Vous êtes sur la page 1sur 604

Comprehensive nuclear materials

EDITOR IN CHIEF
Rudy J. M. Konings
European Commission, Joint Research Centre,
Institute for Transuranium Elements, Karlsruhe, Germany

SECTION EDITORS
Todd R. Allen
Department of Engineering Physics, University of Wisconsin, Madison, WI, USA
Roger E. Stoller
Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA
Shinsuke Yamanaka
Division of Sustainable Energy and Environmental Engineering, Graduate School of Engineering, Osaka University,
Osaka, Japan

Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
225 Wyman Street, Waltham, MA 02451, USA

The following articles are US Government works in the public domain and not subject to copyright:
Radiation Effects in UO2
TRISO-Coated Particle Fuel Performance
Composite Fuel (cermet, cercer)
Metal Fuel-Cladding Interaction
No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic,
mechanical, photocopying, recording or otherwise without the prior written permission of the publisher
Permissions may be sought directly from Elseviers Science & Technology Rights Department in Oxford, UK: phone (44) (0) 1865 843830;
fax (44) (0) 1865 853333; email: permissions@elsevier.com. Alternatively you can submit your request online by visiting the Elsevier web
site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material
Notice
No responsibility is assumed by the publisher 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,
Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Catalog Number: 2011929343
ISBN (print): 978-0-08-056027-4
For information on all Elsevier publications
visit our website at books.elsevier.com
Cover image courtesy of Professor David Sedmidubsky, The Institute of Chemical Technology, Prague
Printed and bound in Spain
12 13 14 15 16 10 9 8 7 6 5 4 3 2 1

Editorial : Gemma Mattingley


Production: Nicky Carter

EDITORS BIOGRAPHIES
Rudy Konings is currently head of the Materials Research Unit in the
Institute for Transuranium Elements (ITU) of the Joint Research Centre
of the European Commission. His research interests are nuclear reactor
fuels and actinide materials, with particular emphasis on high temperature
chemistry and thermodynamics. Before joining ITU, he worked on nuclear
fuel-related issues at ECN (the Energy Research Centre of the Netherlands)
and NRG (Nuclear Research and Consultancy Group) in the Netherlands.
Rudy is editor of Journal of Nuclear Materials and is professor at the Delft
University of Technology (Netherlands), where he holds the chair of
Chemistry of the nuclear fuel cycle.

Roger Stoller is currently a Distinguished Research Staff Member in the


Materials Science and Technology Division of the Oak Ridge National
Laboratory and serves as the ORNL Program Manager for Fusion Reactor
Materials for ORNL. He joined ORNL in 1984 and is actively involved in
research on the effects of radiation on structural materials and fuels for
nuclear energy systems. His primary expertise is in the area of computational modeling and simulation. He has authored or coauthored more than
100 publications and reports on the effects of radiation on materials, as well
as edited the proceedings of several international conferences.

Todd Allen is an Associate Professor in the Department of Engineering


Physics at the University of Wisconsin Madison since 2003. Todds research
expertise is in the area of materials-related issues in nuclear reactors, specifically radiation damage and corrosion. He is also the Scientific Director for
the Advanced Test Reactor National Scientific User Facility as well as the
Director for the Center for Material Science of Nuclear Fuel at the Idaho
National Laboratory, positions he holds in conjunction with his faculty
position at the University of Wisconsin.

vi

Editors Biographies

Shinsuke Yamanaka is a professor in Division of Sustainable Energy and


Environmental Engineering, Graduate School of Engineering, Osaka
University since 1998. He has studied the thermophysics and thermochemistry of nuclear fuel and materials. His research for the hydrogen behavior in
LWR fuel cladding is notable among his achievements and he received
the Young Scientist Awards (1980) and the Best Paper Awards (2004) from
Japan Atomic Energy Society. Shinsuke is the program officer of Japan
Science and Technology Agency since 2005 and the visiting professor of
Fukui University since 2009, and he is also the associate dean of Graduate
School of Engineering, Osaka University since 2011.

PREFACE
There are essentially three primary energy sources for the billions of people living on the earths surface: the
sun, radioactivity, and gravitation. The sun, an enormous nuclear fusion reactor, has transmitted energy to the
earth for billions of years, sustaining photosynthesis, which in turn produces wood and other combustible
resources (biomass), and the fossil fuels like coal, oil, and natural gas. The sun also provides the energy that
steers the climate, the atmospheric circulations, and thus fuelling wind mills, and it is at the origin of
photovoltaic processes used to produce electricity. Radioactive decay of primarily uranium and thorium
heats the earth underneath us and is the origin of geothermal energy. Hot springs have been used as a source
of energy from the early days of humanity, although it took until the twentieth century for the potential of
radioactivity by fission to be discovered. Gravitation, a non-nuclear source, has been long used to generate
energy, primarily in hydropower and tidal power applications.
Although nuclear processes are thus omnipresent, nuclear technology is relatively young. But from the
moment scientists unraveled the secrets of the atom and its nucleus during the twentieth century, aided by
developments in quantum mechanics, and obtained a fundamental understanding of nuclear fission and fusion,
humanity has considered these nuclear processes as sources of almost unlimited (peaceful) energy. The first fission
reactor was designed and constructed by Enrico Fermi in 1942 in Chicago, the CP1, based on the fission of
uranium by neutron capture. After World War II, a rapid exploration of fission technology took place in the United
States and the Union of Soviet Socialist Republics, and after the Atoms for Peace speech by Eisenhower at the
United Nations Congress in 1954, also in Europe and Japan. A variety of nuclear fission reactors were explored for
electricity generation and with them the fuel cycle. Moreover, the possibility of controlled fusion reactions has
gained interest as a technology for producing energy from one of the most abundant elements on earth, hydrogen.
The environment to which materials in nuclear reactors are exposed is one of extremes with respect to
temperature and radiation. Fuel pins for nuclear reactors operate at temperatures above 1000  C in the center of
the pellets, in fast reactor oxide fuels even above 2000  C, whereas the effects of the radiation (neutrons, alpha
particles, recoil atoms, fission fragments) continuously damage the material. The cladding of the fuel and the
structural and functional materials in the fission reactor core also operate in a strong radiation field, often in a
dynamic corrosive environment of the coolant at elevated temperatures. Materials in fusion reactors are
exposed to the fusion plasma and the highly energetic particles escaping from it. Furthermore, in this
technology, the reactor core structures operate at high temperatures. Materials science for nuclear systems
has, therefore, been strongly focussed on the development of radiation tolerant materials that can operate in a
wide range of temperatures and in different chemical environments such as aqueous solutions, liquid metals,
molten salts, or gases.
The lifetime of the plant components is critical in many respects and thus strongly affects the safety as well as
the economics of the technologies. With the need for efficiency and competitiveness in modern society, there is
a strong incentive to improve reactor components or to deploy advanced materials that are continuously
developed for improved performance. There are many examples of excellent achievements in this respect. For
example, with the increase of the burnup of the fuel for fission reactors, motivated by improved economics and a
more efficient use of resources, the Zircaloy cladding (a ZrSn alloy) of the fuel pins showed increased
susceptibility to coolant corrosion, but within a relatively short period, a different zirconium-based alloy was
developed, tested, qualified, and employed, which allowed reliable operation in the high burnup range.

xxi

xxii

Preface

Nuclear technologies also produce waste. It is the moral obligation of the generations consuming the energy
to implement an acceptable waste treatment and disposal strategy. The inherent complication of radioactivity,
the decay that can span hundreds of thousands of years, amplifies the importance of extreme time periods in the
issue of corrosion and radiation stability. The search for storage concepts that can guarantee the safe storage
and isolation of radioactive waste is, therefore, another challenging task for materials science, requiring a close
examination of natural (geological) materials and processes.
The more than 50 years of research and development of fission and fusion reactors have undoubtedly
demonstrated that the statement technologies are enabled by materials is particularly true for nuclear
technology. Although the nuclear field is typically known for its incremental progress, the challenges posed
by the next generation of fission reactors (Generation IV) as well as the demonstration of fusion reactors will
need breakthroughs to achieve their ambitious goals. This is being accompanied by an important change in
materials science, with a shift of discovery through experiments to discovery through simulation. The progress
in numerical simulation of the material evolution on a scientific and engineering scale is growing rapidly.
Simulation techniques at the atomistic or meso scale (e.g., electronic structure calculations, molecular dynamics, kinetic Monte Carlo) are increasingly helping to unravel the complex processes occurring in materials
under extreme conditions and to provide an insight into the causes and thus helping to design remedies.
In this context, Comprehensive Nuclear Materials aims to provide fundamental information on the vast variety of
materials employed in the broad field of nuclear technology. But to do justice to the comprehensiveness of the
work, fundamental issues are also addressed in detail, as well as the basics of the emerging numerical simulation
techniques.
R.J.M. Konings
European Commission, Joint Research Centre,
Institute for Transuranium Elements, Karlsruhe, Germany
T.R. Allen
Department of Engineering Physics,
Wisconsin University, Madison, WI, USA
R. Stoller
Materials Science and Technology Division,
Oak Ridge National Laboratory, Oak Ridge, TN, USA
S. Yamanaka
Division of Sustainable Energy and Environmental Engineering,
Graduate School of Engineering, Osaka University, Osaka, Japan

FOREWORD
Nuclear materials denotes a field of great breadth and depth, whose topics address applications and facilities
that depend upon nuclear reactions. The major topics within the field are devoted to the materials science and
engineering surrounding fission and fusion reactions in energy conversion reactors. Most of the rest of the field
is formed of the closely related materials science needed for the effects of energetic particles on the targets and
other radiation areas of charged particle accelerators and plasma devices. A more complete but also more
cumbersome descriptor thus would be the science and engineering of materials for fission reactors, fusion
reactors, and closely related topics. In these areas, the very existence of such technologies turns upon our
capabilities to understand the physical behavior of materials. Performance of facilities and components to the
demanding limits required is dictated by the capabilities of materials to withstand unique and aggressive
environments. The unifying concept that runs through all aspects is the effect of radiation on materials. In this
way, the main feature is somewhat analogous to the unifying concept of elevated temperature in that part of
materials science and engineering termed high-temperature materials.
Nuclear materials came into existence in the 1950s and began to grow as an internationally recognized field
of endeavor late in that decade. The beginning in this field has been attributed to presentations and discussions
that occurred at the First and Second International Conferences on the Peaceful Uses of Atomic Energy, held in
Geneva in 1955 and 1958. Journal of Nuclear Materials, which is the home journal for this area of materials
science, was founded in 1959. The development of nuclear materials science and engineering took place in the
same rapid growth time period as the parent field of materials science and engineering. And similarly to the
parent field, nuclear materials draws together the formerly separate disciplines of metallurgy, solid-state
physics, ceramics, and materials chemistry that were early devoted to nuclear applications. The small priesthood of first researchers in half a dozen countries has now grown to a cohort of thousands, whose home
institutions are anchored in more than 40 nations.
The prodigious work, Comprehensive Nuclear Materials, captures the essence and the extensive scope of the
field. It provides authoritative chapters that review the full range of endeavor. In the present day of glance and
click reading of short snippets from the internet, this is an old-fashioned book in the best sense of the
word, which will be available in both electronic and printed form. All of the main segments of the field are
covered, as well as most of the specialized areas and subtopics. With well over 100 chapters, the reader finds
thorough coverage on topics ranging from fundamentals of atom movements after displacement by energetic
particles to testing and engineering analysis methods of large components. All the materials classes that have
main application in nuclear technologies are visited, and the most important of them are covered in
exhaustive fashion. Authors of the chapters are practitioners who are at the highest level of achievement
and knowledge in their respective areas. Many of these authors not only have lived through a substantial
part of the history sketched above, but they themselves are the architects. Without those represented here
in the author list, the field would certainly be a weaker reflection of itself. It is no small feat that so many
of my distinguished colleagues could have been persuaded to join this collective endeavor and to make the
real sacrifices entailed in such time-consuming work. I congratulate the Editor, Rudy Konings, and

xxiii

xxiv

Foreword

the Associate Editors, Roger Stoller, Todd Allen, and Shinsuke Yamanaka. This book will be an important
asset to young researchers entering the field as well as a valuable resource to workers engaged in the
enterprise at present.
Dr. Louis K. Mansur
Oak Ridge, Tennessee, USA

Permission Acknowledgments
The following material is reproduced with kind permission of Cambridge University Press
Figure 15 of Oxide Dispersion Strengthened Steels
Figure 15 of Minerals and Natural Analogues
Table 10 of Spent Fuel as Waste Material
Figure 21b of Radiation-Induced Effects on Microstructure
www.cambridge.org
The following material is reproduced with kind permission of American Chemical Society
Figure 2 of Molten Salt Reactor Fuel and Coolant
Figure 22 of Molten Salt Reactor Fuel and Coolant
Table 9 of Molten Salt Reactor Fuel and Coolant
Figure 6 of Thermodynamic and Thermophysical Properties of the Actinide Nitrides
www.acs.org
The following material is reproduced with kind permission of Wiley
Table 3 of Properties and Characteristics of SiC and SiC/SiC Composites
Table 4 of Properties and Characteristics of SiC and SiC/SiC Composites
Table 5 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 5 of Advanced Concepts in TRISO Fuel
Figure 6 of Advanced Concepts in TRISO Fuel
Figure 30 of Material Performance in Supercritical Water
Figure 32 of Material Performance in Supercritical Water
Figure 19 of Tritium Barriers and Tritium Diffusion in Fusion Reactors
Figure 9 of Waste Containers
Figure 13 of Waste Containers
Figure 21 of Waste Containers
Figure 11 of Carbide Fuel
Figure 12 of Carbide Fuel
Figure 13 of Carbide Fuel
Figure 4 of Thermodynamic and Thermophysical Properties of the Actinide Nitrides
Figure 2 of The UF system
Figure 18 of Fundamental Point Defect Properties in Ceramics
Table 1 of Fundamental Point Defect Properties in Ceramics
Figure 17 of Radiation Effects in SiC and SiC-SiC
Figure 21 of Radiation Effects in SiC and SiC-SiC
Figure 6 of Radiation Damage in Austenitic Steels
Figure 7 of Radiation Damage in Austenitic Steels
Figure 17 of Ceramic Breeder Materials
Figure 33a of Carbon as a Fusion Plasma-Facing Material
Figure 34 of Carbon as a Fusion Plasma-Facing Material
i

ii

Permission Acknowledgments

Figure 39 of Carbon as a Fusion Plasma-Facing Material


Figure 40 of Carbon as a Fusion Plasma-Facing Material
Table 5 of Carbon as a Fusion Plasma-Facing Material
www.wiley.com
The following material is reproduced with kind permission of Springer
Figure 4 of Neutron Reflector Materials (Be, Hydrides)
Figure 6 of Neutron Reflector Materials (Be, Hydrides)
Figure 1 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 3 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 4 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 5 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 6 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 7 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 8 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 9 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 10 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 11 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 12 of Properties and Characteristics of SiC and SiC/SiC Composites
Figure 22d of Fission Product Chemistry in Oxide Fuels
Figure 3 of Behavior of LWR Fuel During Loss-of-Coolant Accidents
Figure 14a of Irradiation Assisted Stress Corrosion Cracking
Figure 14b of Irradiation Assisted Stress Corrosion Cracking
Figure 14c of Irradiation Assisted Stress Corrosion Cracking
Figure 25a of Irradiation Assisted Stress Corrosion Cracking
Figure 25b of Irradiation Assisted Stress Corrosion Cracking
Figure 1 of Properties of Liquid Metal Coolants
Figure 5b of Fast Spectrum Control Rod Materials
Figure 3 of Oxide Fuel Performance Modeling and Simulations
Figure 8 of Oxide Fuel Performance Modeling and Simulations
Figure 10 of Oxide Fuel Performance Modeling and Simulations
Figure 11 of Oxide Fuel Performance Modeling and Simulations
Figure 14 of Oxide Fuel Performance Modeling and Simulations
Figure 5 of Thermodynamic and Thermophysical Properties of the Actinide Nitrides
Figure 51 of Phase Diagrams of Actinide Alloys
Figure 6 of Thermodynamic and Thermophysical Properties of the Actinide Oxides
Figure 7b of Thermodynamic and Thermophysical Properties of the Actinide Oxides
Figure 9b of Thermodynamic and Thermophysical Properties of the Actinide Oxides
Figure 35 of Thermodynamic and Thermophysical Properties of the Actinide Oxides
Table 11 of Thermodynamic and Thermophysical Properties of the Actinide Oxides
Table 13 of Thermodynamic and Thermophysical Properties of the Actinide Oxides
Table 17 of Thermodynamic and Thermophysical Properties of the Actinide Oxides
Figure 18 of Radiation Damage of Reactor Pressure Vessel Steels
Figure 7 of Radiation Damage Using Ion Beams
Figure 9b of Radiation Damage Using Ion Beams
Figure 28 of Radiation Damage Using Ion Beams
Figure 34 of Radiation Damage Using Ion Beams
Figure 35 of Radiation Damage Using Ion Beams
Figure 36d of Radiation Damage Using Ion Beams
Figure 37 of Radiation Damage Using Ion Beams
Table 3 of Radiation Damage Using Ion Beams

Permission Acknowledgments

Figure 5 of Radiation Effects in UO2


Figure 9a of Ab Initio Electronic Structure Calculations for Nuclear Materials
Figure 9b of Ab Initio Electronic Structure Calculations for Nuclear Materials
Figure 9c of Ab Initio Electronic Structure Calculations for Nuclear Materials
Figure 10a of Ab Initio Electronic Structure Calculations for Nuclear Materials
Figure 23 of Thermodynamic and Thermophysical Properties of the Actinide Carbides
Figure 25 of Thermodynamic and Thermophysical Properties of the Actinide Carbides
Figure 26 of Thermodynamic and Thermophysical Properties of the Actinide Carbides
Figure 27 of Thermodynamic and Thermophysical Properties of the Actinide Carbides
Figure 28a of Thermodynamic and Thermophysical Properties of the Actinide Carbides
Figure 28b of Thermodynamic and Thermophysical Properties of the Actinide Carbides
Figure 2 of Physical and Mechanical Properties of Copper and Copper Alloys
Figure 5 of Physical and Mechanical Properties of Copper and Copper Alloys
Figure 6 of The Actinides Elements: Properties and Characteristics
Figure 10 of The Actinides Elements: Properties and Characteristics
Figure 11 of The Actinides Elements: Properties and Characteristics
Figure 12 of The Actinides Elements: Properties and Characteristics
Figure 15 of The Actinides Elements: Properties and Characteristics
Table 1 of The Actinides Elements: Properties and Characteristics
Table 6 of The Actinides Elements: Properties and Characteristics
Figure 25 of Fundamental Properties of Defects in Metals
Table 1 of Fundamental Properties of Defects in Metals
Table 7 of Fundamental Properties of Defects in Metals
Table 8 of Fundamental Properties of Defects in Metals
www.springer.com
The following material is reproduced with kind permission of Taylor & Francis
Figure 9 of Radiation-Induced Segregation
Figure 6 of Radiation Effects in Zirconium Alloys
Figure 1 of Dislocation Dynamics
Figure 25 of Radiation Damage Using Ion Beams
Figure 26 of Radiation Damage Using Ion Beams
Figure 27 of Radiation Damage Using Ion Beams
Figure 4 of Radiation-Induced Effects on Material Properties of Ceramics (Mechanical and Dimensional)
Figure 7 of The Actinides Elements: Properties and Characteristics
Figure 20 of The Actinides Elements: Properties and Characteristics
Figure 18a of Primary Radiation Damage Formation
Figure 18b of Primary Radiation Damage Formation
Figure 18c of Primary Radiation Damage Formation
Figure 18d of Primary Radiation Damage Formation
Figure 18e of Primary Radiation Damage Formation
Figure 18f of Primary Radiation Damage Formation
Figure 1 of Radiation-Induced Effects on Microstructure
Figure 27 of Radiation-Induced Effects on Microstructure
Figure 5 of Performance of Aluminum in Research Reactors
Figure 2 of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 3 of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 5 of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 10a of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 10b of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 10c of Atomic-Level Dislocation Dynamics in Irradiated Metals

iii

iv

Permission Acknowledgments

Figure 10d of Atomic-Level Dislocation Dynamics in Irradiated Metals


Figure 12a of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 12b of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 12c of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 12d of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 16a of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 16b of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 16c of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 16d of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 16e of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 17a of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 17b of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 17c of Atomic-Level Dislocation Dynamics in Irradiated Metals
Figure 17d of Atomic-Level Dislocation Dynamics in Irradiated Metals
www.taylorandfrancisgroup.com

5.01

Corrosion and Compatibility

S. Lillard
Los Alamos National Laboratory, Los Alamos, NM, USA

2012 Elsevier Ltd. All rights reserved.

5.01.1

Theory

5.01.1.1
5.01.1.2
5.01.1.3
5.01.1.4
5.01.1.5
5.01.1.6
5.01.2
5.01.2.1
5.01.2.2
5.01.2.3
5.01.2.4
References

Introduction
Half Cell Reactions
Cell Potentials and the Nernst Equation
Reference Electrodes and Their Application to Nuclear Systems
The Thermodynamics of Corrosion from Room Temperature to the PWR
Kinetics of Dissolution and Passive Film Formation
Analytical Methods
Introduction
Potentiodynamic Polarization
Electrochemical Impedance Spectroscopy
MottSchottky Analysis

Abbreviations
BWR
CNLS

Boiling water reactor


Complex nonlinear least squares fitting of
the data
EC
Electrical equivalent circuit
EIS
Electrochemical impedance
spectroscopy
EPBRE External pressure-balanced reference
electrode
FFTF
Fast Flux Test Facility
HIC
Hydrogen-induced cracking
HIFER Hi-Flux Isotope Reactor
IG
Intergranular
LBE
Leadbismuth eutectic
PWR
Pressurized water reactor
SCC
Stress corrosion cracking
SS
Stainless steel

Symbols
A
ai
C
ci
CR
E
EW
Ecorr
f

Surface area
Activity of species i
Capacitance
Concentration of species i
Corrosion rate
Potential
Equivalent weight
Corrosion potential
Mass fraction

F
i
icorr
ji
k
L
M
MM
n
ND
Q
r
R
Rp
RV
S
t
ti
T
Vo
z
Z
Z0
Z00
jZj
b
ba
bc
d

2
2
3
4
6
8
10
10
11
12
14
16

Fugacity
Faradays constant
Current density
Corrosion current density
Square root of 1
Rate constant
Oxide thickness
Molecular weight
Metal cation
Number of electrons
Donor concentration
Reaction quotient
Rate of reaction
Gas constant
Polarization resistance
Solution resistance
Entropy of transport
Time
Transport number of species i
Temperature
Oxygen vacancy
Charge
Impedance
Real part of the impedance
Imaginary part of the impedance
Magnitude of the impedance
Symmetry factor
Anodic Tafel slope
Cathodic Tafel slope
Double layer thickness

Corrosion and Compatibility


0

DCp Change in standard partial molar heat


capacity
DE0 Standard reduction potential
DG Change in Gibbs energy
DG0 Standard Gibbs energy
DS0 Standard entropy change

Electronic charge
Permittivity of space
0
f
Applied potential
h
Overpotential
u
Phase angle
r
Material density
v
Frequency

5.01.1 Theory
5.01.1.1

Introduction

Mars Fontana identified eight forms of corrosion in


his book Corrosion Engineering1 and it is quite easy
to find examples of almost all of these in nuclear
reactors in both the primary and secondary cooling
water systems. For example, galvanic corrosion in
zirconiumstainless steel couples,2,3 crevice corrosion in tube sheets4 and former baffle bolts,5 and
pitting corrosion in alloy 600 steam generator
tubes.6,7 Perhaps the most infamous form of corrosion
observed in nuclear reactors is stress corrosion cracking (SCC), or environmental fracture, as we shall
refer to it here, which has numerous examples in
the literature. Environmental fracture includes both
intergranular SCC (IG), such as that which occurs in
austenitic stainless steel, and hydrogen-induced cracking (HIC), frequently observed in nickel base alloys.
Failure by one of these mechanisms results from an
interplay between stress, microstructure, and the environment (e.g., the electrochemical interface). The goal
of this chapter is not to address each of the corrosion
mechanisms outlined by Fontana individually, that will
be accomplished in the following chapters. Rather,
this chapter is meant to provide the reader with the
fundamental electrochemical theory necessary to
critically evaluate the data and discussions in the
corrosion chapters that follow.
In this section, we will review the fundamental
theory of the electrochemical interface. In the first
three subsections, we review Half Cell Reaction, Cell
Potentials and the Nernst Equation, and Reference
Electrodes in Nuclear Systems. In these sections, we
develop the theory necessary to understand the role of
electrochemical potential in environmental fracture

and corrosion mechanisms. For example, intergranular


stress corrosion cracking (IGSCC) is only observed at
potentials more positive than a critical value while HIC
is only observed at potentials more negative than a
critical value. In the remaining two sections, we review
the Thermodynamics from Room Temperature to the
pressurized water reactor (PWR) and Kinetics of Dissolution and Passive Film Formation. These sections
should help the reader to understand the role of the
passive film in the corrosion mechanism and the competition that occurs between film formation and metal
dissolution rate. As the fundamental role of irradiation
in corrosion and environmental fracture mechanism is
far from well established, in each section, we incorporate empirical irradiation data as examples and discuss
concepts that are more broadly important to nuclear
systems.
5.01.1.2

Half Cell Reactions

The electrochemical interface is characterized by an


electrode (in this case a metal such as a cooling pipe)
and an electrolyte (e.g., the cooling water in a reactor). While the bulk electrolyte contributes to variables such as solution chemistry and ohmic drop
(solution resistance is discussed later in this chapter),
it is the first nanometer of electrolyte that plays the
most important role in electrochemistry. In this short
distance, referred to as the electrochemical double
layer, a separation of charge occurs. It is this separation of charge that provides the driving force
(potential drop) for corrosion reactions. For example,
a 100 mV-applied potential across a typical double
layer will result in an electric field on the order of
106 V cm2. In the model proposed by Helmholtz,8
the double layer may be thought of as capacitor, with
positive charge on the metal electrode and the
adsorption of negatively charged cations on the solution side (Figure 1). The capacitance of the double
layer is equal to that in its electrical analog e0D/d,
where e0 is the permittivity of space, D is the dielectric, and d is the thickness of the layer. For most
electrochemical double layers, C is on the order of
106 F cm2.
Electrochemical reactions that take place in the
double layer are reactions in which a transfer of
charge (electrons) occurs. There are two different
types of cells in which electrochemical reactions
may occur9:
Electrolytic cells in which work, in the form of electrical
energy, is required to bring about a nonspontaneous reaction.

Corrosion and Compatibility

Metal

Because the system cannot store charge, the electrons


produced during the anodic reaction must be used.
This occurs at the cathode where typical reactions
may include oxygen reduction:

Excess negative
charge

Excess positive
charge

Bulk solution

Acid: O2 4H 4e ) 2H2 O

Base: O2 2H2 O 4e ) 4OH


+

II
III

or hydrogen reduction:
2H 2e ) H2

fmetal

IV

From eqns [I] and [III], the general corrosion of an


Fe surface in basic solution may then be written as:
2Fe O2 2H2 O ) 2FeOH2

fsolution
Double
layer
Figure 1 A diagram depicting the separation of charge at
the electrochemical double layer and the associated
potential drop (f).

H2O
2H+

ClH2

H2O
H+

H2O

ClH2O

Cl-

H+

Oxide
Fe

For any chemical reaction the driving force, the


Gibbs energy, may be written as10:

Figure 2 Diagram of what the anodic and cathodic


reactions may look like on an iron surface depicting the
separation of reactions and ionic conduction.

DG DG 0 RT lnQ

Voltaic cells in which a spontaneous reaction occurs


resulting in work in the form of electrical energy.
Electrolytic cells cover a fairly large number of
electrochemical reactions but may generally be
thought of as plating or electrolysis type reactions
and will not be treated here. Corrosion reactions are
voltaic cells and will be the focus of this chapter. As in
an electrolytic cell, voltaic cells are characterized by
two separate electrodes, an anode and a cathode. In
corrosion, reactions at the anode take the form of
metal dissolution, the formation of a soluble metal
cation:
Fe ) Fe2 2e

where Fe(OH)2 is the corrosion product.


An example of what the anodic and cathodic reactions on Fe electrode might look like is presented in
Figure 2. Though the anodic and cathodic reactions
occur at physically separate locations, as shown in
this figure, the reactions must be connected via an
electrolyte (aqueous solution). Figure 2 also suggests
that corrosion reactions are controlled by variables
such as mass transport (diffusion, convection, migration), concentration, and ohmic drop (resistivity of
the electrolyte). These variables will be considered in
our discussion of corrosion kinetics.
5.01.1.3 Cell Potentials and the Nernst
Equation

Fe2+

e-

V

I

1

where DG0 is the standard Gibbs energy, Q is the


reaction quotient equal to the product of the activities (assumed to obey Raoults Law for dilute solutions and, thus, equal to the concentration) of the
products divided by the reactants, R is the gas constant, and T is temperature. The electrical potential,
E, is related to the Gibbs energy of a cell by the
relationship:
nFE DG

2

where n is the number of electrons participating


in the reaction and F is Faradays constant. For the
reduction of hydrogen on platinum:
1

e Pt , H2g
Haq
2

VI

Corrosion and Compatibility

The reaction quotient, Q (starting conditions) becomes:


Q

fH2 1=2
H 

3

where fH2 is the fugacity of hydrogen gas. Substituting


eqns [2] and [3] into eqn [1], we find for the reduction
of hydrogen on platinum that:
"
#
RT fH2 1=2
0
4
ln
E DE
H 
F
where F is Faradays constant and DE0 is the standard
reduction potential for the reaction in eqn [VI]. Equation [10] is commonly referred to as the Nernst equation and defines the equilibrium reduction potential of
the half cell and is pH dependent. The Nernst equation
is commonly expressed in its generalized form as:
E DE 0

RT
lnQ 
F

5

5.01.1.4 Reference Electrodes and Their


Application to Nuclear Systems
In Equation [4], all of the parameters are easily
calculated with one exception, DE0. Therefore, we
define DE0 0 in eqn [4] for a set of specific parameters and refer to this cell as the standard hydrogen
electrode (SHE): H2 pressure of 1 atm, a pH 0, and
a temperature of 25  C. This provides a reference
from which we can calculate the standard potentials
for all other reduction reactions using eqn [5]. These
are referred to as standard reduction potentials and a
few examples are provided in Table 1.
While the SHE is the accepted standard, from
a practical standpoint, this reference electrode is
difficult to construct and maintain. As such, experimentalists have taken advantage of a number of other
reduction reactions to construct reference electrodes
for laboratory use. The reaction selected typically
depends on the application. One common reference
electrode is the silversilver chloride electrode
(Ag/AgCl) which is based on the reduction of Ag
in a solution of potassium chloride:
Ag e , Ags

VII

Ag Cl , AgCls

VIII

and the overall reaction being:


Ags Cl , AgCls e

XI

Table 1
Standard reduction potentials for several reactions important to the nuclear power industry
Reduction reaction

Standard reduction
potential (V)

Au3 3e Au
Cl2 2e 2Cl
O2 4H 4e 2H2O
Ag e Ag
Fe3 e Fe2
O2 2H2O 4e 4OH
AgCl e Ag Cl
2H 2e H2 (NHE)
Ni2 2e Ni
Fe2 2e Fe
Cr3 3e Cr
Zr4 4e Zr
Al3 3e Al
Li e Li

1.52
1.36
1.23
0.80
0.77
0.4
0.22
0.0
0.25
0.44
0.74
1.53
1.66
3.04

The Nernst equation for eqn [XI] is equal to:




aAgCl 
RT
0
E DEAg=AgCl
ln
aAg aCl 
F
0
DEAg=AgCl
 lnaCl 

6

where aCl is the activity of chloride and for which


the concentration (mCl ) in molal (mol kg1) is frequently substituted. In the corrosion lab, the reference electrode is constructed by electrochemically
depositing an AgCl layer onto a silver wire. This
wire is then placed in a glass capillary filled with a
solution of potassium chloride the concentration of
which then defines the cell potential (aCl in eqn [6]).
One end of the capillary is sealed using a porous frit
(typically a porous polymer) that acts as a junction
between the solution of the reference electrode and
the environment of the corrosion experiment.
While the Ag/AgCl reference electrode construction described above is straightforward for the lab,
there are several obstacles that must be overcome
before it can be used in a nuclear power plant setting,
namely, radiation flux, pressure, and temperature. As it
turns out, the primary impact of ionizing radiation
on laboratory reference electrodes relates to damage
of the cotton wadding and polymer frits used in their
construction and no change in cell potential occurs.11
As such, two approaches based on the Ag/AgCl
reference electrode have been used to measure electrode potential in nuclear power reactors. In the first
approach, an internal reference electrode operates in
the same high-temperature environment as the reactor. In this case, one must consider the solubility of

Corrosion and Compatibility

Sapphire lid
AgCl pellet

Rulon adapter
Compression fitting

Sapphire
container

Pt cap
Ni wire
Alumina
insulators

Ceramic to
metal braze
Restrainer

Ag/AgCl

Kovar
TIG weld
304SS

Seal

Coaxial
cable
Figure 3 Diagram of an internal Ag/AgCl reference
electrode used in BWRs. Top end is inserted into the cooling
loop, while the coaxial cable provides electrical connection.
Reprinted from Indig, M. E. In 12th International Corrosion
Congress, Corrosion Control for Low-Cost Reliability; NACE
International: Houston, TX, 1993; p 4224, with permission
from NACE International.

Ag/Cl complexes that form as a function of temperature eqn [6].12,13 That is, reactions in addition to eqns
[VIII] and [XI] must be considered. From a construction viewpoint, the internal reference electrode
consists of a silver chloride pellet on a platinum foil
(Figure 3).14 External potential measurement is made
via contact with a nickel wire which is connected to an
electrometer via a coaxial cable. The electrode is
housed in a sapphire tube that is sealed via a porous
sapphire cap. In this configuration, there is no internal
electrolyte per se. Upon placing the electrode in a
boiling water reactor (BWR), the porous cap allows
the cooling water to penetrate the electrode and the
potential is determined from eqn [6] and the solubility
of AgCl in high purity water as a function of
temperature.15
In the second approach, an external pressurebalanced reference electrode is used (EPBRE).
In the EPBRE, the reference electrode is maintained
at room temperature and pressure and the corresponding constants are used in eqn [6]. The
reference is connected to the high-temperature environment via a nonisothermal salt bridge sealed with a
porous zirconia plug (Figure 4).16 As a result of this
configuration, the EPBRE is not susceptible to

Compression fitting
1/4 NPT
Pure water or
0.01 M KCl
Glass wick
1/4 OD SS
tube

Heat-shrinkable
PTFE tube

SS nut
Rulon sleeve
Zirconia plug

Figure 4 A diagram of a pressure-balanced reference


electrode is used in BWRs. Bottom of figure is sealed into
pressure vessel via compression fitting while Ag/AgCl
electrode (top) remains at room temperature and pressure.
Reprinted from Oh, S. H.; Bahn, C. B.; Hwang, I. S.
J. Electrochem. Soc. 2003, 150, E321, with permission from
The Electrochemical Society.

potential deviations owing to the solubility of AgCl


and its complexes as a function of temperature. However, the temperature gradient between the reactor
and the reference electrode results in a junction
potential that must be subtracted from eqn [6]. The
corresponding thermal liquid junction potential
(ETLJ)17 is given by:


1 T2 tM SM tCl SCl
dT
7
ETLJ 

zM
zCl
F T1
where t, S, and z represent the transport number, the
entropy of transport, and the charge on the cation,
respectively. The symbol M in eqn [7] represents the
metal in the chloride salt, MCl, and is commonly Li, Na,
or K. In addition to ETLJ, there is also the isothermal
liquid junction potential, EILJ, which arises due to the
differences in cation and anion mobilities through
the porous frits and the fact that the electrolyte in the
external reference (typically KCl) is vastly different
from the reactor cooling water in which it is immersed17:

RT T2 ti
EILJ 
dlnai 
8
F T1 zi

Corrosion and Compatibility

where the subscript i denotes a species that may be


transported through the zirconia plug and for a nuclear
power reactor may include species ions as Ag, Cl,
H, OH, K, and B(OH4).
It has been shown that both ETLJ and EILJ each
increase by as much as 0.15 V over the temperature
range of 25350  C. The result is a decrease in the
measured potential of 0.30 V at 350  C. While these
junction potentials can be calculated and used to
correct eqn [7], it has been shown that there is
some deviation at higher temperatures (>200  C)
and an experimental fitting procedure is the preferred method for calibration of the reference
electrode.

reactions with two soluble species

 3 
Fe
11
Fe2 Fe3 e E 0 0:771 0:059log
Fe2

solubility of iron and its oxides


Fe Fe2 2e E 0 0:440 0:0295logFe2

2Fe2 3H2 O Fe2 O3 6H 2e


E 0 0:728  0:177pH  0:059logFe2

An atlas of electrochemical equilibria has been created by M. Pourbaix for metals in aqueous solution at
room temperature.18 This atlas contains potential
pH diagrams, so-called Pourbaix diagrams, which
define three equilibrium thermodynamic domains
for metals in aqueous solutions: immunity, passivity,
and corrosion. Immunity is defined as the state where
the base metal is stable while corrosion is defined
as the formation of soluble metal cations and passivity the formation of a stable oxide film. Pourbaixs
derivation requires that the values of the standard
chemical potential, m0, for all of the reacting substances are known for the standard state at the
temperature and pressure of interest. For chemical
reactions at room temperature, the equilibrium conditions are defined by the relationship18:
P 0
nm
9
logK
5708
and for electrochemical reactions at room temperature (Table 1) equilibrium is defined by:
P 0
nm
0
10
E
96485n
where K is the equilibrium constant for the reaction,
m0 is in Joules per mole, v is the stoichiometric coefficient for the species, n is the number of electrons, 5708
is a conversion constant equal to RT/(log10e) where
T is temperature (298.15 K) and R the ideal gas constant (8.314472 J (K mol)1), and 96 485 is Faradays
constant in J (mol V)1.
As an example of these diagrams, consider the
ironwater system and the solid substances Fe,
Fe3O4, and Fe2O3. Pourbaix18 defined the relevant
equations for this system as:

13


Fe 2H2 O HFeO
2 3H 2e

E 0 0:493  0:089pH 0:0295logHFeO


2

5.01.1.5 The Thermodynamics of Corrosion


from Room Temperature to the PWR

12

14


3HFeO
2 H Fe3 O4 2H2 O 2e

E 0 1:819 0:029pH  0:088logHFeO


2

15

reaction of two solid substances


3Fe 4H2 O Fe3 O4 8H 8e
E 0 0:085  0:059pH

16

2Fe3 O4 H2 O 3Fe2 O3 H 2e


E 0 0:221  0:059pH

17

An example of a simplified Pourbaix diagram for Fe


at room temperature based on the reactions in eqns
[11][17] is presented in Figure 5, where Eq. [12]
corresponds to figure line 23, [13] to line 28, [14] to
line 24, [15] to line 27, [16] to line 13 and [17] to line
17. Note that Eq. [11] is the boundary between Fe2
and Fe3 and was not drawn in the original figure.
In addition to the lines separating the domains for Fe,
Pourbaix diagrams will typically include the domains
associated with water stability (oxidation and reduction) represented by the dashed lines marked a and b
in Figure 5. Upon inspection of this diagram one
would conclude what is know from experience with
Fe: that iron is passive in alkaline solutions and at
higher applied potentials owing to oxide film formation while at more acidic solutions Fe is susceptible to
corrosion owing to Fe2. It is worth noting again that
these potentialpH domains are defined solely by the
thermodynamic stability of the species within them
and these diagrams do not consider kinetics which
will be addressed later in this chapter. This is important as while a species/reaction may be thermodynamically stable it may be kinetically hindered.
While the use of Pourbaix diagrams to characterize room temperature corrosion reactions is

Corrosion and Compatibility

1.5

Fe3+

1.5

Fe3+
20

20

1.0

1.0
b

Fe2O3

Fe2+

EH2 (200 C) (V)

28

EH (25 C) (V)

0.5

a
26

-0.5

17
Fe3O4

23

0.5

Fe2+

26

13

-1.0

HFeO-2

17

Fe3O4

23
27
24

Fe

Fe2O3
a

-0.5

13

-1.0

28

Fe

HFeO22-

-1.5

-1.5
0

10

15

pH

widespread, these diagrams and the method for generating them as presented thus far cannot be used at
the higher temperatures associated with nuclear power
reactors. This is due to the lack of standard potentials
at elevated temperature as required by eqns [9] and
[10] (e.g., the application of Table 1 to higher temperature). In the absence of these high-temperature thermodynamic data, Townsend19 used an extrapolation
method introduced by Criss and Coble (the correspondence principle). The method allows for empirical
entropy data of ionic species at 25  C to be extrapolated to higher temperatures. In this method, the standard Gibbs free energy is calculated from the
relationship:

T

T
250

250

DC p T dlnT

DC p T dT
18
0

where DS is the standard entropy change and DC p is


the change in standard partial molar heat capacity. The
potentialpH diagram for the FeH2O system and the
solid substances Fe, Fe3O4, and Fe2O3 at 200  C calculated by Townsend is presented in Figure 6. In comparison with the diagram at 25  C (Figure 5), the
Fe2O3 and Fe3O4 regions are extended to lower pH
and potentials. As a result the area associated with
corrosion at lower solution pH is decreased. However,
0

24

10

15

pH

Figure 5 A simplified potentialpH diagram for the


FeH2O system and the solid substances Fe, Fe3O4, and
Fe2O3 at 25  C based on the reactions in eqns [11][17].
Reprinted from Townsend, H. E. Corrosion Sci. 1970,
10, 343, with permission from Elsevier.

DDG 0  DT DS 0 250

29

27

Figure 6 The potentialpH diagram for the FeH2O


system and the solid substances Fe, Fe3O4, and Fe2O3 at
200  C. Most dramatic influence of increased temperature is
the presence of a large region of soluble species (corrosion)
at high pH. Reprinted from Townsend, H. E. Corrosion Sci.
1970, 10, 343, with permission from Elsevier.

the most notable change in the diagram is at high pH


where the area associated with corrosion owing to the
soluble HFeO
2 has increased dramatically. The Criss
and Coble method is limited, however, to the
150200  C range and, to extend the Pourbaix to the
temperatures of power reactors, Beverskog used a
HelgesonKirkhamFlowers model to extend the
heat capacity data to 300  C.20
Thus far, we have described a method for generating electrochemical equilibria diagrams and regions
of passivity, corrosion, and immunity for pure metals
from 25 to 300  C. From an engineering standpoint,
we would like to know this information for structural
alloys such as austenitic stainless steels and super
nickel alloys. At temperatures near 25  C, the predominant oxide responsible for passivity is Cr2O3 and it is
sufficient to rely only on the Cr potentialpH diagram
for alloys with a high Cr content. However, at higher
temperatures other oxides form such as Fe(Fe,Cr)2O4,
(Cr,Fe)2O3, (Cr,Fe,Ni)3O4, and (Cr,Fe,Ni)2O3, and it is
desirable to know the thermodynamic stability of the
alloy. Beverskog has developed the ternary potential
pH diagrams for the FeCrNiH2OH2 system for
temperatures up to 300  C using heat capacitance data
and the revised HelgesonKirkhamFlowers model
described above.21 However, FeCrNi phases lack
thermodynamic data and the ternary oxides were,

Corrosion and Compatibility

npH
2

H2CrO4(aq)
HCrO4-

Potential (VSHE)

CrO42Cr(OH)2+
0
NiCr2O4(cr)
Cr2+

FeCr2O4(cr)

-1

Cr2O3(cr)
Cr(cr)
-2

6
pH300 C

10

Figure 7 PotentialpH diagram for chromium species in


FeCrNi at 300  C. Concentration of aqueous species is
106 molal. Reprinted from Beverskog, B.; Puigdomenech, I.
Corrosion 1999, 55, 1077, with permission from NACE
International.

thus, not considered. The diagrams assumed that the


metallic elements in the alloy had unit activity, that is,
equal amounts of iron, chromium, and nickel. An
example of the potentialpH diagram for chromium
species in FeCrNi at 300  C and aqueous species
with a concentration of 106 molal is presented in
Figure 7. Unlike the Fe diagram, where the presence
of soluble HFeO2
2 species increased with temperature
(Figure 6), the diagram for Cr in FeCrNi is dominated by passive region where the stable oxides of
Cr2O3, FeCr2O4, and NiCr2O4 are formed.
5.01.1.6 Kinetics of Dissolution and
Passive Film Formation
The study of dissolution kinetics, corrosion rate,
attempts to answer the question: What are the relationships that govern the flow of current across a
corroding interface and how is this current flow
related to applied potential? Consider the anodic
dissolution of a metal with an activation barrier
equal to G1a nFE (eqn [2]). If we increase the driving
force (potential) from its equilibrium condition, E0, to
a new value, f, the new barrier is given by the
relationship22:

decreasing the barrier, that is, not all of the applied


potential is dropped across the electrochemical double layer. The rate (ra) of this reaction is expressed in
the same, Arrhenius, form as for chemical reactions:


DGa
20
ra ia =nF ka co exp 
RT
where ia is the anodic current density, ka is the rate
constant, co is the concentration of oxidized species,
and DGa is the change in free energy for the anodic
reaction. Substituting G2a  G1a in eqn [20] for DGa in
eqn [19], we express the anodic reaction rate as22:


1  bnF

21
ia nFka cR exp
RT
where  (the overpotential) represents a departure
from equilibrium and is equal to f  E0. We can derive
a similar expression for the cathodic reaction22:


bnF

22
ic nFkc co exp 
RT
where ic is the cathodic current density, kc is the rate
constant, and co is concentration of oxidized species.
Combining eqns [21] and [22] and rearranging them,
we can write an expression for the total current, i:
 



1  bnF
bnF
  exp 

23
i io exp
RT
RT
where io is the exchange current density and is equal to
b b
nFkcc1b
o ka cR . This expression is commonly referred
to as the ButlerVolmer equation.
To apply eqn [23] to corrosion reactions, we need
to be able to relate  to the corrosion potential, Ecorr ,
that is, as it stands the ButlerVolmer equation is
derived for equilibrium conditions. Returning to our
definition of the overpotential  f  E0, by both
subtracting and adding Ecorr from the right side of
this definition, inserting the resulting expression back
into eqn [23] and rearranging we find22:


1  bF
Ecorr  Ea
icorr ia exp
RT


bF
ic exp  Ecorr  Ec
25
RT

19

For small applied potentials around Ecorr , the Stern


Geary approximation of eqn [24] is used23:


ba bc
f  Ecorr
26
i 2:303icorr
ba bc

where b is the symmetry factor and reflects the fact


that not all of the increase in potential goes to

where ba and bc are defined as the anodic and


cathodic Tafel slopes (discussed in Section

Ga2 Ga1  1  bnF f  E0

0.4

0.1

0.35

0
Current density (A m-2)

Potential (V vs. SCE)

Corrosion and Compatibility

0.3
0.25
0.2
Beam on at 100 nA
~540 s

0.15
0.1

500

1000
Time (s)

1500

2000

5.01.2.2) having units of volts and are empirical factors related to the symmetry factor by the relationships22:
RT
1  bnF

bc 2:303

RT
bnF

-0.1
-0.2
Increasing radiation flux
Increasing cathodic reaction rate

-0.3
-0.4
-0.5

Figure 8 Influence of proton irradiation on the Ecorr of a


SS 304L electrode in dilute sulfuric acid, pH 1.6. The
increase is caused by the production of water radiolysis
products.

ba 2:303

Beam = 35 na
Beam = 62 na
Beam = 100 na

27
28

As it relates to the nuclear power industry, eqn [24] not


only relates the corrosion rate (icorr) to the applied
potential, f, but it can also help us to rationalize
other processes such as the influence of water radiolysis
products on corrosion rate. For example, it is generally
observed that ionizing radiation (g, neutron, proton,
etc.) increases Ecorr potential (Figure 8) and corrosion
rate at Ecorr .24 The flux of ionizing radiation on the
cooling water results in radiolysis, the breaking of
chemical bonds. During the course of water radiolysis,
a wide variety of intermediate products are formed,
such as O2, eaq, and the OH radical.25 The vast majority of these species have very fast reaction rates so that
the end result is a handful of stable species. These
stable products are typically oxidants, such as O2, H2,
and H2O2. That is, these products readily consume
electrons (eqns [II][IV]) and increase cathodic reaction rate (Figure 9). From eqn [25] we see that an
increase in the cathodic reaction rate, ic, necessarily
results in an increase in corrosion rate, icorr , consistent
with the observation described.
While the development of dissolution kinetics is
straightforward, the kinetics associated with passive

-0.6
-0.2

-0.1

0
0.1
0.2
0.3
Potential (V vs. SCE)

0.4

0.5

Figure 9 Influence of proton irradiation on the cathodic


reactions on a Au electrode in dilute sulfuric acid, pH 1.6.
The increase is caused by the production of water radiolysis
products.

film formation and breakdown are less well understood yet equally as important to our understanding
of corrosion mechanisms. One such example is the
case of localized corrosion where the probability for
a pit to transition from a metastable to a stable state is
governed by the ability of the active surface to repassivate. Another example is the initiation of SCC where
passive film rupture results in very high dissolution
rates and, correspondingly, crack advance rate which is
controlled by the activation kinetics described above as
the bare metal dissolves.2628 During the propagation
stage of SCC, the crack tip must propagate faster than
(1) the oxide film can repassivate the surface and (2)
the corrosion rate on the unstrained crack sides so that
dissolution of the walls does not result in blunt notch.
To evaluate the role of repassivation kinetics
in SCC and corrosion mechanisms in general, investigators set about measuring three critical experimental parameters, namely film: ductility,29,30 bare
surface dissolution rates,3133 and passive film formation rates34,35 for various alloys. Each of these techniques involves the depassivation of a metal electrode
using a tensile frame or a nano-indenter (in the case
of ductility studies) or scratching/breaking an electrode (bares surface current density and repassivation
studies) and measuring the resulting current transient
as a function of time. An example of a current transient for SS 304L in chloride solution is presented
in Figure 10. The surface was under potentiostatic
control and was bared using a diamond scribe. The
data were collected using a high-speed oscilloscope.

