladding was (in 2018) under irradiation exposure in the Go¨sgen reactor in Switzerland according to the IMAGO project (Bischoff et al., 2016, 2018a). They have also tested coated M5 in the Halden reactor before this reactor was permanently shut down in 2018. AREVA also has rodlets made of M5 Cr-coating under irradiation in the Advanced Test Reactor (ATR) at the Idaho National Laboratory (INL) under PWR environmental conditions. Some of the in-pile studies will include not only the Cr-coated M5 but also the urania modified with chromia fuel concept (described in Chapter 7: Alternative fuels to urania). The irradiation studies will provide crucial information on the pelletcladding interaction both diametrically and axially in the rod, and the effect of pellet swelling on the external behavior of the thin chromium coating. The behavior of MAX phases under irradiation were explored and it was found that they generally can withstand high doses without suffering significant damage in the form of amorphization or swelling (Tallman et al., 2015). It was shown that, besides resistance to heavy iron and He irradiation, MAX phases such as Ti3SiC2, Ti3AlC2, and Ti2AlC have high tolerance to neutrons comparable to that of SiC and three times better than alloy 617 (Tallman et al., 2015). The microstructural stability and electrical resistivity of polycrystalline Ti3SiC2, Ti3AlC2, Ti2AlC, and Ti2AlN were studied under neutron irradiation at the 6-MW Massachusetts Institute of Technology reactor. Irradiation was performed at 360 C and 695 C in an inert gas (He 1 Ne) atmosphere to a total damage of 0.1 dpa (Tallman et al., 2015). They reported less damage at the higher irradiation temperature owing to dynamic recovery and found that the Ti2AlC and Ti3SiC2 phases are the most resistant to damage by neutrons (Tallman et al., 2015). It was noted that irradiation in the reactor may activate the element chromium to Cr-51, which is a radioisotope with a half-life of 28 days (Geelhood and Luscher, 2019). However, the formation of Cr-51 can be monitored in each nuclear power station and these species may be removed from the effluents before it becomes a safety problem. Chromium is not going to be a new element in the reactor. Austenitic stainless steels such as type 304 and nickel-based alloy such as Alloy 600 have been used as core internal

80

Accident-Tolerant Materials for Light Water Reactor Fuels

component materials for many decades. Both austenitic alloys rely on chromium for corrosion resistance in reactor environments. General Electric, working with Southern Nuclear, uploaded in Cycle 29 of the Edwin Hatch Unit 1 Power Plant in Georgia (USA) lead test assemblies (LTA) of ARMOR-coated Zircaloy-2 fueled segmented rods (Lin et al., 2018). The objective of this insertion is to obtain irradiated materials for post irradiation examination (PIE) studies to determine performance of the coated cladding under normal neutron irradiation conditions of an operating commercial power reactor. The first harvesting for PIE will be conducted in the spring of 2020.

Coated zirconium licensing for reactor use The zirconium coating concept for enhanced ATF (EATF) seems to be the most straight forward evolution from the point of view of licensing for reactor use. This is because the fuel rods remain essentially all zirconium alloy with only a thin surface modification. For example, the neutronic impact of the presence of chromium, or CrN on the surface of the cladding maybe directly incorporated into slight modifications of the core design. A metallic coating may not add any significant barrier to heat transfer from the fuel to the coolant; therefore little or no impact is anticipated to the fuel centerline temperature. If the tube wall thickness of the current zirconium alloy cladding is between 0.6 and 0.8 mm, the coating may add only 10-µm thickness or a 1.25% to the original zirconium tube wall thickness. Basically, for neutronic calculations and structural considerations, the coated zirconium concept is highly like the original configuration (Younker and Fratoni, 2016). One significant advantage of coatings or surface treatments is that most of the cladding remains zirconium alloys, which have been used for the past 60 years (NEA, 2018). Accordingly, cladding properties are modified only on its surface and therefore licensing of these concepts may likely largely benefit and justified from the already-known zirconium alloy behavior. The thinner the intended coating, the smaller the modification of the existing cladding, which may infer enabling the licensing process (NEA, 2018). A coating applied on current zirconium alloys may modestly increase the cost of the current cladding. General Electric working with their ARMOR coating program considers that coated zirconium alloy is a variation of the current bare Zircaloy cladding evolution, and closer to implementation that GE’s other concept called IronClad (Lin et al., 2018). It is envisioned that ARMOR-coated Zircaloy-2 can be deployed in light water power reactors as a fretting resistance solution before regulatory recognition of the incremental accident safety benefit (Lin et al., 2018). Pacific Northwest Laboratory was commissioned by the US Nuclear Regulatory Commission to prepare and publish a comprehensive evaluation on the anticipated performance of chromium-coated zirconium alloys, on the

Accident-tolerant fuels cladding concept Chapter | 4

81

basis of a list of cladding material properties (Geelhood and Luscher, 2019). The state-of-the-art report on the chromium coating of zirconium alloys, which is pursued by three fuel vendors in the United States, is meant to increase the NRC preparedness to receive the licensing topical reports from future license applicants. Some of the properties needed by the NRC would include (1) thermal conductivity, (2) thermal expansion, (3) emissivity, (4) enthalpy and specific heat, (5) mechanical properties such as yield stress and elastic modulus, (6) thermal and irradiation creep rate, (7) axial irradiation growth, (8) oxidation rate, (9) hydrogen pickup, and (10) ballooning behavior at high temperature and oxidation rate in steam at 800 C 1200 C (Geelhood and Luscher, 2019). For example, on the oxidation rate of the coating, it is stated that chromium would offer an excellent protection to the cladding by the formation of a chromium oxide on the surface that will offer extremely low dissolution rates both under normal oxidation conditions and anticipated operational occurrences. However, the authors recommended the acquisition of actual oxidation rate of chromium coatings in prototypical reactor performance using fueled rods (Geelhood and Luscher, 2019). The presence of the chromium oxide on the coating surface will also be a barrier for hydrogen entrance into the underlying zirconium alloy substrate. However, the authors recommended to confirm the lower hydrogen pickup prediction via destructive examination of cladding segments removed from lead test assemblies. Increased dosage release from power stations because of the presence of the chromium coating is unlikely, but plants should monitor the presence of the radioisotope Cr-51 and report the information to the NRC (Geelhood and Luscher, 2019). Some of the data gaps that may be needed for licensing of the chromiumcoated zirconium alloys include (1) irradiation creep tests, (2) power ramp tests, and (3) reactivity initiated accident tests. Other data that may also be needed for licensing but may be acquired in the shorter term from evaluations of the LTA segments include (1) axial growth, (2) rod bow, (3) fission gas release, (4) hydrogen pickup, and (5) overall oxidation rate (Geelhood and Luscher, 2019).

Chapter 5

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter Outline Overview What are FeCrAl alloys? Metallurgy and microstructure of FeCrAl Earlier considerations of FeCrAl alloys for nuclear applications Why are FeCrAl considered for accident-tolerant fuel cladding? Benefits and challenges Thermal properties of FeCrAl Mechanical properties of FeCrAl Oxidation resistance of FeCrAl under LWR’s normal operation conditions Composition of the oxide films on FeCrAl coupons Electrochemical behavior of FeCrAl alloys in high-temperature water Shadow corrosion Galvanic corrosion Resistance to crud deposition under normal operation conditions Resistance to EAC of ferritic alloys under LWR normal operation conditions Resistance to fretting under normal operation conditions Resistance of monolithic FeCrAl cladding to thermal shock

84 84 85 87

88 92 93 96 102

106 108 111 112

113 115 116

Accident-Tolerant Materials for Light Water Reactor Fuels. DOI: https://doi.org/10.1016/B978-0-12-817503-3.00005-5 © 2020 Elsevier Inc. All rights reserved.

Interaction between the urania fuel and the FeCrAl cladding Oxidation resistance of FeCrAl in high-temperature gas environments Mechanism of protection at accident condition temperatures The Roles of metal oxides on the surface of FeCrAl Normal operation oxidation to accident oxidation scenario and vice versa Scenario 1: Water-oxidized APMT tubes exposed to superheated steam Scenario 2: Steam-oxidized APMT tubes exposed to high-temperature water The versatile oxidation behavior of FeCrAl Alloys Fabrication and implementation of cladding tubes Welding of FeCrAl alloys Mitigation measures to parasitic neutron absorption of FeCrAl Mitigation measures to increased tritium release into the coolant Irradiation behavior of FeCrAl Corrosion behavior of used FeCrAl cladding in cooling pools Licensing for reactor use

118 119 121 124

124 125

126 127 129 131 134 134 137 139 140

83

84

Accident-Tolerant Materials for Light Water Reactor Fuels

Overview The use of monolithic FeCrAl alloys to replace zirconium alloys is the second nearest term for the accident-tolerant fuel claddings after coated zirconium alloys. Currently a single layer of less than 1-mm-thick zirconium alloy metal separates the fuel cavity and its toxic compounds from the coolant water. The simplest and most elegant improvement would be to replace the current alloy for another alloy, obviously as resistant to corrosion as zirconium under normal operation conditions but a thousand times more resistant to attack by steam under severe accident conditions. An alloy down selection was conducted in steam under severe accident scenarios ( . 1200 C) and the highest environmental accident resistance was always offered by FeCrAl alloys because of the formation of an adherent, thin, and shielding film of alumina, which protected the FeCrAl from steam attack until its melting point. FeCrAl are ferritic materials that have a low corrosion rate under normal operation conditions by the formation of a protective chromium oxide on the surface. Ferritic FeCrAl are also resistant to environmental cracking from the coolant side. FeCrAl alloys can be made into fulllength fuel rod tubes with a wall thickness of less than 0.5 mm and they can be industrially welded to hermetically seal the fuel cavity by a solid-state method. The thermal neutron absorption cross section of FeCrAl is approximately ten times higher than that of the current zirconium alloys, but this fact can be partially mitigated by making the cladding wall thickness half of the current values for Zircaloy owing to the stronger mechanical properties at temperature of the FeCrAl alloys. The use of FeCrAl cladding may initially increase the tritium release into the coolant but is anticipated that the oxidation of the cladding from the fuel cavity side and the coolant side will be effective barriers for hydrogen diffusion across the tube wall. The use of FeCrAl cladding will eliminate the occurrence of debris fretting and shadow corrosion, two current Zircaloy cladding environmental degradation modes. FeCrAl tube cladding concepts are currently under irradiation both at the advanced test reactor in Idaho National Laboratory and in two BWR commercial nuclear power stations in the United States. Even though FeCrAl materials were never used in light water reactors before, their characterization in reactor anticipated environments is advancing rapidly through international development efforts.

What are FeCrAl alloys? Iron
chromium
aluminum (FeCrAl) alloys are a family of ferritic or bodycentered cubic steel-like alloys, which were first commercialized by Hans von Kantzow in Hallstahammar, Sweden in the early 1930s. The name of the original company that commercialized FeCrAl alloys is Kanthal, which is a combination of the last name of the inventor and the city of origin. The

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

85

FeCrAl alloys are mostly used for heating elements and in catalytic converters owing to their unparalleled resistance to high-temperature ( . 1000 C) oxidation, even under cyclic thermal conditions. The basic composition of FeCrAl are iron (Fe) base, with approximately 20
30 mass percentage of chromium (Cr) and 4
7.5 mass percentage of aluminum (Al). If the amount of aluminum in the alloy is higher than 8 mass percentage, the aluminum oxide on the surface may not be protective. It was suggested that the amount of aluminum should be: %Al 5 (40-%Cr)/6 and the minimum amount of chromium is recommended at 12% (Geanta et al., 2016). Over the years some of the alloys were modified with 2
3 mass percentage of molybdenum (Mo) and other minor alloying elements (in the parts per million range) such as yttrium (Y), hafnium (Hf), titanium (Ti), Tantalum (Ta), and zirconium (Zr). For high-temperature applications, the FeCrAl alloys were first produced in the traditional melting, casting and forging method but eventually powder metallurgy was used to manufacture some versions of FeCrAl alloys. The powder metallurgy processing was first suggested in the mid-1960s to obtain better mechanical properties and especially to avoid the low elongation to failure at a near-ambient temperature of the cast products (General Electric, 1965; Wilson et al., 1978). The use of powder metallurgy allowed for the alloy to have not only exceptional oxidation resistance but also much-needed manufacturability at a near-ambient temperature and hightemperature strength and creep resistance (Quadakkers et al., 1991). The increase in mechanical properties of the powder-produced materials was mainly a result of the smaller grain size and the presence of dispersed oxides, which are obstacles for dislocation propagation.

Metallurgy and microstructure of FeCrAl Fig. 5.1 shows the typical grain size and the second-phase precipitates of alloys manufactured by the powder process advanced powder metallurgy tubing (APMT) and by traditional melting, casting, and forging (C26M). As for other family of alloys, the microstructure of FeCrAl alloys would vary considerably depending on their manufacturing path and their thermomechanical processing history. In Fig. 5.1 the microstructure of APMT shows small grains in the order of 2
10 μm diameter with the size of numerous dispersed oxides being less than 500 nm (shown as white spots), which are rich in Zr-, Hf-, Ti-, and other minor-oxide-forming elements. The APMT microstructure may also have sporadic porosity typical for powder metallurgy products. In the current heat of APMT the porosity was in the order of 1 μm in size. On the other hand, the microstructure of C26M shows a cleaner microstructure, with fewer dispersed oxides, no porosity, and larger grains. The larger grains obtained in C26M are not desirable because they offer little or no ductility. For the cladding application, the manufacturing of thin wall tubes for both APMT and C26M alloys will undergo a series of

86

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.1 Microstructure of iron
chromium
aluminum (FeCrAl) alloys.

thickness-reduction processes and intermediate annealing. The tube fabrication and resulting microstructure is discussed later in this chapter. Detailed microstructure studies of two heats of powder metallurgy APMT and three heats of traditionally melted alloys such as C26M were reported by Sun et al. (2018). One of the heats of APMT (Heat #1) that was from a larger 10 cm 3 10 cm billet showed a bimodal microstructure with big elongated grains (up to 1 mm long and 100
400 μm wide) surrounded by small grains (,10 μm). The second heat of APMT (Heat #2) was from a 12 mm OD rod or wire and it had a more uniform microstructure with grains in the order of 10- to 16-μm diameter (Sun et al., 2018). The different microstructure of the two heats of APMT produced different mechanical properties. The elongation to failure at an ambient temperature of the bimodal grain microstructure was only (0.8 6 0.2)%, while the elongation to fracture of the APMT heat with the more uniform grain microstructure was (25.9 6 1)% (Sun et al., 2018). The manufacturability of thin wall tubes will be facilitated for the stock material with small initial grains.

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

87

Earlier considerations of FeCrAl alloys for nuclear applications In the 1960s General Electric Company, working under contract with the US Atomic Energy Commission, studied for several years the family of FeCrAl alloys for nuclear power applications (General Electric, 1965, 1966, 1967a,b). The GE studies showed superior resistance of FeCrAl to oxidation and steam corrosion resistance up to 1300 C (General Electric, 1965). As cladding materials, both Fe-25Cr-4Al-1Y (alloy 2541) and Fe-15Cr-4Al-1Y (alloy 1541) showed excellent stability in air and steam for up to 10,000 h at 1100 C (General Electric, 1966). In-pile tests showed that the FeCrAl alloys had good retention of fission gas products. In 1965 it was reported that embrittlement at low temperature was caused by the precipitation of a Crrich phase; therefore the reduction on chromium addition was effective in decreasing the embrittlement of the alloys. Some of the studies carried out to determine the oxidation and the irradiation stability of Fe-25Cr-4Al-1Y (alloy 2541) clad focused on developing fuel elements for elevated temperature nuclear applications, which included coolants of steam, air, and carbon dioxide. In 1965 researchers understood that in order to maintain the gaseous oxidation resistance up to 1300 C, as the chromium content was decreased, the concentration of aluminum need to be increased (Fig. 5.2) (General Electric, 1965). To increase the strength of the traditionally melted FeCrAl alloys, molybdenum up to 5% was added. The addition of 5% Mo to a Fe15Cr-4Al-1Y (alloy 1541) doubled the tensile strength of the alloy at 870 C. The alloys produced by powder metallurgy showed a higher strength than the traditionally melted alloys did, even with a moderate addition of molybdenum in the traditional alloys (General Electric, 1966). The ductility of the alloys was also improved by the powder metallurgy technique. Increasing the strength of the FeCrAl further by using oxide dispersion strengthening (ODS) is also proposed. Out-of-pile exposure data showed that FeCrAl alloys were compatible with UO2 after 10,000 h exposure at 1100 C (General

ht pe rc

en

t

Oxidation resistance at 900º–1300ºC Poor Good

8

eig ,w nu m Alu mi

10

GE-NMPO

6 4

2 Fe

4

8

12

16

20

24

26

Chromium, weight percent

FIGURE 5.2 Chromium (Cr) and aluminum (Al) needed for oxidation resistance up to 1300 C (General Electric, 1965).

88

Accident-Tolerant Materials for Light Water Reactor Fuels

Electric, 1965, 1966). Two in-pile tests of Fe-Cr-Al-Y-clad Fe-UO2 indicated excellent containment of fission products up to 1100 C (General Electric, 1966). As part of the General Electric studies that ended in 1966 the results of studies on the chemical
physical interaction between FeCrAl cladding and urania fuel between 500 C and 1200 C were reported (Edwards and Bohlander, 1969). Thirteen alloy combinations containing from 5Cr to 25Cr and from 4Al to 10Al, as well as Fe-Al and Fe-Cr alloys were tested for compatibility with the urania fuel. The FeCrAl specimens were coupled to urania and isothermally exposed for up to 3450 h. Results showed that from 1000 C to 1200 C, the FeCrAl reduced some of the urania in the cermet fuels to form alumina on the surface of the cladding, thereby creating free uranium (U), which was initially transferred to the oxide in the cladding. That is, the equilibrium oxygen partial pressure on urania is higher than on alumina, and a displacement reaction occurred. The urania get reduced and the aluminum in the alloy was oxidized. Once an alumina layer is formed on the surface of the FeCrAl cladding, the transfer or diffusion of uranium stopped (General Electric, 1968; Edwards and Bohlander, 1969). There was no uranium transfer to the cladding at temperatures below 1000 C. Fe-CrAl-Y alloys were also extensively tested for their potential application as fast breeder reactor cladding materials (General Electric, 1968). It was reported that the presence of yttrium in the alloy decreased the ductility and notched-impact properties of Fe-Cr-Al-Y (General Electric, 1966). By decreasing Y in the alloy, the ductile-to-brittle transition temperature (DBTT) did not decrease but the values of toughness (impact energy) above the DBTT increased. The atomized product offered improved impact properties (and a decrease of the DBTT) mainly because of the optimized microstructure in the alloy (General Electric, 1966). The mechanical properties (such as yield strength and elongation to failure) of FeCrAl are highly dependent on the microstructure of the alloys.

Why are FeCrAl considered for accident-tolerant fuel cladding? Benefits and challenges FeCrAl alloys have been used in industrial applications since the early 1930s, when the company Kanthal was established in Sweden to commercialize its properties. Due to their extraordinary resistance to high-temperature oxidation, FeCrAl alloys are mainly used for heating elements in furnaces, catalytic converters, hot plates, toasters, and even e-cigarettes. The development of FeCrAl alloys was not targeted for the manufacturing of a reactor’s internal components, neither for light water reactors (LWRs) or Generation IV reactors. FeCrAl alloys have an uncommon high-temperature oxidation resistance because of the collaborative effect of chromium and aluminum (Battelle, 1970). At lower temperatures and up to approximate 1000 C, the alloy protects itself from the environment by the formation of a chromium oxide

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

89

(chromia) film on the surface. However, as the temperature increases further above 1000 C, chromia may volatilize or evaporate by reacting with water molecules in the gas environment. However, while the chromium oxide is still present on the surface of the alloy, it allows for an aluminum oxide (alumina) film to develop and grow between the metallic FeCrAl substrate and the chromium oxide. This aluminum oxide is generally in the order of 1-μm thick and shows continuous, adherent, and protective characteristics (Rebak, 2018). FeCrAl alloys generally contain, besides the three basic elements iron (Fe), chromium (Cr) and aluminum, smaller proportions of other elements such as molybdenum (Mo), rare earths (RE), zirconium (Zr), titanium (Ti), and tantalum (Ta). Oak Ridge National Laboratory published a comprehensive manual on the properties of FeCrAl, which is mainly geared toward nuclear applications (ORNL, 2018a,b). The manual describes not only physical properties and environmental resistance to oxidation by also processing routes of manufacturing FeCrAl alloys, including traditional wrought methods and powder metallurgy processes, as well as final product forms, for example, thin-walled tube making. Table 5.1 shows the composition of a selected group of FeCrAl alloys, which have been recently investigated in the laboratory with the aim of nuclear reactor applications. Legacy alloys not included in Table 5.1 are MA956 (S67956) (Fe 1 20Cr 1 4.5Al 1 0.5Y2O3 1 0.5Ti), PM2000 (Fe 1 19Cr 1 5.5Al 1 0.5Y2O3 1 0.5Ti), and several earlier versions of Kanthal. Some of the FeCrAl alloys such as Aluchrom YHf and C26M are produced in the traditional way of melting, casting, forging, rolling, etc., and other alloys such as APMT and FeCrAl ODS are produced using powder metallurgy and extrusion (Jo¨nsson et al., 2004). C26M alloy in Table 5.1 was recently developed at Oak Ridge National Laboratory (ORNL, 2018a,b) and FeCrAl ODS was recently developed in Japan (Sakamoto, 2017). Both alloys were developed for nuclear applications. Two other FeCrAl powder metallurgy alloys, namely, Incoloy MA956 and PM2000, were used

TABLE 5.1 Nominal composition of FeCrAl in mass percentage, the remaining is iron. Alloy

Cr

Al

Others

APMT

21

5

3Mo

C26M

12

6

2Mo 1 0.05Y

FeCrAl NFD ODS

12

6

0.5Ti 1 Zr 1 Y2O3 1 Fe2O3

Aluchrom YHf

21

5.5

0.05Y, 0.03Hf, 0.3Si

APMT, Advanced powder metallurgy tubing; FeCrAl, Iron
chromium
aluminum; NFD, Nippon fuel development; ODS, oxide dispersion strengthening; Y, yttrium; Hf, hafnium.

90

Accident-Tolerant Materials for Light Water Reactor Fuels

sporadically in the past in the industry at large and were also briefly considered for fuel cladding in Generation IV reactors mainly as a result of their mechanical stability and oxidation resistance (Floreen et al., 1980; Kalvala et al., 2009; Soler-Crespo et al., 2009; Chen and Hoffelner, 2009; Chen et al., 2013). MA956 and PM2000 are now deemed obsolete or discontinued from the market owing to the lack of enough industrial demand. The main benefit of the FeCrAl family is their extraordinary resistance to high-temperature degradation by air and steam. The worldwide accidenttolerant fuel (ATF) programs were started because of a plant black out at the Fukushima Daiichi nuclear power stations and the rapid attack of the zirconium-based materials inside the reactor at accident condition temperatures above 1200 C. FeCrAl alloys do not experience the high-temperature attack by air or steam that the zirconium alloys undergo even at temperatures as low as 600 C. Fig. 5.3 of iron
chromium
aluminum (FeCrAl) alloys shows some of the desirable properties of these alloys for nuclear applications, including properties for LWRs.

FIGURE 5.3 Beneficial characteristics of iron
chromium
aluminum (FeCrAl) for nuclear applications.

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

91

1. Ferritic alloys FeCrAl (such as APMT and C26M) have low coefficients of thermal expansion (CTE), which is a beneficial property to avoid changes in geometric shapes during heating up and cooling down in the reactor core. Lower CTE will also reduce the generation of thermal stresses in the reactor components. 2. Ferritic FeCrAl alloys have higher thermal conductivity than zirconium alloys, allowing for a faster and more efficient transfer of the heat generated in the oxide fuel pellets in the ID to the water in the OD. The faster heat removal allows for maintaining a lower centerline temperature in the fuel pellets, which may render a longer fuel pellet life without crumbling or cracking. A positive outcome of this could be the possibility to extend the current allowed burn up of 62 GWd/MTU to 75 GWd/MTU. 3. Ferritic alloys such as FeCrAl are highly resistant to environmentally assisted cracking (EAC) from the coolant side, which is needed to avoid the release of fissile toxic elements into the coolant during breaching of the cladding hermeticity. Before the widespread use of zirconium alloys for fuel cladding, the commercial nuclear power plants used austenitic type 304SS alloy for the cladding of the fuel, which sometimes became sensitized during welding and suffered environmental cracking in service from the coolant side (Terrani, 2018). The waterside cracking of the type 304 SS fuel rods was the main reason the utilities shifted from type 304SS to zirconium alloys for the cladding. 4. Like other ferrous alloys containing chromium (i.e., stainless steels type 304 and nickel alloy Inconel 600), the FeCrAl alloys such as APMT have low general corrosion rates under normal operation and anticipated operational occurrences by the formation of a protective chromium oxide on the surface. The corrosion resistance is offered by the formation of a thin layer of chromium oxide on the surface of the reactor components even under hydrogen gas injection in the reactor. 5. FeCrAl alloys have higher mechanical properties than traditional zirconium alloys, which would allow for making tubes with thinner wall thickness to compensate for the ten times higher thermal neutron absorption of FeCrAl when compared with that of zirconium. 6. Owing to its open ferritic or body-centered cubic microstructure, FeCrAl alloys are more resistant to irradiation damage than typical austenitic alloys such as type 304SS and Inconel 600, which are face-centered cubic materials. The higher resistance to irradiation damage such as void swelling is desirable to avoid geometric distortion and decreased ductility in the reactor structural materials. 7. Ferritic FeCrAl alloys are manufactured using common and abundantly inexpensive elements such as iron (Fe), chromium (Cr), and aluminum (Al). These alloys do not contain the more expensive nickel element like many other current LWR structural materials have. Moreover, the presence of nickel may produce activated species, which could be released in

92

Accident-Tolerant Materials for Light Water Reactor Fuels

the coolant or as part of the turbine steam. The lack of nickel in FeCrAl is another positive attribute. 8. FeCrAl alloys offer an unparalleled resistance to high-temperature attack by steam at temperatures characteristic of design base accidents (DBA) or beyond design basis accidents (BDBA) (such as the Fukushima Daiichi accident). Up to approximately 1000 C
1100 C the FeCrAl alloys are protected by chromium oxide on the surface, in the same manner as other traditional materials such as type 304SS. However, at 1200 C and higher temperatures the synergistic effect of Cr and Al promotes the surface formation of a thin and protective alumina layer that is stable until the melting point of the alloy at around 1500 C. This high-temperature resistance to oxidation greatly reduces the formation of ignitable hydrogen gas and the release of exothermic oxidation heat, which may further accelerate the melting of the fuel cladding. The main challenges of the FeCrAl family of alloys is the ten times higher thermal neutron absorption cross section when compared with that of zirconium alloys (George et al., 2014, 2015; Powers, 2016) (Fig. 5.4). This higher neutron absorption cross section of FeCrAl is commonly referred as “neutron penalty” or “parasitic neutron capture.” Another challenge that many envision for FeCrAl is their transparency to atomic hydrogen across the cladding wall owing to their bcc structure. This may allow for the tritium generated in the fuel cavity to diffuse through the cladding wall into the coolant. Others do not see the tritium issue to be relevant because the presence of oxides in either the ID or the OD of the cladding may greatly minimize the release of tritium into the coolant.

Thermal properties of FeCrAl Table 5.2 shows the CTE, the thermal conductivity, and the melting point of the proposed FeCrAl materials and those of other traditional nuclear materials such as type 304SS and Zircaloy-2. More data are also available from the ORNL Manual on FeCrAl (ORNL, 2018a,b). The CTE for the ferritic FeCrAl is lower than the CTE for austenitic type 304 SS, which is a desirable property for minimizing dimensional changes of the near five-meterlong fuel rod. The CTE of FeCrAl is approximately double the value for Zircaloy alloys (ORNL, 2018a,b; Sakamoto et al., 2018) but lower than the CTE values for austenitic stainless steels. Table 5.2 also shows that the thermal conductivity of the ferritic FeCrAl at 500 C is higher than for the current material of Zircaloy-2, meaning that the heat generated by the fission reactions in the fuel can be more readily extracted and transferred across the cladding wall to the coolant water.

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

93

FIGURE 5.4 Benefits and detriments of iron
chromium
aluminum (FeCrAl) for accidenttolerant fuel (ATF) cladding.

Mechanical properties of FeCrAl Figs. 5.5
5.7 show the mechanical properties of APMT and FeCrAl ODS as well as traditional fuel cladding Zircaloy-2. The mechanical data for Zircaloy 22 were taken from reports published in 1961
62 (Whitmarsh, 1962; Mehan and Wiesinger, 1961), and the data for APMT was obtained at the University of Idaho using APMT bars or rods (Guria and Charit, 2016). One data point for APMT was taken from the Kanthal datasheet (Kanthal, 2018). Kanthal did not publish mechanical properties at temperatures higher than ambient. Data for FeCrAl ODS are from Sakamoto et al. (2018). Sun et al. (2018) measured the mechanical properties of two heats of APMT material at an ambient temperature. APMT Heat #1 had a bimodal grain microstructure of large grains surrounded by 10-μm-sized grains, while APMT Heat #2 had a more uniform grain size of 10
16 μm. They reported the yield stress for Heats #1 and #2 as 579 6 4 and 727 6 9 MPa, respectively. The ultimate

94

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 5.2 Coefficient of thermal expansion and thermal conductivity of FeCrAl and other nuclear materials. Alloy

CTE [ 3 1026 (K21)]

Thermal conductivity (W m21 K21)

Melting point ( C)

Zircaloy-2 R60802 (Whitmarsh, 1962; Murabayashi et al., 1975; Wah Chang, 2003)

27 C
100 C 5 5.83 27 C
200 C 5 5.99 27 C
300 C 5 6.16 27 C
400 C 5 6.3 27 C
500 C 5 6.46 27 C
600 C 5 6.61

0 C
100 C 5 13.8 0 C
300 C 5 15.3 0 C
500 C 5 17.9

1850

Type 304SS S30400 (ASM Metals Handbook, 1990)

0 C
100 C 5 17.3 0 C
315 C 5 17.8 0 C
650 C 5 18.7

0 C
100 C 5 16.2 0 C
500 C 5 21.5

1400
1455

APMT (Kanthal Datasheet, 2012)

20 C
250 C 5 12.4 20 C
500 C 5 13.1 20 C
750 C 5 13.6 20 C
1000 C 5 14.7 20 C
1200 C 5 15.4

50 C 5 11 600 C 5 21 800 C 5 23 1000 C 5 27 1200 C 5 29

1500

C36M (ORNL, 2018a,b)

100 C 5 11.6 300 C 5 12.2 500 C 5 12.8

0 C
100 C 5 13 0 C
300 C 5 16 0 C
500 C 5 20

1500

FeCrAl ODS (Sakamoto et al., 2018)

Like C36M

Like C36M

1503 6 1.5

Aluchrom YHf (VDM, 2008 Aluchrom Datasheet)

RT-100 C 5 12.2 RT-200 C 5 12.4 RT-300 C 5 12.6 RT-400 C 5 12.9 RT-500 C 5 13.3 RT-600 C 5 13.6 RT-700 C 5 13.8 RT-800 C 5 14.3 RT-900 C 5 14.8

30 C 5 9.8 100 C 5 10.9 200 C 5 12.4 300 C 5 13.9 400 C 5 15.5 500 C 5 16.9 600 C 5 18.2 700 C 5 19.7 800 C 5 21.1 900 C 5 22.5

1500

FeCrAl, Iron
chromium
aluminum; CTE, coefficients of thermal expansion; ODS, oxide dispersion strengthening; Y, yttrium; Hf, hafnium.

tensile strengths were 597 6 11 and 837 6 22 MPa for APMT Heats #1 and #2, respectively (Sun et al., 2018). The values of the mechanical properties by Sun et al. (2018) agree with the other published values, as shown in Fig. 5.5 and 5.6. Fig. 5.5 shows that at a temperature of 300 C
400 C, relevant to fuel cladding performance for LWRs, the yield stress of APMT and FeCrAl ODS

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

95

FIGURE 5.5 Yield stress for APMT and Zircaloy-2.

FIGURE 5.6 Ultimate tensile strength for APMT and Zircaloy-2.

is approximately six times higher than the yield stress of Zircaloy-2. Fig. 5.6 shows that the value of the ultimate tensile strength of APMT and FeCrAl ODS is approximately four times higher than for Zircaloy-2. The higher strength of APMT would allow for a thinner wall of the cladding tube to compensate for the higher thermal neutron absorption of FeCrAl alloys compared with that of Zircaloy-2. Fig. 5.7 shows that the elongation to failure of APMT, FeCrAl ODS and Zircaloy-2 in the temperature range 300 C
400 C are similar and in the order of 20%
30% elongation. The elongation

96

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.7 Elongation to failure for APMT and Zircaloy-2.

experienced by the ferritic materials is enough to allow for thin wall tube fabrication either by pilgering or by warm drawing.

Oxidation resistance of FeCrAl under LWR’s normal operation conditions Currently the tubes for the fuel claddings in LWRS are made with zirconium alloys. The aim of the accident-tolerant fuel programs is to replace the zirconium alloys for materials more resistant to oxidation to high-temperature steam and air (Terrani, 2018). Fig. 5.8 shows a configuration of the fuel components in a boiling water reactor (BWR) system. There are approximately 800 bundles of fuel per BWR core, and each bundle contains approximately 100 rods. Each fuel rod is manufactured using a zirconium alloy tube with urania pellets being stacked inside. The rods are bundled in groups of 100 using approximately 8 spacer grits (commonly made of alloy X-750), which keep the rods at equal distance from each other through their entire length of approximately 4 m. Each bundle is also enveloped at four sides by a box or channel that is also made of zirconium alloys. The channel boxes allow for the vertical flow of the cooling water from the bottom to the top of the bundles. The Zircaloy wetting area by the cooling water is approximately 1 3 108 cm2 for the rods and approximately 40 3 106 cm2 for the channels. The mass of Zircaloy inside a BWR reactor core is approximately 33 metric tons (Terrani, 2018). The operation conditions of a BWR are generally pure water at 288 C and a pressure of 7.2 MPa and for a pressurized water reactor (PWR) primary circuit the conditions are approximately 330 C water and 15 MPa

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

97

FIGURE 5.8 Boiling water reactor (BWR) fuel components.

pressure. In general, for laboratory testing, the BWR conditions are separated into normal water chemistry (NWC)—which may contain oxidizing species such as 1000 ppb of dissolved oxygen—and hydrogen water chemistry (HWC)—which may contain 300 ppb of dissolved hydrogen. The PWR coolant generally contains dissolved hydrogen gas (3.7 ppm) and additions of lithium hydroxide to regulate the pH of the water and additions of boric acid for reactivity control. It has been known since 1930 that FeCrAl alloys develop an alumina layer on their surface when exposed to high temperatures ( . 1000 C) in presence of either air or water vapor, but there was no information on what would happen to the FeCrAl if exposed to light water power reactor environments at temperatures near 300 C because these alloys were not intended initially for nuclear use. Therefore it was important to evaluate how FeCrAl would behave under normal operation conditions, in the case of a loss of coolant accident, and after the rods are quenched with fresh water. The general corrosion characteristics of FeCrAl alloys in out-of-pile simulated normal BWR and PWR operation conditions have been extensively investigated (Terrani et al., 2016; Rebak, 2018). Investigations on the normal operation conditions of FeCrAl alloys at General Electric were conducted by using the mass loss method, analysis of surface oxide films, and electrochemical methods. Table 5.3 shows a standard set of autoclaves for determining the mass change in FeCrAl coupons, both flat and tube coupons. All the tested FeCrAl seem to have a minimal mass loss in the first few weeks

98

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 5.3 Out-of-pile immersion tests for APMT alloys (Rebak, 2018). Autoclave

Test conditions

Oxide films on the surface

S-2

Simulated PWR, highpurity water, 3.75 ppm H2, 330 C

Single layer, Cr rich, 10
100-nm thick

S-5

Simulated BWR, HWC (0.3 ppm H2), 288 C

Single layer, Cr rich, 100
120-nm thick

S-6

Simulated BWR, NWC (1 ppm O2), 288 C

Double layer, total 150- to 180-nm thick, external layer is Fe and Cr rich and internal layer is Cr rich and only 10- to 15-nm thick.

of testing until a protective barrier of chromium oxide is established on the surface of the coupons. After the chromium oxide film is formed on the surface of the FeCrAl alloys, the mass change does not significantly change in time for immersions periods of one year. Under hydrogen conditions, the surface oxide is a 10-nm-thick layer of pure chromium oxide without the presence of any iron, molybdenum, or aluminum (Rebak, 2018). This means that for the chromium to form a pure chromium oxide on the surface, some aluminum, molybdenum, and iron must dissolve into the water. In the presence of oxygen, the surface oxide seems to be a bilayer structure, with an external thicker layer containing iron and chromium and an internal thinner layer containing only chromium oxide. Under normal operation conditions, in hydrogen or oxygen atmosphere the oxide on the surface does not contain either aluminum or molybdenum (Rebak, 2018). PWR, Pressurized water reactor; BWR, boiling water reactor; HWC, hydrogen water chemistry; NWC, normal water chemistry. For conditions containing either 300-ppb dissolved hydrogen or 1000-ppb dissolved oxygen, Fig. 5.9 shows the mass change of two flat coupons of APMT made from billet material for either oxygen or hydrogen containing water as a function of the immersion time. Fig. 5.9 shows that at the same temperature, the mass change in the coupons was different depending on whether the environment is reducing (excess hydrogen) or oxidizing (excess oxygen). Under oxidizing conditions, the APMT coupons gained initially some mass by the development of a mixed oxide containing iron and chromium. However, as the internal Cr-rich protective oxide layer develops, the external oxide layer may slowly dissolve in the 288 C water, losing mass as the immersion time increases. In the excess hydrogen environment at 288 C, the coupons lost mass in the first 90 days of exposure but, as the protective chromium oxide film is developed on the surface, the mass loss under reducing conditions stopped and remained constant for the remaining one-year

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

99

FIGURE 5.9 Mass change in boiling water reactor (BWR) testing conditions for APMT flat coupons.