10

Corrosion and Compatibility

7.0  104
6.0  104

Current (A)

5.0  104

td

tr

0.002 0.004 0.006 0.008


Time (s)

4.0  104
3.0  104
2.0  104
1.0  104
0.0
0.01

0.012

Figure 10 Scratch test current transient from a SS 304L


electrode in 0.1 M NaCl. The transient is characterized by a
growth period, td, and a repassivation period, tr.

The transient is characterized by two separate processes, anodic dissolution and repassivation represented by td and tr in Figure 10. To analyze the
repassivation rates, the period tr is typically fit to an
expression and evaluated as a function of solution pH
or electrode potential. The most prolific work in this
field is probably on the alloy SS 304L. For this alloy, it
has been proposed that the kinetics of film growth are
controlled by ion migration under high electric
field.3638 The kinetics of high-field film growth
were first proposed by Cabrera and Mott39 to obey
the kinetic relationship:
 
BV
29
i Aexp
L
where i is the current density, V is the voltage, L is the
oxide thickness, and A and B are constants.

5.01.2 Analytical Methods


5.01.2.1

Introduction

In this section, we will review the principle analytical


methods used to probe the electrochemical interface.
In the Section 5.01.2.2 Potentiodynamic Polarization, we discuss linear polarization resistance and
the practical application of corrosion kinetics, eqns
[25] and [26]. In that section, we also describe the
salient points of the anodic polarization curve. In the
Section 5.01.2.3 Electrochemical Impedance Spectroscopy, we introduce an ac method for interrogating
the electrochemical interface. This technique is

probably the most versatile experimental method


available to scientists and researchers. As it relates
to nuclear reactors, this technique has the ability to
subtract out the contribution of the solution resistance to polarization resistance measurements which,
if not accounted for in highly resistive cooling water
measurements will result in nonconservative corrosion rates. In the final section, we introduce a more
seldom used technique, MottSchottky analysis.
While this is by no means a common experimental
method, it provides a conduit for the reader to
become familiar with defects in the oxide film, their
transport and ways to quantify it. This has particular
interest here as ionizing radiation may promote corrosion rates by increasing transport of these defects
through the passive oxide film.
Regretfully, the scope of this chapter is limited,
and we are not able to discuss the step-by-step details
of the experimental methods that are used to make
corrosion measurements. A comprehensive guide
to experimental methods in corrosion has been
published by Kelly et al.40 as well as Marcus and
Mansfeld41 while a more broad description of electrochemical methods has been published by Bard and
Faulkner.42 The reader is also encouraged to become
familiar with the equipment that is used to make
electrochemical measurements and a good introductory chapter on this topic has been presented by
Schiller.43 The most important instrument is, no
doubt, the potentiostat. While this instrument is the
cornerstone of corrosion science, it does have its pitfalls including bandwidth limitations and the potential for ground loop circuits when used in conjunction
with other equipment such as load frames, autoclaves
and cooling loops. The latter can be overcome using
proper instrumentation such as potentiostat with a
floating ground, or isolation amplifiers.
To investigate the influence of the neutron flux
on corrosion rates and mechanisms, real-time in-situ
corrosion measurements are often made in-reactor
or at neutron facilities such as Oak Ridge National
Labs Hi-Flux Isotope Reactor (HIFER) and Argonne
National Labs Fast Flux Test Facility (FFTF). Alternately, neutron damage can be simulated using ion
beams. As it relates to ion irradiation, this method
provides opportunities for studying the interaction
of the components of reactor environments (radiation, stress, temperature, aggressive media) that are
not possible with in-reactor or neutron irradiation
facilities. For a full discussion of this topic, see
Chapter 1.07, Radiation Damage Using Ion
Beams. To summarize these experiments, controlled

11

Corrosion and Compatibility

environmental cells are coupled to accelerator beamlines to study the interaction of the environment and
irradiation on structural materials. Corrosion at the
substrateenvironment interface is studied in real
time by numerous electrochemical techniques
including those described below. With respect to
dose, light ions can be used to reach doses up
to 10 dpa in several days. However, the depth of
penetration is low (tens or micrometers), which puts
unrealistic limitations on electrochemical cell construction. Heavy ion irradiation can reach several
hundred displacements per atom in a matter of days
but the penetration depth is much less.
5.01.2.2

Potentiodynamic Polarization

During our development of ButlerVolmer reaction


kinetics, we introduced the SternGeary approximation and defined Tafel slopes within the context of the
symmetry factor without much further explanation.
To understand the empirical source of the Tafel
slopes, we rearrange eqn [26] to define the polarization resistance Rp in units of O cm2:
Rp

ba bc
DE

2:303icorr ba bc Di

30

The Tafel slopes can be obtained from a plot of


potential as a function of the logarithm of current
density as shown for the cathodic curve in Figure 11
where bc has units of volts. A similar anodic plot can
be generated to obtain ba. It is important to realize
that these slopes are frequently not equal as they are
related to separate mechanisms on the electrode surface. From a plot of both the anodic and cathodic
Tafel slopes, we can also obtain icorr (Figure 11).
With knowledge of icorr , ba, and bc, the polarization
iOH2
EH

20

10
Slope =

Applied current
curve

Ecorr
+

20

20

iapp(cathodic)

icorr

E
iOM

h(mv)

40
H+

/H2

resistance can then be determined from eqn [30]. For


small voltage perturbations, the slope of a plot of
applied potential (DE ) versus the change in current
density (Di ) is equal to Rp (Figure 12), often referred
to as the linear polarization resistance as the slope is
only linear for small voltage perturbations around Ecorr .
Errors not accounted for in eqn [30] include
uncompensated solution resistance (RO) and choosing
a scan rate that is too fast. From a reactor standpoint,
the value of RO in the cooling water or a simulant
where the resistivity is high may be a significant
contribution to the measured Rp and, therefore, result
in a nonconservative corrosion rate, that is, the calculated corrosion rate will be too low. For a complete
standardized method of conducting potentiodynamic
polarization resistance measurements and data analysis, the reader is referred to the appropriate ASTM
standards G3, G59, and G102.4446
Potentiodynamic polarization curves can also be
used to assess corrosion rate as well as determine if a
material is passive, active, or susceptible to pitting
corrosion in a given environment. Consider the curve
in Figure 13. It plots the applied potential as a
function of the log of the absolute value of the current
density. The corrosion potential and corrosion current density are shown at the intersection of the
cathodic and anodic curves. Mass loss (m, in grams)
for a given period of exposure can then be

DE
Diapp

40
iapp(anodic)

-10

Tafel region

- EM/M+
log iapplied

Figure 11 Diagram depicting the cathodic polarization of


an electrode near Ecorr. Tafel extrapolation showing the
determination of icorr is also presented. Reprinted from
Fontana, M. G. Corrosion Engineering; McGraw Hill:
New York, 1986, with permission from McGraw Hill.

-20
h(mv)
Figure 12 Diagram depicting the linear polarization for an
electrode about Ecorr. Slope, DE/Di, is equal to the
polarization resistance. Reprinted from Fontana, M. G.
Corrosion Engineering; McGraw Hill: New York, 1986, with
permission from McGraw Hill.

12

Corrosion and Compatibility

positive hysteresis in the reverse portion of the curve


that is not observed in the case of solution oxidation
or transpassivity.

Epit
Potential

Erepass

5.01.2.3 Electrochemical Impedance


Spectroscopy
Eflade

ba
ipass

icorr

bc

Ecorr

Log current density


Figure 13 Diagram depicting the potentiodynamic
polarization of an electrode far from Ecorr. Relevant
potentials and currents are defined in the text.

determined from the corresponding icorr using


Faradays Law, for a pure metal40:
icorr t EW
31
FA
where t is time in seconds, icorr has units of A (cm2),
A is surface area, and EW is equivalent weight and is
equal to molecular weight (M) divided by the number of electrons in the reaction. Similarly, for an alloy:
m

EW P

1
n
i
i fi =Mi

32

where the subscript i denotes the alloying element of


interest and f is the mass fraction of that element
in the alloy. From Faradays law, it is also possible to
calculate corrosion rate, the penetration depth owing
to corrosion over a period of time in units of mm
year1 (CR):
CRmm year1

3:27  103 icorr EW


r

33

where r is the material density in g cm3.


Other critical parameters in Figure 13 include
the Flade potential (EFlade) which marks the critical
potential necessary for passive film formation, the
passive current density (ipass), the pitting potential
(Epit), and the repassivation potential (Erepass). With
respect to EFlade, this active to passive transition is not
observed for materials that are spontaneously passive
in a given environment such as SS in a BWR. In
such a case, the current would be limited by the
film dissolution current (ipass). As it relates to Epit,
the onset of localized corrosion is characterized by a
sharp increase in current at a given potential. As such,
materials that are more susceptible to pitting have
lower Epit values. Pitting is also characterized by a

While potentiodynamic polarization is typically considered a destructive technique, that is, it alters the
surface of the corrosion sample, electrochemical
impedance spectroscopy (EIS) is a powerful nondestructive technique for obtaining a wealth of data
including Rp.47 Further, this technique has the ability
to subtract out the uncompensated solution resistance (RO) from the measurement which, for highly
resistive reactor cooling water environments, is a
significant advantage.
In EIS, a small sinusoidal voltage perturbation
(10 mV) is applied across the electrode interface
over a broad frequency range (mHz to MHz). By
measuring the transfer function of the applied voltage
to the system current, the system impedance may be
obtained. For corrosion systems, the impedance (Z)
is a complex number and may be represented in
Cartesian coordinates by the relationship:
Zo Z0 Z00

34

where o is the applied frequency in radians, Z0 is the


real part of the impedance, and Z00 is the imaginary
part of the impedance, and the magnitude of the
impedance jZj is given by:
q
35
jZj Z0 2 Z00 2
In its simplest form, the electrochemical interface can
be thought of as an electrical equivalent circuit (EC):
a resistor (R) with an impedance Z(o) R and a
capacitor (C) with Z(o) 1/joC where j is the square
root of 1.48 Thus, the impedance of a resistor is
purely real and independent of frequency while the
impedance of a capacitor is purely imaginary and
inversely proportional to frequency. With respect to
the electrochemical interface, the polarization resistance is in parallel with the double layer capacitance
(Cdl, owing to adsorption of charged anions/cations
in the electrolyte). These two components act in
series with the solution resistance, RO as seen in the
EC in Figure 14. This circuit is referred to as a
simple Randles circuit and represents an ideal interface. Commonly, however, Cdl does not behave as an
ideal capacitor and its impedance is better represented by the expression

13

Corrosion and Compatibility

-6  104

Rp
RW

-5  104
-4  104

Figure 14 Simplified Randles equivalent circuit of an


electrochemical interface where Rp is the polarization
resistance, Cdl is the double layer capacitance, and RO is the
geometric resistance associated with the solution
resistance.

Z (W)

Cd1

-3  104

-2  104
-1  104

105

1  104 2  104 3  104 4  104 5  104 6  104


Z (W)

Rp + RW

Figure 16 Nyquist format for data in Figure 15 where


Rp 5  104 O, Cdl 4  106, and RO 200 O.

104
|Z| (W)

Zo 1=C j oa
103

RW
102
10-3 10-2 10-1 100 101 102
(a)
Frequency (Hz)

103

104

105

0
-10

Q ()

-30
-40
-50

-70

wmax

-80
-90
10-3 10-2 10-1 100 101 102
(b)
Frequency (Hz)

where a is typically found to be between 0.5 and 1.


The element that represents this behavior is known
as a constant phase element (Ccpe).49
The response of a simple Randles circuit as a
function of frequency is shown in Figure 15(a) and
15(b). These plots are referred to as the Bode magnitude plot (eqn [35]) and the Bode phase plot50 where
the phase angle, y, is equal to:
 00 
Z
37
y tan1
Z0
This phase angle is a result of the double layer
capacitance where the current leads the applied ac
voltage perturbation as is the case for pure capacitors.
The parameters Rp and RO may be determined graphically from the Bode magnitude plot as shown in
Figure 15(a) while Cdl is determined from the graphical parameter omax (converted to radians, 2pf) in
Figure 15(b) and the relationship40:
1 q
1 Rp =RO
38
Cdl
omax Rp

-20

-60

36

103

104

105

Figure 15 (a) Bode magnitude and (b) Bode phase plots.


Data were generated from an electrical equivalent of a
simplified Randles circuit (Figure 13) where Rp 5  104 O,
Cdl 4  106, and RO 200 O.

Alternately, the data may be presented using the


Nyquist format which plots the imaginary impedance
as a function of the real impedance as seen in
Figure 16 (sometimes referred to as a ColeCole
plot). As graphical analysis is somewhat imprecise,
commercially available complex nonlinear least
squares fitting of the data (CNLS) is commonly
used to obtain these parameters (Figure 17).

14

Corrosion and Compatibility

|Z| exper.
|Z| fit

102

ROX

RW

-10

|Z| (W m2)

-30

Q exper.
Q fit

-40
-50

100

Q (degrees)

-20
101

-60

COX
Figure 19 Equivalent circuit of an oxide-covered metal in
liquid PbBi eutectic (LBE) where Rox is the resistance of the
passive film, Cox is the double layer capacitance of the
oxide, and RO is the geometric resistance associated with
the LBE.

-70
10-1
10-3

10-2

10-1

100

101

102

-80
103

Frequency (Hz)
Figure 17 Bode magnitude and Bode phase plots for SS
304L during proton irradiation in a pressurized deionized
cooling water loop at 125  C showing experimental data and
complex nonlinear least squares fit of equivalent circuit in
Figure 14.

Corrosion rate (mm year 1)

6.0

5.0

4.0

3.0
Photons
Neutrons
2.0

1.0

Protons

20

100 120
40
60
80
Flux (particles m2 per proton)

140

Figure 18 Corrosion rate as a function of particle flux


for a SS 304L electrode during proton irradiation in a
pressurized deionized cooling water loop at 125  C.

EIS has been used successfully to investigate the


passive films on Zr alloys,5153 SS 304L,54 and nickel
base alloys55 in environments that simulate reactor
cooling water systems. In addition, it has also been
used to measure the real-time corrosion rates of
materials during irradiation. Lillard et al.56,57 measured the corrosion rate of materials as a function of
immersion time in a deionized water cooling loop
during proton irradiation. In that work, radiationhardened probes made from Alloy 718, SS 304L,
and Al 60 601 were exposed to a proton beam at

various current densities. The impingement of the


beam on the probes resulted in a mixed neutron,
photon, and proton flux. It was shown that corrosion
rate was almost linear with photon and neutron flux
as compared to proton flux where anomalies existed
at intermediate fluxes (Figure 18). The data were
ultimately used to extrapolate lifetimes for accelerator materials.
In other studies, EIS has also been applied to
investigate passive films on metals in sodium (Na)cooled and leadbismuth (LBE) systems that simulate reactor environments. The equivalent circuit
(EC) used to model the data is similar to the simplified Randles circuit; however, in this case, there is no
electrochemical double layer, only the impedance
and capacitance associated with the oxide
(Figure 19). In this EC, Rox is the dc resistance of
the oxide, Cox is the oxide capacitance, and RO is the
geometric resistance associated with the liquid metal
(Na or LBE). In one such study, Isaacs investigated
the capacitance of anodized films on Zr in liquid Na.
In that study, it was shown that both Rox and Cox were
a function of Na temperature between 50 and 400  C.
Lillard et al.58 reported similar trends for HT-9 in
LBE. At constant temperature, Rox and Cox for HT-9 in
LBE were a function of immersion time (Figure 20).
The data were used to calculate oxide thickness as a
function of time. In addition, Rox was related to ionic
transport through the film and corrosion rates were
calculated using Wagners oxidation theory. Upon irradiation in a proton beam, this rate fell even further.59
Additional information about leadeutectic coolants
may be found in Chapter 5.09, Material Performance in Lead and Lead-bismuth Alloy.
5.01.2.4

MottSchottky Analysis

The formation and growth of passive oxide films is


driven, fundamentally, by interfacial reactions and
defect, electron and ion transport processes. Yet the

Corrosion and Compatibility

105

ROX

15

5.5  1010

101

5.0  1010

104

Prior to irradiation

Fit

Beam on

Fit

101

101

1 C-2 (cm4 F-2)

ROX (W cm2)

Rp = 0.9 t1.7
102

COX (nF cm2)

100
103

4.5  1010
4.0  1010
3.5  1010

102
COX

100
101
50

50
100
Immersion time (h)

150

3.0  1010
103
200

2.5  1010
-0.2

0.2

0.4

0.6

0.8

Potential (V vs. SCE)

Figure 20 Impedance capacitance for the oxide on


an HT-9 steel as a function of immersion in liquid PbBi
eutectic. Reprinted from Lillard, R. S.; Valot, C.;
Hanrahan, R. J. Corrosion 2004, 60, 1134, with
permission from the author.

Figure 21 MottSchottky plots for a SS 304L electrode in


dilute sulfuric acid, pH 1.6. Before and after proton
irradiation. Reprinted from Lillard, R. S.; Vasquez,
G. J. Electrochem. Soc. 2008, 155, C162, with permission
from the author.

nature and relative importance of these processes


are still far from being understood. Key to oxide
growth, and, therefore, passivation, is the mobility of
these defects, specifically vacancies. Under irradiation, however, the defect properties of the oxide
are undoubtedly changed, and the extent of corrosion is related both to the microstructure and transport properties of defects. One way to probe the
transport properties of the oxide film is by using
MottSchottky theory.
Returning to our discussion of EIS above, the
oxide capacitance may be obtained at high frequency
from the relationship

depends on what type of semiconductor the oxide is


(p vs. n) and this effects the sign of the slope.
Lillard used MottSchottky analysis to examine
the influence of proton irradiation on defect generation and transport in the oxide film on SS 304L.61
The passive film on SS 304L is an n-type semiconducting oxide and the major defect is the oxygen
vacancy which acts as an electron donor. According
to the Point Defect Model62 for oxide growth, oxygen
vacancies are produced at the metalfilm interface by
the injection of a metal atom into the oxide lattice:

Z00 o 1=oC or C 1=oZ00 olim

39

By measuring Z00 as a function of applied dc voltage


at the high frequency limit (olim) and calculating the
film capacitance from eqn [39], it is possible to evaluate donor concentration (ND) in the oxide film from
the well-known MottSchottky relationship:


1
kT
40

2=ee
N
U

U
0
D
fb
C2
q
where e0 is the permittivity of space and is equal to
8.854  1014 F cm1, e is the electronic charge and
is equal to 1.602  1019 C, U is the applied potential
in V, Ufb is the flatband potential in V, kT/q is equal to
25 mV at 25  C, and ND is the donor concentration
(oxygen vacancies) in cm3.60 ND is the slope of a plot
of 1/C2 versus applied potential. The type of defect

m ! MM x=2Vo xe

41

where m is a metal atom, MM is a metal cation, Vo is a


oxygen vacancy, e is an electron, and x is the stoichiometric coefficient. In pH 1.6 H2SO4, it was found
that the cation vacancy concentration in the oxide
increased from 2.94  1021 in the absence of irradiation to 3.41  1021 during irradiation (Figure 21).
It was proposed that the film on SS 304L is composed
of an inner Cr-rich p-type semiconductor and an
outer Fe-rich n-type semiconductor. This bilayer
film results in a pn junction where the two layers
meet. On the p side of the junction, there is a surplus
of holes, while on the n side of the junction, a surplus
of electrons exists. The energy bands are such that
it is uphill (an increase in energy) for electrons
moving across the junction (from the n side to the
p side). On a related topic, a discussion of defects in
bulk oxides can be found in Chapter 1.02, Fundamental Point Defect Properties in Ceramics.

16

Corrosion and Compatibility

References
1.
2.
3.
4.
5.

6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.

28.
29.
30.
31.
32.

Fontana, M. G. Corrosion Engineering; McGraw Hill:


New York, 1986.
Cox, B. J. Nucl. Mater. 2005, 336, 331.
Lysel, G.; Nystrand, A. C.; Ullberg, M. J. ASTM Int. 2005, 2,
355.
Abella, J.; Balachov, I.; Macdonald, D. D.; Millet, P. J.
Corrosion Sci. 2002, 44, 191.
Scott, P. M.; Meunier, M. C.; Deydier, D.; Silvestre, S.;
Trenty, A. In American Society for Testing and Materials
Special Technical Publication; Kane, R. D., Ed.; ASTM:
Conshohocken, PA, 2000; p 210.
Hur, D. H.; Choi, M. S.; Lee, D. H.; Song, M. H.; Han, J. H.
Corrosion 2006, 62, 905.
Hwang, S. S.; Kim, H. P.; Kim, J. S. Corrosion 2003,
59, 821.
Helmholtz, H. Pogg. Ann. 1853, LXXXIX, 211.
Whitten, K. W.; Gailey, K. D. General Chemistry; Saunders
College: Philadelphia, PA, 1981.
Castellan, G. W. Physical Chemistry; Addison-Wesley:
Reading, MA, 1983.
Danielson, M. J. Corrosion 1995, 51, 450.
Macdonald, D. D. Corrosion 1978, 34, 75.
Oijerholm, J.; Forsberg, S.; Hermansson, H. P.; Ullberg, M.
J. Electrochem. Soc. 2009, 156, P56.
Indig, M. E. In 12th International Corrosion Congress,
Corrosion Control for Low-Cost Reliability; NACE
International: Houston, TX, 1993; p 4224.
Indig, M. E.; Nelson, J. L. Corrosion 1991, 47, 202.
Oh, S. H.; Bahn, C. B.; Hwang, I. S. J. Electrochem. Soc.
2003, 150, E321.
Lvov, S. N.; Macdonald, D. D. J. Electroanal. Chem. 1995,
403, 25.
Pourbaix, M. Atlas of Electrochemical Equilibria in
Aqueous Solutions; NACE/Cebelcor: Houston, TX, 1974.
Townsend, H. E. Corrosion Sci. 1970, 10, 343.
Beverskog, B.; Puigdomenech, I. Corrosion Sci. 1996,
38, 2121.
Beverskog, B.; Puigdomenech, I. Corrosion 1999, 55,
1077.
Prentice, G. Electrochemical Engineering Principles;
Prentice Hall International Series; Prentice-Hall: Upper
Sadddle River, NJ, 1991.
Stern, M.; Geary, A. L. J. Electrochem. Soc. 1957, 104, 56.
Marsh, G. P.; Taylor, K. J.; Bryan, G.; Worthington, S. E.
Corrosion Sci. 1986, 26.
Lewis, M. B.; Hunn, J. D. J. Nucl. Mater. 1999, 265, 325.
Hoar, T. P.; Hines, J. G. J. Iron Inst. 1956, 182, 124.
Diegle, R. B.; Boyd, W. K. In Stress Corrosion Cracking
The Slow Strain-Rate Technique; Ugiansky, G. M.,
Payer, J. H., Eds.; ASTM: West Conshohocken, PA,
1979; p 26.
Ford, F. P. Corrosion 1996, 52, 375.
Bubar, S. F.; Vermilyea, D. A. J. Electrochem. Soc. 1966,
113, 892.
Bubar, S. F.; Vermilyea, D. A. J. Electrochem. Soc. 1967,
114, 882.
Hoar, T. P.; Scully, J. C. J. Electrochem. Soc. 1964,
111, 348.
Hoar, T. P.; West, J. M. Nature 1958, 181, 835.

33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.

57.
58.
59.
60.
61.
62.

Frankel, G. S.; Jahnes, C. V.; Brusic, V. V.; Davenport, A. J.


J. Electrochem. Soc. 1995, 142, 2290.
Ford, F. P.; Burstein, G. T.; Hoar, T. P. J. Electrochem.
Soc. 1980, 127, 1325.
Marshal, P.; Burstein, G. T. Corrosion Sci. 1983, 23, 1219.
Burstein, G. T.; Davenport, A. J. J. Electrochem. Soc.
1989, 136, 936.
Burstein, G. T.; Marshall, P. I. Corosion Sci. 1984, 24, 449.
Davenport, A. J.; Burstein, G. T. J. Electrochem. Soc.
1990, 137, 1496.
Cabrera, N.; Mott, N. F. Rep. Prog. Phys. 1948, 12, 163.
Kelly, R. G.; Scully, J. R.; Shoesmith, D. W.; Buchheit, R. G.
Electrochemical Techniques in Corrosion Science and
Engineering; Marcel Dekker: New York, 2003.
Marcus, P.; Mansfeld, F. Analytical Methods in Corrosion
Science and Engineering; Taylor & Francis: New York,
2006.
Bard, A. J.; Faulkner, L. R. Electrochemical Methods:
Fundementals and Applications; Wiley: New York, 2001.
Schiller, C. A. In Analytical Methods in Corrosion Science
and Engineering; Marcus, P., Mansfeld, F., Eds.; Taylor &
Francis: New York, 2006; p 361.
ASTM, G 3. Standard practice for conventions applicable
to electrochemical measurements in corrosion testing;
ASTM International; 1989.
ASTM, G 102. Standard practice for calculation of
corrosion rates and related information from
electrochemical measurements; ASTM International; 1989.
ASTM, G 59. Standard test method for conducting
potentiodynamic polarization resistance measurements;
ASTM International; 1997.
MacDonald, J. R. Impedance Spectrocopy; Wiley:
New York, 1987.
Mansfeld, F. In Analytical Methods in Corrosion Science
and Engineering; Marcus, P., Mansfeld, F., Eds.; Taylor &
Francis: Boca Raton, FL, 2006; p 463.
Hsu, C. H.; Mansfeld, F. Corrosion 2001, 57, 747.
Mansfeld, F. Corrosion 1988, 44, 558.
Ai, J.; Yingzi, C.; Uriquidi-Macdonald, M.; Macdonald, D. D.
J. Nucl. Mater. 2008, 379, 162.
Barberis, P.; Frichet, A. J. Nucl. Mater. 1999, 273, 182.
Nagy, G.; Kerner, Z.; Battistig, G.; Csordas, A. P.;
Balogh, J. J. Nucl. Mater. 2001, 297, 62.
Kim, Y. J. Corrosion 2000, 56, 389.
Fulger, M.; Ohai, D.; Mihalache, M.; Pantiru, M.;
Malinovschi, V. J. Nucl. Mater. 2009, 385, 288.
Lillard, R. S.; Gac, F.; Paciotti, M.; et al. In Effects
Radiation on Materials; American Society for Testing and
Materials Special Technical Publication #1045;
Rosinski, S. T., Grossbeck, M. L., Allen, T. R., Kumar, A. S.,
Eds.; ASTM: West Conshohocken, PA, 2001.
Lillard, R. S.; Willcutt, G. J.; Pile, D. L.; Butt, D. P. J. Nucl.
Mater. 2000, 278, 277.
Lillard, R. S.; Valot, C.; Hanrahan, R. J. Corrosion 2004, 60,
1134.
Lillard, R. S.; Paciotti, M.; Tcharnotskaia, V. J. Nucl. Mater.
2004, 335, 487.
Sikora, E.; Sikora, J.; Macdonald, D. D. Electrochim. Acta
1996, 41, 283.
Lillard, R. S.; Vasquez, G. J. Electrochem. Soc. 2008, 155,
C162.
Macdonald, D. D. J. Electrochem. Soc. 1992, 139, 3434.

5.02

Water Chemistry Control in LWRs

C. J. Wood
Electric Power Research Institute, Palo Alto, CA, USA

2012 Elsevier Ltd. All rights reserved.

5.02.1

Introduction

18

5.02.2
5.02.2.1
5.02.2.2
5.02.2.3
5.02.2.4
5.02.2.5
5.02.3
5.02.3.1
5.02.3.2
5.02.3.3
5.02.3.4
5.02.4
5.02.4.1
5.02.4.2
5.02.4.3
5.02.4.4
5.02.5
5.02.6
References

BWR Chemistry Control


Evolution of BWR Chemistry Strategies
Mitigating Effects of Water Chemistry on Degradation of Reactor Materials
Radiation Field Control
Fuel Performance Issues
Online Addition of Noble Metals
PWR Primary Water Chemistry Control
Evolution of PWR Primary Chemistry Strategies
Materials Degradation
PWR Radiation Field Control
Fuel Performance
PWR Secondary System Water Chemistry Experience
Evolution of PWR Secondary Chemistry Strategies
Chemistry Effects on Materials Degradation of SGs
Control of Sludge Fouling of SGs
Lead Chemistry
Chemistry Control for FAC in BWRs and PWRs
Water Chemistry Control Strategies

19
19
20
23
26
27
27
27
29
33
35
37
37
40
43
44
45
45
46

Abbreviations
AO

AVT

BOP
BRAC

BWR
CGR
CRUD
DMA
DZO
EBA
ECP

Axial offset, referring to localized


flux depression in reactor core
caused by buildup of boroncontaining deposits. Originally
called AOA for axial offset anomaly.
All-volatile treatment, suing
ammonia for pH control in steam
generators
Balance of plant
BWR radiation and control,
referring to designated standard
points in BWR reactors for
radiation field measurements
Boiling water reactor
Crack growth rate
Corrosion product deposits on
fuel element surfaces
Dimethylamine
Depleted zinc oxide (BWRs)
Enriched boric acid (PWRs)
Electrochemical corrosion
potential

ETA
FAC
GE

HWC
HWC-L
HWC-M
IGA
IGSCC
LWR
MOX
MPA
MRC
NDE
NMCA
NWC
OD IGA/SCC

Ethanolamine
Flow-assisted corrosion
General electric, the vendor for
BWRs in the United States and
some other countries
Hydrogen water chemistry
HWC (low) with 0.20.5 ppm
hydrogen
HWC (moderate) with 1.62.0 ppm
hydrogen
Intergranular attack
Intergranular stress corrosion
cracking
Light water reactor
Mixed oxide fuel
3-methoxypropylamine
Molar ratio control (PWR
secondary side)
Nondestructive examination
Noble metal chemical addition
Normal water chemistry (BWRs)
Outside diameter IGA/SCC in
steam generator tubes

17

18

Water Chemistry Control in LWRs

OLNC
OTSG
PAA
PbSCC
PWR
PWSCC
SCC
SG
SHE

On-line noble chemistry


Once through steam generator
Poly acrylic acid
Lead assisted stress corrosion
cracking
Pressurized water reactor
Primary water stress corrosion
cracking
Stress corrosion cracking
Steam generator
Standard hydrogen electrode (for
ECP measurements)

5.02.1 Introduction
Other chapters of this comprehensive describe the various degradation processes affecting the structural materials used in the construction of nuclear power plants
(see Chapter 5.04, Corrosion and Stress Corrosion
Cracking of Ni-Base Alloys; Chapter 5.05, Corrosion and Stress Corrosion Cracking of Austenitic
Stainless Steels; and Chapter 5.06, Corrosion and
Environmentally-Assisted Cracking of Carbon and
Low-Alloy Steels). This chapter describes the influence
of water chemistry on corrosion of the most important
materials in light water reactors (LWRs). In particular,
alloys susceptible to intergranular attack (IGA) and stress
corrosion cracking (SCC) are significantly impacted
by water chemistry, most notably, sensitized 304 stainless steel in boiling water reactors (BWRs) and nickelbased alloys in pressurized water reactors (PWRs).
Excellent water quality is essential if material
degradation is to be controlled. In the early days of
nuclear power plant operation, impurities in the
coolant water were a major factor in causing excessive
corrosion. Chlorides and sulfates are particularly
aggressive in increasing intergranular stress corrosion
cracking (IGSCC) and other corrosion processes.
Transient increases of impurities in the coolant that
occur during fault conditions (e.g., condenser leaks
and ingress of oil or ion exchange resins) proved to be
particularly damaging. Thus, water chemistry was
traditionally regarded as a key cause of material degradation. Initial efforts to improve water quality
brought about a slow but steady reduction in impurities through improved design and operation of purification systems. Not only were the average
concentrations of impurities reduced over time, but
the frequency and magnitude of impurity spikes
from transient fault conditions were also diminished.

However, excellent water chemistry alone was not


sufficient to control corrosion. Hence, programs to
modify water chemistry were introduced, including
minimizing oxygen to reduce the electrochemical
corrosion potential (ECP) in BWRs, and oxygen and
pH control in PWRs. More recently, additives to
further inhibit the corrosion process have been developed and are now in widespread use. As a result,
water chemistry advances are now an important part
of the overall operating strategy to control material
degradation.
Primary system water chemistry also affects fuel
performance through the deposition of corrosion
products on fuel pin surfaces, and influences radiation fields outside the core. Core uprating through
increased fuel duty has reduced margins for tolerating corrosion products (CRUD) on BWR fuel pin
surfaces. In PWRs, increasing fuel cycle duration
has increased the challenge of controlling pH within
the optimum range. At the same time, regulatory
limits on worker radiation exposure are tending to
be tightened worldwide, putting pressure on the
operators to reduce radiation dose rates. Successful
operation of PWR steam generators (SGs) and the
remainder of the secondary system demand strict
water chemistry control in secondary side systems if
corrosion problems are to be avoided.
Other operating parameters also influence the
optimization process, for example, life extension (to
60 years) has emphasized the importance of controlling
degradation of circuit materials. Therefore, although
control of structural material degradation remains the
highest priority, water chemistry must be optimized
between the sometimes-conflicting requirements affecting other parts of the reactor.
Advances in water chemistry have enabled plant
operators to respond successfully to these technical
challenges, and the overall performance has steadily
improved in recent years.1 Plant-specific considerations sometimes influence or indeed limit the options
for controlling water chemistry, so we see different
chemistry specifications at different plants. This is especially true internationally and significant differences
between countries are noted. The US industry started
developing water chemistry guidelines 2530 years
ago, and these now provide the technical basis for
guidelines in many other countries. The early editions
of these guidelines presented impurity specifications
and required action if limits were exceeded. When
advanced water chemistries were developed and qualified, the guidelines evolved into a menu of options
within an envelope of specifications that should not be

Water Chemistry Control in LWRs

exceeded. Guidance is now provided on how to select


a plant-specific water chemistry strategy.2
The basis for water chemistry control was discussed in detail by Cohen.3 The remainder of this
chapter describes more recent water chemistry developments for BWRs, PWR primary systems, and
PWR secondary systems including SGs, with a
short section on flow-assisted corrosion (FAC) in
both BWRs and PWRs.

5.02.2 BWR Chemistry Control


5.02.2.1 Evolution of BWR Chemistry
Strategies
BWR water chemistry has to be optimized between
the requirements to minimize material degradation,
avoid fuel performance issues, and control radiation
fields. These factors are depicted in Figure 1,4 which
also includes the main chemistry changes involved in
the optimization process.
Plant-specific considerations sometimes influence
or indeed limit the options for controlling water chemistry, so we see different chemistry specifications at
different plants. This is especially true internationally
and significant differences in chemistry strategies
between countries are noted. Design features are an
important reason for these different chemistry regimes,
to which must be added the effects of different operational strategies in recent years. For example, a key
issue facing BWRs in the United States concerns
IGSCC of reactor internals, as discussed in other
chapters. The occurrence of IGSCC resulted in the

Clad corrosion
crud deposition:
Limits on
feedwater zinc

Impurity control:
Monitoring/analysis
required

implementation of hydrogen water chemistry, with or


without noble metal chemical addition (NMCA), to
ensure that extended plant lifetimes are achieved.
German plants use 347 stainless steel, which is less
susceptible to IGSCC than sensitized 304 stainless
steel used originally in US-designed plants. Some
Swedish and Japanese plants have replaced 304 stainless steel reactor internals with 316 nuclear grade
material to minimize potential problems, as this material is less susceptible to IGSCC. As a result, many of
these plants continue to use oxygenated normal water
chemistry, whereas all US plants control IGSCC
through the use of hydrogen water chemistry (HWC)
with or without normal metal chemical addition to
improve the efficiency of the hydrogen in reducing
ECP. Second, BWRs in United States undoubtedly
have greater cobalt sources than plants in most other
countries, despite strong efforts to replace cobalt
sources. This resulted in higher out-of-core radiation
fields, leading all US plants to implement zinc injection
to control fields, whereas only a small number of plants
of other designs use zinc. Third, the move to longer fuel
cycles and increased fuel duty at US plants, while
having major economic benefits, has led to new constraints on chemistry specifications in order to avoid
fuel performance issues.
Figure 2 depicts the changing chemistry strategies over the past 30 years, showing the focus on
improving water quality in the early 1980s and the
move to educing chemistry to control IGSCC, which
in turn resulted in increased radiation fields, subsequently controlled by zinc injection.

Materials
degradation
and mitigation

Water
chemistry
guidelines

Fuel
performance

Chemistry
control issues

Figure 1 Boiling water reactor chemistry interactions.

19

BWR internals
IGSCC, IASCC:
HWC or NMC
required

Radiation
exposure

Radiation fields
crud bursts:
Zinc required

20

Water Chemistry Control in LWRs

Increasing concerns about core internals cracking led to the need to increase hydrogen injection
rates, which in turn resulted in the introduction
of NMCA to reduce operating radiation fields from
N-16. Figure 3 shows the rate of implementation of
HWC, zinc and NMCA, and online noble metal
addition (OLNC). The rationale and implications of
these developments are discussed in greater detail in
subsequent sections.
The goal for BWRs is therefore to specify chemistry
regimes that, together with the improved materials
used in replacement components (e.g., 316 nuclear
grade stainless steel), will ensure that the full extended
life of the plants will be achieved without the need for
further major replacements. At the same time, radiation dose rates, and hence worker radiation exposure,
must be closely controlled, and fuel performance must
not be adversely affected by chemistry changes.

The first requirement of plant chemistry is to maintain high-purity water in all coolant systems, including
the need to avoid impurity transients, which are
beyond the scope of this paper. The performance of
all plants has improved steadily over the years, as
shown by the trend for reactor water conductivity for
GE-designed plants, given in Figure 4. This figure
shows that conductivity now approaches the theoretical minimum for pure water. In fact, deliberately added
chemicals, such as zinc (discussed in the following
section), account for much of the difference between
measured values and the theoretical minimum.
The conductivity data are consistent with the
reactor water concentrations for sulfate and chloride.
In fact, sulfate is the most aggressive impurity from
the viewpoint of IGSCC, and much effort has gone
into reducing it.
5.02.2.2 Mitigating Effects of Water
Chemistry on Degradation of Reactor
Materials

1977: Neutral,
oxygenated water

Corrosion, radiation buildup issues

1980s: Purer is better

IGSCC was first observed in small bore piping using


sensitized 304 stainless steel fairly soon after BWRs
started operation. Laboratory studies showed that
impurities increased IGSCC rates, and in fact water
quality in BWRs gradually improved in the early
1980s. However, the same studies found IGSCC in
high-purity oxygenated water typical of good BWR
operations. The key parameter affecting IGSCC was
found to be ECP, as shown in Figure 5. In this laboratory test, the change from oxidizing conditions
typical of normal water chemistry (NWC) operation

Chemistry guidelines

Late 1980s1990s:
HWC, zinc

Controlling IGSCC,
radiation buildup

2000s: Noble metal


chemical addition

Core internals cracking


control with lower fields

Promising new option

20062008:
Online Noblechem

Figure 2 Evolution of Boiling water reactor chemistry


options from 1977 to 2008.

40

Number of BWRs

35

Zn injection

NMCA

HWC (no NMCA)

OLNC

30
25
20
15
10
5
0
1983

1988

1993

1998

2003

2008

Figure 3 Implementation of zinc injection, hydrogen water chemistry, noble metals chemical addition, and online noble
metal at US boiling water reactors.

21

Water Chemistry Control in LWRs

0.40
0.35
EPRI action level 1

Conductivity ( S cm1)

0.30
0.25
0.20
0.15
0.10
0.05
Theoretical conductivity limit, 25 C

0.00
1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

Figure 4 Boiling water reactor mean reactor water conductivity at US boiling water reactor.

250

0.4950

0.4945
200

2.7 108 mm s1
1 106
mm s1

150

0.4935

0.4930

0.4925

100
CT2 #7-304SS 4 dpa
Constant load, 19 ksiin.

Dissolved O2

Outlet cond: 0.30 S cm1

50

Inlet cond: 0.27 S cm1 Na2SO4

0.4920

0.4915
1488

Dissolved oxygen (ppb)

Crack length (in.)

0.4940

1508

1528

1548

1568

1588

0
1608

Test time (h)


Figure 5 Laboratory results showing the effect of reducing oxygen concentration on crack growth of 304 stainless steel.

to reducing conditions greatly reduced the rate of


crack growth.
Furthermore, hydrogen injection was effective at
reducing the ECP in BWRs, as shown in Figure 6.
In this figure, it can be seen that crack growth rates
(CGR) for Alloy 182 were low in hydrogen water

chemistry (HWC), but increased greatly when the


plant reverted to normal water chemistry (NWC).
These results indicated that continuous hydrogen
injection was required to fully mitigate cracking.
Examination of extensive inspection data from several
plants indicated that no IGSCC was observed with an

22

Water Chemistry Control in LWRs

901.00
900.00

HWC
ECP = 510 mV (SHE)

NWC
ECP = +110 mV (SHE)

Crack length

174 miles year-1

HWC

< 5 miles year-1

899.00
898.00
897.00
< 5 miles year -1

Alloy 182

896.00
895.00
800

900

1000

1100

1200

1300
Time (h)

1400

1500

1600

1700

Figure 6 Effect of hydrogen water chemistry on crack growth of Alloy 182.

ECP of 230 mV or lower, using a standard hydrogen


electrode (SHE). This is the basis for the 230 mV
requirement used by US plants for IGSCC control.
In BWRs, the radiation field in the core decomposes
water to hydrogen and oxygen species, most of which
immediately recombine back to water. But some
remain as oxygen or hydrogen peroxide, because some
hydrogen is stripped into the steam phase before it can
recombine. These same radiolysis reactions cause
hydrogen to react with oxygen or peroxide to reduce
ECP. These reactions occur mainly in the downcomer,
and relatively low hydrogen concentrations are effective at lowering ECP in out-of-core regions of the
system. More than half the BWRs in the United States
adopted low hydrogen injection rates of 0.20.5 ppm
(called HWC-L), which, coupled with the replacement
of recirculation piping using 316 stainless steel, mitigated IGSCC of recirculation piping.
In the 1990s, concerns about the cracking of core
internals increased, but the low concentrations of
hydrogen used to protect out-of-core regions were
not sufficient to reduce ECP enough to mitigate
IGSCC of in-core materials, because of the radiolysis
of water occurring in the core. As a result, it was
necessary to increase hydrogen concentrations to
1.62.0 ppm to lower the in-core ECP sufficiently to
provide protection in the reactor vessel (termed
HWC-M for moderate concentrations of hydrogen).
Although this approach was effective in protecting
core internals, it also increased radiation fields in the
steam side of the circuit, including the turbines, as a
result of carryover of nitrogen-16 under reducing
chemistry. (Under the oxidizing conditions of NWC,
most of the N-16 remains in the water as soluble

species such as nitrate, and only a small percent is


transported with the steam.) In some plants, local
shielding of turbine components has reduced the
impact of the gamma radiation to acceptable levels,
but the projected 46-fold increase did in fact curtail
plans for increased hydrogen injection rates at many
plants. Note that these N-16 radiation fields are a
problem only when the plant is at power, as rapid
decay occurs at shutdown because of the short halflife of N-16. (By contrast, out-of-core radiation fields
from Cobalt-60 persist after shutdown and impact on
maintenance work during outages.)
NMCA was developed to increase the efficiency
of hydrogen in BWR cores, to avoid high N-16 fields.
In this process, a nanolayer of platinum rhodium is
deposited on the wetted surfaces of the reactor. These
treated surfaces catalyze the hydrogen redox reaction, converting oxygen back to water. When the
addition of hydrogen to the feedwater raises the
molar ratio of H2 to O2 to 2 or higher, the ECP of
the treated surfaces drops to the hydrogen/oxygen
redox potential, which is about 450 mV (SHE). This
can be achieved with hydrogen concentrations of
only about 0.2 ppm, and under these conditions, the
main steam radiation level is not increased to an
unacceptable level. The first plant used NMCA successfully in 1997, and over 25 plants have already
followed, with excellent results. Field measurements
show that NMCA has been effective in providing
mitigation against IGSCC by lowering the ECP
below the 230 mV (SHE) threshold with relatively
low hydrogen injection rates.
The NMCA process is typically applied at refueling outage, before new fuel is inserted into the core,

Water Chemistry Control in LWRs

additional benefit with NMCA on the upper, outer


shroud regions, as indicated by the additional shading
in the left-hand side of the figure5. It is estimated that
noble metals protect slightly more of the outer core
region than does moderate HWC (HWC-M), but the
difference is not significant.
Figure 8 shows the dramatic benefit of noble
metals in reducing the rate of stub tube cracking at
Nine Mile Point 1 since the application in 2000.
Before 2000, several stub tubes had to be repaired
or replaced at each outage, but since the application,
only one tube leaked, and this was believed to have
already cracked before NMCA.
Recently, attention has been focused on the online
application of noble metals, with the first application
at the KKM plant in Switzerland. By April 2008,
there were four applications in the United States.
This is discussed in a later section.

HWC protected regions

NMCA protected regions

and is effective for about three fuel cycles, before


reapplication is necessary. The regions of the reactor
vessel internals that are protected by HWC-M or
NMCA are shown in Figure 7. While both techniques offer significant areas of mitigation, there is an

5.02.2.3

Radiation Field Control

Corrosion products deposited on the fuel become


activated, are released back into the coolant, and
may be deposited on out-of-core surfaces. Both soluble and insoluble species may be involved, the latter
tending to deposit in stagnate areas (crud traps). The
chemistry changes to control IGSCC resulted in
increased out-of-core radiation fields, and the implementation by most plants of depleted zinc injection to

Figure 7 Mitigated regions of the boiling water reactor


core.

Number of stub tubes identified with IGSCC


throughwall cracking based on leakage

12

10

Noble metal applied


mid cycle
may 2000

0
19841985 19861987 19881990

23

1991

1993

1995

1997

1999

2001

2003

2005

2007

RFO-11

RFO-12

RFO-13

RFO-14

RFO-15

RFO-16

RFO-17

RFO-18

RFO-19

Year
Figure 8 Mitigation of stub tube cracking at Nine Mile Point Unit 1.

24

Water Chemistry Control in LWRs

control dose rates, as discussed later in this section.


During shutdowns, the major radiation source for
personnel exposure is activated corrosion products,
deposited on primary system surfaces. Exposures are
generally accumulated at high-radiation field locations
where maintenance work is frequently needed.
Although improvement of maintenance equipment
and procedures, reduction of maintenance requirements, increased hot-spot shielding, and control of
contamination dispersion have significantly reduced
total exposure, further reduction of radiation fields is
a major goal in programs for minimizing occupational
radiation exposure.
The primary source of radiation field buildup on
out-of-core surfaces in BWRs is 60Co, which in
mature plants usually accounts for 8090% of the
total dose. 60Co has a relatively long half-life of
5.27 years. The higher the soluble 60Co concentration
in the coolant, the more 60Co is incorporated and
deposited on out-of-core systems and components,
resulting in higher dose rates on recirculation piping,
the reactor water cleanup system, dead legs, and
other crud traps in the system. Other activated transition metals such as 54Mn, 58Co, 59Fe, and 65Zn
contribute the remainder of the dose. 51Cr also contributes significantly to the piping dose in some
NMCA plants. The radiation fields commonly
measured in a BWR at the straight vertical section
of recirculation pipes are considered to be more
representative for the purposes of radiation buildup
trending and comparison with other plants. These
measurements are done in a prescribed manner
developed under the EPRI BWR Radiation and Control program and are called BRAC point measurements. These measurements represent primarily the
incorporation of soluble 60Co into the corrosion film
on the piping surfaces and tend to be a fairly good
predictor of drywell dose rates. The deposition of
particulate oxides that contain 60Co and other activated species can also contribute significantly to outof-core radiation levels in BWRs, especially in hot
spots. The particulate oxides, which vary in size,
originate primarily from corrosion of the steam/condensate system and are introduced via the feedwater.
The sole precursor of the gamma-emitting 60Co
isotope is 59Co. 59Co is present as an impurity in the
nickel in structural alloys used in BWRs (e.g., Type
304 stainless steel) and is the main constituent of
wear-resistant alloys (e.g., Stellite), used as hard facing in valves and other applications requiring outstanding wear resistance. Corrosion and wear lead to
release of 59Co into the coolant from these sources,

which is transported to the core and incorporated


into the crud that deposits on the fuel rods. The
59
Co is activated to 60Co by neutron activation,
released back into the coolant, and incorporated as
a minor constituent into the passive films that form
on components that are inspected, repaired, and
replaced by maintenance personnel. Components in
the neutron flux (e.g., the control blades) directly
release 60Co. Cobalt source removal is clearly important if radiation fields are to be minimized. Another
gamma-emitting isotope, 58Co, is formed by the activation of nickel from stainless steel and nickel-based
alloys. 58Co has a shorter half-life and is not as major
a contributor to radiation fields as 60Co in BWRs, but
is much more significant in PWRs.
Shutdown drywell dose rates increase when coolant chemistry is changed for the first time from
oxidizing (NWC) to reducing (HWC) conditions.
This results from a partial restructuring of the oxides
formed under the oxidizing conditions of NWC
(Fe2O3 type) to a more reducing spinel type oxide
compound (Fe3O4 type). The oxides affected are the
fuel deposits, the corrosion films on stainless steel
piping, and out-of core deposits. This results in an
increase in the chemical cobalt (and 60Co) concentration in the oxide because of the higher solid-state
solubility of transition metals in the spinel structure.
The presence of a higher soluble reactor 60Co concentration released from fuel crud while this conversion is occurring only aggravates the situation. The
processes are depicted in Figure 9. The net result at
most plants is a temporary increase in reactor water
60
Co, both soluble and insoluble forms, which leads to
significantly increased shutdown dose rates because
of both the increased reactor water concentrations
and the increased capacity for transition metal uptake
by the spinel phases.6

Oxide stable
under normal
water chemistry
Fe2O3
(containing 60Co,
58Co, 54Mn, etc.)