FIGURE 5.10 Mass change in boiling water reactor (BWR) hydrogen water chemistry (HWC) for iron
chromium
aluminum (FeCrAl) tube coupons.

immersion time (Rebak, 2018). The mass loss for AMPT flat coupons under HWC in BWR type environment stabilized at 20 μm/cm2 (Fig. 5.9). Fig. 5.10 shows the mass change for tube coupons as a function of the immersion time for up to 9 months (39 weeks) immersion in BWR HWC conditions at 288 C. The tube coupons included three types of FeCrAl alloys (as shown in Table 5.1) and Zircaloy-2. While Zircaloy-2 tube coupons gained approximately 14 mg/dm2 after 9-months immersion, all the FeCrAl

100

Accident-Tolerant Materials for Light Water Reactor Fuels

alloys lost small amounts of mass in the same hydrogenated water environment. The lowest mass loss was noted for APMT coupons (B18 mg/dm2) probably because APMT had the highest chromium content of the three FeCrAl alloys. The mass loss for APMT tube coupons were slightly higher than for the flat coupons made from billet material in Fig. 5.9. Fig. 5.10 shows that the C26M and FeCrAl ODS tube coupons lost approximately between 600 and 800 μg/cm2 after 9-months immersion in the same electrolyte. After 9-months immersion, it appears that the mass loss for the three FeCrAl tube coupons is not linear in time. However, the change in mass loss for APMT is much lower than the change for the lower chromium C26M and ODS Nippon fuel development (NFD) tube coupons. Fig. 5.11 shows the mass change for three FeCrAl alloys tube coupons in Table 5.1 as well as for Zircaloy-2 tube coupons in the BWR pure water containing 1 ppm dissolved oxygen at 288 C. The Zircaloy-2 tube coupons gained higher mass in the oxygenated BWR water (NWC) than in the hydrogenated BWR water (HWC) (Fig. 5.10). On the other hand, in the NWC, the FeCrAl tube coupons experienced smaller mass change than in the HWC. In BWR NWC both APMT and C26M lost small amounts of mass, while the ODS NFD coupons gained a negligible amount of mass. The highest amount of mass loss for the C26M coupons was 10 mg/dm2. It is likely that the oxygen present in the water facilitated the early passivation of the FeCrAl coupons and therefore the mass loss by dissolution of elements such as Mo and Al into the water was minimized. According to Table 5.1 the main difference between the ODS NFD coupons and the C26M coupons is the absence of molybdenum in the ODS material. It may seem apparent that molybdenum

FIGURE 5.11 Mass change in boiling water reactor (BWR) normal water chemistry (NWC) for iron
chromium
aluminum (FeCrAl) tube coupons.

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

101

in the alloy may initially promote mass loss until a protective layer of chromium oxide is formed on the surface. The formation of the protective chromium oxide layer may be delayed in the HWC when compared with the NWC at the same temperature. Fig. 5.12 shows the mass change for three FeCrAl alloys in Table 5.1 as well as for Zircaloy-2 tube coupons in the PWR type of water containing 3.75 ppm dissolved hydrogen at 330 C. The Zircaloy-2 tube coupons again gained mass to approximately 24 mg/dm2 after 9-months immersion, which was higher than for the BWR hydrogen atmosphere at 288 C (Fig. 5.10). All the FeCrAl tube coupons lost mass in the hydrogenated PWR water at 330 C. The lowest mass loss was approximately 12 mg/dm2 for APMT probably because of its higher content in chromium. The mass loss for APMT tube coupons was similar between the BWR (288 C) and PWR (330 C) hydrogen water chemistries. However, for the C26M tube coupons, the mass loss in PWR hydrogenated environment was lower than in the BWR hydrogenated environment probably because higher temperature accelerated the formation of the protective chromium oxide on the surface of the coupons and therefore reducing the dissolution rate. Fig. 5.13 shows the appearance of two coupons per each alloy in each of the three tested conditions in Table 5.3. The top row shows coupons for Zircaloy-2 and the four FeCrAl tubes from autoclave S-2 or PWR simulated hydrogenated water at 330 C. The middle row is for the tube coupons for the hydrogenated BWR type of water at 288 C and the lower row is for the BWR oxygenated water at 288 C. Most of the coupons do have a shiny dark gray or black appearance except for the C26M and two type of ODS coupons

FIGURE 5.12 Mass change in pressurized mium
aluminum (FeCrAl) tube coupons.

water

reactor

(PWR)

for

iron
chro-

102

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.13 Six-month immersion tube coupons.

in the oxygenated 288 C environment (BWR NWC), which show a golden type of appearance.

Composition of the oxide films on FeCrAl coupons Fig. 5.14 shows the composition of the oxide film formed on APMT flat billet coupons after 1-year immersion in BWR NWC conditions or pure water containing 1 ppm of dissolved oxygen at 288 C. The oxide film shows two oxide layers of approximately 200-nm thickness. The external layer is approximately 180-nm thick and contains iron and chromium. The internal layer contains approximately 10% of the total thickness (15 nm) and is composed solely of chromium oxide. The dual layer oxide film does not contain either Mo or Al, suggesting that these two elements were preferentially leached out or dissolved into the water. Fig. 5.15 shows the composition of the oxide on APMT flat billet coupons immersed in PWR-type environment at 330 C for one year. The surface oxide is a single layer of chromium oxide with approximately 10- to 100-nm thickness. In the PWR environment the oxide does not contain Fe, Mo, or Al, which may have preferentially dissolved into the water. Fig. 5.16 shows the oxide characteristics formed on the OD the APMT tube coupon after exposure for 6 months in simulated hydrogenated PWR

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

103

FIGURE 5.14 Oxide composition of APMT in boiling water reactor (BWR) normal water chemistry (NWC).

water at 330 C shown in Fig. 5.13. The maximum thickness of the oxide formed on APMT was 100 nm and it contained mostly chromium. That is, the iron, aluminum, and molybdenum probably dissolved preferentially in the circulating water in the autoclave system. Fig. 5.17 shows the oxide characteristics formed on the C26M coupon after exposure for 6 months in simulated hydrogenated PWR water at 330 C shown in Fig. 5.13. The thickness of the oxide formed on C26M was 1
1.5 μm and it contained mostly chromium. The iron and molybdenum probably dissolved in the circulating water in the autoclave. The C26M tube coupon contained oxide crystal deposited on the surface, which could have originated from the dissolution of the C26M or from other components in the autoclave system. The oxide in the C26M tube was ten times thicker and may contain more aluminum than the oxide on APMT. Another difference between the behavior of APMT and C26M tubes in PWR type of water at 330 C is the surface characteristics shown in Fig. 5.18. AMPT has practically no precipitates on the surface, while C26M has many iron-rich crystals.

104

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.15 Oxide composition of APMT in pressurized water reactor (PWR).

FIGURE 5.16 Oxide APMT tube coupon pressurized water reactor (PWR).

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

105

FIGURE 5.17 Oxide C26M tube coupon pressurized water reactor (PWR).

FIGURE 5.18 Oxide top surface pressurized water reactor (PWR) APMT and C26M.

Immersion corrosion tests shows that APMT and C26M will resist general corrosion under normal operation conditions in both BWR and PWR environments by the formation of a protective chromium oxide film on the surface. This is the same mechanism that protects other commonly used alloys in LWRs for the last six decades such as type 304SS, X-750, and Inconel 600 (Rebak et al., 2017a,b). Both C26M and APMT will be resistant to general corrosion under normal operation conditions in both BWR- and PWR-type reactors.

106

Accident-Tolerant Materials for Light Water Reactor Fuels

Electrochemical behavior of FeCrAl alloys in hightemperature water Inside the reactor, all metallic components adopt their own electrochemical or corrosion potential, which is controlled by the nature or chemical composition of the alloy and the environment including temperature, pH, and presence of ionic species in the water and dissolved gases (oxidizing or reducing). Because FeCrAl alloys have never been used before in LWRs, it was important to characterize their electrochemical behavior relatively to the well-known behavior of legacy alloys such as Zircaloy and austenitic chromium-containing materials such as type 304SS and nickel alloy X-750 (Rebak et al., 2019a,b). Electrochemical measurement studies were performed in pure water at 288 C and 10-MPa pressure in a 1-gal (3.78 L) stainless steel autoclave at a water flow rate of 100 cm3/min. The specimens for the electrochemical testing were strips (0.5 cm wide and 2.5 cm long) polished using a wet 600-grit emery paper. Water in the loop was purified to 18 MΩ  cm and it was equilibrated with appropriate mixtures of argon (Ar), oxygen gas (O2), or hydrogen gas (H2) to establish the desired water chemistry. Oxygen and hydrogen monitors were used for measuring the dissolved oxygen and hydrogen in the inlet and outlet streams, respectively. The studied aqueous environments were simulated reactor environments with the addition of only hydrogen or oxygen to high-purity water and without the addition of lithium or boron. The pH of high-temperature water (near 300 C) is approximately 5.6. A platinum (Pt) electrode was used to monitor the effective redox potential of the water. The materials used to manufacture the specimens for the electrochemical measurements are listed in Table 5.4. Fig. 5.19 shows the corrosion potential behavior of five engineering alloys and platinum for six weeks in high-temperature water at 288 C. The potential measurements were performed using a zirconia high-temperature pH sensor with a copper/copper oxide (Cu/Cu2O) internal junction as a reference electrode and an Agilent Model 34970A electrometer. The measurements were taken every 10 min. The measured potential values were converted to the standard hydrogen electrode (SHE) scale. For the first approximately 480 h of immersion, the water contained dissolved hydrogen gas (300-ppb H2), for the second period (480
713 h) the water contained dissolved oxygen gas (1000-ppb O2), and for the last period the water was reversed back to 0.3 ppm of hydrogen. Under hydrogen conditions, the corrosion potential for the six materials was low and between 2600 and 2800 mV (SHE). The maximum separation among the corrosion potential for all the materials was generally in a band of less than ,50 mV. At approximately 480 h of immersion, the gas was changed from hydrogen to oxygen, and a sudden increase in the corrosion potential for all the materials was observed (Fig. 5.19). While in the hydrogen environment, the potential

TABLE 5.4 Alloys for electrochemical measurements. Element

Zircaloy-2 HCP

304SS Austenitic

Alloy 33 Austenitic

X-750 Austenitic

C26M Ferritic

APMT Ferritic

, TB . Zr

B 98
















Fe

0.07
0.2

B 70

32

7

B 80

B 70

Ni

0.03
0.08

8

31

B72







Cr

0.05
0.15

19

B 33

15

12

22

Al










0.7

6

5

Mo







1.6




2

3

Si




0.75 max

0.5 max

0.5 max

0.2

0.7 max

Mn




2 max

2 max

1 max




0.4 max

C




0.08 max

0.015 max

0.08 max

0.01 max

0.08 max

Other

1.5 Sn




0.6 Cu, 0.4 N

2.5 Ti, 1 Nb

0.05 Y

Y, Hf

Notes: Compositions are nominal in mass percentage. HCP, Hexagonal close-packed; Ni, nickel; Mo, molybdenum; Si, silicon; Mn, manganese; C, carbon; Sn, tin; Cu, copper; N, nitrogen; Ti, titanium; Nb, niobium; Y, yttrium; Hf, hafnium.

108

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.19 Corrosion potential of iron
chromium
aluminum (FeCrAl).

of all the metallic materials was approximately the same, in the oxygencontaining water, there was a clear separation of the corrosion potential values. The highest (noblest) potential was for platinum and the lowest (less noble) was for Zircaloy-2. The potentials of all the other materials—including nickel-based X-750, austenitic stainless-steel type 304, and the two newer ferritic FeCrAl materials (APMT and C26M)—were similar to each other and near 0 V SHE. This is not surprising because these four alloys contain chromium (Cr) for passivation, and therefore the four alloys develop on the surface a chromium-rich film. On the basis of the corrosion potential data, it is evident that the redox kinetics on the ferritic FeCrAl are like the wellknown behavior of traditional austenitic nuclear reactor materials, such as Type 304SS and X-750, in high-temperature water. Figs. 5.20 and 5.21 (Kim et al., 2015b) show earlier and shorter term similar data as that given in Fig. 5.19. Again, in the oxidizing environment, there was a separation of the corrosion potentials (Pt highest and Zircaloy-2 lowest), and the other materials—including APMT, nanoferritic 14Cr alloy (NFA), Alloy 33, nickel alloy X-750, and Type 304SS—were all in a narrow band of less than 100 mV. In the hydrogen atmosphere, the corrosion potential of all the tested materials (including Pt) were grouped in a narrow band of less than 50 mV and in the order of 20.6 V (SHE).

Shadow corrosion In BWRs, an enhanced growth of zirconium oxides may occur on areas of zirconium alloy components that are “facing” other metals such as nickelbased alloys and stainless steels. This anomalous growth in the oxide of the

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

109

FIGURE 5.20 Corrosion potential in oxygen.

FIGURE 5.21 Corrosion potential in hydrogen.

Zr alloy is called “shadow” corrosion because the shape of the higher corrosion area on the zirconium alloy component resembles the imprint of the nearby stainless steel or nickel alloy metallic component (Kim et al., 2010). Two hypotheses have been used to explain this phenomenon: (1) galvanic

110

Accident-Tolerant Materials for Light Water Reactor Fuels

corrosion and (2) local radiolysis; however, there is evidence that these two hypotheses may not explain all the occurrences of shadow corrosion. Lysell et al. (2005) argue that the most likely cause of shadow corrosion is the theory of galvanic corrosion. Shadow corrosion in reactors has been explained by a separation in the corrosion potential of the materials under irradiation. Under irradiation, the corrosion potential of nickel alloys such as X-750 and Zircaloy-2 go in separate directions, the potential of the Zircaloy becomes more cathodic, and the potential of the nickel alloy becomes more anodic (Kim et al., 2010). In current reactors, the fuel cladding is Zircaloy, and other components such as the separation grits and bundle handles generally are stainless steels or nickel-based alloys. FeCrAl (APMT or C26M) is proposed to be used for the cladding of the fuel rods instead of Zircaloy; however, the current stainless steels and nickel-based alloys components in the reactor are expected to remain the same. Therefore it would be important to determine how the effect or irradiation might affect the behavior of FeCrAl alloys when compared with that of the current Zircaloy materials. The plant irradiation behavior has been simulated in laboratory autoclaves using ultraviolet (UV) illumination (Kim et al., 2010, 2015a). However, the energy of UV irradiation used was too low to produce radiolytic species of the water, which do exist in an actual reactor environment. The present UV treatment was meant to study only the irradiation response of the oxide films on the surface of the alloys. The test specimens were irradiated through a sapphire window from a penetration in the bottom of the autoclave. A thin diamond disk made by chemical vapor deposition was placed on top of the sapphire window to protect it from dissolution in high-temperature water and did not affect the UV intensity (Kim et al., 2015a; Rebak et al., 2019a,b). Fig. 5.22 shows the corrosion potential behavior of three alloys in high-purity water with 1 ppm oxygen at 288 C. In the absence of illumination, the corrosion potential of APMT and X-750 was very close at near 0.1 V (SHE), while the corrosion potential of Zircaloy-2 was approximately 100 mV lower. During the period of UV illumination, the corrosion potential of both X-750 and APMT became more anodic (noble); that is, they both moved in the same direction. However, the corrosion potential of Zircaloy-2 became more cathodic (less noble). This separation in the corrosion potential in the presence of irradiation may explain the occurrence of shadow corrosion on zirconium in boiling water reactor nuclear power plants (Kim et al., 2010). The changes in the corrosion potential in the presence of UV light are due to the photo-excitation of n-type zirconium dioxide (ZrO2) formed on Zircaloy-2 and p-type nickel oxide (NiO) on X-750. If a FeCrAl tube is used for the ATF cladding, shadow corrosion is not expected to occur in the proximity of a X-750 component because under irradiation the corrosion potential of FeCrAl and nickel alloys such as X-750 would be practically the same (Fig. 5.4).

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

111

FIGURE 5.22 Corrosion potential under ultraviolet (UV) irradiation showing a separation in the potentials of Zircaloy and X-750, but there is no separation between APMT and X-750.

Galvanic corrosion Galvanic corrosion may occur when two different metals (alloys) enter into contact (or are connected by an electrical conductor) while exposed to an ionic corrosive electrolyte. If a difference in the corrosion potential exists between the two metals, this may be the driving force for and increased corrosion of one of the metals in the couple. Tests were conducted in which APMT and Zircaloy-2 were each connected to X-750 in a typical 288 C water with 1-ppm dissolved oxygen to simulate reactor conditions. The coupled electrodes for galvanic corrosion measurement had a gap of approximately 1 mm between an anode electrode (Zircaloy-2 or APMT) and the cathode electrode (X-750) (Kim et al., 2015a). Fig. 5.23 shows that an anodic galvanic current was measured when Zircaloy-2 was galvanically coupled with X-750 in a hot-water environment. This anodic galvanic current is mainly attributed to a larger and positive difference in the corrosion potential between Zircaloy-2 and X-750 in water with oxygen (Fig. 5.19). The anodic galvanic current for the dissolution of Zircaloy-2 was increased when the UV illumination was on, showing the increased oxidation of Zircaloy in the presence of irradiation (Kim et al., 2015a). On the contrary, the galvanic current between APMT and X-750 was cathodic; that is, the X-750 was slowly preferentially corroding in the presence of APMT, however with a smaller absolute current than when X-750 was coupled to Zircaloy. This is a result of the closer values of corrosion potentials between APMT and X-750,

112

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.23 Galvanic Current of Zirc-2 and APMT coupled to X-750 in 288 C water plus 1 ppm O2 with and without ultraviolet (UV) illumination. Under UV irradiation, there is an increase in the anodic current for Zircaloy-2 but no change in the cathodic current for APMT.

as shown in Fig. 5.19. The galvanic current between X-750 and APMT was not affected by the UV illumination. These galvanic corrosion results suggest that the FeCrAl alloy APMT as an ATF cladding material would be less susceptible to enhanced shadow or galvanic corrosion when it is electrically connected to a X-750 spacer grid or any other nickel alloy or stainless steel component —such as the fuel bundle handles—in the reactor.

Resistance to crud deposition under normal operation conditions In the zirconium and uranium dioxide fuel rod system, it may be likely that a scale buildup would happen on the wall of the cladding in contact with the coolant owing to the presence of iron, chromium, and nickel ions in the water. The buildup on the OD of the cladding could be linked to the heat transfer across the cladding wall (Olander, 2009). Crud deposition is undesired because it decreases the heat removal rate from the rods by the coolant. The crud deposition in power plants has been curtailed in the last couple of decades by the strict control of the chemical composition of the coolant. At this moment it is not known how the magnetic ferritic alloys APMT and C26M would behave regarding crud deposition from the coolant side. Testing is necessary to determine whether FeCrAl materials would be more or less prone to crud deposition than the current zirconium alloys. In reactors with zirconium alloy fuel rods, the main dissolved elements in the water that later contribute

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

113

to crud deposition are nickel and iron. If the cladding material is changed to FeCrAl, besides nickel and iron, it is expected that the water may also contain dissolved molybdenum and aluminum. At this moment, it is not known how these newer elements dissolved in the coolant may impact the environmental performance of the metallic components in the reactor.

Resistance to EAC of ferritic alloys under LWR normal operation conditions EAC, also known as stress corrosion cracking (SCC), is the main failure mode of current LWR’s internal components such as austenitic stainless steels and nickel-based alloys. Current fuel rods with zirconium alloy cladding do not fail by EAC from the waterside. EAC is a mechanism by which normally ductile materials may fail in a brittle manner when exposed to a specific environment in the presence of tensile stresses. EAC of engineering reactor’s internal component alloys at B300 C light water environment of nuclear reactors has been studied extensively for many decades. Austenitic stainless steels such as types 304, 321, 347, and 316 as well as nickel-based alloys 600 and X-750 all suffer from EAC in LWR environments, especially in the cold worked condition and in the irradiated condition (Arioka et al., 2006; Ford et al., 2006; Scott and Combrade, 2006; Andresen and Morra, 2008; Hojn´a, 2013). EAC of austenitic stainless steels fuel cladding from the coolant side seems to be the main reason the use of austenitic stainless steels for fuel cladding was discontinued in the 1960s (Terrani, 2018). The EAC failure of cladding mainly happened in the areas of welding and it was attributed to the sensitization of the then (decades old) higher carbon containing austenitic stainless steels. In the 1960s, when the price of zirconium became affordable, power plant operators shifted from type 304SS to zirconium alloys for fuel cladding material mainly because of the zirconium lower neutron absorption cross section. Recently, the susceptibility to EAC of ferriticchromium-containing steels was tested in a simulated reactor environment and it was found that ferritic steels are highly resistant to cracking, mainly as a result of their bcc microstructure (Andresen et al., 2012, 2014; Rebak, 2013). Andresen (2014) also showed that even a previously high-dose irradiated ferritic HT9 specimen was resistant to EAC/SCC in high-temperature water. Ferritic stainless steels are likewise resistant to the traditional chloride cracking (Sedriks, 1996), which is a troubling failure mechanism for many austenitic stainless steels in chloride-containing environments. As part of the accident-tolerant fuel project at General Electric, Andresen et al. (2014) tested the susceptibility to cracking of FeCrAl alloys such as APMT and found that this family of ferritic alloys were also resistant to cracking in typical simulated light water environments. Fig. 5.24 shows the result of the APMT CT specimen c648, which was 23% cold worked (to increase its susceptibility to cracking) in 288 C water containing 30-ppb

114

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.24 Stress corrosion cracking (SCC1) test of APMT in 288 C oxygenated water.

sulfate and 2 ppm dissolved oxygen and at a stress intensity of 30 ksi  Oin (33 MPa  Om) and a high R ratio of 0.5. These conditions of testing are aggressive, considering previous SCC testing for austenitic alloys in nuclear power environments. The blue line in Fig. 5.24 shows the crack length of APMT specimen c648 as a function of the testing time. The pink and red lines are the corrosion potentials of platinum and of the APMT specimen, respectively. The corrosion potentials are high and between 1200 mV and 1300 mV SHE because the water contained dissolved oxygen, which may simulate the presence of radiolytic species in the upper region of a BWR. The green line shows the outlet conductivity of approximately 0.3 μS/cm of the aqueous solution before filtering. The testing water recirculated through the autoclave at a flow of 100 cm3/min (Andresen et al., 2012, 2014). Fig. 5.24 shows that at a loading frequency of 0.03 Hz at 2733 h of testing, the crack growth rate was 1.5 3 1027 mm/s and as the frequency of the applied load was decreased from 0.03 Hz to 0.01 Hz at 2766 h and to 0.001 Hz at 3324 h, the crack propagation rate decreased from to 1.1 3 1028 mm/s to 3 3 1029 mm/s, and to an undetectable value when the load was made constant. This means that crack growth is not sustainable under noncyclic or constant load conditions. Fig. 5.25 shows test results for the same specimen c648 for a testing period between 8500 and 14500 h in which the stress intensity factor has been increased to 40 ksi  Oin (44 MPa  Om) to promote crack growth. Again, the crack growth rate was reduced from 4 3 1028 mm/s to 2.3 3 1028 mm/s to 1.7 3 1028 mm/s to 7.7 3 1029 mm/s, and finally to 5 3 10210 mm/s as the frequency was decreased from 0.4 mHz

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

115

FIGURE 5.25 Stress corrosion cracking (SCC) test of APMT in 288 C oxygenated water.

at time 8436 h to 0.1 mHz at 12583 h and to constant load hold at 12918 h. For comparison, Rebak (2013) reported that under a constant load testing condition, austenitic type 304SS had a crack growth rate of 3 3 1027 mm/s for a stress intensity of 25 ksi  Oin (27.5 MPa  Om). That is, FeCrAl alloys as well as other ferritic alloys are resistant to EAC, even in the cold worked condition, under a high applied load, and in the presence of oxidizing conditions in high-temperature water.

Resistance to fretting under normal operation conditions The fuel rods (cladding tubes filled with uranium pellets) are assembled in bundles of approximately 100 rods (Fig. 5.8). In each bundle the tubes are spatially organized using grid spacers which keep the tubes equidistant from each other (Olander, 2009). The grid spacers should allow for tube sliding owing to normal thermal expansion. The grids are generally made using nickel alloy X-750 (N07750) and sometimes during the upward flow of the coolant, debris such as wires and washers may be trapped in the grids and touch the cladding repeatedly because of the typical vibration of the tube bundles during normal operation of the plant. This repetitive touching between debris and the tube can lead to the perforation of the tube to cause the release of fission products into the coolant. The perforation of the tube during vibration is called fretting failure, which is the most common fuel rod failure in the USA according to the Electric Power Research Institute (EPRI; Schneider et al., 2018).

116

Accident-Tolerant Materials for Light Water Reactor Fuels

Sakamoto et al. (2018) performed comparative 1-mm sliding wear tests for FeCrAl ODS (Table 5.1) and Zircaloy-2 in dry and wet environments both against type 304 SS using a contact force of 19.6 N and a frequency of 10 Hz. For the 90-min wet test, they reported a considerably higher wear depth of the Zircaloy-2 specimen (with a depth of 0.2 mm) when compared with the FeCrAl ODS specimen (with a depth of ,0.01 mm). That is, the use of FeCrAl cladding in replacement of the Zircaloy cladding may greatly reduce the fuel cladding perforation failures by fretting (Fig. 5.4).

Resistance of monolithic FeCrAl cladding to thermal shock In the case of a loss of coolant accident in a LWR, the cladding material may be exposed to temperatures in the order of 1200 C or higher and then fresh water may be pumped in to quench the core using the emergency core cooling system (ECCS). The cladding material will suffer a thermal shock during the reflooding with fresh water. It is important to determine whether the IronClad or FeCrAl material keeps the structural integrity and the mechanical properties upon the emergency quenching. That is, it is important to know if the cladding would shatter on the ingress of ambient temperature water or if the cladding will contain the fuel pellets without the release of radionuclides into the water. Mechanical test specimens were exposed to 1200 C for 2 h in air and in steam and then quenched in fresh water at ambient temperature (Dolley et al., 2018; Schuster et al., 2019). Fig. 5.26 shows the stress strain data for APMT tensile specimens at ambient temperature in the following conditions: (1) As machined or as-received, (2) After preexposure to air at 1200 C for 2 h and then quenched in water at ambient temperature, and (3) After pre-exposure to steam at 1200 C and then quenched in ambient temperature water. The mechanical testing of the as-received APMT material (blue lines) shows high strength and a good elongation to failure ( . 25%). When the specimens were pre-exposed to air at 1200 C for 2 h and then quenched in water (orange lines), the elongation to failure remained high (higher than 25%) but the mechanical properties decreased by approximately 100 MPa. When the specimens were pre-exposed to steam at 1200 C for 2 h and then water quenched (green lines), the elongation to failure decreased to approximately 10% but the mechanical strength remained the same as for the specimens exposed to air. One of the reasons for the decreased mechanical strength of the 1200 C exposed specimens was probably the grain growth experienced by the APMT during the 1200 C excursion. However, even after the 2 h treatment at 1200 C, the APMT material had enough strength and ductility (toughness) to be able to contain the fuel pellets during a fresh water flooding following a LOCA accident scenario. Fig. 5.27 shows the fracture surface of APMT after the tensile tests at ambient temperature, both for as-received specimens and for specimens that were exposed to steam at 1200 C for 2 h and then water quenched. The

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

117

FIGURE 5.26 Stress strain for APMT in quenching tests.

FIGURE 5.27 APMT fracture surface.

as-received top row of images shows a ductile failure, with considerable necking and a dimpled fracture surface. However, for the specimens preexposed to steam the tensile failure happened with little or no necking and

118

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.28 Grain growth for APMT.

the fracture surface exhibited a cleavage like brittle fracture appearance. It is likely that some of this decrease in total strength and reduced elongation was due to grain growth during the 1200 C 2 h treatment (Fig. 5.28). In the asreceived APMT tensile tested material the grain size was approximately 10 μm diameter and after the 2 h exposure to 1200 C steam the grain size increased to approximately B40 μm. It has also been shown that after quenching in ambient temperature water, the alumina layer formed during the residence in steam at 1200 C stayed on the surface of the quenched specimen (Schuster et al., 2019). That is, there is no spallation of the protective alumina during initial flooding with fresh water.

Interaction between the urania fuel and the FeCrAl cladding Urania fuel has been used in zirconium alloys tubing to make fuel rods for LWR since the mid-1950s. During its life in the reactor, the fuel may swell and exert hoop stresses onto the cladding which may or may not relax the stresses through creep. This phenomenon is called pellet-cladding interaction (PCI) (Cox, 1990). Moreover, the fission products from the fuel may react chemically with the cladding wall as well. A typical example of the mechanical and chemical interaction between the fuel and the cladding is the cracking of zirconium alloys in the presence of liquid iodine formed during the fission process in the fuel space (Cox, 1990). In BWRs the iodine-induced cracking of Zircaloy-2 has mostly been resolved by the introduction of a softer zirconium liner in the ID of the tube, which can better accommodate the swelling of the fuel (especially during power ramping) without introducing too high of a strain into the cladding wall. Currently there is no information on the effect of fission products elements such as iodine, cadmium, or cesium on the integrity of the FeCrAl cladding. Irradiation exposures both at commercial nuclear power plants and at the advanced test reactor (ATR) in the Idaho National Laboratory may

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

119

demonstrate the compatibility of FeCrAl IronClad with any of the fission products. Legacy data from the 1960s showed no direct interaction between urania and FeCrAl cladding at temperatures highly than 1000 C, mainly because an alumina film formed on the cladding wall and avoided a direct reaction of the underlying FeCrAl metal with the urania elements. Current out-of-pile tests showed similar findings related to the protective alumina layer in the ID of the cladding acting as a barrier between the cladding and the fuel. Sakamoto et al. (2018) performed compatibility tests between urania and FeCrAl ODS (Table 5.1) both at 1450 C (or just 50 C below the melting point of FeCrAl) and at 1580 C (or 80 C above the melting point of FeCrAl ODS). The urania pellets were in direct physical contact with the FeCrAl ODS specimen in an alumina boat in an inert oxygen-free environment. After 25-h exposure at 1450 C the sandwiched specimen of FeCrAl ODS and urania showed no interaction because a 4-μm-thick layer of alumina formed on the surface of FeCrAl ODS, which prevented any reaction of FeCrAl with urania. In the test performed at 1580 C, the FeCrAl ODS melted and there was some diffusion of Al and Fe into the urania pellet (Sakamoto et al., 2018).

Oxidation resistance of FeCrAl in high-temperature gas environments FeCrAl and Kanthal alloys have been used for high-temperature applications (above 1000 C) for nearly nine decades. FeCrAl alloys are the typical example of a material designed for a highly specific niche application. Very few engineering alloys contain metallic elements that give them intrinsic thermodynamic resistance to corrosion. Most of the alloys rely on kinetics effects, for example, by suppressing the cathodic reaction or by the formation of a thin oxide layer on their surface to considerably slow down the attack by the environment. This strategy takes advantage of the protection given by an initial controlled corrosion (or the formation of a protective oxide layer) that later minimizes the effect of the environment on the component. It is not that common stainless steels or galvanized steel do not corrode in typical condensed aqueous solutions; it is that the overall corrosion rate has been slowed down to imperceptibly low numbers for practical engineering lifetime components’ applications. The same principle of protection by surface oxides in high-temperature applications applies to the FeCrAl alloy system in dry gaseous environments (General Electric, 1965; Stott et al., 1995). The initial formation of chromium oxide on the surface facilitates the formation of an inner alumina film underneath the chromia because aluminum needs lower activity of oxygen to form its oxide than chromium does. This synergistic cooperation between chromium and aluminum is generally referred as the third element effect.

120

Accident-Tolerant Materials for Light Water Reactor Fuels

Chromium oxide provides the right environment for alumina to develop on the entire surface of the component even though the concentration of aluminum in the alloy may be only 5
6%. The surface oxides (e.g., alumina) have high thermodynamic stability compared to the elements in the alloy (e.g., aluminum), they have high melting points, and are a powerful barrier for diffusion of aggressive species to continue the attack (Stott et al., 1995). It is understood that the alumina that first forms under the chromia scale (especially at T , 1000 C) is a transient type and may suffer a solid-state transformation as a function of time, eventually becoming alpha alumina or corundum, which is the protective scale (Go¨tlind et al., 2007; Li et al., 2018; Yamamoto et al., 2015). The early transient or unstable alumina types are classified as cubic γ-, tetragonal δ-, and monoclinic θ-Al2O3 and they contain a larger number of defects than the stable α-alumina. The unstable alumina phases transform to the alpha stable phase for longer times at 1000 C and lower temperatures or rapidly at 1200 C. The alpha phase is a hexagonally close-packed oxygen-ion crystal structure and with aluminum ions filling two-thirds of the octahedral interstices (Stott et al., 1995). The transformation of the lower temperatures’ unstable forms to the alpha alumina occurs more rapidly in ferritic alloys (such as the FeCrAl) than in the austenitic alloys such as the Haynes 214 nickel
chromium
aluminum alloy. Once alumina forms on the surface, and as the temperature rises above 1000 C, the alloy maintains its resistance against attack by the steam or air environment until the melting point of the FeCrAl alloys (B1500 C). One of the concerns about the use of FeCrAl alloys in high-temperature ( . 1000 C) to ambient-temperature cyclic applications is that if spallation of the protective alumina happens during the cooling down and then the alumina layer reforms upon heating up, the aluminum in the alloy may be consumed and reduced to a concentration below the critical needed value to be able to form the protective alumina layer (Go¨tlind et al., 2007). However, this issue of impoverishing the aluminum on the surface of the alloy is of no consequence for the nuclear application intended and described here. Aluminum is only needed in the alloy in the case that there is a loss of coolant accident. Otherwise, aluminum never participates on the composition of the oxide under normal operation conditions. If the fuel has its three two-year cycles in the reactor without experiencing an accident, aluminum would not participate in the inherent overall environmental resistance of FeCrAl. Quadakkers et al. studied the oxidation behavior of MA956 powder metallurgy alloy (Fe-20Cr-4.5Al-0.5Y2O3) and a Fe-20Cr-5Al alloy for several exposure times at 900 C, 1000 C, and 1100 C using air enriched with 18O isotopes (Quadakkers et al., 1991). They reported similar oxidation kinetics for the two alloys and little or no effect of testing temperature in the range 900 C
1100 C. There is abundant information in the literature about the effects of water vapor on the oxidation kinetics of FeCrAl alloys; however, some of the literature may be confusing or contradictory because the effect

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

121

of water vapor on the oxidation of FeCrAl seems to be temperature dependent. The most comprehensive results at temperatures above 1000 C suggest that the effect of steam may modify the external layer of the oxide, where generally traces of iron and chromium are present. That is, water molecules react with chromium and iron oxides and evaporate them, but the inner protective layer of columnar alpha alumina remains unaltered in air, in air plus water vapor, and in pure steam conditions (Rebak et al., 2018; Al-Badairy and Tatlock, 2000). That is, in presence of steam the oxide on the surface of FeCrAl alloys is a clean columnar alpha alumina, while in the presence of air, the surface oxide may have two layers—an internal columnar alumina layer and an external equiaxed grains oxide with embedded traces of chromium and iron. It is reported elsewhere that although water vapor has no substantial effect on the isothermal oxidation of FeCrAl alloys at 1200 C, it may have significant effects in cyclic oxidation (Onal et al., 2003). But this cyclic oxidation issue is of nonconsequence for the intended current nuclear application because the fueled cladding would not experience this type of cyclic temperature service. The use of ferritic alloys FeCrAl as monolithic cladding for LWRs urania fuel started as a down selection of alloys which could resist attack by steam at temperatures higher than 1200 C (Rebak, 2015; Terrani, 2018). Terrani (2018) published side by side parabolic oxidation rate constants for several alloys and silicon carbide in 100% steam as a function of the inverse of the absolute temperature. At 1200 C, the resistance to oxidation of APMT was three orders of magnitude lower than for zirconium alloys. Only silicon carbide had a lower oxidation rate than did APMT. There is not another alloy than APMT to have the extraordinary resistance to attack by air and steam at temperatures higher than 1200 C. Fig. 5.29 shows the appearance of typical cladding tubing of Zircaloy-2 and APMT before and after exposure to 100% steam for 2 h. The mass gain for the Zircaloy-2 tubing was 60.6 mg/cm2 and the mass gain for the APMT tubing was 0.176 mg/cm2 showing the unparalleled resistance of APMT to attach by steam. The Zircaloy-2 tube metal was completely consumed in the steam, while the APMT tube barely showed a loss of shine.

Mechanism of protection at accident condition temperatures Beyond 1100 C, the alloys must contain approximately 4%
6% Al to offer protection. The way the FeCrAl alloys work is by the initial formation of a Cr2O3 oxide on the surface. As the temperature increases, a continuous thin alumina layer (Al2O3) develops underneath the Cr2O3 film because the partial pressure of oxygen needed to form alumina is lower than the equilibrium partial pressure with chromia. Eventually the external Cr2O3 oxide evaporates and the underlying alumina layer protects the material up to its melting point of approximately 1500 C (Fig. 5.30). The evaporation of chromia from

122

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.29 Steam oxidation of Zircaloy-2 and APMT tubing.

FIGURE 5.30 Mechanism of iron
chromium
aluminum (FeCrAl) oxidation at high temperature in air or steam.

the surface of high-temperature alloys, especially in presence of water molecules in the gas, has been recognized in the last few decades (Holcomb, 2008, 2009). Water vapor increases the evaporation loss by allowing the formation of CrO2(OH)2(g), which for these conditions has a higher vapor

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

123

pressure than CrO3(g). CrO3(g) is the predominate Cr gas species in dry air or oxygen (Holcomb, 2008). /2Cr2 O3 ðsÞ 1 H2 OðgÞ 1 3/4O2 ðgÞ 5 CrO2 ðOHÞ2 ðgÞ

1

The combined synergistic action between Cr and Al in FeCrAl at very high temperatures have been known for almost a century. The initial formation of chromia on the surface protects the alloy while alumina is developing underneath. Eventually alumina will be responsible for protection of the alloy after the chromia reacts with water and oxygen and evaporates as volatile hydroxides. Fig. 5.31 shows that when a coupon of FeCrAl APMT was exposed to steam at 1200 C for 4 h, it developed an approximately 1-μm-thick oxide, which was a pure columnar gamma alumina or Al2O3. The surface film did not contain Cr, Fe, or Mo. FeCrAl alloys such as APMT generally contain small amounts (less than 1%) of rare earth elements that may help peg the protective oxide to the alloy substrate. The oxide in Fig. 5.31 shows a dark spot, which is rich in the minor alloying elements Zr, Hf, Ti, and Ta. The FeCrAl producers such as Sandvik-Kanthal for APMT may sell the alloy in the preoxidized condition. Fig. 5.32 shows that an APMT production mill preoxidized strip in air at 1050 C for 8 h also contained an approximately 0.5- to 1.0-μm-thick alumina layer on the surface. The alumina that formed at

FIGURE 5.31 Oxide formed on APMT tube at 1200 C.

124

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.32 Oxide formed on APMT at 1050 C in air for 8 h.

1050 C in air consisted of the typical two layers: the external layer was equiaxed alumina containing residual particles of Ta, Zr and Hf; an internal columnar alumina that was mostly free of precipitates. Because the oxide was formed in air, it still contained residual amounts of Cr2O3 on the surface probably because the air did not hold enough moisture to completely evaporate the chromia from the surface of the coupon while exposed to 1050 C.

The Roles of metal oxides on the surface of FeCrAl Normal operation oxidation to accident oxidation scenario and vice versa This section considers two scenarios about the oxidation and oxides behavior on FeCrAl and the role that the oxides play on the surface of the material either in condensed water or in steam, or both. Scenario 1: After a FeCrAl cladding is in high-temperature water for several months, with a protective Cr oxide developed on its surface. Would this FeCrAl cladding be able to grow a protective alumina layer on the surface if exposed to superheated steam at T . 1000 C in the case of an accident? Scenario 2: After a high-temperature pre-oxidation treatment is applied to the FeCrAl tube cladding to grow a compact alumina layer on the surface. What happens to the alumina layer in the OD of the tube when it enters in contact with B300 C water? Does it dissolve? Would a Cr oxide be able to

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

125

develop on the surface after the alumina dissolves? Similarly, if there is a LOCA event and the cladding of FeCrAl develops a layer of alumina on the surface, what happens when the accident is quenched with fresh water? Does the alumina dissolve? Does chromia form to protect the cladding surface again?

Scenario 1: Water-oxidized APMT tubes exposed to superheated steam Fig. 5.33 shows the oxidation resistance for 4 h in superheated steam at 1200 C of APMT tube specimens, both in the as-received and pre-exposed to high-temperature water for 73 days. The figure shows that the specimens that were immersed in high-temperature water for 73 days had the same resistance to superheated steam than fresh as received nonwater preoxidized specimens. That is, fuel rods clad with FeCrAl alloys in an operating plant will resist attack by superheated steam in the unlikely event of a loss of coolant accident. Both, for as-received tube specimens and for specimens preexposed to high-temperature water, the oxidation rate in superheated steam was in the order of 0.25
0.35 mg/cm2. Fig. 5.34 shows the alumina layer that forms on the surface of an APMT tube which was first exposed to 330 C PWR-type hydrogenated (3.75 ppm) water for 73 days and then tested in 100% Steam for 4 h at 1200 C. The naturally formed chromium oxide layer in the PWR water environment mostly evaporated and a protective alumina film is formed on the surface of the APMT tube. The alumina film was approximately 1-μm thick and contained two layers: an equiaxed external thinner layer and a thicker internal layer of columnar alumina. The steam formed oxide does not contain Cr, Fe, or Mo.

FIGURE 5.33 Oxidation in steam of water preoxidized APMT.

126

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.34 Oxidation in steam of pressurized water reactor (PWR) preoxidized APMT tube.

Scenario 2: Steam-oxidized APMT tubes exposed to hightemperature water Fig. 5.35 shows the mass change (loss) as a function of immersion time for APMT tube specimens exposed to both BWR (288 C) and PWR (330 C) hydrogenated water. Each point is the average value for two or more specimens. The mass loss for the as-received tube specimens was the same as for the as-received flat specimens. The tube specimens that were preoxidized (PO) in steam at 1200 C for 2 h had initially a higher mass change (loss) than the tube specimens that were as-received (AR) or nonpreoxidized. However, once the alumina layer, which previously formed in the steam preoxidation test, dissolved in the hydrogenated water, the mass loss of the PO specimens practically stopped, and the mass loss rate became the same as for the AR specimens. The surface area of each PO specimen tube in Fig. 5.35 was approximately 3.7 cm2. When the specimens were preoxidized in steam at 1200 C for 2 h they developed a 1-μm-thick alumina layer on the surface. The volume of the alumina layer is then 3.7 cm2 3 0.0001 cm 5 0.00037 cm3 (volume of aluminum oxide). The mass of aluminum oxide is m 5 V 3 density alumina 5 0.00037 3 3.95 g/cm3 5 1.46 mg. Dividing the mass of 1.46 mg by the surface area of 3.7 cm2 5 0.4 mg/cm2, which is

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

127

FIGURE 5.35 Mass loss in hydrogenated water of steam preoxidized APMT.

exactly what Fig. 5.35 plot shows as mass loss of the PO tube specimens due to the dissolution of the alumina layer in the water. This is a clear proof that the initial higher mass change is just the dissolution of the alumina and then the dissolution stops because a Cr2O3 would develop on the surface. Rebak (2018) has shown that the oxide formed on the surface of a steam preoxidized APMT tube specimen after immersion in PWR hydrogenated water at 330 C for 284 days consisted only of chromium oxide, similarly to the asreceived APMT specimens in hydrogenated water at 330 C. That is, the alumina layer from the surface of the preoxidized tube specimen dissolved in water and that a chromium-rich oxide developed to protect the APMT alloy from further corrosion in the B300 C water.

The versatile oxidation behavior of FeCrAl Alloys When freshly fabricated specimens of FeCrAl alloys are exposed to B300 C water they develop a protective Cr2O3 oxide on the surface. This is the same mechanism by which well-known 300 series austenitic stainless steels protect themselves against corrosion (Rebak et al., 2017a,b; Stachowski et al., 2017; Gupta et al., 2018). In hydrogenated high-temperature water, the oxide is a single layer of chromium oxide (chromia), and in excess oxygen conditions the oxide is a double layer, an external oxide containing iron and chromium, and a thinner internal layer that is only chromium oxide. In B300 C water the oxide does not contain either molybdenum or aluminum. It is likely that as the oxide film develops on the surface of these stainless materials, the molybdenum and aluminum in the alloy dissolve into the water until a protective chromium oxide grows and the dissolution rate of the alloy slows down or stops. If freshly fabricated tube specimens are exposed to B1200 C

128

Accident-Tolerant Materials for Light Water Reactor Fuels

steam, they form a protective alpha alumina film on the surface. The oxide that develops in steam does not contain Mo, Fe, or Cr. It has been demonstrated through basic research that as the material heats up, first a chromium oxide grows on the surface, which allows for an aluminum oxide to develop under the chromium oxide. Eventually the iron, chromium, and molybdenum oxides evaporate in the steam environment and the alloy is covered by a thin film of alumina, which is stable until the melting point (B1500 C) of the FeCrAl alloys (Rebak, 2018). Fig. 5.36 shows schematically the oxide development in the scenarios described (normal operation or accident). If a specimen with a chromium oxide formed in water under normal operation conditions is subsequently exposed to steam, the chromium oxide evaporates, and the alloy develops an alumina layer on the surface. If a tube of FeCrAl, which suffered an accident and therefore has an alumina layer on the surface, is flooded with fresh water (quenched), the alumina oxide dissolves in the water and the alloy develops or regrows a chromium oxide on the surface. That is, FeCrAl APMT is a smart alloy because it develops the oxide that it needs for its protection in the conditions to which it is exposed (i.e., B300 C water vs B1200 C steam). Stainless steels (e.g., austenitic type 304 SS) contain at least 13% of chromium so they can maintain passivity in most industrial applications. The

FIGURE 5.36 Versatile oxides on iron
chromium
aluminum (FeCrAl).

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

129

long-lasting passivity is a slowing down of the corrosion rate of the steel by the formation of a thin protective and adherent layer of chromium oxide on the surface. The passivity provided by chromium oxide of the stainless steels and nickel-based alloys also applies to ferritic FeCrAl alloys in LWR environments at temperatures near 300 C (Rebak et al., 2017a,b). The nuclear power reactors have been successfully using austenitic stainless steels and nickel alloys with chromium in hot water for over sixty years. Chromium is the element that provides the resistance in hot water by the formation of a protective oxide film on the surface under reactor normal operation condition environments. As long as the alloy contains 12% in mass of chromium, it would passivate in LWRs. Currently used alloys that fulfill this chromium content requirement or passivity are, for example, type 304/316 austenitic stainless steels, and nickel-based alloys such as 600, 690, X-750, 718, and 800. Since 2012 the use of ferritic FeCrAl alloys as an alternative for fuel cladding has been recommended to provide resistance to attack by steam in the case of a severe accident situation. Ferritic FeCrAl have never been used in LWRs before. The FeCrAl also contain chromium and it was expected and demonstrated here and elsewhere that they will also develop a thin chromium oxide on the surface to protect against high-temperature water attack (Rebak et al., 2017a,b). FeCrAl alloys also have approximately 4%
6% of aluminum, which will only “act” in the case of a severe accident. If a plant never has an accident, the aluminum in the FeCrAl will just sit and wait. Aluminum does not participate in the passive film formed in regular B300 C normal operation conditions water. Aluminum may never be needed because for the normal operation conditions only the chromium is needed. However, it has also been shown here that if an unlikely loss of coolant accident (LOCA) happens, the aluminum will be called into action to develop the protective alumina layer required to resist the attack by steam. Chromium can protect the cladding tubes against steam approximately until near 1100 C. At higher temperatures, aluminum offers the protection to the tubes until their melting point. The synergistic effect between Cr and Al in FeCrAl makes this material ideal for both situations, normal operation conditions and prepared in the case of an accident condition.

Fabrication and implementation of cladding tubes In reports prepared by General Electric for the Atomic Energy Commission (General Electric, 1965, 1966) the FeCrAl alloy fabrication procedures were described. The alloys were vacuum induction melted in alumina crucibles and then cast into copper or graphite molds. Sound ingots of up to 200 kg were produced. The ingots were then extruded at 1000 C following 14 to 1 and 20 to 1 reduction ratios. Powder metallurgy alloys of composition Fe13Cr-6Al-1.5Y were also produced by induction melted tin argon material and then atomized in argon to produce less than 100-mesh size particles. The

130

Accident-Tolerant Materials for Light Water Reactor Fuels

powder was canned in mild steel and then extruded at 1000 C to later being processed to sheet and rod (General Electric, 1966). The powder metallurgy product had improved strength and notch-impact properties than did the traditionally produced melted and cast material. The mid-1960s tubes were produced by warm drawing at 300 C of tube blanks produced from drilled rods or by hot extruding a billet over a mandrel (General Electric, 1965, 1966). Oak Ridge National Laboratory produced a traditionally melted, cast, and thermo-mechanically processed alloy C26M (ORNL, 2018a,b), which is similar in composition to some of the alloys cast by GE in the 1960s (General Electric, 1966
68). General Electric adopted the ORNL C26M chemistry and grouped the C26M with Sandvik APMT and NFD FeCrAl ODS (Table 5.1) and named this FeCrAl family IronClad. Fig. 5.37 shows the microstructure of 0.4-mm-thick wall tubes of APMT and C26M. The APMT tube was fabricated following a pilgering process of a powder metallurgy tube hollow. The C26M tube was fabricated by a warm drawing process from a tube hollow produced from a cast ingot. That is, the

FIGURE 5.37 Tubes’ microstructures.

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

131

most basic production of the source hollow material was different, powder metallurgy for APMT and traditional melting, casting, and forging for C26M. Both tube fabrication methods were able to yield the 5-m-long cladding tube in the required geometry of 10.26 mm outside diameter (OD) and 0.4-mm wall thickness. However, Fig. 5.37 shows that the resulting microstructures in the transverse direction were different depending on the fabrication route. APMT had a much finer grain size of 10
20 μm, with many dispersed oxides (showing as white nanosized dots in the images) and with some occasional micro size porosity. The grain size for the C26M was slightly larger in the order of 50 μm, and the material was mostly free from oxides and porosity. Two alloy fabrication methods are generally used for FeCrAl. For example, APMT and FeCrAl ODS were produced by powder metallurgy, and alloy C26M was produced by traditional melting, ingot pouring, and then forging. Both types of alloys were ended in bar stock via extrusion and finally the bars were gun drilled to make hollows for tube manufacturing. The APMT tubes were manufactured by Sandvik Materials Technology in Sweden using a pilgering method, the FeCrAl ODS were fabricated by Nippon Fuel Development in Japan using pilger cold rolling, and the C26M tubes were manufactured by Century Tubes in San Diego, CA following a warm drawing process.

Welding of FeCrAl alloys Fig. 5.38 shows the results of two initial trials of tungsten inert gas (TIG) welding of thin-walled C26M and APMT tubes to C26M and APMT caps, respectively (Rebak, 2017). It is apparent from Fig. 5.38 that FeCrAl

FIGURE 5.38 Thin-walled C26M and APMT tubing industrially welded to caps using TIG.

132

Accident-Tolerant Materials for Light Water Reactor Fuels

IronClad tubes can be welded without difficulty using standard plant production techniques to make hermetic seals. In the weld seams, where localized melting occurred, there was significant grain growth mostly on the tube side. However, the final welded products were without distortion, cracking, or porosity. IronClad alloys can likewise be readily welded using pressure resistance welding (PRW), also in a fueled rod industrial production setting. Pressure resistance welding is a preferred method because it does not involve melting which could be disruptive for the powder metallurgy ODS-type materials. Fig. 5.39 shows schematically from left to right the sequential steps for the fabrication of fueled rods. To make one rod, a tube and two caps of FeCrAl are needed plus enough urania pellets to fill the tube. Then two critical orbital welds are performed to hermetically seal the fuel rod elements. First an initial or bottom end cap is welded to the tube and, after introducing the fuel pellets into the tube, there is a final or closure weld performed to the second or top end cap. The welds need to be hermetic to withstand pressure and to contain the gaseous fission products inside the pressurized rod. In a BWR nuclear power plant, the rods need to stand an internal pressure of 6.3 MPa, hoop stresses of 110 MPa, and axial stresses of 55 MPa at the location of the welded caps (Kimura et al., 2017). It is likely that the welds in the industrial production will be a solid-state pressure-resistance welding. Several welding or joining methods of FeCrAl has been investigated at GE Research, Idaho National Laboratory, and Oak Ridge National

FIGURE 5.39 Fuel rod fabrication steps.

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

133

Laboratory. ORNL (2018a,b) reported that several welding methods have been used for FeCrAl, including laser beam welding, pressure resistance welding, and gas tungsten arc welding (GTAW or TIG). Not major issues with welding processes of FeCrAl have been identified and reported. Rebak (2017) described that thin-walled tubes of APMT were welded to APMT end caps using the same industrial TIG production setting for fuel rod manufacturing currently used for zirconium alloy rods. The resulting welds were free from cracks, distortion, or porosity (Rebak, 2017). Idaho National Laboratory (INL, 2015) prepared several reports on the laser welding and pressure resistance welding (PRW) of thin-walled tubes made using two commercial powder metallurgy ODS FeCrAl alloys (MA956 and Kanthal-D) and the ORNL traditionally melted and mold cast FeCrAl alloy C35M (INL, 2015). The laser beam welding (LBW) involves melting at the weld seam, while the pressure resistance welding is a solid-state process that does not involve melting during the joining of the cap to the tube. INL reported excellent bonding between the tubing and the end plugs or caps using both EBSD mapping as well as tensile tests of the weld seam (Gan et al., 2017, 2018). There was some grain growth at the weld seam using the LBW method, but in the PRW resulted joint, the grain size practically remained unchanged and it was difficult to determine the bonding line (Gan et al., 2018). The PRW of tubes to end caps had an excellent resistance to bursting tests (Gan et al., 2018) because the welded tubes pressurized at an ambient temperature and at 180 C burst at pressure values that were three times higher than the values of the design pressure of 2500 psi. For both the PRW and LBW joints, the overpressure burst always happened in the tube section, never in the weld seam, showing the excellent bonding strength by both joining methods (Gan et al., 2018). The mechanical behavior of nonirradiated FeCrAl welds was also studied at Oak Ridge National Laboratory (Field et al., 2014; Gussev et al., 2017). Field et al. (2014) prepared sub-sized tensile specimens of three traditionally melted FeCrAl model alloys containing a laser-welded seam in the middle of the gage. They reported grain growth in the fusion zone and lower tensile strength but larger elongations to failure when compared with the nonwelded specimens. The failure of the welded specimens was always in the fusion zone probably because of the softer larger grain microstructure (Field et al., 2014). Six model FeCrAl alloys made by the traditional melting and thermomechanical process were ‘welded’ using autogenous bead on plate pulsed laser welding method (not an actual joining) (Gussev et al., 2017). They reported the weld seams free from cracks, porosity, or other defects. The FeCrAl alloy development in Japan studied the weldability of their ODS material (Kimura et al., 2017, 2018). The initial trials involved welding 430 steel caps to FeCrAl ODS tubing using the following three methods: (1) PRW, (2) electron beam (EB) welding, and (3) TIG (Kimura et al., 2017). They characterized the soundness of the weldments using X-ray computed

134

Accident-Tolerant Materials for Light Water Reactor Fuels

tomography (CT) and tensile tests at an ambient temperature. For the TIG welding they reported grain growth at the weld seam and a softening, which caused the tubes to fail at stresses lower than the nonwelded tube. For the EB welding, the tensile tests showed that the specimens always failed in the nonwelded portion of the tube; that is, the weld seams were stronger than the rest of the tube (Kimura et al., 2017). Overall it has been shown that several methods may be used to weld the FeCrAl caps to the FeCrAl tubes to obtain leak proof rods free from joining defects.

Mitigation measures to parasitic neutron absorption of FeCrAl By its own nature, FeCrAl alloys offer a larger parasitic neutron absorption compared with zirconium alloys (Zinkle et al., 2014; Rebak, 2015; George et al., 2015) (Fig. 5.4). But because FeCrAl alloys such as APMT, FeCrAl ODS, and C26M are mechanically stronger than zirconium alloys at near 400 C, the FeCrAl material for the cladding can be made approximately half the thickness of the current zirconium alloys for the cladding of the fuel (Fig. 5.40). The thinning of the wall to 0.4 mm mitigates the higher neutron capture of IronClad (2.47 barns) compared with zirconium alloys (0.2 barns). At the same time, the thinning of the wall increases the volume of the urania pellet inside the rod, which partially compensates the thermal neutron availability in the system. Additional design changes (such as in the fuel channel) may be required to meet bundle design requirements, further impacting fuel cycle economics. However, potential mitigation strategies have been identified that may partially or fully offset these neutron penalties, which would increase the fuel cycle cost. Such mitigation strategies include alternate materials (e.g., silicon carbide composite channel materials), higher allowable heat generation rates, as well as relaxation of regulatory requirements due to much improved fuel cladding performance under normal/off-normal, design basis and beyond design basis accident conditions, which in turn will result in improved economics of plant operation. Other alternatives to the higher neutron penalty of FeCrAl would be the use of a higher density fuel such as U3Si2 (Chen and Yuan, 2017).

Mitigation measures to increased tritium release into the coolant A second issue that requires attention is the potential to increased release of tritium into the coolant. EPRI reported that when austenitic stainless steel cladding were used for power generation, the amount of tritium in the coolant water was approximately 10 times higher than when zirconium cladding were used (EPRI, 1982). Since FeCrAl are ferritic (bcc) in nature, the

FIGURE 5.40 Mechanical and neutron absorption properties of IronClad and Zircaloy-2.

136

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.41 Oxide barriers to tritium release into the coolant.

diffusion of tritium through the cladding wall into the coolant could be even higher than when the austenitic (fcc) iron-based material was used. It is known that hydrogen intake into iron alloys decreases in the presence of oxide films on the surface of the alloys (Levchuck et al., 2008; Strehlow and Savage, 1974). Therefore one potential mitigation strategy, currently under investigation, is by the formation of oxide barriers on the cladding wall both from the ID and OD of the cladding (Fig. 5.41). Under normal operation conditions, the cladding may develop an alumina layer in the ID, which will greatly reduce the atomic hydrogen (tritium) entrance and diffusion through the tube wall and into the coolant (Rebak and Kim, 2016; Levchuk et al., 2008). The partial pressure of oxygen in the fuel cavity due to the thermodynamic dissociation of urania may be enough to allow for a layer of alumina to form on the ID of the FeCrAl cladding wall. Similarly, after exposure of the OD of the cladding to the coolant water at close to 300 C, the OD tube wall will develop in a few weeks a protective compact chromia layer, which is also a barrier for hydrogen diffusion (Strehlow and Savage, 1974; Van Deventer and Maroni, 1983), thereby minimizing the presence of tritium in the coolant (Fig. 5.41). Tritium permeation studies were conducted on FeCrAl ODS alloys (Sakamoto et al., 2017, 2018). They studied the permeation of hydrogen as a function of the temperature on noncorroded FeCrAl specimen and on an FeCrAl ODS specimen immersed for 30 days in water with 8 ppm of dissolved oxygen at 290 C. They reported a modest decrease by a factor of less than 10 in the tritium permeation through the preoxidized specimen when compared with the nonoxidized FeCrAl ODS specimen (Sakamoto et al., 2018). One of the common failure mechanisms of the current cladding of zirconium alloys is the embrittlement by hydrides. Atomic hydrogen would form on the surface of the zirconium cladding and diffuse into the cladding wall

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

137

and react with the metal to form stable zirconium hydrides. The hydrides tend to develop near the OD of the zirconium alloy fuel rod owing to the temperature gradient across the cladding wall. Hydrogen is less soluble in the zirconium alloy matrix near the cooler OD and reacts with zirconium to form hydrides. By using FeCrAl alloys for the cladding, the hydride in the cladding issue is eliminated because none of the elements in FeCrAl (Fe, Cr, Al, or Mo) react with hydrogen (or tritium) to precipitate stable metal hydrides in the manner that Zr does.

Irradiation behavior of FeCrAl Because ferritic FeCrAl alloys have never been used before in nuclear reactors, information regarding their behavior under irradiation is limited. Only recently exposure studies on FeCrAl have been conducted in experimental reactors. In 2020 it is expected to have FeCrAl neutron irradiation results from the commercial LWR Hatch Unit 1 in Georgia, USA (Rebak, 2017). In general, it is anticipated that FeCrAl would have low radiation-induced swelling (Little and Stow, 1979; Ko¨gler et al., 2012) (Fig. 5.4). Irradiation to up to 30 dpa of commercial ferritic steels including FeCrAl in the temperature range 380 C
615 C showed swellings below the experimental detection limit of 0.1% (Little and Stow, 1979). Ko¨gler et al. (2012) studied the response of powder metallurgy ODS alloy PM2000 to dual beam ion implantation at ambient temperature and at 300 C for a total dose of 52 dpa. They reported that the specimens irradiated at 300 C showed even a reduction of defects around yttrium oxide particles when compared with the as-received materials and very little hardness increase upon irradiation (Ko¨gler et al., 2012). There was some concern that neutron irradiation would promote in FeCrAl ferritic alloys a similar phenomenon like thermal aging in the vicinity of 475 C, when there is a phase separation between an iron-rich phase (α) and a chromium-rich phase (α’). Chao et al. (2014) conducted thermal aging at 475 C of PM2000 FeCrAl powder metallurgy ODS for 20 and 200 h and they measured the mechanical properties at ambient temperature after aging at 475 C. They reported that the yield stress (YS) increased as the aging time increased from approximately 1000 MPa for a nonaged material to 1100 MPa for 20 h of aging to 1300 MPa for 200 h of aging; however, the total elongation to failure did not significantly decrease because it was 15% for nonaged, 14.5% for 20 h of aging and 12% for 200 h of aging (Chao et al., 2014). It was also reported that the largest increase in the yield stress of PM2000 at ambient temperature upon aging at 475 C happened for the first 100 h of thermal aging and little additional effect was observed for aging times of 320 h and 520 h (Chao et al., 2014). Meanwhile the elongation to failure was always higher than 10% for all the tested conditions. Ejenstam et al. (2015) produced four laboratory induction melted FeCrAl

138

Accident-Tolerant Materials for Light Water Reactor Fuels

alloys with compositions Fe 1 10Cr 1 4Al, Fe 1 10Cr 1 6Al, Fe 1 10Cr 1 8Al, and Fe 1 21Cr 1 5Al and studied their thermal stability after aging at 450 C, 500 C, and 550 C for 10 h, 100 h, 1000 h, and 10,000 h. The alloys were studied in the form of 4-mm diameter wires or as 8 3 1 mm strips hot rolled from the wires (Ejenstam et al., 2015). They reported a hardness increase for the 21Cr alloy aged at 450 C and 500 C but no change in hardness for the 21Cr material aged at 550 C. The increase in hardness for the 21Cr alloy leveled off after 1000 h of aging. None of the three 10Cr alloys suffered any increase of hardness for any of the three aging temperatures. Neutron irradiation tests of specimens of four noncommercial FeCrAl alloys containing from 3%Al 1 17%Cr to 5% Al 110%Cr were performed at 320
382 C to a nominal dose level of 1.8 dpa (Field et al., 2015). Field et al. (2015) found that even though the yield stress slightly increased upon irradiation owing to the formation of dislocation loops and alpha prime, the total elongation to failure was always higher than 10% and it remained more or less constant or actually increased due to the annealing or recovery of the cold work in the nonirradiated material. Field et al. (2017) also studied the irradiation response of the commercial alloy Alkrothal 720 (Fe 1 13Cr 1 4.2Al). The neutron irradiation was carried at 382 C to a total dose of 1.8 dpa. They reported an increased formation of dislocation loops and other defects but did not correlate the findings to the mechanical behavior of the irradiated material. They noted that the initial metallurgical state of the alloy is important to understand how it behaves under irradiation (Field et al., 2017). Aydogan et al. (2018) performed heavy ion irradiation for a dose of up to 16 dpa at 300 C of two laboratory-made FeCrAl, namely C06M2 (Fe 1 10Cr 1 6Al 1 2Mo 1 0.2Si 1 0.05Y) and C36M3 (Fe 1 13Cr 1 6Al 1 2Mo 1 0.2Si 1 0.05Y) alloys in the form of 0.38-mm wall thickness tube strips. These alloys were initially vacuum induction melted, homogenized at 1200 C, hot extruded at 800 C, gun drilled, and made into thin-walled tubing by drawing. They reported a gradual increase in the surface hardness (to a depth of 200 nm) as the dose of irradiation increased; however, the hardness leveled off between 3.4 dpa and 16 dpa (Aydogan et al., 2018). The authors did not discuss the implications of their findings on the use of FeCrAl for fuel cladding applications. One of the failure mechanisms of austenitic materials in current LWRs is linked to irradiation damage, more specifically to radiation-induced segregation (RIS) (Hojn´a, 2013). RIS has also been studied in the classical ferritic Fe-Cr alloys such as T91 (Fe 1 8.37Cr 1 0.9Mo) and HT9 (Fe 1 11.6Cr 1 1Mo) using proton irradiation at 400 C to up to 3 dpa (for HT9) and 10 dpa (for T91) (Wharry and Was, 2013). It was reported that Cr modestly (,2%) enriched upon irradiation at prior austenite grain boundaries, and the enrichment was higher for T91 than for HT9. General Electric has an accelerated plan to obtain commercial reactor irradiation data on FeCrAl (Lin et al., 2018). In February 2018 GE teamed

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

139

up with Southern Nuclear to upload in Cycle 29 of the Edwin Hatch Unit 1 Power Plant in Georgia (USA) lead test assemblies of FeCrAl components containing nonfueled segmented tubes of C26M and connecting components of APMT (Lin et al., 2018). GE calls their FeCrAl concept IronClad. The objective is to obtain irradiated IronClad material for postirradiation examination (PIE) studies to determine performance of the cladding under normal irradiation conditions of an operating commercial power reactor. The first harvesting for PIE will be conducted before Cycle 30 in the spring of 2020. The fact that the C26M tubes are nonfueled will facilitate the initial PIE studies. In October 2019 GE is planning to insert fueled IronClad tubes into the Clinton Unit 1 plant operated by Exelon (Hayes, 2019). The insertion at Clinton will provide integral data, including pellet-cladding interaction (PCI). In parallel, GE is working with Idaho National Laboratory in the testing of the IronClad concept in the ATR both in the programs called ATF-1 (dry capsules) and ATF-2 (wet PWR typical water chemistry), which are ongoing tests in mid-2019.

Corrosion behavior of used FeCrAl cladding in cooling pools The fuel cycle involves a long life for the fuel bundles, from the fabrication of the fueled rods, their residence in the reactor for probably three cycles of two years each, and then a cooling period of a minimum of five years in used fuel storage pools to remove most of the fission heat. After a residence of 5
20 years in the cooling pools the fuel may end up in dry cask storage for up to 100 years until reprocessing or final disposal in a geologic repository (Schuster et al., 2019). It was mentioned before that FeCrAl were not designed to be used in wet environments; therefore little is known on the corrosion performance of FeCrAl in aqueous environments. Since the used fuel rods will remain immersed in water for at least five years at a maximum temperature of 60 C (Rebak and Huang, 2015), it is important to assess their corrosion performance in aqueous electrolytes. Fig. 5.42 shows the anodic behavior of FeCrAl alloys such as C26M, APMT, and ODS FeCrAl as well as type 304 austenitic stainless steel in a seawater-like environment of 3.5% NaCl solution at 45 C. The chloride concentration in these tests is much higher than the actual chloride concentration in the cooling pools system. The objective of the tests in Fig. 5.42 is to determine the relative localized corrosion resistance of the ferritic materials when compared with that of the well-known type 304 stainless steel using the ASTM G 61 guidelines. The four IronClad specimens were prepared from the OD of actual cladding tubes, with representative microstructure and geometry. For the four tested materials the corrosion potential was similar and approximately 2700 mV SCE in deaerated 3.5% NaCl at 45 C. The anodic passive current density was the lowest for APMT and the higher was for

140

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 5.42 Cyclic potentiodynamic polarization.

C26M and ODS FeCrAl inversely proportional to the chromium content in the alloys. The higher chromium content in APMT (21%) may have established an initially more protective and compact passive film on the surface than the 12%Cr content in the C26M tube specimen. The highest breakdown or pitting potential corresponded to APMT (again because of the highest chromium content). However, all four tested materials showed a similar value of repassivation potential or the potential at which the reverse current crosses over the forward current. Current results show that ferritic FeCrAl alloys behave similarly like their austenitic stainless steels cousins. That is, the resistance to corrosion and localized corrosion is a function of the alloying elements and not of the bcc or fcc microstructure of the steels. The localized corrosion resistance of stainless alloys can generally be explained by the pitting resistance equivalent (PRE) number 5 %Cr 1 3.3%Mo 1 16%N. The higher the PRE number, the higher the resistance to localized corrosion. For APMT the PRE is 31, for C26M is 19, for FeCrAl NFD ODS alloy PRE is 12 (since it does not contain molybdenum), and for type 304SS PRE is 18. The PRE ranking shows that the alloy most resistant to localized corrosion should be APMT and the least resistant would be FeCrAl ODS with a PRE of 12.

Licensing for reactor use The use of monolithic FeCrAl cladding for urania in commercial power reactors has the attractiveness of its simplicity because it involves the replacement of one metal (zirconium) for another (FeCrAl). The coolable geometry or the surface area that the cooling water enters in contact with

FeCrAl—iron
chromium
aluminum monolithic alloys Chapter | 5

141

does not change inside the reactor core. The thermal hydraulic behavior would remain practically the same than for the current concepts (Rebak et al., 2015). General Electric and the US Department of Energy as well as the Nuclear Energy Institute and the EPRI are working together to gather data that may be used by the US Nuclear Regulatory Commission (NRC) to grant exemptions to current zirconium centric licensing guidelines. The use of IronClad cladding is being accepted as the second near-term accident-tolerant fuel after the coated zirconium alloy concepts. General Electric through its nuclear fuel company (Global Nuclear Fuels) has partnered with two utilities in the United States to insert articles of FeCrAl into two commercial power reactors to assess their behavior. One utility is Southern Nuclear, which has inserted FeCrAl components into the Hatch Unit One reactor in February 2018, and the second utility is Exelon, which is planning to insert FeCrAl concepts into the Clinton Unit plant in October 2019 in its Cycle 20 (Hayes, 2019). General Electric is working with the NRC as it obtains their data to prepare the licensing topical reports so that the internalization of information by the NRC scientists may happen concurrently as the data is obtained and examined in safety analyses.

Chapter 6

Silicon carbide and ceramics metal composite Chapter Outline Overview Why do we consider silicon carbide composites for accident tolerant fuel? Benefits and challenges Thermal properties and permeability of SiC/SiC fuel cladding SiC/SiC fuel cladding, fabrication, and implementation

143

144 148

Environmental behavior of SiC/SiC under normal operation conditions 149 Environmental behavior of SiC/SiC under accident conditions 153 Irradiation behavior 155 Licensing for reactor use 156

148

Overview The third and the last concept for ATF fuel cladding is the use of silicon carbide (SiC) composite materials. The implementation of SiC for cladding is further down the road than the other two most near-term concepts, namely coated zirconium alloys, and FeCrAl alloys (Chapter 4, Accident tolerant fuel cladding concept: coatings for zirconium alloys, and Chapter 5, FeCrAl—iron
chromium
aluminum monolithic alloys). The main reason SiC is considered for nuclear application is because of its high transparency to neutrons, its resistance to attack by air and steam at temperatures as high as 1700 C, and its high mechanical strength at the high temperatures. One of the handicaps for SiC is their high solubility in condensed water at approximately 300 C typical of light water reactor operation conditions. Another challenge is related to the fabrication of the rod to produce a reliable hermeticity to avoid the release of toxic compounds into the coolant. Other issues that may need to get resolved are the need for a consistent production route with controlled stoichiometry of the SiC compound and containing low impurities, which may harm not only their resistance to oxidation but also their resistance to neutron irradiation. Other unknown for nuclear applications is the SiC cladding response to pellet cladding interaction. Since hermeticity and fabrication issues need to be resolved before SiC is used for cladding application, it may be used earlier as channel material for BWR Accident-Tolerant Materials for Light Water Reactor Fuels. DOI: https://doi.org/10.1016/B978-0-12-817503-3.00006-7 © 2020 Elsevier Inc. All rights reserved.

143

144

Accident-Tolerant Materials for Light Water Reactor Fuels

reactors which does not need the hermeticity of the fuel rods. The more near term application for channels will provide familiarity on the use of ceramics in a nuclear environment which could be a facilitation path for later use in fuel rods. For channel boxes, the issue of SiC high dissolution rate in water may be solved by coating it with a thin layer of chromium like for the zirconium cladding.

Why do we consider silicon carbide composites for accident tolerant fuel? Benefits and challenges As mentioned in the previous chapters, there are three main areas of material studies for accident tolerant fuel (ATF) claddings, which in likely chronological order of implementation are listed as: 1. coated zirconium alloys, 2. monolithic FeCrAl alloys, and 3. silicon carbide (SiC) composite materials. Fig. 3.4 shows that the application of silicon carbide ceramics in light water reactors (LWRs) may be not earlier than in 10
15 years (year B2030). Even before the Fukushima accident of March 2011, silicon carbide was considered for nuclear applications mainly because of its high transparency to thermal neutrons, its tolerance to radiation damage, its mechanical strength, creep resistance plus its extraordinary resistance to oxidation in air and steam at elevated temperatures or under reactor severe accident conditions. Several countries are studying the feasibility of adopting silicon carbide composites as structural materials for LWRs ATF. The three main challenges for silicon carbide metal matrix composites are (1) how to make them compatible with B300 C water, (2) how to seal the end caps to the tubes, and (3) how to make the cladding hermetic or how to avoid the presence of microcracks to avoid fission products diffusion into the water. Silicon carbide composites are being studied for two main applications: (1) for channel material in boiling water reactor (BWR) and (2) for fuel cladding material in all water reactors. For channel material, SiC needs to be compatible with hot water but it does not need to be hermetically sealed, so channel material is a more near-term application for ceramics in LWRs than for fuel cladding (Fig. 3.4). Implementation of SiC for cladding not only would require a material compatible with water and being hermetic but the use of SiC could also impact the fuel rod and fuel assembly designs (NNL, 2018). The deviations needed to optimize the performance of SiC may be connected to the thickness of the cladding wall and the diameter of the fuel pellet. SiC is ideal for the fuel economy due to the high transparency to thermal neutrons but it may be costlier during the fabrication of the cladding as compared to the more amenable and simpler monolithic metallic claddings such as Zircaloy-2 and FeCrAl.