Corrosion films
Vessel crud
Fuel crud

Restructuring under
HWC conditions

Fe3O4
form of oxide

Small insoluble
particles containing
60Co, 54Mn, etc.
Soluble 60Co, etc.
released during
restructure

Figure 9 Boiling water reactor oxide behavior under


reducing conditions.

Water Chemistry Control in LWRs

25

0.8
Before Zn addition
After Zn addition

RxW 60Co (Ci kg 1)

0.6

0.4

0.2

0
Brunswick-1

Brunswick-2

Dresden-2

Figure 10 Hydrogen water chemistry plant RxW

60

Duane Arnold

Fitz patrick

Monticello

Pilgrim

Co response to zinc addition.

As mentioned earlier, zinc addition reduces radiation field buildup. The mechanism of the zinc ion effect
is complex, as release of 60Co from fuel crud is reduced,
and deposition out-core is also reduced. Overall, reactor water 60Co is decreased significantly after zinc
addition, as shown by plant data in Figure 10.
Aqueous zinc ion promotes the formation of a
more protective spinel-structured corrosion film on
stainless steel, especially when reducing conditions
are present. Second, both cobalt and zinc favor tetrahedral sites in the spinel structure, but the site preference energy favors zinc incorporation. Thus, the
available sites have a higher probability of being filled
with a zinc ion than a cobalt ion (or 60Co ion), and
hence the uptake of 60Co into the film will be significantly less if zinc ion is present in the water. The
60
Co remains longer in the water and is eventually
removed by the cleanup system.
The zinc was originally added to the feedwater as
ZnO, but it was quickly found that the 65Zn that was
created by activation of the naturally occurring 64Zn
isotope in natural zinc created problems. With the
use of zinc oxide depleted in the 64Zn isotope, called
depleted zinc oxide (DZO), this drawback was eliminated. Because of the high cost of DZO, feedwater
zinc injection was not implemented widely until
HWC shutdown dose issues emerged.
For the case of plants treated with NMCA and
injecting hydrogen, the oxidant concentration on the
surface of the stainless steel is zero (due to the Pt

and Rh catalyzing the reaction of any oxidant with


the surplus hydrogen). The net result is that the ECP
is at or very near the hydrogen redox potential,
typically about 490 mV (SHE) for neutral BWR
water. This low potential causes a much more thorough restructuring of the oxides to the spinel state
than observed under moderate hydrogen water
chemistry (HWC-M).
Feedwater iron ingress has a significant influence
on the effectiveness of zinc injection. As discussed in
the next section, deposits on fuel cladding surfaces
(called CRUD) are mainly composed of iron oxides,
with other constituents. Therefore, reducing iron
ingress from the feedwater has the benefit of minimizing crud buildup, which is important for fuel
reliability (next section). For these reasons, extensive
efforts have been made to reduce iron ingress, with
significant success. Furthermore, fuel crud has a large
capacity for incorporating zinc and is in fact where
most of the zinc ends up. The lower the amount of
crud on the fuel, the greater the proportion of zinc
that remains in solution and can subsequently be
incorporated in out-of-core surfaces. Therefore, at
plants with low feedwater iron, less zinc is captured
by the crud on the fuel, so a relatively greater amount
remains in solution and is available to control out-of
core radiation fields. This is very important, as zinc
injection rates are limited by fuel performance concerns, and hence lowering feedwater iron is essential
for maintaining lower radiation fields.

26

Water Chemistry Control in LWRs

5.02.2.4

Fuel Performance Issues

Fuel durability has improved over the years, and


failures have declined, helped by improvements in
water purity. In operation, zircaloy fuel cladding
develops a thin oxide layer (ZrO2), which typically
does not adversely affect performance. However, an
increase of deposition of corrosion product deposits
(crud) on this oxide film is undesirable because it
can reduce heat transfer and increase fuel pin temperatures, with resultant increased corrosion of the
fuel cladding, ultimately increasing the risk of fuel
failure. Moreover, the addition of additives to control
corrosion may increase the risk of crud buildup on
the fuel. For example, zinc and noble metals in BWRs
tend to increase the adherence of crud deposits on
the fuel, which can result in undesirable oxide spalling in higher-rated cores. In fact, corrosion-related
fuel failures occurred at four plants in the United
States between 1999 and 2003. Although the precise
root cause of fuel failures is often difficult to determine, it is clear that excessive crud buildup played a
role in these failures. With progressive uprating of
fuel duty in both PWRs (and BWRs), the margin to
tolerate crud has been reduced and additional care
has to be taken in specifying the water chemistry to
avoid undesirable fuel performance issues. Despite
these more demanding conditions, fuel failures have
decreased in recent years.
Concern about the possibility of adverse effects of
NMCA on fuel has prompted imposition of a strict
limit on the amount of noble metal that can end up on
the fuel and guidance on the injection of zinc. Plant
data indicate that spalling of the corrosion layer from

fuel cladding, which is often regarded as a precursor


to cladding failure, is prevented if the cycle average
feedwater zinc is maintained below 0.4 ppb in NMCA
plants (0.6 ppb for non-NMCA plants). More recent
data indicate that quarterly averages may be as high
as 0.5 ppb for NMCA plants, without occurrence of
spalling.5
These feedwater zinc data are the basis for limits
in the water chemistry guidelines. The 2008 chemistry guidelines7 retain the cycle average feedwater
zinc limit of 0.4 ppb (0.6 ppb for non-NMCA plants)
but enable a slight increase in the quarterly average
to 0.5 ppb, which may allow flexibility in controlling
radiation buildup in parts of the cycle.
The tighter control of water chemistry in recent
years has been successful in controlling crud formation on fuel cladding, and Figure 118 shows failures
from pelletclad interaction causing SCC, fabrication
defects, debris, and crud/corrosion. Note that there
have been zero crud/cladding related fuel failures in
US BWRs since 2004 (although assessment of 2007
failures is not yet complete, crud/corrosion is not
believed to be a factor here).
Analysis of recent plant data confirms that control
of feedwater iron ingress has the positive benefit of
reducing the amount of crud on the fuel. Control of
copper, which generally originates from admiralty
brass alloys, is also beneficial; not only can copper
have detrimental effects on the fuel, but it also limits
the ability of hydrogen to reduce the ECP, and it also
leads to higher radiation fields. As a result, most US
plants have replaced condensers containing brass
tubing.

Number of failed assemblies

30
25
20

PCI-SCC
Unknown
Fabrication
Debris
Crud/corrosion

15
10
5
0
2000

2001

2002

2003
2004
EOC year

2005

2006

Figure 11 US boiling water reactor fuel failures by mechanism for each end-of-cycle (EOC) year.

2007

Water Chemistry Control in LWRs

5.02.2.5

Online Addition of Noble Metals

As discussed earlier, the classic NMCA process is


generally applied during refueling outages before
the new fuel is loaded into the core. Reapplication
after about three cycles of operation takes approximately 2 days, while the plant maintains 107154  C
as it enters the refueling outage. To reduce this outage
time, GE-Hitachi developed OLNC, first demonstrated at KKM (a GE design of plant in Switzerland)
in 2005, with several more additions subsequently.
Preliminary results indicate that there have been no
unexpected chemistry effects during the first OLNC
applications, and shutdown radiation fields actually
decreased at KKM after OLNC.5 Subsequently, CGR
of susceptible welds decreased significantly, as shown
by the decrease in slopes in Figure 12 after OLNC
initiation for two welds that have been monitored for
several years.
The effects of OLNC on fuel have been extensively studied in fuel removed from KKM, and no
adverse effects have been observed. The jury is still
out on this concern, but the general assessment is that
OLNC will have no more impact than the classic
application, and may well prove to be of less concern.
More IGSCC and fuel measurements are planned,
but with no issues emerging to date, it appears that
OLNC applications about every 12 months would be
effective and economical, avoiding the critical path
time necessitated for the classic NMCA application
during refueling outages. Initial OLNC applications
have been carried out at plants that had previously
applied noble metals in the classic off-power manner.

However, the first OLNC application at a plant that


has not used noble meals previously occurred in late
2008, but no results are available.

5.02.3 PWR Primary Water


Chemistry Control
5.02.3.1 Evolution of PWR Primary
Chemistry Strategies
In the very early days of PWR operation, heavy crud
buildup on fuel cladding surfaces was caused by the
transport of corrosion products from the SGs into the
reactor core. As a result, activated corrosion products
caused high-radiation fields on out-of-core surfaces
(Figure 13), fuel performance was compromised, and
even coolant flow issues were observed in some
plants.
These problems were initially mitigated by imposing
a hydrogen overpressure on the primary system, to
reduce the ECP, and raising the primary chemistry
pH. Materials degradation in primary systems was
then not a major concern, with most of the emphasis
focused on secondary side corrosion issues in the
SGs. Commercial PWR power plants use a steadily
decreasing concentration of boric acid as a chemical
shim (for reactor control) throughout the fuel cycle,
which results in the use of lithium hydroxide to
control pH. Some 30 years ago, the concept of
coordinated boron and lithium was developed,
whereby the concentration of LiOH was gradually
reduced in line with the boric acid reduction to
maintain a constant pH.

300
NC
appl.

HWC

Indication length (mm)

250
Ind 9

Ind10

Not inspected in 00, 01,04

200
150
100
50

27

OLNC
37 g
OLNC
98 g
OLNC
198 g
OLNC
199 g

Indications 9,10
may be seeing
mitigation by OLNC

0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Year
Figure 12 Ultrasonic inspection results after online noble metal chemical addition.

28

Water Chemistry Control in LWRs

Corrosion products
deposit out of core
Pressurizer
Steam generator
Corrosion
products
activated in
reactor core

Corrosion products
released from SG tubing
Coolant pump
Reactor
Primary loop

Figure 13 Transport and activation of corrosion products in pressurized water reactor primary systems.

Corrosion products released from the steam generator tubes are transported, dissolved, or deposited
by the coolant on the basis of solubility differences.
The solubility of nickel and iron depend on pH,
temperature, and redox potential, all of which vary
with location around the nonisothermal system.
Originally, a constant at-temperature pH of 6.9 was
recommended, based on the minimum temperature
coefficient of solubility of magnetite. In fact, it was
determined that heavy fuel crud buildup was avoided
if a constant pH of at least 6.9 was maintained. This
was possible with 12-month fuel cycles, but fuel cladding corrosion concerns limited the maximum LiOH
concentration to 2.2 ppm. Consequently, plants often
started the fuel cycle with pH below 6.9, which
resulted in deposition of corrosion products on the
fuel, activation of cobalt and nickel, and subsequent
transport to out-of-core surfaces, resulting in radiation fields remaining relatively high.
Even though detailed studies of fuel crud showed
that the prime constituent of the crud was nickel
ferrite (for which the optimum pH is 7.4), this coordinated chemistry had remained the standard for
many years, until higher pHs became the norm in
the 1990s. Although research and plant demonstrations showed that the 2.2 ppm limit was excessively
conservative, the move to higher Li concentrations
has been slow. However, detailed fuel examinations
from a recent plant demonstration (that will be discussed later) have indicated that Li can be raised to as
high as 6 ppm.

About 25 years ago, primary water stress corrosion cracking (PWSCC) of Alloy 600 SG tubes was
observed in a few plants, leading to studies on mitigating this effect. Following successful demonstration of zinc injection in BWRs, initial field tests at
PWRs showed that radiation fields were reduced,
and laboratory studies indicating that PWSCC was
reduced were eventually confirmed. As a result, zinc
injection is being implemented at an increasing rate,
although concerns about fuel performance at highduty plants have not been completely resolved. Most
recently, buildup of boron-containing crud in areas
of subcooled nucleate boiling leading to localized
flux depression has encouraged the use of higher Li
concentrations to minimize corrosion product transport. Concerns about the potential adverse effects of
zinc deposited in high-crud regions have resulted in
several highly rated plants applying in situ ultrasonic
fuel cleaning before implementing zinc injection.
Although zinc injection was developed for radiation field control, laboratory studies showed that it
also inhibited SCC under PWR conditions. The
identification of PWSCC in reactor vessel penetrations in the last 1520 years has encouraged the use
of zinc injection, but has also focused attention on
the effects of dissolved hydrogen, for which the
recommended range has remained 2550 ml kg1
for 30 years. It now appears that raising hydrogen
will reduce PWSCC rates, while lowering it may
delay initiation of PWSCC. The interactions of
materials, radiation fields, and fuels in PWR primary

Water Chemistry Control in LWRs

PWSCC:
pH (Li, B) minimal effect
Zn beneficial
dissolved H2 effect

Plant
operations

29

Dissolved H2
control range

Materials
degradation

PWR
chemistry
control

Fuel
performance

Plant
dose
rates
Radiation fields:
pH (Li, B), Zn
beneficial

Crud deposition:
Zn concern for
highly rated cores

Figure 14 Pressurized water reactor primary chemistry optimization. Reproduced from Fruzzetti, K.; Perkins, D. PWR
chemistry: EPRI perspective on technical issues and industry research. In VGB NPC08 Water Chemistry Conference, Berlin,
Sept 1418, 2008.

Dissolved (H2) range changes

Elevated constant pH (7.3/7.4)


Ultrasonic fuel cleaning
Elevated constant pH (7.1/7.2)
Zinc injection
Modified elevated lithium program
EPRI water chemistry guidelines
Elevated lithium program
Constant pH 6.9
1975

1980

1985

1990

1995

2000

2006 2008

Figure 15 Pressurized water reactor primary chemistry


changes at US plants.

chemistry and optimization issues covered in the


water chemistry guidelines, which are discussed
later, are depicted in Figure 14.
The evolution of water chemistry control in PWR
primary systems in the United States over the last
30 years is shown in Figure 15.
The following sections address the three main
factors pH control, zinc injection, and dissolved
hydrogen control that have dominated PWR primary chemistry strategies in the past and continue to
do so today.9 Each of these factors is considered from
the viewpoint of materials degradation, radiation
field control, and fuel performance concerns.

5.02.3.2

Materials Degradation

Materials degradation has been covered in detail in


Chapter 5.04, Corrosion and Stress Corrosion
Cracking of Ni-Base Alloys and Chapter 5.05, Corrosion and Stress Corrosion Cracking of Austenitic
Stainless Steels, and here only the specific effects of
water chemistry variables on materials in PWR primary systems will be reviewed, particularly those that
may affect the chemistry of optimization process.
Recent papers by Andresen et al.10,11 provide
detailed results of a comprehensive study of the effects
of PWR primary water chemistry on PWSCC of
nickel-based alloys. Extensive studies have been carried out to determine the effect of lithium, boron, and
pH on PWSCC, and the generally held conclusion is
that any effects are minimal, especially compared to
material susceptibility, stress state and temperature,
and other operational issues. Crack initiation tests
using the most reliable types of reverse U-bend specimens indicate that pH has a relatively small effect on
crack initiation (generally less than a factor of 2).
Although the most rapid crack initiation occurred at
pH310  C 7.25, with slower rates at higher or lower pHs,
CGR tests generally confirm that pH has minimal
effect.
The effect of lithium is even smaller than the pH
effect, and the influence of boron is minor or nonexistent. Andresen et al. concluded that the effects of relevant variations in PWR primary water chemistry (B, Li,

30

Water Chemistry Control in LWRs

and pH) have little effect on the SCC growth rate in


Alloy 600, and thus provide little opportunity for mitigation of PWSCC. Plant data have found no adverse
effects from increasing lithium and pH in primary
systems. As a result, it is considered that adjusting
pH, lithium, or boron to minimize crack initiation
may be of minimal value. The 2007 edition of the
PWR Primary Water Chemistry Guidelines12 reviewed
the most recent data and concluded that pH strategy
changes based on PWSCC considerations are not warranted. This means that plants have the flexibility to
pursue B/Li/pHt chemistry adjustments to minimize
crud transport and radiation buildup without concern
for negative effects on PWSCC susceptibility of nickelbased alloys, although of course chloride and sulfate
impurities should continue to be minimized.
Following good experience in BWRs, zinc injection has been implemented in the primary systems of
PWRs, both to reduce primary side cracking of nickelbased alloys and to control dose rates. The qualification
work for BWRs showed that zinc inhibited SCC, but
the benefit was not sufficient to avoid the need for
hydrogen water chemistry to mitigate IGSCC. Thus,
the motivation for BWR zinc injection was exclusively
radiation field control.
The situation in PWRs is different, as laboratory
work13 showed that initiation of PWSCC was significantly delayed by zinc injection, and hence the motivation for the initial applications of zinc in most US
PWRs at the 1030 ppb level was to control PWSCC
of SG tubing. Additionally, German-designed PWRs
and a few US plants used 5 ppb depleted zinc for
radiation control.

Figure 16 shows the rate of introduction of zinc


injection at PWRs worldwide.
Zinc injection produces thinner, more protective
oxides on stainless steel and Alloy 600, with zinc
displacing Co2, Ni2, and Fe2 from normal spinels
to give ZnCr2O4, which is very stable. The benefits
of PWR zinc injection have been clearly demonstrated in reducing PWSCC degradation (especially
growth rate) of Alloy 600 SG tubes, and in controlling
radiation fields. Evaluation of currently available
laboratory data2 indicates that PWSCC initiation
will be reduced, and PWSCC CGR may be reduced
in thicker cross-section components, depending upon
other factors such as the stress intensity factor of the
specimen. Andresen et al.11 concluded that crack
growth mitigation by adding Zn requires further
study, although two of four tests show a decrease
in growth rate of >3. Molander et al.14 also found
that the effect of zinc on CGR was minor. Hence,
more work is needed before making definitive conclusions from laboratory studies regarding the benefit
of zinc in mitigating CGR.
SG tube nondestructive examination (NDE) data
from eight plants injecting zinc indicated reduction
in the incidence of PWSCC by a factor of 210.9 An
example from a 2-unit PWR showing the effect of
zinc on SG tubing over successive cycles is given in
Figure 17.
The largest effect of zinc appears to be on initiation of cracking, with a smaller effect on CGR, with
the data indicating a factor of 210 reduction for
initiation and about a factor of 1.5 reduction in CGR,
consistent with the extensive laboratory work,11

Application of zinc in world PWRs

Number of plants and percentage


of PWR injecting zinc

50
45

Percent of PWRs injecting

Number of units injecting

40
35
30
25
20
15
10
5
0
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Year

Figure 16 Application of zinc injection in pressurized water reactors worldwide.

Water Chemistry Control in LWRs

31

140
X indicates last refueling outage
before
start of zinc injection

Number of new tubes affected

120
100
80

Unit 1
Unit 2
60
40
20
0

X3

X2

X1

X+1
X
Refueling outage

X+2

X+3

X+4

Figure 17 Effect of zinc on steam generator tube degradation at a US pressurized water reactor.

70
60
Number of plants

indicating that zinc inhibits mainly by delaying the


initiation of PWSCC. However, the SG NDE data
also showed that zinc reduced the rate of crack propagation (depth) by 1760%.
These results are consistent with initial laboratory
test data indicating that zinc reduced crack propagation by a factor of approximately 3 at low stress
intensities, but had no effect at higher stress intensities. In addition, the lack of cracking in the Farley
PWR pressure vessel head penetrations (exposed to
zinc for over 12 years), compared to PWSCC indications in similar pressure vessel heads in other plants,
suggests that zinc addition is beneficial for Alloy 600
(and possibly Alloy 82/182) thick-section components under PWR primary service conditions.
Recent work has studied the influence of dissolved
hydrogen on PWSCC. In the early days of PWR
operation, the lower limit on hydrogen was set at
25 ml kg1, to provide adequate margin against radiolysis and heavy crud formation. Plant tests in France
showed that this limit was excessively conservative
and that less than 10 ml kg1 would be satisfactory,
provided good control of oxygen was maintained in
makeup water. Several workers have found that the
maximum in PWSCC CGR occurs close to the ECP
corresponding to the Ni/NiO thermodynamic equilibrium condition.15 Although this potential is unaffected by lithium/boron/pH (consistent with the fact
that these do not greatly influence PWSCC over the
range of practical relevance), the equilibrium potential is significantly affected by the dissolved hydrogen

50
40
30
20
10
0
2530
3035
3540
4045
4550
Cycle average hydrogen concentration (cm3 kg1)

Figure 18 US plant data for dissolved hydrogen.

concentration. Andresen et al. found that the peak in


SCC growth rate versus H2 fugacity was temperature
dependent, but generally fell within the hydrogen
concentration range used in PWRs. This provides
an opportunity for mitigation, by perhaps a factor of
2 in Alloy 600 and a factor of 5 in Alloys 182, 82,
and X750, as the median value of the dissolved
hydrogen concentration for US plants is approximately 35 ml kg1.
US PWRs currently operate within dissolved
hydrogen within the recommended 2550 ml kg1
range, with the majority in the 3040 ml kg1 range,
but none with more than 44 ml kg1 (Figure 18).
The lower limit is set conservatively to provide an
operating margin over the level of hydrogen required

32

Water Chemistry Control in LWRs

to suppress water radiolysis in the reactor core. Somewhat lower concentrations are used in other countries.
The dissolved hydrogen concentrations corresponding to the peak CGR for a typical range of
PWR primary operating temperatures are 4.3 ml kg1
at 290  C, 10.4 ml kg1 at 325  C, and 16.5 ml kg1
at 343  C.15 Andresen et al.11 published Figure 19,
which indicates the proposed factors of improvement
on changing from an initial hydrogen concentration
of 25 ml kg1. It can be seen that raising the hydrogen provides benefit, but lowering it is detrimental
below 330  C.

In response to the data showing the benefit of


increasing hydrogen in reducing CGR, the US industry program in progress focuses on the extent to
which dissolved hydrogen can be increased without
adverse consequences to other parts of the system.
Other countries, including Japan, are also investigating lowering hydrogen, because laboratory data
suggest that the initiation of cracking is delayed at
lower hydrogen concentrations. This is depicted in
Figure 20, as discussed by Molander.14
The lower line in this figure shows the time to
initiate cracking, based on laboratory tests using

Factor of improvement from H2

4.0
Based on Alloy 182,
a current H2 level of 25 cm3 kg1

3.5

70 cm3 kg-1 H2

25

3.0

45 cm3 kg-1 H2

25

2.5
2.0
1.5

Good
1.0
25
0.5
0.0
270

280

290

1 cm3 kg-1 H2

300
310
320
Temperature (C)

4 cm3 kg-1 H2

25

330

340

Bad

350

Figure 19 Effect of dissolved H2 on primary water stress corrosion cracking crack growth rate at different temperatures.

ml H2/kg H2O (330 C)


10

15

20

25

30

Crack initiation time (h)

Jenssen data on Alloy 600

35

1E07

Growth
8E08

20 000

6E08

15 000

4E08

10 000
Initiation

2E08

5000

0
0

10
Hydrogen activity (kPa)

15

Crack growth rate (mm s1)

5
25 000

0E+00
20

Figure 20 Dependencies between the dissolved hydrogen content in pressurized water reactor primary coolant on the
crack initiation time observed on initially smooth surfaces and on the crack propagation rate.

Water Chemistry Control in LWRs

reverse U-bend specimens, whereas the upper line


shows crack growth data over a similar concentration
range. Thus, the lowering of hydrogen appears feasible. However, the relative importance of crack initiation
and crack propagation is very dependent on material
and plant conditions. In the United States, concern
about increased crack propagation at low hydrogen
and low temperatures, as shown in Figure 19, has
resulted in moving to higher hydrogen being preferred
to the alternative of reducing hydrogen. Several factors
combine to make higher H2 the preferred way to mitigate SCC, including the importance of bottom-head
penetrations (which are exposed to 290  C water)
and the recent observation that the CGR in coldworked Alloy 600 is not mitigated at low H2.11
The preferred strategy in the United States is
to gradually increase hydrogen to the upper end of
the existing range, with the potential to move higher
(say to 60 ml kg1) when the ongoing qualification
work is completed. This will include evaluation of
the effects of dissolved hydrogen on radiation fields
and fuel performance, although any such effects are
expected to be minimal.16
5.02.3.3

PWR Radiation Field Control

Corrosion products released from out-of-core materials (primarily SG tubing) deposit on the fuel and
become activated, are released back into the coolant,
and may be deposited on out-of-core surfaces. Both
soluble and insoluble species may be involved, with the
latter tending to deposit in stagnate areas (crud traps).
In addition to the chemistry items discussed later in
this section, it must be stressed that other factors are
important to the goal of reducing radiation fields. In
particular, the success of the later German-designed
plants in eliminating cobalt sources in hardfacing alloys,
thereby achieving very low radiation fields, demonstrates the benefits of cobalt source reduction. With
many plants replacing SGs, a correlation between
recontamination rates and surface finish of the new
SG tubing has been noted by Hussey et al.17
Typical PWR fuel cycles start with a relatively high boric acid concentration, which gradually
reduces to zero at the end of the cycle. Lithium
hydroxide is added to maintain an approximately
constant pH. As the duration of fuel cycles increased,
more boric acid was required at the start of cycle,
which in turn necessitated increased LiOH to maintain the desired pH (Figure 21).
As mentioned earlier, radiation field buildup can
be controlled by minimizing corrosion product

33

transport and activation. Initially, coordination of


lithium hydroxide with boron to maintain a constant
at-temperature pH of 6.9 was recommended, based
on the minimum solubility of magnetite. In fact, the
prime constituent of the crud turned out to be nickel
ferrite, requiring a pH of 7.4 for minimum solubility. Fruzzetti et al.15 have recently reviewed the data
on elevated pH, which provides a number of benefits
including decreased general corrosion (and thus
reduced corrosion product transport to the core).
Field-tests of pHs greater than 6.9 confirmed that
radiation fields were lower. Although no adverse
effects were observed on the fuel, many plants were
slow to abandon a 2.2 ppm limit, established to avoid
excessive zircaloy corrosion. However, there were
indications of heavier crud formation after long periods operating below pH 6.9, and as fuel concerns
relaxed, a gradual move toward a maximum of 3 ppm
lithium resulted. Moreover, pHs in the range 7.17.2
became more popular in the late 1990s, with 7.37.4
eventually gaining favor.
Figure 22 shows the maximum lithium concentrations reported by US PWRs in recent years. It can
be seen that 95% are now using greater than 3 ppm
at full power: a significant change from earlier in
the decade.
A demonstration of elevated Lithium/pH is in
progress at Comanche Peak PWR.18 The goal was to
reduce radiation fields and reduce susceptibility to the
Axial Offset Anomaly (AOA) by reducing crud
buildup. This test involved increasing the primary
system pH from 7.1/7.2 to 7.3 and then two cycles at
7.4. No significant adverse trends have been noted,
either in the area of chemistry or core performance.
Radiation fields measured have shown a modest but
continued improvement. On the basis of the positive
trends and absence of any negative effects, Comanche
Peak has established elevated constant pHTave 7.4 as
the primary chemistry regime for both units.
Without the increases in pH/lithium that have
taken place, radiation fields would have been
expected to increase significantly for longer fuel
cycles. The increase in boiling in localized regions
of the core (called subcooled nucleate boiling)
in PWRs resulting from power uprating has resulted
in higher crud buildup on the upper fuel surfaces, and
there is growing evidence from US PWRs that radiation fields are indeed higher for the highest rated cores.
Enriched boric acid (EBA), that is boric acid
enriched with B-10, enables a given pH to be
achieved with less lithium hydroxide, as the required
concentration of B-10 can be obtained with less total

34

Water Chemistry Control in LWRs

Constant pH 7.2
6

Lithium
Li high limit
Li low limit

5
Li target = 6.0 E7 B2 + 0.0023B + 0.4413

Lithium (ppm)

4
3.5 ppm limit
3
2.2 ppm limit
2

Start of 18-month
cycle

Start of 12-month cycle

20

80

140

200

260

320

380

440

500

560

620

680

740

800

860

920

980

1040

1100

1160

1220

1280

1340

1400

1460

1520

1580

1640

1700

Boron (ppm)
Figure 21 Lithium concentrations required to maintain pH 7.2 for different fuel cycle lengths.

boric acid. EBA is used at several plants in Europe,


typically to increase shutdown margin when using
mixed oxide fuel (MOX), but has not been applied
to date in the United States. However, consideration
is being given to using EBA at some plants that will
use MOX fuel in the future. Despite the transition to
the use of EBA in operating plants, designing for it in
new plants is recommended.19
As discussed earlier, the motivation for the initial
applications of zinc in most US PWRs was to control
PWSCC of SG tubing. However, German-designed
PWRs and a few US plants used 5 ppb depleted
zinc for radiation control, mostly with depleted zinc
to avoid zinc-65 formation. A recent paper Understanding the zinc behavior in PWR primary coolant: a
comparison between French and German experience
by Tigeras et al.20 provides a European perspective on
this topic.
This paper concludes that zinc injection seems to
present the most positive and clearest results: in all
the units injecting zinc, a dose rate reduction has
been detected after a certain period of exposure
without leading to any negative impact on plant

systems, components, and operation. Thus zinc


injection should be considered as a strategy with
benefits in short, medium, and long term. Its application as soon as possible in the life of nuclear power
plants and especially before SG replacement and fuel
cycles modifications seems to be an excellent decision to contribute to ensuring the passivation process
of new components, the fuel performance, the full
power operation of the units, and the long life of
materials and components.
Figure 23 shows the effect of zinc in reducing
radiation dose rates at several plants. It can be seen
that the reduction factor approximately correlates
with the cumulative zinc exposure in ppb months
(the product of the average zinc concentration and
the duration of zinc addition). As little as 5 ppb
zinc has been shown to reduce radiation fields by
3550% at operating plants, based on zinc exposures
of 700 ppb months. There is relatively little difference between plants with Alloy 600/690 SG tubing
and those with Alloy 800 tubing, but plants using
depleted zinc show greater benefit than those using
natural zinc, as shown in the figure.

Water Chemistry Control in LWRs

35

Percentage of units within range

70

<3 ppm
3.03.5 ppm
>3.5 ppm

60
50
40
30
20
10
0

2000

2001

2002

2003

2004

2005

2006

2007

EOC year
Figure 22 Maximum reported coolant lithium (full power) at US pressurized water reactors.

Cumulative dose rate reduction fraction

1.2
Alloy 800 w/depleted zinc
Alloy 600 and 690 w/depleted zinc
Alloy 600 and 690 w/natural zinc
Log Alloy 800 plants
Log Alloy 600 and 690 w/depleted zinc
Log Alloy 600 and 690 w/natural zinc

0.8

0.6

0.4

0.2

0
0

200

400

600
800 1000 1200 1400 1600
Cumulative zinc exposure (ppb months)

1800

2000

Figure 23 Effect of zinc injection on radiation dose rates.

5.02.3.4

Fuel Performance

With progressive uprating of fuel duty, the margin to


tolerate crud has been reduced and additional care
has to be taken in specifying the water chemistry to
avoid undesirable fuel performance issues. Figure 24
shows the root causes of PWR fuel failures since
2000, including failures from pelletclad interaction
causing SCC, fabrication defects, debris, grid fretting, and crud/corrosion. In contrast to the BWR

situation, shown in Figure 11, very few failures in


recent years have been attributed to crud/corrosion
(the exceptions to this comment are discussed in a
following section).
A phenomenon called axial offset (AO) has caused
concern over the past 10 years.21 AO is a measure of
the relative power produced in the upper and lower
parts of the core and is normally expressed as a
percent, with a positive percent indicating that

36

Water Chemistry Control in LWRs

Number of failed assemblies

120
Unknown
Debris
Crud/corrosion

100

Fabrication
PCI-SCC
Grid fretting

80
60
40
20
0
2000

2001

2002

2003

2004

2005

2006

2007

EOC year
Figure 24 US pressurized water reactor fuel failures by mechanism.

more power is produced in the upper part of the core.


AOA occurs when boron concentrates in corrosion
product deposits (crud) on the upper spans of fuel
assemblies undergoing subcooled nucleate boiling,
causing a reduction in neutron flux. AOA has affected
at least 20 PWRs in the United States, as well as
several in other countries.
Clearly, fuel crud is involved in the AO phenomenon, and water chemistry effects must be considered
in controlling AO. Besides their axial asymmetry, the
composition of fuel deposits in boiling cores is different from nonboiling fuel. The nickel-rich deposits on
boiling cores tend to be removed much less effectively
by conventional chemistry shutdown evolutions than
the nickel-ferrite deposits on nonboiling cores. Alternative methods are therefore required for removing
corrosion product deposits from reload fuel from highduty cores, including ultrasonic fuel cleaning.
An important difference exists between plants
with Alloy 600 or 690 SG tubing and those (such as
German-designed plants) with Alloy 800 tubing.
The latter have a much lower proportion of nickel
in fuel crud and have not experienced the AO
phenomenon.22
Early work showed that lithium increased zircaloy oxidation rates, although the adverse effects
were reduced in the presence of boric acid. As a
result, a limit of 2.2 ppm lithium was generally
imposed to reduce zircaloy corrosion, although
excessive crud formation at low pHs was likely to
be more detrimental to the cladding than higher
lithium concentrations, especially as the resistance

to corrosion of zircaloy improved. This was confirmed by one of the few failures in recent years
that was uniquely attributed to crud buildup. In this
example, a move to a longer fuel cycle necessitated
increasing the boron concentration at start of cycle;
however, the 2.2 ppm lithium limit was retained,
resulting in the pH being well below 6.9 for the
initial period of the cycle. This in turn caused
heavy crud formation, to which subsequent fuel failures were attributed.
The move in the past ten years to greater fuel duty,
with operation of fuel at higher temperatures (with
localized subcooled nucleate boiling), has caused
crud-related problems to reappear, particularly the
localized flux depression as a result of buildup of
boron-containing crud, which were discussed earlier.
This in turn has renewed interest in elevated pH/
lithium to minimize corrosion product transport, the
use of EBA and the more immediate mitigation that
can be obtained from fuel cleaning.
Fuel performance is always a concern with changes
in water chemistry, such as zinc injection. On the basis
of current experience, the impact appears to be minimal for the majority of plants, but insufficient data
exist for plants with the highest fuel duties to allow
application without postexposure fuel inspections.
Data from US plants suggest little or no fuel concerns
for coolant zinc levels up to 40 ppb for plants with
less-highly rated cores. Extended experience at these
plants, over at least 10 years of operation, indicates no
adverse effects on fuel at zinc concentrations from
15 to 25 ppb. However, there have been no data

Water Chemistry Control in LWRs

available until recently for higher zinc concentrations


in higher duty cores where significant subcooled
nucleate boiling occurs on the fuel clad surface.23
Perkins et al.24 comment that fuel performance must
be considered prior to injecting zinc and additional
monitoring and fuel surveillances to understand and
evaluate the impact and the role of zinc may be
required in some circumstances.

5.02.4 PWR Secondary System Water


Chemistry Experience
5.02.4.1 Evolution of PWR Secondary
Chemistry Strategies
The objectives of PWR secondary water chemistry
control are to maximize secondary system integrity
and reliability by minimizing impurity ingress and
transport, minimizing SG fouling, and minimizing corrosion damage of SG tubes. Since secondary side
corrosion damage of SG tubes is primarily caused by
impurities in boiling regions, where high concentrations of impurities occur in occluded regions of
the SG formed by corrosion product deposits, new
approaches are continually sought to control corrosion
product transport to and fouling within the SGs.25
PWRs have experienced IGA on both the primary
and secondary sides of the Alloy 600 SG tubing, which
has been a major contributing cause of the replacement of most of the SGs with mill-annealed tubing,
not only in the United States but internationally.
Figure 25 illustrates the various corrosion processes
found in different locations in a recirculating SG.26
PWR secondary system water chemistry has
evolved through many changes over the years, largely
in response to emerging technical issues associated
with this degradation of structural materials in SGs.
In the early days of PWR operation, wastage became
a problem in the secondary side of PWR SGs, resulting in a switch from the use of sodium phosphate
inhibitor to all-volatile treatment (AVT) using
ammonia, which in turn brought about the denting
phenomenon. Tighter control of impurities, oxidizing potential, and pH were necessary to mitigate
the denting problem. Despite continued chemistry
improvements, many plants have had to replace
SGs of earlier designs (e.g., those tubed with Alloy
600MA), as shown in Figure 26.
Newer generation SGs are performing well,
although there remain concerns about the adverse
effects of lead impurity, causing Pb-assisted stress
corrosion cracking (PbSCC), which is discussed later.

37

Lead has been observed in various flow streams (final


feedwater, heater drains, etc.) in the secondary systems
of PWRs. Lead is detected at some concentration in
nearly all deposit analyses (SG and other locations).
Lead is present in trace concentrations in secondary
system materials of construction, as well as in chemical
additives such as hydrazine.15
Figure 27 shows the worldwide causes of SG
repairs through 2004. It can be seen that IGA is
currently the most prevalent form of degradation.
Figure 28 compares the behavior of three types of
SG tubing, Alloy 600MA (mill-annealed material used
in early plants, Alloy 600TT (thermally treated material used in later plants), and Alloy 690TT (an
improved alloy used in most replacement SGs). This
diagram is taken from the 2008 PWR Secondary Water
Chemistry Guidelines,27 which contains a much more
detailed account of corrosion processes. 600TT has
reduced susceptibility under mildly oxidizing highalkaline conditions, that is, SCC is not observed until
higher pH than for 600MA, and 600TT has approximately the same susceptibility as 600MA under acidic
conditions. 690TT is indicated as having a still smaller
region of susceptibility in the high-alkaline region and
as having no susceptibility in the acid region except
under highly oxidizing conditions that are unlikely to
occur in plants. However, other work indicates that
SCC can occur in 690TT at an acidic pH, especially
if lead is present. Also, SCC occurs in both 600MA and
600TT in the mid pH region if lead is present.
In the 1990s, improved pH control using amines
became a regular practice, and fine-tuning, including
using mixtures of different amines to control pH
throughout the circuit and coordination with resin
utilization, continues today. Hydrazine is used to
remove oxygen from the system. Hydrazine levels
have continually been reviewed and optimized, with
due regard to any impact on FAC in secondary systems, as FAC rates increase at very low oxygen concentrations. Molar ratio control (MRC) describes a
control strategy that adjusts the bulk water chemistry,
generally sodium and chloride, such that the solution
that is developed in the flow-occluded region is targeted to be near neutral. MRC can involve the addition of chloride ions to balance the cations that
cannot be reduced via source term reduction programs. MRC was widely practiced to minimize SCC
concerns, but has not been actively employed at plants
replacing to SGs tubed with Alloy 690TT. With
more plants replacing their SGs, less plants are adopting the MRC program. Only ten plants were doing
MRC in 2007,28 and they are all with original SGs with

38

Water Chemistry Control in LWRs

U-bend cracks
(PWSCC)

Fatigue

Free span
ODSCC
IGA

ODSCC

PWSCC

Expansion
transition

PWSCC

PWSCC or
ODSCC

ODSCC
Denting

Fretting, wear, corrosion,


thinning

Tubesheet
Pitting
IGA

Expansion
transition

Tubesheet

ODSCC
Sludge

Tubesheet

PWSCC tube-end
cracking

Tubesheet

Figure 25 Corrosion processes in recirculating steam generators, showing primary water stress corrosion cracking and
outside diameter stress corrosion cracking on the secondary side.

Water Chemistry Control in LWRs

39

140
Operating plants
Plants w/replacement SGs
120

134

84
88

81

134
134

134
72

64
67

59

52

45
51

37

22
29

17

10

3
5
7

78

134

134

132

131
132

134
131
131

132

132
133

134

12

132

134

12

133

133

130

121

63
68

55

46

1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007

22

20

31

41

40

11

93
99

60

108
115

127

80

79

Number of plants

100

Year
Figure 26 Steam generator replacement status worldwide.

100
90
80

Percent

70
60
50
40
30
20
10
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004

0
Year
IGA
Impingement Pitting
Other
Wear
Thinning

Fatigue
Unknown
SCC
Preventive

Figure 27 Worldwide causes of steam generator tube repair.

600MA and 600TT tubing. Currently, no plants with


replaced SGs are believed to be using MRC.
Titanium-based inhibitors to minimize corrosion are also employed at some plants. Boric acid

treatment (BAT) involves the addition of boric acid


to feedwater. Such approaches are worthy of consideration, on the basis of plant-specific degradation mechanisms, operational considerations, and

40

Water Chemistry Control in LWRs

1.0

TT690

Potential (V vs. Ec )

0.8
Some tests indicate that 690 TT may be
susceptible in the low pH region, especially if lead is
690 TT U-bend cracked
present
in near neutral AVT
with lead and oxidizing sludge
TT690
TT600
TT600
MA600

0.6

0.4

MA600
0.2
600 MA and 600 TT can be susceptible
in mid pH range if lead or reduced
sulfur is present.

6
7
8
pH 300 C (572 F)

10

11

12

Figure 28 Corrosion mode diagram for Alloys 600MA, 600TT and 690TT (based on Constant Extension Rate Tensile
Tests at 300  C), showing regions where materials are susceptible to attack.

interactions. The most recent developments are


aimed at reducing deposit buildup in crevices,
including the use of dispersants, such as polyacrylic
acid (PAA), that is discussed in more detail later.
The historical trends in PWR secondary chemistry are shown in Figure 29.
5.02.4.2 Chemistry Effects on Materials
Degradation of SGs
Corrosion of SG tubes has been the major issue
affecting selection of secondary water chemistry
parameters. However, corrosion and FAC of SG
internals and other secondary system components
are also important concerns.
Corrosion of SG tube materials is mainly affected
by the following water chemistry related factors, in
addition to nonwater chemistry factors such as material susceptibility, temperature, and stress:
 pH Corrosion of several different types, including IGA/SCC and pitting, are strongly affected
by the local pH. High pH (caustic conditions)
and low pH (acidic conditions) accelerate the
rates of IGA/SCC.
 ECP The ECP is a measure of the strength of the
oxidizing or reducing conditions present at the

metal surface. The rate of corrosion processes are


strongly affected by the ECP. Secondary side SCC in
tube alloys tends to be accelerated by increases in
ECP, that is, by the presence of oxidizing conditions.
 Specific species Some impurity species accelerate
corrosion of tubing alloys as a result of their effects
on pH and ECP. In addition, lead and reduced
sulfur species (e.g., sulfides) appear to interfere
with formation of protective oxide films on the
tube metal surfaces, and thereby increase risks of
IGA/SCC, independent of influences on pH or
potential. Similarly, chlorides tend to increase the
probability of pitting.
These factors have been most thoroughly explored
for mill-annealed Alloy 600 (600MA). As discussed
in Chapter 5.04, Corrosion and Stress Corrosion
Cracking of Ni-Base Alloys, tests indicate that the
other tubing alloys, that is, stress-relieved Alloy
600 (600SR), thermally treated Alloy 600 (600TT),
nuclear grade Alloy 800 (800NG), and thermally
treated Alloy 690 (690TT), exhibit similar tendencies, but have increased resistance to corrosive attack,
in the order listed, with 690TT having the highest
resistance. Laboratory tests and plant experience indicate that 690TT has very high resistance to IGA/SCC
on the outside diameter on tubing (OD IGA/SCC) in

Water Chemistry Control in LWRs

41

Pb remediation
Dispersants
Titanium
Molar ratio control
MPA, DMA
ETA chemistry
Morpholine chemistry
Boric acid addition
EPRI water chemistry guidelines
Ammonia chemistry
Phosphate
1975

1980

1985

1990

1995

2000

2005

Figure 29 Evolution of water chemistry for pressurized water reactor secondary systems.

normally expected crevice conditions, but OD IGA/


SCC could possibly occur as a result of upsets or as a
result of long-term fouling and accumulation of aggressive species in deposit-formed crevices. Alloy 800NG
also has high resistance to OD IGA/SCC, but laboratory tests indicate that it is about twice as susceptible as
Alloy 690TT, and it has experienced limited amounts
of IGA/SCC in plants, while no operation-related
corrosion of 690TT has been reported. Laboratory
tests and some plant experience indicate that 600TT
is significantly more resistant than 600MA but less
resistant than 800NG and 690TT.
Water chemistry selected to protect SG tubes
appears to be satisfactory for most balance-of-plant
(BOP) components such as turbines. The main corrosion concerns in the BOP that affect secondary
system water chemistry are FAC of carbon steel piping, tubing, and heat exchanger internals and shells,
and ammonia attack of copper and copper alloy
tubes. In addition, FAC has also affected some recirculating SG internal components (e.g., feedrings,
swirl vanes). FAC is mainly influenced by the
at-temperature pH and oxygen content around the
secondary system. Ammonia attack of copper alloys
is mainly influenced by the concentrations of ammonia and oxygen at the copper alloy locations, but is
also accelerated by increases in concentrations and
pH associated with other amines, although not as
strongly as by increases in ammonia.
Once-through steam generators (OTSGs) have
different thermal hydraulics and (in original SGs)

tube materials than recirculating steam generators


(RSGs). These differences have led to OTSGs
having somewhat different tube corrosion experience than RSGs of the same vintage. For the most
part, OTSGs have experienced somewhat lower
rates of tube degradation. However, significant
IGA/IGSCC has been detected in the upper bundle
free spans of several units, especially at scratches,
and SG replacement has been performed or is
planned at all units.
The locations in SGs that are most affected by
IGA/IGSCC are those where free circulation of secondary water is impeded by the local geometry, for
example, in crevices formed by tube support plates or
by sludge piles that can accumulate on the tube sheet.
Impurities in the secondary water can concentrate in
these locations by boiling and evaporation in a process called hideout.
The key issue influencing water chemistry regimes
in PWR secondary system is to minimize SG degradation by controlling sludge buildup, reducing (and
balancing, e.g., MRC) the concentration of impurities
(i.e., sodium, chloride and sulfate) in deposits at the
tube-tubesheet and tube-tube support plate interfaces. The use of advanced amines to control pH has
increased significantly in the past few years, as discussed in a following section. Figure 30 shows the
main approaches used in typical chemistry control
strategies.
Impurities are removed from SGs by blowdown of
the coolant. Over the past 20 years or so, average

42

Water Chemistry Control in LWRs

Key issue: Mitigating IGA/IGSCC


in concentrating regions

Approach: Control local chemistry

Molar ratio control

Reduce Na
increase Cl

Reduce iron

Redox potential

Amines
dispersants

Reduce Cu
increase N2H4

Inhibitors

Boric acid
TiO2

Figure 30 Pressurized water reactor secondary chemistry control strategies.

blowdown impurity concentrations in US SGs have


been reduced from several ppb to the sub-ppb range.
Many PWRs today have SG blowdown concentrations near or below the analytical detection limit.
Minimization of impurities is recommended but has
been insufficient to prevent or completely mitigate
IGA/IGSCC at most plants with susceptible tube
material and design, as it can result in sodium-rich
feedwater. Cations such as sodium can be more effectively retained by boiling in a crevice than chloride.
Hence, excess cations over anions or anions over
cations result in specific corrosion issues because of
concentration processes in local environments.
The original all-volatile treatment used ammonia
to control pH, but a less-volatile chemical than ammonia would improve pH control throughout the circuit.
Early work employed morpholine, but now several
other amines are used. Since the initial application of
advanced amine chemistry about 15 years ago, there
has been tremendous success in reducing the transport
of corrosion products to the SGs by improving the attemperature pH around the BOP, especially in the
two-phase regions. This has resulted in mitigation of
FAC and thus reduced generation of corrosion products that ultimately get transported to the SGs. Ethanolamine (ETA) remains the most used amine at US
plants, with 75% of the US plants using ETA or
ETA with other amines, such as dimethylamine
(DMA) or 3-methoxypropylamine (MPA), to control
secondary cycle pH, as shown in Figure 31.17 Several
plants now use a mixture of amines to achieve the
optimum pH throughout the secondary system, with
25% of the US plants using MPA or MPA with other
amines while 12% of the plants use morpholine or
morpholine with other amines.

0%

4%
5%

4%

16%
56%

2%
7%
6%

ETA
MPA

ETA/DMA
ETA/MPA
MPA/DMA MPA/Morph
Morph/DMA

ETA/Morph
Morph

Figure 31 Amines used in the secondary systems of


US pressurized water reactors.