Silicon carbide and ceramics metal composite Chapter | 6

145

One of the main reasons SiC has always been considered for nuclear power applications is because of its low thermal neutron absorption cross section (low barns) and the low activation of elements during reactor irradiation (Fig. 6.1). Table 6.1 shows the neutron absorption or capture cross section (and other properties) of a list of elements and SiC. The lower the value of the barns, the easier is for thermal neutrons to migrate through the material. Table 6.1 shows that the neutron cross section for SiC (0.086 barns) is even lower than the cross section for the current cladding material based on the element zirconium (0.18
0.20 barns) and lower than for the FeCrAl alloy concept which would have a neutron cross section of 2.4 barns (George et al., 2015). SiC has also been identified as an excellent option as an accident tolerant material mainly because it can maintain good mechanical properties and exceptional resistance to attack by steam up to 1700 C (Singh et al., 2019). The use of SiC ceramic components for LWRs would be a revolutionary concept as opposed to the more evolutionary concepts of metallic coated

FIGURE 6.1 Benefits and detriments of SiC. ATF, Accident tolerant fuel; CMC, ceramic matrix composite; SiC, silicon carbide.

146

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 6.1 Thermal neutron absorption cross section (barns). Element

Neutron cross section (barns)

Melting point ( C)

Density (g/cm3)

Al

0.22

660

2.7

B

760

2075

2.46

C

0.0035

3550

2.26

Cr

2.9

1850

7.14

Fe

2.56

1538

7.87

Mo

2.48

2617

10.28

N

1.91

2 210




Nb

1.1

2415

8.57

Ni

4.5

1455

8.91

O

0.00028

2 218




Si

0.166

1414

2.33

SiC

0.086

2730

2.58

Sn

0.63

232

7.31

Ti

5.6

1670

4.51

Y

1.28

1526

4.47

Zr

0.184

1855

6.51

zirconium alloys or FeCrAl cladding discussed in previous chapters. SiC for LWRs appears as the most impactful technology for LWR, probably with a two-step application, first for channel boxes in BWR and guide tubes for pressurized water reactor (PWR) and in a second step, as the hermetic cladding for the fuel (Xu et al., 2018; Terrani, 2018). The greatest challenge in using SiC/SiC for cladding is how to seal the rod after the fuel has been loaded (Yueh and Terrani, 2014). The use of SiC materials for nuclear application has been studied since the 1960s for high-temperature gas-cooled reactor fuels, and is currently being investigated, in its composite form for structural application in light water fission power reactors (Deck et al., 2015). In other industries, such as aviation, the use of SiC material is generally referred as a ceramic matrix composite (Zok, 2016). For most industries, the attractiveness of SiC is due to its resistance to degradation at temperatures in the order of 1500 C and higher, where metals and alloys cannot perform without coatings and/or cooling. Another attractive property of SiC materials is the high melting

Silicon carbide and ceramics metal composite Chapter | 6

147

point and their high-temperature mechanical properties. On the down side for nuclear applications, current technology SiC may not be able to maintain a complete hermeticity as cladding material, that is needed to avoid the release of fission products from the fuel into the coolant. Another issue with the use of SiC for LWR is their instability in hot (B300 C) water. It was suggested that for fuel cladding, the SiC/SiC may need three layers: an internal monolithic layer to avoid diffusion of the fission products, a middle one reinforced with fibers to provide mechanical strength, and an outer one to provide corrosion resistance in water (Park et al., 2013). Monolithic SiC is a brittle ceramic, but if it is engineered with reinforcing SiC fibers, it can be reliably used for structural applications. Therefore, for engineering purposes, SiC materials are generally referred as a SiC/SiC composite since the composite is proposed as SiC fibers embedded in a SiC matrix which may include a chemically vapor infiltrated (CVI) medium (Katoh et al., 2014a). While SiC/SiC engineered systems do not fail by brittle fracture, it may undergo microcracking under stress. The microcracking may occur at only 0.1% strain levels and lead to the loss of hermeticity (Deck et al., 2015). The presence of porosity in the SiC/SiC fiber 1 matrix composite is in the order of 10%
20%, and this porosity can negatively affect the thermal conductivity and the strength of the material, specially sliding and debonding between the fibers and the matrix (Katoh et al., 2014a). The SiC for cladding application may need low porosity (,5%) and be optimized to have a strong SiC/SiC structure combined with a monolithic SiC layer to serve as the impermeable barrier for fission products (Deck et al., 2015). The optimized SiC structure for LWR application should also have improved corrosion resistance in the coolant. Even before the accident at the Fukushima Daiichi nuclear power stations, the application of SiC composite was considered and tested for PWR LWR cladding (Carpenter and Kazimi, 2010). Just a few months after the Fukushima Daiichi accident of March 2011, SiC composite materials were considered for BWR channel applications in replacement of zirconium alloys (Yueh et al., 2012). Besides the Fukushima accident, zirconium channel boxes already needed a redesign due to reactor performance issues such as the bowing partially caused by neutron flux and partly by shadow corrosion (Yueh et al., 2012). The consideration of SiC for BWR channel boxes is not limited by the restriction of hermetic seal requirements in the cladding application for the fuel (Yueh and Terrani, 2014; Singh et al., 2019). Therefore, the use of SiC/SiC for channel boxes in BWR seems like the first logical step since it will act like an in situ test in an actual reactor environment and provide real-time data regarding corrosion and irradiation resistance. The channel boxes are a rather simple structure to provide physical separation between the fuel bundles and the control rods. The only concerning damage that the channels may suffer is swelling and distortion due

148

Accident-Tolerant Materials for Light Water Reactor Fuels

to irradiation which may interfere with the operation of the control rods (Singh et al., 2019). Ray et al. (2012) discussed the use of SiC as cladding materials in LWRs, which would provide the best economic performance and an increased temperature safety margin. Under accident conditions, the use of SiC would not generate either hydrogen gas or heat in reaction with steam. Sauder (2014) listed some of the attractive attributes of using SiC composites for nuclear applications as: (1) high thermomechanical strength over a wide range of temperatures, in pile and out of pile, (2) high in pile thermal conductivity, (3) low activation under irradiation, (4) low swelling under irradiation even at doses higher than 40 dpa, and (5) high melting point. In the United States, the pioneer for the investigation of SiC for nuclear applications has been General Atomics, which has been partnering with Westinghouse in its development, including fabrication and testing (Deck et al., 2015).

Thermal properties and permeability of SiC/SiC fuel cladding A nuclear fuel cladding should have a high thermal diffusivity and thermal conductivity to remove the fission heat in an efficient manner and avoid thermal gradients across the cladding wall. For SiC/SiC composites, the thermal diffusivity will not be equal in all directions since fiber and monolithic material conducts heat differently and because the porosity distribution is not uniform in all directions (Deck et al., 2015). Helium permeability studies were conducted to examine the influence of the configuration of the cladding tube wall, from just SiC/SiC composite to composite with inner or outer layer of monolithic SiC material (Deck et al., 2015). As fabricated, the tubes with a monolithic layer passed the helium permeation tests. However the permeability efficiency was not known for an actual cladding application after thermal cycling and irradiation.

SiC/SiC fuel cladding, fabrication, and implementation There are two components in the use of SiC composites for ATF in LWRs: (1) SiC/SiC for BWR channels and (2) SiC/SiC for fuel cladding. The first application is a rather simple box, which does not need to be sealed hermetically, the channels are just panels that separate the fuel bundles from the control rods. However, for the second application, the fabrication and joining techniques need to be more elaborate. For fuel cladding, the SiC/SiC material needs to be made into 4
5 m long tubes, with a perfect circularity and free from paths that would allow the release of radionuclides from the fuel cavity to the coolant across the cladding wall. It is also important that the SiC/SiC would be compatible with the fuel and the coolant and that the material would undergo little or no dimensional changes due to thermal and irradiation progressions (Khalifa et al., 2015). To fabricate the fuel rods, the

Silicon carbide and ceramics metal composite Chapter | 6

149

tubes need to be capped with end plugs and then hermetically sealed. In the case of monolithic metallic cladding such as zirconium alloys, the end plugs are orbital welded to the tubes either using tungsten inert gas welding of pressure resistance welding. However, welding cannot be used for SiC/SiC materials and the ability to join SiC to itself is a critical issue that needs to be resolved if SiC are going to be applied for fuel claddings. Four end-plug designs for several joint geometries into monolithic and composite SiC tubes were produced and evaluated for mechanical and permeability performance (Khalifa et al., 2015). Monolithic SiC end plugs were inserted into the mating SiC tubes and the joints were sealed via a CVI. Katoh et al. (2014b) list several techniques that have been used to join a SiC tube to a SiC plug, including: (1) solid-state diffusion bonding, (2) glass ceramic joining, (3) transient eutectic phase (TEP) joining, (4) Si
C reaction bonding, (5) selective area chemically vapor deposited (CVD), (6) MAX phase joining, and (7) metallic braze-base joining (Katoh et al., 2014b). The development of an optimized SiC/SiC tube for LWRs fuel cladding may need first the development of a reliable nuclear-grade product with a reproducible and consistent fabrication path which needs to be feasible and scalable to industrial manufacturing (Deck et al., 2015). This fabricated tube should also meet operational performance requirements. The technology to produce a SiC/SiC product for cladding performance does not seem to exist yet (Xu et al., 2018). Up to 3-ft (91 cm) long prototypical LWR diameter SiC/SiC tube structures were fabricated for characterization (Deck et al., 2015). There were several variations of the tube structures including solely SiC/SiC reinforced composites and other tubes with a monolithic layer of SiC in the internal and external surfaces of the tubes. While the internal or external (or both) monolithic layer provides hermeticity and oxidation resistance while the composite fiber-containing layers provide improved strength and toughness to the tube (Deck et al., 2015). Besides resistance to hoop and axial stresses, other requirements from the fabrication process were the straightness and ovality (roundness) of the tubes. The 91 cm long tubes were able to meet the straightness metrics but not roundness, mainly because of excessive surface roughness (Deck et al., 2015).

Environmental behavior of SiC/SiC under normal operation conditions As mentioned before, SiC/SiC composite materials are desirable for ATF applications due to their low neutron absorption cross section, excellent resistance to high-temperature attack by steam with low hydrogen gas production, and good mechanical properties at high temperature. However, the corrosion resistance of SiC in coolant water typical of PWR operating conditions is one of the challenges (Hirayama et al., 1989; Terrani et al., 2015; Xu

150

Accident-Tolerant Materials for Light Water Reactor Fuels

et al., 2018; Terrani, 2018) (Fig. 6.1). The hydrothermal corrosion of SiC was known for more than 40 years and it was understood that high-purity SiC was more resistant to corrosion than industrial-grade material. The degradation process of SiC in condensed water appears to be the same as in water gas, that is, it is a two-step mechanism involving the initial oxidation of SiC to silica (SiO2) and then the removal of silica from the surface by the formation of soluble compounds (in condensed water) or the formation of volatile compounds in a gas atmosphere. The rate of removal of silica from the surface of SiC is much slower in gas atmosphere than in condensed water; therefore, SiC can be labeled as degradation resistant in gas water while not stable in condensed water. In condensed water, the rate limiting step is the oxidation of SiC to SiO2, since all formed SiO2 immediately dissolves in water as hydrated silicic acid Si(OH)4. If the condensed water at approximately 300 C contains driving forces (such as dissolved oxygen) for the SiC oxidation to SiO2, the entire degradation process will be faster. Therefore, the presence of oxygen or radiolytic species such as hydrogen peroxide in the hot water will only increase the dissolution process of SiC. Current LWR plants such as BWR and PWR rely on a hydrogen gas overpressure in the water at near 300 C to depress the redox potential in the system to minimize metal degradation processes such as stress corrosion cracking. The hydrogen water chemistry will also improve the performance of SiC in the reactor core, that is, it will decrease the dissolution rate of SiC in the water. In a high-temperature environment, SiC may react with oxygen or water gas to form a protective SiO2 (silica) on the surface according to the following equations (Hirayama et al., 1989): SiC 1 2O2 -SiO2 1 CO2 SiC 1 2H2 O-SiO2 1 CH4 and then the methane may combust according to CH4 1 2O2 -CO2 1 2H2 However, when condensed water is present, the protective silica which forms in air or water gas gets dissolved immediately in the water and diminishing the protection to the SiC substrate. Hirayama et al. (1989) prepared coupons of sintered SiC for corrosion immersion tests for up to 200 h in three environments at 290 C: (1) pure water, (2) 0.045 M Na2SO4 1 0.005 M H2SO4, pH 4 at temperature, and (3) 0.1 M LiOH pH 10 at temperature. The tests were performed in a refreshed autoclave, with either purged oxygen gas (oxygen concentration of 32 ppm) or nitrogen gas (having an oxygen concentration of ,20 ppb). After 72 h immersion, they reported that the mass loss of the coupons increased with the pH and that the mass loss was higher in the oxygenated environment than in the nitrogen gas

Silicon carbide and ceramics metal composite Chapter | 6

151

environment. At pH 10, the mass loss in the oxygenated environment was two orders of magnitude higher than in the nitrogen environment, but the difference in mass loss between the two environments decreased as the pH decreased. In oxygenated environments, the mass loss had a higher dependency with pH than in nitrogen environments (Hirayama et al., 1989). They also reported that as the immersion time increased, the mass loss in kg/m2 increased, and the slope was higher at pH 10 than at pH 4. Hirayama et al. (1989) postulated that the silica which forms on the surface of SiC due to reaction with water and oxygen in the water is dissolved generating protons in the process. Therefore, in alkaline solutions, the dissolution reaction of silica in water will be favored 1 22 1 SiO2 1 H2 O-H2 SiO3 -HSiO2 3 1 H -SiO3 1 2H

Since Hirayama et al. (1989) examined the corroded coupons and did not detect silica on the surface, they proposed the following dissolution reaction: SiC 1 4H2 O-SiðOHÞ4 1 CH4 1 22 1 SiðOHÞ4 -H3 SiO2 4 1 H -H2 SiO4 1 2H

They explained that the higher dissolution rate of SiC in oxygencontaining water was due to the release of four protons as the final reaction products: SiC 1 2O2 1 2H2 O-SiðOHÞ4 1 CO2 1 22 22 1 SiðOHÞ4 1 CO2 1 H2 O-H3 SiO2 4 1 HCO3 1 2H -H2 SiO4 1 CO3 1 4H

Hirayama et al. (1989) reported that at the higher pH, the attack also progressed faster because of preferential intergranular attack. Kim et al. (2003) studied the corrosion behavior of sintered SiC (SSiC) and CVD SiC in distilled water at 360 C for up to 10 days. They reported that as the exposure time increased the mass loss in mg/cm2 for both types of SiC increased. At each testing time the mass loss for the SSiC was higher than for the CVD SiC (of higher purity). Observations of the corroded coupons for both materials showed that the dissolution was not uniform, but it progressed preferentially through more active paths like grain boundaries. Since the SSiC had lower purity than the CVD SiC, it had higher mass losses. The authors could not detect either SiO2 or Si(OH)4 on the surface of the tested coupons, showing that all oxidized silicon immediately dissolved in water (Kim et al., 2003). Henager et al. (2008) tested high-purity CVD coupons of SiC for up to 4000 h in high-purity pH 4.6 water at 300 C and 10 MPa containing less than 10 ppb dissolved oxygen and 140 kPa of hydrogen overpressure. They reported a nonuniform dissolution of Si from the SiC specimen via a watersoluble silicon hydroxide. Park et al. (2013) performed immersion corrosion

152

Accident-Tolerant Materials for Light Water Reactor Fuels

testing for up to 90 days of CVD SiC in simulated normal operation conditions of LWRs at 360 C, both using circulating loops and static autoclaves. One autoclave recirculating loop had pure water (pH 7/7.5) and the other had PWR type water with 2.2 ppm Li and 650 ppm B (pH 6.8). The total mass loss of the CVD SiC coupons as a function of testing time was practically the same for the pure water and the PWR type water loops. They reported higher dissolution rates in the static autoclaves since the water may have had a higher oxygen content (Park et al., 2013). The corrosion attack progressed mostly in an intergranular manner. None of the tested coupons had any oxides (silica SiO2) on the surface even after 90 days of testing, suggesting the high solubility of silica in the water. Terrani et al. (2015) reported that the hydrothermal dissolution of SiC in typical BWR and PWR normal operation conditions environments progressed via the hydrolysis of silica to aqueous silicic acid. The dissolution of silica in water is highly dependent on the temperature, the crystal structure of the tested material and the pH of the environment (the solubility increases at pH . 8), and it was shown that amorphous silica has the highest solubility in water (Terrani et al., 2015). In addition to silica coupons, coupons of SiC were also tested by exposing to 200 cm3/min recirculating typical BWR and PWR water environments for up to 3 months. The BWR environments were at 288 C and contained either an excess of hydrogen (0.3 ppm H2) or oxygen (1 ppm O2), while the simulated PWR 330 C water was free of boron and lithium and was hydrogenated (3.57 ppm H2). Two types of SiC coupons were tested: (1) CVD monolithic SiC and (2) CVI SiC matrix reinforced with SiC fiber composites (SiC/SiC) (Terrani et al., 2015). They reported that the dissolution of CVD SiC was higher in the BWR water with oxygen than in the BWR water with hydrogen. The mass loss for the PWR (with hydrogen) was slightly higher than for the BWR (with hydrogen). In the three environments, the mass loss increased as the testing time increased. For the CVI SiC composite material, the coupons in the hydrogenated environments (BWR and PWR) gained mass for immersion times of 2 months. Fused silica coupons were also tested along the SiC coupons; however, the fused silica dissolved completely in the three environments only after 1 month of immersion (Terrani et al., 2015). Observation of the exposed coupons after the tests showed that in the hydrogenated environments (BWR and PWR), the CVD SiC coupons underwent selective dissolution and the CVD coupons in the BWR water with oxygen underwent intergranular corrosion. In the case of the CVI SiC/SiC coupons, the dissolution pattern in the three tested environments was more uniform than for the CVD coupons. Immersion corrosion test results suggest that SiC dissolves in B300 C water by forming silica (SiO2) as an intermediate step. Then the silica dissolved in water as silicic acid (Si(OH)4). The overall rate limiting step for the recession of SiC is not the dissolution in water of silica (SiO2) but rather the rate limiting step is the formation of silica from SiC. Silica was never observed

Silicon carbide and ceramics metal composite Chapter | 6

153

on the surface of any tested SiC coupon, meaning that as soon as it forms from SiC, it dissolves in water. Under irradiation, the dissolution rate of SiC in LWR environment will only increase due to water radiolysis which increases the oxidizing power in the environment and because irradiation makes the SiC more defective and therefore more prone to corrosive attack (Terrani et al., 2015; Kondo et al., 2015). Kondo et al. (2015) studied the effect of irradiation on the corrosion behavior of high-purity polycrystalline coupons of CVD SiC in 320 C simulated LWR environments at 20 MPa. The irradiation was performed before the coupons were immersed for corrosion besting by 5.1 MeV Si21 ions at 400 C and 800 C to dose levels between 0.14 and 2.6 dpa. The corrosion tested coupons had irradiated areas and nonirradiated areas; therefore, the effect of irradiation on the amount of corrosion damage could be straightforward. After the immersion test for 168 h, the irradiated portion of the coupon had higher recession than the nonirradiated portion of the coupon (Kondo et al., 2015). The corrosion attack in general was highly guided by high misorientation crystallographic patterns in the coupons such as grain boundaries and stacking faults. As mentioned earlier, SiC/SiC composite materials are desirable for ATF applications due to their low neutron absorption cross section, their excellent resistance to high-temperature attack by steam with low hydrogen gas production, and their good mechanical properties at high temperature. However, the corrosion resistance of SiC in coolant water typical of PWR operating conditions is one of the challenges (Terrani et al., 2015; Xu et al., 2018). Two issues of concern are: (1) can the microstructure and composition of the SiC/SiC be optimized to resist environmental degradation and (2) what would be the effect of in reactor irradiation on the corrosion resistance of the composite. Several types of SiC/SiC configured specimens including CVI, CVD, and TEP were tested out of pile at 343 C and 15.5 MPa in PWR type water containing B800 ppm boron, B3.1 ppm lithium, and dissolved hydrogen gas (100 mL H2/kg water) (Xu et al., 2018). The amount of dissolved oxygen was less than 2 ppb. SiC dissolved in the high-temperature water and the imposed limit was not more than 0.034 mg/dm2/day and it was reported that some specimens had corrosion rates lower than the maximum allowed (Xu et al., 2018). Immersion corrosion tests were also performed under gamma irradiation. Results showed poor performance for the TEP specimens but better performance by the CVD specimens. The authors suggested that irradiation may increase the corrosion rates of SiC/SiC as compared to outof-pile tests (Xu et al., 2018).

Environmental behavior of SiC/SiC under accident conditions One of the main reasons why SiC/SiC composites have been selected for ATF cladding is due to their unparalleled resistance to high-temperature

154

Accident-Tolerant Materials for Light Water Reactor Fuels

attack by air and steam (Fig. 6.1). The reactivity between monolithic or composite SiC and steam at high temperatures has been studied initially for aerospace applications and later for nuclear applications in the case of a loss of coolant accident (Terrani et al., 2012). The parabolic oxidation rate of SiC coupons in 100% steam at 1200 C is approximately three orders of magnitude lower than the rate of oxidation for zirconium alloys and approximately six times lower than the rate of oxidation for FeCrAl APMT (advanced powder metallurgy tubing) alloy (Terrani, 2018). This very slow oxidation rate of SiC in steam will release less hydrogen to the environment and will produce less heat of reaction if compared to the current zirconium alloy system. The attack of SiC in high-temperature steam follows a two-step process: first with the formation of a scale of silica (SiO2) on the surface following a parabolic kinetic law and then the subsequent linear kinetic evaporation of silica from the surface by the formation of volatile hydroxides (Nguyen et al., 2004; Terrani et al., 2012). As said, the degradation of SiC at high-temperature ( . 1200 C) steam occurs in two steps (Angelici Avincola et al., 2015): SiC 1 3H2 O-SiO2 1 3H2 1 CO followed by SiO2 ðsÞ 1 2H2 OðgÞ-SiðOHÞ4 ðgÞ or SiO2 ðsÞ 1 H2 OðgÞ-SiOðOHÞ2 ðgÞ or SiO2 ðsÞ 1 0:5H2 OðgÞ-SiOðOHÞ ðgÞ SiC/SiC tube specimens were exposed at 1600 C for up to 64 h in argon and argon plus 10, 30, and 60 kPa water gas. The tubes had three layers, the internal monolithic SiC, the intermediate fibers, and the external CVD layer (Angelici Avincola et al., 2015). As the testing time and the amount of water gas increased from 10 to 30 kPa, the mass gain by the SiC specimens increased. The mass increase behavior for 60 kPa of water gas was irregular. SiC tube specimens were also resistant to cracking upon quenching in 90 C water after exposures to 1600 C and 2000 C in environments of argon plus steam. That is, the SiC tubes maintained their coolable shape after quenching (Angelici Avincola et al., 2015). It was also stated that during hightemperature oxidation of SiC specimens, the amount of hydrogen gas generated was at least 40 times lower than for Zircaloy coupons. Monolithic tubes of sintered SiC from two different vendors and bar-shaped CVD SiC were tested in pure steam at 1 atm in the temperature range 1140 C
1500 C for up to 72 h (Lee et al., 2016). They reported that the mass loss rates increased with the temperature and with the steam flow rate in the test chamber. The effect of the steam flow rate was higher at the higher temperatures. They remarked that the degradation of SiC by steam was a surface thinning phenomenon which did not change the mechanical integrity of the rest of the specimen.

Silicon carbide and ceramics metal composite Chapter | 6

155

Irradiation behavior Before nuclear-grade SiC/SiC composite was considered for structural applications in LWRs, their neutron irradiation response was evaluated in the TRISO (tristructural-isotropic) fuel program (Snead et al., 2007; Yueh and Terrani, 2014; Terrani et al., 2015). The effect of irradiation is highly dependable on the type of SiC product that is being irradiated, since a single crystal would not have the same response as a CVD material or the higher defect materials produced by industrial processes such as sintering, or polymer derived. The effect of irradiation in the temperature range from ambient to 1000 C was not significant when the SiC was close to stoichiometry (Snead et al., 2007). Neutron irradiation produced swelling, which was temperature dependent. However, the swelling saturated at relatively low doses of 1 dpa for temperatures of less than 1000 C (Yueh and Terrani, 2014). A decrease in the thermal conductivity as the neutron irradiation dose increased has been reported for CVD SiC (Snead et al., 2007) (Fig. 6.2). The most significant damage produced by neutron irradiation of SiC is swelling, since it correlates with the amount of introduced effects during

FIGURE 6.2 Decrease in thermal conductivity of SiC (Snead et al., 2007). CVD, Chemically vapor deposited; SiC, silicon carbide.

156

Accident-Tolerant Materials for Light Water Reactor Fuels

irradiation (Ben-Belgacem et al., 2014). The irradiation induced defects will affect properties such as thermal conductivity. For nuclear applications, stoichiometric and crystalline SiC materials have shown unparalleled stability in high radiation conditions (Katoh et al., 2014a). The closer to stoichiometry of both the fibers and the matrix, the better the resistance to irradiation. The presence of impurities in the SiC/SiC tends to be detrimental for irradiation tolerance, especially on thermal conductivity. Katoh et al. (2014a) found minimal effect of neutron irradiation on several of the nuclear-grade SiC/SiC properties, including coefficient of thermal expansion, density, electrical conductivity, specific heat, mechanical properties, but not for thermal conductivity.

Licensing for reactor use The use of SiC composite materials for channel and cladding application in LWRs is a revolutionary concept since for six decades only metals and alloys have been used inside the reactor. The first evolutionary step for the ceramic ATF concept is to apply SiC for components that are not critical in the release of radionuclides to the environment. For example, channel boxes in BWR are the perfect first step since it would allow to obtain data on the interaction of SiC composites with the coolant under actual commercial reactor irradiation. The second step would be the use of SiC composites to contain the fuel. There are two current challenges to this application, one is the proper sealing of the container after introducing the fuel and the second is the inherent typical microcracking, which was engineered into the composite. The microcracking is a characteristic of the SiC composite to make it more resistant to catastrophic failure, but the microcracking can offer a path for the radioactive species to migrate from the fuel side to the coolant side of the fuel rod (Terrani, 2018). One of the solutions to avoid a full pathway from the fuel side to the coolant side is to apply an external coating to the SiC composite cladding. This external coating may serve to minimize radionuclide release and also to minimize hydrothermal corrosion. By its own nature, a SiC composite is heterogeneous and anisotropic. It is anticipated that before SiC composites find approval for use as structural components inside the reactor, a consensus may need to appear in the materials community as what is a SiC/SiC composite, and what is the architecture of the cladding wall.

Chapter 7

Alternative fuels to urania Chapter Outline Overview Introduction The urania nuclear fuel The urania excellent performance Accident tolerant fuels under consideration Improved urania fuels by doping Modified urania performance under normal operation conditions Modified urania performance under accident conditions Higher density fuels: uranium silicide

157 158 159 159 160 160 163 164 164

Reactivity of uranium disilicide Reactivity of U3Si2 with the cladding Fabrication and implementation of U3Si2 fuels Higher density fuels: uranium nitride Reactivity of uranium mononitride fuel Fabrication paths for uranium mononitride Behavior of uranium mononitride under irradiation

165 167 167 168 169 169 170

Overview The fueling of the power reactors happens at relatively long intervals of up to 2 years, which is a convenient way of reducing operation costs by cutting down on fueling shut down periods. To be able to accommodate long runs, the fuels must comply with strict characteristics and properties. Uranium dioxide (UO2) or urania has been used reliably in light water reactors for several decades; therefore newer proposed fuels must meet a high standard to be seriously considered. The nearest term accident tolerant fuel (ATF) is a slight modification of the current urania by doping it with chromia to increase its thermal conductivity and to make it more resistant to crumbling at high irradiation doses. Like the case of the chromium-coated zirconium alloys for cladding, the doped urania fuel may have the easiest and fastest path for regulatory licensing approval. The second family of ATFs considered is called the high uranium density fuel because due to their specific stoichiometry these compounds contain more atoms of uranium per unit mass of fuel than urania. The two most advanced fuels in the high-density family are uranium silicide and uranium nitride. The largest handicap of the two high-density fuels is their high

Accident-Tolerant Materials for Light Water Reactor Fuels. DOI: https://doi.org/10.1016/B978-0-12-817503-3.00007-9 © 2020 Elsevier Inc. All rights reserved.

157

158

Accident-Tolerant Materials for Light Water Reactor Fuels

reactivity with water and steam in the case of a cladding breach. Not much is known of their response to an actual nuclear reactor environment. The uranium silicide fuel is currently under irradiation in the advanced test reactor and in a commercial reactor as part of a fuel vendor evaluation.

Introduction The aim of nuclear power is to convert heat energy from the fission of uranium (U-235) into high-quality steam to turn steam turbines and generate clean reliable electricity. There are large intervals (18 24 months) between the refueling of the nuclear power plants; therefore the fuel that is used in these plants needs to be safe and well characterized to last at least for the 2 years between fueling outages of the reactor core. The performance integrity of the fuel is important not only under normal operation of the reactors but also under anticipated operational occurrences (AOOs). The safety of the fuel is a direct result of its stability in the fuel cavity, including resistance to fragmentation and the avoidance of pellet cladding interaction (PCI) that may lead to stress corrosion cracking from the ID of the cladding. The newer accident tolerant fuels (ATFs) must show a robust performance not only under design basis accidents (DBAs) but also beyond design DBAs (BDBAs) (Zinkle et al., 2014). Monolithic urania fuel served the nuclear power industry well for over six decades since first implemented with the Zircaloy cladding in the mid-1950s at the Shippingport power station. If a new fuel needs to be realized, it should show definitive benefits over urania. These benefits may include enhanced fission product retention at higher temperatures and better thermal, mechanical, and physical properties (Zinkle et al., 2014). The disadvantages of implementing newer fuels include: (1) unknown fabrication costs, (2) possible changes in the reactor physics design, (3) compatibility between the fuel and the cladding and stability of the fuel in the coolant, (4) optimized fuel enrichment level, (5) irradiation behavior under reactor conditions, and (6) determination of the behavior of the fuel under loss-of-coolant accidents (LOCAs), reactivity initiated accidents, and BDBA scenarios (Zinkle et al., 2014). As part of the ATF studies, there is an international effort to revisit studies on the selection of not only the cladding for the fuel rods (as in the previous chapters) but also of the form and nature of the fuel itself, as alternatives to the current well-established monolithic urania (UO2). The two main areas of the alternative fuels are: 1. improved or doped urania and 2. higher density fuels The improved or modified urania fuel may be the result of adding less than 0.2% of species (usually called doping) such as chromia to improve

Alternative fuels to urania Chapter | 7

159

properties such as thermal conductivity, change the mechanical properties of the fuel or increase its grain size to better retain fission products within the fuel after in-reactor irradiation. The limited improvement that the dopants may produce via an increase in thermal conductivity may be nullified by the decrease in the density of uranium in the fuel and therefore decreasing the fuel efficiency (NNL, 2018). The higher density uranium fuels imply a fuel compound that has a higher stoichiometric proportion of uranium in its stoichiometric formula. The studied fuels which have a higher uranium density than urania include uranium carbides, borides, nitrides, and silicides (NNL, 2018). Some of the disadvantages of these higher density fuels are their poor compatibility with water in case of a breach in the cladding, meaning that more fuel would be dissolved into the coolant. Another uncertainty about the higher density fuels is the lack of a proven reliable and reproducible industrial fabrication path (NNL, 2018).

The urania nuclear fuel Commercial nuclear power around the world started with the development of power reactors for submarines in the United States and in the Soviet Union. Over the years, several fuel materials were investigated for use in power reactors including metals and alloys, oxides, carbides, hydrides, nitrides, and fluorides (Simnad, 1981). However, from the very beginning in the late 1940s, there was a favorite fuel which met the requirements to produce electricity using nuclear fission, the monolithic uranium dioxide or urania. When in the 1950s and 1960s, the commercial nuclear light water power reactors were starting to be connected to the grid, the fuel of choice of the reactors was always urania. And urania continued to be the optimal fuel of until the present day of 2019 (NNL, 2018). Urania has a high melting point of 2847 6 30 C, it has a good compatibility with water, and its behavior under reactor irradiation conditions was well investigated (NNL, 2018).

The urania excellent performance Urania (UO2) has been used as fuel for light water power reactors in the United States since the Shippingport nuclear power station started to generate electricity in 1957. One of the most outstanding properties of UO2 as a light water reactor fuel (LWR) fuel form is its thermodynamic stability. Until today, urania is the fuel that had the lowest reactivity with water in the 300 C 400 C temperature range of all the proposed uranium fuel forms. Decades of experience with urania in commercial power plants has shown that even during cladding breaches of many forms, the urania pellets were stable enough in water to avoid serious consequences to the reactor or the environment.

160

Accident-Tolerant Materials for Light Water Reactor Fuels

The further oxidation of urania in air, steam, and water has been investigated for decades, not only regarding its performance under normal operation condition in the reactor but as spent or used fuel after removal from the reactor (McEachern and Taylor, 1997). The oxidation of urania in air follows a two-step process: UO2 -U3 O7 =U4 O9 -U3 O8 If water is present, it may aid to the oxidation by the formation of hydrated species or by helping on the diffusion of oxygen through the external oxide layer. In the case of spent fuel or used fuel, water may facilitate enhanced kinetics by the formation of radiolytic species such as hydrogen peroxide.

Accident tolerant fuels under consideration The required properties or attributes of the ATFs are those that would mitigate the consequences of an accident when compared with the current system of zirconium alloy and monolithic urania. These mitigating attributes include: (1) an enhanced retention of fission products, (2) a higher rate of heat dissipation, and (3) a diminishing interaction with the cladding (BraggSitton et al., 2014; NEA, 2018). Similarly, as with the cladding, the proposed ATFs may be evolutionary or revolutionary. The evolutionary concept will imply a small change to the existing fuel, with the likely positive benefit of a faster regulatory implementation. A revolutionary or total change of the fuel may carry a higher burden in the likelihood of commercial application. As mentioned before, there are two main branches in the alternative ATFs under study, including: 1. improved or modified urania (evolutionary) and 2. higher density uranium-based fuels (revolutionary) such as U3Si2 or UN.

Improved urania fuels by doping The improved urania fuel is an evolutionary concept. The improvement can be achieved by addition of dopants which may facilitate the retention of fission products or increase their thermal conductivity. Studies of adding dopants to urania started years before the Fukushima accident with the objective of increasing the grain size in the fuel and to increase the plasticity of the material in order of diminishing the mechanical interaction with the cladding (Delafoy et al., 2015; NEA, 2018). The objective of modifying the urania fuel with chromia was to improve the fuel economics and performance without reduction in safety margins (Delafoy et al., 2015). They reported that adding 0.16 wt.% chromia to an amount near its solubility limit in urania favors grain growth and increases the mechanical properties of the fuel,

Alternative fuels to urania Chapter | 7

161

which may provide margins on the operational flexibility of light water reactors (Cardinaels et al., 2012; Delafoy et al., 2015). A larger grain size in the order of 50 µm may delay the release of fission gases from the fuel. Delafoy et al. (2015, 2018) describe the advantages of a chromia-doped urania fuel for M5 tube cladding. One advantage is the increased mechanical dependability and integrity which will ease manufacturability by reducing the likelihood of missing pellet surface developments during production. The chromia addition also increases the resistance to interaction with water by the formation of an external barrier film which reduces further attack by the coolant (Delafoy et al., 2015). Chromia-doped urania fuels with up to 4.95 wt.% of U-235 have been undergoing exposures in both boiling water reactor (BWR) and pressurized water reactor (PWR) power reactor irradiation studies since 1997 (Delafoy et al., 2015). In PWR conditions, the doped pellets were clad with M5 material, and in BWR reactors, the cladding was performed with Zircaloy-2, including softer zirconium lined and nonlined tubes. Burnups higher than 75 GWd/tU have been reached (Delafoy et al., 2015). The focus of the postirradiation analyses was mainly to acquire data on fission gas release and to characterize the PCI behavior. By measuring the thermal conductivity as a function of temperature between 200 C and 1600 C of PWR irradiated doped urania fuel with a burnup value of 65 GWd/tU, they concluded that there was insignificant effect on the addition of chromia as compared to the behavior of traditionally undoped urania (Delafoy et al., 2015). They also reported a high dimensional and structural integrity of the doped urania, especially having a high resistance to densification. Changes introduced during manufacturing allowed for the optimization of grain size, total porosity, porosity distribution, and swelling rate under irradiation. When the chromia-doped urania fuel is irradiated up to 50 GWd/tU, the amount of fission gas release was like for the nondoped urania fuel. However, for higher irradiation dosage, the amount of fission gas release was approximately 15% 25% lower for the doped than for the nondoped fuel (Delafoy et al., 2015). When a microstructural examination of the irradiated fuel was conducted, it was determined that an important proportion of the fission gas in the doped fuel was precipitated solid within the fuel. The fact that the fission products were trapped or captured by the fuel reduced the total pressure in the ID of the tube at the end of life of the rod in the reactor (Delafoy et al., 2015). This relief in the pressure built-up inside tubes loaded with doped urania was more evident for the higher fuel burnups ( . 50 GWd/tU). Power ramp tests were performed to determine the comparative PCI behavior between traditional urania fuel and chromia-doped urania fuel for rods irradiated both under PWR and BWR conditions (Delafoy et al., 2015). The parameters employed in the ramp tests were optimized to take full advantage of failure risks regarding the PCI mechanism effect. For example, the total irradiation burnup of the ramped rods was in the range

162

Accident-Tolerant Materials for Light Water Reactor Fuels

of 20 45 GWd/tU since these rods are the most susceptible to PCI failures. The ramp tests were also performed in the range of expected not only under normal operation conditions but also under AOOs. For burnup values higher than 45 GWd/tU, the most likely mechanism of failure would be related to hydrogen uptake by the cladding (Delafoy et al., 2015). Two different power ramp tests were performed: (1) starting at a lower level of 10 kW/m and ending at 40 kW/m and (2) starting at a higher power level of 20 25 kW/m and ending at 30 kW/m. The power ramp tests were single-step ramps (as opposed to multiple smaller staircase ramps) to make the tests more aggressive by avoiding relaxation of the cladding during a longer incremental staircase ramp. The conducted tests would not only determine the PCI effect but also the consequence of increased pressure in the rods due to fission gas release inside the cladding (Delafoy et al., 2015). Power ramp results showed that Zircaloy-2 BWR irradiated rods containing chromia-doped fuels had an increment in power of 7 10 kW/m to produce a similar PCI-related damage when compared with both newer lined and legacy nonlined irradiated rods containing nonchromia-doped urania fuel. In the case of PWR type irradiated rods made with M5 cladding, the rods with chromia-modified urania had a power increment of 4 kW/m to fail by PCI as compared to similar irradiation level rods with nondoped fuel (Delafoy et al., 2015). The overall technological limit for PCI is improved by 35% by using the chromia-doped fuel instead of the urania fuel without doping (Delafoy et al., 2018). Postirradiation examinations (PIE) studies were conducted to determine which material properties of the chromia-doped urania rods had the responsibility of providing the power gain during the single ramp tests. The PIE tests showed a similar behavior between BWR and PWR tested rods with chromia-doped fuel, since for both types of rods, an equivalent linear relationship was found between the fraction of fission gas release and the final power level used in kW/m (Delafoy et al., 2015). However, for the PWR rods with traditional nondoped fuel, the fraction of fission gas release had an exponential relationship with the final power used in kW/m, meaning much larger fractions of fission gas release at lower final powers in the ramping studies. For final power ramp values in the order of 45 50 kW/m, there was a 50% lower fission gas release when the irradiated rods contained chromiadoped urania fuel pellets (Delafoy et al., 2015). PIE studies also revealed a larger creep of the tubes during the power ramps when chromia-doped urania was used in the rods. A second observation from PIE studies was that the chromia-doped pellets had numerous smaller radial cracking in the periphery of the pellet, suggesting a more uniform distribution of strain during the power ramps thus avoiding larger localized strains that may lead to PCI failures. Delafoy et al. (2015) also reported that during the PIE studies the fuel contained metallic chromium, that is, some of the initial chromia in the fuel was reduced releasing oxygen into the cavity of the fuel, and this extra oxygen may have contributed to PCI mitigation measures.

Alternative fuels to urania Chapter | 7

163

Ja¨derna¨s et al. (2015) reported that the properties of urania fuel can be adjusted by incorporating additives or dopants as well by changing the fabrication procedures such as sintering variables. The additives may change the gas diffusion rate within the fuel pellet due to a change in the defect structure in the lattice. The additives can also change the creep rate of the fuel by promoting grain growth in the pellet. These chemical and mechanical changes in the fuel pellets may minimize PCI failures of the rods (Ja¨derna¨s et al., 2015). During one investigation, three different additives were incorporated into urania fuel and later irradiation and power ramp tested in the Halden reactor in Norway. The additives were Al2O3 SiO2, Al2O3 Cr2O3, and Gd2O3. The average burnup of the experimental and baseline fuels ranged from 44 to 58 MWd/kgU and the power ramp applied ranged from 31 to 48 kW/m. They reported that the fuel containing Al Si best retained the volatile fissile products and did not allow them to reach the cladding (Ja¨derna¨s et al., 2015). However, a conclusive explanation on the interaction between iodine and the cladding was not provided. Regarding reprocessing of the fuel, comprehensive scoping analysis has shown there is no significant difference between chromia-doped urania and standard nondoped urania. A urania fuel doped with alumina and chromia (named ADOPT or advanced doped pellet technology) has exhibited several benefits under normal operation conditions including: (1) a 50% reduction in fission gas release, (2) an improved PCI performance mainly due to the increased plasticity of the pellet, and (3) higher resistance against postfailure degradation (reduced rate of fuel washout) (Arborelius et al., 2006). Other possible beneficial effects such as the changes of the oxygen potential and formation of alternate/modified fission product secondary phases as a result of the dopants in the pellets are under investigation.

Modified urania performance under normal operation conditions Delafoy et al. (2018) reported in their studies of fuel rods made with M5 cladding and chromia-modified urania fuel pellets will improve the performance of the fuel under normal operation conditions when compared with the traditional rods made with nondoped urania fuel. They listed the benefits as (1) increase in fission gas retention which is trapped inside large grains of the modified fuel and would overall decrease the pressure in the fuel cavity by approximately 20 bars, and an (2) increase in PCI resistance with the modified fuel which would allow for extended power limits and larger degree of freedom in the operation of the light water reactors. If there is an unforeseen contact between the fuel and the coolant, the chromia-doped urania fuel will have a five times better resistance of release of fuel material into the coolant than the nondoped fuel. Since the chromia-doped fuel would

164

Accident-Tolerant Materials for Light Water Reactor Fuels

have better mechanical properties and stability, the pellet can be processed with fewer defects and therefore the operational cost may be reduced, and the safety margin may be increased (Delafoy et al., 2018).

Modified urania performance under accident conditions Since one of the attributes of the chromia-doped urania is to decrease the internal pressure of the rod, this development will at the same time decrease the susceptibility of cladding ballooning during an accident. The larger grain and better initial integrity of the modified fuel would reduce fragmentation during an accident and therefore limit the fuel segregation in the balloon during LOCA (Delafoy et al., 2018). A reduced fragmentation would be positive since less amount of fuel debris may be ejected from the cladding into the coolant.

Higher density fuels: uranium silicide Compounds of uranium and silicon, and especially uranium disilicide (U3Si2) have been considered for light water ATF application mainly because of its: (1) higher density of the element uranium in the formula and (2) higher thermal conductivity as compared to monolithic urania (White et al., 2015; Middleburgh et al., 2016; Sooby Wood et al., 2017b; Hoggan et al., 2018; Antonio et al., 2018). One of the detrimental aspects of the silicide fuels is its lower melting point as compared to urania (Table 7.1) (Harp et al., 2015). However, since the disilicide has also higher thermal conductivity, it may not need the highest melting point requirement. The density of uranium in U3Si2 is 11.3 gU/cm3; compared with 9.7 gU/cm3 in the monolithic urania fuel. This represents a 16.3% increase in the mass of U-235 in the silicide fuel compared with the urania fuel. The higher load of uranium

TABLE 7.1 Properties of fuels (Harp et al., 2015). Property

Urania

Uranium disilicide (U3Si2)

Melting point ( C)

2847

1665

Thermal conductivity (W/mK) (400 C 1200 C)

6 2.5 (it decreases with the temperature)

13 22.3 or 16.1 28.2 (it increases with the temperature)

Density of U-235 (gU/ cm3)

9.7

11.3

Theoretical fuel density (g/cm3)

10.96

12.2

Alternative fuels to urania Chapter | 7

165

atoms per unit volume of the fuel could be important in the case of using a higher neutron absorption ATF cladding such as a FeCrAl alloy (Harp et al., 2015; He et al., 2018; Hoggan et al., 2018). The larger uranium content per unit volume may enable power uprates, extend cycle length in LWRs, or reduce enrichment, which result in positive economic outcomes for the operation of the commercial power reactors (Harp et al., 2015). The thermal conductivity of traditional urania is rather low and it promotes large temperature gradients between the fuel and the coolant. The thermal conductivity of urania decreases as the temperature increases, while the opposite is true with uranium disilicide (Hoggan et al., 2018). The low thermal conductivity of urania results in high fuel centerline temperatures during reactor startup and operation (White et al., 2015). A low fuel centerline temperature is desirable: (1) to reduce induced thermal stresses, (2) to minimize the diffusion of fission products to the OD (outside diameter) of the fuel pellet, and (3) to decrease fuel fracturing. One of the advantages of uranium disilicide is that the thermal conductivity increases as the temperature increases, that is, the heat dissipation from the fuel will be higher at the higher temperatures (Middleburgh et al., 2016). A higher thermal conductivity is desired to keep the radial temperature more uniform to reduce interactions with the cladding. The stoichiometric stability of U3Si2 was evaluated using models and calculations for simulated irradiated material and it was determined to be resistant to transitions to hypo or hyper stoichiometry (Middleburgh et al., 2016). Modeling studies were conducted considering the following combinations: (1) uranium disilicide fuel coupled to a FeCrAl cladding, (2) urania fuel coupled to Zircaloy-4 cladding, and (3) uranium disilicide fuel coupled to Zircaloy-4 cladding (He et al., 2018). Results showed that in the uranium disilicide/FeCrAl pair the fuel would have a lower centerline temperature by approximately 300 C, and have a flat radial temperature, which was attributed to the higher conductivity of the fuel. This lower temperature in the fuel may result in avoidance of crumbling of the fuel, a reduction of fission gas release, and a decrease in gap pressure, which would minimize interactions between the fuel pellet and the cladding (He et al., 2018).

Reactivity of uranium disilicide Some of the detrimental properties for uranium disilicide as ATF are the low melting point and the susceptibility to oxidation (Sooby Wood et al., 2017a). Results reported indicate that uranium disilicide would oxidize in air rather rapidly at temperatures as low as 400 C following a w 5 ktn law which maybe parabolic or linear depending on the temperature range. The corrosion products maybe a layer of urania; however, silica (SiO2) was not found on the oxidized specimens, suggesting instant evaporation as soon as it forms (Sooby Wood et al., 2017a). The oxidation kinetics of U3Si2 in air was

166

Accident-Tolerant Materials for Light Water Reactor Fuels

evaluated as a function of the temperature between ambient and 1000 C. The authors used thermogravimetric, X-ray diffraction, and scanning electron microscopy analysis to show that uranium disilicide behaved practically like uranium metal in its reaction with air, that is, there is no discernible effect of silicon on the oxidation rate of the fuel compound (Sooby Wood et al., 2017a). The oxidation of the disilicide progressed mainly by inter- and intragranular paths which lead to pulverization of the tested specimens. The authors mentioned that these tests do not represent LWR operating or accident conditions since there was no water, steam, or hydrogen involved. During normal operation conditions the fuel is protected from the coolant water by the hermetic cladding, which allows for heat transfer from the fuel to the cladding but avoids intimate contact between them. However, in the case of a cladding breach, the coolant may enter in contact with the fuel (Nelson et al., 2018). Therefore it is important to determine the stability of the fuel in contact with water to determine the likelihood of the water dispersing radioactive elements to the environment. Monolithic specimens of U3Si2 of 94% 96% theoretical density were exposed to flowing steam and to an argon-6% hydrogen atmosphere (Sooby Wood et al., 2018). The mass change in the tested specimens was monitored using thermogravimetric analysis either under isothermal tests between 290 C and 400 C or during temperature ramping up to 1000 C. In the ramping test to 1000 C in steam, the U3Si2 pellet specimen was totally consumed when the temperature reached 459 C. In another similar test in steam, using a urania pellet specimen, the temperature reached 1000 C and a dwell time of 1 h was added. The urania pellet specimen gained less than 0.1% of mass after 1 h exposure to 1000 C steam (Sooby Wood et al., 2018). When the U3Si2 pellet specimen was exposed to a steam flow under isothermal conditions, it darkened and crumbled after 6 h at 350 C, and it sustained a total mass loss at 375 C after 4 h and at 400 C after 1.5 h (some of the mass loss may have occurred not by consumption of the specimen but sometimes by physical ejection from the test crucible due to a forceful reaction with the environment). When the isothermal exposures of the U3Si2 pellet specimen were exposed to the argon 1 6% H2 stream environments, the reaction was even faster in H2 gas than in the steam stream (Sooby Wood et al., 2018). In the hydrogen stream, the pulverization of the U3Si2 pellet specimen occurred fast, since it took 1 h at 350 C, 0.7 h at 375 C, and 0.5 h at 400 C for the complete reaction. Nelson et al. (2018) immersed three types of fuel pellet specimens (urania, U3Si2, and uranium mononitride UN) in hydrogenated water from 300 C to 350 C. A pulverization time of less than 50 h was observed for U3Si2 at 350 C. When the temperature was 300 C and the hydrogen content in the water was between 1 and 5 ppm, the U3Si2 pellet specimen lasted approximately 1 month (Nelson et al., 2018). In comparison, the urania pellet specimen that was tested in parallel did not suffer any appreciable damage under

Alternative fuels to urania Chapter | 7

167

all hydrogen concentrations at 300 C 350 C. In general, in any type of water contact test, the behavior of U3Si2 was always inferior of that of traditional urania (Nelson et al., 2018).

Reactivity of U3Si2 with the cladding The out of pile reactivity between U3Si2 and FeCrAl couples was investigated under vacuum by exposing the two materials in tight contact at 500 C, 600 C, 800 C, and 1000 C for 30 and 100 h (Hoggan et al., 2018). They reported that at the higher temperatures (800 C 1000 C), there was some diffusion of uranium into the cladding material and some iron into the U3Si2 fuel. A B10 µm thick bond layer developed between the cladding and the fuel at 1000 C for 100 h and this layer had an enrichment in molybdenum and silicon (Hoggan et al., 2018).

Fabrication and implementation of U3Si2 fuels The fabrication process of urania fuel has been proven through decades of commercial manufacturing and plant performance. The following fabrication paths have been investigated for U3Si2: (1) powder metallurgy, (2) ball milling, and (3) arc melting. Some U3Si2 fuel products may end with leaner amounts of silicon (hypostoichiometric) or containing secondary phases (Noordhoek et al., 2016). Moreover, the fabrication processes for uranium silicide are still in laboratory or pilot plant scale since an industrial costeffective path for the production of U3Si2 has not been identified yet. Harp et al. (2015) manufactured pellets of uranium disilicide using a laboratory setup by mixing powders of the elements uranium and silicon in the required stoichiometric amounts. The added amount of silicon was above the stoichiometric requirement to compensate for any silicon loss during the arc melting step. The two elements mixed powder were pressed into a compact and then annealed at 1450 C for 30 min followed by three times arc melting to complete the reaction of both elements into the U3Si2 compound. The first arc melting was at the lowest current and then the current was gradually increased for the next two melting steps. The premixing of the powders method in the stoichiometric ratio provided a final fuel ingot which was 97% U3Si2 when compared with other manufacturing methods in which bulk pieces of uranium and silicon were arc melted, given way to a fuel that contained 10% of U3Si2 (Harp et al., 2015). After the third arc melting was completed, the U3Si2 ingot product was converted into powder by a comminution planetary milling process using several steps with decreasing zirconia milling media sizes. Finally, the powder was pressed into pellets, sintered, and machined using centerless grinding to their desired final dimension (Harp et al., 2015). The pressing into pellets step needs to be optimized for the

168

Accident-Tolerant Materials for Light Water Reactor Fuels

values of pressure used since higher pressures did not necessarily result in better pellets. Sintering was performed in an argon controlled atmosphere (with low oxygen concentration) for times ranging from 2 to 8 h and sintering temperatures from 1200 C to 1550 C. After centerless grinding, the pellets had an average diameter of 8.19 mm and an average height of 6.12 mm with a density above 94% of the theoretical density (Harp et al., 2015).

Higher density fuels: uranium nitride As mentioned in previous sections, urania (UO2) fuel has been in use for civilian light water power reactors, since it was initially recommended in the early 1950s at the Bettis Laboratory in Pittsburgh. One of the main reasons for the use of urania was its low reactivity with the coolant in the case of a cladding breach. For many decades, other types of fuel or compounds of uranium were also explored and studied even though none of them gained the traction needed to be implemented in a utility power reactor (Simnad, 1981). One of the many compounds was uranium mononitride (UN) (Taylor and McMurtry, 1960; Evans and Davies, 1963; Simnad, 1981). One of the reasons the uranium nitride compound was ever considered is because the stoichiometry allows for a higher density of the element uranium than in urania, at 71% in UN versus 53% in UO2 (Table 7.2) (Evans and Davies, 1963). Another parameter for evaluating the performance of a fuel rod is the thermal power per unit length, which is connected to the thermal conductivity of the fuel material from the allowed centerline temperature to the intended temperature at the edge of the fuel pellet (Simnad, 1981). The use of uranium nitride in a light water reactor may increase the residence time of the fuel in the reactor without changing the enrichment (Zakova and Wallenius, 2012). Besides its higher uranium density, recently it was suggested to use

TABLE 7.2 Properties of uranium compounds. Compound

Density (g/cm3)

% U content

Melting point ( C)

U

19.13

100

1128

UN

14.32

71

2650

UC

13.63

69

2350

U3Si2

12.20

59

1665

UO2

10.96

53

2750

UAl2

8.26

35

1590

After Evans, P.E., Davies, T.J., 1963. Uranium nitrides. J. Nucl. Mater. 10 (1), 43 55.

Alternative fuels to urania Chapter | 7

169

uranium nitride as ATF because of its high thermal conductivity, and high melting point (NEA, 2018; Lawrence Bright et al., 2019). The high density of uranium in the UN fuel (Table 7.2) could be beneficial for operating the fuel in the reactor for longer cycles in between refueling. However, the higher density of the theoretical UN fuel can be offset by the needed smear to accommodate the higher UN swelling under irradiation and to compensate the higher neutron absorption cross section of nitrogen (1.75 barns) as compare to other uranium companion elements such as carbon (0.0038 barns) or oxygen (0.4 barns).

Reactivity of uranium mononitride fuel It is important to understand the reactivity of the fuel in presence of the coolant (water at 300 C) in the case of a breach in the cladding housing. Regarding their stability in the presence of water, uranium nitride fuels are less reactive than uranium carbides (Sugihara and Imoto, 1969). However, UN may be still oxidized rapidly by water above 200 C (Sugihara and Imoto, 1969; Paljevi´c and Despotovi´c, 1975; NEA, 2018) forming explosive hydrogen gas according to two likely reactions: UN 1 2 H2 O-UO2 1 NH3 1 1/2 H2 3 UN 1 2 H2 O-UO2 1 U2 N3 1 2 H2 One of the ways to control the high reactivity of uranium nitride with water is to dope it with zirconium oxide (Zakova and Wallenius, 2012). It has also been mentioned that the reactivity of UN with water can be controlled by reducing porosity and by lessening the content of carbon in the produced fuel (NEA, 2018).

Fabrication paths for uranium mononitride A cost-effective process for the fabrication of UN fuel has not been optimized yet (NEA, 2018). The following several processes have been used in the past to prepare UN (Evans and Davies, 1963): (1) by reacting nitrogen or ammonia with uranium metal, (2) by reacting higher nitrides with either hydrogen or high pressure nitrogen and uranium metal, and (3) by nitriding urania and carbon. This last method called the carbothermic reduction of urania in a nitrogen atmosphere seems the most promising for a large production of uranium mononitride fuel (NEA, 2018). However, the production method still needs to be optimized since the produced fuel quality may not be consistent due to residual carbon and oxygen contents in some of the raw materials and in the processing environment.

170

Accident-Tolerant Materials for Light Water Reactor Fuels

Behavior of uranium mononitride under irradiation Uranium nitride fuels may experience a high swelling rate under irradiation, which could be a concern for pellet cladding mechanical interaction (NEA, 2018). The swelling of UN is a function of the burnup experienced by the fuel. There is no experience about UN irradiation in a light water reactor environment (NEA, 2018). Therefore the behavior of UN under AOOs and design-based accidents (DBAs) can only be inferred, but it is expected that the UN high swelling rate, high Young’s modulus, and low creep rate could be detrimental properties (NEA, 2018). At temperatures higher than 2000K, the UN fuel may decompose releasing nitrogen in the fuel space inside the cladding, which may overpressurize the cladding leading to a possible breach. Once the breach occurs, the UN will rapidly react with steam releasing ignitable hydrogen gas (NEA, 2018).

Chapter 8

Maturity of the accidenttolerant fuel concepts: the fuel cycle and used fuel disposition Chapter outline Overview Assessment on accident-tolerant fuel maturity concepts NEA assessment on maturity of ATF cladding concepts

171 172

Assessment on maturity of ATF fuel concepts The nuclear fuel cycle

180 185

174

Overview This chapter presents an educated, yet subjective maturity evaluation of the five accident-tolerant fuel (ATF) concepts discussed in Chapters 4 7. The five evaluated fuel concepts include three cladding concepts (chromiumcoated zirconium alloy, FeCrAl alloys, and silicon carbide composites) and two fuel concepts including modified urania fuel and high-density fuels. The rating of the maturity level of the concepts has been conducted using metrics from technical readiness level principles by a panel of international nuclear materials experts, mainly from the nuclear energy agency. Not surprisingly the highest maturity level corresponded to the two evolutionary concepts which may have a better chance to obtain near-term licensing from the regulatory entities. These two nearest term concepts include chromium-coated zirconium alloy for cladding and chromia-doped urania for fuel. The second nearest term cladding concept was the monolithic iron chromium aluminum FeCrAl alloy. The other two concepts, namely silicon carbide for cladding and higher uranium density for fuel, have a much longer term possibility of implementation in a commercial power reactor owing to the many technical issues that still need to be solved. A positive twist is that a ceramic silicon carbide design can be first implemented for channel boxes in BWRs as a near- to mid-term ATF concept. This will allow for data gathering in an actual reactor environment, which may accelerate the subsequent implementation of SiC/SiC CMC as cladding for the fuel. Accident-Tolerant Materials for Light Water Reactor Fuels. DOI: https://doi.org/10.1016/B978-0-12-817503-3.00008-0 © 2020 Elsevier Inc. All rights reserved.

171

172

Accident-Tolerant Materials for Light Water Reactor Fuels

Assessment on accident-tolerant fuel maturity concepts Chapters 4 7 have a detailed description of five accident-tolerant fuel (ATF) concepts, three types of ATF claddings, and two types of proposed fuels. The claddings were organized in function of their technical readiness and likely implementation schedule, being (1) coated zirconium alloys (Chapter 4: Accident-tolerant fuels cladding concept: coatings for zirconium alloys), (2) monolithic FeCrAl alloy (Chapter 5: FeCrAl—iron chromium aluminum monolithic alloys), and (3) SiC/SiC CMC composite material (Chapter 6: Silicon carbide and ceramics metal composite). The two basic ATF concepts (doped urania and higher density fuels) are in Chapter 7, Alternative fuels to urania. A condensed evaluation on the maturity of the five ATF concepts (three claddings and two fuel concepts) is presented in this chapter. Most of the data cited here can be referenced to the document prepared by the Expert Group on Accident-Tolerant Fuels for Light Water Reactors (EGATFL) of the Nuclear Energy Agency (NEA) (NEA, 2018). The assessment has also been evaluated based on guidelines provided by the US Department of Energy (DOE) in their Technology Readiness Assessment (TRA) (Carmack et al., 2017). The technology readiness level (TRL) can be grouped into nine categories (Table 8.1) (Carmack et al., 2017; NEA, 2018). The three stages that were followed to develop the TRL lists include (1) identification of the critical technology elements, (2) assessment of the TRL for each technology element, and (3) development of a technology maturation plan for each element based on the TRL. The methodical following of these steps facilitates the estimation of the overall effort associated with the development as well as allows for prioritization for competing technologies (Carmack et al., 2017). The development of nuclear materials is advanced following an iterative process of sequential activities. Each TRL level has its own requirements—that is, completing requirements related to TRL 3 does not automatically make the process a TRL 4. ATF, Accident-tolerant fuel; ATR, Advanced test reactor; PCMI, Pellet cladding mechanical interaction; TRL, Technology readiness level. One of the outcomes of the NEA EGATFL document was the ranking and rating of the several ATF concepts based on their technical and industrial maturity as well as the technical readiness level (TRL) of the concepts (NEA, 2018). The need to evaluate and rank the ATF concepts using the same set of rules or criteria was presented before by Bragg-Sitton et al. (2014, 2016) under the designation of metrics or attributes. The NEA (2018) prepared attribute qualitative charts to compare the maturity level of several ATF concepts, including cladding and fuel using a set of colored fuel rods describing knowledge categories (Table 8.2): The ranking of the ATF concepts was performed for several cladding and fuel concepts regarding (1) their fabrication and manufacturability including economics, (2) their performance under normal operation conditions and AOOs, (3) their performance

Maturity of the accident-tolerant fuel concepts Chapter | 8

173

TABLE 8.1 Technology readiness level (NEA, 2018). Technology readiness level

Relative level of development

Definition and description regarding accident-tolerant fuels

TRL 1

Basic principles observed and reported

Basic principles observed; for example, FeCrAl and SiC are highly resistant to attack by steam; scientific research begins to be translated into applied R&D.

TRL 2

Potential applications validated

Speculative applications are proposed; experimental applied work is designed to complement the scientific observations.

TRL 3

Proof-of-concept, research for feasibility

Active R&D initiated, laboratory testing to measure needed parameters for analytical work. Different components in the system are validated (e.g., fuel and cladding).

TRL 4

Technology development, fuel design parameters

Basic technological components are integrated. Is the fuel compatible with the cladding? This is the first step for the bridge between scientific research and engineering.

TRL 5

Between technology development and demonstration

Near prototypical laboratory-scale system testing in simulated environments such as out-of-pile normal reactor operation (autoclave) and accident conditions (steam). Demonstration of full rod fabrication capability.

TRL 6

Technology of fuel safety basis demonstration

Prototype rodlet testing in relevant environments, i.e., exposure of rodlets to research reactors (e.g., ATR) to determine coolant compatibility and PCMI.

TRL 7

Field scale demonstration for ATF assemblies

Demonstration of a prototype in an actual field environment such as insertion of segmented rods in commercial power plants.

TRL 8

Full core reactor conversion with ATF

Final design complete, fueled rods demonstrated to work under all reactor expected conditions.

TRL 9

Reactor operation with licensed fuel

The technology is in its final form, multiple reactors with ATF concepts in routine operation.

174

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 8.2 Attributes for cladding and fuels (NEA, 2018). White color

Lack of knowledge does not allow for color identification, property not addressed.

Green color

The concept is mature, data available, and the results are positive.

Yellow color

Data are available, but the results are not fully positive, optimization may be needed.

Blue color

There are no data, but the lack of data does not appear challenging.

Purple color

There are no data, and this lack of data may be troubling.

Gray color

A potential show stopper has been identified.

under accident conditions such as design basis accidents (DBA) and beyond design basis accidents (BDBA), and finally (4) their likely properties and performance at the end of the fuel cycle such as the likelihood of the repurposing of the used fuel. The assessment by the NEA EGATFL was comprehensive addressing as many combinations and permutations brought up by the country members of NEA. In this chapter a simplified and condensed version of the NEA rankings is presented.

NEA assessment on maturity of ATF cladding concepts The assessment of the maturity of the ATF concepts was based on the knowledge of the members of the EGATFL (NEA, 2018). A slightly different ranking may have been produced if other technologists were members of the panel. Therefore, the evaluation of maturity is a snapshot in time and this evaluation is subjective and it was meant to evolve. There is no absolute quantitative measure of maturity for the proposed accident-tolerant fuels (Carmack et al., 2017). During the NEA assessment most technologies could not avoid comparing the ATF-proposed concepts with the well-known TRL 9 fuel rod concept, which is the pair urania fuel 1 zirconium alloy cladding, currently in safe use in over 400 civilian power reactors all over the globe (Carmack et al., 2017). In the cladding assessment three options are ranked and classified here based on the original efforts from NEA (2018). The three concepts include: (1) chromium-coated zirconium alloys, (2) monolithic FeCrAl, and (3) SiC/ SiC ceramic composites. The assessments were made for attributes such as manufacturing, reactor performance, and end of fuel cycle considerations. Table 8.3 shows a comparison between the three cladding concepts regarding

Maturity of the accident-tolerant fuel concepts Chapter | 8

175

TABLE 8.3 Maturity level for cladding concepts regarding fabrication and manufacturability. Attribute

Cr-coated zirconium

Monolithic FeCrAl

SiC/SiC composites

Compatibility with large-scale production needs Compatibility with quality and uniformity standards Cost

Impact on the industrial network (suppliers and subcontractors)

fabrication, overall cost, and details such as the qualification of the product and associated processes, the security of the supply chain for all materials for the cladding, likelihood of making actual size (length: B5 m) tubes, ease of producing hermetic sealing or welding, level of effort during fuel assembly manufacturing, quality control procedures, inspection, quality assurance program, and anticipated overall reject ratio of components. According to the attributes in Table 8.3 and from the point of view of manufacturing, the most advanced concept is the monolithic FeCrAl, followed by the coated zirconium. The SiC/SiC cladding concept is the one that needs more discovery and development, especially regarding reproducibility of manufacturing processes and consistent quality issues. A SiC/SiC cladding may need an embedded metallic liner (sandwich design) to avoid the diffusion of toxic elements into the coolant. Table 8.4 shows a condensed version of the three ATF cladding concepts based on the document prepared and published by the NEA (2018). The table shows a comparison between the three cladding concepts regarding their known and predicted behavior during normal operation conditions of light water reactors. This is a rough approximation because the reactor operation conditions changes from plant to plant. Under the attribute Reactor Operation, the following issues were considered in the assessment of the maturity of the concepts (NEA, 2018): (1) performance during normal operation and in situations of anticipated operation occurrences (AOOs), (2) the length of the useful fuel cycle (e.g., 18 or 24 months), (3) interaction of the cladding with the reactivity control systems, (4) likely behavior of leakers after irradiation and ability to contain fuel fragments inside the cladding,

176

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 8.4 Maturity level of three cladding concepts for normal operation conditions and anticipated operation occurrences. Attribute

Cr-coated zirconium

Monolithic FeCrAl

SiC/SiC composites

Reactor operation

Mechanical properties Thermal behavior

Fuel/cladding interaction Irradiation behavior

Coolant interaction

Licensibility

(5) how fabrication defects may impact fuel rod performance, and (6) neutron penalty due to parasitic absorption. Under the attribute Mechanical Properties, the following topics were evaluated: (1) strength, ductility and toughness of the cladding material, (2) creep behavior as a function of stress and temperature, (3) fatigue performance, (4) debris and fretting wear resistance, (5) circumferential buckling, rod and assembly bow, (6) ramping behavior, and (7) coating adhesion (for the coated concept). In the Thermal Behavior attribute the following issues were considered: (1) thermal conductivity, (2) specific heat, and (3) melting temperature. For Fuel to Cladding Interaction attribute, the topics evaluated included (1) compatibility or lack of chemical reactivity between the fuel and the cladding, (2) resistance to pellet cladding mechanical interaction (PCMI), (3) resistance to damage of cladding ID from the presence of fission products such as cadmium or iodine, (4) permeation of tritium from the fuel cavity to the coolant across the cladding wall, and (5) type and size of a possible perforation damage for leaking of fissile products. For the attribute on Irradiation Behavior, subjects considered included (1) effect of irradiation on geometrical stability of the

Maturity of the accident-tolerant fuel concepts Chapter | 8

177

cladding, including swelling and growth, (2) embrittlement of the cladding during irradiation, (3) what is the limit of the irradiation dose, and (4) irradiation-influenced microstructural and chemical composition changes. For the attribute cladding coolant interaction, the following issues were assessed: (1) chemical compatibility between the coolant and the cladding, such as oxide formation and depth of recess of the cladding thickness, (2) impact of cladding dissolved species on the coolant chemistry (e.g., Mo, Al, and SiO2 in the water), (3) hydriding and effect of hydride formation on the mechanical stability of the cladding, (4) shadow corrosion, cladding compatibility with spacer grids and control rods, (5) crud formation on the cladding OD, and (6) thermal hydraulic interaction of cladding with the coolant. For the attribute on Licensibility, matters considered included (1) repeatability and reproducibility of the experimental data to support licensing, (2) ability of existing codes and models to simulate and predict cladding behavior, and (3) how removed is the evaluation methodology compared with considerations for the current zirconium alloys behavior. Table 8.4 shows that from the point of view of normal reactor operation and AOOs, the most advanced cladding concept is the chromium-coated zirconium alloy, because it is the concept that most resemble the current technology. The three US-based fuel producers are considering coated zirconium alloy as the nearest term and most likely licensable concept as a first step into ATF implementation (Hayes, 2019) (Chapter 4: Accident-tolerant fuels cladding concept: coatings for zirconium alloys). The coating to be applied on the zirconium alloy is generally assumed to be less than 10-µm thick and therefore the coating would not alter or modify substantially the current monolithic zirconium alloy concept. What is still unknown for the coated zirconium concept is the susceptibility to crud deposition and the critical heat flux. For normal operation and AOOs, the SiC/ SiC cladding concept is the one that has the largest knowledge gaps and therefore the fuel vendors are pursuing it as a longer term initiative (Chapter 6: Silicon carbide and ceramics metal composite). The two current most challenging issues for the SiC/SiC concept are (1) its hydrothermal corrosion or dissolution of the SiC composite in the coolant and (2) the difficulty to manufacture a hermetic sealing of the cladding that would stand thermal cycles and power ramps for a multiyear performance between fuel reloads. Table 8.5 shows a condensed visually descriptive version of what was originally prepared and published by the NEA (2018). The table shows a comparison between the three cladding concepts regarding their known and predicted behavior during accident conditions of light water reactors. Accident conditions may include DBA, BDBA or Severe Accidents (temperatures higher than 1200 C). To assess the properties in Table 8.5, many topics were evaluated. For the Reactor Operation attribute, the issues considered included (1) performance during accidental transients, (2) impact of

178

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 8.5 Maturity level for claddings under accident conditions. Attribute

Cr-coated zirconium

Monolithic FeCrAl

SiC/SiC composites

Reactor operation

High-temperature mechanical properties Thermal behavior, including quenching and melting Fuel/cladding interaction

Loss of coolant accident and reactivity insertion accident

core loading, which depends on the type of cladding, and (3) behavior of leakers that existed during normal operation or prior to the accident, which would perform worse during a DBA. For the attribute of High-temperature Mechanical Properties, the following issues were evaluated: (1) thermal creep; (2) elongation or ductility; and (3) toughness, strength, and fatigue resistance. For the attribute of Thermal Behavior, the items evaluated included: thermal conductivity, specific heat, and melting temperature. For the attribute on Fuel/cladding Interaction, the issues considered included (1) chemical reactivity between the fuel and the cladding, or stability of the fuel and cladding, (2) fission gas generation and fission products behavior, (3) susceptibility to pellet cladding interaction, and (4) in the case of a breach, reaction between the environment and the ID of the cladding. For the attribute on Loss of Coolant Accident (LOCA) and reactivity insertion accident (RIA), the topics evaluated included (1) resistance of the cladding to steam attack at temperatures in the order of 1200 C 1500 C and likelihood of breakaway oxidation, (2) resistance of cladding to quenching and post-quench mechanical properties, (3) degradation mode of the cladding, including phase transformation of the non-oxidized matrix, (4) mechanical performance such as PCMI under high speed ramp, (5) hydrogen gas production, and (6) interaction of cladding with molten material. The analysis shown in Table 8.5 shows the good behavior and state-ofthe-art knowledge of the chromium-coated zirconium alloys as a nearest term ATF concept solution. The major handicap of the chromium-coated

Maturity of the accident-tolerant fuel concepts Chapter | 8

179

concept is that the maximum temperature during LOCA may be limited to 1000 C because eventually the thin layer (10 µm) of the chromium coating will react with the steam and volatilize, exposing the underlying zirconium alloy to the steam. It is known that FeCrAl and SiC/SiC have at least a 1000-times lower reactivity with steam than current zirconium alloys at temperatures higher than 1000 C (Rebak, 2015; Terrani, 2018). However, there are other unknowns for the FeCrAl and the SiC/SiC concepts because these materials were never used inside a light water reactor like the Zircaloy alloys were consistently used for over 60 years. The comparative behavior between FeCrAl and SiC/SiC in accident conditions maybe equivalent at the current time, considering the current data until more data are available for in-reactor conditions and likely RIA tests at the TREAT reactor in the Idaho National Laboratory. Table 8.6 shows a list of attributes to be evaluated if the entire fuel cycle is considered. Each attribute in the left column has several issues that need to be assessed. For example, considering Transportation, the topics analyzed include mechanical behavior and ductility during transportation, which includes drop and punch tests, and the ability to withstand a fire. All likely fabrication defects need to be evaluated for transportation requirements. For the attribute on Long-term storage, issues that need to be considered include (1) impact of corrosion suffered during reactor performance, including hydride (if any) reorientation, (2) impact of possible fabrication imperfections such as coating defects, (3) effect of residual radioactivity, and

TABLE 8.6 Maturity level for fuel cycle issues of three cladding concepts. Attribute U-235 enrichment limit Transportation

Long-term used fuel storage Reprocessing

Cost

Cr-coated zirconium

Monolithic FeCrAl

SiC/SiC composites

180

Accident-Tolerant Materials for Light Water Reactor Fuels

(4) long-term microstructural and chemical composition evolution. For the attribute on Reprocessing the topics that need to be evaluated include (1) tritium accumulation, (2) behavior during crushing and shearing of the used fuel, and (3) chemical reactivity with reprocessing media such as nitric or hydrofluoric acid. Many of the issues considered in the evaluation were a result of the knowledge that the NEA technologists had and that it was centered in the six-decades use of zirconium alloys. The analysis of the attributes in Table 8.6 shows many white areas for which data are not available. The easiest concept to evaluate is the FeCrAl monolithic cladding because it is the simplest concept. The raw materials for FeCrAl are inexpensive and the only cost associated with making the rods is that for the manufacturing of the tube for the fuel rods. Iron, chromium, and aluminum elements are well known regarding their compatibility with reprocessing steps, such as reactivity with dissolving acids such as nitric. For the long-term storage, FeCrAl cladding should be stable because none of these elements react with hydrogen to produce stable brittle hydrides like the zirconium element does. Also, hydrogen is unlikely to accumulate in the lattice of FeCrAl because these alloys are ferritic or bcc structure, allowing for fast easy diffusion of the hydrogen across the cladding wall. From the viewpoint of plant operation, the FeCrAl have an extra cost associated to the neutron penalty owing to the approximately ten-times higher thermal neutron absorption of FeCrAl when compared with that of the zirconium alloys. Chromium-coated zirconium alloys may have unknowns at the end of fuel cycle because the adhesion of the coating as a function of time is not known. Also, how the coating may affect hydrogen diffusion and hydride behavior in the zirconium alloy substrate is not known. Maturity unknowns for the fuel cycle are mainly for the longer term SiC/SiC concept. Also, during reprocessing, SiC may have the issue of being insoluble in acids.