The proper control of oxygen in pressurized water


reactor (PWR) secondary feedwater, using an oxygen
scavenger such as hydrazine and/or carbohydrazide,
has been an enduring issue. The requirements for
oxygen concentration necessitate that some optimization take place. Maintaining reducing conditions
that is, low electrochemical potential in the SG is
essential to minimize SCC. On the other hand, some
oxygen in the feedwater counteracts corrosion of

Water Chemistry Control in LWRs

1985

< 1992

1993

1994

1995

1996

1998

43

2005

% of US PWRs using advanced water chemistries


120
99

100
80
60

76
62

73

40
58

40
68

44

44

20
23

Using advanced
amines

57

30

23

30 30

50

41

31 33

28

On molar ratio
control

Using >100 ppb


FW N2H4

35

19
37 41

17

Using boric acid


treatment

Figure 32 Pressurized water reactor secondary chemistry trends.

carbon steel surfaces and the transport of corrosion


products to the SG. Recent work has investigated the
effect of hydrazine and oxygen on the ECP of SG
tubing materials (Alloys 600 and 690) as well as
stainless steel (304 and 316) and carbon steel during
PWR startup conditions.
These laboratory studies have shown that changes
in the concentration of hydrazine, used to ensure a
reducing potential in the SG, within the typical range
allowed and employed (e.g., 20150 ppb) have no
discernable effect on FAC at feedwater temperatures
(e.g., 180235  C).
Figure 32 shows the trends in using advanced
water chemistry regimes in the secondary systems
at US plants.
5.02.4.3

Control of Sludge Fouling of SGs

Corrosion products in the secondary side of PWR


SGs primarily deposit on the SG tubes. These deposits can inhibit heat transfer, lead to thermalhydraulic
instabilities through blockage of tube supports, and
create occluded regions where corrosive species can
concentrate along tubes and within tube-to-tube support plate crevices. The performance of the SGs can
be compromised not only through formation of an
insulating scale, but also through the removal of tubes
from service due to corrosion.
Although the application of various amines to
control the at-temperature pH (pHT) in specific
locations of PWR secondary systems has been

successful in reducing the corrosion of BOP metals


and thus reducing corrosion product transport to
SGs, a complementary strategy now exists for significantly reducing SG fouling through online application of dispersant, which inhibits deposition. By
inhibiting deposition of the corrosion products, the
dispersant facilitates more effective removal from the
SGs via blowdown. This strategy has been employed
at fossil boilers for many decades. However, due to the
use of inorganic polymerization initiators (containing
sulfur and other impurities), polymeric dispersants had
not been utilized in the nuclear industry. Only recently
has a PAA dispersant been available that meets the
criteria for nuclear application, and progress has been
made in reducing SG fouling by application of an
online dispersant to substantially improve the efficiency of blowdown iron removal. Dispersant application is proving to be a highly promising technology for
markedly decreasing SG fouling, delaying (or possibly
eliminating) the need for expensive chemical cleaning
and effectively reducing the frequency of sludge lancing for SG maintenance.
Online application of PAA to the feedwater system has been successfully demonstrated to greatly
increase the efficiency of the blowdown system
in eliminating feedwater corrosion products from
fouling the SGs. The first application occurred at
Arkansas Nuclear One Unit 2 for a three-month
trial in early 2000, which demonstrated a significant
improvement in the blowdown iron removal efficiency from 2% to 60% with 46 ppb PAA in

44

Water Chemistry Control in LWRs

With dispersant

Iron removal efficiency

100

10
Without dispersant

1
0

0.5
1
1.5
2
Dispersant concentration (ppb)

Figure 33 Iron removal efficiency during dispersant test


at McGuire pressurized water reactor.

the feedwater. The second application, in 20052006,


at McGuire Unit 2 for a 69 month trial in their
replacement SGs tubed with Alloy 690TT showed
a similar significant improvement., as shown in
Figure 33.15
The following conclusions are evident from the
McGuire 2 demonstration described in the above
reference:
 PAA is an effective dispersant. A feedwater PAA
concentration approximately equal to the feedwater iron concentration (2 ppb in this case)
appears to effectively remove approximately 50%
of the influent feedwater iron under steady-state
operating conditions.
 Although blowdown copper spikes with initial
PAA application (albeit to a much lesser extent
than iron), it quickly returns to normal levels and
remains there.
 Filter element consumption is manageable.
 Blowdown cation conductivity and ammonia behavior changed during the trial, but these changes are
believed to be mainly due to changes in plant configuration and not PAA.
 The SG thermal performance level has improved
slightly with dispersant application, most likely
due to slight beneficial changes in the tube deposit
thermal properties.
The long-term trial at McGuire 2 demonstrated
the significant improvement in blowdown iron
removal efficiency with application of PAA dispersant (a follow up to the successful short-term trial
at ANO-2 in 2000). Based on the success of the
McGuire 2 long-term trial, evaluations are in progress with SG vendors looking toward technical
concurrence for long-term use in their fleet of recirculating SGs.

5.02.4.4

Lead Chemistry

PbSCC is a serious concern that can affect all SG


tubing materials currently employed. A better understanding of lead behavior is needed at SG and
feedwater temperatures before possible mitigation
techniques can be successfully developed. It is well
known that soluble lead at very low concentrations
can contribute to SCC of nickel alloys. Likewise, it
is well known that some locations on the secondary
side of PWR SGs will accumulate lead in the solid
state (i.e., deposit) with local concentrations considerably in excess of those observed to accelerate
cracking in laboratory testing. Recent investigations
using analytical transmission electron microscopy15
have identified lead in the cracks of many tubes
pulled from PWRs. However, the absence of extensive operating SG tube failures at rates comparable to
what might be predicted based on laboratory studies
of PbSCC indicates that some mitigating phenomenon could be present.15
EPRI has published a sourcebook on lead29 that
summarizes the state-of-knowledge regarding PbSCC
and its effects on PWRs. It incorporates PbSCC laboratory testing, the current understanding of lead
transport and other physical chemistry aspects of
lead, and the accumulated industry experience
regarding PbSCC and its mitigation.
Three clearly understood and accepted facts regarding lead in PWR secondary water systems became clear
as this sourcebook was being put together:
 In laboratory testing, the presence of lead accelerates SCC of mill-annealed 600MA, stress-relieved
600SR, and thermally treated 600TT stainless
steels as well as thermally treated Alloy 690
(690TT).
 In operating PWRs, lead is present in the secondary system.
 In two cases, a large ingress of lead to the secondary system has occurred as a result of lead blankets
having been left behind in SGs; the tubes in the
affected SGs cracked sooner and faster than in
the other SGs at the same units.
Set against these known facts are the following four
points:
 The mechanisms by which lead is transported from
its ultimate source to the SG tube and into a crack
are not well understood, and a comprehensive
evaluation of possible mechanisms has not been
performed.

Water Chemistry Control in LWRs

 The threshold concentration at which lead will


accelerate SCC in SGs is not well defined.
 No definitive indicator of PbSCC is available.
 There is no well-characterized mechanism by
which lead accelerates SCC.
Recent work has shown that adsorption/desorption
of Pb on corrosion products and SG tubing surfaces
could potentially be a major sink/source, respectively,
for Pb microscopy.15 There is no direct evidence of
adsorption in SGs; however, there is sufficient potential
for this mechanism that direct high-temperature measurements under SG conditions have been performed.
As a result of ongoing laboratory studies, microscopy15 speculates that formation of a lead layer slows
repassivation, after a passive film at the crack tip is
disrupted, potentially to an extent to which a crack
can initiate and propagate.

5.02.5 Chemistry Control for FAC in


BWRs and PWRs
FAC causes wall thinning of carbon steel piping,
vessels, and components, as discussed in Chapter
5.06, Corrosion and Environmentally-Assisted
Cracking of Carbon and Low-Alloy Steels. The
wall thinning is caused by an increased rate of dissolution of the normally protective oxide layer, for
example, magnetite, that forms on the surface of
carbon and low-alloy steels when exposed to highvelocity water or wet steam. The oxide layer reforms
and the process continues. If the thinning is not
detected in time, the reduced wall cannot withstand
the internal pressure and other applied loads. The
result can be either a leak or a complete rupture.
The rate of wall loss (wear rate) of a given component is affected by temperature, fluid bulk velocity,
the effect of component geometry on local hydrodynamics, the at-temperature pH, the liquid phase
dissolved oxygen concentration, and the alloy composition. The addition of chromium to steels decreases
the rate of FAC. Materials used to replace piping
damaged by FAC include low-alloy steels containing
chromium and molybdenum (P11, 1.25% Cr0.5%
Mo and P22, 2.25% Cr1% Mo) and carbon or lowalloy steels clad with stainless steel. Corrosion models
are used to estimate wall thinning and determine where
monitoring is required. An example of the approach
commonly used in the United States is described by
Chexal and Horowitz.30

45

The main chemistry factors that affect the rate


of FAC are pH and dissolved oxygen concentration. FAC is not an issue for PWR primary systems.
As indicated earlier, laboratory studies have shown
that changes in the concentration of hydrazine in the
PWR secondary system feedwater, used to ensure a
reducing potential in the SG, have no discernable
effect on FAC at feedwater temperatures, within the
typical range allowed and employed.
The chemistry parameter that a BWR plant has
some degree of control over is dissolved oxygen.
Oxygen affects the form and solubility of the oxide
layer, the dissolution of which is inherent in FAC.
Several plants inject oxygen into the system, as the
rate of FAC increases dramatically if the oxygen
concentration is less than about 25 ppb. Plant data
are shown in Figure 34.
Use of HWC in a BWR can significantly reduce
the amount of oxygen in the main steam, extraction
steam, and heater drain systems, thus potentially
increasing the FAC rates in these areas of the plant.
The effect of NMCA on the corrosion behavior
of carbon steel in 550  F (288  C) water containing
various amounts of oxygen and hydrogen has been
studied and the data confirm that there is no adverse
effect of NMCA on FAC.7
The carbon steel segments of the BWR vessel
bottom-head drain line have been identified as
being FAC susceptible because of the flow conditions
and the potential for low dissolved oxygen concentrations. However, a significant number of inspections
have been performed recently at US plants and little
thinning has been observed.
The 2008 edition of the BWR Water Chemistry
Guidelines7 recommends that feedwater oxygen
should be maintained above 30 ppb to minimize FAC
of carbon and low-alloy steels.

5.02.6 Water Chemistry Control


Strategies
Sometimes, step changes in chemistry strategy are
unavoidable, as with the move to reducing chemistry
in BWRs. In these cases, the operators must be
prepared to deal with adverse effects. Some BWRs
adopting reducing conditions experienced a large
jump in out-of-core radiation fields, which may be
avoided with prior zinc injection. Addition of new
chemicals requires extensive qualification. For example,
the successful demonstrations of BWR online noble

Water Chemistry Control in LWRs

Relative FAC wear rate


(expressed as percentage of the average wear
rate of components at 10 pbb oxygen)

46

160
140
Plant A
120

Plant A

100

Plant B
Plant C

80

Plant D
60

Average

40
20
0
0

10

20

30

40

50

60

70

80

Dissolved oxygen (ppb)


Figure 34 Plant data showing the relationship between flow-assisted corrosion and dissolved oxygen. (Oxygen values
are localized, calculated by the CHECWORKS codes from measured values at condensate or feedwater locations.)

chemistry and PAA dispersants in PWR SGs resulted


from detailed monitoring and evaluation during the
first injections. If possible, changes in chemistry
should be made in baby steps, with monitoring at
each step, before further changes are implemented.
Examples of this strategy are the gradual increases in
lithium/pH and dissolved hydrogen in PWR primary
systems. These incremental changes minimize adverse
side-effects and allow a planned approach to the optimum plant-specific chemistry control program.
The US nuclear power industry produces guidance documents to assist plant personnel in determining a plant-specific chemistry control strategy. The
early versions of these documents, developed in the
1980s, listed water chemistry specifications and
actions to be taken if the limits were exceeded. As
more chemistry options became available, the guidelines evolved into providing guidance on selecting the
most appropriate chemistry for a specific plant. Thus,
the 2008 BWR Water Chemistry Guidelines7 offers
recommendations on controlling ECP, zinc injection,
and feedwater iron control. Likewise, the 2007 PWR
Primary Water Chemistry Guidelines12 provides
guidance on pH control and zinc injection, and the
2008 PWR Secondary Water Chemistry Guidelines27
discusses impurity control, amines, and dispersants.
Theses documents are used by all US nuclear power
plants and provide the technical basis for similar
guidelines used in many other countries. Development of a strategic water chemistry plan, as discussed
in these documents, is seen as crucial to controlling
material degradation in the future.

References
1.

2.

3.
4.
5.
6.

7.
8.
9.

10.

11.

12.

Swan, T.; Wood, C. J. In Developments in Nuclear Power


Plant Water Chemistry, VIIIth International Conference on
Water Chemistry of Nuclear Reactor Systems, Oct 2326,
2000; BNES: Bournemouth, UK, 2000.
Fruzzetti, K.; Wood, C. J. In Developments in Nuclear
Power Plant Water Chemistry. International Conference on
Water Chemistry of Nuclear Reactor System, Jeju Island,
Korea, Oct 2326, 2006.
Cohen, P. Water Coolant Technology of Power Reactors;
Gordon and Breach: New York, 1969.
Jones, R. L. In International Water Chemistry Conference,
San Francisco, Oct 1115, 2004; EPRI: Palo Alto, CA,
2004.
Garcia, S.; Wood, C. Recent advances in BWR water
chemistry. In VGB NPC08 Water Chemistry Conference,
Berlin, Sept 1418, 2008.
Cowan, R.; Hussey, D. Radiation field trends as related to
chemistry in United States BWRs. In 2006 International
Conference on Water Chemistry of Nuclear Reactor
Systems, Jeju Island, Korea, Oct 2326, 2006.
EPRI. Boiling Water Reactor Water Chemistry Guidelines
2008 Revision; EPRI: Palo Alto, CA, 2008.
Edsinger, K. In Nuclear News; Tompkins, B., Ed.; 2008;
pp 3436.
Fruzzetti, K.; Perkins, D. PWR chemistry: EPRI
perspective on technical issues and industry research.
In VGB NPC08 Water Chemistry Conference, Berlin,
Sept 1418, 2008.
Andresen, P.; Ahluwalia, A.; Hickling, J.; Wilson, J.
Effects of PWR primary water chemistry on PWSCC of
Ni alloys. In 13th International Conference on
Environmental Degradation of Materials in Nuclear Power
Systems, Whistler, Canada, Aug 1923, 2007.
Andresen, P.; Ahluwalia, A.; Wilson, J.; Hickling, J.
Effects of dissolved H2 and Zn on PWSCC of Ni alloys.
In VGB NPC08 Water Chemistry Conference, Berlin,
Sept 1418, 2008.
EPRI. Pressurized Water Reactor Primary Water Chemistry
Guidelines: Revision 6; EPRI: Palo Alto, CA, 2007.

Water Chemistry Control in LWRs


13. Pathania, R.; Yagnik, S.; Gold, R. E.; Dove, M.;
Kolstad, E. Evaluation of zinc addition to PWR
primary coolant. In 7th International Symposium on
Environmental Degradation of Materials in Nuclear
Power Systems, Breckenridge, CO, NACE: Houston,
TX, 1995; pp 163176.
14. Molander, A.; Jenssen, A.; Norring, K.; Konig, M.;
Andersson, P.-O. Comparison of PWSCC initiation
and crack growth data for Alloy 600. In VGB
NPC08 Water Chemistry Conference, Berlin,
Sept 1418, 2008.
15. Fruzzetti, K.; Rochester, D.; Wilson, L.; Kreider, M.;
Miller, A. Dispersant application for mitigation of steam
generator fouling: Final results from the McGuire
2 long-term trial and an industry update and EPRI
perspective for long-term use. In VGB NPC08 Water
Chemistry Conference, Berlin, Sept 1418, 2008.
16. Haas, C.; Ahluwalia, A.; Kucuk, A.; Perkins, D. PWR
operation with elevated hydrogen. In VGB NPC08 Water
Chemistry Conference, Berlin, Sept 1418, 2008.
17. Hussey, D.; Perkins, D.; Choi, S. Benchmarking
radioactivity transport and deposition in PWRs. In VGB
NPC08 Water Chemistry Conference, Berlin, Sept 1418,
2008.
18. Stevens, J.; Bosma, J. Elevated RCS pH program at
Comanche peak. In VGB NPC08 Water Chemistry
Conference, Berlin, Sept 1418, 2008.
19. Nordmann, F. Worldwide chemistry objectives and
solutions for NPP. In VGB NPC08 Water Chemistry
Conference, Berlin, Sept 1418, 2008.

20.

21.

22.
23.
24.

25.
26.
27.
28.
29.
30.

47

Tigeras, A.; Stellwag, B.; Engler, N.; Bretelle, J.; Rocher, A.


Understanding the zinc behavior in PWR primary coolant:
A comparison between French and German experience.
In VGB NPC08 Water Chemistry Conference, Berlin,
Sept 1418, 2008.
Frattini, P. L.; Blok, J.; Chauffriat, S.; Sawicki, J.; Riddle, J.
In VIIIth International Conference on Water Chemistry of
Nuclear Reactor Systems, Oct 2326, 2000; BNES:
Bournemouth, UK, 2000.
Riess, R. Personal communication, 2008.
Byers, W.; Wang, G.; Deshon, J. Limits of zinc addition in
high duty PWRs. In VGB NPC08 Water Chemistry
Conference, Berlin, Sept 1418, 2008.
Perkins, D.; Ahluwalia, A.; Deshon, J.; Haas, C. An EPRI
perspective and overview of PWR zinc injection. In VGB
NPC08 Water Chemistry Conference, Berlin, Sept 1418,
2008.
Millett, P. J; Hundley, F. Nucl. Energ. 1997;
36, pp 251258.
EPRI. Personal communication from K. Fruzzetti, 2009.
EPRI. Pressurized Water Reactor Secondary Water
Chemistry Guidelines Revision 6; EPRI: Palo Alto, CA, 2008.
Choi, S. Personal communication, 2009.
EPRI. Pressurized Water Reactor Lead Sourcebook; EPRI:
Palo Alto, CA, 2006.
Chexal, V.; Horowitz, J. ChexalHorowitz flowaccelerated corrosion model Parameters and influences.
In ASME PVP-Vol B, Current Perspectives of International
Pressure Vessels and Piping Codes and Standards, Book
No. H0976B, 1995.

5.03

Corrosion of Zirconium Alloys

T. R. Allen
University of Wisconsin, Madison, WI, USA

R. J. M. Konings
European Commission, Joint Research Centre, Institute for Transuranium Elements, Karlsruhe, Germany

A. T. Motta
The Pennsylvania State University, University Park, PA, USA

2012 Elsevier Ltd. All rights reserved.

5.03.1
5.03.2
5.03.2.1
5.03.2.2
5.03.2.3
5.03.3
5.03.3.1
5.03.3.2
5.03.3.3
5.03.3.4
5.03.3.4.1
5.03.3.4.2
5.03.3.4.3
5.03.4
5.03.5
5.03.5.1
5.03.5.2
5.03.5.3
5.03.5.4
5.03.5.5
5.03.6
5.03.7
References

Introduction
General Considerations
Oxidation
Hydrogen Uptake
Controlling Factors for Corrosion
Uniform Oxidation
Mechanism
Temperature and Heat Flux
Coolant Chemistry
Irradiation Effects
Radiolysis
Irradiation effects in the oxide layer
Changes in the metallurgical state of the metal
Nodular Oxidation
Hydrogen Embrittlement
Hydrogen Production During Aqueous Corrosion of Zirconium-Base Materials
Hydrogen Absorption
Hydride Formation
Hydride Formation Rates
Formation of Hydride Rim
Delayed Hydride Cracking
Summary and Outlook

Abbreviations
BWR
CANDU
CRUD
DHC
IAEA
M5TM
PWR
tHM
VVER
ZIRLOTM
Zry

Boiling water reactor


Canadian Deuterium Uranium
Chalk River unidentified deposits
Delayed hydride cracking
International Atomic Energy Agency
Zirconium alloy material with niobium
(AREVA)
Pressurized water reactor
Ton heavy metal
Voda Voda Energy Reactor
Zirconium alloy material with niobium,
tin, and iron (Westinghouse)
Zircaloy

49
50
50
51
52
53
53
57
57
59
59
60
60
61
61
62
62
62
63
64
65
66
66

5.03.1 Introduction
Zirconium alloys are widely used for fuel cladding
and in pressure tubes, fuel channels (boxes), and fuel
spacer grids in almost all water-cooled reactors: light
water reactors such as the pressurized water reactor
(PWR) and the boiling water reactor (BWR) as well
as the Canadian designed Canadian Deuterium
Uranium (CANDU) heavy water reactor. Since
its employment in the first commercial nuclear
power plant (Shippingport) in the 1960s, Zircaloy, a
zirconiumtin alloy, has shown satisfactory behavior
during many decades. However, degradation due to
waterside corrosion can limit the in-reactor design life
of the nuclear fuel. The critical phenomenon is the

49

50

Corrosion of Zirconium Alloys

hydrogen ingress into the cladding during corrosion,


which can cause cladding embrittlement. As utilities
are striving to achieve higher fuel burnups, the
nuclear industry has made several efforts to understand the mechanisms of corrosion and to mitigate
its effects.
In striving for increased burnup of the nuclear fuel
from 33 000 to 50 000 MWd/tHM and beyond in
PWRs, associated studies have shown that the corrosion of the Zircaloy-4 cladding accelerates under
these higher burnup conditions. Although alloys
that are more modern have not yet shown evidence
of this high-burnup acceleration, this is a potential
concern. Also, the efforts to increase the thermalcycle efficiency in PWRs by operating at higher
temperatures (power uprates), combined with the
more aggressive chemistry (introduction of B and Li
for example) related to the use of high-burnup fuel,
have resulted in increased fuel duty,1 and in increased
corrosion rates. This has led to the introduction
of cladding tubes of new zirconium alloys such as
zirconiumniobium, which are much more corrosion
resistant.2,3 With the introduction of these materials,
the nuclear industry aims at zero tolerance for fuel
failure in the future.4
Many reviews on the corrosion of zirconium alloys
both out- and in-reactor, have been published.511
The extensive reviews made by an international
expert group of the International Atomic Energy
Agency (IAEA) and published as IAEA-TECDOCs
684 and 99612,13 are major references in this respect.
As mentioned by Cox,6,7 the number of publications
on this topic is so enormous that it is impossible for a
short review to be comprehensive. This also applies
to the current chapter, which therefore focuses on the
main issues, naturally relying on the above-mentioned
existing reviews and updating the information where
possible with new results and insights.

protective, thus limiting the access of oxidizing species


to the bare metal. Much evidence exists to indicate
that Zr oxidation occurs by inward migration of oxygen ions through the oxide layer, either through
grain boundaries or through the bulk.5,12,13
Zr O2 ZrO2
As shown in Figure 1, the growth of the oxide
layer on the metal surface depends on the kinetics of
the oxygen diffusion through this layer. Because the
corrosion kinetics slow down as the oxide thickness
increases, it has been argued that the rate controlling
step in the oxidation process is the transport of atomic
species in the protective oxide, by either oxygen diffusion through the oxide film14,15 or diffusion of electrons through the oxide film. These processes are
necessarily coupled to maintain electroneutrality.
Electron transport is, however, difficult in zirconium
dioxide, as it is an electrical insulator when undoped.
Although this is not positively confirmed, it is likely
that the role of doping elements in the determination
of corrosion kinetics is done through their influence
on the electron or oxygen transport in the oxide layer.
Several types of corrosion morphologies have
been observed in nuclear reactors and in autoclave
experiments, of which the most important are
1. Uniform: The formation of a thin uniform layer of
zirconium dioxide on the surface of a zirconium
alloy component (see Figure 2).
2. Nodular : The formation of local, small, circular
zirconium oxide blisters (see Figure 3).
3. Shadow: The formation of local corrosion regions
that mirror the shape (suggestive of a shadow)
of other nearby noble reactor core components
(Figure 4).

H2O O2 + 2H+

Coolant

5.03.2 General Considerations


5.03.2.1

Oxidation

Corrosion of zirconium alloys in an aqueous environment is principally related to the oxidation of the
zirconium by the oxygen in the coolant, dissolved
or produced by radiolysis of water. A small amount
of oxygen can be dissolved in the metal, but once the
thermodynamic solubility limit is exceeded, ZrO2 is
formed on the metal. (All zirconium components normally have a thin oxide film (25 nm) on their surface
in their as-fabricated state.) The oxide formed is

H+
O2

H+

Oxide
H+ + e

Zr + 2O2 ZrO2 + 4e

H0

Metal

Figure 1 Schematic presentation of the corrosion of the


zirconium alloys. Corrosion of zirconium alloys in nuclear
power plants; TECDOC-684; International Atomic Energy
Agency, Vienna, Austria, Jan 1993.

Corrosion of Zirconium Alloys

51

Zr + H2O = ZrO2 + H2

ZrH2x
ZrO2

Figure 2 Uniform oxide layer formation and hydride precipitation in Zircaloy cladding. European Atomic Energy
Commission.

The occurrence of these morphologies is strongly


dependent on the reactor operating conditions and
chemical environment (particularly the concentration of oxygen in the coolant), which are distinctly
different in PWRs, BWRs, and CANDU (Table 1).
In both BWRs and PWRs, a uniform oxide layer is
observed, although its thickness is normally greater in
PWR than in BWR, primarily because of the higher
operating temperature. Nodular corrosion occurs
occasionally in BWRs because a much higher oxygen
concentration occurs in the coolant because of water
radiolysis and boiling. Shadow corrosion is also occasionally observed in BWRs and is a form of galvanic
corrosion.

1 mm

5.03.2.2

100 m

Figure 3 General appearance of nodules formed on


zirconium alloy following a 500  C steam test at 10.3 MPa. In
the bottom, a cross-section view of a nodule is shown,
exhibiting circumferential and vertical cracks. Photo courtesy
of R. Ploc and NFIR (Nuclear Fuel Industry Research Group).
Reproduced from Lemaignan, C.; Motta, A. T. Zirconium
Alloys in Nuclear Applications, Materials Science and
Technology, Nuclear Materials Pt. 2; VCH
Verlagsgesellschaft mbH, Weinheim, Germany, 1994.

Hydrogen Uptake

The formation of an oxide layer would not bring


severe consequences to cladding behavior were it
not for the fact that in parallel with the corrosion
process, a fraction of the hydrogen, primarily produced by the oxidation reaction as well as by radiolysis of water, diffuses through the oxide layer into the
metal. Zirconium has a very low solubility for hydrogen (about 80 wt ppm at 300  C and 200 wt ppm at
400  C) and once the solubility limit is exceeded, the
hydrogen precipitates as a zirconium hydride phase
(Figure 2):
ZrH; sln H2 ZrH1:6 or ZrH2
As a result, the following effects have been reported
(although not all confirmed) to occur in the cladding:
hydrogen embrittlement due to excess hydrogen
or its localization into a blister or rim,16,17 loss of

52

Corrosion of Zirconium Alloys

fracture toughness, delayed hydride cracking (DHC),


and acceleration of corrosion and of irradiation
growth. Hydrogen embrittlement impacts the
mechanical resistance of the Zircaloy cladding to

failure and it is thus of key importance to understand


its underlying mechanisms. The ductility reduction
due to hydrogen embrittlement is dependent on the
volume fraction of hydride present, the orientation of
the hydride precipitates in the cladding, and their
degree of agglomeration.18,19

Oxide

5.03.2.3

Oxide

(b)

(a)
Figure 4 Zirconium oxides near (b) and away from
(a) a stainless steel control blade bundle, showing the effect
of shadow corrosion. Reproduced from Adamson, R. B.;
Lutz, D. R.; Davies, J. H. Hot cell observations of shadow
corrosion phenomena. In Proceedings Fachtagung der
KTG-Fachgruppe, Brennelemente und Kernbautelle,
Forschungszentrum Karlsruhe, Feb 29Mar 1, 2000.
Table 1

Controlling Factors for Corrosion

The oxidation and hydrogen uptake of Zircaloy is of


course determined by many factors. First of all, the
chemical and physical state of the material: composition, metallurgical condition, and surface condition.
These conditions are often specific to the material
and sometimes batch-specific and also related to the
fabrication process, as discussed in detail in Chapter
2.07, Zirconium Alloys: Properties and Characteristics. This is evident from the different behavior
of Zircaloy and ZrNb alloys, as shown in Figure 5
for two different zirconium alloys employed in the
French PWRs, Zircaloy and Zr1% Nb (M5). The
peak oxide layer thickness of Zircaloy-4 (oxide thickness at the hottest fuel grid span) increases significantly with burnup (i.e., residence time in the
reactor), whereas that of Zr1%Nb shows a moderate
increase.
In addition, a number of environmental factors
affecting the corrosion of zirconium alloys must be
considered:
1. Coolant Chemistry: It is obvious that the dissolved
oxygen and hydrogen play a major role in the
corrosion process, but other dissolved species
must also be taken into account. To control the
pH of the coolant at slightly alkaline conditions,

Typical reactor environments to which the zirconium alloys are exposed

Coolant
Inlet temperature ( C)
Outlet temperature ( C)
Pressure (MPa)
Neutron fluxa (n cm2 s1)
Coolant chemistry
[O2] (ppb)
[H2] (ppm)
pH
B (as H3BO3) (ppm)
Li (as LiOH) (ppm)
Na (as NaOH) (ppm)
K (as KOH) (ppm)
NH3 (ppm)

BWR

PWR

VVER

CANDU

H2O
272278
280300
7
47  1013

H2O
280295
310330
15
69  1013

H2O
290
320
15
57  1013

D2O
255
300
10
2  1012

200
0.03
7

<0.05
25
6.97.4
02200
0.55

< 0.1

<5
0.51
10.210.8

0140016
0.050.6
0.030.35
520
630

a
E > 1 MeV.
Corrosion of zirconium alloys in nuclear power plants; TECDOC-684; International Atomic Energy Agency, Vienna, Austria, Jan 1993.

Corrosion of Zirconium Alloys

53

60
50
Oxide thickness (mm)

M5
Zirc-4

40
30
20
10
0
0

20

40

60

80

Burnup (MWd kgU1)


Figure 5 Peak oxide layer thickness as a function of burnup for Zircaloy-4 and Zr1%Nb (M5). Reproduced from Bossis, P.;
Pecheur, D.; Hanifi, K.; Thomazet, J.; Blat, M. J. ASTM Int. 2006, 3(1), Paper ID JAI12404.

LiOH is added and H3BO3 (boric acid) is added


for reactivity control in PWRs. Furthermore,
impurities (Cl, F) and coolant-borne species (Cu,
Ni, etc.) must be considered.
2. Radiation: In reactor, the Zircaloy and the coolant
are subjected to the effects of energetic particles.
The principal effect is the production of oxidizing
species such as O2 in the coolant.
3. Temperature : In the range of water reactor operation
(240330  C), the combined effect of temperature
and radiation on zirconium alloy oxidation and
hydriding have been characterized extensively,
varying from almost no effect to acceleration of
oxidation by factors of up to two orders of magnitude, depending on environment and radiation level.
4. In addition, the presence of boiling and CRUD
(the term CRUD stands for Chalk River unidentified deposits, the nuclear power plant in which the
effect was observed for the first time) deposition in
PWR can enhance corrosion.

5.03.3 Uniform Oxidation


5.03.3.1

Mechanism

Uniform corrosion is defined as a process that occurs


approximately with the same speed on the entire
surface of an object (ISO 8044). It can be considered
as an electrochemical cell process, in which the metal
is anodically oxidized:
Zr 2O2 ZrO2 2Vo 4e
O

o
where VO
indicates a lattice vacancy in the ZrO2
layer. The corresponding cathodic reaction at the
oxide/coolant interface can be the reduction of
water:

2H2 O 4e 2Vo 2O2 4H


O

or, when the water contains dissolved oxygen:


2H2 O O2 2VOo 4e 4OH
The oxygen ions diffuse preferentially via the oxide
crystallite boundaries to the oxide/metal interface,
whereas the vacancies diffuse in the opposite direction. The hydrogen can combine with electrons to
form atomic/molecular hydrogen that dissolves in
the coolant water or diffuses to the metal.
Uniform corrosion is a passivating event since a
protective layer of zirconium oxide is formed as a
result of the reaction with the O2 ions or the OH
radicals. Electron microscopy shows that the oxide
layer is microcrystalline, initially equiaxed, later
growing into columnar grains that are formed in a
dense packing, of which the mean crystallite size
increases as the oxide thickens.15 Figure 6 shows
typical microstructures of the oxide layer for several
zirconium-based cladding materials. Figure 6(c), in
particular, shows the columnar grains extending right
near the oxide/metal interface.
The corrosion kinetics have been studied extensively. As mentioned above, because the corrosion
rate slows down with oxide thickness, the rate
controlling step is thought to be the transport of
oxidizing species in the layer.15,20 During corrosion,

54

Corrosion of Zirconium Alloys

Zircaloy-4

ZIRLO

(a)

Zr2.5Nb

(b)

(c)

150 nm

150 nm

150 nm

Figure 6 Grain size, shape, and orientation comparison near the oxide/metal interface of (a) Zircaloy-4, (b) ZIRLO, and
(c) Zr2.5Nb alloy oxides formed in 360  C pure water environments. The hand-drawn sketch below each bright-field image
shows oxide crystallite grain boundaries. Black arrows indicate oxide growth direction. Reproduced from Yilmazbayhan, A.;
Breval, E.; Motta, A. T.; Comstock, R. J. J. Nucl. Mater. 2006, 349, 265281.

a potential develops across the oxide layer. The negative potential at the oxide/metal interface accelerates the electron migration process and retards the
O2 diffusion until both operate at the same rate.
Bossis et al.22 argue that the surface reactions are
rate-determining in some Nb-containing alloys.
The measurements of the weight gain kinetics for
zirconium and its alloys (the weight gain is due to
oxygen ingress and follows the overall corrosion
kinetics) were found to fall into two stages, referred
to as pre- and posttransition. For constant temperature and pressure, the pretransition corrosion kinetics
are independent of pH between about 1 and 13 (if no
specifically aggressive species such as LiOH are present) and of the source of the oxygen. The kinetics of
the pretransition oxide layer formation, as measured
from weight gain (DW ), have been found to approximately follow a cubic rate law21:
3

DW k1 t

1

where k1 is the preexponential factor and t is time.


More recent results have shown that the rate law

depends on the alloys according to (DW)n kt, with


n between 2 and 5.22 The temperature dependence of
k1 follows an Arrhenius-type equation:


Q1
2
k1 B1 exp
RT
where B1 is an empirical constant, R is the universal gas constant, T is the absolute temperature,
and Q1 is the activation energy for pretransition
oxidation. The values for B1 are Q1 are obtained empirically from fitting of experimental data, for example,
B1 6.36  1011 (mg dm2)3 per day and Q1/R
13640 K, as found by Kass21 for Zry-2 and Zry-4.
The posttransition kinetics, on the contrary, are
approximately linear (n 1) in time23:
DW k2 t C
with
k2 B2 exp



Q2
RT

3
4

and C the weight gain at transition. B2 is the empirical


constant and Q2 the activation energy for posttransition

Corrosion of Zirconium Alloys

oxidation. Hillner et al.23 discussed the results of


numerous analyses of experimental corrosion studies
on Zircaloy with varying time and temperature to
derive B2 and Q2. As discussed by these authors,
most studies suffer from paucity of data for extended
exposures. Their own results for Zry-2 and Zry-4
cover a wide range of time and weight gain and the
posttransition kinetics were interpreted to consist of
two linear stages (with a change at about 400 mg dm2
or about 30 mm) with B2 2.47  108 mg dm2 day1
and Q2/R 12880 K for stage 1, and B2 3.47  107
mg dm2day1 and Q2/R 11452 K for stage 2.
Whether or not Hillners interpretation of a change
in mechanism is correct, certainly the data is best
described by a two-stage empirical fit.
A schematic representation of these pre- and posttransition kinetics is shown in Figure 7 as the dashed
lines. Also shown in this graph is the more recent
view that three stages can be discriminated for zirconium alloy corrosion processes23:
1. The early pretransition regime, characterized by
the formation of a thin, black, tightly adherent
corrosion film that grows thicker in accordance
with a nearly cubic rate law.
2. The intermediate stage that lies between the preand posttransition stages. As initially shown by
Bryner,24 this region appears to comprise a series
of successive cubic curves, similar to the initial
cubic kinetic curve. This linear rate results from
the superposition of various regions of the oxide
layer following pretransition growth rate but
slightly out of phase with each other.

Oxide thickness

Posttransition
(linear)
Transitory
(cyclic)

Pretransition
(cubic)

Time

Figure 7 Schematic representation of the zirconium alloy


corrosion showing the pretransition, transitory, and
posttransition regions. The dashed lines indicate early
models that recognized only the pre- and posttransition
regimes. Reproduced from Hillner, E.; Franklin, D. G.;
Smee, J. D. J. Nucl. Mater. 2000, 278, 334.

55

3. The linear posttransition kinetic regime.


In the very early stages of the oxide formation, the
layer is dense and composed of grains that have a
predominantly tetragonal or cubic structure. As the
grains grow, columnar grain growth is established and
the tetragonal grains tend to transform to monoclinic
oxide, which constitutes the majority of the oxide
formed.20 Although the tetragonal phase has often
been associated with protective behavior, this correlation is noncausal and in fact, oxides with lower tetragonal fraction have been found to be more protective.26,27
The diffusion of oxygen takes place along the grain
boundaries in the oxide layer,4 the kinetics of which
are given by eqn [1]. The size of the columnar grains
and their grain-to-grain misorientation (Figure 6)
have been related to the transition thickness.
Studies of Zircaloy corrosion in autoclaves clearly
reveal the cyclic corrosion kinetics,20,24 the oxide
layer appearing to be composed of successive layers
of 23 mm thickness (Figures 810), for which the
oxidation kinetics progressively decrease as a result
of the growth of the oxide layer, in accordance with
eqn [1]. The cycles are separated by transitions during which the kinetics appears to accelerate. The
transitions are caused by the destabilization of the
oxide layer, as a result of which the passivating layer
becomes porous and fractured at the end of the cycle,
losing its protective role, and reopening for rapid
oxidation. A new oxidation cycle then starts. Several
processes have been suggested for the destabilization
of the oxide layer, such as7,2527:
(a) Cracking of the oxide as a result of the accumulation of compressive stresses in the oxide from
imperfect accommodation of the volume expansion attendant upon oxide formation.
(b) Cracking of the oxide as a result of the transformation of initially tetragonal ZrO2 to the monoclinic modification,10 or as a result of the
oxidation of intermetallic precipitates initially
incorporated in metallic form, both of which
result in a volume increase.
(c) The porosity formed in the oxide reaches a percolation condition, leading to easy access of the
coolant to the underlying metal.
The first factor is normally considered to be the main
driver, although the other factors have also been
proposed to contribute. The levels of stress accumulation depend on the phase transformation tensor
(various levels of accommodation of the PillingBedworth strains in the in-plane directions), which

56

Corrosion of Zirconium Alloys

Oxide thickness (m)

15
12
Zircaloy-4
9
M5

6
3
0

200

400

600

800

Time (days)
Figure 8 Results of oxidation tests of Zircaloy-4 and of M5 in autoclaves, at 360  C, with 10 ppm Li and 650 ppm B,
showing the cyclic nature of the oxidation. Redrawn from Bataillon, C.; Feron, D.; Marchetti, L.; et al. E-DEN Monograph
Corrosion Commissariat a` lEnergie Atomique; 2008.

Zr4

Zr4

20 mm

20 mm

ZIRLO

ZIRLO

Figure 9 Optical micrographs of oxide layers formed in Zircaloy-4 and in ZIRLO, in reflected (left) and transmitted
light (right). The regular periods formed during the cyclic corrosion process correspond to the oxide transitions in the two
alloys. Photo courtesy of G. Sabol, Westinghouse Electric Co.

has been shown to vary from alloy to alloy, thus likely


causing the consistent differences seen among the
oxide thicknesses at transition for various alloys.
Thus, each alloy has a reproducible transition thickness
in a given environment. This cyclic process has been
shown to reproduce itself with remarkable regularity
upward of 17 transitions,26,27 as shown in Figure 9. This
can also be seen in the SEM micrograph in Figure 10
which suggests that cracking occurs at transition.

As discussed by Battaillon et al.,25 the kinetics of the


cyclic process can be described by a succession of
equations similar to [1] and [2], each representing a
specific cycle. The length of the cycle seems to
be material dependent as shown in Figure 8. Also,
Zircaloy contains second phase precipitates of Zr(Cr,
Fe)2 and tin as a dissolved element (see Chapter 2.07,
Zirconium Alloys: Properties and Characteristics).
The intermetallic precipitates are known to have a

Corrosion of Zirconium Alloys

57

1000

Weight gain (mg dm2)

800

10 m
Figure 10 The oxide layer formed on M5 in autoclaves
at 360  C, with 10 ppm Li and 650 ppm B dissolved in
the water showing the layered nature of the oxide, with
periodic cracking. Bataillon, C.; Feron, D.; Marchetti, L.;
et al. E-DEN Monograph Corrosion Commissariat a`
lenergie atomique, 2008. From DEN Monographs
Corrosion and Alteration of Nuclear Materials, ISBN
978-2-281-11369-3 (2010), editions du Moniteur, CEA.

600

400

200

280

300

320
Temperature (C)

340

360

Figure 12 The effect of temperature on the oxidation


kinetics of Zircaloy-4, as derived from autoclave test in
water for 2500 days. Reproduced from Hillner, E.; Franklin,
D. G.; Smee, J. D. J. Nucl. Mater. 2000, 278, 334.

experimentally. As shown in Figure 12, the corrosion


kinetics accelerate above about 310  C. An increase
of 5  C for a typical cladding temperature of 335  C
results in a 26% increase in weight gain.
The temperature of the metaloxide interface (Ti)
is, however, not only dependent on the temperature of
the coolant, but also on the heat flux (f in W cm2):

higher oxidation resistance than the zirconium


matrix.28,29 When the oxidation of the zirconium progresses, the Zr(Cr,Fe)2 precipitates are incorporated
in metallic form into the oxide layer (Figure 11).
However, the iron is progressively dissolved in the
zirconium oxide. Tin is present in the oxide layer as
nanoparticles of b-Sn, SnO, or Sn(OH)2. The slower
oxidation kinetics of ZrNb alloys have been attributed to the absence of the second phase precipitates.7

fe
l
where Ts is the temperature at the wateroxide
layer boundary, e the oxide layer thickness (in cm),
and l the thermal conductivity of the oxide layer
(W cm1 K1). Considering that zirconium oxide is
a poor thermal conductor, the oxide layer will act
as an insulator increasing the temperature of the
metaloxide interface. For typical values for a
PWR (f 55 W cm2) and a thermal conductivity
of 0.022 W cm1 K1, the interface temperature
increases 1 K for an oxide layer of 4 mm.25
As a related effect, nucleate boiling can occur at the
oxidewater boundary, once this boundary reaches
the saturation temperature (344.5  C at 15.5 MPa in
a PWR). As a result, an enrichment of Li in the liquid
phase near the oxidewater boundary can occur
(Figure 13), which can reach a factor of 3.25 This is
not expected to increase the corrosion significantly
for conditions typical for PWRs.

5.03.3.2

5.03.3.3

100 nm
Figure 11 Zr(Cr,Fe)2 precipitates incorporated in metallic
form into the oxide layer on Zircaloy-4. Adapted from
Pecheur, D.; Lefebvre, F.; Motta, A. T.; Lemaignan, C.;
Charquet, D. Oxidation of Intermetallic Precipitates in
Zircaloy-4: Impact of Irradiation. In 10th International
Symposium on Zirconium in the Nuclear Industry, ASTM
STP 1245; Baltimore, MD, 1994; 68770; Pecheur, D.;
Lefebvre, F.; Motta, A. T.; Lemaignan, C.; Wadier,
J. F. J. Nucl. Mater. 1992, 189, 2318332.

Temperature and Heat Flux

An increase of temperature increases the oxidation


kinetics, as is evident from eqn [1], and confirmed

Ti  Ts

Coolant Chemistry

The corrosion of Zircaloy is influenced by the chemical composition of the coolant. The PWR coolant

58

Corrosion of Zirconium Alloys

activation.) In addition, the coolant may contain


small concentrations of anionic impurities that play a
role in the corrosion mechanism (Figure 14).
Extensive research has been performed to understand the role of lithium hydroxide and boric acid on
the kinetics of the corrosion of zirconium alloys.
Experiments in autoclaves have shown that the rate
of oxidation of Zircaloy-4 increases significantly
when boric acid is absent.25 After an initial stage
where the corrosion kinetics are as expected, corrosion is accelerated in conjunction with a decrease of
the thickness of the protective oxide layer,30,31 as
derived from microscopic observations, especially
by the ingress of Li into the oxide (Figure 15).
Enhanced dissolution of the crystallite grain boundaries has been suggested as the mechanism.32 This
effect was absent in the presence of boric acid, and no
significant difference was observed for the oxidation
kinetics for LiOH concentrations between 70 and
1.5 ppm (Figure 14). The protective effect of boric
acid has been suggested to be related to the plugging
of the porosity in the oxide by a borate compound.33
The coolant chemistry also influences the solubility of coolant-borne metallic impurities (e.g., iron,
nickel, copper, etc. arising from corrosion release
from circuit surfaces), which may deposit on fuel
rod surfaces as CRUD, which is composed of metal
oxides such as Fe2O3 (hematite), Fe3O4 (magnetite),
FeOOH (goethite), or (Ni,Co)xFe3-xO4 (spinel).3436
Such CRUD deposits are occurring specifically at
positions with sub-cooled boiling and may have, in
some cases, appeared to contribute to accelerated

contains boron and lithium. Boron, present as boric


acid (10002000 ppm at the beginning of the cycle,
depending on the cycle length, and about zero at the
end of the cycle), is added to control the core reactivity through neutron absorption of 10B. The boric acid
is weakly dissociated, particularly at high temperature,
which could lead to a slightly acidic environment. To
counteract this, small quantities of lithium hydroxide
(510 ppm) are added in the water, to obtain a slightly
alkaline pH, to avoid deposition of corrosion products
on the cladding and limit the corrosion of core structures made of stainless steel or Inconel alloys. (Lithium
enriched over 99% of 7Li is used, as the use of
6
Li produces the undesirable tritium through
Oxide

Water

Steam
bubble

Enrichment of species
of low volatility

Figure 13 Schematic representation of the enrichment of


species at the oxidewater boundary during nucleate
boiling. Adapted from DEN Monographs Corrosion and
Alteration of Nuclear Materials, ISBN 978-2-281-11369-3
(2010), editions du Moniteur, CEA.
20
70 ppm Li
(B=0)

18

Oxide thickness (m)

16
10 ppm Li
650 ppm B

14
12

3.5 ppm Li
1000 ppm B

10
70 ppm Li
650 ppm B

1.5 ppm Li
650 ppm B

6
4
2
0
0

100

200

300

400

500

Time (days)
Figure 14 The effect of Li and B on the oxidation kinetics of Zircaloy-4. Bataillon, C.; Feron, D.; Marchetti, L.; et al. E-DEN
Monograph Corrosion Commissariat a` lEnergie Atomique, 2008. From DEN Monographs Corrosion and
Alteration of Nuclear materials, ISBN 978-2-281-11369-3 (2010), editions du Moniteur, CEA.

Corrosion of Zirconium Alloys

increased significantly above the coolants nominal


level, the increased corrosion caused by CRUD
deposits is thought to be due to the role of Li, in
combination with the increased metal/oxide interface temperature.37
Fluorine is produced in the coolant water by neutron capture in 18O to give 19F. Laboratory experiments have shown that the corrosion of Zircaloy-4
begins to accelerate between 19 and 190 ppm fluorine
at 360  C, which is well above the coolant specification in most reactors (0.15 ppm).

16
14
12
Oxide thickness (mm)

59

10
8
6
4

5.03.3.4

A wealth of information exists on the in-reactor


behavior of Zircaloy from worldwide fuel monitoring
programs as well as from experimental research programs, from which information about the radiation
effects on zirconium alloy corrosion can be deduced.
These effects can be of multiple origin and include
radiolysis of the coolant, changes in the metallurgical
condition, displacement damage, or phase transformations. As discussed by Cox,7 no in-reactor effects
are evident for an oxide layer thickness below
56 mm. Above that thickness, a departure from data
with radiation field and for in-reactor conditions
suggests an irradiation-induced acceleration of the
oxide breakdown. Bataillon et al.25 suggest a factor
of 2 for the oxidation rate of Zircaloy-4 between
in-reactor and autoclave experiments during the
first two reactor cycles. This increases to about 4
during cycles 5 and 6. This effect is shown in
Figure 17, in which the thickness of the oxide layer
on Zircaloy-4 as a function of exposure time is compared for three cases: (a) autoclaves without thermal
gradient, (b) corrosion loop with thermal gradient
generated by an electrical heating inserted in the
cladding tube, and (c) in-reactor, with the effects of
thermal gradient and irradiation.38

0
0

50

100

150

200

Time (days)
Figure 15 The evolution of the microstructure of the oxide
layer on Zircaloy-4 after oxidation in an autoclave with
70 ppm Li, without boron (360  C). The blue line shows the
total oxide thickness, whereas the red line shows the
thickness of the protective inner layer. Reproduced from
Bataillon, C.; Feron, D.; Marchetti, L.; et al. M. E-DEN
Monograph Corrosion Commissariat a` lEnergie
Atomique; 2008.

Loose deposit
Adherent deposit

Oxidized cladding
Figure 16 Schematic representation of the CRUD
morphology.

oxidation of both BWR and PWR cladding. CRUD


can have a wide variety of morphologies, from dense
to porous, thus having very different thermal conductivity. The CRUD structure can generally be
described an inner deposit that is tightly adherent
to the oxide layer and an outer deposit that has a
loose structure (Figure 16). However, the thermal
conductivity of CRUD is generally better than that
of zirconium oxide and therefore its added effect on
deterioration of the heat flux through the corrosion
layer rarely results in excessive cladding temperatures. Because the lithium concentration in the
CRUD, where it is deposited as lithium borate, is

Irradiation Effects

5.03.3.4.1 Radiolysis

The ionizing radiation will interact with water molecules producing a variety of reaction products:
hv
H2 O ! e ; H; OH; H2 O2 ; H2

Figure 18(a) shows the results of a typical computer


simulation of the speciation as a function of time.15
As one can see, numerous oxidizing species such as
O2, O., HO2, and H2O2 that could accelerate the
corrosion are formed. For this reason, the coolant
water in PWRs is hydrogenated. This effect is

60

Corrosion of Zirconium Alloys

shown in Figure 18(b), which indicates that the


presence of hydrogen significantly reduces the
steady-state concentration of the oxidizing species.
5.03.3.4.2 Irradiation effects in the oxide layer

Oxide thickness (mm)

Because the growth of the oxide layer on zirconium


alloys is strongly related to the diffusion of oxygen ions
through the layer, as discussed above, the displacement

5.03.3.4.3 Changes in the metallurgical


state of the metal

PWR, 346 C
40

Corrosion loop, 346 C


Autoclave, 354 C

0
0

300

600

Time (days)

Figure 17 The thickness of the oxide layer on Zircaloy-4


as a function of exposure time for three cases: (a)
autoclaves without thermal gradient, (b) corrosion loop with
thermal gradient generated by an electrical heating inserted
in the cladding tube, and (c) in-reactor, with the effects of
thermal gradient and irradiation. Reproduced from Gilbon,
D.; Bonin, B. E_DEN Monograph Les Combustibles
Nucleaires, Commissariat a` lEnergie Atomique; 2008.