Assessment on maturity of ATF fuel concepts The current assessment on the maturity of the ATF concepts is based on the document published by the NEA in 2018. Only three fuels concept are evaluated: (1) chromia-doped urania, (2) uranium silicide U3Si2, and (3) uranium nitride (UN). The maturity concepts are represented in four tables in the form of fuel pellets in the same color scale used before to demonstrate knowledge or knowledge gaps for the cladding. Table 8.7 shows four attributes by which the three ATF concepts were evaluated and compared regarding their industrial fabrication ability (NEA, 2018). The four attributes are (1) compatibility with large-scale production needs for the fuel, (2) compatibility with quality and uniformity standards during production, (3) cost, and (4) impact on the industrial network—Can suppliers and subcontractors be allocated? For the attribute on Compatibility with large-scale production, the following issues were considered:

Maturity of the accident-tolerant fuel concepts Chapter | 8

181

TABLE 8.7 Maturity level for three fuel concepts regarding fabrication and manufacturability. Attribute

Chromiadoped urania

Uranium silicide (U3Si2)

Uranium nitride (UN)

Compatibility with large-scale production needs Compatibility with quality and uniformity standards Cost

Impact on the industrial network (suppliers and subcontractors)

(1) qualification of the product and the accompanying process, (2) market availability of each material component of the fuel, (3) technical database and streamlined manufacturing criteria, (4) large-scale pellet fabrication, (5) welding or hermetic sealing of the fuel cavity, and (6) fuel assembly manufacturing. For the attribute on Compatibility with quality and uniformity standards the following topics were evaluated: (1) quality control, ability to inspect, and (2) reject ratio, which may impact cost. Table 8.7 shows that the nearest term development regarding ATF is the chromia-doped urania, for which most of the attributes have currently a clear resolution in the fabrication path. The higher density fuels (uranium silicide and uranium nitride) have only been fabricated in the laboratory and a clear industrial manufacturing process is not currently available. In some cases, optimization of a process route is required but in other cases there are altogether no available fabrication process data. Some specifications have been developed for uranium silicide fuels, and it seems that the production costs of the silicides can be controlled. Table 8.8 shows the attributes of the three ATFs regarding their performance in the reactor under normal operation conditions and AOOs (NEA, 2018). The attributes evaluated for the three type of fuels were similar to those used for the cladding in the previous pages, and included (1) reactor operation, (2) mechanical properties of the fuel, (3) thermal behavior of the fuel, (4) fuel cladding interaction, (5) irradiation effects on the fuel, (6) compatibility between the fuel with the coolant, and (7) licensibility considerations. For the attribute on Reactor Operation, some of the associated

182

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 8.8 Maturity level for three fuel concepts regarding normal operation conditions and anticipated operation occurrences. Attribute

Chromiadoped urania

Uranium silicide (U3Si2)

Uranium nitride (UN)

Reactor operation

Mechanical properties of the fuel Thermal behavior of the fuel Fuel cladding interaction

Irradiation effects on the fuel Compatibility of the fuel with the coolant Licensibility considerations

issues included (1) impact of the reactor load related to the overall cladding properties and performance, (2) actual behavior of leakers during irradiation and the probability of fuel fragments dispersion in the coolant, (3) effect of specific fabrication defects on the pellet, and (4) neutron penalty. For the attribute on Mechanical Properties, the issues of thermal expansion and creep were mostly considered. For the attribute on Thermal Behavior, the following topics were evaluated: (1) thermal conductivity, (2) specific heat, and (3) melting temperature for the fuel. For the attribute on fuel/cladding interaction the issues considered were (1) chemical reactivity between fuel and cladding, (2) resistance to PCMI, and (3) resistance to environmentalassisted cracking due to fission products such as iodine or cadmium. For the attribute on Irradiation Effects, the topics evaluated were (1) irradiation limit, (2) fission product behavior, (3) embrittlement, (4) irradiation-induced microstructural and/or chemical composition changes, and (5) dimensional stability, including growth and swelling. For the attribute on Compatibility between the Fuel and the Coolant, the issues considered included

Maturity of the accident-tolerant fuel concepts Chapter | 8

183

(1) chemical compatibility with the coolant, (2) impact of fuel corrosion on the coolant chemistry, and (3) resistance to oxidation of the fuel. For the attribute on Licensibility, the issues evaluated included (1) capability of existing models and codes to simulate the behavior of the fuel, (2) the repeatability and reproducibility of the experimental data to support licensing but the regulatory body, and (3) ability to use current processes and methods. Table 8.8 shows the highest viability of the chromia-doped urania fuel when compared with the higher density fuels. The most problematic issues with the higher density fuels are their high reactivity with the coolant in case of a breach in the cladding. During the reaction with water and steam, the nitride fuel is expected to generate hydrogen while suffering crumbling or fragmentation. During normal operation conditions, there could be issues related to the swelling, and low irradiation creep of the nitride fuel. There is no experience of UN fuels in light water reactor conditions (NEA, 2018). For the silicide fuel, there are irradiation studies currently being conducted in the Idaho advanced test reactor (ATR). During AOOs, the nitride fuel could be responsible for PCMI owing to its high Young’s modulus and its lower creep rate. On the other hand, the silicide fuel is expected to cause less PCMI effect than the current urania fuel (NEA, 2018). The use of both silicide and nitride fuel may increase the fuel cycle compared with UO2, mainly because of their higher content in U-235. Both silicide and nitride fuels do have higher thermal conductivity than urania, which is a positive outcome to transfer the heat faster to the coolant through the cladding wall. The high thermal conductivity of the silicide is especially important because the silicide has a lower melting point compared with the nitride and the urania fuels. Reactivity between the silicide fuel and cladding elements such as zirconium, chromium, and iron has also been expected (NEA, 2018). But the silicide would have a good compatibility with the silicon carbide cladding regarding both reactivity and PCMI. The effect of typical light water reactor conditions irradiation on swelling of either nitride or silicide is not known at this time. Table 8.9 shows that little is known in the realm of accident effects on the three proposed ATF concepts. Again, the chromia-doped urania fuel seems the most advanced concept for which the highest confidence exists on its probably behavior under accidents conditions. The higher density fuels need more research for behavior under accident conditions; therefore, these fuels can be labeled as longer term consideration when compared with the advanced or doping-modified urania fuel. The upper temperature limit of the nitride fuel maybe lower than for the current urania fuel, not because of melting but because the nitride fuel may dissociate into the metal and nitrogen in a helium atmosphere when the temperature reaches 1730 C (NEA, 2018). At the same time the nitride swelling during LOCA may contribute to cladding breach. The upper temperature limit of the silicide fuel could also be an issue owing to its relatively low melting point of 1665 C.

184

Accident-Tolerant Materials for Light Water Reactor Fuels

TABLE 8.9 Maturity level for three fuel concepts regarding accident conditions. Attribute

Chromiadoped urania

Uranium silicide (U3Si2)

Uranium nitride (UN)

Reactor operation

Mechanical properties of the fuel at high temperatures Thermal behavior of the fuel

Fuel cladding interaction

LOCA and RIA

Compatibility of the fuel with superheated steam Licensibility considerations

LOCA, Loss of coolant accident; RIA, Reactivity insertion accident.

Table 8.10 shows the maturity status of the fuels, taking into consideration the entire fuel cycle (NEA, 2018). Owing to its lowest departure from the current urania fuel, the doped urania has the least amount of changes; therefore, it has the fewer number of unknowns, showing the fuel as most likely for near-term application or most advanced in its implementation. Both nitride and silicide fuels have a higher density of uranium, which can compensate for higher neutron absorption cross section of N-14 and U3Si2 compared with the current urania fuel. As mentioned before, both the nitride and silicide would have a lower chemical stability than the current urania fuel owing to their higher reactivity with air and water (NEA, 2018). Regarding reprocessing, both nitride and silicide score well because they are soluble in nitric acid. Both high-density fuels may have additional cost compared with urania because of special infrastructure needed for processing and unusual storage conditions.

Maturity of the accident-tolerant fuel concepts Chapter | 8

185

TABLE 8.10 Maturity level for fuel cycle issues of three fuel concepts. Attribute

Chromiadoped urania

Uranium silicide (U3Si2)

Uranium nitride (UN)

U-235 enrichment limit Thermal behavior

Chemical stability, fission product Reprocessing

Cost

The nuclear fuel cycle The nuclear fuel required to generate electricity in nuclear power plants has several stages of its life, assuming a once-through fuel cycle (Fig. 8.1). The fuel is manufactured by filling metallic tubes with urania pellets of about 4-m high; the combination of a tube and the filling of the fuel pellets is called a fuel rod. The fuel rods are then assembled in bundles containing perhaps 100 rods each. For example, in each BWR power reactor there could be in the order of 800 bundles. The fuel bundles are especially designed for each reactor and transported from the fuel vendors to the utilities that operate the reactors. The nuclear fuel sits in the reactor core for approximately six years, undergoing a progressive degradation though a process of nuclear fission. Approximately every two years one-third of the fuel is removed from the reactor and replaced by fresh one. After its life in the reactor the fuel becomes used or spent fuel, and it may be classified as highly radioactive nuclear waste. From the reactor core the fuel is then transferred to cooling pools for approximately 5 20 years to remove most of the decay heat from the used fuel. After their residence in the pools the used fuel may be transferred to dry cask storage for probably near 100 years, where additional heat is removed using natural air convection. At any point the used fuel may be reprocessed to remove usable fissile material and reduce the amount of the actual waste. Eventually, the used or waste fuel containing long-lived radioactive products is intended to be buried in geologic nuclear waste

186

Accident-Tolerant Materials for Light Water Reactor Fuels

FIGURE 8.1 Life cycle of fuel bundle.

repositories for probably many thousands of years. All the countries with nuclear power are planning to build a geologic repository for their waste (Rebak, 2011). In Fig. 8.1, it is shown that the time of residence in the reactor of a fuel bundle is the shortest time of their entire life. After its brilliant performance in the reactor the used fuel is transported to water pools and subsequently to dry cask storage to finally end up in a geologic repository (Rebak et al., 2009). It is necessary to demonstrate that the specific ATF concept does not only survive the environments of the reactor (normal operation conditions, and unlikely accident conditions) but also there is a requirement for them to survive the residence in the pools and in the dry cask.

Chapter 9

Licensing and the increased safety of power reactors’ operation Chapter Outline Overview Licensing process in the United States Increased safety of nuclear power plant operation

187 188

Evolutionary trend of the nuclear fuel Safety analysis and source term

188 190

188

Overview The implementation of accident tolerant fuel concepts will increase the safety of nuclear power plant operation. This increased safety will occur not only because of the actual performance of the material concepts under normal and accident conditions but also because of the full evaluation of plant operation. This assessment can be accomplished by considering the concept of source term, which is defined as the type and number of radioactive contaminants that can be released for a nuclear power station owing to an accident. The analyses of several areas of improvements by use of ATF concepts in the reactor showed a decrease in the source term when compared with the current fuel designs. The range of materials that are being proposed to ensure safe operation of reactors has been discussed in the previous chapters. Before these materials can be used in the civilian generation of electricity, they must be approved by the regulatory entities such as the Nuclear Regulatory Commission (NRC) in the United States. Materials such as the coated zirconium alloys for fuels and channels, as well as the chromia-doped urania fuels may get a fast-track regulatory approval than concepts that are removed from the presently available reactor materials. For example, the use of SiC/ SiC ceramic matrix composite for fuel cladding may need longer documentation and testing for obtaining final approval.

Accident-Tolerant Materials for Light Water Reactor Fuels. DOI: https://doi.org/10.1016/B978-0-12-817503-3.00009-2 © 2020 Elsevier Inc. All rights reserved.

187

188

Accident-Tolerant Materials for Light Water Reactor Fuels

Licensing process in the United States Operating nuclear power stations in the United States require a license from the NRC, which is described in the Title 10 of the Code Federal Regulations (10 CFR). The process has two parts: (1) building a reactor and (2) operating a reactor. For the subject of this book of accident tolerant fuel (ATF) safer materials for current light-water reactors, the process will imply a refurbishing or retrofit of the presently available materials, under the topic of reactor operation.

Increased safety of nuclear power plant operation As mentioned in the previous chapters, the civilian nuclear power generation has been relying on practically the same structural materials for the fuel rods and other reactor components since the late 1950s. The unfortunate events of North East Japan involving explosions of accumulated hydrogen gas in the containment buildings of four reactors made the international nuclear materials community reevaluate the use of zirconium as a cladding material for the urania fuel. Zirconium alloys served the nuclear industry well for six decades because engineers understood the behavior of zirconium and were able to manage and resolve real-world failure modes over time. Engineers also knew that under loss of coolant accident (LOCA) conditions, zirconium alloys would not deliver a stellar performance because they undergo an autocatalytic exothermic reaction with steam, which may generate a huge amount of combustible hydrogen gas. The explosions at the Fukushima Daiichi nuclear power stations was a justification and an inflexion point for the engineers to pursue materials that would be less prone to undergo the catastrophic failures experienced by the Fukushima reactors.

Evolutionary trend of the nuclear fuel From the mid-1950s, the pair of uranium dioxide fuel (urania or UO2) and zirconium-based alloys cladding have been identified as reliable materials to build fuel rods for light-water power reactors. For the last 60 years, the fuel system has been witnessing an evolutionary innovation (including small improvements in alloy composition) or new fabrication methods, but the fuel rod pair zirconium/urania remained unchanged (Fig. 9.1). In 2019, the international nuclear fuel community is looking for a step change in the type of fuels under consideration. This step change in nuclear fuel may help nuclear power generation to be reimagined for the public to accept and support nuclear energy as a clean form of producing electricity. Fig. 9.1 shows that for the ATF concept fuels to be successful as a step or revolutionary change, it would

Licensing and the increased safety of power reactors’ operation Chapter | 9

189

FIGURE 9.1 Step change.

need to have a lower cost and to provide an increase in plant operation safety. One way of reducing fuel cost is to increase enrichment of U-235 and increase residence in the reactor (increase burn up to probably 75 GWd/MTU). The ATF would also need to minimize reliability events such as fuel rod leaking due to debris fretting, crud deposition, and channel bowing because of hydrogen entrance. Like in many other industries, changes and innovations for ATF concepts to be accepted by the utilities would be a typical balance between cost and benefits. Costs include (1) research and development to mature the ATF concept, (2) licensing, and (3) utility implementation. Benefits include (1) fuel cycle economics (i.e., increased burn up), (2) increased fuel reliability, and (3) plant operational flexibility. The National Nuclear Laboratory (NNL, 2018) from the UK published an ATF position paper where a comparison on the reactor performance and economic feasibility table was discussed for ATF claddings and fuels compared with the current zirconium alloys and urania fuel couple. Most of the improvements for both the cladding and the ATF concept fuels were made appropriately for severe accident situations. In the claddings area, the one concept that has the fewest detrimental attributes is the coated zirconium alloy, with the weakest point in its manufacturability. For the FeCrAl cladding concept the most significant handicap is the parasitic neutron absorption. For the SiC/SiC concept the most negative attributes were listed as the manufacturability and the hydrothermal corrosion under normal operation conditions (NNL, 2018). For the ATF concepts, the modified UO2 offered fewer challenges. The other two fuels (UN and U3Si2) may have unacceptable reactivity with water and difficulties in industrial-scale manufacturability, which may render them economically unattractive.

190

Accident-Tolerant Materials for Light Water Reactor Fuels

Safety analysis and source term The Electric Power Research Institute, EPRI (2019) published a report containing the opinion of the nuclear industry in the United States, mainly regarding economics and safety. The experts argued that in the case of an accident, the use of ATF concepts would increase the coping time by 1 3 h, which could be relevant for an accident path similar to that of Fukushima Daiichi stations. Under normal operation conditions, the ATF concepts may increase nuclear plant safety while potentially reducing operational costs. Moreover, ATF concepts in replacement of the current Zr/UO2 performance may boost plant efficiency, for example, (1) by enhancing core design capabilities, (2) by allowing for maneuverability flexibility, (3) by increasing the allowable burn up of the fuel rods, and (4) by reducing outage times. The EPRI team (2019) also visited five United States nuclear power stations [both boiling-water reactor (BWR) and pressurized water reactor (PWR)] and assessed the viewpoint of each plant regarding foreseeable benefits on the implementation of prevailing ATF concepts. The seven areas where improvements can be reached with ATF concepts were listed as (1) flexible power operations, (2) probabilistic risk assessment (PRA), (3) fuel cycle economics, (4) crud deposition, (5) design-basis accidents, (6) source term improvements, and (7) equipment classification, and life extension. Moreover, each of the seven benefit areas was evaluated under two identifiers: (1) potential savings and (2) improved technical benefits (EPRI, 2019). 1. Flexible power operations: For the last 60 years, it was always accepted that nuclear power plants (NPPs) will provide the base load to the electrical grid, often operating at full load. In recent years, with the access to renewable electricity sources, it was suggested that if NPP would be able to operate with a flexible load, providing the energy that the grid may require at different times of the day or the week, or in different seasons. It is envisioned that NPP may be more amenable to flexible power operations if ATF concepts are instituted because the ATF concepts are more robust and have resilient operation changes than the traditional fuel forms (EPRI, 2019). 2. Probabilistic risk assessment: One of the missions of the United States NRC is to guarantee that civilian NPPs operate with minimal risk to public health and safety. This is addressed in 10 CFR (NRC, 2019). For NPP the aim is to reduce the possibility of damaging the fuel in the reactor core and to potentially avoid the release of radioactive elements to the environment, where they may harm plants, animals, and humans. Risk is determined by two factors: (1) how often might a hazard occur and (2) how much harm is likely to result from said hazard (Rebak, 2017). For example, in the case of the Fukushima accident, the root cause was the very tall tsunami that disabled the diesel generators and caused total plant

Licensing and the increased safety of power reactors’ operation Chapter | 9

191

black out. The tsunami had an incredibly low frequency of occurrence (the first in the 40 years of operation of the generating station), but the consequences were catastrophic. The enforcement by the NRC is to use PRA and oversight to reduce the risk of the NPP, that is, by controlling that the event would occur (taller sea walls for the Fukushima case), and by putting a sturdier barrier (such as ATF) between the fuel and the environment. Another way to reduce the consequences in the Fukushima case of dispersing radioactive elements to the environment would be to avoid the hydrogen accumulation in the reactor housing buildings. That is, one of the best strategies to avoid the release of toxic radioactivity to the environment is to use the Defense in Depth evaluation to confirm sufficient layers of defense against both core damage and loss of containment functions. Current evaluations regarding the use of PRA to reduce risks in operation of NPP that would employ ATF concepts seem to be relatively small (B10% 20%) (EPRI, 2019). One of the reasons of the less-thananticipated effect of ATF in minimizing risk is that some of the benefits of ATF cannot be fully assessed and quantified at the present time. The most important effect seems to be the extra coping time for bringing additional power or repairing equipment to mitigate the effects of the accident. 3. Fuel cycle economics: The use of ATF concepts in light-water reactors may open the possibility of reevaluating and optimizing the fuel cycle economics. One of the possibilities is to increase the fuel cycle to 24 months and the other is to extend the discharge burn up from 62 GWd/ MTU to 74 GWd/MTU (EPRI, 2019). These two measures will allow the NPP to use the fuel more efficiently. A longer fuel length will decrease the total fuel cost and the outage costs. Similarly, by extending the fuel burn up, fewer fresh fuel assemblies will be required at each fuel cycle, which will lower the fabrication costs, the cost of storage, and the cost of disposal of the used fuel. One of the utilities surveyed by EPRI (2019) conveyed that currently they may have up to eight plant outages in their fleet in 1 year, which is an enormous drain on staffing and costs associated with reload activities. That is, this utility could benefit from longer fuel cycles and fewer outages, which could require increased burn ups and/or increased fuel enrichments. It is projected that ATF designs may have the potential to offer the needed performance to achieve the desired higher fuel duties. For example, one of the limitations of the current monolithic zirconium alloy cladding is their shorter needed residence to satisfy requirements of maximum allowable corrosion across the wall thickness of the cladding of 17%. This maximum corrosion allowance is also tied to the limit of maximum hydrogen absorbance by the cladding during the corrosion process. For example, if chromium-coated zirconium alloy cladding is

192

Accident-Tolerant Materials for Light Water Reactor Fuels

used as the ATF concept, the amount of total corrosion during normal operation conditions in the reactor will be lower than for monolithic zirconium alloy. Therefore the time for maximum oxidation allowable will be increased with an increase in needed ductility (less hydrogen absorption) in the case of a beyond design-basis accident (BDBA) LOCA scenario. Moreover, studies show that the coating with chromium of the zirconium alloy would minimize clad ballooning and decrease the size of the opening in case that there is clad failure, which will result in a decreased source term dosage. Moreover, if a doped urania fuel is used as a companion fuel in the coated zirconium cladding, this will result into a lower tendency for pellet fragmentation, which will further reduce the dispersion propensity of the pellet into the cavity of the coolant. Another area of improvement in the fuel cycle economics is to modify the criterion for the departure from nucleate boiling (DNB). The NRC defines DNB as the point at which the heat transfer from a fuel rod across the cladding wall rapidly decreases owing to the insulating effect of a steam blanket that forms on the fuel rod’s outer surface. The maximum heat flux immediately before the boiling transition from nucleate to film is called the critical heat flux. EPRI (2019) has assessed the influence of adopting ATF concepts on current criteria regarding DNB and summarized that both PWR and BWR could provide an additional benefit under anticipated operational occurrences and design-basis accidents. The newer criteria of DNB for ATF concepts should consider the higher strength and the higher thermal conductivities of the proposed ATF concepts. This newer criteria for DNB would translate into economical gains by using ATF. 4. Crud deposition: Dissolved and suspended solid corrosion products from structural elements in light-water reactors are called crud. Primarily in boiling-water reactors, a large fraction of crud products may deposit or solidify on the outer surfaces of fuel rods, where they can affect the thermal efficiency of the fuel rods, which could lead to temperature increases and potential cladding breach and fuel rod failure. Likewise, activated crud debris may detach from the locations in which they solidify, producing radioactive contamination in the cooling water system and even in the used fuel storage systems (Orlov et al., 2011). The buildup of crud on fuel rods may create operational issues, constrains on core designs, and potentially be responsible for some of the recent fuel failures in LWR zirconium-alloy-based claddings, such as (a) reduced thermal conductivity of the overall cladding plus the added crud, (b) accelerated clad corrosion with increased hydrogen uptake, (c) crud-induced localized corrosion, (d) crud-induced power shifts, and (e) increased doses in the primary coolant loop (EPRI, 2019). It is anticipated that the use of ATF concepts will reduce the volume of crud deposition on the fuel rods;

Licensing and the increased safety of power reactors’ operation Chapter | 9

193

therefore it will decrease the events of crud-driven power shifts or crudinduced localized corrosion concerns (EPRI, 2019). The lower amount of crud deposition would also minimize or eliminate the need for fuel cleaning; therefore it will reduce outage times, which will translate into cost savings. It is expected that the use of chromium-coated zirconium alloy cladding and monolithic FeCrAl cladding would result in a lower amount crud formation on the surface. However, studies of crud coverage mechanisms are needed to determine the prospective betterment of ATF concepts over current bare zirconium alloys, both in out-of-pile laboratory autoclaves and through pool-side inspections of the segmented rods being irradiated by fuel vendors at commercial power stations. 5. Design-basis accidents: During the power reactor useful life, there may be events that can be divided in two large groups, namely (a) normal operation conditions and (b) out-of-normal operation conditions, which would include situations such as anticipated operational occurrences (AOOs), design-basis accidents (DBAs), and BDBAs. According to the Department of Energy, AOO is an abnormal event that is expected to occur during the lifetime of the facility, such as small radioactive materials spills or small fires. According to the NRC, design-basis accident is a postulated accident that a nuclear facility must be designed and built to withstand without loss to the systems, structures, and components necessary to ensure public health and safety. NRC defines BDBAs as a technical way to discuss accident sequences that are possible but were not fully considered in the design process because they were judged to be too unlikely. In that sense, they are considered beyond the scope of design-basis accidents that a nuclear facility must be designed and built to withstand. Under DBA conditions the use of ATF concepts could provide savings by relaxing or eliminating selected reactor trip setpoints, by reducing surveillance, by reducing instrumentation, by decreasing calibration costs, by increasing thermal design limits, and by relaxing the time needed for critical operator actions (EPRI, 2019). The use of ATF concepts will allow the following improvements: (a) Potential replacement of DNB limits with a time-at-temperature limit and (b) improved DBA margins will enable thermal limit relaxation by relaxing emergency core cooling system injection requirements. In some specific ATF concepts, if SiC/ SiC cladding is used, DNB may no longer be a concern because extended time-at-temperature without fuel failure is expected. If a monolithic FeCrAl cladding is used, there is possibility for the replacement of DNB limits with dry-out. The anticipated improved DBA margins would enable thermal limit relaxation by relaxing emergency core cooling system injection (EPRI, 2019). 6. Source term improvements: As mentioned before, source term is used to measure radioactive contamination because of an accident. The implementation of ATF concepts into light-water reactors will result in savings

194

Accident-Tolerant Materials for Light Water Reactor Fuels

owing to broad improvements in plant operation, maintenance, planning, etc. It will also enable the relaxation of operational constraints and procedures established to address accident scenarios (EPRI, 2019). The use of ATF concepts will enable the following improving technologies for specific ATF concepts. For coated cladding, an improved fuel reliability is expected, with fewer leakers and reduced fuel fragmentation and dispersal because of reduced opening gaps when the cladding is breached. For doped pellets and U3Si2, alternate source term methodology improvements would be expected owing to improved fission gas retention (EPRI, 2019). For U3Si2 pellets, there will be a reduced likelihood of cladding rupture, which improves the impact of source term. For the SiC/SiC cladding, the absence of ruptures would reduce radioactive releases. BDBA tolerance would reduce source terms without hydrogen generation. For the monolithic FeCrAl cladding concept an improved fuel reliability is anticipated, with fewer leakers, and reduced fuel fragmentation and dispersal in the coolant because of a reduced opening gap in the case of a high temperature high pressure breach. 7. Equipment classification and life extension: If the plants adopt the ATF concept, it is expected that saving can be accomplished by relaxing equipment qualification (EQ) life requirements and by reclassifying equipment (EPRI, 2019). The use of ATF may provide significant technical improvement in the following areas. For the coated cladding concept, it is expected that the coating will potentially reduce crud adherence, which will result in reduced radioactivity levels. For doped pellets, a potential improvement in fission gas retention is expected, which will theoretically reduce source term and improve postaccident conditions. For the SiC/SiC cladding, an improved fission product barrier would reduce EQ demands. For the monolithic FeCrAl cladding, an improved fission product barrier that will also reduce EQ demands is anticipated. EPRI (2019) published calculations of the comparative reactor core peak temperatures between ATF concepts and the existing zircaloy/UO2 reference materials as a function of time following the sequencing events at the Unit 1 of the Fukushima Daiichi plant. That is, this is an exercise to assess what would have happened if the Fukushima Unit 1 had ATF materials rather than the traditional fuel system. The comparative metric parameter was the time needed under each fuel concept for the temperature in the reactor to reach the melting point of nonirradiated UO2 (3138K). Therefore the difference in time to reach 3138K between the ATF concept and the traditional fuel was the additional coping time for recovering the reactor (EPRI, 2019). It was estimated that the additional coping time for Unit 1 was a little over 1 h for the fuel concept FeCrAl/UO2 (EPRI, 2019). With the traditional fuel system, the core damage of Unit 1 occurred very fast, taking between 4 and 5 h, and if FeCrAl/UO2 had been used it would have taken 6.5 h.

Licensing and the increased safety of power reactors’ operation Chapter | 9

195

To rate the severity of an accident, one of the parameters in which the NRC is interested in is the source term. As defined by the NRC, source term is the types and amounts of radioactive or hazardous material released into the environment following an accident. The inventory of the radioactive materials is a function of the U-235 fuel enrichment in the reactor before the accident, on thermal power of the reactor core and the final burn up at the accident time (EPRI, 2019). Many of the proposed ATF concepts still rely on the use of urania (UO2); therefore the use of ATF would not impact the total amount and the number of radioactive species when compared with the current fuel. Nevertheless, an impact in the source term may be anticipated because the use of ATF may involve the enrichment of the fuel (e.g., from 5% to 6%) and the increase in the total burn up from 62 GWd/MY to B75 MWd/MTU (EPRI, 2019). Evaluations of ATF concepts involving nonurania fuels such as U3Si2 or UN are still more uncertain than the ATF concepts with urania fuel probably involving doping or higher enrichment. Nonetheless, considering the parameter of source term, the use of ATFs may provide additional benefits with respect to the current fuel in the delayed time for the release of radioactive species. This postponement in the release would be a positive impact on the deployment of equipment and measures in the plant related to dose consequences.

Chapter 10

Looking to the future Chapter Outline Overview

197

Overview In the previous nine chapters, a description of the accident-tolerant fuel (ATF) programs around the globe has provided. The description presented in this book was not meant to be comprehensive since the field of the ATF literature is vast and increasing every week. As many as 30 countries may be working in the ATF field of research. The recent increase in activity in the field of nuclear materials may save the near future use of nuclear energy to generate civilian electricity. Little or no innovation in nuclear materials for power plants has happened in the last three decades (since 1990); the industry had a complacent status quo and preferred to go unnoticed rather than to innovate. The tsunami that hit northeast Japan in March 2011 was a catastrophic event that killed approximately 20,000 people on the coast and caused complete station black out at the Fukushima Daiichi site. No human losses were reported as a consequence of the plant building explosions and due to the release of radioactivity to the environment. Nevertheless, the hydrogen explosions witnessed live on TV all around the world put a stain in nuclear power for the uneducated public even though the event was without casualties. Nuclear energy is clean and safe and does not contribute to the release of greenhouse gases to the environment. Nuclear energy can work with a constant delivery of power day and night and does not rely on atmospheric events to generate electricity. The Fukushima disaster was an unfortunate event but at the end, it acted as a catalyst for the international materials’ community to finally get together and try something new. Zirconium alloys have been used successfully for over six decades in civilian nuclear power generation. There was little incentive for the industry to change or innovate, looking for something bolder but safer. It is now accepted that if the nuclear power business does not adapt to changes and embrace innovation, it may

Accident-Tolerant Materials for Light Water Reactor Fuels. DOI: https://doi.org/10.1016/B978-0-12-817503-3.00010-9 © 2020 Elsevier Inc. All rights reserved.

197

198

Accident-Tolerant Materials for Light Water Reactor Fuels

completely disappear, at least in the western world. In the Americas and Europe, there were no newer reactors connected to the grid for decades, and the existing power reactors are slowly shutting down. The ATF materials revolution projects may be the jolt that the industry needed to wake up and defend this clean and safe source of energy. The public seems to accept without questioning the process of burning fossil fuels to generate electricity and its terrible safety record and its contribution to pollution, but the public seems to have an irrational fear of nuclear power, even though historical data show that the latter is orders of magnitude safer for humans and the environment compared to the burning of fossil fuels. For nuclear energy to be sustainable and survive as a source of clean and safe electricity, the public may need to change their perception of its beneficial qualities. The next 10 years will be crucial in determining whether the world will embrace nuclear power or nuclear power will be a curiosity of the past. The ATFs programs around the globe may not only make reactors safer to operate but also be the catalyst to save the entire industry from extinction.

References Adamson, R., 2007. Shadow corrosion. Corrosion Mechanisms in Zirconium Alloys. A. N. T. International, Skultuna, Sweden, pp. 4 12 (October). Adamson, R.B., 2000. Effects of neutron irradiation on microstructure and properties of Zircaloy. In: Sabol, G.P., Moan, G.D. (Eds.), Zirconium in the Nuclear Industry: Twelfth International Symposium, ASTM STP 1354. ASTM International, pp. 15 31. Adamson, R.B., 2010. Zirconium Production and Technology: The Kroll Medal Papers 1975 2010, ASTM Stock RPS2. ASTM International, West Conshohocken, PA, November. Alat, E., Motta, A.T., Comstock, R.J., Partezana, J.M., Wolfe, D.E., 2015. Ceramic coating for corrosion (c3) resistance of nuclear fuel cladding. Surf. Coat. Technol. 281, 133 143. Al-Badairy, H., Tatlock, G.J., 2000. The influence of moisture content of the atmosphere on alumina scale formation and growth during high temperature oxidation of PM2000. Mater. High Temp. 17 (1), 133 137. Available from: https://doi.org/10.1179/mht.2000.020. Allen, T., Busby, J., Meyer, M., Petti, D., 2010. Materials challenges for nuclear systems. Mater. Today 13 (12), 14 23. Andresen, P.L., Morra, M.M., 2008. Stress corrosion cracking of stainless steels and nickel alloys in high-temperature water. Corrosion 64 (1), 15 29. Andresen, P.L., Jurewicz, T.J., Rebak, R.B., 2012. Stress corrosion cracking resistance of ferritic chromium steel in high temperature water. In: Paper C2012-0001181, Corrosion/2012 Conference, NACE International, Houston, TX. Andresen, P.L., Rebak, R.B., Dolley, E.J., 2014. SCC resistance of irradiated and unirradiated high Cr ferritic steels, In: Paper 3760, Corrosion/2014 Conference, NACE International, Houston, TX. Ang, C.Y., Burkhammer, E.W., 1960. Sintering of high density uranium dioxide bodies. J. Nucl. Mater. 2 (2), 176 180. Angelici Avincola, V., Grosse, M., Stegmaier, U., Steinbrueck, M., Seifert, H.J., 2015. Oxidation at high temperatures in steam atmosphere and quench of silicon carbide composites for nuclear application. Nucl. Eng. Des. 295, 468 478. Antonio, D.J., Shrestha, K., Harp, J.M., Adkins, C.A., Zhang, Y., Carmack, J., et al., 2018. Thermal and transport properties of U3Si2. J. Nucl. Mater. 508, 154 158. Arborelius, J., Backman, K., Hallstadius, L., Limba¨ck, M., Nilsson, J., Rebensdorff, B., et al., 2006. Advanced doped UO2 pellets in LWR applications. J. Nucl. Sci. Technol. 43 (9), 967 976. Available from: https://doi.org/10.1080/18811248.2006.9711184. Arioka, K., Yamada, T., Terachi, T., Staehle, R.W., 2006. Intergranular stress corrosion cracking behavior of austenitic stainless steels in hydrogenated high-temperature water. Corros. Sci. 62, 74 83. ASM International Handbook Committee, 1990. tenth ed. ASM International Handbook Committee, vol. 1. ASM International, Materials Park, OH.

199

200

References

Aydogan, E., Weaver, J.S., Maloy, S.A., El-Atwani, O., Wang, Y.Q., Mara, N.A., 2018. Microstructure and mechanical properties of FeCrAl alloys under heavy ion irradiations. J. Nucl. Mater. 503, 250 262. Available from: https://doi.org/10.1016/j.jnucmat.2018.03.002. Basu, S., Obando, N., Gowdy, A., Karaman, I., Radovica, M., 2012. Long-term oxidation of Ti2AlC in air and water vapor at 1000 1300 C temperature range. J. Electrochem. Soc. 159 (2), C90 C96. Battelle, 1970. Topical Report. In: Drennen, D.C., Jackson, C.M., Crites, N. (Eds.) Sorenson, Liquid Metal Engineering Center, November 13, 1970, Battelle Memorial Institute, Columbus, OH. Ben-Belgacem, M., Richet, V., Terrani, K.A., Katoh, Y., Snead, L.L., 2014. Thermo-mechanical analysis of LWR SiC/SiC composite cladding. J. Nucl. Mater. 447, 125 142. Betova, I., Bojinov, M., Kinnunen, P., Saario, T., 2011. Zn injection in pressurized water reactors laboratory tests, field experience and modelling. Research Report VTT-R-05511-11. Bischoff, J., Vauglin, C., Delafoy, C., Barberis, P., Perche, D., Guerin, B., et al., 2016. Development of Cr-coated zirconium alloy cladding for enhanced accident tolerance. In: Paper 17561, Proceeding of TopFuel 2016, American Nuclear Society, Boise, ID, September 11 14, 2016. Bischoff, J., Delafoy, C., Vauglin, C., Barberis, P., Roubeyrie, C., Perche, D., et al., 2018a. AREVA NP’s enhanced accident-tolerant fuel developments: focus on Cr-coated M5 cladding. Nucl. Eng. Technol. 50, 223 228. Available from: https://doi.org/10.1016/j. net.2017.12.004. Bischoff, J., Delafoy, C., Chaari, N., Vauglin, C., Buchanan, K., Barberis, P., et al., 2018b. Crcoated cladding development at framatome. In: Paper A0152, TopFuel 2018, European Nuclear Society, Prague, September 30 to October 4, 2018. Brachet, J.C., Le Saux, M., Lezaud-Chaillioux, V., Dumerval, M., Houmaire, Q., Lomello, F., et al., 2016. Behavior under LOCA conditions of enhanced accident tolerant chromium coated Zircaloy-4 claddings. In: Paper 17821, Proceeding of TopFuel 2016, American Nuclear Society, Boise, ID, September 11 14, 2016. Brachet, J.C., Dumerval, M., Lezaud-Chaillioux, V., Le Saux, M., Rouesne, E., Hamon, D., et al., 2017. Behavior of chromium coated M5 claddings under LOCA conditions. In: Paper A-047, Proceedings of WRFPM Conference, Jeju-do, Republic of South Korea, September 10 14, 2017. Brachet, J.-C., Idarraga-Trujillo, I., Le Flem, M., Saux, M. Le, Vandenberghe, V., Urvoy, S., et al., 2019. Early studies on Cr-coated Zircaloy-4 as enhanced accident tolerant nuclear fuel claddings for light water reactors. J. Nucl. Mater. 517, 268 285. Bragg-Sitton, S., Merrill, B., Teague, M., Ott, L., Robb, K., Farmer, M., et al., 2014. Advanced Fuels Campaign: Enhanced LWR Accident Tolerant Fuel Performance Metrics. INL/EXT13-29957, FCRD-FUEL-2013-000264, Idaho National Laboratory. Bragg-Sitton, S.M., Todosow, M., Montgomery, R., Stanek, C.R., Montgomery, R., Carmack, W.J., 2016. Metrics for the technical performance evaluation of light water reactor accident tolerant fuel. Nucl. Technol. 195 (2), 111 123. Buongiorno, J., 2014. Can corrosion and CRUD actually improve safety margins in LWRs? Ann. Nucl. Energy 63, 9 21. Cardinaels, T., Govers, K., Vos, B., Van den Berghe, S., Verwerft, M., de Tollenaere, L., et al., 2012. Chromia doped UO2 fuel: investigation of the lattice parameter. J. Nucl. Mater. 424, 252 260. Carmack, J., Goldner, F., Bragg-Sitton, S.M., Snead, L.L., 2013. Overview of the U.S. DOE Accident Tolerant Fuel Development ProgramSeptember 15 19 TopFuel 2013. American Nuclear Society, Charlotte, NC, pp. 734 739.