Fast neutron irradiation can change the relative concentration of alloying elements between precipitate
and matrix by a variety of mechanisms including
ballistic mixing by a primarily knock-on effect.41 Fe
and Cr can be dissolved from the intermetallic Zr(Cr,
Fe)2 particles into the surrounding a-Zr matrix.42,43
The dissolution is linked to precipitate amorphization and the modified equilibrium between the
amorphous precipitate and the matrix.44 As a result
of the precipitate dissolution, the smallest particles
dissolve completely, and the largest are significantly
reduced in size. The ultimate location of the Fe
is determined by thermal diffusion effects in the
vicinity of the intermetallic particles where c-type
dislocations may have formed. Postirradiation45
corrosion of such specimens shows progressive

-5

-3

-6

-4

H2

Dissolved H2 0 ppb
Dissolved O2 0 ppb

-8

O2

O-2

-9

OH

-10

HO2

-11
-12
-13

HO-2

H2O2

(a)

-6
-7

O2
Dissolved O2 200 ppb
Dissolved H2 500 ppb

H2O2

-8

O-2

-9

HO2

-10

OH

-11

HO-2

-12

-14
-15
-4

H2

-5
H2O2

Conc. (mol l1) in log scale

-7
Conc. (mol l1) in log scale

of ions from their lattice sites by fast neutron damage


could lead to enhanced point defect concentrations,
enhanced diffusion, and hence enhanced corrosion
in-reactors.39,40 However, experimental studies have
shown no clear evidence for this.
Formation of electronhole pairs and Compton electrons by b and g radiation could also theoretically lead to a significant increase in the
electron conduction (electrical conductivity). The
experimental evidence for this is, however, not
conclusive.

-3

-2

-1

Time (s) in log scale

-13
-4

(b)

-3

-2

-1

Time (s) in log scale

Figure 18 Typical result of a computer simulation or radiolysis of water at 250  C for a dose rate of 4.5  102 Gy h1; (a) pure
water; (b) with dissolved O2 and H2. Reproduced from Waterside Corrosion of Zirconium Alloys in Nuclear Power Plants;
IAEA-TECDOC-996; International Atomic Energy Agency, Vienna, Austria, 1998.

Corrosion of Zirconium Alloys

degradation of the posttransition corrosion rates


with increasing dose, although other studies have
shown nodular corrosion improvement of irradiated
material.46 Cox7 concluded that this effect is now
seen as a primary contributor to enhanced corrosion
in PWRs, since alloys that do not contain Fe or in
which the Fe is incorporated in radiation-resistant
particles, show lower in-reactor corrosion kinetics.

5.03.4 Nodular Oxidation


In BWR conditions (Table 1), nonuniform, so-called
nodular oxidation can also take place (Figure 19).
The mechanism for nodular oxidation is yet to be
fully understood, as it has proved to be challenging to
study in laboratory experiments. Oxidation studies in
500  C steam demonstrated the dependence of nodular corrosion on second phase precipitate size and
distribution,47 typical for various cladding batches
and their metallurgical structure. However, other
factors also affect nodule formation, as batches of
the same cladding can behave quite differently.
Such differences can be related to in-reactor phenomena such as galvanic effects, impurities, radiolytic species, and local power and flux.7,15 The
question of whether nodular corrosion nucleates at
intermetallic particles, between intermetallic particles,
or as a collective property of a group of grains, is yet to
be resolved.6 As discussed by the IAEA Expert group,6
the experimental evidence points toward the fact that
nodules form away from intermetallic precipitates in

61

the alloys. In his review, Cox7 concluded that the


redistribution of Fe from secondary phase particles
diminishes nodular corrosion, but enhances uniform
corrosion.

5.03.5 Hydrogen Embrittlement


Absorption of hydrogen is a major contributor
to degradation of zirconium alloys during service
in nuclear systems.22 This degradation is primarily
attributed to the formation of zirconium hydrides, a
brittle phase that can embrittle cladding, and reduce
its fracture toughness, thus enhancing the susceptibility to cracking.48 Recent studies have also shown
that the addition of hydrogen can increase the creep
rates in Zr2.5Nb49 and possibly irradiation growth.
The hydrogen can come from a variety of sources,
some of them detailed in a 1998 report from an IAEA
Expert group14:
(i)

Hydrogen left over in the Zircaloy tubing after


fabrication or from residual moisture due to
surface preparation (the initial concentrations
of hydrogen in the cladding, postfabrication
but prereactor service, are on the order of
10 wt ppm).
(ii) Desorption of water from incompletely dried
up fuel.
(iii) Hydrogen produced by (n,p) reactions in the
cladding.
(iv) Hydrogen ingress from the coolant water into
the cladding during reactor exposure.

Uniform oxide

Nodular oxide

Zircaloy
20 m
Figure 19 Typical appearance of nodular corrosion in visual inspection and metallographic examination.
Figures courtesy of Ron Adamson.

62

Corrosion of Zirconium Alloys

(v)

Absorption during the normal corrosion processes that occur in high-temperature aqueous
solutions.
(vi) Direct reaction of a clean (no species other
than zirconium) surface with gaseous hydrogen. This hydrogen nominally could come
from three sources: protons released by oxidation that form hydrogen gas, hydrogen produced by radiolysis of the water exposed to
a high-energy neutron flux, and hydrogen specifically added to the cooling water to control
stress corrosion cracking (see Chapter 5.02,
Water Chemistry Control in LWRs and
Chapter 5.08, Irradiation Assisted Stress
Corrosion Cracking).
(vii) Diffusion of hydrogen through a metallic bond
with a dissimilar metal in which hydrogen has a
higher activity.
(viii) Cathodic polarization of zirconium in an electrolyte (typical for low-temperature reactors).
By far, the largest source comes from the normal
corrosion processes (v).
5.03.5.1 Hydrogen Production During
Aqueous Corrosion of Zirconium-Base
Materials
The reaction of Zr with water to form zirconium
oxide produces atomic hydrogen (a proton)
Zr 2H2 O ! ZrO2 2H2

Ia

Zr 2H2 O ! ZrO2 4H

Ib

The proton released in the oxidation of Zr either


combines with another proton to form gaseous hydrogen (eqn I(a)) or diffuses into the zirconium (eqn I(b)),
where it can form zirconium hydrides. The majority of
the protons formed during oxidation combine to form
hydrogen gas but a fraction enters the metal. The term
hydrogen pickup fraction fH is used to relate the
hydrogen absorbed to the hydrogen liberated during
the corrosion reaction.
fH

H absorbed in cladding
H generated in corrosion

Although the total amount of hydrogen absorbed is


proportional to oxide thickness, the proportionality
constant, the hydrogen pickup fraction, changes from
alloy to alloy such that alloy design can significantly
improve cladding performance. The pickup fraction

also changes with corrosion temperature and with


corrosion time, but the mechanisms by which this
occurs are not yet resolved.
5.03.5.2

Hydrogen Absorption

A detailed overview of the process of the absorption of


hydrogen into zirconium-base materials is provided
in the IAEA Technical Document Waterside Corrosion of Zirconium Alloys in Nuclear Power Plants.15
The oxide itself generally presents an effective barrier
to the absorption of hydrogen such that the structure
of the oxide and the electron transport mechanism
can be linked to hydrogen uptake. There is some
evidence that the nickel content in Zircaloy-2
increases the absorption of hydrogen (Figure 20),
either by supporting direct dissociation of water or
by mitigating recombination of hydrogen and oxygen.5052 This was one reason why the nickel was
removed in the formulation of Zircaloy-4. The Zr
(Fe,Ni)2 intermetallic particles that exist in Zircaloy
may provide a significant pathway for hydrogen
uptake. Specifically, the electron current flows primarily at sites where intermetallic particles partially,
or completely, short-circuit the oxide.53 Additionally,
flaws have been found to exist in the oxide that are not
associated with existing intermetallic particles but are
at pits that may have resulted from intermetallic dissolution during pickling. These flaws allow the
cathodic process to proceed. These locations are evidenced by cracks or small holes visible in the oxide.54
For some alloys, the amount of absorbed hydrogen varies as a function of oxide thickness. For
example, Cox55 has shown that for Zircaloy-2, an
initially high hydrogen absorption rate decreases as
the oxide thickness increases, but then picks up
again after the oxide reaches the transition region.
Other researchers have shown an increase in
H uptake just before the oxide transition.56 The additional porosity in the oxide, after transition, makes
hydrogen pickup more likely. For Zircaloy-4, the
pickup appears to be constant with the growth of
the oxide. Oxygen additions to the water normally
reduce the hydrogen uptake and hydrogen additions
increase it.57
5.03.5.3

Hydride Formation

Hydrogen absorbed into the zirconium alloy cladding


at levels more than the terminal solid solubility
can embrittle the cladding through the formation
of hydrides. The stress concentration at the ends of

Corrosion of Zirconium Alloys

. Zircaloy-2, as rolled
. Zircaloy-2, -treated
. Zircaloy-4, as rolled
. Zircaloy-4, -treated

Hydrogen pickup (PPM)

60

63

~100-mil Thick specimens


Zircaloy-2, -treated

50
40

Zircaloy-2 as-rolled
30
Zircaloy-4,
as-rolled or -treated

20
10
0
0

500

1000

1500

2000

2500

3000

Hydrogen overpressure (PSI)


Figure 20 Hydrogen pickup in Zircaloy-2 and Zircaloy-4 as a function of hydrogen overpressure after 14 days in 343  C
water. Reproduced from Hillner, E. Hydrogen absorption in Zircaloy during aqueous corrosion, Effect of Environment,
U.S Rep. WAPD-TM-411, Bettis Atomic Power Lab., W Mifflin, PA, 1964.

larger plate-type hydrides, as well as the localized


deformation in the ligaments between the hydrides,
leads to material weakness. As confirmed by Kerr
et al.,58 performing recent in situ fracture work at the
Advanced Photon Source at Argonne National Laboratory, in materials with large pregrown hydrides
(100 mm), the residual stress field of the zirconium
matrix governs the residual stress state of the hydride
and load is shed to the notch tip hydride phase on
increasing applied load.
The hydrides, if formed, can be circumferential
or radial (see Figure 21). The embrittlement is
influenced by the orientation of hydrides relative to
the stress. Hydrides that are oriented normal to the
tensile load enhance embrittlement by providing
an easy path for the growth of cracks through the
hydrides.59 Radial hydrides are of greater concern,
as they are oriented perpendicular to the hoop stress
that arises during operation of the cladding tube.
As one example, the stress ratio (hoop stress/axial
stress sy/sz) from gas pressurization anticipated
during a loss of coolant accident has an approximate
value of two and most of the deformation is in the
hoop direction.60
The hydrides observed in fuel cladding exposed to
reactor environment are most often fcc delta hydrides
ZrHx (where x  1.6). For a fixed amount of hydrogen
uptake, the density and size of the formed hydrides is
a strong function of the material microstructure. Initial hydride orientation has been shown to be a function of the texture of the Zircaloy-4 that develops
during fabrication.61,62 As an example, as reported by

Singh and coworkers,63 Zr2.5Nb that has been


quenched and aged forms a higher density of smaller
hydrides than Zr2.5Nb that was cold-rolled and
stress relieved, and the difference was attributed to
the underlying grain structure, which acted as the
nucleation host for the formation of hydrides. If sufficient stress is applied during the formation of the
hydrides, the hydride platelets will form perpendicular to the applied stress.
A specific example of deleterious hydride orientation is from DHC, in which circumferentially oriented hydrides dissolve and reprecipitate at the crack
tip, parallel to the crack orientation. Understanding
DHC is of specific concern for CANDU pressure
tubes and ensuring the long-term stability of spent
fuel during storage and is discussed later in this
section.6469
5.03.5.4

Hydride Formation Rates

At reactor operating temperatures, the stable phases


in the ZrH phase diagram are (i) hcp-Zr with dissolved hydrogen and the delta hydride ZrHx, where x
varies between 1.45 and 1.2 at high temperature.
The terminal solid solubility of H in hcp-Zr is
H
CaZr
A expEH =T

5

where A is a constant equal to 1.2  105 wt ppm (or


0.8 mole H per cm3), and the activation energy for
solid solution is 4300 K (the temperature validity for
this equation is up to 865  C, the limit of the alpha
phase region).

64

Corrosion of Zirconium Alloys

250 mm

(a)

100 m

Figure 22 Hydride-rim formation in cladding on


high-burnup (67 GWd/t) PWR fuel. (cladding from the
H. B. Robinson plant, courtesy R. Daum ANL).

hydride precipitation), the hydrogen concentration in


the cladding is essentially homogeneous.
5.03.5.5
(b)

100 m

Figure 21 (a) Circumferential and (b) radial hydrides.


Figures courtesy of Ron Adamson.

Hydrogen is very mobile in a-Zr and once it is


absorbed in the cladding, it migrates easily in
response to concentration, temperature, and stress
gradients. The diffusion coefficient of H in Zr is
H
D0H expEmH =kB T
DZr

EmH
3

6

0:47eV and the preexponential factor


where
is 7  10 cm2 s1. This results in a high diffusion
coefficient at the reactor operating temperatures (at
355  C (average cladding temperature), the diffusion
coefficient is 1.1  106 cm2 s1), so that hydrogen
responds quickly to changed conditions to establish
a new steady state. The characteristic time to attain
significant hydrogen ingress by diffusion through the
thickness of the cladding at this temperature is about
12 min, which is much smaller than normal reactor
exposure times. This means that at any given time,
the hydrogen distribution in the cladding can be
considered to be in quasi-steady state, that is, temporal variations need not be considered. Because of
this, when the hydrogen is in solid solution (before

Formation of Hydride Rim

As, from eqn [5], the hydrogen solubility in Zircaloy


decreases with decreasing temperature, the outer
cladding arrives at the solubility limit before the
inner cladding does. For an outer cladding temperature of 325  C and an inner cladding temperature of
385  C, the hydrogen solubilities are respectively 90
and 170 wt ppm. This causes hydrides to form preferentially at the outer cladding diameter.
Metallographic examinations performed on cladding hydrided below the solubility limit show a more
or less homogeneous hydride distribution through
the thickness of the cladding. These hydrides presumably have precipitated out during the cooling
from operation temperature, so that at reactor temperature, the hydrogen is in solution.
As the overall hydrogen content increases as a
result of increased corrosion, eventually the outer
layer of the cladding reaches saturation and a hydride
rim starts to form, whose thickness will increase
with increasing reactor exposure. Figure 22 shows a
metallograph of high-burnup cladding showing enhanced hydride formation near the outer diameter of
the cladding.
The hydride distribution response to stress and
temperature gradients is at the root of several degradation mechanisms, such as DHC, secondary hydriding, and the degradation of cladding ductility from
oxide spalling.

Corrosion of Zirconium Alloys

5.03.6 Delayed Hydride Cracking


A detailed summary of the DHC phenomena is available in the report of an IAEA (International Atomic
Energy Agency) Coordinated Research Project.70 An
overview is presented here.
The classic theory of DHC comes from the work
of Dutton and Puls.71 The basis of the Dutton and
Puls theory is sketched in Figure 23. The basis of the
theory is that the crack tip hydride grows as hydrogen
migrates from hydrides in the bulk of the material to
the crack tip. The driving force for the diffusion of
the hydrogen is the difference in the chemical potential of hydrogen between the bulk material and the
crack tip hydride in response to hydrostatic stress.70,72
Dutton and Puls theory follows the following
logic. The partial molar volume of hydrogen in the
hydrides is positive. An increasing hydrostatic tensile
stress reduces the chemical potential of hydrogen in
the hydride relative to the bulk. This chemical potential difference causes hydrogen in solution to diffuse
to the crack tip where it precipitates. In this model,
the hydrides in the bulk dissolve to maintain the
hydrogen concentration at the hydride interface at
the solubility limit for the temperature at which the
cracking is occurring. As the precipitate grows at the
crack tip, it will eventually reach a critical size that is

dependent on the stress intensity factor at the crack


tip and the crack will then progress an additional
distance that is related to the hydride size. The relationship between crack velocity and stress intensity
factor for this type of cracking is shown in Figure 24.
The stable crack growth velocity is a function of temperature since that temperature affects both hydrogen
solubility and hydrogen transport to the crack tip.
A detailed update of the Dutton-Puls theory,
along with a response to challenges to the theory
by the group of Kim73 was recently released.74
The review by McRae supports the validity of the
Dutton and Puls models rather than the alternatives
promoted by Kim.
Recently, Colas et al.75 used high-energy synchrotron X-rays at the advanced photon source, to perform in situ characterization of hydride dissolution,
reprecipitation, and reorientation during thermal
cycles under load. Their results, from samples initially charged with hydrogen at concentrations up to
600 ppm, indicate that the reorientation occurred
above a threshold stress of 7580 MPa. In another

Unstable
crack
growth

Hydride
Plastic
enclave

dg

Hydride

r=L

r = li
r=l

Log crack velocity

Free
surface

65

Vc
Stable crack
growth

KIC

No crack
growth

rg

KH

Stress intensity factor (KI)

s
Figure 23 Crack and diffusion geometry assumed in
the DHC model of Puls. Reproduced from Puls, M. P.
J. Nucl. Mater. 2009, 393, 350367.

Figure 24 Schematic diagram of crack propagation by


DHC in hydrided Zircaloy. Reproduced from Delayed
hydride cracking in zirconium alloys in pressure tube
nuclear reactors. IAEA-TECDOC-1410; Nuclear Power
Technology Development Section, International Atomic
Energy Agency, Vienna, Austria, 2004.

66

Corrosion of Zirconium Alloys

recent study related to hydride reorientation, Daum


et al.76 found that the threshold stress is approximately
7580 MPa for both nonirradiated and high-burnup
stress-relieved Zry-4 fuel cladding cooled from
400  C. Within the uncertainty of the experiment,
the irradiation was not critical to setting the threshold
stress. Under ring compression at both room temperature and 150  C, Daum found that radial-hydride
precipitation embrittles Zry-4. Interestingly, for
nonirradiated Zircaloy-4, samples with lower hydrogen concentration (300 vs. 600 ppm) appeared to be
more susceptible to radial-hydride related embrittlement. In a separate work, Daum76 also showed that
failure is sensitive to hydride-rim thickness. Zircaloy-4
cladding tubes with a hydride-rim thickness >100 mm
(700 wt ppm total hydrogen) exhibited brittle behavior, while those with a thickness <90 mm (600 wt
ppm) remained ductile.

Many challenges remain to be addressed in understanding the corrosion and hydriding of Zr alloys in
nuclear environments. Although mechanistic understanding has been developed over the past decades,
still much needs to be understood, in particular, on
the role of alloy design on corrosion rate, hydrogen
pickup, and oxide transition. In addition, the behavior
of cladding under the more challenging conditions of
severe fuel duty associated with longer exposures,
higher temperatures, aggressive chemistry, and incidence of boiling needs to be studied.

References
1.
2.
3.

5.03.7 Summary and Outlook


The extensive research on the corrosion of zirconium alloys has resulted in an enormous flow of
information during several decades. On the basis
of this, a basic understanding of the processes leading to oxidation and hydriding of zirconium alloys
now exists.
1. Zirconium alloy fuel cladding undergoes corrosion when subjected to the reactor environment.
The corrosion is related to the oxidation of the
metal by coolant water and is associated with
hydrogen uptake into the cladding. The latter
phenomenon is the main contributor to in-reactor
degradation of cladding performance.
2. Uniform corrosion is alloy dependent and environment dependent. The higher fuel duty now
imposed on nuclear fuel can lead to accelerated
corrosion, which can limit fuel lifetime.
3. Corrosion is accelerated under irradiation relative
to out-of-pile results. The corrosion rates can
increase after a given exposure.
4. The hydrogen pickup fraction can vary from alloy
to alloy and for different environments and corrosion times. The hydrogen pickup mechanisms and
the influence of the alloy on the process are still
under study.
5. Modern alloys such as M5 and ZIRLO provide
much improved corrosion performance and show
the potential for significant benefits from careful
alloy design.

4.
5.
6.
7.
8.

9.

10.
11.

12.
13.
14.
15.
16.

17.
18.

Kestersson, R.; Yueh, K.; Shah, H.; et al. In 2006


International Meeting LWR Fuel Performance (Top Fuel
2006), Salamanca, Spain, 2006; pp 6771.
Bossis, P.; Pecheur, D.; Hanifi, K.; Thomazet, J.; Blat, M.
J. ASTM Int. 2006, 3, paper IDJAI 12404.
Sabol, G. P. ZIRLO: an alloy development success. In 14th
ASTM International Symposium on Zr in the Nuclear
Industry, STP 1467, Stockholm, 2005; pp 324.
Yang, R.; Ozer, O.; Rosenbaum, H. In Light Water Reactor
Fuel Performance Meeting; ANS: Park City, UT, 2000.
Cox, B. J. Nucl. Mater. 1968, 28, 1.
Cox, B. In Advances in Corrosion Science and
Technology; Fontana, M. G., Staehle, R. W., Eds.; Plenum:
New York, 1976; Vol. 5, p 173.
Cox, B. J. Nucl. Mater. 2005, 336, 205.
Lim, D.; Graham, N. A.; Northwood, D. O. The degradation
of zirconium alloys in nuclear reactors a review;
Canadian Report, Atomic Energy Control Board, Ottawa
(Now Canadian Nuclear Safety Commission) INFO-0174;
Jan 1986.
Lemaignan, C.; Motta, A. T. Zirconium Alloys in Nuclear
Applications, Materials Science and Technology, vol. 10.
Nuclear Materials Pt. 2; Frost, B. R. T., Ed.; VCH
Verlagsgesellschaft mbH, Weinheim, Germany, 1994.
Franklin, D. G.; Lang, P. G. In 9th International Symposium
on Zirconium in the Nuclear Industry, ASTM STP 1132,
1990; pp 332.
Bailly, H.; Menessier, D.; Prunier, C., Eds. The Nuclear
Fuel of Pressurized Water Reactors and Fast Reactors:
Design and Behaviour, Collection du Commissariat a
LEnergie Atomique; English Edition: Intercept, Andover,
Hants, UK, 1999.
Cox, B. J. Nucl. Mater. 1969, 29, 50.
Cox, B.; Pemsler, J. P. J. Nucl. Mater. 1968, 28, 7378.
Corrosion of zirconium alloys in nuclear power plants;
IAEA TECDOC-684; International Atomic Energy Agency,
Vienna, Austria, Jan 1993.
Waterside Corrosion of Zirconium Alloys in Nuclear Power
Plants; IAEA-TECDOC-996; International Atomic Energy
Agency, Vienna, Austria, 1998.
Garde, A. M.; Smith, G. P.; Pirek, R. C. Effects of
hydride precipitate localization and neutron fluence on the
ductility of irradiated Zircaloy-4. In 11th International
Symposium on Zr in the Nuclear Industry, STP 1295,
Garmisch-Partenkirchen, ASTM, 1996; pp 407430.
Pierron, O. N.; Koss, D. A.; Motta, A. T.; Chan, K. S.
J. Nucl. Mater. 2003, 322, 2135.
Raynaud, P. R.; Motta, A.; Koss, D. A.; Chan, K. S.
Fracture toughness of hydrided Zircaloy-4 sheet under

Corrosion of Zirconium Alloys

19.

20.
21.
22.

23.
24.
25.
26.
27.
28.

29.
30.

31.

32.

33.
34.
35.
36.
37.
38.
39.

through-thickness crack growth conditions. In 15th


International Symposium on Zr in the Nuclear Industry,
ASTM STP 1505, Sunriver, OR 2007; pp 163177.
Daum, R. S.; Majumdar, S.; Bates, D. W.; Motta, A. T.;
Koss, D. A.; Billone, M. C. On the embrittlement of
Zircaloy-4 under RIA-relevant conditions. In Thirteenth
International Symposium on Zirconium in the Nuclear
Industry, ASTM STP 1423, 2002; pp 702718.
Motta, A. T.; Gomes-da-Silva, M. J.; Yilmazbayhan, A.;
Comstock, R. J.; Cai, Z.; Lai, B. J. ASTM Int. 2008, 5;
paper ID# JAI10125.
Kass, S. J. Nucl. Mater. 1969, 28, 315321.
Bossis, P.; Lelievre, G.; Barberis, P.; litis, X.; LeFebvre, F.
Multi-scale characterisation of the metaloxide interface
of zirconium alloys. In 12th Internation Symposium on Zr
in the Nuclear Industry, ASTM STP-1354, 2000;
pp 918940.
Hillner, E.; Franklin, D. G.; Smee, J. D. J. Nucl. Mater.
2000, 278, 334.
Bryner, J. S. J. Nucl. Mater. 1979, 82, 84.
Bataillon, C.; Feron, D.; Marchetti, L.; et al. E-DEN
Monograph Corrosion Commissariat a` lEnergie
Atomique; 2008.
Yilmazbayhan, A. PhD. Thesis in Nuclear Engineering,
Penn State University, 2004.
Yilmazbayhan, A.; Motta, A. T.; Comstock, R. J.;
Sabol, G. P.; Lai, B.; Cai, Z. J. Nucl. Mater. 2004, 324,
622.
Pecheur, D.; Lefebvre, F.; Motta, A. T.; Lemaignan, C.;
Charquet, D. Oxidation of intermetallic precipitates in
Zircaloy-4: Impact of irradiation. In 10th International
Symposium on Zirconium in the Nuclear Industry,
Baltimore, MD, ASTM STP 1245; 1994; pp 687705.
Pecheur, D.; Lefebvre, F.; Motta, A. T.; Lemaignan, C.;
Wadier, J. F. J. Nucl. Mater. 1992, 189, 23182332.
Pecheur, D.; Godlewski, J.; Billot, P.; Thomazet, J.
Microstructure of oxide films formed during the
waterside corrosion of the Zircaloy cladding in
lithiated environment. In 11th International
Symposium on Zr in the Nuclear Industry, ASTM STP
1295, Garmisch-Partenkirchen, 1995; pp 94113.
Pecheur, D.; Godlewski, J.; Peybernes, J.; Fayette, L.;
Noe, M.; Frichet, A.; Kerrec, O. Contribution to the
understanding of the effect of water chemistry on the
oxidation kinetics of Zircaloy-4 cladding. In 12th
International Symposium on Zr in the Nuclear Industry;
ASTM STP-1354, Toronto, 1998; pp 793811.
Cox, B.; Ungurelu, M.; Wong, Y. M.; Wu, C. In the 11th
International Symposium on Zr in the Nuclear Industry,
ASTM STP 1295; American Society for Testing of
Materials, 1996, p 114.
Britten, C. F.; Arthurs, J. V.; Wanklyn, J. N. J. Nucl. Mater.
1865, 15, 263.
Roy A. Castelli, Nuclear Corrosion Modelling. The Nature
of CRUD. Elsevier ISBN: 978-1-85617-802-0.
Orlov, A.; Degueldre, C.; Wiese, H.; Ledergerber, G.;
Valizadeh, S. J. Nucl. Mater. 2011. doi:10.1016/
j.jnucmat.2010.12.033.
Janney, D. E.; Porter, D. L. J. Nucl. Mater. 2007, 362,
104115.
Adamson, R.; Garzarolli, F.; Cox, B.; Strasser, A.;
Rudling, P. Corrosion Mechanism in Zirconium Alloys.
ZIRAT12 Special Topic Report; ANT International, 2007.
Gilbon, D.; Bonin, B. E_DEN Monograph Les
Combustible Nucleaires, Commissariat a` lEnergie
Atomique; 2008.
Maguire, M. A. In the 9th International Symposium on
Environmental Degradation of materials in Nuclear Power
Systems, Newport Beach, CA, 1999.

40.
41.

42.
43.
44.
45.

46.

47.

48.

49.
50.

51.
52.

53.
54.
55.

56.
57.

58.
59.

60.

61.

67

Cox, B.; Alcock, A.; Derrick, F. D. J. Electrochem. Soc.


1961, 108, 129.
Motta, A. T.; Lefebvre, F.; Lemaignan, C. In the 9th
International Symposium on Zr in the Nuclear Industry,
ASTM STP 1132; American Society for Testing of
Materials, 1991; p 718.
Griffiths, M.; Gilbert, R. W.; Carpenter, G. J. C. J. Nucl.
Mater. 1987, 150, 53.
Wang, W. J. S.; Tucker, R. P.; Cheng, B.; Adamson, R. B.
J. Nucl. Mater. 1986, 138, 185.
Motta, A. T.; Lemaignan, C. J. Nucl. Mater. 1992, 195,
277285.
Cheng, B. C.; Kruger, R. M.; Adamson, R. B. In the 10th
International Symposium on Zr in the Nuclear Industry,
ASTM STP 1245; American Society for Testing of
Materials, 1994; p 400.
Etoh, Y.; Kikuchi, K.; Yasuda, T.; Koizumi, S.; Oishi, M.
Neutron irradiation effects on the nodular corrosion on
Zircaloy-2. In International Topical Meeting on LWR Fuel
Performance; American Nuclear Society: Avignon, 1991;
pp 691700.
Garzarolli, F.; Stehle, H.; Steinberg, E.; Weidinger, H.
Progress in the Knowledge of Nodular Corrosion,
ASTM-STP-939; American Society for Testing and
Materials: Philadelphia, PA, 1987; p 417.
Sawatzky, A.; Ells, C. E. In Twelfth International
Symposium on Zirconium in the Nuclear Industry, ASTM
STP 1354; Sabol, G. P., Moan, G. D., Eds.; American
Society for Testing and Materials: West Conshohoken, PA,
2000; pp 3248.
Kishore, R. J. Nucl. Mater. 2009, 385, 591594.
Hillner, E. Hydrogen absorption in Zircaloy during
aqueous corrosion, Effect of Environment, U.S Rep.
WAPD-TM-411; Bettis Atomic Power Lab., W Mifflin, PA,
1964.
Kass, S.; Kirk, W. W. ASM Trans. Quart. 1962, 56, 77.
Cox, B. The effect of some alloying additions on the
oxidation of zirconium in steam, U K. Report,
AERE-R4458; United Kingdom Atomic Energy Authority,
AERE, Harwell. Berks, 1963.
Ramasubramanian, N. J. Nucl. Mater. 1975, 55, 134154.
Cox, B.; Wong, Y. M. J. Nucl. Mater. 1999, 270, 134146.
Cox, B. Some factors which affect the rate of oxidation
and hydrogen absorption of Zircaloy-2 in steam, U K. Rep.
AERE-R4348; United Kingdom Atomic Energy Authority,
AERE, Harwell, Berks, 1963.
Harada, M.; Wakamatsu, R. J. ASTM Int. 2008, 5; Paper ID
JAI101117.
Kass, S. The development of the Zircaloys, Corrosion of
Zirconium Alloys, ASTM STP-368; American Society for
Testing and Materials, W. Conshohocken, PA, 1964;
pp 327.
Kerr, M.; Daymond, M. R.; Holt, R. A.; Almer, J. D.;
Stafford, S.; Colas, K. B. Scripta Mater. 2009, 61,
939942.
Davies, P. H.; Sterns, C. P. In Fracture Toughness
Testing of Zircaloy-2 Pressure Tube Material with Radial
Hydrides Using Direct-Current Potential Drop;
Underwood, J. H., Chait, R., Smith, C. W., Wilhem, D. P.,
Andrews, W. A., Newman, J. C., Eds.; ASTM STP 905,
1986, p 379.
Daum, R. S.; Majumdar, S.; Tsai, H.; et al. Mechanical
Property Testing of Irradiated Zircaloy Cladding Under
Reactor Transient Conditions, Small Specimen Test
Techniques, ASTM STP 1418; Sokolov, M. A., Landes, J. D.,
Lucas, G. E., Eds.; American Society for Testing and
Materials: West Conshohocken, PA, 2002; Vol. 4.
Vaibhaw, K.; Rao, S. V. R.; Jha, S. K.; Saibaba, N.;
Jayaraj, R. N. J. Nucl. Mater. 2008, 383, 7177.

68

Corrosion of Zirconium Alloys

62. Kiran Kumar, N. A. P.; Szpunar, J. A.; He, Z. J. Nucl. Mater.


2010, 403, 101107.
63. Singh, R. N.; Roychowdhury, S.; Sinha, V. P.; Sinha, T. K.;
De, P. K.; Banerjee, S. Mater. Sci. Eng. A 2004,
374, 342.
64. Marshall, R.; Louthan, M. Trans. ASM 1963, 56, 693700.
65. Nagase, F.; Fuketa, T. J. Nucl. Sci. Technol. 2004, 41(12),
12111217.
66. Singh, R.; Kishore, R.; Singh, S.; et al. J. Nucl. Mater. 2004,
325, 2633.
67. Choubey, R.; Pulse, M. Met. Mat. Trans. A 1994, 25A,
9931004.
68. Barraclough, K.; Beevers, C. J. Mater. Sci. 1969, 4, 518525.
69. Daum, R. S.; Majumdar, S.; Liu, Y.; Billone, M. C. J. Nucl.
Sci. Technol. 2006, 43(9), 10541067.
70. Delayed hydride cracking in zirconium alloys in pressure
tube nuclear reactors. IAEA-TECDOC-1410; Nuclear

71.
72.
73.
74.
75.
76.

Power Technology Development Section, International


Atomic Energy Agency, Vienna, Austria, 2004.
Dutton, R.; Puls, M. P. Effect of Hydrogen on Behavior of
Materials, TMS-AIME: New York, 1976; pp 512525.
Eadie, R. L.; Coleman, C. E. Scripta Metall. 1989, 23,
18651870.
Kim, Y. S. Mater. Sci. Eng. A 2010, 527, 74807483.
Puls, M. P. J. Nucl. Mater. 2009, 393, 350367.
Colas, K. B.; Motta, A. T.; Almer, J. D.; et al. Acta Mater.
2010, 58, 65756583.
Daum, R. S.; Majumdar, S.; Bates, D. W.; Motta, A. T.;
Koss, D. A.; Billone, M. C. On the embrittlement of
Zircaloy-4 under RIA-relevant conditions. In Zirconium in
the Nuclear Industry: 13th International Symposium. ASTM
STP 1423, 2002; pp 702718.

5.04 Corrosion and Stress Corrosion Cracking


of Ni-Base Alloys
S. Fyfitch
AREVA NP Inc., Lynchburg, VA, USA

2012 Elsevier Ltd. All rights reserved.

5.04.1

Introduction

70

5.04.2
5.04.2.1
5.04.2.2
5.04.2.3
5.04.3
5.04.3.1
5.04.3.2
5.04.3.3
5.04.3.4
5.04.4
5.04.4.1
5.04.4.1.1
5.04.4.1.2
5.04.4.1.3
5.04.4.1.4
5.04.4.2
5.04.4.3
5.04.4.4
5.04.4.4.1
5.04.4.4.2
5.04.4.4.3
5.04.4.4.4
5.04.4.4.5
5.04.4.5
5.04.4.5.1
5.04.4.5.2
5.04.4.5.3
5.04.4.5.4
5.04.4.5.5
5.04.4.6
5.04.4.7
5.04.5
References

Ni-Base Alloy Use in PWRS/BWRS


Wrought NiCrFe Alloys
Age-Hardenable Ni-Base Alloys
Ni-Base Welding Alloys
General Corrosion
Water Chemistry
Flow Rates
Crevices
Mitigation
Stress Corrosion Cracking
Environmental Conditions
Temperature
Water chemistry
Sulfur intrusions
Electrochemical potential
Flow Rates
Crevices
Material Susceptibility Factors
Heat treatment
Microstructure
Grain size
Chemical composition
Product form
Stress
Operating stress
Residual stress
Surface effects
Weld geometry
Stress relief annealing
Irradiation
Mitigation
Outlook

70
70
72
73
73
75
75
75
75
75
77
77
77
80
81
81
82
82
82
83
84
84
85
85
86
86
87
87
87
88
90
90
90

Abbreviations
ASME
ASTM
B&PV
BWR
CEDM
CMTRs

American Society of Mechanical Engineers


American Society for Testing and Materials
Boiler and Pressure Vessel
Boiling water reactors
Control element drive mechanism
Certified material test reports

CRDM
ECP
EPRI
GMAW
GTAW
HAZ
IASCC

Control rod drive mechanism


Electrochemical potential
Electric Power Research Institute
Gas-metal-arc welding
Gas-tungsten-arc welding
Heat-affected zone
Irradiation-assisted stress corrosion
cracking

69

70

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

IGSCC
INCO
LM
LWR
MSE
NRC
PWR
PWSCC
RCS
RUBs
SAW
SCC
SEM
SMAW
TT

Intergranular stress corrosion cracking


International Nickel Company
Light microscopy
Light water reactor
Mechanical surface enhancement
Nuclear Regulatory Commission
Pressurized water reactors
Primary water stress corrosion cracking
Reactor coolant system
Reverse U-bends
Submerged-arc welding
Stress corrosion cracking
Scanning electron microscope
Shielded-metal-arc welding
Thermal treatment

5.04.1 Introduction
Nickelchromiumiron alloys (i.e., nickel-base alloys)
are widely used in the power industry in both fossil
(e.g., coal and gas) power stations and light water
reactor (LWR) nuclear power stations (i.e., pressurized
and boiling water reactors (PWRs and BWRs)). As a
result, the service behavior of these alloys has been
extensively studied,1 especially their susceptibility to
corrosion and stress-induced corrosion phenomena.
The power industry is concerned with the occurrence
of such failure phenomena because of their effect on
the safety and availability of equipment.
Corrosion and, in particular, stress corrosion failures
are not new. The power industry is well acquainted
with stress corrosion cracking (SCC) of stainless steel
in BWR piping and nickel-base alloys in PWR steam

Table 1

generators and its effect on equipment availability.2


SCC of these austenitic alloys has been known for
more than 50 years.

5.04.2 Ni-Base Alloy Use in PWRS/


BWRS
5.04.2.1

Wrought NiCrFe Alloys

The wrought nickel-base alloys that are typically used


for nuclear applications are Alloy 600 and, more
recently, Alloy 690, which contain approximately
twice the chromium content. These materials are
used primarily for their inherent resistance to general
corrosion (i.e., oxidation resistance), strength at elevated temperatures, and a coefficient of thermal expansion very close to carbon and low-alloy steels. The
typical chemical composition and mechanical properties of these alloys are summarized in Tables 1 and 2,
respectively.
Both Alloy 600 and Alloy 690 are non-agehardenable, austenitic solid-solution strengthened
materials. No precipitation reaction is possible with
either alloy to increase strength; however, strength
can be increased by cold-working the material. They
are normally used in the annealed condition; however, a low-temperature heat treatment, or thermal
treatment (TT), is also generally used with these
alloys, which tends to improve the resistance to
SCC in primary water chemistry conditions, which
is typically known as primary water SCC (PWSCC)
(see later sections of this chapter). This improvement
is clearly shown to be more pronounced, at least for
Alloy 600 material, in crack initiation testing.3

Chemical composition of wrought nickel-base alloys used in nuclear applications

Alloying element

Alloy 690

Alloy 600

Alloy X-750

Alloy 718

Alloy 800

Ni Co
C
Mn
Fe
S
Si
Mo
Cu
Cr
Ti
Al
P
Nb Ta
Others

58.0 min.
0.04 max.
0.5 max.
7.011.0
0.015 max.
0.50 max.

0.50 max.
28.031.0

72.0 min.
0.15 max.
1.00 max.
6.0010.00
0.015 max.
0.50 max.

0.50 max.
14.017.0

70.0 min.
0.08 max.
1.00 max.
5.09.0
0.01 max.
0.50 max.

0.50 max.
14.017.0
2.25 2.75
0.401.0

0.701.20

50.055.0
0.08 max.
0.36 max.
Bal.
0.015 max.
0.35 max.
2.83.3
0.30 max.
17.021.0
0.651.15
0.200.80
0.015 max.
4.755.50
B 0.006 max.

3235
0.03 max.
0.41.0
Bal.

0.300.70

<0.75
20.023.0
<0.60
0.150.45

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

These alloys are widely used in LWRs. In BWRs,


applications include such locations as reactor vessel
nozzle safe ends, core support structures, and shroud
bolts. The PWR applications are typically within the
reactor coolant system (RCS) such as steam generator
tubing, penetrations and nozzles, control rod drive
mechanism (CRDM) and control element drive mechanism (CEDM) nozzles in reactor vessel heads, and
instrument nozzles in pressurizers and RCS piping, but
may also be found in selected non-Class 1 components
such as the Core Flood Tanks. In addition, Alloy 600
has also been used in a number of fastener applications.

Table 2

71

Figures 14 show typical applications of Alloy 600


material as used in the reactor coolant systems of the
four major BWR and PWR vendor designs.
Alloy 600 and 690 materials are available as plate,
barstock, tube/pipe, or forged material. The majority
of these materials were procured for the American
Society for Testing and Materials (ASTM)4 or American Society of Mechanical Engineers (ASME) Boiler
and Pressure Vessel (B&PV) Code5 specifications
(e.g., ASTM B 166 and B 167 or ASME SB-166 and
SB-167). Table 3 lists the various industry specifications that are used to procure these materials.

Typical room temperature mechanical properties of wrought nickel-base alloys used in nuclear applications

Mechanical property

Alloy 600

Alloy 690

Alloy X-750a

Alloy 718b

Alloy 800c

Yield strength, min. MPa (ksi)


Ultimate tensile strength, min. MPa (ksi)
Elongation, min. (%)

242 (35)
552 (80)
30

242 (35)
586 (85)
30

655 (95)
1103 (160)
20

1241 (180)
1034 (150)
10

334 (48)
572 (83)
30

Alloy X-750 HTH: Solution annealing at 1093  C (2000  F) and age-hardening at 704718  C (13001324  F).
Alloy 718: Solution annealing at 11001400  C (18322000  F) and age-hardening at 720 and 620  C (1328 and 1148  F).
c
Alloy 800: Solution annealing at 10381066  C (19001950  F) and age-hardening at 760  C (1400  F) for 10 h, furnace cool to 649  C
(1200  F), hold for 20 h.
b

Feedwater recirculation
inlet/outlet welds
RPV attachments/
brackets

Shroud support
structure

Courtesy GE nuclear

CRD in-core housing


instrumentation penetrations

Figure 1 Typical applications of Alloy 600 materials in the reactor coolant systems of a General Electric Design
boiling water reactor. RPV, Reactor Pressure Vessel.

72

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

Instrument and
vent penetrations
(both hot legs)

Decay
heat
line
weld
CRDM nozzles
Pressure
relief nozzle
safe ends/
welds (both
tanks)

PZR vent, spray,


and relief line
welds
PZR steam and
water instrument
penetrations

CRDM
motor
housings
Leak-off
monitor
lines
PZR heater
sleeves
and
diaphragm
plates

Core flood
tank
instrument
penetration
(both tanks)

PZR surge
nozzle weld

HPI/MU
nozzle
welds
(all cold legs)

Hot legsurge
nozzle
weld

Core flood
line welds
(both lines)
Core
guide
lugs
(ID)

Piping-RC pump
suction and
discharge welds
(all pumps)

RV bottom
head
instrument
penetrations

Primary drain
nozzles (both
SGs)

Instrument nozzles and


drain penetrations (all
cold legs)

SG nozzle dam
rings (both SGs)

Figure 2 Typical applications of Alloy 600 materials in the reactor coolant systems of a Babcock & Wilcox Design
pressurized water reactor. RV, Reactor Vessel; RC, Reactor Coolant; SG, Steam Generator; PZR, Pressurizer.

5.04.2.2

Age-Hardenable Ni-Base Alloys

Alloy X-750, a high-strength precipitation-hardening


alloy originally developed for gas turbines and the
aerospace industry, is widely used in internal applications for both BWR and PWR designs, such as fuel
assembly hold-down springs, control rod guide tube
support pins, jet pump beams, and reactor internals
structural bolting. This alloy is very similar in composition to Alloy 600, but contains additions of titanium and aluminum, which combine with nickel to
form the g0 precipitates, Ni3Al and Ni3 (Al, Ti), for
strengthening.6
Alloy 718 is another age-hardenable austenitic
nickel-base alloy, originally developed for the aerospace industry, that has seen much use in the nuclear
industry as a structural material due to its high strength
and corrosion resistance.7 A significant increase in
strength can be achieved by two precipitation reactions from solid solution involving g0 and g00 (Ni3Nb)

secondary phases within the austenitic matrix.8 The


addition of niobium sets this alloy apart from other
high-strength nickel-base alloys (e.g., Alloy X-750)
that are strengthened by g0 alone. Both the g0 and g00
precipitates are quite small and can only be resolved
with an SEM (scanning electron microscope) unless
gross over-aging has taken place. The microstructures
of the solution-annealed and age-hardened conditions
are indistinguishable with light microscopy (LM).
Alloy 718 has also been used extensively in PWR
primary coolant systems, predominantly for fuel assembly hardware.9 Alloy 718 is utilized for fuel assembly
hold-down springs, bolts, and spacer grids. It has been
shown to possess superior SCC initiation resistance
compared to Alloy X-750. Although Alloy 718 has
experienced some isolated failures in PWRs due to
fatigue/fretting cracking, it is considered highly resistant to intergranular SCC (IGSCC) initiation and
other forms of corrosion.

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

73

PZR
instrument
nozzles
PZR and RC pipe-surge
line connections
CEDM motor
housing

Spray nozzlepipe weld


Safety and
relief valve
nozzle-pipe
welds and/or
flanges

CEDM/ICI nozzles to
RPV head welds
RPV top head
vent nozzle

PZR heater
sleeves

RPV head
leak monitor
tubes (2)
Surge
nozzle-pipe
welds
Charging inlet
nozzles (2 cold
legs)

Instrument nozzles
(all hot and cold legs)
Safety injection and
SDC inlet nozzle
(all hot and cold legs)
Let-down and drain
nozzles (all hot and
cold legs)

Shutdown
cooling
inlet
nozzle
(all cold
legs)

Spray
nozzles
(2 cold legs)

Primary nozzle closure


rings and welds
(both SGs)
Bottom channel head
drain tube and welds
(both SGs)

Guide lugs
flow skirt
ICI nozzles-ICI guide
tubes (system 80 plants)

RCP suction
and discharge
(all cold legs)

Shutdown cooling outlet


nozzle (1 hot leg)

Figure 3 Typical applications of Alloy 600 materials in the reactor coolant systems of a Combustion Engineering
Design pressurized water reactor. RPV, Reactor Pressure Vessel; RC, Reactor Coolant; RCP, Reactor Coolant Pump;
SDC, Shutdown Cooling; PZR, Pressurizer.

In addition, a modified Alloy 800 material, with


carbide-forming elements added to limit the solid
solution carbon content, has been successfully used for
many years in Germany for steam generator tubing.
The typical chemical compositions, mechanical
properties, and industry procurement specifications
are provided in Tables 13.
5.04.2.3

Ni-Base Welding Alloys

Welding of nickelchromiumiron alloys is typically


performed using arc-welding processes such as gastungsten-arc welding (GTAW), shielded-metal-arc
welding (SMAW), and gas-metal-arc welding
(GMAW).7 Submerged-arc welding (SAW) may also
be used provided the welding flux is carefully selected.
Alloy 82, 182, and 132 are typical filler metals used to
join Alloy 600 components to carbon or low-alloy steel
vessels and other component items. These weld alloys
are also used as cladding in selected components
within the reactor coolant system. In addition, Alloy
52 and 152 (and newly developed variants such as
Alloy 52M and 152M) are filler metals that have

recently become the preferred materials used to join


Alloy 690 component items to carbon or low-alloy
steel vessels in the reactor coolant system.
Occasionally, there is a need for welded Alloy
X-750 items and the filler metals Alloy 82 or 69
(ERNiCrFe-8) were used during original fabrication.
The Alloy 69 material is no longer produced and
filler metal Alloy 718 is the currently recommended
material for welding.
The typical chemical compositions and industry
procurement specifications of these alloys are summarized in Tables 3 and 4.

5.04.3 General Corrosion


One of the main reasons that nickel-base alloys were
chosen for LWR applications is that they have the
ability to withstand a wide variety of severe operating
conditions involving corrosive environments, high
temperatures, high stresses, and combinations of these
factors. General corrosion can be defined as uniform
deterioration of a metal surface by chemical or

74

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

Spray nozzlepipe weld

Safety and relief


nozzle-pipe
welds

Head
vent pipe
CRDM
motor
housings

CRDM
nozzles

RPV
head
leak
monitor
tube

Surge nozzlepipe welds

Thermowells (all
hot and cold legs)

RV nozzle
pipe weld
(all hot and
cold legs)

SG nozzlepipe weld
(all hot and
cold legs)

Core support
block
Bottom-mounted
instrument nozzles

Bottom channel
head drain tube
and welds (all SGs)

Primary nozzle
closure rings and
welds (all SGs)

Figure 4 Typical applications of Alloy 600 materials in the reactor coolant systems of a Westinghouse Design pressurized
water reactor. RPV, Reactor Pressure Vessel; RV, Reactor Vessel; SG, Steam Generator.

Table 3

Typical nickel-base alloy specifications used in nuclear applications

ASME B&PV Code

ASTM Standard

Material

Product form

SB-163
SB-166
SB-167
SB-168
SB-637
SB-670
SFA 5.11
SFA 5.14

B 163
B 166
B 167
B 168
B 637
B 670

Alloys 600/690/800
Alloys 600/690
Alloys 600/690
Alloys 600/690
Alloys 718/X-750
Alloy 718
Alloys 182/152
Alloys 82/52

Seamless tubing
Rod and bar
Seamless pipe and tube
Plate, sheet, and strip
Rod, bar, and forgings
Plate, sheet, and strip
Covered welding electrodes
Bare welding rods and electrodes

electrochemical reaction with the environment. Nickel


has good resistance to corrosion in the normal atmosphere, in freshwaters, and in deaerated nonoxidizing
acids, and it has excellent resistance to corrosion by
caustic alkalies. The high nickel content of these alloys
gives them resistance to corrosion by many organic and
inorganic compounds and also makes them virtually

immune to chloride-ion SCC. Chromium additions


provide resistance to sulfur compounds and also provide resistance to oxidizing conditions at high temperatures or in corrosive solutions. Details of the
corrosion resistance in these types of environments
can be found elsewhere (i.e., see also Chapter 2.08,
Nickel Alloys: Properties and Characteristics).2,10

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

Table 4

75

Chemical composition of nickel-base welding alloys used in nuclear applications

Alloying
element

Alloy 52a
filler metal

Alloy 69
filler metal

Alloy 72
filler metal

Alloy 82
filler metal

Alloy 132
electrode

Alloy 152a
electrode

Alloy 182
electrode

Ni Co
C
Mn
Fe
S
Si
Mo
Cu
Cr
Ti
Al
P
Nb Ta
Al Ti
Zr
B
Others

Bal.
0.04 max.
1.0 max.
7.011.0
0.015 max.
0.50 max.
0.50 max.
0.30 max.
28.031.5
1.0 max.
1.10 max.
0.030 max.
0.10 max.
1.5 max.

0.50 max.

70.0 min.
0.08 max.
1.0 max.
5.09.0
0.015 max.
0.50 max.

0.50 max.
14.017.0
2.002.75
0.401.00
0.030 max.
0.701.20

55.0 max.
0.05
0.1 max.
0.2 max.
0.008 max.
0.1 max.

0.20 max.
44.0 max.
0.6 max.

67.0 min.
0.10 max.
2.53.5
3.0 max.
0.015 max.
0.50 max.

0.50 max.
18.022.0
0.75 max.

0.030 max.
2.03.0

0.50 max.

62.0 min
0.08 max
3.5 max
11.0 max
0.02 max
0.75 max

Bal.
0.05 max.
5.0 max.
7.012.0
0.015 max.
0.75 max.
0.50 max.
0.50 max.
28.031.5
0.50 max.
0.50 max.
0.030 max.
1.02.5

0.50 max.

59.0 min.
0.10 max.
5.09.5
10.0 max.
0.015 max.
1.0 max.

0.50 max.
13.017.0
1.0 max.

0.030 max.
1.02.5

0.50 max.

0.50 max
13.017.0

0.030 max.
1.54.0

0.50 max

Alloys 52M and 152M have controlled additions of boron and zirconium.

5.04.3.1

Water Chemistry

The water chemistry of LWRs is discussed in detail


in Chapter 5.02, Water Chemistry Control in LWRs.
Nickel-base alloys are essentially immune to general
corrosion in LWR environments due to the formation
of an adherent Cr-rich oxide on the surface.
5.04.3.2

Flow Rates

The inherent passivity of nickel-base alloys provides


them with excellent resistance to flow-assisted corrosion. They are able to withstand very high flow rates,
on the order of 18.3 m s1 (60 ft s1), without concern.
Corrosion rates in such flowing conditions for nickelbase materials are expected to be <2.5 mm year1
(<0.1 mil year1).11
5.04.3.3

Crevices

General corrosion of nickel-base alloys in crevices


is not anticipated to be of great concern in LWRs
because of the passive nature of these materials. Pitting
may occur occasionally in the presence of impurities,
which could lead to SCC (see Section 5.04.4.3),
particularly in the more oxidizing conditions of BWRs.
No failures in PWRs have been directly attributed to
creviced locations.
5.04.3.4

Mitigation

As general corrosion is minimal in LWR environments, mitigation is not really necessary. As noted

in this section on LWR Structural Materials, PWR


environments have reducing conditions and general
corrosion is not of concern. However, mitigation of
corrosion concerns in the more oxidizing environment of BWRs has been through the use of hydrogen
water chemistry (i.e., to make the environment more
reducing, similar to a PWR) and noble metal chemical
additions.12,13

5.04.4 Stress Corrosion Cracking


SCC of nickel-base alloys is an important age-related
phenomenon affecting LWRs. This type of failure
mechanism for nickel-base alloys typically occurs
intergranularly and is generally termed intergranular
stress corrosion cracking (IGSCC). In PWRs, IGSCC
is typically termed primary water stress corrosion
cracking (PWSCC). The occurrence of SCC of nickelbase alloys has been extensively studied since the first
reported observation of cracking in laboratory tests
using Alloy 600 in high-purity water by Coriou et al.14
in 1959. Over the last three decades, IGSCC has
been observed numerous times in LWRs and it has
affected both the safe and economic operation of
the reactors. In BWRs, cracking of nickel-base
components such as safe ends, shroud bolts, and access
hole covers has occurred; however, the predominant
failures have been identified in Alloy 182 welds. In
PWRs, PWSCC of Alloy 600 component items has
been observed in steam generators, pressurizers, and

76

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

CRDM nozzles, and most recently in Alloy 182 and


Alloy 82 welds. The mechanism of this cracking
phenomenon is not completely understood, and prediction of crack initiation time has proven to be
extremely difficult, if not impossible, due to the uncertainty of numerous variables (e.g., heat treatment and
residual stress). In this section, emphasis will be given to
the SCC of Alloy 600 materials in PWRs, given the fact
that it has been the most prevalent; however, as noted
above, BWR conditions are not immune to IGSCC of
nickel-base alloys.
It is known, however, that SCC of nickel-base materials occurs as a result of the following three factors:
 susceptibility of the material
 a tensile stress (including both operating and residual stress)
 a corrosive environment
The synergistic effect of these three factors is
typically shown on a Venn-type of diagram (Figure 5).
As an example, the susceptibility of Alloy 600
material to PWSCC depends on several factors,
including the chemical composition, heat treatment
during manufacture of the material, heat treatment during fabrication of the component, and
operating parameters of the component. Chemical
composition and heat treatment are interrelated in
several ways. For example, one reason for annealing
Alloy 600 is to solutionize the carbon in the alloy.
As the material cools, chromium carbides precipitate from the solution at both intragranular and
intergranular locations. If the cooldown from the

Mechanical
Operational tensile stresses
residual tensile stress

SCC
Corrosive
Susceptible material
Chemical composition
microstructure

Electrochemical
corrosion potential
temperature
pH-value

Figure 5 Synergistic factors affecting stress corrosion


cracking of nickel-base materials.

anneal is sufficiently slow, a greater number of carbides


will precipitate at the grain boundaries (i.e., intergranularly) and the resistance to PWSCC will be
improved. Well-decorated grain boundaries are an
indication that an Alloy 600 material has received
proper heat treatment and that sufficient carbon
was available in the solution to combine with chromium. If adequate amounts of carbon and chromium
exist, but the anneal is not at a high enough temperature or sufficient time is not allowed to solutionize the
carbon, an adequate amount of carbon will not be
available to precipitate intergranularly as chromium
carbides, leading to minimal grain boundary decoration. Most precipitation occurs during cooldown
following annealing; however, stress relief treatments
can lead to additional precipitation. The primary goal
of stress relief, however, is to allow a local realignment
of highly strained regions to reduce internal stresses.
Carbon and chromium concentration gradients are
also reduced given the extended time at the temperature. Thus, if the anneal has not adequately solutionized carbon for chromium carbide precipitation at
the grain boundaries, stress relief treatment will not
reduce susceptibility to PWSCC.
Tensile stresses, resulting from both residual and
operating stresses, can be significant for some Alloy
600 component items. Operating stresses are produced from mechanical and thermal loading, while
residual stresses are generated as a result of fabrication, installation, and welding processes. Residual
stresses are more difficult to quantify than operating
stresses and, in many instances, are of a higher magnitude than operating stresses.
PWSCC is a thermally activated degradation
mechanism, that is, as the temperature increases, the
rate of PWSCC increases exponentially. Thus, the
hot leg temperature of the RCS creates a more
aggressive environment in which the Alloy 600 components must operate.
The cracking observed in PWRs to date is typically axially oriented (although circumferentially oriented cracks have been observed) and occurs in
an area, such as a weld heat-affected zone (HAZ),
that has high residual tensile stresses. In a cylindrically
shaped component (e.g., piping, vessel, and nozzles),
the circumferential stresses are inherently higher than
axial stresses. Thus, in a homogeneous material with no
initial flaws, cracking would be expected to occur
axially because of the higher circumferential stresses.
PWSCC has been the subject of much research
and analyses in recent years as a result of the many
failures that have been attributed to it. However,

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

(5070  F) temperature differences between hot and


cold legs are enough to significantly influence the
time to initiation and subsequent crack growth rate.
Temperature is generally believed to affect the
rate of SCC attack in accordance with an activation
model for thermally controlled processes (Arrhenius
equation), exp(Q/RT), where Q is the activation
energy, R is the ideal gas constant, and T is the
absolute temperature. The current consensus is that
the activation energy for crack initiation falls in the
range of 188230 kJ mol1 (4555 kcal mol1) and
many predictions are based on 210 kJ mol1 (50 kcal
mol1). There is also evidence that the activation
energy varies with material carbon content.15

a reliable crack initiation model has yet to be developed. PWSCC of Alloy 600 components in the RCS
can lead to through-wall cracking and thus leakage of
primary water. (Catastrophic failure is not expected as
circumferentially-oriented cracks do not occur
unless very high axial stresses are generated in the
component, e.g., from roll expansion methods.)
5.04.4.1

Environmental Conditions

The major environmental conditions affecting SCC


of nickel-base material in LWR environments appear
to be temperature, water chemistry (oxygen, hydrogen, lithium, boron, and sulfur content), and electrochemical potential (ECP). Each of these factors is
evaluated as follows.

5.04.4.1.2 Water chemistry

The water chemistry of LWRs can generally be


described as essentially pure water. PWRs primarily
include hydrogen, boron, and lithium to produce reducing conditions. BWRs primarily operate with low levels
of oxygen, but in recent times, hydrogen additions have
been introduced to limit the oxidizing potential of
the environment. Additional details are included in
Chapter 5.02, Water Chemistry Control in LWRs.

5.04.4.1.1 Temperature

By far, temperature is the single most significant


environmental factor influencing the initiation of
SCC in LWR environments. This is evidenced by
the fact that the vast majority of SCC of PWR steam
generator roll expansion transitions have occurred
on the hot leg side of the tube sheet. The 2839  C

MA + drawn
35 % area reduction

4000

SCC initiation time (h)

77

3000

2000

1000

Hydrogen

200 ppm B
0.7 ppm Li
hydrogen

500 ppm B
1.0 ppm Li
hydrogen

1100 ppm B
2.0 ppm Li
hydrogen

Water chemistry
Figure 6 Stress corrosion cracking initiation times for Alloy 600 steam generator tubing at 360  C (680  F). Stress
corrosion cracking initiation time as a function of primary water chemistry for as-drawn (35% area reduction) mill-annealed
tubing. Reproduced from Airey, G. P. The stress corrosion cracking performance of Inconel Alloy 600 in pure and primary
water environments. In Proceedings: 1983 EPRI Workshop on Primary-Side Stress Corrosion Cracking of PWR Steam
Generator Tubing; EPRI NP-5498, Project S3035, with permission from Electric Power Research Institute.

78

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

5.04.4.1.2.1 Hydrogen

The effect of dissolved hydrogen on SCC susceptibility of nickel-base alloys (e.g., Alloys 600 and
X-750) has been evaluated by numerous researchers.
Pathania and McIlree16 reviewed the influence of
hydrogen on PWSCC in 1987. At that time, the
emphasis of the work was on initiation at temperatures of 360  C (680  F) and above. The authors concluded that the susceptibility of Alloy 600 increased
when the amount of dissolved hydrogen increased.
Airey17 has shown that dissolved hydrogen increases
the rate of PWSCC of steam generator tubing in
autoclave tests at 360  C (680  F). An example of his

Temperature (C)
10-5

290

325

345

365

Crack growth rate (mm s-1)

4 pts

10-6
4 pts
Ave. of
4 pts

data, shown in Figure 6, shows that the SCC initiation time for pure water is decreased dramatically
when hydrogen is added. However, in the primary
water of PWRs, the effect of hydrogen appears to be a
function of the boron and lithium content. Bandy and
Van Rooyen18 have shown a similar effect with boron
alone versus pure water and primary water (Figure 7).
They showed that 83% of the specimens cracked in
pure water with hydrogen versus only 2% in pure
water without hydrogen.
More recently, Norring has reported on tests to
determine the effect of hydrogen overpressure in
330  C (626  F) water.19 Results of these tests, shown
in Figure 8(a), suggest that the rate of PWSCC
increases with increasing hydrogen overpressure.
The most recent update on the influence of
hydrogen on PWSCC was prepared by Cassagne
et al.20 in 1997. All the data seem to indicate that the
susceptibility of Alloy 600 decreases drastically for
low hydrogen values (<10 kPa (1.45 psi)) regardless of
temperature. For hydrogen partial pressure above
100 kPa (14.5 psi), a more progressive decrease in
susceptibility seems to occur between 360 and
400  C (680 and 752  F). Data are not available in
this range for lower temperatures. Between 10 and
100 kPa (1.45 and 14.5 psi), it appears that PWSCC
initiation is not greatly affected for all temperatures
between 400 and 310  C (752 and 590  F). At 290  C
(554  F), the influence of hydrogen cannot be
assessed because of a lack of data.
5.04.4.1.2.2

10-7
Flattened specs, I.E.
Cold worked, pure H2O
As received, pure H2O + H2
As received, pure H2O + H3BO3
As received, pure H2O
As received, (0.03 % C),
pure H2O
Primary H2O
10-8

1.75

1.70

1.65
1
T(K)

1.60

1.55

] 1000

Figure 7 Effect of hydrogen on Alloy 600 cracking in pure


and primary water environments. Reproduced from Bandy,
R.; Van Rooyan, D. Quantitative examination of stress
corrosion cracking of Alloy 600 in high temperature
water Work in 1983. In Proceedings: 1983 EPRI Workshop
on Primary-Side Stress Corrosion Cracking of PWR Steam
Generator Tubing; EPRI NP-5498, Project S3035, with
permission from Electric Power Research Institute.

Boron and lithium

Evaluations of boron and lithium on PWSCC of Alloy


600 have been performed by numerous investigators.
Norring et al.19 concluded that increasing the lithium
content from 2.4 to 3.5 ppm significantly decreased
the time to crack initiation (see Figure 8(b)).
The most complete evaluation was performed by
Ogawa et al.21. In these tests, hydrogen overpressure was
kept at a constant level of 30 cm3 kg1 H2O. The results
of this work are shown in Figures 9 and 10. It appears
that maintaining a constant pH of 7.17.3 (at 285  C
(545  F)) will produce a range of crack initiation times
(Figure 9) and that increasing the boron content
greater than 1200 ppm (at any lithium level) decreases
the crack initiation times (Figure 10). Therefore,
the beginning of cycle boron concentrations appears
to be the worst condition for PWSCC initiation. Maintaining a pH level of 7.3 (at 285  C (545  F)) also
appears to be better than a pH level of 7.1.
Follow-up tests were performed at high boron
concentrations with varying lithium concentrations

Temperature...........

329.9 C

329.6 C

Boron......................

1471 ppm

1441 ppm

Lithium....................

2.42 ppm

2.38 ppm

Hydrogen content...

13.1 ml kg-1

25.2 ml kg-1

7.0 kPa

13.6 kPa

7.354

7.351

Activity...

99
90

50
25
10
5
1
100 200

10
5

(a)

99
90

B&W 1700 F 7/8 (6A)

25
10
5

500 1000 2000 5000 10 000


Exposure time (h)

The influence of hydrogen on the tendency to PWSCC

Temperature...........

328.9 C

329.6 C

Boron......................

1241 ppm

1441 ppm

Lithium....................

3.54 ppm

2.38 ppm

Hydrogen content...

24.7 ml kg-1

25.2 ml kg-1

13.6 kPa

13.6 kPa

7.575

7.351

Activity...
pH...........................

B&W 1700 F 7/8 (6)

90
50
25
10
5
1
100 200

500 1000 2000 5000 10 000


Exposure time (h)

B&W 1700 F 3/4

90
50
25
10
5
1
100 200

99
Cracked specimens (%)

Cracked specimens (%)

99

Cracked specimens (%)

99

(b)

500 1000 2000 5000 10 000


Exposure time (h)

50

1
100 200

500 1000 2000 5000 10 000


Exposure time (h)

B&W 1700 F 3/4

25

B&W 1700 F 7/8 (6)

99
90

79

50

1
100 200

Cracked specimens (%)

Cracked specimens (%)

pH...........................

Cracked specimens (%)

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

500 1000 2000 5000 10 000


Exposure time (h)
B&W 1700 F 7/8 (6A)

90
50
25
10
5
1
100 200

500 1000 2000 5000 10 000


Exposure time (h)

The influence of lithium on the tendency to PWSCC

Figure 8 Effect of hydrogen and lithium on time to stress corrosion cracking of Alloy 600 steam generator tubing.
Reproduced from Norring, K.; Rosborg, B., Engstrom, J., Svenson, J. Influence of LiOH and H2 on primary side IGSCC of
Alloy 600 steam generator tubes, Colloque International Fontevraud II, Sept 1014, 1990; Societe Francaise dEnergie
Nucleaire, Paris, France, 1990; pp 243249, with permission from Societe Francaise dEnergie Nucleaire.

80

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

Li (ppm)

4
1.7 1.6
1.5

1.4

1.3

1.2
1.1

1.0

0
6.3

6.4

6.5

6.6

6.7

6.8

6.9 7.0
pH 285 C

7.1

7.2

7.3

7.4

7.5

7.6

Figure 9 Isosusceptibility diagram for primary water stress corrosion cracking of Alloy 600 as a function of lithium and pH at
285  C (545  F). The numbers (1.0, 1.1, 1.2, . . .) represent the ratio of the percent intergranular fracture referenced to the
response at B:280/Li:2.0 ppm. Reproduced from Ogawa, N.; et al. Nucl. Eng. Des. 1996, 165, 171180.

Li (ppm)

1.7

1.6

1.5

1.4

1.3

1.2

1.1

1.0

0
2000

1500

1000
B (ppm)

500

Figure 10 Isosusceptibility diagram for primary water stress corrosion cracking of Alloy 600 as a function of lithium and
boron. The numbers (1.0, 1.1, 1.2, . . .) represent the ratio of the percent intergranular fracture referenced to the response at
B:280/Li:2.0 ppm. Reproduced from Ogawa, N.; et al. Nucl. Eng. Des. 1996, 165, 171180.

to simulate beginning of fuel cycle conditions. Ogawa


et al.22 report that there is little effect of lithium content from 2 to 10 ppm at boron concentrations greater
than 1200 ppm (see Figure 11) and PWSCC susceptibility at 1600 ppm boron (210 ppm lithium) was
higher than that at 500 or 280 ppm boron concentrations (with 2 ppm lithium).
5.04.4.1.3 Sulfur intrusions

Sulfur intrusions by themselves will not produce


SCC in nickel-base material; however, sulfate will
promote intergranular attack and intergranular SCC.
A sensitized material microstructure is much more

susceptible (in terms of initiation time) to this type of


attack, although all nickel-base materials will be
attacked by sulfate.
Andresen23 has shown that the time to failure
decreased by 23 orders of magnitude in constant
load tests between pure water and sulfate impurities
(conductivities ranging from 5 to 55 mS cm1 due to
sulfuric acid additions).
Bandy et al.24 have shown that sulfates are very
potent cracking agents for Alloy 600 materials. Temperature significantly accelerates the cracking and
most likely decreases the threshold stress for cracking
to occur.

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

81

11
(1.53)
10
1.5

(1.46)

Li (ppm)

5
1.4
4
1.3
1.2
1.1

1.0
1.1

1.2

0
2000

1500

1000
B (ppm)

500

Figure 11 Revised isosusceptibility diagram for primary water stress corrosion cracking of Alloy 600 as a function of lithium
and boron. The numbers (1.0, 1.1, 1.2, . . .) represent the ratio of the % intergranular fracture referenced to the response at
B:280/Li:2.0 ppm. Values in parentheses are extrapolated or interpolated from test result values. Reproduced from
Ogawa, N.; Nakashiba, T.; Yamada, M..; Umehara, R.; Okamoto, S.; Tsuruta, T. In Proceedings of the Eighth International
Symposium on Environmental Degradation of Materials in Nuclear Power Systems; American Nuclear Society: La Grange
Park, IL, 1997; pp 395401. Copyright 1997 by the American Nuclear Society, La Grange, IL, USA.

5.04.4.1.4 Electrochemical potential

100
PH2 = 0.005 MPa

IGSCC (%)

80

PH2 = 0.1 MPa

60
40
20

Alloy 600
350 C
0.01 m H3BO3 + 0.001 m LiOH soln

0
-1000

-300
0
(E )
Potential (mV) Corr

300

Figure 12 Effect of electrochemical potential on rate


of primary water stress corrosion cracking in
Alloy 600 material. Reproduced from Smialowka,
S. Hydrogen induced IGSCC of Alloy 600 in high
temperature aqueous environments. In Proceedings:
1987 EPRI Workshop on Mechanisms of Primary Water
Intergranular Stress Corrosion Cracking; EPRI
NP-5987SP, with permission from Electric Power
Research Institute.

SCC is also significantly affected by the ECP.


As shown in Figure 12 by the work of Smialowska,25
small changes in the ECP can have a large effect on SCC.
Test results have demonstrated that a crack growth
rate maximum, with respect to coolant hydrogen
variation, is observed in proximity to a key phase transition, the nickel (Ni) to nickel oxide (NiO) phase transition, as shown bya Pourbaixdiagram (see Figures 13 and
14).26,27 The SCC hydrogen dependency is fundamentally described by the extent that the alloys corrosion
potential deviates from the potential of the Ni/NiO
phase transition. This potential difference represents
the relative stability of the SCC controlling oxide films
(e.g., crack tip oxides are often of a NiO structure).
5.04.4.2

Flow Rates

Flow rates in LWRs do not appear to have any effect


on SCC susceptibility and no testing data are known
to be available.

82

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

Potentials typical of
deaerated environments

Oxide film and IGSCC


arrows indicate an increasing
sensitivity to cracking

RHE potential

Potential

Ni/Nio equilibrium

Potentials typical of environments


with dissolved hydrogen (10/30 bars)

No oxide film
no IGSCC

Neutral environments

Caustic environments

Stable Cr oxides

Cr very soluble
9/10
pH at 300/320 C

Figure 13 Effect of pH and potential on the surface films and sensitivity to stress corrosion cracking of Alloy 600 material.
Reproduced from Scott, P. M.; Le Calvar, M. In Proceedings of the Sixth International Symposium on Environmental
Degradation of Materials in Nuclear Power Systems; The Minerals, Metals, and Materials Society: Warrendale, PA, 1993;
pp 657667, with permission from The Minerals, Metals, and Materials Society.

5.04.4.3

Crevices

Nickel-base materials have been found to be very


susceptible to IGSCC in crevice areas of BWR applications, especially at welds where both weld-induced
sensitization and high residual stresses are present.
Stress corrosion crack initiation and growth of nickelbase materials in BWRs have been attributed to the
development of an acidic environment within crevices
because of the oxygen in normal BWR water chemistry, combined with high residual and applied stresses
resulting from the geometry and nearby welds. The
first widely reported occurrence of SCC of Alloy 600
material in BWRs was at the Duane Arnold unit in
1978.28 This occurred in the crevice area of a recirculation inlet nozzle safe end and was initiated in the
creviced area between the thermal sleeve and the safe
end. However, no known failures of nickel-base materials have been directly attributed to environmental
conditions within crevices in PWRs.

5.04.4.4

Material Susceptibility Factors

The PWSCC susceptibility of a particular heat of


Alloy 600 material is dependent upon a variety of
factors. The most important factors, based on results
from a number of investigators, appear to be material
microstructural features, chemical composition, and
manufacturing process (or product form). Each of
these has been evaluated below.
5.04.4.4.1 Heat treatment

Considerable research efforts have been made to


identify the mechanism responsible for PWSCC of
Alloy 600. These investigations, reviewed by Was,29
have shown that PWSCC is primarily dependent
on the heat treatment received by the material. For
instance, a heat treatment of mill-annealed (MA)
Alloy 600 in the temperature range 650750  C
(12001380  F) has been found to cause a drastic
improvement in the resistance to PWSCC. This has

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

83

1.50
O2

1.00

H2O

0.50

NiO2
1 ppb

V (SHE)

0.00

Acid
SO4
cracking

H2
1 atm
8.2 ppm

Ni3O4

Ni++
Ni
-0.50

NiO

Fe++
Fe

PWR
Secondary side
Primary side

Fe3O4
Fe

Caustic
cracking

PWSCC

-1.00

Ni(OH)3
Caustic
IGA

-1.50

-2.00

10

12

14

pH

Figure 14 Main domains of intergranular attack and stress corrosion cracking of Alloy 600 in aqueous solutions at
300 C (572 F) relative to the Pourbaix diagram for nickel. (Note the stability boundary for iron relative to its lowest
oxidation states is also shown to indicate that iron can be oxidized under all likely PWR primary and secondary conditions
whereas elemental nickel can be stable under certain PWR primary conditions depending on the corrosion potential
fixed by the hydrogen partial pressure.) Reproduced from Scott, P. M.; Combrade, P. In Proceedings of the Eighth
International Symposium on Environmental Degradation of Materials in Nuclear Power Systems; American Nuclear Society:
La Grange Park, IL,1997; pp 6573. Copyright 1997 by the American Nuclear Society, La Grange Park, IL, USA.

been termed thermal treatment (TT). Although the


exact mechanism of enhanced resistance to PWSCC
is unclear and still under debate, there is general
agreement that the chemical composition and structure of grain boundaries are of crucial importance.
In this connection, chromium depletion, segregation
of impurities to grain boundaries, intergranular carbides and their mechanical effect on stress concentrations, and grain boundary misorientation appear to
be essential to the corrosion behavior of the material.
Annealing temperature and time are critical to the
precipitation reactions that occur. The penultimate
and final anneals appear to be the most critical
for PWSCC resistance. Stiller et al.,30 Briant et al.,31
Hall and Briant,32 EPRI NP-507233 all agree that a

high-temperature final anneal, sufficient to solutionize all carbon precipitates, followed by a slow
enough cooling rate to precipitate copious carbides
on the grain boundaries (intergranular carbides) will
produce a reasonably PWSCC-resistant material.
Norring34 tested Alloy 600 reverse U-bends (RUBs)
in high-purity water containing hydrogen. The time
to crack initiation at 320  C (608  F) increased by a
factor of 8 when the annealing temperature was
increased from 925  C (1697  F) to 1025  C (1877  F).
5.04.4.4.2 Microstructure

The precipitation obtained in the final material


microstructure is dependent upon heat treatment
and chemical composition (mainly carbon, chromium,

84

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

titanium, and nitrogen content). During the cooling


that occurs after annealing, carbon combines with
chromium to form chromium carbides. In addition,
titanium carbides and titanium nitrides are formed.
These precipitate within the grains (intragranularly)
or at grain boundaries (intergranularly) depending
largely upon the temperature reached during annealing
and thermal treatments, the presence of prior precipitates (e.g., undissolved carbides), the time at temperature, the carbon content, and the cool down rate.
A study was conducted by the Electric Power
Research Institute (EPRI) in 1981 to evaluate carbide
dissolution and precipitation kinetics of Alloy 600.35
This study showed that final annealing temperatures
between 982 and 1010  C (1800 and 1850  F) (for
carbon contents between 0.02% and 0.048%) provide
a consistent thermal treatment response and an adequate precipitation of chromium carbides.
The most widely accepted hypothesis for the beneficial effect of intergranular carbides on PWSCC
resistance has been proposed by Bruemmer,36 who
suggests that grain boundary carbides promote crack
blunting due to their effectiveness as dislocation
sources. Another possible explanation proposed by
Smialowska25 is that Alloy 600 material passivates more
readily in the presence of grain boundary carbides.
Most recently, Fish et al.37 have proposed that intergranular carbides may improve passivity at the crack
tip by reducing the amount of carbon segregated along
the grain boundaries.
Whatever the mechanism, or mechanisms, involved,
there is general agreement that a microstructure with
copious intergranular carbides and few intragranular
carbides correlates with good resistance to PWSCC.
A range of intergranular and intragranular carbide
precipitation has been used by many investigators to
quantify PWSCC susceptibility.15 Scott et al.38 have
developed a range of material susceptibility indices,
based on minimum times to failure, for Alloy 600
material used in French steam generators. These factors
have also been used for CRDM nozzles.39,40
5.04.4.4.3 Grain size

Grain size can be related to yield strength and tensile


strength of Alloy 600 material by the general type of
HallPetch relationship.41 That is, a lower yield
strength material will typically have a larger grain
size. Data for PWSCC42 show that very small grain
sizes (typically ASTM grain size numbers >9) are
more prone to PWSCC than larger grain sizes (typically ASTM 48). Other data43 show that steam
generator tubing with a grain size number larger

than ASTM 6 is less susceptible to PWSCC. However, the effect of grain size is most likely a secondorder effect, while carbide precipitation tends to be
more important.
5.04.4.4.4 Chemical composition

The chemical composition of Alloy 600 has mainly


been correlated to PWSCC susceptibility in terms of
the carbon content. Many investigators have shown
that carbon contents near the high end of the typical
mill range (0.020.06 wt%) result in increased
PWSCC susceptibility. Electricite de France (EdF)
conducted a series of RUB specimen tests on steam
generator tubing, using a low mill anneal temperature
with carbon content ranging from 0.01 to 0.07 wt%.44
The tests were conducted in elevated temperature
pure water and primary water with hydrogen overpressure. These tests show that low carbon (<0.018%)
had the lowest PWSCC susceptibility. Pichon and
others45 have shown that carbon contents that are
relatively high (>0.063%) or very low (<0.012%)
are more susceptible to PWSCC.
The carbon content of Alloy 600 weld materials
(Alloys 182 and 82) has also been investigated by
several researchers.46,47 It appears that low carbon
contents (<0.02%) are most susceptible, medium
carbon contents (0.020.04%) are intermediate, and
high carbon contents (>0.04%) have the most resistance to PWSCC.
However, correlating the actual carbon content of
the material may not be the best approach. This is
also shown in the test data reported by Norring et al.34
A better approach may be to determine the available
carbon content. This would entail knowledge of the
titanium and nitrogen content of the heat of material.
If it is assumed that all the nitrogen is precipitated as
titanium nitrides (TiN), then the remaining titanium
can be assumed to react with the carbon in the
material. With these assumptions (which are thermodynamically reasonable), the remaining carbon would
be the available carbon amount to precipitate as
chromium carbides (M7C3 or M23C6 precipitates).
Unfortunately, titanium and nitrogen are not normally
reported in the certified material test reports (CMTRs);
thus, this approach cannot be used without archive
material or sampling of the heats in service.
Buisine et al.48 evaluated the PWSCC resistance of
nickel-based weld metals with various chromium
contents. The tests clearly demonstrated that weld
metals with 30% chromium were resistant to
PWSCC. The threshold for PWSCC resistance
appears to be between 22% and 30% chromium.

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

There have been discussions in the literature of


the role of chromium depletion along the grain boundaries (sensitization), as well as boron and phosphorous segregation along the grain boundaries.49 The
degree of sensitization is tied to the heat treatments
and there have been conflicting reports on the benefits
and disadvantages of sensitization. It has been shown
that a sensitized material is more susceptible to caustic
attack. In addition, boron and phosphorous may be
beneficial for retarding PWSCC, but the experimental
evidence is weak at present.
5.04.4.4.5 Product form

Alloy 600 component items in the RCS are fabricated


by press forging, hammer forging, hot rolling, and
forming and machining from bar stock in accordance
with ASME SB-166 or cold drawn and hot finished
tubing or piping in accordance with ASME SB-167.
In addition, it has been suggested that some ASME
SB-167 material was actually produced from SB-166
bar stock and recertified as SB-167 material. The
fabrication processes affect the material microstructure, yield strength, and hardness, which in turn
affect the PWSCC resistance of the material.
The PWSCC susceptibility of various product
forms has been evaluated to some degree. The most
susceptible product form appears to be cold worked
and low temperature MA steam generator tubing, as
evidenced by the vast majority of data available in the
literature. However, the other product forms (i.e.,
forgings, piping, and bar stock) have not been assessed
as extensively. One study, performed by Webb50
found differences in susceptibilities of cold-worked
and hot-worked materials with similar microstructures. The cold worked and annealed tubing was
more susceptible than hot worked and annealed
forging material. The different behavior remains
even when yield strength differences between the
materials are taken into account.
Another study on the relative susceptibility of
high and low yield strength bar and tubing, performed for EPRI,51 concluded that PWSCC susceptibility ranked in decreasing order as (1) high-yield
strength bar, (2) high-yield strength tubing, and
(3) low-yield strength bar. However, no low-yield
strength tubing was tested in this study.
Microstructural evaluation in the study described
above indicated that grain boundary carbide decoration is generally poor in bar products and better in
tubing products. However, the susceptibility of Alloy
600 material also depends on surface cold work due
to machining, grinding, and reaming. A material with

85

highly susceptible microstructure when subjected


to a large amount of cold work on the surface (i.e.,
reaming or grinding) becomes very susceptible to
PWSCC. Therefore, component items that were
machined from bar stock and with weld root grinding
should be considered highly susceptible to PWSCC.
A machined surface without reaming or grinding is
considered to have undergone moderate cold work
and somewhat less susceptible to PWSCC. The least
susceptible material is for cold drawn tubing with a
high mill-anneal temperature. This type of approach
has been adopted by Consumers Power Company
at the Palisades Nuclear Plant.52
The most recent work was performed by Briceno
et al.,53 who tested a variety of product forms.
Two groups of materials appear to exist, as shown in
Figure 15. The first group contains cold-worked
material, thick wall tube, and steam generator tubing;
whereas, the second group contains hot-worked thick
wall tubes. The forged bars showed higher crack initiation times than the tubes tested. Specimens made of
forged bars and hot-worked tube with similar grain
boundary carbide distributions (60%) and the same
grain size (ASTM 4.04.5) showed a significant difference in initiation times. The same tendency is seen
for a low density of intergranular carbides when coldworked tube and forged bar with the same grain size
were compared.
5.04.4.5

Stress

Laboratory tests and operating experience in PWRs


suggest that significant PWSCC should not occur
in Alloy 600 component items at stresses less than
about 242 MPa (35 ksi) for temperatures up to around
324  C (615  F). As operating design stresses permitted by the ASME B&PV Code are much less than
242 MPa (35 ksi), PWSCC failure would not be
expected to occur due to applied pressure and thermal loadings. This is supported by the fact that
essentially all PWSCC failures have occurred at
locations where (1) high residual stresses are produced during fabrication (e.g., pressurizer nozzle
J-groove welds), (2) high stresses are produced as
a result of strains induced during operation (e.g.,
steam generator-dented tube support plate intersections), or (3) high stresses are produced by geometric abnormalities (e.g., excessive steam generator
tube ovality).
The following paragraphs address the operating
and residual stresses that may act in RCS pressure
boundary component items.

86

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

Specimens cracked/specimens tested

1.0

0.8

0.6
FC no cracks
FB
FA
FX
V1V2
W1
W2 no cracks
GW
GV

0.4

0.2

0.0
0

Bar

Thick
wall
tube

S.G.
tube

1000

2000

3000
4000
Time (h)

5000

6000

7000

Heat

Working
method

Final thermal
treatment

YS (MPa)
after
forging

YS (MPa)
after
thermal
treatment

%C

FC

Forged bar

745 C/2 h (a.c)

428

413

0.021

FB

Forged bar

800 C/2 h (a.c)

491

489

0.021

FA

Forged bar

800 C/2 h (a.c)

550

412

0.024

FX

Hot worked

(*)

291

0.03

V1

Cold rolled 1000 C/3 min (w.c)

301.5

0.051

V2

Cold rolled 1000 C/3 min (w.c)

274

0.052

W1

Hot worked

980 C

320.6

0.081

W2

Hot worked

980 C

244.7

0.067

GVW

Cold worked 927 C/35 min (a.c)

393

0.04

GV

Cold worked 927 C/35 min (a.c)

389

0.038

( * ) no heat treatment after extrusion of the tube.


% C = Percentage carbon content
Figure 15 Stress corrosion cracking initiation time at 330  C (626  F) for up to 10 500 h. Reproduced from Briceno, D.;
Blazquez, F; Hernandez, F. In Proceedings of the Eighth International Symposium on Environmental Degradation of Materials
in Nuclear Power Systems; American Nuclear Society: La Grange Park, IL,1997; pp 249256. Copyright 1997 by the American
Nuclear Society, La Grange Park, IL, USA.

5.04.4.5.1 Operating stress

5.04.4.5.2 Residual stress

Operating stresses, on the order of 69138 MPa


(1020 ksi) in the hoop direction of J-groove, partial
penetration welds, as well as full penetration welds,
are anticipated in LWR component items. Stress
levels on this order are not high enough to cause
concern with SCC. Therefore, residual stresses must
be taken into account.

The magnitude and sign of residual stresses in most


typical RCS component items are considered to be a
function of (1) surface layer hardness produced by cold
working (e.g., rolling, machining, bending, or reaming)
and (2) residual stresses produced by welding.
Test data by Bandy and Van Rooyen (Figure 16)
indicate that the initiation time for PWSCC in

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

87

Fraction of yield strength (RT)

10

AR

CW

345 C (AR)

1.0

0.1
1

100

10

1000

Failure (days)
Figure 16 Correlation between stress and time to stress corrosion cracking for steam generator tubing tested in pure water
at 300  C (tfailure K  Stressb, b 4). CW: cold work; AR: as-received. Reproduced from Bandy, R.; Van Rooyan D.
Quantitative examination of stress corrosion cracking of Alloy 600 in high temperature water Work in 1983. In Proceedings:
1983 EPRI Workshop on Primary-Side Stress Corrosion Cracking of PWR Steam Generator Tubing; EPRI NP-5498, Project
S3035, with permission from Electric Power Research Institute.

elevated temperature water varies inversely with the


fourth power of the applied stress.18 Test data by
Yonezawa et al.54 (Figure 17(a) and 17(b)) suggest
that the initiation time varies inversely with the sixth
or seventh power of the applied stress. In addition,
Figure 18 shows data from Yonezawa that suggest
that the absolute value of the stress raised to a
power is not the only significant factor. Other significant factors include the amount of cold work and the
ratio of the applied stress to the material yield
strength. In other words, the PWSCC initiation
time at a given applied stress level should increase
as the material strength is increased. Conversely, for
strain-controlled situations, such as at a J-groove
weld, where the residual stress is a fixed percentage
of the material yield strength, increasing the yield
strength can actually decrease the initiation time.
5.04.4.5.3 Surface effects

The residual stress in a component item is a function of the applied mechanical forces and the heat
input. Mechanical forces applied during cold work
tend to produce compressive residual stresses in
the surface layer. Conversely, heat input has the
potential to produce tensile residual stresses. Specifically, heat input causes expansion of the surface
layer yielding, if the thermal stresses exceed material yield strength and residual tensile stresses in
the surface layer upon cooling. This suggests that
high heat input operations such as drilling, turning,

or grinding have the potential to develop higher


tensile residual stresses than low heat input operations such as reaming. Note that these high tensile
residual stresses produced during machining are
limited to a few mils depth.
Berge et al.55 have shown, in crack initiation testing
performed in France, that the time to crack initiation with a base material tensile stress of 500 MPa
(72.5 ksi) is four times that of a material with a
cold-worked surface stress of 1000 MPa (145 ksi).
5.04.4.5.4 Weld geometry

The welding of small penetration nozzles into larger


vessel component items, by either partial penetration
or full penetration welds, can produce tensile stresses
in the nozzle. The magnitude of these stresses is
generally on the order of the yield strength of the
material. The cooling and subsequent shrinking of
the weld material pull the nozzle wall toward the
much stiffer component item to which it is welded.
This shrinkage results in residual tensile hoop stresses in the nozzle near the weld. This weld shrinkage
also results in axial bending of the nozzle. The magnitude of the bending stress, however, is typically
much lower than that of the hoop stress.
5.04.4.5.5 Stress relief annealing

The primary goal of stress relief heat treatment is to


allow a local realignment of highly strained regions
to reduce internal stresses. As mentioned above,

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

Stress measured by X-ray diffraction analysis on the


top of U-bend specimens (kgf mm-2)

88

100

90
s-7

80

70

60

Crack

50
0.2% Offset proof stress at room temperature: 3.3 kgf mm-2
40

102

103

Applied stress (kgf mm-2)

(a)

104

Stress corrosion cracking testing time (h)

Crack
No crack
s-6
70
60
50
40
30
20

0.2% offset proof stress at 360 C

10
0
(b)

102

103

104

Stress corrosion cracking testing time (h)

Figure 17 Correlation between stress and time to stress corrosion cracking for Alloy 600 steam generator tubing.
(a) Effect of stress on the stress corrosion cracking resistance of mill-annealed Alloy 600 at 360  C in high temperature
water. (b) Effect of stress on the stress corrosion cracking resistance of mill-annealed Alloy 600 at 360  C in high
temperature water, using constant load stress corrosion cracking test. Reproduced from Yonezawa, T.; Onimura, K.; Saito, I.;
Takamatsu, H. In Materials for Nuclear Reactor Core Applications; BNES: London, 1987; pp 7783, with permission from
British Nuclear Energy Society.

improvement to the intergranular carbide precipitation is also almost always a benefit.


The degree of stress relief is known to be a function of time and temperature. Available data from the
International Nickel Company (INCO) are given in
Figure 19.56 It can be seen that at a temperature of
482  C (900  F) for 4 h, 21% of the stress is relieved.
Also, at a temperature of 870  C (1600  F) for 4 h,
88% of the stress is relieved. It should be noted that
essentially all the stress relief occurs within 1 h at
the temperature of interest. Residual stresses are

generally reduced from values near the room temperature yield strength to values that are near zero stress.
5.04.4.6

Irradiation

Irradiation-assisted stress corrosion cracking


(IASCC) is an age-related degradation mechanism
where materials exposed to neutron radiation become
more susceptible to SCC with increasing fluence.57
IASCC, like PWSCC, is a distinctive subset of SCC.
Despite numerous investigations and research efforts,

89

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

Applied
stress
(kgf mm-2)

Material
Heat
treat
ment

Stress corrosion cracking


testing time (h)

Applied
stress
-2
Frac- (kgf mm )
ture
0.80.2

Cold
working

1.0sO2

24

8%
tensile
strained

45

15%
tensile
strained

57

(2.793)
(2.713)

IGSCC

975 C 25%
0.07 annea- tensile
ling strained

72

(159)
(3.231)

IGSCC

C (%)

20%
cold
rolled

87

2.000 4.000 6.000 8.000 10.000 12.000 14.000

(1.717)

Fracture

2.000 4.000 6.000 8.000 10.000 12.000 14.000

9.505
(3.704)
(5.772)

Stress corrosion cracking


testing time (h)

IGSCC
IGSCC

IGSCC

IGSCC
IGSCC

45

(12.8481)

IGSCC

(2.374)
(4.191)

57

IGSCC

No crack
Crack

Figure 18 Effect of applied stress to yield strength ratio on time to stress corrosion cracking for Alloy 600 steam
generator tubing. Effect of cold working on the stress corrosion cracking resistance for the same ratio of applied stress to
yield strength of mill-annealed Alloy 600 at 360  C in high temperature water. Reproduced from Yonezawa, T.; Onimura, K.;
Saito, I.; Takamatsu, H. In Materials for Nuclear Reactor Core Applications; BNES: London, 1987; pp 7783, with permission
from British Nuclear Energy Society.

30
482 C (900 F)
593 C (1100 F)
25
538 C (1000 F)
Residual stress (1000 psi)

649 C (1200 F)
20

15
704 C (1300 F)
10
760 C (1400 F)
5
871 C (1600 F)

2
Time (h)

Figure 19 Effect of heating time and temperature on residual stress of cold-drawn, annealed Alloy 600 rod.
Reproduced from Inconel Alloy 600, International Nickel Company: Huntington, WV, 1978, with permission from Special
Metal Corporation.

details of the IASCC mechanism remain hypothetical.5861 The current consensus is that IASCC results
from a synergistic effect of irradiation damage to the
material, water environment with possible radiolysis

effects, and a stress state. At present, interactions


between these variables have not adequately been
quantified and no primary IASCC controlling mechanism has been identified.

90

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

Nickel-base alloys, such as Alloy 600 and 690,


are not typically utilized in areas of high fluence
because of their high nickel content. Nickel will
become highly radioactive when exposed to neutron
radiation. No known IASCC-related studies have
been reported in the open literature for these alloys.
There are, however, some data available for Alloy
X-750 at low neutron fluence levels. Results obtained
from swelling capsule tests indicate that IASCC
resistance of Alloy X-750 is approximately the same
as Types 304 and 316 stainless steel.62,63 Some tensile
test data of small scale Alloy X-750 HTH Condition
bolts irradiated in a PWR to 1.4  1019 n cm2
(E > 1.0 MeV) performed without any failure up to
0.95% strain.64 Simulated BWR testing demonstrated
an irradiation-enhanced IGSCC susceptibility of
Alloy X-750 HTH Condition as a function of fluence
and boron content.65 Fluence at 1019 n cm2 versus
10141018 n cm2 (E > 1.0 MeV) showed an increase
in susceptibility.
Alloy 718 has an excellent record in PWR primary
water applications often at operating stresses close to or
exceeding the yield point that can be 1000
1100 MPa (145160 ksi). Alloy 718 is used in fuel applications where very high neutron fluxes are also experienced. The few failures that have occurred have been
attributed to a manufacturing defect that allowed components to enter service with preexisting intergranular
defects. Alloy 718 is known to be highly resistant to
crack initiation, but IGSCC will propagate rapidly in
PWR primary water from preexisting defects.
5.04.4.7

Mitigation

A number of techniques have been evaluated and are


available to delay or eliminate the occurrence of SCC
in LWRs. Many of the available and future mitigation
techniques were evaluated in a paper presented at a
conference held by the US Nuclear Regulatory Commission (NRC) in 2003.66 In general, these techniques
fall into three main categories:
1. Mechanical surface enhancement (MSE)
2. Environmental barriers or coatings
3. Chemical or electrochemical corrosion potential
(ECP) control
MSE techniques represent processes that reduce surface tensile residual stresses or induce compressive
surface stresses on a component item or weld. Examples of MSE techniques include shot peening and
electropolishing. Environmental barrier or coating
techniques represent processes that protect the

material surface from an aggressive environment.


Coating examples include nickel plating and weld
deposit overlays or inlays. Chemical or ECP control
techniques represent changes to the environment
that alter the corrosion process or produce corrosion
potentials outside the critical range for SCC. Examples of chemical or ECP control include zinc additions
to the reactor coolant and modified water chemistry
(e.g., hydrogen water chemistry and noble metal chemical additions to BWR coolant; variations in dissolved
hydrogen levels, lithium concentrations, and boron
concentrations for PWR coolant).

5.04.5 Outlook
Wrought nickel-base alloys and their weld metals
were originally used in LWRs due to the materials
inherent resistance to general corrosion in a number
of aggressive environments and because of a coefficient of thermal expansion that is very close to that of
low alloy and carbon steel. Over the last 40 years,
SCC has been observed in numerous component
items and associated welds, sometimes after relatively
long incubation times. The occurrence of SCC has
been responsible for significant downtime and
replacement power costs.
Nickel-base materials will continue to be used both
for repair and replacement activities in currently
operating nuclear units and in the next generation of
LWRs. Although the continued use of Alloy 600 material, for example, has essentially been eliminated in
LWRs, acceptance and use of the higher chromium
material, Alloy 690, is widely recognized by the industry. These higher chromium materials have been
shown to be highly resistant to SCC in laboratory
experiments and component repairs. Their use has
proven to be an effective decision, since replacements
have been free from cracking in operating reactors over
periods up to about 20 years. Improvements in the
weldability of the higher chromium weld metals (i.e.,
variants of Alloys 52 and 152) have been identified
through significant research and the results have been
put into application today.

References
1.
2.

Shah, V. N.; Macdonald, P. E. Eds. Aging and Life


Extension of Major Light Water Reactor Components;
Elsevier: Amsterdam, 1993.
Corrosion in the nuclear power industry. Metals
Handbook, Vol. 13, Corrosion, 9th ed.; ASM International:
Metals Park, OH, 1987; p 927.

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys


3.

4.
5.
6.
7.

8.

9.
10.
11.
12.

13.

14.

15.
16.

17.

18.

19.

20.

Knorring, K.; Stiller, K.; Nilsson, J. In Proceedings of 5th


International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems Water Reactors,
Monterey, CA, 1991; American Nuclear Society:
La Grange, IL, 1991; pp 482487.
American Society for Testing and Materials Annual
Book of ASTM Standards; ASTM: West Conshohocken,
PA, 2009; Vol. 02.04.
ASME Boiler and Pressure Vessel Code, Section II, Part B,
Nonferrous Materials; ASME: New York, NY, 2009.
INCONEL Alloy X-750 Product Handbook, 5th ed.; Inco
Alloys International: Huntington, WV, 1988.
Miglin, M. T.; Domian, H. A.; Baty, D. L. Microstructure and
Stress Corrosion Resistance of Alloy X-750, 718, and
A-286 in LWR Environments; Research Project 2181-1,
NP-6392-M; Electric Power Research Institute: Palo Alto,
CA, June 1989.
Keiser, D. D.; Brown, H. L. A review of the physical
metallurgy of Alloy 718; ANCR-1292; Aerojet Nuclear
Company, Idaho Falls, ID, Feb 1976; UC-25, Materials,
TID-4500, R63.
Mager, T. R.; Wilson, I. L. W. Improved stress corrosion
resistance of NiCrFe alloys; EPRI Research Project
2181-2, Interim Report, NP-6908-M; July 1990.
Resistance to Corrosion, Product Brochure 5M 5-85 S-37,
4th ed.; Inco Alloys International: Huntington, WV, 1985.
Copson, H. R.; Berry, W. E. Corrosion 1960, 16(2), 123129.
Kass, J. N.; Cowan, R. L. In Proceedings of Second
International Symposium on Environmental Degradation
of Materials in Nuclear Power Systems Water Reactors;
American Nuclear Society: La Grange Park, IL, 1985;
pp 211218.
Andresen, P. L. In Proceedings of Seventh International
Symposium on Environmental Degradation of Materials in
Nuclear Power Systems Water Reactors; National
Association of Corrosion Engineers: Houston, TX, 1995;
pp 563577.
Coriou, H.; Grall, L.; Pelras, M.; Vettier, S. In Third
Metallurgical Symposium on Corrosion (Aqueous and
Gaseous), 1959; North Holland: Amsterdam, 1960;
pp 161169.
Report of Joint PWR Owners Group Alloy 600 Meeting;
prepared by Dominion Engineering Inc., available from
EPRI, Palo Alto, CA, Apr 2728, 1995; Attachment #2.
Pathania, R. S.; McIlree, A. R. In Proceedings of the Third
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems; National Association
of Corrosion Engineers: Houston, TX, 1987.
Airey, G. P. The stress corrosion cracking performance of
Inconel Alloy 600 in pure and primary water environments.
In Proceedings: 1983 EPRI Workshop on Primary-Side
Stress Corrosion Cracking of PWR Steam Generator
Tubing; EPRI NP-5498, Project S3035.
Bandy, R.; Van Rooyan D. Quantitative examination of
stress corrosion cracking of Alloy 600 in high
temperature water Work in 1983. In Proceedings: 1983
EPRI Workshop on Primary-Side Stress Corrosion
Cracking of PWR Steam Generator Tubing; EPRI NP-5498,
Project S3035.
Norring, K.; Rosborg, B.; Engstrom, J.; Svenson, J.
Influence of LiOH and H2 on primary side IGSCC of Alloy
600 steam generator tubes, Colloque International
Fontevraud II, Sept 1014, 1990; Societe Francaise
dEnergie Nucleaire: Paris, France, 1990; pp 243249.
Cassagne, Th; Fluery, B.; Vaillant, F.; de Bouvier, O.;
Combrade, P. In Proceedings of the Eighth International
Symposium on Environmental Degradation of Materials in
Nuclear Power Systems; American Nuclear Society:
La Grange Park, IL, 1997; pp 307321.