References

201

Carmack, W.J., Braase, L.A., Wigeland, R.A., Todosow, M., 2017. Technology readiness levels for advanced nuclear fuels and materials development. Nucl. Eng. Des. 313, 177 184. Carpenter, D.M., Kazimi, M.S., 2010. An Assessment of Silicon Carbide as a Cladding Material for Light Water Reactors, MIT-ANP-TR-132. Massachusetts Institute of Technology, November. Chao, J., Capdevila, C., Serrano, M., Garcia-Junceda, A., Jimenez, J.A., Miller, M.K., 2014. Effect of a a0 phase separation on notch impact behavior of oxide dispersion strengthened (ODS) Fe20Cr5Al alloy. Mater. Des. 53, 1037 1046. Available from: https://doi.org/ 10.1016/j.matdes.2013.08.007. Cheadle, B.A., 2010. The development of Zr 2.5Nb pressure tubes for CANDU reactors. J. ASTM Int. 7 (8), 1 15. Available from: https://doi.org/10.1520/JAI103057. Chen, J., Hoffelner, W., 2009. Irradiation creep of oxide dispersion strengthened (ODS) steels for advanced nuclear applications. J. Nucl. Mater. 392, 360 363. Chen, S., Yuan, C., 2017. Neutronic analysis on potential accident tolerant fuel-cladding combination U3Si2-FeCrAl. Sci. Technol. Nucl. Ins. 2017, 12 pp. Article ID 3146985. Available from: https://doi.org/10.1155/2017/3146985. Chen, J., Jung, P., Hoffelner, W., 2013. Irradiation creep of candidate materials for advanced nuclear plants. J. Nucl. Mater. 441, 688 694. Cheng, B., Kim, Y.-J., Chou, P., 2016. Improving accident tolerance of nuclear fuel with coated Mo-alloy cladding. Nucl. Eng. Technol. 48, 16 25. Cox, B., 1990. Pellet-clad interaction (PCI) failures of zirconium alloy fuel claddings. A review. J. Nucl. Mater. 172, 249 292. Cox, B., 2005. Some thoughts on the mechanisms of in-reactor corrosion of zirconium alloys. J. Nucl. Mater. 336, 331 368. Csontos, A., 2019. EPRI sponsored ATF R&D activities. In: 8th EPRI/INL/DOE Joint Workshop on Accident Tolerant Fuel, Tampa, FL, February 20 21, 2019. Cummins, W.E., Matzie, R., 2018. Design evolution of PWRs: shippingport to generation III 1 . Prog. Nucl. Energy 102, 9 37. Daub, K., Van Nieuwenhove, R., Nordin, H., 2015. Investigation of the impact of coatings on corrosion and hydrogen uptake of Zircaloy-4. J. Nucl. Mater. 467, 260 270. Available from: https://doi.org/10.1016/j.jnucmat.2015.09.041. Daub, K., Persaud, S.Y., Rebak, R.B., Van Nieuwenhove, R., Ramamurthy, S., Nordin, H., 2017. Investigating potential accident tolerant fuel cladding materials and coatings. In: Jackson, J.H., Paraventi, D., Wright, M. (Eds.), Proceedings of the 18th International Conference on Environmental Degradation of Materials in Nuclear Power Systems - Water Reactors. Springer International Publishing, Cham, 2018, pp. 215 234. Deck, C.P., Jacobsen, G.M., Sheeder, J., Gutierrez, O., Zhang, J., Stone, J., et al., 2015. Characterization of SiCeSiC composites for accident tolerant fuel cladding. J. Nucl. Mater. 466, 667 681. Delafoy, C., Arimescu, V.I., Hengstler-Eger, R.M., Landskron, H., Moeckel, A., Bellanger, P., 2015. Areva Cr2O3-doped fuel: increase in operational flexibility and licensing margins. In: Paper A0099, TopFuel 2015, Zurich, Switzerland, September 13 17, 2015. Delafoy, C., Bischoff, J., Larocque, J., Attal, P., Gerken, L., Nimishakavi, K., 2018. Benefits of Framatome’s E-ATF evolutionary solution: Cr-coated cladding with Cr2O3-doped fuel. In: Paper A0149, TopFuel 2018, Prague, Czechia, September 30 to October 4, 2018. Deshon, J., Hussey, D., Kendrick, B., McGurk, J., Secker, J., Short, M., 2011. Pressurized water reactor fuel crud and corrosion modeling. JOM 63, 64. Available from: https://doi.org/ 10.1007/s11837-011-0141-z.

202

References

DOE-NE, 2019. Department of Energy Office of Nuclear Energy, DOE Awards $111 Million to US Vendors to Develop Accident Tolerant Nuclear Fuels, January 31, 2019. Dolley, E.J., Schuster, M., Crawford, C., Rebak, R.B., 2018. Mechanical behavior of FeCrAl and other alloys following exposure to LOCA conditions plus quenching. In: Jackson, J.H., Paraventi, D., Wright, M. (Eds.), Proceedings from the 18th Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Springer, December 20, 2018, p. 1401. Edwards, H.S., Bohlander, K.M., 1969. Physico-Chemical Studies of Fe-Cr-Al-Clad Fuel Systems. United States, N.P. Available from: https://doi.org/10.2172/4823602. Ejenstam, J., Thuvander, M., Olsson, P., Rave, F., Szakalos, P., 2015. Microstructural stability of Fe Cr Al alloys at 450 550 C. J. Nucl. Mater. 457, 291 297. Available from: https:// doi.org/10.1016/j.jnucmat.2014.11.101. EPRI, 2019. Electric Power Research Institute, Accident-Tolerant Fuel Valuation: Safety and Economic Benefits, Technical Report 3002015091, Palo Alto, CA. Erbacher, F.J., Leistikow, S., 1987. Zircaloy fuel cladding behavior in a loss-of-coolant accident: a review. In: Adamson, R.B., Van Swam, L.F.P. (Eds.), Zirconium in the Nuclear Industry: Seventh International Symposium, ASTM STP 939. American Society for Testing and Materials, Philadelphia, PA, pp. 451 488. Evans, P.E., Davies, T.J., 1963. Uranium nitrides. J. Nucl. Mater. 10 (1), 43 55. Field, K.G., Gussev, M.N., Yamamoto, Y., Snead, L.L., 2014. Deformation behavior of laser welds in high temperature oxidation resistant Fe Cr Al alloys for fuel cladding applications. J. Nucl. Mater. 454 (1 3), 352 358. Available from: https://doi.org/10.1016/j. jnucmat.2014.08.013. Field, K.G., Hu, X., Littrell, K.C., Yamamoto, Y., Snead, L.L., 2015. Radiation tolerance of neutron-irradiated model FeCrAl alloys. J. Nucl. Mater. 465, 746 755. Available from: https://doi.org/10.1016/j.jnucmat.2015.06.023. Field, K.G., Briggs, S.A., Hu, X., Yamamoto, Y., Howard, R.H., Sridharan, K., 2017. Heterogeneous dislocation loop formation near grain boundaries in a neutron-irradiated commercial FeCrAl alloy. J. Nucl. Mater. 483, 54 61. Available from: https://doi.org/10.1016/j. jnucmat.2016.10.050. Floreen, S., Kane, R.H., Kelly, T.J., Robinson, M.L., 1980. An evaluation of incoloy alloy MA 956 for high temperature gas cooled reactor systems. J. Mater. Energy Syst. 2 (3), 15 24. Available from: https://doi.org/10.1007/BF02833434. Ford, F.P., Gordon, B.M., Horn, R.M., 2006. Corrosion in boiling water reactors. In: Cramer, S. D., Covino Jr., B.S. (Eds.), ASM Handbook, Vol. 13C. Corrosion: Environments and Industries, pp. 341 361. Available from: https://doi.org/10.1361/asmhba0004145. Franklin, D.G., 2011. Performance of zirconium alloys in light water reactors with a review of nodular corrosion. J. ASTM Int. 7 (6). Available from: https://doi.org/10.1520/JAI103032. Gan, J., Jerred, N., Perez, E., Haggard, D.C., Wen, H., 2017. Weld Development of FeCrAl thinwall cladding for LWR accident tolerant fuel. In: 2017 Water Reactor Fuel Performance Meeting, Paper A027, Jeju-do, Republic of Korea, September 10 14, 2017. Gan, J., Jerred, N., Perez, E., Haggard, D.C., 2018. Laser and pressure resistance weld of thinwall cladding for LWR accident-tolerant fuels. JOM 70, 192 197. Available from: https:// doi.org/10.1007/s11837-017-2697-8. Garzarolli, F., Stehle, H., Steinberg, E., 1996. Behavior and properties of Zircaloys in power reactors: a short review of pertinent aspects in LWR fuel. In: Bradley, E.R., Sabol, G.P. (Eds.), Zirconium in the Nuclear Industry: Eleventh International Symposium, ASTM STP 1295. ASTM International, pp. 12 32.

References

203

Geanta, V., Voiculescu, I., Stanciu, E.-M., 2016. Hafnium influence on the microstructure of FeCrAl alloys. In: International Conference of Innovative Research 2016, IOP Conference Series: Materials Science and Engineering, 133 pp. Available from: https://doi.org/10.1088/ 1757-899X/133/1/012016. Geelhood, K.G., Luscher, W.G., 2019. Degradation and Failure Phenomena of Accident Tolerant Fuel Concepts Chromium Coated Zirconium Alloy Cladding. Report PNNL-28437, January 2019. General Electric, 1965. Report GEMP-334A “Fourth Annual Report High-Temperature Materials and Reactor Component Development Programs,” ‘Volume I Section 7. Oxidation-Resistant Fuel Element Materials Research. Materials. Advanced Technology Services, Nuclear Materials & Propulsion Operation, United States Atomic Energy Commission, Contract No. AT (40-1) 2847, Cincinnati 15, OH, February 26, 1965. https://digital.library.unt.edu/ark:/67531/metadc1033470/m2/1/high_res_d/4637943.pdf. General Electric, 1966. Report GEMP-400A “Fifth Annual Report High-Temperature Materials Programs,” Part A Section 5. Oxidation-Resistant Fuel Element Materials Research. Materials. Atomic Products Division, Nuclear Materials & Propulsion Operation, United States Atomic Energy Commission, Contract No. AT (40-1) 2847, Cincinnati, OH, February 28, 1966. Available from: https://doi.org/10.2172/4537915. General Electric, 1967a. Report GEMP-475A “Sixth Annual Report High-Temperature Materials Programs,” Part A. Nuclear Technology Department, Nuclear Materials & Propulsion Operation, Nuclear Energy Division, United States Atomic Energy Commission, Contract No. AT (40-1) 2847, Cincinnati, OH, March 31, 1967. Available from: https:// doi.org/10.2172/4365010. General Electric, 1967b. Report GEMP-66 “AEC Fuels and Materials Development Program Progress Report No. 66,” Chapter 5 (57020), Advanced Long-Life Reactor Fuel Element, Moderator, Control, and Shield Materials Development by G.R. Van Houten and M.R. Broz, Nuclear Technology Department, Nuclear Energy Division, Nuclear Materials & Propulsion Operation, AEC Research and Development Report, GEMP-66, Cincinnati, OH, May 31, 1967. General Electric, 1968. Report GEMP-1004 “Seventh Annual Report AEC Fuels and Materials Development Program. High-Temperature Materials Programs,” Nuclear Technology Department, Nuclear Materials & Propulsion Operation, Nuclear Energy Division. United States Atomic Energy Commission, Contract No. AT (40-1) 2847, March 31, 1968. Available from: https://doi.org/10.2172/4823748. George, N.M., Powers, J.J., Maldonado, G.I., Terrani, K.A., Worrall, A., 2014. Neutronic analysis of candidate accident-tolerant cladding concepts in light water reactors. Transactions of the American Nuclear Society, vol. 111, Anaheim, CA, November 9 13, 2014. George, N.M., Terrani, K.A., Powers, J.J., Worrall, A., Maldonado, I., 2015. Neutronic analysis of candidate accident-tolerant cladding concepts in pressurized water reactors. Ann. Nucl. Energy 75, 703 712. Go¨tlind, H., Liu, F., Svensson, J.-E., Halvarsson, M., Johansson, L.-G., 2007. The effect of water vapor on the initial stages of oxidation of the FeCrAl alloy kanthal AF at 900 C. Oxid. Met. 67, 251. Gray, D.M., White, D.W., Andresen, P.L., Kim, Y.J., Lin, Y.-P., Curtis, T.C., et al., 2007. Fuel rod with wear-inhibiting coating. US Patent Application US 2009/0022259 A1. Filed in 2007 & Published on January 22, 2009. Guria, A., Charit, I., 2016. Tensile properties of accident-tolerant aluminum-bearing ferritic steels. Annals Nucl. Energy 100. Available from: https://doi.org/10.1016/j.anucene.2016.09.018.

204

References

Gupta, V.K., Larsen, M., Rebak, R.B., 2018. Utilizing FeCrAl oxidation resistance properties in water, air and steam for accident tolerant fuel cladding. ECS Trans. 85 (2), 3 12. Gussev, M.N., Field, K.G., Yamamoto, Y., 2017. Design, properties, and weldability of advanced oxidation-resistant FeCrAl alloys. Mater. Des. 129, 227 238. Hache, G., Chung, H.M., 2001. The history of LOCA embrittlement criteria. In: Conference: 28th Water Reactor Safety Information Meeting, Bethesda, MD, 10/23/2001--10/25/2001; Report ANL/ET/CP-102703, Published December 6, 2001. Han, X., Wang, Y., Peng, S., Zhang, H., 2019. Oxidation behavior of FeCrAl coated Zry-4 under high temperature steam environment. Corros. Sci. 149, 45 53. Harp, J.M., Lessing, P.A., Hoggan, R.E., 2015. Uranium silicide pellet fabrication by powder metallurgy for accident tolerant fuel evaluation and irradiation. J. Nucl. Mater. 466, 728 738. Hartmann, C., Totemeier, A., Holcombe, S., Liverud, J., Limi, M., Hansen, J.E., et al., 2018. Measurement station for interim inspections of Lightbridge metallic fuel rods at the Halden boiling water reactor. EPJ Web Conf. 170, 04011. Available from: https://doi.org/10.1051/ epjconf/201817004011. Hayes, S.L., 2019. DOE laboratory overview on ATF. In: 8th EPRI/INL/DOE Joint Workshop on Accident Tolerant Fuel Tampa, FL, February 20 21, 2019. He, Y., Chen, P., Wu, Y., Su, G.H., Tian, W., Qiu, S., 2018. Preliminary evaluation of U3Si2FeCrAl fuel performance in light water reactors through a multi-physics coupled way. Nucl. Eng. Des. 328, 27 35. Henager Jr., C.H., Schemer-Kohrn, A.L., Pitman, S.G., Senor, D.J., Geelhood, K.J., Painter, C. L., 2008. Pitting corrosion in CVD SiC at 300 C in deoxygenated high-purity water. J. Nucl. Mater. 378, 9 16. Hillner, E., 1977. Corrosion of zirconium-base alloys an overview. In: Lowe Jr., A.L., Parry, G.W. (Eds.), Zirconium in the Nuclear Industry, ASTM STP 633. ASTM International, pp. 211 235. Hirayama, H., Kawakubo, T., Goto, A., Kaneko, T., 1989. Corrosion behavior of silicon carbide in 290 C water. J. Am. Ceram. Soc. 72 (11), 2049 2053. Hofmann, P., 1999. Current knowledge on core degradation phenomena, a review. J. Nucl. Mater. 270, 194 211. Hoggan, R.E., He, L., Harp, J.M., 2018. Interdiffusion behavior of U3Si2 with FeCrAl via diffusion couple studies. J. Nucl. Mater. 502, 356 369. Hojn´a, A., 2013. Irradiation-assisted stress corrosion cracking and impact on life extension. Corrosion 69, 964 974. Holcomb, G.R., 2008. Superalloys for Ultra Supercritical Steam Turbines Oxidation Behavior. Superalloys 2008, TMS 2008, pp. 601 608. Holcomb, G.R., 2009. Steam oxidation and chromia evaporation in ultra-supercritical steam boilers and turbines. ECS Trans. 16 (44), 81 92. Available from: https://doi.org/10.1149/ 1.3224746. Holtzman, B., 2019. Industry ATF Working Group Update. 8th EPRI/INL/DOE Joint Workshop on Accident Tolerant Fuel Tampa, FL, February 20 21, 2019. Hwasung, Y., Maier, B., Mariani, R., Bai, D., Xu, P., Sridharan, K., April 2016. Development of zirconium-silicide coatings for accident tolerant zirconium-alloy fuel cladding. Proceedings of International Congress on Advances in Nuclear Power Plants. ICAPP 2016, San Francisco, CA, pp. 2126 2131. IAEA, 2019. International Atomic Energy Agency, Nucleus, Power Reactor Information System (PRIS), May 2019.

References

205

Idaho National Laboratory (INL), 2015. In: Gan, J., Perez, E., Haggard, D.C. (INL), Jerred, N. (University Space Research Association) (Eds.), Status Report on Weld Development for FeCrAl Thin-Wall Cladding for LWR Accident Tolerant Fuels, INL/LTD-15-36583, September 2015. Ja¨derna¨s, D., Corleoni, F., Puranen, A., Tejland, P., Granfors, M., Matsunaga, J., 2015. PCI imitigation using fuel additives. In: Paper A0121, TopFuel 2015, Zurich, Switzerland, September 13 17, 2015. Jones, R.L., Nelson, J.L., 1990. Hydrogen water chemistry for BWRs: a status report on the EPRI development program. Nucl. Eng. Des. 124, 33 42. Jo¨nsson, B., Berglund, R., Magnusson, J., Henning, P., Ha¨ttestrund, M., 2004. High temperature properties of a new powder metallurgical FeCrAl alloy. Mater. Sci. Forum 461 464, 455 462. Available from: https://doi.org/10.4028/www.scientific.net/MSF.461-464.455. Kalvala, P.R., Raja, K.S., Misra, M., Tache, R.A., 2009. Overview of welding of oxide dispersion strengthened (ODS) alloys for advanced nuclear reactor applications. In: Proceedings of 2009 International Congress on Advances in Nuclear Power Plants, Japan, p. 2572. Kanthal, 2018. Datasheet for APMT. kanthal.com/en/products/material-datasheets/tube/kanthalapmt/. Karoutas, Z., Brown, J., Atwood, A., Hallstadius, L., Lahoda, E., Ray, S., et al., 2018. The maturing of nuclear fuel: past to accident tolerant fuel. Prog. Nucl. Energy 102, 68 78. Kass, S., 1964. The Development of the Zircaloys. ASTM STP368-EB, November 1964. Katoh, Y., Ozawa, K., Shih, C., Nozawa, T., Shinavski, R.J., Hasegawa, A., et al., 2014a. Continuous SiC fiber, CVI SiC matrix composites for nuclear applications: properties and irradiation effects. J. Nucl. Mater. 448, 448 476. Katoh, Y., Snead, L.L., Cheng, T., Shih, C., Lewis, W.D., Koyanagi, T., et al., 2014b. Radiation-tolerant joining technologies for silicon carbide ceramics and composites. J. Nucl. Mater. 448, 497 511. Khalifa, H.E., Deck, C.P., Gutierrez, O., Jacobsen, G.M., Back, C.A., 2015. Fabrication and characterization of joined silicon carbide cylindrical components for nuclear applications. J. Nucl. Mater. 457, 227 240. Kim, W.-J., Hwang, H.S., Park, J.Y., Ryu, W.-S., 2003. Corrosion behaviors of sintered and chemically vapor deposited silicon carbide ceramics in water at 360 C. J. Mater. Sci. Lett. 22, 581 584. Kim, Y.-J., Rebak, R.B., Lin, Y.-P., Lutz, D., Crawford, D., Kucuk, A., et al., 2010. Photoelectrochemical investigation of radiation-enhanced shadow corrosion phenomenon. JAI 7 (7), 1 18. Available from: https://doi.org/10.1520/JAI102952. Kim, H.-G., Kim, I.-H., Park, J.-Y., Koo, Y.-H., 2015a. Application of coating technology on zirconium-based alloy to decrease high-temperature oxidation. In: Comstock, R., Barberis, P. (Eds.), Zirconium in the Nuclear Industry: 17th International Symposium, STP 1543. ASTM International, West Conshohocken, PA, pp. 346 369. Available from: https://doi. org/10.1520/STP154320120161. Kim, Y.J., Wagenbaugh, F., Jurewicz, T.B., Blair, R.J., Rebak, R.B., 2015b. Environmental behavior of light water reactor accident tolerant candidate cladding materials under design conditions. In: Paper C2015-5817, Proceedings of Corrosion 2015 Conference (NACE International), Houston, TX, March 15 19, 2015. Kim, H.-G., Yang, J.-H., Kim, W.-J., Koo, Y.-H., 2016a. Development status of accidenttolerant fuel for light water reactors in Korea. Nucl. Eng. Technol. 47, 1 15. Kim, H.-G., Yang, J.-H., Kim, W.-J., Koo, Y.-H., 2016b. Development status of accidenttolerant fuel for light water reactors in Korea. Nucl. Eng. Technol. 48, 1 15.

206

References

Kimura, A., Yuzawa, S., Sakamoto, K., Hirai, M., Kusagaya, K., Yamashita, S., 2017. Welding technology R&D of Japanese accident tolerant fuel cladding of FeCrAl-ODS steel for BWRs. In: Paper A-059, 2017 Water Reactor Fuel Performance Meeting, WRFPM 2017, Jeju-do, Korea, September 10 14, 2017. Kimura, A., Yuzawa, S., Yamasaki, Y., Yabuuchi, K., Sakamoto, K., Ukai, S., et al., 2018. Welding technology R&D of Japanese accident tolerant FeCrAl-ODS fuel cladding for BWRs. In: Paper A0073, TopFuel 2018, Prague, September 30 to October 4, 2018. Ko¨gler, R., Anwand, W., Richter, A., Butterling, M., Ou, X., Wagner, A., et al., 2012. Nanocavity formation and hardness increase by dual ion beam irradiation of oxide dispersion strengthened FeCrAl alloy. J. Nucl. Mater. 427, 133 139. Kondo, S., Lee, M., Hinoki, T., Hyodo, Y., Kano, F., 2015. Effect of irradiation damage on hydrothermal corrosion of SiC. J. Nucl. Mater. 464, 36 42. Lawrence Bright, E., Rennie, S., Siberry, A., Samani, K., Clarke, K., Goddard, D.T., et al., 2019. Comparing the corrosion of uranium nitride and uranium dioxide surfaces with H2O2. J. Nucl. Mater. 518, 202 207. Lee, Y., McKrell, T.J., Montecot, A., Pantano, M., Song, Y., Kazimi, M.S., 2016. Oxidation behavior of sintered tubular silicon carbide in pure steam I: experiments. Ceram. Int. 42, 1916 1925. Lemaignan, C., 2002. Physical phenomena concerning corrosion under irradiation of Zr alloy. In: Moan, G.D., Rudling, P. (Eds.), Zirconium in the Nuclear Industry: Thirteenth International Symposium, ASTM STP 1423. ASTM International, pp. 20 29. Lemaignan, C., 2006. Corrosion or zirconium alloy components in light water reactors. ASM Metals Handbook, vol. 13C, Corrosion: Environments and Industries. ASM International, Metals Park, OH, p. 415. Levchuk, D., Bolt, H., Do¨beli, M., Eggenberger, S., Widrig, B., Ramm, J., 2008. Al-Cr-O thin films as an efficient hydrogen barrier. Surf. Alumina Coat. Technol. 202, 5043 5047. Li, N., Parker, S.S., Wood, E.S., Nelson, A.T., 2018. Oxide morphology of a FeCrAl alloy, kanthal APMT, following extended aging in air at 300 C to 600 C. Metall. Mater. Trans. A 49 (7), 2940 2950. Available from: https://doi.org/10.1007/s11661-018-4649-5. Lin, Y.P., Fawcett, R.M., Desilva, S.S., Lutz, D.R., Yilmaz, M.O., Davis, P., et al., 2018. Path towards industrialization of enhanced accident tolerant fuel. Paper A0141 TopFuel 2018, European Nuclear Society, Prague, Czech Republic, September 30 to October 4, 2018. Little, E., Stow, D., 1979. Void-swelling in irons and ferritic steels: II. An experimental survey of materials irradiated in a fast reactor. J. Nucl. Mater. 87, 25 39. Lustman, B., 1979. Zirconium technology twenty years of evolution. In: Zirconium in the Nuclear Industry (Fourth Conference), ASTM STP 681, pp. 5 18. Lustman, B., 1981. Development of the Zircaloy clad UO2 fuel element for shippingport. J. Nucl. Mater. 100, 72 77. Lysell, G., Nystrand, A.-C., Ullberg, M., 2005. Shadow corrosion mechanism of Zircaloy. In: Rudling, P., Kammenzind, B. (Eds.), Zirconium in the Nuclear Industry: Fourteenth International Symposium, ASTM STP1467. ASTM International, West Conshohocken, PA, pp. 445 461. Available from: http://dx.doi.org/10.1520/STP37520S. Maier, B.R., Garcia-Diaz, B.L., Hauch, B., Olson, L.C., Sindelar, R.L., Sridharan, K., 2015. Cold spray deposition of Ti2AlC coatings for improved nuclear fuel cladding. J. Nucl. Mater. 466, 712 717. Available from: https://doi.org/10.1016/j.jnucmat.2015.06.028. Mardon, J.-P., Charquet, D., Senevat, J., 2000. Influence of composition and fabrication process on out-of-pile and in-pile properties of M5 alloy. In: Sabol, G.P., Moan, G.D. (Eds.),

References

207

Twelfth International Symposium on Zirconium in the Nuclear Industry, ASTM STP 1354. ASTM International, pp. 505 524. McCullum, R., 2019. Reactor Redo, Scientific American, May 2019, pp. 72 75. McEachern, R.J., Taylor, P., 1997. A Review of the Oxidation of Uranium Dioxide at Temperatures Below 400 C. AECL-11335, COG-95-281-I, Whiteshell Laboratories, Pinawa, Manitoba, Canada, 1997. Mehan, R.L., Wiesinger, F.W., 1961. Mechanical Properties of Zircaloy-2, KAPL-2110. Knolls Atomic Power Laboratory, Schenectady, NY. Meng, C., Yang, L., Wu, Y., Tan, J., Dang, W., He, X., et al., 2019. Study of the oxidation behavior of CrN coating on Zr alloy in air. J. Nucl. Mater. 515, 354 369. Available from: https://doi.org/10.1016/j.jnucmat.2019.01.006. Mervin, B., 2019. EPRI Fuel Reliability Program, Core-TAC, Technical Advisory Committee Meeting - Winter 2019. Tampa, FL, February 21, 2019. Middleburgh, S.C., Grimes, R.W., Lahoda, E.J., Stanek, C.R., Andersson, D.A., 2016. Nonstoichiometry in U3Si2. J. Nucl. Mater. 482, 300 305. Motta, A.T., Couet, A., Comstock, R.J., 2015. Corrosion of zirconium alloys used for nuclear fuel cladding. Annu. Rev. Mater. Res. 45, 311 343. Murabayashi, M., Tanaka, S., Takahashi, Y., 1975. Thermal conductivity and heat capacity of Zircaloy-2, -4 and unalloyed zirconium. J. Nucl. Sci. Technol. 12 (10), 661 662. Available from: https://doi.org/10.1080/18811248.1975.9733170. National Nuclear Laboratory, 2018. Accident Tolerant Fuel: A UK Perspective. https://www.nnl. co.uk/innovation-science-and-technology/121-2/position-papers/#8476. Nelson, A.T., Migdisov, A., Sooby Wood, E., Grote, C.T., 2018. U3Si2 behavior in H2O environments: Part II, pressurized water with controlled redox chemistry. J. Nucl. Mater. 500, 81 91. Nguyen, Q.N., Opila, E.J., Robinson, R.C., 2004. Oxidation of ultrahigh temperature ceramics in water vapor. J. Electrochem. Soc. 151 (10), B558 B562. National Nuclear Laboratory (NNL), 2018. Boiling Water Reactor Technology International Status and UK Experience, Position Paper. Birchwood Park, Warrington, UK. Noordhoek, M.J., Besmann, T.M., Andersson, D., Middleburgh, S.C., Chernatynskiy, A., 2016. Phase equilibria in the U-Si system from first-principles calculations. J. Nucl. Mater. 479, 216 223. Nuclear Energy Agency NEA, 2018. State-of-the-art report on light water reactor accident tolerant fuel. In: Prepared by the Expert Group on Accident-Tolerant Fuels for Light-Water Reactors, September 2018. Nuclear News, 2018. Hatch Unit Restarts With Accident-Tolerant Fuel. March 7, 2018. Ocken, H., 1980. An improved evaluation model for Zircaloy oxidation. Nucl. Technol. 47 (2), 343 357. Available from: https://doi.org/10.13182/NT80-A32437. Oelrich, R., Ray, S., Karoutas, Z., Xu, P., Romero, J., Shah, H., et al., 2018. Overview of westinghouse lead accident tolerant fuel program. TopFuel 2018. European Nuclear Society, Prague, Czech Republic, p. A0151. Olander, D., 2009. Nuclear fuels present and future. J. Nucl. Mater. 389, 1 22. Onal, K., Maris-Sida, M.C., Meier, G.H., Pettit, F.S., 2003. Water vapor effects on the cyclic oxidation resistance of alumina forming alloys. Mater. High Temp. 20 (3), 327 337. Available from: https://doi.org/10.1179/mht.2003.039. Orlov, A., Degueldre, C., Wiese, H., Ledergerber, G., Valizadeh, S., 2011. Corrosion product deposits on boiling-water reactor cladding: experimental and theoretical investigation of magnetic properties. J. Nucl. Mater. 416 (1 2), 117 124. September.

208

References

ORNL - Oak Ridge National Laboratory, 2018a. In: K.G., Field (Ed.), Handbook on the Material Properties of FeCrAl Alloys for Nuclear Power Production Applications, ORNL/ SPR-2018/905, M2NT-18OR020202091, August 2018. ORNL - Oak Ridge National Laboratory, 2018b. Handbook on the Material Properties of FeCrAl Alloys for Nuclear Power Production Applications. Rev. 1.1, ORNL/SPR-2018/905, In: Field, K.G., Snead, M.A., Yamamoto, Y., Terrani, K.A. (Eds.), M2NT-18OR020202091, Nuclear Technology Research and Development, ORNL (August 2018). Paljevi´c, M., Despotovi´c, Z., 1975. Oxidation of uranium mononitride. J. Nucl. Mater. 57, 253 257. Park, J.-Y., Kim, I.-H., Jung, Y.-I., Kim, H.-G., Park, D.-J., Kim, W.-J., 2013. Long-term corrosion behavior of CVD SiC in 360 C water and 400 C steam. J. Nucl. Mater. 443, 603 607. Pickman, D.O., 1994. Zirconium alloy performance in light water reactors: a review of UK and Scandinavian experience. In: Garde, A.M., Bradley, E.R. (Eds.), Zirconium in the Nuclear Industry: Tenth International Symposium, ASTM STP 1245. ASTM International, pp. 19 32. Powers, J.J., Worrall, A., Robb, K.R., George, N.M., Maldonado, G.I., 2016. Analysis of Operational and Safety Performance for Candidate Accident Tolerant Fuel and Cladding Concepts. Report IAEA-TECDOC-1797, International Atomic Energy Agency (IAEA). Quadakkers, W.J., Elschner, A., Speier, W., Nickel, H., December 1991. Composition and growth mechanisms of alumina scales on FeCrAl-based alloys determined by SNMS. Appl. Surf. Sci. 52 (4), 271 287. Ray, S., Lahoda, E., Franceschini, F., 2012. Assessment of different materials for meeting the requirement of the future fuel designs. In: TopFuel 2012, Reactor Fuel Performance Meeting, Paper A0115, Manchester, UK, September 2 6, 2012. Rebak, R.B., 2011. Environmental degradation of engineered barrier materials in nuclear waste repositories. In: Revie, W. (Ed.), Uhlig’s Handbook, Chapter 36. Wiley & Sons, pp. 503 516. Rebak, R.B., 2013. Resistance of ferritic steels to stress corrosion cracking in high temperature water. In: Paper PVP2013-97352, pp. V06AT06A085; 10 pp. ASME 2013 Pressure Vessels and Piping Conference, vol. 6A: Materials and Fabrication, Paris, France, July 14 18, 2013. Available from: https://doi.org/10.1115/PVP2013-97352. Rebak, R.B., 2015. Alloy selection for accident tolerant fuel cladding in commercial light water reactors. Metall. Mater. Trans. E 2, 197 207. Rebak, R.B., 2017. Iron-chrome-aluminum alloy cladding for increasing safety in nuclear power plants. EPJ Nucl. Sci. Technol. 3, 34. Rebak, R.B., 2018. Versatile oxide films protect FeCrAl alloys under normal operation and accident conditions in light water power reactors. JOM 70, 176. Rebak, R.B., Huang, S., 2015. Anticipated improved performance of advanced steel cladding under long term dry storage of spent fuel. In: Paper No. PVP2015-45643, pp. V007T07A042; 7 pp. ASME 2015 Pressure Vessels and Piping Conference, vol. 7: Operations, Applications and Components, Boston, MA, July 19 23, 2015. Available from: https://doi.org/10.1115/PVP2015-45643. Rebak, R.B., Kim, Y.-J., 2016. Hydrogen diffusion in FeCrAl alloys for light water reactors cladding applications. In: Paper PVP2016-63164, 2016 ASME PVP Conference, MF-7 Materials and Technologies for Nuclear Power Plants, July 17 21, 2016, Vancouver, BC. Rebak, R.B., Lin, Y.-P., Kim, Y.-J., August 2009. Review on electrochemical corrosion of zirconium alloys in high temperature water, The Proceedings of the 14th International

References

209

Conference on Environmental Degradation of Materials in Nuclear Power Systems-Water Reactors, 2. American Nuclear Society, Virginia Beach, VA, pp. 1400 1406. Rebak, R.B., Stachowski, R.E., Lin, Y.-P., Brown, N.R., Terrani, K.A., 2015. Assessment of advanced steels as accident tolerant fuel cladding for commercial light water reactors. In: 17th International Conference on Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, August 9 13, 2015, Ottawa, Ontario, Canada. Rebak, R.B., Larsen, M., Kim, Y.-J., 2017a. Characterization of oxides formed on ironchromium-aluminum alloy in simulated light water reactor environments. Corros. Rev. 35, 177. Rebak, R.B., Terrani, K.A., Gassmann, W.P., Williams, J.B., Ledford, K.L., 2017b. Improving nuclear power plant safety with FeCrAl alloy fuel cladding. MRS Adv. 2(21 22), 1217 1224 (Energy and Sustainability). Rebak, R.B., Gupta, V.K., Larsen, M., 2018. Oxidation characteristics of two FeCrAl alloys in air and steam from 800 C to 1300 C. JOM 70 (8), 1484 1492. Available from: https://doi. org/10.1007/s11837-018-2979-9. Rebak, R.B., Jurewicz, T.B., Kim, Y.-J., 2019a. Electrochemical behavior of accident tolerant fuel cladding materials under simulated light water reactor conditions. In: Papavinasam, S., Rebak, R.B., Yang, L., Berke, N.S. (Eds.), Advances in Electrochemical Techniques for Corrosion Monitoring and Laboratory Corrosion Measurements, ASTM STP1609. ASTM International, West Conshohocken, PA, pp. 231 243. Available from: http://dx.doi.org/ 10.1520/STP1609201701453. Rebak, R.B., Blair, R.J., Gupta, V.K., 2019b. Corrosion Evaluation of Iron-ChromiumAluminum alloys in spent fuel cooling pools. In: Paper 12944, Corrosion/2019, NACE International, Houston, TX. Rickover, H.G., Geiger, L.D., Lustman, B., 1975. History of the development of zirconium alloys for use in nuclear reactors. U.S. Report, Division of Naval Reactors, Washington, DC, TID-26740, March 1975. Rippon, S., 1984. History of the PWR and its worldwide development. Energy Policy 12 (3), 259 265. Roberts, D.A., 2016. Magnetron Sputtering and Corrosion of Ti-Al-C and Cr-Al-C coatings for Zr-alloy Nuclear Fuel Cladding (Master of Science Thesis). The University of Tennessee, Knoxville, May 2016. Sakamoto, K., Hirai, M., Ukai, S., Kimura, A., Yamaji, A., Kusagaya, K., et al., 2017. Overview of Japanese Development of Accident Tolerant FeCrAl-ODS Fuel Claddings for BWRs. 2017 Water Reactor Fuel Performance Meeting, 10 14 September 2017, Jeju Island, Korea. Sakamoto, K., Miura, Y., Ukai, S., Kimura, A., Yamaji, A., Kusagaya, K., et al., 2018. Progress on Japanese Development of Accident Tolerant FeCrAl-ODS Fuel Cladding for BWRs. Paper A0011, TopFuel 2018, Prague, Czechia, September 30 to October 4, 2018. Sauder, C., 2014. Ceramic matrix composites: nuclear applications. Ceram. Matrix Comp. Mater. Model. Technol. 609 646 (Chapter 22). Savchenko, A.M., Ivanov, V.B., Novikov, V.V., Skupov, M.V., Kulakov, G.V., Orlov, V.K., et al., 2015. Review of A. A. Bochvar Institute Activivites in Developing Potentially Accident Tolerant Fuel for Light Water Reactors. TopFuel 2015, Zurich, Switzerland, September 2015. Sawatzky, A., Ells, C.E., 2000. Understanding hydrogen in zirconium. In: Sabol, G.P., Moan, G. D. (Eds.), Zirconium in the Nuclear Industry: Twelfth International Symposium, ASTM STP 1354. ASTM International, pp. 32 48.