21.
22.

23.

24.
25.

26.

27.

28.
29.
30.
31.
32.
33.
34.
35.
36.

37.

38.

39.

40.
41.
42.
43.

91

Ogawa, N.; et al. Nucl. Eng. Des. 1996, 165, 171180.


Ogawa, N.; Nakashiba, T.; Yamada, M.; Umehara, R.;
Okamoto, S.; Tsuruta, T. In Proceedings of the Eighth
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems; American Nuclear
Society: La Grange Park, IL, 1997; pp 395401.
Andresen, P. L. The effects of sulfate impurities in 288  C
water on IGSCC of Inconel 600 in constant load and SSRT
experiments. In Corrosion 84, Apr 26, 1984; available
from National Association of Corrosion Engineers:
Houston, TX; Paper Number 177.
Bandy, R.; Roberge, R.; Newman, R. C. Corrosion 1983,
39(10), 391398.
Smialowka, S. Hydrogen induced IGSCC of Alloy 600 in
high temperature aqueous environments. Proceedings:
1987 EPRI Workshop on Mechanisms of Primary Water
Intergranular Stress Corrosion Cracking; EPRI NP5987SP.
Scott, P. M.; Le Calvar, M. In Proceedings of the Sixth
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems; The Minerals, Metals,
and Materials Society: Warrendale, PA, 1993; pp 657667.
Scott, P. M.; Combrade, P. In Proceedings of the Eighth
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems; American Nuclear
Society: La Grange Park, IL, 1997; pp 6573.
Pipe Crack Study Group. Investigation and evaluation of
stress-corrosion cracking in piping of light water reactor
plants; NUREG-0531; NRC, Feb 1979.
Was, G. S. Corrosion 1990, 46(4), 319330.
Stiller, K.; et al. Metall. Mater. Trans. 1996, 27A, 327341.
Briant, C. L.; OToole, C. S.; Hall, E. L. Corrosion 1986,
42(1), 1527.
Hall, E. l.; Briant, C. L. Metall. Trans. 1985, 16A,
12251236.
Specially prepared Alloy 600 tubing; EPRI NP-5072,
Project S303-17; Feb 1987.
Carbide dissolution and precipitation kinetics of Inconel
600; EPRI NP-2093, Project 1708-1, Oct 1981.
Bruemmer, S.; et al. Corrosion 1988, 44(11), 782788.
Fish, J. S.; Lewis, N.; Yang, W. J. S.; Perry, D. J.;
Thompson, C. D. In Proceedings of the Eighth
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems; American Nuclear
Society: La Grange Park, IL, 1997; pp 266273.
Scott, P.; Meyzaud, Y.; Benhamou, C. In Proceedings of
International Symposium on Plant Aging and Life
Predictions of Corrodible Structures, Sapporo, Japan,
May 1518, 1995; pp 18.
Boudot, R.; Vidal, P.; Gelpi, A. A method for the evaluation
of the nickel based alloys susceptibility to PWSCC.
In Proceedings: 1992 EPRI Workshop on PWSCC of Alloy
600 in PWRs, EPRI TR-103345, Paper F6, Dec 1993.
Cattant, F. Metallurgical investigations performed on
CRDM nozzles removed from power plants.
In Proceedings: 1992 EPRI Workshop on PWSCC of Alloy
600 in PWRs, Dec 1993; EPRI TR-103345, Paper B5.
Park, H. B.; Kim, Y. H.; Lee, B. W.; Rheem, K. S. J. Nucl.
Mater. 1996, 231(3), 204212.
Proceedings: 1985 Workshop on Primary-side SCC of
PWR S/G Tubing; EPRI NP-5158, Project S3035, EPRI.
Shah, V. N.; Lowenstein, D. B.; Turner, A. P. L.; et al. Nucl.
Eng. Des. 1992, 134, 199215.
Majo, D. G.; Gelpi, A.; Dallery, D.; Rouillon, Y.; Van
Duysen, J. C.; Zacharie, G. Prediction of the in-service
behavior of Alloy 600 tubes used in steam generators of
pressurized water reactors. In Colloque International
Fontevraud II Sept 1014, 1990; Societe Francaise
dEnergie Nucleaire: Paris, France, 1990; pp 281290.

92

Corrosion and Stress Corrosion Cracking of Ni-Base Alloys

44. Pichon, C.; Boudot, R.; Benhamou, C.; Gelpi, A. Residual


life assessment of French PWR vessel head penetrations
through metallurgical analysis. In American Society of
Mechanical Engineers, Pressure Vessels and Piping
Division, 1994 Pressure Vessels and Piping Conference;
Service Experience and Reliability Improvement: Nuclear,
Fossil, and Petrochemical Plants, 1994; Vol. 288,
pp 4147.
45. Seman, D. J.; Webb, G. L.; Parrington, R. J. Primary water
stress corrosion cracking of Alloy 600 Effects of
processing parameters. In Proceedings: 1991 EPRI
Workshop on PWSCC of Alloy 600 in PWRs, EPRI
TR-100852, Paper E3, July 1992.
46. Mullen, J. V.; Parrington, R. J. Stress corrosion of Alloy
600 weld metal in primary water. In Proceedings: 1992
EPRI Workshop on PWSCC of Alloy 600 in PWRs, Dec
1993;EPRI TR-103345, Paper F3.
47. Buisine, D.; Vaillant, F.; Vidal, P.; Gimond, C. PWSCC
resistance of nickel based weld metals with various
chromium contents. In Proceedings: 1994 EPRI Workshop
on PWSCC of Alloy 600 in PWRs, EPRI TR-105406, Paper
D5, Aug 1995.
48. Airey, G. P. Metallography 1980, 13, 2141.
49. Webb, G. L.; Burke, M. G. In Proceedings of the Seventh
International Symposium on Environmental Degradation
of Materials in Nuclear Power Systems; National
Association of Corrosion Engineers: Houston, TX, 1995.
50. Sarver, J. M.; Pathania, R.; Stuckey, K.; Fyfitch, S.;
Gelpi, A.; Foucault, M. PWSCC of Alloy 600 penetrations
(EPRI RP 3223). Presentation E1 in Proceedings: 1994
EPRI Workshop on PWSCC of Alloy 600 in PWRs, EPRI
TR-105406, Project 3223-1, Aug 1995.
51. Hoang, P. H. Primary water stress corrosion cracking
inspection ranking scheme for Alloy 600 components.
In Proceedings of the Sixth symposium: Current Issues
Related to Nuclear Power Plant Structure, Equipment and
Piping, Dec 46, 1996; available from N. Carolina State
University Center for Nuclear Power Plant Structures,
Equipment, and Piping: Raleigh, NC.
52. Briceno, D.; Blazquez, F.; Hernandez, F. In Proceedings of
the Eighth International Symposium on Environmental
Degradation of Materials in Nuclear Power Systems;
American Nuclear Society: La Grange Park, IL, 1997;
pp 249256.
53. Yonezawa, T.; Onimura, K.; Saito, I.; Takamatsu, H.
Materials for Nuclear Reactor Core Applications; BNES:
London, 1987; pp 7783.

54.

55.
56.
57.
58.

59.

60.

61.
62.

63.
64.

65.

Berge, P.; Buisine, D.; Gelpi, A. PWSCC effect of initial


surface preparation. In Proceedings: 1997 EPRI Workshop
on PWSCC of Alloy 600 in PWRs, EPRI TR-109138, paper
E5, Nov 1997.
Inconel Alloy 600, International Nickel Company:
Huntington, WV, 1978.
McNeil, M. B. Nucl. Eng. Des. 1998, 181, 5560.
Scott, P. A review of irradiation assisted stress corrosion
cracking of austenitic materials for PWR core internals.
Eurocorr, Framatome, France, Sept. 1996.
Nelson, J. L.; Andresen, P. L. In Proceedings of Fifth
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems Water Reactors,
Monterey, CA, Aug, 1991; American Nuclear Society:
La Grange Park, IL, 1991; pp 1026.
Andresen, P. L.; Ford, F. P.; Murphy, S. M.; Perks, J. M.
In Proceedings of 4th International Symposium on
Environmental Degradation of Materials in Nuclear Power
Systems Water Reactors, Jekyll Island, GA, Aug, 1989;
National Association of Corrosion Engineers: Houston, TX,
pp 1-831-121.
Hanninen, H.; Aho-Mantilla, I. In Proceedings of 3rd
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems Water Reactors,
Traverse City, MI, Aug 30Sept 3, 1987; The Minerals,
Metals, and Materials Society: Warrendale, PA, 1987;
pp 7792.
Scott, P. J. Nucl. Mater. 1994, 211, 101122.
Andresen, P. L. In Stress Corrosion Cracking Materials
Performance and Evaluation; Jones, R. H., Ed.;
American Society for Metals: Metals Park, OH, 1992;
pp 182210.
Materials Reliability Program: Stress Corrosion Cracking
of High Strength Reactor Vessel Internals Bolting in PWRs
(MRP-88); EPRI Report 1003206; 2003.
Bajaj, R.; Mills, W. J.; Lebo, M. R.; Hyatt, B. Z.; Burke, M.
G. In Proceedings of Seventh International Symposium on
Environmental Degradation of Materials in Nuclear Power
SystemsWater Reactors; American Nuclear Society:
La Grange Park, IL, 1997.
Conference on Vessel Penetration Inspection, Crack
Growth and Repair (NUREG/CP-0191). Compiled by:
Mintz, T. S., Cullen, W. H., Sr.; US Nuclear Regulatory
Commission, Mar 2004.

5.05 Corrosion and Stress Corrosion Cracking of Austenitic


Stainless Steels
U. Ehrnsten
VTT Technical Research Centre of Finland, Espoo, Finland

2012 Elsevier Ltd. All rights reserved.

5.05.1

Introduction to Austenitic Stainless Steels

5.05.1.1
5.05.1.2
5.05.1.3
5.05.2
5.05.2.1
5.05.2.1.1
5.05.2.1.2
5.05.2.1.3
5.05.2.1.4
5.05.2.1.5
5.05.2.2
5.05.2.3
5.05.3
5.05.4
References

Types, Mechanical Properties, and Microstructures


Welding
Components Made of Stainless Steels in BWRs and PWRs
Stress Corrosion Cracking
IGSCC in BWR Environment
Degree of sensitization
Deformation
Environment
Stress
Components at risk
IGSCC in PWR Environment
TGSCC in BWR and PWR Environments
Pitting Corrosion
Microbiologically Induced Corrosion

Abbreviations
BWR
CGR
ECP
EPR
HAZ
HWC
IGSCC
K
KISCC
LWR
MIC
NG
NMC
NMCA
NWC
PLEDGE
PWR
RBMK
SCC
TGSCC
VVER

Boiling water reactor


Crack growth rate
Electrochemical corrosion potential
Electrochemical (potentiokinetic)
reactivation
Heat-affected zone
Hydrogen water chemistry
Intergranular stress corrosion cracking
Stress intensity factor
Threshold stress intensity for SCC
Light water reactor
Microbiologically influenced corrosion
Nuclear grade
Nobel metal chemistry
Noble metals chemistry addition
Normal water chemistry
Plant life extension and diagnosis by
GE (General Electric)
Pressurized water reactor
Channel type graphite moderated
reactor
Stress corrosion cracking
Transgranular stress corrosion cracking
Water-water energetic reactor

93
94
94
94
96
96
96
98
98
100
101
101
101
102
102
103

5.05.1 Introduction to Austenitic


Stainless Steels
Austenitic stainless steels have rendered their vast
use because of their good performance in corrosive
environments, in addition to their excellent ductility,
formability, toughness, and weldability. The good
corrosion resistance of austenitic stainless steels is
mainly due to chromium alloying, resulting in a protective, chromium-rich passive film on the material
in many environments. Molybdenum, used as an
alloying element in Type 316 stainless steels, further
increases the corrosion resistance. Chromium and
molybdenum are, however, both ferrite-forming elements, and to maintain a fully austenitic structure,
a balance between austenite-stabilizing elements
(C, N, Ni, Mn, and Co) and ferrite-stabilizing elements (Cr, Mo, Si, Ti, Nb, Al, V, and W) in solution
must be established.1 To compensate for the molybdenum addition in Type 316 stainless steels, the
amount of nickel is increased. In materials for nuclear
environments, the cobalt content is kept as low as
possible (<0.02%), because of its strong influence
on radioactivity buildup.
Stress corrosion cracking (SCC) is taken into account
in the design codes for light water reactors (LWRs,
93

94

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

i.e., boiling and pressurized water reactors (BWRs


and PWRs)) through a statement that SCC should not
occur.2 Intergranular stress corrosion cracking (IGSCC)
is still by far the largest damage mechanism for austenitic stainless steels in oxidizing BWR conditions, and
work on avoiding IGSCC is still going on.3,4 Several
factors affect IGSCC, one of them being sensitization.
Sensitization is a result of nucleation and growth of
chromium-rich carbides on grain boundaries, causing
grain boundary chromium depletion. Chromiumdepleted grain boundaries are prone to corrosion
and, in combination with a large enough stress and
suitable environment, to IGSCC. All measures to
prevent sensitization are therefore taken in all steps
of component manufacturing and plant operation.
Concerning the chemical composition, this can be
done by reducing the carbon content to levels below
0.03%, as is done in Types 316L, 316LN, and 316NG
stainless steels, where the carbon content is typically in the order of 0.02%. Since carbon is a
strengthening element, nitrogen is added to these
steels to still achieve good mechanical properties.
Nitrogen also reduces chromium-rich carbide formation, a concept that is utilized in the French RCCM
norms, which allow a carbon content of 0.035% in
their nitrogen-strengthened Type 316 stainless steel
with 0.08% N. The other approach to avoiding sensitization is to tie up the carbon into precipitates.
This is utilized in stabilized stainless steels, which
are of two main categories, titanium- and niobiumstabilized stainless steels, Type 321 and 347, respectively. To ensure that the carbon is tied into Ti(C,N)
or Nb(C,N) precipitates, a high enough stabilization
ratio, that is, Ti/C or Nb/C above 5 or 10, respectively, is specified.
Good corrosion resistance is ensured by restricting the amount of harmful elements, especially sulfur
and phosphorus, which may cause intergranular
corrosion when segregated to grain boundaries. Several standardized grain boundary corrosion tests,
such as the Strauss test5 and the electrochemical
potential reactivation (EPR) test,6 are employed routinely as part of acceptance tests for materials and
components.
5.05.1.1 Types, Mechanical Properties, and
Microstructures
The chemical compositions and mechanical properties of the most common stainless steels are presented
in Table 1 and the main mechanical properties are
summarized in Table 2.

In addition to compositional, mechanical, and corrosion resistance requirements, several other requirements are put on austenitic stainless steel materials
for nuclear components. These include, for example,
requirements on grain size. A small enough grain
size is needed to enable reliable nondestructive inspection requirements, using ultrasonic techniques. A common requirement is that the grain size must not exceed
ASTM number 4.0, which corresponds to an average
grain size of 90 mm. The grain size of stabilized stainless
steel components is typically smaller than this.
5.05.1.2

Welding

Stainless steels are generally welded with a slightly


over-alloyed filler metal to ensure good corrosion
resistance of the final joint. The weld shall contain
a small amount (>3% but <10%) of d-ferrite to
avoid solidification and liquation cracking.7 Welding
induces residual stresses, which together with the
operational stresses enhance crack initiation and
growth. The aim is, naturally, always to minimize
the residual stresses by a proper choice of welding
parameters, by securing a good fit between the parts
to be welded, etc. The use of a narrow-gap welding
technique has increased remarkably during the last
few decades. The narrow-gap welding method has
many advances as it results, for example, in a lower
level of residual stresses, a reduced weld volume, a
narrower heat-affected zone (HAZ) with lower risk of
sensitization, and less grain growth.8
5.05.1.3 Components Made of Stainless
Steels in BWRs and PWRs
Austenitic stainless steel is the main construction
material in nuclear power plants (NPPs) owing to
its good corrosion resistance, ease to manufacture different shapes, and good weldability. In BWRs, stainless
steels are used for piping and reactor pressure vessel
cladding and structures inside the pressure vessel,
including the core shroud (which separates the primary water upward flow through the core from the
downward flow in the annulus), the core plate (which
supports the bottom of the fuel), the top guide
(which aligns the top of the fuel bundles), the shroud
dome, the steam separators, etc. Austenitic stainless
steel is also largely used for other components such
as pumps, valves, shafts, sleeves, and in auxiliary systems such as water tanks.
The material choices for PWRs are essentially the
same. The steam generator (SG) tubes are made of

Chemical composition of common stainless steel alloys

Type

C (max.%)

Mn (max.%)

Si (max.%)

P (max.%)

S (max.%)

Cr

304

0.08

2.00

1.00

0.045

0.030

18.020.0

8.0010.5

304L

0.03

2.00

1.00

0.045

0.030

18.020.0

8.0012.0

304LN

0.03

2.00

1.00

0.045

0.03

18.020.0

8.0012.0

Z2CN19-10
316

0.035
0.08

2.00
2.00

1.00
1.00

0.040
0.045

0.030
0.030

18.520.0
16.018.0

9.010.0
10.014.0

2.003.00

316L

0.030

2.00

1.00

0.045

0.030

16.018.0

10.014.0

2.003.00

316LN

0.030

2.00

1.00

0.045

0.030

16.018.0

10.014.0

2.003.00

Z2CN18-12
321

0.038
0.08

2.00
2.00

1.00
1.00

0.040
0.045

0.030
0.030

17.018.2
17.019.0

11.512.5
9.0012.0

2.252.75

347

0.08

2.00

1.00

0.045

0.030

17.019.0

9.0013.0

Source: Wegst, C. E. Stahlschluessel Key to Steel; Stahlschluessel Wegst GmbH: Marbach, 1995.
a
The corresponding alloys according to the Swedish SS and the German DIN standards are also given.

Ni

Mo

Others

SS 2333/SS 2332
DIN 1.4301
SS 2352 DIN
1.4306
SS 2371 DIN
1.4311

0.100.16
0.08

SS 2343/SS 2347
DIN 1.4436/DIN
1.4401
SS 2348/SS 2353
DIN 1.4404/DIN
1.4435
SS 2375 DIN
1.4429

0.100.16
0.08

Corresponding
alloysa

5 C
min. Ti
10 C
min. Nb

SS 2337 DIN
1.4541
SS 2338 DIN
1.4550

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

Table 1

95

96

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

Table 2

Minimum room temperature mechanical properties of stainless steels (hot finished and/or annealed forging)

Type

Tensile strength
(min. MPa)

0.2% yield strength


(min. MPa)

Elongation
(min.%)

Reduction
in area
(min.%)

304
304L
304LN
Z2CN1910
316
316L
316LN
Z2CN1812
321
347

515
450
515
510
515
450
515
510
515
515

205
170
205
210
205
170
205
210
205
205

40
40
40
35
40
40
40
35
40
40

50
50
50
50
70
50
50

Source: Davis, J. ASM Specialty Handbook; ASM International: Materials Park, OH, 1994; ASM Standard A 240.

Ti-stabilized stainless steel in the Russian designed


Waterwater energetic reactors (VVERs), which are
PWRs with slightly different water chemistry and
horizontal instead of vertical SGs, as in Western
PWRs. The SG tubes in western PWRs are made
of nickel-based materials (Alloy 600, 690, or 800).
Further, both the SG vessel and the pressurizer are
clad with austenitic stainless steel.

5.05.2 Stress Corrosion Cracking


SCC is a failure mode caused by a combination of a
susceptible material, stresses, and an aggressive environment (Figure 1). There are two modes of SCC in
austenitic stainless steel, namely intergranular and
transgranular stress corrosion cracking (IGSCC and
TGSCC). IGSCC in austenitic stainless steel is the
major failure mode in BWRs, while it has not been
considered as a plausible failure mode in PWR primary water under normal operation conditions.
However, the number of IGSCC cases in PWRs has
increased by time, showing that PWRs are not totally
immune to IGSCC.
Most TGSCC cases are due to chloride-induced
SCC. TGSCC is rare in the primary system, and
the failure cases are typically observed in auxiliary
systems.
5.05.2.1

IGSCC in BWR Environment

In this section, the main factors affecting IGSCC in


BWR environment are reviewed, that is, degree of
sensitization, deformation, electrochemical corrosion
potential (ECP), water purity, and stress. Also several

Region of
potential stress
corrosion cracking

Stress

Environment

Material
susceptibility

Figure 1 The classic presentation of stress corrosion


cracking includes the three circles: material, environment,
and stress. Reproduced from General Electric Company.
Alternative Alloys for BWR Pipe Application; NP-2671-LD,
Final Report; San Jose, CA, 1982, with permission from
BWR Owners Group.

other parameters affect IGSCC susceptibility, but a


comprehensive description of these is out of the scope
of this chapter. Among these parameters are temperature, hydrogen, mechanisms related to localization
of deformation and to corrosion deformation interactions, such as effect of strain rate, subtle differences in
chemical composition, dynamic strain aging, dynamic
recovery, vacancy injection, selective dissolution, grain
boundary segregation, and relaxation.921
5.05.2.1.1 Degree of sensitization

Sensitization is the result of nucleation and growth of


chromium-rich carbides M23C6 at the grain boundaries, which results in a depletion of chromium at the
grain boundaries because of faster diffusion rate

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

along the grain boundaries compared to that within


the grain interior. Chromium-rich carbides form
within the temperature range of 500750  C, but
continue to grow down to much lower temperatures.
Sensitization can therefore occur as thermal sensitization during heat treatment and welding or as lowtemperature sensitization during long-time exposure
to LWR temperatures, below the chromium-rich carbide precipitation temperature.22 In the latter case,
the nucleation of carbides must have occurred previously, and the nucleated carbides grow during the
long-time exposure and deplete the grain boundaries
of chromium. The degree of sensitization is typically
measured using the EPR test, which is sensitive to the
area where the grain boundary chromium content is

below 15%, and is, thus, not a true measure of the


grain boundary chromium content.
The time for carbide precipitation increases as the
carbon level decreases as seen in Figure 2. Nitrogen
alloying delays carbide precipitation (Figure 3),
while deformation accelerates diffusion and precipitation. The obvious remedy to avoid sensitization
is, thus, to decrease the amount of free carbon, as
explained earlier, by reducing the carbon content, or
by tying carbon to Ti- or Nb-carbides and by
restricting the degree of deformation.
IGSCC in sensitized stainless steels occurs typically in the weld HAZ at a distance of 48 mm from
the fusion line, at the location where a high degree of
sensitization combined with high residual stresses

1000
1800
900

800
1400
316L (0.027% C)

316 (0.07% C)

Temperature (F)

Temperature (C)

1600
304 (0.053% C)

700

1200
600
1000
500
100

101

102

103

104

Time (min)

900
800

1832

0.069% N
0.145% N
0.247% N

1652

0.039% N

1472

700

1292

600

1112

500
0.01

0.1

10
Time (h)

100

1000

932

Precipitation temperature (F)

Precipitation temperature (C)

Figure 2 Timetemperatureprecipitation diagram for stainless steels with different carbon contents. Reproduced
from Shah, V. N.; MacDonald, P. E. Aging and Life Extension of Major Light Water Reactor Components; Elsevier:
Amsterdam, 1993.

1000

97

Figure 3 Effect of nitrogen on precipitation of M23C6 in a 0.05C17Cr13Ni5Mo stainless steel. Reproduced from
Peckner, D.; Bernstein, I. M. Handbook of Stainless Steels; McGraw-Hill: New York, 1977; pp. 4-354-53, pp 751757.

98

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

results in most severe conditions for IGSCC. The


typical location of IGSCC is different in nonsensitized stainless steels, where IGSCC occurs very close
to the fusion line, within the first few grains.
5.05.2.1.2 Deformation

Deformation increases the susceptibility of stainless


steels to IGSCC, in sensitized as well as in nonsensitized stainless steels, where the role of strain is decisive. Deformation occurs as bulk cold work from
rolling, bending, grinding, etc., as surface cold work
from machining, grinding, etc. and as weld shrinkage
from welding. Weld shrinkage can lead to up to 25%
equivalent room temperature strain in the weld HAZ
(Figure 4). The highest degree of deformation occurs
very close to the fusion boundary, and this is also the
location of observed IGSCC cracking in nonsensitized stainless steels pipes.2327 The importance of
cold deformation in IGSCC is shown in Figure 5.
Deformation is estimated to be the main affecting
parameter in 50% of all IGSCC cases covered in
the survey (including sensitized and nonsensitized
stainless steels). The effect of deformation has been
studied using bulk-deformed materials,2835 and the
results show a correlation between IGSCC crack
growth rate (CGR) and yield strength (Figure 6).

20

Various BWR stainless steel


weld HAZs

% Strain

15

10

10

15

20

25

30

35

Distance from weld fusion line (mm)


Figure 4 Deformation versus distance from the weld
fusion line in various stainless steel weld HAZs. Deformation
is expressed in terms of equivalent tensile strain at room
temperature, and results from weld shrinkage strains
during welding. Reproduced from Andresen, P. L.; et al.
In Corrosion 2000, NACE 55th Annual Conference,
Orlando, FL, Mar 2631, 2000; p 12, Paper No. 00203,
with permission from BWR Owners Group.

Much effort is nowadays put on deformation in


terms of restrictions on bulk and surface deformation
and on the development of sophisticated surface treatment procedures to remove surface cold work at critical locations.36,37 Application of narrow-gap welding
results both in a decrease in the degree of deformation
in the HAZ and in lower residual stresses. It should be
pointed out that some components, such as bolts, can
be made of intentionally cold-worked stainless steel to
increase the material strength.
5.05.2.1.3 Environment

As mentioned earlier, one of the main reasons for the


good behavior of austenitic stainless steels in LWR
conditions is the formation of a protective passive
film in high-temperature water (around 300  C).
The oxide film formed in high-temperature water
has a double-layered structure. The inner layer
grown on the metal surface consists of a chromium
spinel or magnetite and is covered by an outer layer
of magnetite or FeNi spinel precipitated from the
aqueous phase.38,39 The double-layered oxide structure forms so that faster diffusing elements pass
through the inner layer to the outer layer while the
slower diffusing elements, such as chromium, remain
in the inner layer and therefore the outer layer contains mainly of iron and the inner layer is enriched
with chromium. Although consensus is not yet reached
on the mechanistic details for corrosion and SCC in
LWR environments, breakage of the passive film is
generally considered to be of major importance
because of the fact that if the oxide film breaks, the
corrosion rate is high until passivation occurs.28,40,41
The CGR of IGSCC is highly dependent on the
oxidizing power of the environment, that is, the ECP
(Figure 7). The corrosion potential increases as the
oxygen content increases in the high-temperature

Cold work
Chemistry
Material
Nuclear grade
Ni-base alloy
Weld repair
Residual stresses
Sensitization

Figure 5 Cause of IGSCC in Swedish nuclear power


plants. Cold work is the biggest singular parameter affecting
IGSCC. Gott, K. Personal communication, April 2010.

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

99

Crack growth rate (mm s1)

1E 06
Very high
martensite

Unsensitized 304, 304L and 316L SS and A600


288 C high purity water, 2000 ppb O2
CT tests at 27.530 MPa m
Circles = High martensite SS
Triangles = Alloy 600
Predicted
response
Two sensitized points
for comparison

1E 07

Very low or no
martensite

Annealed cold worked

1E 08
0

100

200

300
400
500
Yield strength (MPa)

600

700

800

Figure 6 Effect of yield strength (and martensite content) on stress corrosion crack growth rate of unsensitized stainless
steels in oxygenated, high purity water. The predicted response is based on the PLEDGE model (plant life extension and
diagnosis by GE). Reproduced from Andresen, P. L.; et al. In Proceedings of 11th International Symposium on Environmental
Degradation of Materials in Nuclear Power Systems Water Reactors, Stevenson, WA, Aug 1114, 2003; American Nuclear
Society: La Grange Park, IL, 2003; p 435, with permission from BWR Owners Group.

water, but it is not a linear relationship, and small


changes in oxygen concentration can result in large
changes of ECP and CGR. Important to notice is
that the correlation between CGR and ECP is different for sensitized stainless steels and deformed, nonsensitized stainless steels, which show much higher
CGRs compared to sensitized materials at low potentials (although lower than at high ECP).
Two solutions to lower the ECP in BWRs have
been developed, that is, hydrogen water chemistry
(HWC) developed by General Electric and Asea
Brown Boveri, and noble metal chemistry addition
(NMCA, NobleChem), developed by General Electric. The ECP in a BWR recirculation circuit during
normal operation and using normal water chemistry
(NWC), that is, 200 ppb oxygen, is above 100 mVSHE.
The ECP is still higher in the core because of radiolytic decomposition of water forming hydrogen peroxide, H2O2. The ECP is remarkably lower in plants
using either HWC, where 40250 ppb hydrogen is
added to the feed water, or noble metal chemistry
(NMC), where a small amount of platinum is
added to the reactor water either at about 130  C
during startup or during full power operation
(OnLine NobleChem), creating an electrocatalytic
surface layer.42,43 The ECP of the buffered PWR
environment is in the lower range of the ECP curve,
that is, about 600 mVSHE. The trend, especially for
the US BWR fleet is toward HWC and NMCA. None

of the US BWR plants operates on NWC, and 75%


apply NMCA.44 The majority of the European BWRs
operate on NWC.
Water purity has a profound effect on both crack
initiation and CGR in oxidizing environments. The
main concerns are chlorides and sulfates for SCC and
additionally copper for pitting corrosion (although Cu
also has a synergistic effect on SCC). Sulfate and/or
chloride levels already in the ppb-range increase the
IGSCC susceptibility. Power plants monitor online the
conductivity, which is a mirror for water purity, and
analyze the amounts of impurities on regular basis
from grab samples. The conductivity of BWR primary
water of today has been reduced from a typical range
of about 0.4 mS cm1 in the 1970s to 0.10.2 mS cm1
(the conductivity of theoretically pure water is
0.056 mS cm1).
Dissolved oxygen is consumed inside cracks and
crevices, and the local ECP is reduced to low levels,
creating a potential gradient between the outer surface and the crack tip. This results in migration of
anions into the lower potential area, which results in
very high anion levels in the crack despite low levels
in the surrounding environment.45 Further, the local
environment in a crack can remain aggressive for a
long time after, for example, short periods of higher
impurity levels in the bulk environment.
It is not only the environment during steadystate operation that needs attention, but also the

100

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

Sensitized 304 stainless steel


30 MPa m, 288 C water
0.060.4 S cm1, 025 ppb SO4
SKI Round Robin Data
Filled triangle = Constant load
Open squares = Gentle cyclic

200 ppb O2
500 ppb O2
2000 ppb O2

1E-05
4 dpa
304SS

106
316L stainless steel
25 mm CT specimen
constant load
288 C water
Test conditions:
-2
0 C cm EPR
27.5 MPa m
200 ppb O2

20% CW
A600

316L (A14128, square)


304L (Grand Gulf, circle)
non-sensitized SS
50%RA 140C (black)
10%RA 140C (grey)

1E-07

42.5
28.3
14.2
mh1

20% CW A600
GE PLEDGE
predictions
30 MPa m
sens SS

0.5
2000 ppb O2
Ann. 304SS
200 ppb O2

0.25

1E-08

Crack growth rate (mm s1)

Crack growth rate (mm s1)

1E-06

107

108

0.1 S cm
Means from analysis of
120 L sens SS data
0.06 S cm1
0.06 S cm1

0.1

Predicted curves from


PLEDGE code for
typical range in ECP

GE PLEDGE predictions for


unsens. SS (upper curve for 20% CW)

1E-09
0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.1 0.2
(a)
Corrosion potential (VSHE)

109
0.3 0.4
(b)

101

100

101

Solution conductivity (S cm1)

Figure 7 Summary of crack growth rates of sensitized stainless steels versus corrosion potential, ECP (a) (reproduced
from Andresen, P. L.; Morra, M. M. J. Nucl. Mater. 2008, 383(12), 97111) and for nonsensitized stainless steels versus
solution conductivity (b) (reproduced from Andresen, P. L. Corrosion 1988, 44(7), 450). The prediction curves for different
water conductivity levels are according to the PLEDGE model, with permission from BWR Owners Group. RA reduction
in area; CW cold work.

environment during shutdown, downtime periods,


and during startup. The possible role of these will
increase with plant age and amount of shutdowns and
startups.
5.05.2.1.4 Stress

The stresses causing IGSCC are a combination of


residual and operational stresses, although the first is
considered more decisive in IGSCC failures. This is
because operational stresses are kept low by design
and components are usually designed to operate
below 80% of their yield strength. The CGR of
intergranular stress corrosion cracks increases with
increasing stress intensity factor (K ) (Figure 8). The
effect of stress intensity on CGR varies depending on
the material and environment. Knowledge of the
dependency between K and CGR is very important
for structural integrity calculations, which are made
to show that flaws, either postulated or detected using
nondestructive inspections, are tolerable and do not
pose a safety risk. Huge efforts have been put on the
production of high-quality laboratory CGR data and

efforts are still going on. Approved relationships (i.e.,


agreement reached between national safety authority
and plant operators) are called disposition lines, and
examples of published lines are shown in Figure 8.
Several methods to mitigate IGSCC have been
applied over the years, such as last pass heat sink
welding, mechanical stress improvement, and weld
overlay cladding.46 All these aim at producing a compressive stress state in the HAZ. However, these
methods are usually applied as temporary remedies.
Measurement of residual stresses is an area of
increased focus nowadays, and lack of knowledge can
result in excessive under- or over-conservatism in
design and in structural integrity calculations. Also
other stress-related factors affect IGSCC, such as
vibratory loading, thermal loads from, for example,
stratification, as well as load cycles during shutdowns
and startups. Much effort was earlier put on defining
the KISCC; that is, the stress intensity, below which SCC
would not occur. With improved laboratory testing
techniques, lower and lower KISCC values have been
measured and a true threshold value may not exist.

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

101

1E 04
NRC 0313, BWR NWC
MD-01, BWR NWC
JSME, BWR NWC
JSME, BWR HWC

Crack growth rate (mm s )

1E 05
1E 06
1E 07
1E 08
1E 09
1E 10
1E 11

10

20

30
40
50
60
Stress intensity (MPa m)

70

80

90

Figure 8 Crack growth rate versus stress intensity according to dispositions lines for sensitized stainless steels in
normal water chemistry boiling water reactors environment. Compiled by author from NRC Generic Letter GL880011. NRC
Position on IGSCC in BWR Austenitic Stainless Steel Piping; January 25, 1988; http://www.nrc.org; Jansson, C.; Morin, U.
In Proceedings of 8th International Symposium on Environmental Degradation of Materials in Nuclear power Systems Water
Reactors, Amelia Island, FL, Aug 1014; American Nuclear Society: La Grange Park, IL, 1997;
pp 667674; Kobayashi, H.; Kashima, K. Int. J. Press. Vess. Pip. 2000, 77, 937944. JSME is the Japan Society of
Mechanical Engineers; NRC is the Nuclear Regulatory Commission.

5.05.2.1.5 Components at risk

The earliest incidents of SCC in BWRs occurred in


stainless steel fuel cladding, before zirconium alloys
were used.47 IGSCC plagued the BWRs in the 1970s
and caused a clear reduction in capacity factors. Cracking was first observed in the recirculation and water
cleanup systems in pipes with small diameter and later
also in larger diameter pipes. The material was mainly
Type 304 with a high carbon content of 0.6%. Owing
to large efforts to solve the problem, including the
development of Type 316NG, narrow-gap welding
technique, as well as low-potential water chemistries,
the number of IGSCC incidents has remarkably
reduced.3 In the late 1980s, cracking in Ti-stabilized
stainless steel piping was detected.4850 Robust mitigation measures, including adoption of narrow-gap
welding, change of material to Nb-stabilized stainless
steel with higher stabilization ratio requirements, and
reduction of the amount of welds, were applied in
Germany to solve the problem. Also from Russian
channel type graphite moderated reactors (RBMKs),
which operate under BWR-like conditions, numerous
IGSCC cases have been reported.51,52 In the 1990s,
the first cases with IGSCC in nonsensitized stainless
steels were reported in BWRs,23,25 first in pipings,
and later numerously in core shrouds.27 Deformation
(weld shrinkage in piping and surface grinding in

the core shrouds) is considered to be of major importance in these cases.


5.05.2.2

IGSCC in PWR Environment

PWRs operate at low corrosion potentials and very


low oxygen levels, <30 ppb. The risk of IGSCC
in austenitic stainless steels in nonoxidizing environment is, thus, much lower than in NWC BWR
environment.31,53 Incidences with IGSCC under nominal PWR conditions have not been reported. Oxygen
can, however, be enclosed in certain situations, such as
startups, and can lead to IGSCC in austenitic stainless
steels. Although the number of IGSCC cases in PWRs
is still very low, the number seems to be increasing.54,55
IGSCC has been observed in pressurizer heater
sleeves, canopy seals in the control rod drives, SG
safe-ends, etc. Laboratory tests on cold-worked stainless steels show that IGSCC is possible also in normal
PWR environment, indicating (although not generally accepted) that more IGSCC failures may occur in
the operating PWR plants with time.
5.05.2.3 TGSCC in BWR and PWR
Environments
Austenitic stainless steels are prone to TGSCC when
exposed to aggressive oxidizing water, for example,

102

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

containing chlorides, under sufficiently high stresses.


As chloride levels in BWR and PWR primary systems
are kept low, TGSCC is rare under normal operating
conditions. The exception is TGSCC from the secondary side in VVER SG tubes, manufactured of
Ti-stabilized stainless steels. Condenser leakages, frequently reported in VVERs, can result in chloride
contamination of the secondary water and eventually
in TGSCC in the SG tubes. Copper release from brass
condensers and crevices formed by iron deposits further enhance both TGSCC and pitting corrosion on
the secondary side of the SG tubes.
A risk of TGSCC exists at locations where (a slow)
buildup of aggressive conditions can occur. The risk
of TGSCC increases with plant age, as the buildup of
aggressive conditions can be very slow and can occur
at unknown (uninspected) locations. Known chloride
sources are old insulation and sealing materials (e.g.,
asbestos), leakage from cables, polymers, paints, concrete, etc. Wet insulation is the worst of these, as
the insulation provides crevice conditions in addition
to a chloride source. Strict regulations for expendables
(grease, cleaning agents, sealing materials, etc.) allowed
in NPPs are applied to reduce the risk of buildup of
aggressive conditions. All bare outer surfaces of austenitic stainless steel components, where humidity
may exist, can be at risk for TGSCC.
TGSCC has been reported in valves,56 for example, where the source for the chlorides is assumed to
be asbestos sealing used early in time and in water
tanks, where chlorides probably stem from humidity
and concentration buildup at the waterline. Stainless
steel bellows in the BWR reactor containment are, in
principle, at risk because of the high degree of cold
work in the bellows. However, no SCC has been
reported in these. New components can also be at
risk for TGSCC, if proper measures are not taken to
avoid contamination of components during transportation, storage, and installation.
TGSCC can occur in oxidizing concentrated boric
acid solutions although laboratory results are not fully
conclusive whether chloride is also needed or not.5658

is very often observed at same locations as TGSCC, but


pitting corrosion can also occur without SCC and vice
versa. The risk of pitting corrosion under normal BWR
conditions is extremely low. However, pitting corrosion
can occur in pressure boundary systems at locations
where (slow) buildup of aggressive local conditions can
occur. Such locations are, for example, areas with low
water flow, dead ends, and valves with sealing.
As pitting occurs only in oxidizing conditions, it is
not a plausible degradation mechanism in PWR primary water under nominal environmental conditions.
However, the environment may be oxidizing both
locally and/or temporarily because of startups, for
example. Different systems during shutdown may be
filled with air, and this may cause air pockets during
startup. The oxygen from air will then dissolve into
the primary water and local oxidizing conditions
temporarily emerge until the oxygen is consumed by
the oxidation of metal surfaces. The risk of pitting
corrosion (and TGSCC) is, however, highest in auxiliary systems, for example, at outer surfaces, where the
temperature is low enough for condensation to occur.
Thus, pitting corrosion can occur at nominally dry
locations. Accumulation of aggressive local conditions
is enhanced by crevices.
The sources of chlorides were listed earlier.
Sulfate sources have been introduced earlier, for
example, in molybdenum disulfide greases, but
since the harmful influence of this material was identified, it is not an allowed expendable material. Again,
copper can enter the system from copper-containing
structural components.
Pitting corrosion is seldom considered to pose a
safety problem, as the wall thicknesses of pressure
boundary components are usually large enough to
sustain pitting corrosion for long times without leakage. However, pitting corrosion is always an indication of a harmful environment existing at the location
and is often associated with the risk of TGSCC,
which can cause wall cracking in short time periods.
Pitting corrosion enhances the risk of SCC as the pits
increase the local stress concentration and thus act as
crack initiators. Observation of pitting corrosion shall
therefore not be omitted as insignificant.

5.05.3 Pitting Corrosion


Pitting corrosion occurrence has several similarities to
TGSCC, that is, it requires oxidizing conditions and
presence of water with harmful ions, such as chlorides,
fluorides, sulfates, and/or copper, but no stress is needed.
The Type 304 stainless steel is more prone to pitting
corrosion than Type 316 stainless steel. Pitting corrosion

5.05.4 Microbiologically Induced


Corrosion
A rather rare corrosion mode is microbiologically
induced corrosion, or nowadays, microbiologically influenced corrosion (MIC). MIC is normal

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

electrochemical corrosion where the microorganisms


either chemically or physically change the conditions
on the metal surface to be favorable to corrosion.59
MIC appears as localized corrosion rather than as
uniform corrosion, and in welds rather than in base
materials. Pitting corrosion in the weld metal can
cause preferential attack of either the austenite or
the ferrite phase of the weld metal. The microorganisms of interest in MIC are mostly bacteria and fungi.
The highest risk of MIC is at temperatures from 15
to 45  C and near neutral pH, that is, in the range
from 6 to 8. MIC has been observed in fire-fighting
systems, for example.
MIC is stopped with great difficulty once it is
established due to the high sustainability of the microorganisms involved. The quality of the water in all
phases of the lifetime of the equipment is, thus, very
important at locations with risk of MIC. Water of high
quality must be used, not only during normal operation,
but also during hydrotesting of the system, for example.

13.
14.

15.

16.

17.
18.
19.
20.
21.
22.

References
1.
2.

3.
4.
5.

6.

7.
8.

9.
10.
11.
12.

Peckner, D.; Bernstein, I. M. Handbook of Stainless Steels;


McGraw-Hill: New York, 1977; pp 4-354-53, 751757.
Ford, P. F. In Proceedings of 13th International
Symposium on Environmental Degradation of Materials in
Nuclear Power Systems Water Reactors, Whistler, BC,
Canada, Aug 1923, 2007; Allen, T. R., et al. Eds.;
Canadian Nuclear Society: Toronto, ON, 2007.
NRC. Expert Panel Report on Proactive Management of
Materials Degradation; NUREG/CR-6923; US Nuclear
Regulatory Commission: Washington, DC, 2007.
Shah, V. N.; MacDonald, P. E. Aging and Life Extension of
Major Light Water Reactor Components; Elsevier:
Amsterdam, 1993.
ASTM Standard A 262-02a. Standard Practices for
Detecting Susceptibility to Intergranular Attack in
Austenitic Stainless Steels; ASTM International: West
Conshohocken, PA, 2008; p 17.
ASTM Standard G 108-94. Standard Test Method for
Electrochemical Reactivation (EPR) for Detecting
Sensitization of AISI Type 304 and 304L Stainless Steels;
ASTM International: West Conshohocken, PA, 2004.
Folkhard, E. Welding Metallurgy of Stainless Steel;
Springer-Verlag: Wien, 1988.
Korhonen, M.; Luukas, M.; Hanninen, H. In International
Conference on Efficient Welding in Industrial Applications
(ICEWIA), Lappeenranta, Finland, Aug 2527, 1999;
Martikainen, J., Eskelinen, H., Eds.; Lappeenranta
University of Technology: Finland, 1999; pp 244255
Aaltonen, P.; Saario, T.; Karjalainen-Roikonen, P.; et al.
In Corrosion 96; Denver, CO, Mar 2429, 1996, NACE
International: Houston TX, 1996; p 12 , Paper No. 81.
Andresen, P. L.; Briant, C. L. Corrosion 1989, 45, 448463.
Arioka, K.; Yamada, T.; Terachi, T.; Miyamoto, T.
Corrosion 2008, 64(9), 691706.
Birnbaum, H. K. In Hydrogen Effects on Material Behavior;
Moody, N. R., Thompson, A. W., Eds.; TMS: Warrendale,
PA, 1990; pp 639660.

23.

24.
25.

26.

27.
28.
29.

30.
31.

103

Boursier, J. M.; Desjardins, D.; Vaillant, F. Corros. Sci.


1995, 37(3), 493508.
Briant, C. L.; Andresen, P. L. In Proceedings of 3rd
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems Water Reactors,
Traverse City, MI, Aug 30Sept 3, 1998; Theus, G. J.,
Weeks, J. R., Eds.; TMSAIME: Warrendale, PA, 1988;
pp 371382.
Couvant, T.; Vaillant, F.; Boursier, J. M. Effect of strainpath on stress corrosion cracking of AISI 304L stainless
steel in PWR primary environment at 360  C. In
Proceedings of Eurocorr 2004, Nice, France, Sept 1216,
2004; p 11, Event No. 226.
Ehrnsten, U.; Ivanchenko, M.; Nevdacha, V.;
Yagodzinskyy, Y.; Toivonen, A.; Hanninen, H.
In Proceedings of 12th International Symposium on
Environmental Degradation of Materials in Nuclear Power
Systems Water Reactors, Salt Lake City, UT, Aug 1418,
2005; Allen, T. R., King, P. J., Nelson, L., Eds.; TMS:
Warrendale, PA, 2005; pp 14751482.
Ferreira, P. J.; Robertson, I. M.; Birnbaum, H. K. Acta
Mater. 1998, 46(5), 17491757.
Hall, M. M., Jr. Corros. Sci. 2008, 50, 29022905.
Hall, M. M., Jr. Corros. Sci. 2009, 51, 225233.
Hong, S. G.; Lee, S.-B. J. Nucl. Mater. 2004, 328,
232242.
McDonald, D. D. J. Electrochem. Soc. 1992, 139,
34343449.
Kekkonen, T.; Aaltonen, P.; Hanninen, H. Corros. Sci.
1985, 25, 821836.
Angeliu, T. M.; Andresen, P.; Hall, E.; Sutliff, J.;
Sitzman, S.; Horn, R. In Proceedings of 9th International
Symposium on Environmental Degradation of Materials in
Nuclear Power Systems Water Reactors. Newport
Beach, CA, Aug 15, 1999; Ford, F. P., Bruemmer, S. M.,
Was, G. S., Eds.; TMSAIME: Warrendale, PA, 1999.
Angeliu, T. M.; Hall, E.; Sutliff, J.; Sitzmen, S.; Andresen, P.
In Corrosion 2000; NACE International: Houston, TX, 2000;
Paper No. 00186.
Ehrnsten, U.; Hanninen, H.; Aaltonen, P.; Jansson, C.;
Nenonen, P.; Angeliu, T. In Proceedings of 10th
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems Water Reactors,
Lake Tahoe, NV, Aug 59, 2001; Ford, F. P., Was, G., Eds.;
NACE International: Houston, TX, 2001; p 10.
Ooki, S.; Tanaka, Y.; Takamori, K. In Proceedings of
12th International Symposium on Environmental
Degradation of Materials in Nuclear Power Systems
Water Reactors, Salt Lake City, UT, Aug 1418, 2005;
Allen, T. R., King, P. J., Nelson, L., Eds.; TMS: Warrendale,
PA, 2005; pp 365376.
Yamashita, H.; Ooki, S.; Tanaka, Y.; Takamori, K.;
Asano, K.; Suzuki, S. Int. J. Press. Vess. Pip. 2008, 85,
582592.
Andresen, P. L. Corrosion 1988, 44(7), 450.
Andresen, P. L. In Proceedings of 10th International
Symposium on Environmental Degradation of Materials in
Nuclear Power Systems Water Reactors, Lake Tahoe,
NV, Aug 59, 2001; Ford, F. P., Was, G., Eds.; NACE
International: Houston TX, 2001; p 9.
Andresen, P.; Angeliu, T.; Catlin, W.; Young, L.; Horn, R.
In Corrosion, 2000, NACE 55th Annual Conference,
Orlando, FL, Mar 2631, 2000; p 12, Paper No. 00203.
Andresen, P. L.; Emigh, P.; Morra, M.; Horn, R.
In Proceedings of 11th International Symposium on
Environmental Degradation of Materials in Nuclear Power
Systems Water Reactors, Stevenson, WA, Aug 1114,
2003; American Nuclear Society: La Grange Park, IL,
2003; pp 816833.