210

References

Schneider, R., Lutz, D.R., McCumbee, P., Cantonwine, P.E., 2018. GNF Fuel Reliability and Channel Performance: 2018 Update. Paper A0111 TopFuel 2018, European Nuclear Society, Prague, Czech Republic, September 30 to October 4, 2018. Schuster, M., Dolley, E.J., Jurewicz, J.B., Rebak, R.B., 2019. Environmental degradation resistance of ATF FeCrAl cladding tube specimens in fuel cycle conditions. In: Jackson, J.H., Paraventi, D., Wright, M. (Eds.). Proceedings from the 19th Environmental Degradation of Materials in Nuclear Power Systems Water Reactors, Boston, MA, August 2019. Scott, P.M., Combrade, P., 2006. Corrosion in pressurized water reactors. In: Cramer, S.D., Covino, Jr., B.S. (Eds.), ASM Handbook, vol. 13C: Corrosion: Environments and Industries, pp. 362 385. Available from: https://doi.org/10.1361/asmhba0004146. Sedriks, A.J., 1996. Corrosion of Stainless Steels, second ed. John Wiley & Sons, Inc, New York. Short, M.P., 2018. The particulate nature of the crud source term in light water reactors. J. Nucl. Mater. 509, 478 481. Silva, R.A., West, J.J., Zhang, Y., Anenberg, S.C., Lamarque, J.-F., Shindell, D.T., et al., 2013. Global premature mortality due to anthropogenic outdoor air pollution and the contribution of past climate change. Environ. Res. Lett. 8 (034005), 11. Available from: https://doi.org/ 10.1088/1748-9326/8/3/034005. Simnad, M.T., 1981. A brief history of power reactor fuels. J. Nucl. Mater. 100, 93 107. Snead, L.L., Nozawa, T., Katoh, Y., Byun, T.-S., Kondo, S., Petti, D.A., 2007. Handbook of SiC properties for fuel performance modeling. J. Nucl. Mater. 371, 329 377. Soler-Crespo, L., Martin-Munoz, F.J., Gomez-Briceno, D., 2009. Corrosion behaviour of PM2000 and MA956 steels in stagnant lead at 700C. 7 p. Eurocorr 2009. The European Corrosion Congress. Corrosion from the nano scale to the plant, Nice (France), September 6 10, 2009. Sooby Wood, E., White, J.T., Nelson, A.T., 2017a. Oxidation behavior of U-Si compounds in air from 25 to 1000 C. J. Nucl. Mater. 484, 245 257. Sooby Wood, E., White, J.T., Nelson, A.T., 2017b. The effect of aluminum additions on the oxidation resistance of U3Si2. J. Nucl. Mater. 489, 84 90. Sooby Wood, E., White, J.T., Grote, C.T., Nelson, A.T., 2018. U3Si2 behavior in H2O: Part I, flowing steam and the effect of hydrogen. J. Nucl. Mater. 501, 404 412. Stachowski, R.E., Fawcett1, R., Rebak, R.B., Gassmann, W.P., Williams, J.B., Terrani, K.A., 2017. Progress of GE development of accident tolerant fuel FeCrAl cladding. 2017 Water Reactor Fuel Performance Meeting, September 10 14, 2017, Jeju Island, Korea. Stephens, W.W., 1984. Extractive metallurgy of zirconium 1945 to the present. In: Franklin, D.G., Adamson, R.B. (Eds.), Zirconium in the Nuclear Industry: Sixth International Symposium, ASTM STP 824, ASTM International, pp. 5 36. Stott, F.H., Wood, G.C., Stringer, J., 1995. The influence of alloying elements on the development and maintenance of protective scales. Oxid. Met. 44, 113 145. Strehlow, R.A., Savage, H.C., 1974. The permeation of hydrogen isotopes through structural metals at low pressures and through metals with oxide film barriers. Nucl. Technol. 22 (1), 127 137. Available from: https://doi.org/10.13182/NT74-A16282. Sugihara, S., Imoto, S., 1969. Hydrolysis of uranium nitrides. J. Nucl. Sci. Technol. 6 (5), 237 242. Sun, Z., Yamamoto, Y., Chen, X., 2018. Impact toughness of commercial and model FeCrAl alloys. Mater. Sci. Eng. A 734, 93 101. Available from: https://doi.org/10.1016/j. msea.2018.07.074.

References

211

Tallman, D.J., Anasori, B., Barsoum, M.W., 2013. A critical review of the oxidation of Ti2AlC, Ti3AlC2 and Cr2AlC in air. Mater. Res. Lett. 1 (3), 115 125. Available from: https://doi. org/10.1080/21663831.2013.806364. Tallman, D.J., Hoffman, E.N., Caspi, E.N., Garcia-Diaz, B.L., Kohse, G., Sindelar, R.L., et al., 2015. Effect of neutron irradiation on select MAX phases. Acta Mater. 85, 132 143. Available from: https://doi.org/10.1016/j.actamat.2014.10.068. Tang, C., Stueber, M., Seifert, H.J., Steinbrueck, M., 2017. Protective coatings on zirconiumbased alloys as accident-tolerant fuel (ATF) claddings. Corros. Rev. 35 (3), 141 165. Taylor, K.M., McMurtry, C.H., 1960. Synthesis and Fabrication of Refractory Uranium Compounds. The Carborundum Company, September Monthly Progress Report No. 10, Niagara Falls, NY, October 13, 1960. Terrani, K.A., 2018. Accident tolerant fuel cladding development: promise, status, and challenges. JNM. American Nuclear Society. Available from: https://doi.org/10.1016/j. jnucmat.2017.12.043. Terrani, K.A.,Chinthaka Silva, G.W., Keiser, J.R., Brady, M.P., Cheng, T., Pint, B.A., et al., 2012. High temperature oxidation of silicon carbide and advanced iron-based alloys in steam-hydrogen environments. In: TopFuel 2012, September 2 6, 2012, Manchester, UK. Terrani, K.A., Yang, Y., Kim, Y.-J., Rebak, R.B., Meyer III, H.M., Gerczak, T.J., 2015. Hydrothermal corrosion of SiC in LWR coolant environments in the absence of irradiation. J. Nucl. Mater. 465, 488 498. Terrani, K.A., Pint, B.A., Kim, Y.-J., Unocic, K.A., Yang, Y., Silva, C.M., et al., 2016. Uniform corrosion of FeCrAl alloys in LWR coolant environments. J. Nucl. Mater. 479, 36 47. United Nations, 1958. Proceedings of the Second United Nations International Conference on the Peaceful Uses of Atomic Energy, Vol. 33, Geneva, Switzerland, September 1 13, 1958. Van Deventer, E.H., Maroni, V.A., 1983. Hydrogen permeation characteristics of some Fe-Cr-Al alloys. J. Nucl. Mater. 113 (1), 65 70. Van Duysen, J.C., Meric de Bellefon, G., 2017. 60th Anniversary of electricity production from light water reactors: Historical review of the contribution of materials science to the safety of the pressure vessel. J. Nucl. Mater. 484, 209 227. VDM, 2008. VDM Aluchrom Yhf. Material Data Sheet No. 4049, VDM Metals GmbH, March 2008. Chang, W., 2003. Zirconium Alloys Data Sheet. ZirAly-019, ATI Wah Chang. Wang, Y., Zhou, W., Wen, Q., Ruan, X., Luo, F., Bai, G., et al., 2018. Behavior of plasma sprayed Cr coatings and FeCrAl coatings on Zr fuel cladding under loss-of-coolant accident conditions. Surf. Coat. Technol. 344, 141 148. Wanklyn, J.N., Jones, P.J., 1962. The aqueous corrosion of reactor metals. J. Nucl. Mater. 6 (3), 291 329. Wharry, J.P., Was, G.S., 2013. A systematic study of radiation-induced segregation in ferritic martensitic alloys. J. Nucl. Mater. 442, 7 16. Available from: https://doi.org/10.1016/j. jnucmat.2013.07.071. White, J.T., Nelson, A.T., Dunwoody, J.T., Byler, D.D., Safarik, D.J., McClellan, K.J., 2015. Thermophysical properties of U3Si2 to 1773K. J. Nucl. Mater. 464, 275 280. Whitmarsh, C.L., 1962. Review of Zircaloy-2 and Zircaloy-4 properties relevant to N.S. Savannah Reactor Design, ORNL-3281, Oak Ridge National Laboratory, Oak Ridge, TN. Wilson, F.G., Knott, B.R., Desforges, C.D., 1978. Preparation and properties of some ODS FeCr-Al alloys. Metall. Mater. Trans. A 9 (2), 275 282. Available from: https://doi.org/ 10.1007/BF02646711.

212

References

Xu, P., Lahoda, E.J., Lyons, J., Deck, C.P., Kohse, G.E., 2018. Status update on Westinghouse SiC composite cladding fuel development. In: TopFuel 2018, Paper A0109, Prague, Czech Republic, September 30 to October 4, 2018. Yamamoto, Y., Pint, B.A., Terrani, K.A., Field, K.G., Yang, Y., Snead, L.L., 2015. Development and property evaluation of nuclear grade wrought FeCrAl fuel cladding for light water reactors. J. Nucl. Mater. 467, 703 716. Available from: https://doi.org/10.1016/ j.jnucmat.2015.10.019. Younker, I., Fratoni, M., 2016. Neutronic evaluation of coating and cladding materials for accident tolerant fuels. Prog. Nucl. Energy 88, 10 18. Available from: https://doi.org/10.1016/j. pnucene.2015.11.006. Yueh, K., Terrani, K.A., 2014. Silicon carbide composite for light water reactor fuel assembly applications. J. Nucl. Mater. 448, 380 388. Available from: https://doi.org/10.1016/j. jnucmat.2013.12.004. Yueh, K., Edsinger, K., Cantonwine, P., Feinroth, H., Griffith, G., Garnier, J., et al., 2012. Silicon carbide composite for BWR channel application. In: TopFuel 2012, September 2 6, 2012, Manchester, UK. Zakova, J., Wallenius, J., 2012. Fuel residence time in BWRs with nitride fuels. Ann. Nucl. Energy 47, 182 191. Zhong, W., Mouche, P.A., Han, X., Heuser, B.J., Mandapaka, K.K., Was, G.S., 2016. Performance of iron-chromium-aluminum alloy surface coatings on Zircaloy 2 under hightemperature steam and normal BWR operating conditions. J. Nucl. Mater. 470, 327 338. Zinkle, S.J., Was, G.S., 2013. Materials challenges in nuclear energy. Acta Mater. 61, 735 758. Available from: https://doi.org/10.1016/j.actamat.2012.11.004. Zinkle, S.J., Terrani, K.A., Gehin, J.C., Ott, L.J., Snead, L.L., 2014. Accident tolerant fuels for LWRs: a perspective. J. Nucl. Mater. 448, 374 379. Zok, F.W., 2016. Ceramic-matrix composites enable revolutionary gains in turbine engine efficiency. Am. Ceram. Soc. Bull. 95, 22 28.

Index Note: Page numbers followed by “f” and “t” refer to figures and tables, respectively.

A Accident tolerant fuels (ATF), 28 30, 43 45, 48 49, 60 64, 158, 171 172 ATF-1, 55 56, 138 139 ATF-2, 55 56, 138 139 claddings, 67, 144, 164 165 FeCrAl alloys for, 88 92 NEA assessment on maturity, 174 180, 176t under consideration, 160 Electric Power Research Institute (EPRI), 44, 60 62 Fukushima nuclear power stations, 45 47 industrial civilian nuclear power participation, 56 Nuclear Energy Institute (NEI), 44, 57 59 program in United States, 53 56 programs, 90, 197 198 safer materials, 188 development for nuclear power plants, 47 49 ten-year plan, 54f timeline for development, 49 51, 50f Accident-tolerant fuel maturity concepts, 51 52 assessment, 172 174, 180 184 attributes for cladding and fuels, 174t technology readiness level (TRL), 173t nuclear fuel cycle, 185 186 Advanced doped pellet technology (ADOPT), 163 Advanced Test Reactor (ATR), 55 56, 79, 118 119, 172, 183 A-group element, 69 Aircraft carrier, 10 11 Alloy 33, 106 108 Alloys manufactured by powder process (APMT), 85 86, 91, 93 94, 103, 106 112, 121, 127 128 elongation to failure, 94 96, 96f fracture surface, 117f

grain growth for, 118f mass change in BWR testing conditions, 98 99, 99f out-of-pile immersion tests, 98t oxide composition/formation, 123f, 124f in BWR NWC, 103f in PWR, 103f oxide top surface PWR, 105f steam oxidation, 122f tube, 130 131 microstructures, 130f ultimate tensile strength, 94 96 yield stress, 94 96, 95f Alternative tolerant fuel zirconium alloys, 65 66 Aluchrom YHf, 89 90 Alumina (Al2O3), 78, 121 123, 127 128 Aluminum (Al), 129 Al-based coatings for zirconium alloys, 72 74 needed for oxidation resistance, 87f Anticipated operational occurrences (AOOs), 60 62, 158, 175 177, 193 APMT. See Alloys manufactured by powder process (APMT) APS-1. See Atomic power station 1 (APS-1) Argon (Ar), 106 ARMOR, 50, 54, 68 69 ARMOR-coated Zircaloy-2, 80 coating, 77, 79 80 As-received tube specimen, 126 127 ATF. See Accident tolerant fuels (ATF) Atomic energy. See Nuclear energy Atomic hydrogen (H ), 35 36 Atomic power station 1 (APS-1), 10 ATR. See Advanced Test Reactor (ATR) Austenitic stainless steels, 13 14, 20, 79 80, 113 tubing, 23 Austenitic type 304SS alloy, 91 92, 106 108

213

214

Index

B

BDBA. See Beyond design basis accidents (BDBA) Bettis atomic power laboratory, 27 Beyond design basis accidents (BDBA), 92, 158, 172 174 LOCA scenario, 191 192 scenarios, 60 62 Black zirconium oxide, 30 31 Boiling water reactors (BWRs), 4 5, 16 17, 19, 45 46, 96 98, 144, 161, 190 194. See also Light water reactors (LWRs); Pressurized water reactors (PWRs) design, 19 fuel assemblies for, 22 23 fuel components, 97f mass change in, 98 99, 99f materials, 18f out-of-pile testing conditions, 32t reactor vessel for, 20 22 Vallecitos, 19 Breakaway corrosion, 30 31 BWRs. See Boiling water reactors (BWRs)

C

C26M. See Melting, casting, and forging (C26M) CANDU reactors, 28 30 Carbothermic reduction of urania, 169 Ceramic coatings, 69 71 Ceramic matrix composite, 146 147 Channels, 22 23, 28 boxes, 147 148 material, 144 Chemical vapor deposition (CVD), 75, 148 149 Chemically vapor infiltrated (CVI) medium, 147 SiC composite material, 152 153 Chromia (Cr2O3), 63 64, 66, 121 123, 158 159. See also Urania (UO2) chromia-doped urania, 171, 180 fuels, 161, 164, 183 layer, 134 136 Chromium (Cr), 17 18, 128 129 chromium-coated zirconium alloys, 171, 174 179 coatings, 63 64, 71 72 Cr-Al-C Zirlo coated coupons, 76 77 CrN coating, 70 71

needed for oxidation resistance, 87f Chromium oxide (CrO), 119 120 CILC. See Crud-induced localized corrosion (CILC) Civilian nuclear power, 8 10 generation, 7 8 participation, 56 Cladding, 16 17, 41, 171. See also Accident tolerant fuels (ATF)—claddings application in LWRs, 155 ballooning, 164 Clean energy, 1 2 Coal mining accidents, 7 Coal plants, 2 Coated zirconium. See also Zirconium alloys under accident conditions performance, 77 79 irradiation behaviour, 79 80 licensing for reactor use, 80 81 under reactor normal operation conditions performance, 76 77 Coefficients of thermal expansion (CTE), 91 92, 94t Combustion Engineering, 19 20 Commercial nuclear power, 8 14, 45 commercial nuclear accidents, 8t Comminution planetary milling process, 167 168 Comprehensive scoping analysis, 163 Computed tomography (CT), 133 134 Concrete containment building, 20 22 Condensed water, 149 150 Coolant, 16 17 Copper/copper oxide internal junction (Cu/ Cu2O internal junction), 106 108 Corrosion potential in FeCrAl alloys, 108f in hydrogen, 109f in oxygen, 109f under UV irradiation, 111f Critical heat flux, 191 192 Crud, 40 41, 192 193 Crud-induced localized corrosion (CILC), 40 41 CT. See Computed tomography (CT) CTE. See Coefficients of thermal expansion (CTE) Cu/Cu2O internal junction. See Copper/copper oxide internal junction (Cu/Cu2O internal junction) Current accident tolerant fuels maturity concepts, 51 52, 53t

Index CVD. See Chemical vapor deposition (CVD) Cyclic potentiodynamic polarization, 139, 140f

D

DBA. See Design-basis accident (DBA) DBTT. See Ductile-to-brittle transition temperature (DBTT) Degradation process of SiC, 149 150 Delayed hydride cracking (DHC), 35 36 Departure from nucleate boiling (DNB), 191 192 Design-basis accident (DBA), 60 62, 65 67, 92, 158, 172 174, 190 194 Deviations, 144 DHC. See Delayed hydride cracking (DHC) Disilicide, 164 165 DNB. See Departure from nucleate boiling (DNB) Doped pellets, 161 Doping, 158 159 improved urania fuels by, 160 163 Ductile-to-brittle transition temperature (DBTT), 88

E

EAC. See Environmentally assisted cracking (EAC) EATF. See Enhanced accident tolerant fuels (EATF) EB welding. See Electron beam welding (EB welding) ECCS. See Emergency core cooling system (ECCS) ECR. See Equivalent cladding reacted (ECR) EGATFL. See Expert Group on Accident Tolerant Fuels for Light Water Reactors (EGATFL) Electric Power Research Institute (EPRI), 45, 60 62, 134 136, 190 boiling water reactor fuel failures, 61f PWR fuel failures, 62f Electricity, 1 7 Electrochemical behavior of FeCrAl alloys, 106 108 Electron beam welding (EB welding), 133 134 Emergency core cooling system (ECCS), 116 Enhanced accident tolerant fuels (EATF), 80 Environmentally assisted cracking (EAC), 16 17, 20, 91, 114f, 115f

215

resistance to EAC of ferritic alloys, 113 115 EPRI. See Electric Power Research Institute (EPRI) EQ. See Equipment qualification (EQ) Equipment qualification (EQ), 194 Equivalent cladding reacted (ECR), 65 Evolutionary trend of nuclear fuel, 188 189, 189f Expert Group on Accident Tolerant Fuels for Light Water Reactors (EGATFL), 51 52, 171, 174 External monolithic layer, 149

F Fabrication of FeCrAl cladding tubes, 129 131 fuel rod fabrication steps, 132f process of urania fuel, 167 Fast reactor power ramping, 36 37 Fe-15Cr-4Al-1Y alloy, 87 88 Fe-25Cr-4Al-1Y alloy, 87 88 FeCrAl alloys. See Iron chromium aluminum alloys (FeCrAl alloys) Fe-Cr-Ni alloys, 15 Ferritic alloys, 91 FeCrAl, 121 Ferritic FeCrAl alloys, 84, 91 92, 129 Ferritic stainless steels, 13 14, 113 Fleet, civilian reactors, 24 Free uranium (U), 88 Fretting failure, 115 FRP. See Fuels Reliability Program (FRP) Fuel cladding, 148 149 Fuel rod, 16 17, 22 Fuels Reliability Program (FRP), 60 Fukushima Daiichi plant, 7 8, 194, 197 198 accident, 44 45 Fukushima nuclear power stations, 45 47 reactors in Fukushima preaccident, 46f, 48f

G Galvanic corrosion, 108 110 of FeCrAl alloys, 111 112 Gas tungsten arc welding (GTAW), 132 133 General Electric (GE), 19, 50 51, 87 88, 113 115, 141 Generation IV reactors, 88 89 Global Nuclear Fuels, 141

216

Index

GTAW. See Gas tungsten arc welding (GTAW) 62 GWd/MTU, 191 192 74 GWd/MTU, 191 192

H Hafnium (Hf), 24, 84 85 Haynes 214 nickel chromium aluminum alloy, 120 Helium permeability studies, 148 Hermetic cladding, 166 Higher density fuels, 157 159 HWC. See Hydrogen water chemistry (HWC) Hydrogen, 106 corrosion potential in, 109f explosions, 197 water chemistry, 149 150 Hydrogen water chemistry (HWC), 96 97 mass change in BWR, 99f

I Idaho National Laboratory (INL), 55 56, 79 Ignitable hydrogen gas, 44 45 Immersion corrosion tests, 105, 153 tube coupons, 101 102, 102f Incoloy MA956 alloy, 89 90 INL. See Idaho National Laboratory (INL) Internal monolithic layer, 149 Ion implantation, 68 Iron (Fe), 17 18 Iron-based alloys, 20 Iron chromium aluminum alloys (FeCrAl alloys), 49, 72 73, 84 85 for accident-tolerant fuel cladding, 88 92 benefits and detriments, 93f alloys, 50 51 alloys for electrochemical measurements, 107t APMT alloy, 153 154 characteristics, 90f cladding, 145 146 coatings, 72 74 considerations for nuclear applications, 87 88 corrosion potential, 108f cyclic potentiodynamic polarization, 139 electrochemical behavior in hightemperature water, 106 108 fabrication and implementation of cladding tubes, 129 131

fuel rod fabrication steps, 132f galvanic corrosion, 111 112 interaction between urania fuel and FeCrAl cladding, 118 119 irradiation behavior of FeCrAl, 137 139 licensing for reactor use, 140 141 mechanical properties, 93 96 metal oxide roles on surface normal operation oxidation to accident oxidation scenario, 124 125 steam-oxidized APMT tubes exposed to high-temperature water, 126 127 water-oxidized APMT tubes exposed to superheated steam, 125 metallurgy, 85 86 microstructure, 85 86, 86f mitigation measures to increased tritium release into coolant, 134 137 to parasitic neutron absorption, 134 monolithic alloys, 171 monolithic cladding, 180 nominal composition, 89t oxidation at high temperature in air or steam, 122f oxidation resistance, 119 121 under LWR’s normal operation conditions, 96 102 oxide films composition on FeCrAl coupons, 102 105 protection mechanism at accident condition temperatures, 121 124 resistance to crud deposition under normal operation conditions, 112 113 to EAC of ferritic alloys under LWR normal operation conditions, 113 115 to fretting under normal operation conditions, 115 116 of monolithic FeCrAl cladding to thermal shock, 116 118 shadow corrosion, 108 110 thermal properties, 92 versatile oxidation behavior, 127 129, 128f welding, 131 134 IronClad, 80, 116, 138 141 mechanical and neutron absorption properties, 135f tubes, 131 132 Irradiation behavior of FeCrAl, 137 139

Index

K Kanthal alloys, 84 85, 119 Kanthal-D, 132 133 Kroll Process, 11 12, 24 25

L Laser beam welding (LBW), 132 133 Laser techniques, 72 LBW. See Laser beam welding (LBW) Lead test assembly (LTA), 53, 80 Legacy alloys, 89 90 Licensing process in United States, 188 for reactor use, 140 141 Light water reactors (LWRs), 15 17, 46 47, 88 89, 94 96, 144. See also Boiling water reactors (BWRs); Pressurized water reactors (PWRs) fuel, 159 and performance of urania, 23 materials, 17 20 nodular corrosion, 34 35, 34f nominal composition of, 21t normal operation conditions, 96 102 nuclear power reactor, 16 17 operational conditions for fuel rods, 28t resistance to EAC of ferritic alloys, 113 115 zirconium alloys, 17, 24 30 LOCA. See Loss of coolant accident (LOCA) Local radiolysis, 108 110 Loss of coolant accident (LOCA), 129, 158, 177 178, 188 Low-carbon economies, 51 52 Lower fuel costs, 23 LTA. See Lead test assembly (LTA) LWRs. See Light water reactors (LWRs)

M MA956 alloy, 132 133 MAX phases, 69 Melting, casting, and forging (C26M), 85 86, 89 91, 103, 106 110 oxide top surface PWR, 105f tube, 130 131 microstructures, 130f Metal oxide roles on FeCrAl surface normal operation oxidation to accident oxidation scenario, 124 125 steam-oxidized APMT tubes exposed to high-temperature water, 126 127

217

water-oxidized APMT tubes exposed to superheated steam, 125 Metallic coated zirconium alloys, 145 146 Metallurgy of FeCrAl alloys, 85 86 Microcracking, 147, 156 Microstructure of FeCrAl alloys, 85 86, 86f Misorientation crystallographic patterns, 153 Mitigation measures of FeCrAl to increased tritium release into coolant, 134 137 to parasitic neutron absorption, 134 Molecular hydrogen (H2). See Atomic hydrogen (H ) pickup by zirconium alloys, 35 36 Molybdenum (Mo), 84 85, 89 Monolithic FeCrAl alloy, 171, 174 175 metallic claddings, 144, 148 149 SiC, 147 specimens of U3Si2, 166 uranium dioxide. See Monolithic urania (UO2) Monolithic urania (UO2), 158 159 fuel, 158

N Nanoferritic 14Cr alloy (NFA), 106 108 National Nuclear Laboratory (NNL), 189 Natural gas, 2 Natural uranium, 8 10 Naval reactors, 24 NEA. See Nuclear Energy Agency (NEA) NEI. See Nuclear Energy Institute (NEI) Neutron irradiation, 155 tests, 138 Neutron penalty, 92 NFA. See Nanoferritic 14Cr alloy (NFA) NFD tube coupons. See Nippon fuel development tube coupons (NFD tube coupons) Nickel (Ni), 17 18 Nickel alloy X-750, 106 108 Nickel oxide (NiO), 108 110 Nickel-based alloys, 20 Niobium (Nb), 64 65 Nippon fuel development tube coupons (NFD tube coupons), 99 100 NNL. See National Nuclear Laboratory (NNL) Nodular corrosion, 34 35, 34f Nonpreoxidized tube specimen, 126 127 Normal operation oxidation to accident oxidation scenario, 124 125

218

Index

Normal water chemistry (NWC), 96 97 mass change in BWR, 100f oxide composition of APMT in BWR, 103f NPPs. See Nuclear power plants (NPPs) NRC. See Nuclear Regulatory Commission (NRC) Nuclear electricity, 1 Nuclear energy, 1 2, 5 10, 197 198 benefits, 5 10 US reactors, 13f Nuclear Energy Agency (NEA), 171 assessment on maturity of ATF cladding concepts, 174 180, 176t EGATFL, 51 52 maturity level for claddings under accident conditions, 178t for fuel cycle issues of cladding concepts, 179t for three fuel concepts regarding fabrication and manufacturability, 181t Nuclear Energy Institute (NEI), 55, 57 59, 59t clean energy, 57f Nuclear fission energy, 2 3 reactions, 20 22 use of, 10 Nuclear fuel cladding, 148 cycle, 185 186 of fuel bundle, 186f Nuclear power, 1, 197 198 applications, 145 benefits of nuclear energy, 5 8 commercial nuclear power, 8 14 reactors under construction, 6f stations, 17, 19 timeline, 9f World energy consumption, 2f Nuclear power plants (NPPs), 2, 5 7, 190 development of safer materials for, 47 49 operation safety, 188 Nuclear Regulatory Commission (NRC), 44, 187 Nuclear-grade SiC/SiC composite, 155 NWC. See Normal water chemistry (NWC)

O Oak Ridge National Laboratory (ORNL), 89 90, 130 Obninsk reactor, 10

ODS. See Oxide dispersion strengthening (ODS) Oregon plant, 24 25 Organisation for Economic Co-operation and Development (OECD), 51 52 ORNL. See Oak Ridge National Laboratory (ORNL) Oxidation, 30 31 of FeCrAl at high temperature, 122f parabolic oxidation rate, 67t performance, 68 protection, 67 68 resistance of FeCrAl alloys, 96 102, 119 121 of zirconium, 30 31 Oxide dispersion strengthening (ODS), 87 88, 94 96, 119, 134 136 Oxide films composition on FeCrAl coupons, 102 105 Oxygen, 106 corrosion potential in, 109f

P Pacific Northwest Laboratory, 80 81 Parasitic neutron capture. See Neutron penalty PCI. See Pellet-cladding interaction (PCI) PCMI. See Pellet-cladding mechanical interaction (PCMI) Pellet-cladding interaction (PCI), 118, 138 139, 158 Pellet-cladding mechanical interaction (PCMI), 26 27, 36 37, 36f, 170, 172, 175 177 Performance of fuel rod, 168 169 Photoconductivity effect, 38 39 Physical vapor deposition process (PVD process), 68 PIE. See Post irradiation examination (PIE) Pitting resistance equivalent (PRE), 139 140 PM2000 alloy, 89 90 PO tube specimen. See Preoxidized tube specimen (PO tube specimen) Post irradiation examination (PIE), 80, 138 139, 162 Power plants, 2 Power ramp tests, 161 162 Power reactors, 3 4. See also Nuclear power plants (NPPs) age, 5f operation, 3f evolutionary trend of nuclear fuel, 188 189

Index licensing process in United States, 188 safety analysis and source term, 190 195 PRA. See Probabilistic risk assessment (PRA) PRE. See Pitting resistance equivalent (PRE) Preoxidized tube specimen (PO tube specimen), 126 127 Pressure resistance welding (PRW), 131 133 Pressure vessel, 20 22 Pressurized water reactors (PWRs), 4 5, 16 17, 19 20, 96 98, 145 146, 161, 190 194. See also Boiling water reactors (BWRs); Light water reactors (LWRs) fuel assemblies for, 22 23 mass change in, 101, 101f materials, 18f out-of-pile testing conditions, 32t oxide APMT tube coupon, 105f oxide composition of APMT in, 103f oxide top surface, 105f reactor vessel for, 20 22 Probabilistic risk assessment (PRA), 190 194 PRW. See Pressure resistance welding (PRW) PVD process. See Physical vapor deposition process (PVD process) PWRs. See Pressurized water reactors (PWRs)

R Radiation-induced segregation (RIS), 138 Radioactivity, 190 191, 197 Radiolytic species, 149 150, 160 Rare earths (RE), 89 Rate limiting step, 152 153 RBMK design, 10 RE. See Rare earths (RE) Reactivity initiated accidents (RIAs), 55 56, 177 178 Reactor irradiation, 41 service, 30 31 Reactor pressure vessel (RPV), 20 22 Revolutionary concept, 145 146 RIAs. See Reactivity initiated accidents (RIAs) RIS. See Radiation-induced segregation (RIS) RPV. See Reactor pressure vessel (RPV)

S S1W, 11 12 Safety analysis and source term, 190 195 crud deposition, 192 193

219

design-basis accidents, 193 equipment classification and life extension, 194 flexible power operations, 190 fuel cycle economics, 191 192 probabilistic risk assessment, 190 191 source term improvements, 193 194 Second phase precipitates (SPPs), 31, 34 35 Shadow corrosion of FeCrAl alloys, 108 110 SHE scale. See Standard hydrogen electrode scale (SHE scale) SiC composite. See Silicon carbide composite (SiC composite) SiC/SiC composites, 147 148, 193 194. See also Silicon carbide composite (SiC composite) ceramic composites, 174 175 cladding concept, 174 175 CMC composite material, 171 engineered systems, 147 environmental behavior under accident conditions, 153 154 under normal operation conditions, 149 153 permeability of fuel cladding, 148 thermal properties of fuel cladding, 148 Silica (SiO2), 152 153, 165 166 Silicon carbide composite (SiC composite), 143 148, 156. See also SiC/SiC composites ceramics, 144 fabrication, 148 149 fuel cladding, 148 149 implementation, 148 149 irradiation behavior, 155 156 licensing for reactor use, 156 metal/materials, 144, 147 148 thermal neutron absorption cross section, 146t Silicon-based coatings, 74 75 Sintered SiC (SSiC), 151 Southern Nuclear, 141 SPPs. See Second phase precipitates (SPPs) SSiC. See Sintered SiC (SSiC) Stainless steels, 11 12, 20, 128 129 Standard hydrogen electrode scale (SHE scale), 106 108 Steam, 16 17 steam-oxidized APMT tubes exposed to high-temperature water, 126 127 turbines, 2

220

Index

Stress corrosion cracking (SCC). See Environmentally assisted cracking (EAC) Stress strain APMT quenching tests, 116, 117f Submarine, 17 20, 24 26 Swelling process of fuel, 36 37 of UN, 170

T Tantalum (Ta), 84 85, 89 Technology Readiness Assessment (TRA), 171 Technology readiness level (TRL), 171 172, 173t TRL 9 fuel rod concept, 174 Tensile tests, 133 134 TEP. See Transient eutectic phase (TEP) Terminal solid solubility (TSS), 35 36 Thermal conductivity, 148, 155 156 decrease in thermal conductivity of SiC, 155f of urania, 148 Thermal diffusivity, 148 304SS alloy, 13 14, 106 Third element effect, 119 120 TIG welding. See Tungsten inert gas welding (TIG welding) Time-at-temperature limit, 193 Titanium (Ti), 84 85, 89 Titanium nitride (TiN), 69 70 Tohoku earthquake, 46 47 TRA. See Technology Readiness Assessment (TRA) Transient eutectic phase (TEP), 148 149 Tristructural-isotropic (TRISO), 155 TRL. See Technology readiness level (TRL) TSS. See Terminal solid solubility (TSS) Tungsten inert gas welding (TIG welding), 131 132

U Ultraviolet illumination (UV illumination), 38 39, 108 110 UN. See Uranium nitride (UN) UO2. See Urania (UO2) Urania (UO2), 11 12, 16 17, 157 159, 188 189, 195. See also Chromia (Cr2O3); Uranium disilicide (U3Si2) ATFs under consideration, 160 excellent performance, 159 160 fuel, 23, 118 119

fuel rod system, 112 113 improved urania fuels by doping, 160 163 modified urania performance under accident conditions, 164 under normal operation conditions, 163 164 nuclear fuel, 159 pellets, 159 uranium nitride, 168 169 uranium silicide, 164 165 Uranium (U-235), 158 Uranium compounds, properties of, 168t Uranium dioxide. See Urania (UO2) Uranium disilicide (U3Si2), 164 165. See also Urania (UO2) fabrication of U3Si2 fuels, 167 168 fuel, 165 implementation of U3Si2 fuels, 167 168 reactivity, 165 167 with cladding, 167 Uranium mononitride, 168 169 behavior under irradiation, 170 fabrication paths for, 169 reactivity of UN fuel, 169 Uranium nitride (UN), 157 158, 168 169, 180 Uranium silicide (U3Si2), 157 158, 164 165, 180 properties of silicide fuels, 164t Uranium-238 isotope (U-238 isotope), 8 10 US Department of Energy (US DOE), 53, 171 US DOE Office of Nuclear Energy (US DOENE), 56 US Nuclear Regulatory Commission (US NRC), 55, 140 141 UV illumination. See Ultraviolet illumination (UV illumination)

V Vallecitos reactor, 13 14 Versatile oxidation behavior, 127 129, 128f

W Water-oxidized APMT tubes exposed to superheated steam, 125 Welding of FeCrAl alloys, 131 134 White zirconium oxide, 30 31

X X-750 nickel alloy, 108 112, 115

Index

Y Yankee-Rowe PWR plant, 12 13 Yield stress (YS), 137 138 Yttrium (Y), 84 85

Z Zircaloy alloys, 25 27, 178 179 channel, 37 38 cladding, 158 fabrication, 13 14 mass gain, 32f zircaloy-2 tube specimens, 31f Zircaloy-1, 25 26 Zircaloy-2, 15 16, 25 26, 108 112, 121 elongation to failure, 94 96, 96f mechanical and neutron absorption properties, 135f steam oxidation, 122f tube coupons, 99 101 ultimate tensile strength, 94 96 yield stress, 94 96, 95f Zircaloy-4, 15 16, 25 26, 70 71 Zirconium (Zr), 64 65, 84 85, 89 clad rods, 23 zirconium-based alloys, 50 51 zirconium-coated rods fabrication and implementation, 74 75 Zirconium alloys, 15 17, 24 30, 63 66, 69, 108 110, 177, 188, 197 198. See also Coated zirconium

221

alternative tolerant fuel, 65 66 aluminum-based coatings for, 72 74 ceramic coatings, 69 71 channel shadow corrosion, 38f chromium coatings for, 71 72 cladding, 26t, 30 31 wall, 35 36 crud deposition on, 40 41 degradation, 29f fretting, 28 30 pervasive, 28 30 fabrication and implementation of zirconium-coated rods, 75 family of candidate coatings for, 68 69 hydrogen pickup by, 35 36 iodine stress corrosion cracking of, 36 37 iron chromium aluminum coatings, 72 74 irradiation damage of, 41 oxidation protection of coatings for, 67 68 PCMI of, 36 37, 36f shadow corrosion, 37 40, 40f silicon-based coatings, 74 75 waterside corrosion, 30 34 Zirconium dioxide (ZrO2), 30 34, 108 110 fuel rod system, 112 113 Zirconium silicide (ZrSi2), 74