104

Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels

32. Andresen, P. L.; Morra, M. M. J. Nucl. Mater. 2008,


383(12), 97111.
33. Horn, R.; Gordon, G.; Ford, P.; Cowan, R. Nucl. Eng. Des.
1997, 174, 313325.
34. Magdowski, R.; Speidel, M. O. In Corrosion 96, NACE,
51st Annual Conference and Exposition, Denver, CO,
Mar 2429 1996; p 6, Paper No. 112.
35. Tahtinen, S.; Hanninen, H.; Trolle, M. In Proceedings of the
6th Symposium on Environmental Degradation of
Materials in Nuclear Power Systems Water Reactors,
San Diego, CA, Aug 15, 1993; Gold, R. E., Simonen, E. P.,
Eds.; TMS: Warrendale, PA, 1993.
36. Prevey, P. S.; Jayaraman, N. In Proceedings of ICSP 9,
Paris, Marne la Vallee, France, Sept 69, 2005; p 7,
Paper No. 260.
37. Offer, H. P.; Morra, M.; Chan, A. In Proceedings of
13th International Symposium on Environmental
Degradation of Materials in Nuclear Power Systems
Water Reactors, Whistler, BC, Canada, Aug 1923, 2007;
Allen, T. R., et al. Eds.; Canadian Nuclear Society:
Toronto, ON, 2007; p 17.
38. Robertson, J. Corros. Sci. 1991, 32, 443465.
39. Stellwag, B. Corros. Sci. 1998, 40, 337370.
40. Ford, P.; Taylor, D.; Andresen, P.; Ballinger, R. Corrosion
Assisted Cracking of Stainless Steel and Low Alloys Steel
in LWR Environments; Report NP5064S; Electric Power
Research Institute: Palo Alto, CA, 1987; p 124.
41. Ford, P. F.; Andresen, P. L. In Corrosion Mechanisms in
Theory and Practice; Marcus, P., et al. Ed.; Marcel
Dekker: New York, 1995; pp 501546.
42. Hettiarachchi, S. In Proceedings of 11th International
Symposium on Environmental Degradation of Materials in
Nuclear Power Systems Water Reactors, Stevenson,
WA, Aug 1114, 2003; American Nuclear Society:
La Grange Park, IL, 2003; pp 477487.
43. Hettiarachchi, S. In Proceedings of 10th International
Conference on Environmental Degradation of Materials in
Nuclear Power Systems-Water Reactors, Lake Tahoe, NV,
Aug 59, 2007; Ford, F. P., Was, G., Eds.; NACE
International: Houston, TX, 2007; p 10.
44. Andresen, P. L. Personal communication, May 2009.
45. Andresen, P. L. In Proceedings of 5th International
Symposium on Environmental Degradation of Materials in
Nuclear Power Systems Water Reactors, Monterey, CA,
Aug 2529, 1995, American Nuclear Society: La Grange
Park, IL, 1995; pp 209218.
46. General Electric Company. Alternative Alloys for
BWR Pipe Application; NP-2671-LD, Final Report;
San Jose, CA, 1982.
47. Hanninen, H.; Aho-Mantila, I. In Proceedings of 3rd
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems Water Reactors,
Traverse City, MI, Aug 30Sept 3, 1987; TMSAIME:
Warrendale, PA, 1987; pp 7792.

48.

49.

50.

51.

52.
53.

54.

55.

56.
57.

58.

59.

Erve, M.; Wesseling, U.; Kilian, R.; et al. In 20th


MPA-Seminar, Stuttgart, Oct 67, 1994; Staatliche
Materialprufungsanstalt (MPA) Universitat Stuttgart:
Stuttgart, Germany, 1994; Vol. 2. p 21, Paper No. 29.
Kilian, R. In Proceedings of 7th International Symposium
on Environmental Degradation of Materials in Nuclear
Power Systems Water Reactors, Breckenridge, CO,
Aug 710, 1995; Airey, G., Ed.; NACE International:
Houston, TX, 1995; pp 529540.
Kilian, R.; Eberle, U.; Brummer, G.; et al. In Proceedings of
the 9th Environmental Degradation of Materials in Nuclear
Power Systems Water Reactors, Newport Beach, CA,
Aug 15, 1999; Bruemmer, S. M., Ford, F. P., Was, G. S.,
Eds.; TMSAIME: Warrendale, PA, 1999; pp 347357.
International Atomic Energy Agency. Mitigation of
Intergranular Stress Corrosion Cracking in RBMK
Reactors; Final Report of the Programmes Steering
Committee; IAEA-EBP-IGSCC; IAEA: Vienna, 2002.
Timofeev, B.; Fedorova, V.; Buchatskii, A. Mater. Sci.
2004, 40(1), 4859.
Scott, P. M. In Proceedings of 9th International
Symposium on Environmental Degradation of Materials in
Nuclear Power Systems Water Reactors, Newport
Beach, CA, Aug 15, 1999; Ford, F. P., Bruemmer, S. M.,
Was, G. S., Eds.; TMSAIME: Warrendale, PA, 1999.
Couvant, T.; Legras, L.; Pokor, C.; et al. In Proceedings of
13th International Symposium on Environmental
Degradation of Materials in Nuclear Power Systems
Water Reactors, Whistler, BC, Canada, Aug 1923, 2007;
Allen, T. R., et al. Eds.; Canadian Nuclear Society:
Toronto, ON, 2007; Vol. 1. pp 499514.
Chynoweth, J.; Hyres, J. In Proceedings of 13th
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems Water Reactors,
Whistler, BC, Canada, Aug 1923, 2007; Canadian
Nuclear Society: Toronto, ON, 2007; Vol. 2, pp 12141225.
Kilian, R.; Wesseling, U.; Wachter, O.; Widera, M.;
Brummer, G.; Ilg, U. In Fontevraud V; Sept 2327, 2002;
SFEN: France, 2002.
Berge, P.; Keroulas, F.; Gras, J.; Noel, D.; Da Vunha
Belo, M. In Proceedings of 4th International Symposium on
Environmental Degradation of Materials in Nuclear Power
Systems Water Reactors, Jekyll Island, GA, Aug 610,
1989; Cubicciotti, D., Ed.; NACE International: Houston,
TX, 1989; pp. 11-7411-87.
McDonald, D.; Cragnolino, G.; Olemacher, J.; Chen, T.;
Dhawale, S. Intergranular Stress Corrosion Cracking of
Austenitic Stainless Steels in PWR Acid Storage Systems;
EPRI NP-2531; Electric Power Research Institute: Palo
Alto, CA, 1982.
Cramer, S., Covino, B., Jr., Moosbrugger, C. Eds.
Handbook Volume 13A: Corrosion Fundamentals, Testing
and Protection; ASM International: Materials Park, OH,
2003.

5.06 Corrosion and Environmentally-Assisted Cracking of


Carbon and Low-Alloy Steels
H.-P. Seifert
Paul Scherrer Institut, Villigen PSI, Switzerland

J. Hickling
Independent Technical Consultant, Prastio-Avdimou, Cyprus

D. Lister
University of New Brunswick, Fredericton, NB, Canada

2012 Elsevier Ltd. All rights reserved.

5.06.1

Introduction

107

5.06.2
5.06.2.1
5.06.2.2
5.06.2.2.1
5.06.2.2.2
5.06.2.2.3
5.06.3
5.06.3.1
5.06.3.2
5.06.3.2.1
5.06.3.2.2
5.06.3.2.3
5.06.3.2.4
5.06.3.2.5
5.06.4
5.06.4.1
5.06.4.2
References

Uniform and Flow-Accelerated Corrosion


Uniform Corrosion
Flow-Accelerated Corrosion
Controlling factors
Mechanisms and models
Service experience and mitigating actions
Localized Corrosion and Environmentally Assisted Cracking
Pitting
Environmentally Assisted Cracking
Basic types of EAC and major factors of influence
Corrosion fatigue and strain-induced corrosion cracking
Stress corrosion cracking
EAC mechanisms and models
Service experience and mitigation actions
Conclusions
Uniform and Flow-Accelerated Corrosion
Localized Corrosion and Environmentally Assisted Cracking

109
109
111
111
114
118
120
120
122
122
123
128
132
136
139
139
139
140

Abbreviations
AC
AGR
ANL
ASME
ASME BPV
ASME III
ASME XI
ASTM
BWR
BWRVIP
BWRVIP-60

Content of Cr, Mo and Cu in alloy in


EPRI CHECWORKS FAC-Code
Advanced gas-cooled reactor
Argonne National Laboratory, USA
American Society of Mechanical
Engineers
ASME Boiler and Pressure Vessel
Code
Section III of ASME BPV Code
Section XI of ASME BPV Code
American Society of Testing and
Materials Standards
Boiling water reactor
Boiling Water Reactor Vessel and
Internals Program
Basis document for SCC
disposition lines for low-alloy
steels

CANDUW

CF
CRDM
CS
DCPD

DH
DL
DO
DSA
EAC
EC
ECP
ECPcrit

CANada Deuterium Uranium,


PHWR developed by Atomic Energy
of Canada Ltd.
Corrosion fatigue
Control rod drive mechanism
(housing)
Carbon steel
(Reversed) direct current potential
drop crack length measurement
method
Dissolved hydrogen (concentration)
Disposition line
Dissolved oxygen (concentration)
Dynamic strain ageing
Environmentally assisted cracking
Erosion corrosion
Electrochemical corrosion potential
Critical cracking potential (e.g., for
SICC)

105

106

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

EPRI

Electric Power Research Institute,


USA
F & A model EAC model for CS & LAS developed
by P. Ford and P. Andresen (GE GR)
FAC
Flow-accelerated corrosion
FRAD
Film rupture anodic dissolution EAC
mechanism
HAEAC
Hydrogen-assisted EAC
mechanism
HAZ
Heat-affected zone of weldment
HCF
High-cycle fatigue
HT
High temperature
HWC
Hydrogen water chemistry
JSME
Japanese Society of Mechanical
Engineers
LAS
Low-alloy steel
LCF
Low-cycle fatigue
LEFM
Linear elastic fracture mechanics
LWR
Light water reactor
MT
Mass transfer in EPRI
CHECWORKS FAC-Code
NDT
Nondestructive testing
NMCA
Noble metal chemical addition
NRC
Nuclear Regulatory Commission,
USA
NWC
Normal water chemistry
PHWR
Pressurized heavy water reactor
PWHT
Postweld heat treatment
PWR
Pressurized water reactor
PWSCC
Primary water stress corrosion
cracking (in PWRs)
QT
Quench and temper heat treatment
RPV
Reactor pressure vessel
SCC
Stress corrosion cracking
SEM
Scanning electron microscope
SHE
Standard hydrogen electrode
SICC
Strain-induced corrosion cracking
SS
Stainless steel
SSR(T)
Slow strain rate (test)
SSY
Small-scale yielding
UTS
Ultimate tensile strength
VGB
German Association of Large Power
Plant Operators
YS
Yield strength

Ceq

Symbols

KI
KI,i

C
Cb

Concentration of Fe(II) species at


the oxidecoolant interface
Concentration of Fe(II) species in
the bulk coolant

CODLL

d
D
da/dN
da/dNAir
da/dNCF

da/dtAir
da/dNAir/
DtR
da/dtCF
da/dNCF/
DtR
da/dtSCC
da/dtSICC
dCODLL/dt

de/dt
de/dtcrit
dKI/dt
EA
Fen

G
h

kc
kd

Thermodynamic equilibrium
concentration of Fe(II) species
Crack-opening displacement at
load line in precracked fracture
mechanics specimen
Pipe diameter
Diffusivity
Crack advance per fatigue cycle
Crack advance per fatigue cycle in
air
Corrosion fatigue crack advance
per fatigue cycle in hightemperature water
Time-based fatigue crack growth
rate in air
Time-based corrosion fatigue
crack growth rate in hightemperature water
SCC crack growth rate
SICC crack growth rate
Crack-opening displacement rate
in slow rising load or displacement
test
Strain rate (sometimes locally at
crack-tip)
Critical strain rate (e.g., for SICC)
Stress intensity factor rate in slow
rising load or displacement test
Arrhenius activation energy of
thermally activated process
Environmental correction factors,
ratio of fatigue life in air at room
temperature to that in water at
service temperature
Geometry factor in EPRI
CHECWORKS FAC-Code
Mass transfer coefficient for Fe(II)
species from the oxidecoolant
interface to the bulk environment
by convection
Geometry factor in Siemens-KWU
WATHEC FAC-Code
Dissolution reaction rate constant
of magnetite at the oxidecoolant
interface
Stress intensity factor (LEFM)
Stress intensity factor at the onset
of SICC crack growth in slow rising
load tests with precracked fracture
mechanics specimens

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

KIJ

Stress intensity factor at the onset


of ductile crack growth
m
Paris law exponent for fatigue and
corrosion fatigue
n
Frequency exponent for corrosion
fatigue
pH  log[H] pH value
R
Load ratio of minimum to
maximum load
Flow-accelerated corrosion rate
RFAC
Dissolution rate of magnetite at the
Rd
oxidecoolant interface
Formation rate of magnetite at the
Rg
metal/oxide interface
Mass transport rate of Fe(II)
Rm
species from the oxidecoolant
interface to the bulk environment
by convection
Re dur/m
Reynolds number
Sc m/( rD)
Schmidt number
Sh hd/D
Sherwood number
t
Time
T
Temperature
u
Coolant velocity
Z
Reduction of area in tensile test
[X]
Concentration of element/species
X in water or in alloy
a
Steam void fraction in EPRI
CHECWORKS FAC-Code
DC Ceq  Cb Undersaturation in dissolved Fe(II)
species
Dd
Wall thinning
DK KI,max  Stress intensity factor range of
KI,min
fatigue cycle
Upper DK threshold for corrosion
DKCF,H
fatigue
Lower DK threshold for corrosion
DKCF,L
fatigue
DK threshold for fatigue in air
DKth, Air
Decline time (down-ramp) of
DtD
fatigue cycle
Hold time at maximum load of
DtH
fatigue cycle
Rise time (up-ramp) of fatigue cycle
DtR
e
Mechanical strain
Critical strain (e.g., for SICC)
ecrit
k
Specific electrical conductivity
m
Viscosity of coolant
n
Loading frequency
Upper critical frequency for
ncrit,H
corrosion fatigue

ncrit,L
r
s
scrit
t

107

Lower critical frequency for


corrosion fatigue
Density
Mechanical stress
Critical stress (e.g., for SCC)
Fluid shear stress at pipe wall

5.06.1 Introduction
Carbon and low-alloy steels (CS & LAS, Table 1)
and their associated weld filler metals are widely
used for pressure vessels and piping in both the
primary and secondary coolant circuits of watercooled reactors (light water reactors (LWRs) and
CANDUs pressurized heavy water reactors
(PHWRs)), as well as in service water systems.1
The main reasons for the use of CS & LAS are
their combination of relatively low cost, good
mechanical strength and toughness properties in
thick sections (hardenability), and good weldability,
as well as their good stress corrosion cracking (SCC)
resistance in primary coolant environments. Compared with austenitic stainless steels and nickel-base
alloys, ferritic CS & LAS exhibit only moderate
corrosion and irradiation resistance. They also show
a ductile-to-brittle transition in toughness properties
at lower temperatures.
CS & LAS components in the primary circuit of
pressurized water reactors (PWRs) are clad (usually
with austenitic stainless steel) and thus do not generally come into direct contact with the reactor coolant.
This is also the case for the reactor pressure vessel
(RPV) in boiling water reactors (BWRs), although
the RPV head is sometimes left unclad and the cladding has been removed from the blend radius of many
RPV feedwater nozzles. In BWRs of German and of
newer General Electric designs, extensive use is also
made of unclad LAS and CS in both the feedwater and
steam lines, as well as in the condensate system. The
primary coolant piping in conventional CANDUs is
made exclusively of unclad CS. In secondary coolant
systems, the steam generator pressure vessel shell is
unclad, as are the feedwater, drain, and steam lines.
CS & LAS pressure-boundary components, in particular in the primary circuit such as the RPV, are
very critical systems with regard to plant safety and
lifetime (extension). Minimizing corrosion improves
plant availability and economics and is also fundamental for safe operation over extended periods of
5060 years.

108

Typical CS & LAS piping and pressure vessel materials in Western LWRs (US designation, according to Section II of ASME BPV Code)


Designation

Type

Product
form

Cmax
(%)

Mn
(%)

Pmax
(%)

Smax
(%)

Simin
(%)

Cumax
(%)

Nimax
(%)

Crmax
(%)

Momax
(%)

Vmax
(%)

YS25 C
(MPa)

Heat
treatment

Microstructure

SA 106 Gr. B

CS
CMn

Pipe drawn

0.30

0.29
1.06

0.035a

0.035a

0.10

0.40a

0.40b

0.40b

0.15b

0.08b

 240
300400c

Normal.

Ferriticpearlitic

SA 333 Gr. 6

CS
CMn

Pipe drawn

0.30

0.29
1.06

0.025a

0.025a

0.10

 240
300400c

Normal.

Ferriticpearlitic

SA 516 Gr. 70

CS
CMn

Vessel plate

0.27d

0.79
1.30

0.03a

0.035a

0.13
0.45

 260
300400c

Normal.

Ferriticpearlitic

SA 533 B Cl.1

LAS
MnMoNi

(R)PV plates

0.25

1.07
1.62

0.12e
(0.35)

0.15e
(0.35)

0.13
0.45

0.10e

0.37
0.73

0.41
0.64

0.05

 345
450550c

Q&T

Bainitic

SA 508 Gr. 3 Cl. 1

LAS
MnMoNi

(R)PV
forging

0.25

1.20
1.50

0.12e
(0.25)

0.15e
(0.25)

0.15
0.40

0.10e

0.40
1.00

0.25

0.45
0.60

0.05

 345
450550c

Q&T

Bainitic

SA 508 Gr.2 Cl. 1

LAS
NiMoCr

(R)PV
forging

0.27

0.50
1.00

0.12e
(0.25)

0.15e
(0.25)

0.15
0.40

0.10e

0.50
1.00

0.25
0.45

0.55
0.70

0.05

 345
450550c

Q&T

Bainitic

In modern steels, these values are less than 0.015%.


Combination shall not exceed 1.0%.
c
Typical range.
d
Carbon varies with thickness up to 0.31%.
e
Requirement for core belt region.
YS yield strength; Normal. normalized; Q & T quenched and tempered.
b

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

Table 1

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

Consideration of both uniform and flowaccelerated corrosion (FAC) behavior for all unclad
surfaces is important for corrosion product transport
and deposition (e.g., crud formation on fuel elements)
but together with the assessment of resistance to
localized corrosion phenomena such as pitting and
environmentally assisted cracking (EAC) is obviously also required for integrity reasons. In the case of
EAC, however, safety considerations furthermore
require that possible defects extending through the
cladding be taken into account, so that the susceptibility of the RPV must be assessed as if no cladding were
present. Sometimes, thick pads of Alloy 182 have been
welded directly onto the RPV to act as attachment
points for internal structures; the higher yield strength
of Alloy 182, the thicker section and its known SCC
susceptibility raise special concerns for these areas.
In such cases, it is possible that SCC or thermal fatigue
of the austenitic alloy will occur such that the crack
tip propagates to the interface between the austenitic
and ferritic alloys. Furthermore, leakage of coolant
from the primary circuit in PWRs poses a special
hazard for CS & LAS components, since the boric
acid it contains can concentrate and lead to uniform
corrosion, or wastage, of external surfaces.
This chapter covers both the uniform and localized corrosion behavior of CS & LAS pressureboundary components in the primary (BWR, PWR,
and CANDU) and secondary (PWR and CANDU)
coolant systems of Western reactors, whereby the
discussion is focused on relevant US nuclear codes
and rules together with material standards in this
area. Special emphasis in Sections 5.06.2 and 5.06.3
is placed on FAC and on EAC, both of which have
resulted in serious pipe ruptures (FAC) or leaks (EAC)
during both nuclear and fossil service in the past.
In Section 5.06.2, the uniform and boric acid
corrosion behavior of CS & LAS, as well as the nature
of the protective oxide film on these materials, are
summarized first, followed by a condensed review of
the FAC behavior of these steels. The major factors
controlling FAC, the underlying mechanism and predictive models, as well as the relevant service experience and possible mitigation actions are discussed.
After a brief overview of pitting in CS & LAS in the
first part of Section 5.06.3, crack initiation susceptibility conditions and crack growth behavior are discussed in detail for the different types of EAC and
compared with the relevant design codes and crack
growth disposition curves for CS & LAS. This is
followed by a review of the mechanistic understanding of EAC and of existing EAC models. LWR service

109

experience and mitigation actions with regard to


EAC are then summarized and compared with this
experimental and theoretical background knowledge.
Finally, Section 5.06.4 summarizes the major conclusions of this review.

5.06.2 Uniform and FlowAccelerated Corrosion


5.06.2.1

Uniform Corrosion

Uniform or general corrosion does not normally


cause a problem for the structural integrity of CS
or LAS components in nuclear coolant systems.
Corrosion rates in typical circuits are generally of
the order of a micrometer per year (1 mm year1) or
less higher than those of stainless steel or nickelbased alloys, for example, but quite acceptable.
Around 300  C, uniform corrosion rates of CS &
LAS are minimal at a slightly alkaline pH300  C of
66.5 (neutral high-purity water has a pH300  C of
5.7) and intermediate dissolved oxygen levels. Under
some shutdown conditions, however, LWR primary
coolant can be aggressive to these materials, in particular in conjunction with increased oxygen levels
(e.g., through oxygen ingress from air); below
100  C, corrosion rates may be high. Compact,
defect-free oxide films grown at higher temperatures
during service are kinetically quite stable at lower
temperatures and usually provide sufficient protection
against uniform corrosion during short shutdown periods. Nevertheless, reactor vessels and LAS piping in
PWRs are clad with stainless steel, which helps reduce
the build-up of crud on fuel and of radiation fields by
ensuring a high degree of water purity with a low level
of dissolved iron.
A particular concern in PWRs arises from the
leakage of borated coolant from joints such as gasketed flanges and its impingement on components
such as flange studs. Up to 2001, some 140 leaks had
been reported publicly.2 Solid boric acid at room
temperature and dilute, deaerated boric acid solutions regardless of temperature have little effect
on CS & LAS, but as the boric acid concentrates, corrosion rates up to about 1 mm year1 may be reached.
Aerated solutions can be much more aggressive,
with the attack increasing with acid concentration.
Note that as hot coolant escapes to the environment,
its boric acid content (which may be nominally
2000 ppm (1 ppm 1 mg kg1; 1 ppb 1 mg kg1) or
more as elemental boron) concentrates by evaporation.
At temperatures in the neighborhood of 100  C, which

110

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

are attained by surfaces impacted by coolant flashing


to steam, corrosion rates can reach 250 mm year1.2
In some situations, flow effects can exacerbate the
attack, as described in Section 5.06.2.2.
The resistance of CS and LAS to corrosion is
dependent upon the protective properties of the
oxide film. Environments such as boric acid that
dissolve or erode the oxide then promote corrosion.
The predominant oxide on CS and LAS in coolant
circuits operating above about 130  C is magnetite
Fe3O4. In deoxygenated alkaline water, the magnetite
forms a double layer that has been well characterized
in terms of materials performance in boiler systems at
temperatures of about 300  C.3 This morphology is
found on CS in CANDU primary circuits, and would
be found on pressure-vessel steel exposed to PWR
primary coolant in the absence of high-alloy cladding.
The layers are formed by the simple oxidation of
the steel by water:
Fe 2H2 O FeOH2 H2

I

The nascent hydrogen is absorbed by the metal and


diffuses to the exterior. Roughly half of the ferrous
species (often as the dissolved hydroxide depending
on the pH) are precipitated oxidatively at the metaloxide interface as small crystallites of magnetite, each
a few tens of nanometers across, also releasing hydrogen to the coolant:
3FeOH2 Fe3 O4 2H2 O H2

II

The precise fraction precipitated is determined by


the density of the oxide relative to that of the metal,
since the inner layer occupies the volume of metal
corroded.3 The remainder of the dissolved iron diffuses through the oxide to the bulk coolant and precipitates according to eqn [II] as an outer layer of
magnetite crystals, each several micrometers across,
again releasing hydrogen to the coolant. If metal
species other than those of iron originate from alloy
components elsewhere in a circulating system, they
may coprecipitate and modify the locally formed
magnetite. An example of double-layer formation is
shown in Figure 1.
The concentration of dissolved iron in the coolant
governs the oxide formation. If the coolant is significantly undersaturated in iron, the outer layer cannot
precipitate and the inner layer may even dissolve at
the oxidecoolant interface. In nonisothermal systems,
temperature gradients create solubility differences and
transport iron around the circuit, modifying the oxide
films accordingly (the same phenomenon transports
different oxides around circuits containing other

Coolant flow
Precipitation
Dissolution
Outer oxide

o/s interface

Inner oxide

m/o interface
Corroding metal

Figure 1 Schematic of double layer oxide formation on


carbon steel in high-temperature water.

materials, such as the nickel-base alloys in PWRs).


Thick films may also spall and release oxide particles
to be distributed by the coolant. In circuits connected
to the reactor core, oxide transport may create deposits
on the fuel, impeding heat transfer and leading to
increased radiation fields around out-of-core components (note that the nickel-base alloys and stainless
steel in PWRs can produce deposits derived from
nickel ferrite, NiFe2O4; on high-burnup fuel undergoing subcooled boiling, these can harbor boron from
the coolant and provoke shifts in the neutron flux, as
well as affect radiation fields).
Evolved hydrogen also affects magnetite solubility
(by the one-third power of the concentration as
indicated by eqn [II]). Such increased solubility at the
metaloxide interface has been invoked as the reason
for the lack of precipitation within pores as iron diffuses to the oxidecoolant interface.4 Magnetite films
formed on steel surfaces that are pressure boundaries,
where the hydrogen evolved by eqn [I] continuously
effuses through the metal, tend to have a more adherent inner layer of larger crystallites than those formed
on totally immersed surfaces such as test coupons,
where the evolved hydrogen can only diffuse through
the oxide to the bulk coolant once the metal becomes
saturated.5
Under neutral oxidizing conditions, magnetite is
still the predominant base oxide formed on steels.6
However, since dissolved oxygen becomes the oxidant rather than water, hydrogen generation is suppressed and the basic oxidation reactions become:
2Fe 2H2 O O2 2FeOH2
6FeOH2 O2 2Fe3 O4 6H2 O

III
IV

The oxide layers especially the outer one then


tend to contain the more-oxidized forms maghemite
and/or hematite (both of formula Fe2O3), particularly in BWR circuits.7 Under reactor coolant conditions, corrosion rates and oxide solubilities under

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

oxidizing conditions are generally substantially lower


than those under reducing conditions. At high oxygen
levels, however, the risk for pitting and EAC increase
significantly (see Section 5.06.3).
The forms of oxide that are thermodynamically
stable under various conditions in coolant circuits are
indicated by Pourbaix diagrams, which plot the equilibrium potentials of the oxidizingreducing reactions
against pH; the higher the potential, the more oxidizing the environment. For dissolved species, the equilibria and therefore the lines in the diagram are
dependent upon the concentration; when illustrating
corrosion situations, a concentration of 106 M or less
is often assumed. It should be borne in mind, therefore,
that such diagrams are mainly indicative in nature and
illustrate the possibilities of species formation without
taking account of reaction kinetics. Figure 2, adapted
from Beverskog and Puigdomenech,8 is an example for
species pertinent to steel at 310  C, where a species
concentration of 108 M is representative. The hydrogen line in the figure represents the equilibrium:
2H2 O 2e 2OH H2
5.06.2.2

V

Flow-Accelerated Corrosion

5.06.2.2.1 Controlling factors

Flow-accelerated (or -assisted) corrosion (FAC), sometimes called erosioncorrosion (EC) in older literature,
0
0.2
Fe2O3

0.4

Fe(OH)4

0.6

Fe3O4

E(v)

Fe(OH)+

Fe(OH)2

0.8

Hydr

ogen

line

Fe(OH)3

1
Fe

1.2
1.4
6.5

7.5

8
pH310 C

8.5

9.5

Figure 2 Pourbaix diagram for iron at 108 m at 310  C.


Reproduced from Beverskog, B.; Puigdomenech, I. Corros.
Sci. 1996; 38(12): 21212135.

111

is essentially the dissolution and erosion of the normally


protective oxide film on CS (or LAS with a Cr-content
< 0.2 wt%), exacerbated by fluid flow effects, resulting in excessive corrosion rates and substantial pipe wall
thinning. Nowadays, the term EC implies the involvement of a significant mechanical component as an abrasive (e.g., by dispersed solid particles in the liquid phase)
or cavitation-induced (mechanical) removal of surface
material; it should therefore be differentiated from FAC,
which is primarily caused by a flow-induced increase in
the mass transfer of dissolving and reacting (corrosive)
species at high-flow or highly turbulent locations,
although fluid shear stress on the oxide film at the
material surface may also make substantial contributions to the damage in some situations.
FAC is a pervasive problem in most types of
steam-raising system and has caused feedwater line
ruptures, occasionally with fatal consequences, in
both fossil and nuclear plants.9,10 In primary coolant
systems also, less serious (though costly) FAC occurs
chronically in the CS outlet feeders of conventional
CANDUs,11 and flow effects are implicated in the
corrosion of PWR pressure-vessel steel by borated
coolant leaking through cracked penetrations in the
RPV head.12 FAC thus occurs in the regions of high
turbulence in both single and two-phase flows, but
never in systems with dry steam.
FAC depends on hydrodynamics (mainly steam
quality, flow rate, fluid shear stress at the wall,
turbulence intensity, and mass transfer coefficient),
environmental factors (mainly temperature, pH,
dissolved oxygen, hydrogen, and iron concentrations)
and material parameters (metal composition Mo,
Cu and, in particular, Cr content).9 The critical
parameter combinations for the occurrence of FAC
in feedwater systems and the main parameter effects
are schematically summarized in Figure 3.
The conditions leading to increased FAC rates
are usually related to regions with turbulent flow, to
low electrochemical corrosion potentials ECP (i.e.,
to chemically reducing conditions), and to low iron
concentrations in the water (Figures 3 and 4).
Depending on the pH, the maximum FAC rates
occur at about 130  C in single-phase flow, and at
about 180  C in two-phase flow (in the latter, it is the
condition in the liquid layer at the steel surface that
controls the FAC rate, but this is difficult to measure
or predict). Note that FAC can still be a problem
at other temperatures, even though rates are lower.
For example, feeder FAC in CANDU primary coolants occurs at 300310  C at the core outlet, and FAC
is also significant in feedwater systems at the low

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

Log FAC rate

112

pH 7

Hydrodynamics
Hydrodynamics
Hydrodynamics
pH 9

Shear stress at surface


Flow rate
Turbulent intensity
Mass transfer coefficient

Flow

Critical conditions
for high FAC risk
in feedwater

150 C

Material

Environment

[Cr] in metal < 0.2%

Log FAC rate

Log FAC rate

pH 7
Low [Fe]
pH < 9.2
120 C < T < 180 C
[O2] < 240 ppb

pH 7

0.2% Cr

pH 9

40 ppb

Log FAC rate

pH 7

12 ppb
pH 9

Log FAC rate

pH 9
[Cr] in metal

130 C

pH 9

pH
[O2]

Figure 3 Critical parameter combinations for flow-accelerated corrosion (derived from Uchida, S. et al. In: Proceedings
of the 13th International Conference on Environmental Degradation of Materials in Nuclear Power Systems, CD-ROM.
Whistler, British Columbia, Canada, 1923 August, King, P., Allen, T., Busby, J., Eds.; Toronto, ON: The Canadian
Nuclear Society, 2007) and major parameter effects on flow-accelerated corrosion under feedwater conditions.

temperature of condensate extraction. Specific geometries like elbows, bends, protruding weld roots,
orifices, and valves cause local turbulence, which
significantly increases FAC rates at, or immediately
downstream of, the location concerned. Systems such
as the moisture-separator/reheater drain lines, where
steam has condensed and relatively iron-free water is
flowing, are particularly susceptible. In primary coolant systems, there is the desire to keep iron concentrations low to prevent crud build-up and radiation
transport problems, hence the frequent use of highalloy materials that are resistant to FAC as cladding.
However, it must be recognized that a recirculating
system will always tend toward equilibrium; in other
words, dissolved iron concentrations on average will
vary around solubility values, depending upon oxide
dissolution and precipitation kinetics, temperature
gradients around the circuit, and the capacity of
sinks such as the purification circuit.
Most studies of FAC have been performed under
feedwater conditions, which generate high rates of
attack that can reach several millimeters per year in
some situations. Neutral chemistry, low-oxygen conditions at about 140  C, as may be found in BWR

feedtrains, can give high FAC rates, so dual-cycle


PWRs or PHWRs routinely add an amine such as
ammonia to raise the pH in the secondary coolant
circuit. The actual pH employed depends upon the
materials of construction; for all-ferrous feedtrains,
a pH25  C from 9.3 to 9.6 is usually specified, but the
value is kept below 9.2 to avoid excessive corrosion of
copper-base alloys, if these are present. Also, to
achieve a more even distribution of additive around
the circuit, an amine (such as morpholine) with a
coefficient of distribution between the steam and
liquid phases closer to unity than that of ammonia
may be used.
Oxygen dissolved in the coolant is also a powerful
inhibitor of FAC; it has been added routinely to feedwater systems in BWRs and certain fossil boilers for
some time. Depending upon the rate of attack, levels
of oxygen from a few ppb to several tens of ppb are
sufficient to stifle FAC completely. Maintaining a
dissolved oxygen content >30 ppb, which raises
the corrosion potential ECP in the feedwater system above the Fe3O4/Fe2O3 phase boundary in the
Pourbaix diagram in Figure 2, is particularly crucial
in BWRs operating on hydrogen water chemistry

113

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

St37.2 (A414 Gr B-carbon steel)

Flow (kg h1)


983

491

15Mo3 (A161 Gr T10.5% Mo)

907
756
605

378
302
227

15NiCuMoNb5
13CrMo44 (A213 Gr T121% Cr, 0.5% Mo)
10CrMo910 (A213 Gr T222.2% Cr, 1% Mo)
3000

4.0
pH = 9.04

pH = 7.0

2.0

Increasing flow

Loss rate (mm year1)

3.0

1.0

Specific material wear rate (g cm2 h1)

1000
300
100
30
10
3.0
1.0
0.3
0
90
(a)

100 110 120 130 140 150 160 170


Temperature (C)

0.1
50
(b)

100

150

200

250

300

Temperature (C)

Figure 4 Effect of temperature, flow rate (a) (data from Bignold, G. J. et al. In: Proceedings of the International Specialists
Meeting on Erosion-Corrosion of Steels in High-Temperature Water and Wet Steam, Les Renardieres, France, 1112 May;
EDF: France, 1982) and material (b) (data from Heitmann, H. G.; Schub, P. In: Proceedings of the Third Meeting on Water
Chemistry of Nuclear Reactors, pp. 243252, Bournemouth, UK, October; British Nuclear Engineering Society (BNES):
London, UK, 1983) on single-phase flow-accelerated corrosion under different flow and chemistry conditions. Reproduced
from Dooley, R. B. Power Plant Chem 2008, 10(2), 6889.

(HWC) with high rates of hydrogen injection into the


feedwater. If HWC is combined with noble metal
chemical addition (NMCA), the FAC risk is reduced,
since much lower hydrogen injection rates are then
adequate to mitigate SCC in stainless steel recirculation piping and reactor internals. (Recombination of
hydrogen and oxygen to lower the ECP requires the
radiation fields present in the RPV.) Oxygen levels
significantly above 50 ppb may increase the risk
of strain-induced corrosion cracking and corrosion
fatigue in CS & LAS feedwater piping (see Section
5.06.3). Furthermore, the deliberate addition of oxygen
to feedwater systems in dual-cycle reactors may
pose problems, since residual oxygen entering the
steam generators can provoke SCC of the high-alloy
steam-generator tubes. Nevertheless, severe FAC of

components in the feed train of advanced gas-cooled


reactors (AGRs) has been successfully mitigated since
the early 1980s by oxygen additions.13
Material properties have a significant impact
on FAC rates, but typically the plant operator has
no control over this (unless a replacement of piping
is an option). Certain elements in the steel can act
to retard FAC, as mentioned earlier; for example,
chromium is particularly effective and a concentration of 0.1% in the metal reduces FAC in 180  C
ammoniated water and watersteam mixtures at
pH25  C 9 by about 70%.14 Moreover, under CANDU
primary coolant conditions of 310  C and pH25  C 10.5
(adjusted with lithium), increasing the chromium content of SA-106 Grade B CS from 0.019% to 0.33%
reduces FAC by about 50%.11

114

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

5.06.2.2.2 Mechanisms and models

As with uniform corrosion (discussed in Section


5.06.2.1), FAC is governed by the ability of the
oxide film to protect the metal. Magnetite forms on
the steel at the metaloxide interface and is degraded
at the oxidecoolant interface by fluid flow effects
and by dissolution according to the general equation
[VI] (which indicates the dependence of the dissolved
species on pH under reducing conditions and which is
equivalent to eqn [II] for b 2). The turbulence
in the coolant and the solubility of the magnetite
are then paramount in determining the severity of
the attack.
Fe3 O4 32  bH H2 3FeOHb 2b
4  3bH2 O
VI
with b 0, 1, 2, or 3.
Mass transfer is often assumed to control the
mechanism.15 This derives from the postulate that
the magnetite film attains a steady-state thickness as
it dissolves at the rate Rd at its outer surface in
coolant undersaturated in dissolved iron and forms
continuously at the metaloxide interface at the
same rate Rg. Since the magnetite formation at
the metaloxide interface accounts for only about
half of the corroded metal, the other half diffuses
through the magnetite to the oxidecoolant interface, and with the iron from the magnetite dissolution is transported to the bulk coolant at the rate Rm.
The FAC rate RFAC is thus twice the dissolution rate
Rd of the magnetite at the oxidecoolant interface.
This concept of two processes in series dissolution
Rd and mass transfer Rm leads to the equation for
the steady-state FAC rate RFAC dm/dt Rm 2Rd
with all the variables in equivalent units of iron per
unit surface and time.
1. Steady-state assumption for the serial process:
Rg Rd 0:5Rm

1

2. Dissolution rate of magnetite at the oxidecoolant


interface according to eqn [VI] (assuming firstorder kinetics):
Rd 0:5  dm=dt kd Ceq  C

2

where kd is the dissolution reaction rate constant, C is


the concentration of Fe(II) species at the oxidecoolant
interface, and Ceq is their equilibrium concentration
according to eqn [VI], which corresponds to their
maximum solubility in the coolant.

3. Transport of Fe(II) species from the oxidecoolant


interface to the bulk environment by turbulent
mass transfer:
Rm h  C  Cb

3

where Cb is the concentration of Fe(II) species in the


bulk coolant and h is the mass transfer coefficient,
which is dependent on flow conditions and geometry.
From eqns [1][3] it follows that:
RFAC

h  kd  DC
0:5  h kd

4

where DC Ceq  Cb is the undersaturation in


iron. Models of FAC are based on the principles
behind eqn [4]. We expect that kd strongly increases
with temperature according to an Arrhenius law
for a thermally activated process (although there are
no data to confirm this over the temperature ranges
of interest), whereas h only shows a moderate
increase through the temperature dependence of the
properties in eqn [6]. If mass transfer controls, h is
small compared with kd (h kd) and the equation
reverts to:
RFAC hDC

5

For a coolant of constant conditions containing little or


no dissolved iron (i.e., Cb  0), the driving force DC
approaches a constant value the solubility of the
oxide, Ceq and RFAC varies as the mass transfer
coefficient (which increases with increasing flow rate
and turbulence). The mass-transfer model then
implies that the effects of materials composition and
coolant chemistry on FAC rate are brought about by
their effects on oxide solubility (Figure 5). According
to eqn [VI], the saturation concentration or solubility
Ceq depends on temperature, pH and H2 concentration
by simple chemical equilibrium thermodynamics.
Accordingly, the effect of chromium in the steel
can be attributed to the relative stability of mixed
oxides containing chromium (iron chromite, FeCr2O4,
for example, is virtually insoluble in reducing coolant
and accounts for the protection afforded by stainless
steel and similar alloys). As corrosion proceeds and
the magnetite dissolves, chromium is not leached out
in concert but continually concentrates in the film. It
is interesting to note that the inhibition occurs immediately at the start of exposure and continues at about
the same level, suggesting that the mechanism is the
rapid formation at the metaloxide interface of a more
protective layer of oxide that is maintained throughout exposure.16 It appears that the higher the chromium content of the steel, the more protective that

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

100
NH3
(mg kg1) pH
0.1
80

Fe (g kg1)

60

8.75

0.2

8.90

0.3

9.00

0.5

9.20

1.0

9.40

2.0

9.60

300

350

40

20

0
0

50

100

150

200

250

Temperature ( C)

Figure 5 Solubility of magnetite/iron as a function of


temperature at various ammonia concentrations.
Reproduced from Dooley, R. B. Power Plant Chem 2008,
10(2), 6889.

oxide layer; hence, the immunity of stainless steel and


certain LAS like 13CrMo44 or 10CrMo910. These
observations cannot be fully explained with the DC
term in the simple mass-transfer model of eqn [5],
which would indicate an increasing inhibition with
time as the chromium concentrates in the film. One
mode of inhibition of FAC by additives in the coolant
may be via incorporation in the magnetite to make it
less soluble. From the experience with chromium,
whatever species is added should be available to affect
the metaloxide interface consistently, presumably by
being kept permanently in solution in the coolant.
So far, titanium has shown promise as a coolantborne inhibitor of FAC under CANDU primary coolant conditions. Its effectiveness has been attributed
to its ability to form a mixed oxide with iron
ulvospinel along with the magnetite that forms
on corroding CS.17 An in-plant demonstration of
titanium addition to a CANDU primary system is
described in Section 5.06.2.3.
The simple mass-transfer model also indicates
that temperature should affect FAC partly through
its influence on magnetite solubility. In ammoniated

115

water at pH25  C 9.0, there is a strong temperature


dependence and the maximum FAC rate occurs
between 130 and 140  C, depending on flow rate.18
The solubility of magnetite under the same conditions increases from the range 515 ppb at 25  C to
a maximum of about 30 ppb that persists over the
range 110150  C,9,19 while mass transfer coefficients
at the same mass flow should approximately double
between 25 and 140  C. Similarly, in neutral water, the
maximum attack for several materials occurs at about
150  C,20 while the magnetite solubility increases from
about 70 ppb at 25  C to a maximum of about 140 ppb
at 120130  C. The rough correspondence between
the temperatures of maximum FAC rate and of maximum magnetite solubility, as well as the effect of
temperature itself on solubility, indicate the strong
influence of oxide film dissolution on the FAC mechanism. It is likely that at low temperatures dissolution
rates of magnetite are low enough for kd to have an
effect through eqn [4] and lower the flow dependence
accordingly.18
The inhibiting effect of amines and high pH at
feedwater temperatures should also be realized
mainly through the solubility of magnetite. Thus, in
neutral water at 140  C, the solubility of magnetite is
about 119 ppb, but if the pH25  C is raised with ammonia to 9.2, the value falls to the range 1426 ppb.9,19
This would suggest that a reduction in FAC rate by a
factor of 8.54.5 might be expected from ammoniating the coolant to pH25  C 9.2; however, experiments
indicate a reduction by a factor of only about 2.16
It is also instructive to consider the mass transfer
implications of the model according to eqn [5]. Mass
transfer in pipe flow in aqueous systems can be
described via a correlation of the mass transfer coefficient h with dimensionless numbers:
h Sh

D D
ARe p Sc q
d
d

6

in which the Sherwood number, Sh (given by hd/D,


where d is the pipe diameter and D is the diffusivity),
characterizes mass transfer in terms of the flow (via the
Reynolds number, Re, given by dur/m, where u is the
coolant velocity, r is the density, and m is the viscosity)
and physical properties (via the Schmidt number, Sc,
given by m/(rD)); A, p, and q are constants.
Typically, experiments on mass transfer of dissolved species yield values between about 0.6 and
0.9 for the exponent p.21,22 Recent experiments in a
water loop on FAC under neutral conditions at
140  C, however, indicated that the FAC rate RFAC
correlated rather weakly with Re.1.2,23 An alternative

116

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

mass transfer analysis gave an excellent correlation


with fluid shear stress at the pipe wall, t:
Ru Pt

7

FAC rate Flow rate (mm year1m s1)

where P is a constant (see Figure 6). Thus, a steel


containing 0.019% chromium gave the correlation
RFACu 0.07t, while a steel containing 0.001% chromium in parallel experiments gave RFACu 0.18t,
where RFAC is in units of millimeters per year, u is
in meters per second, and t is in pascals.16
The predominance of mass transfer in developing
such correlations depends upon the dissolution rate
constant, kd in eqn [4], being large enough to make
the mass transfer coefficient, h, controlling. This
would seem to be valid under neutral chemistry conditions, where the solubility of magnetite is high, but
under high-pH conditions, where the solubility is
reduced, kd may be reduced also and its influence
may become significant. However, although recent
80

indications24 are that FAC in 140  C ammoniated


water at pH 9.2 is not correlated well by the simple
mass-transfer model leading to eqn [6], those experiments also indicated a greater dependence on flow
rate or shear stress, viz., t raised to the power 1.52.0.
This cannot be attributed to an increasing influence
of kd in eqn [4]; apparently, a different mechanism is
involved.
Surface texturing usually accompanies FAC. In
steamwater mixtures, tiger-striping is caused by
the streaming pattern of the liquid film on the surface, while in single-phase water, scalloping sculpts
the attacked surface with grooves, flutes, or shallow
depressions (Figure 7(a) and 7(b)). However, in
experiments in neutral water at 140  C, in which corrosion rates of several millimeters per year were seen
in tubular test sections, a low-chromium steel developed no scallops, even though it corroded at more than
twice the rate of a higher chromium steel that developed distinct scallop patterns.23 The scalloping that
was seen was approximately related to the pipe flow
via a characteristic scallop Reynolds number:

70

Resc 1:55
104 2:6
103

60
50
40
30
20
10
0
0

200

400

600

800

1000

1200

Shear stress (N m2)

Figure 6 Correlation for flow-accelerated corrosion at


140  C in neutral water: carbon steel with 0.019% Cr.

in which the characteristic dimension is the average


scallop spacing. While the scallops were formed by the
corrosion of the metal, it was significant that distinct
oxide forms developed and were related more to scallop crests than to valleys. Those forms, shown in
Figure 7(c), occurred over pearlite grains in the
metal and may be described as coral-like. They provide further confirmation of the importance of oxide
dissolution in the mechanism, since they are no doubt
formed by the different solubilities of the different
compositions of oxide overlaying the lamellae of
cementite and ferrite in the pearlite. As the magnetite

200 m

(a)

8

3 m

(b)

Diameter of piping

(c)

Figure 7 Surface texturing in flow-accelerated corrosion. (a) Scalloping or horseshoe pits in single-phase flow. (b) Tiger
striping in two-phase flow. (c) Coral-like oxide on carbon steel undergoing flow-accelerated corrosion in neutral water at
140  C. (a) and (b) courtesy of COMSY with permission of AREVA NP.

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

generally dissolves, that over the cementite lamellae is


less soluble and is left standing proud. It was noted in
the experiments that the coral oxide concentrated
about 50% more chromium than the surrounding
oxide on the ferrite grains, possibly because the underlying pearlite contained more chromium as carbide
associated with the cementite.
Loop studies using tubular test sections of the
material of interest16 under reactor feedwater conditions establish the basis for adding oxygen with minimal residual concentrations left at the end of piping
systems. At 140  C in neutral water, about 40 ppb of
dissolved oxygen are required to stifle FAC, whereas at
pH25  C 9.2 with ammonia, only 12 ppb are required.
The concentrations required for stifling are related to
the measured FAC rates and it is clear that as oxygen
is progressively added to the corroding system the
cathodic reaction of water being reduced to hydrogen
is replaced by oxygen reduction; at the stifling concentration, the oxygen sink disappears and with continuing
addition its concentration in the loop jumps sharply.
However, although there is an obvious relationship
between the FAC rate at stifling and the stoichiometric
flux of oxygen by mass transfer to the surface, a
straightforward linear correspondence may not apply.13
While several mechanistic models of FAC in feedwater systems based mostly on the principles behind
eqn [4] have been developed, empirical models have
been applied extensively for some time. In the 1980s, for
example, parametric studies at the laboratories of the
then Siemens-KWU led to the formulation of a correlation between pipe wall thinning Dd and the system
variables u (flow velocity), T (temperature), pH, O2
(oxygen concentration), M (materials composition
Cr, Mo, and Cu), and t (exposure time):
Dd kc f u; T ; pH; O2 ; M; t 

9

where kc is a geometry factor. The correlation


was developed initially from data for single-phase
water flow, but was adapted to two-phase steam-water
flows, with the bulk velocity u substituted by the mean
velocity of the annular film of water covering the pipe
wall. The resulting computer code, WATHEC, was
restricted to steels with the content of Cr plus Mo less
than 5% and exposure times greater than 200 h. The
predictions of wall thinning for a large number of
situations were equal to or greater than the measured
values in 85% of the cases in other words, the code
was considered to be suitably conservative.25 Later, the
data management tool DASY was added to the code.
The EPRI-sponsored computer code CHECWORKS combined an empirical equation, which

117

had some basis in mechanisms such as that leading to


eqn [4], with a comprehensive data management
scheme.26 The data management includes analysis
of ultrasonic test data, calculation of critical wall
thickness for components at risk, and organization
of pertinent databases. The FAC rate RFAC is written
as a function of the system variables:
RFAC f T ; AC; MT; O2 ; pH; G; a

10

where T is temperature; AC is alloy content of Cr, Mo,


and Cu; MT is mass transfer; O2 is concentration of
dissolved oxygen; G is a geometry factor; and a is the
steam void fraction. The factors in eqn [10] are interrelated and the equation is nonlinear. While the absolute
predictions of RFAC in CHECWORKS are not generally of high precision, iterations incorporating plant
measurements can identify the locations of risk and
can rank components in the order of vulnerability.27
The FAC of CS is most pronounced under feedwater conditions, but it also occurs at higher temperatures in the primary coolant systems of PHWRs.
The phenomenon was identified in the late 1990s at
the Point Lepreau CANDU in New Brunswick,
Canada, where surfaces of affected outlet feeders of
CS were scalloped and the wall thinning rates plotted
against coolant velocity indicated a dependence on
the velocity raised to the power 1.5.28 Regions of
high turbulence, such as the tight-radius bends close
to the reactor face, were more severely affected.
It was also noted that the coolant at the core outlet
was unsaturated in dissolved iron, since it entered the
core at 265  C saturated after its passage through the
steam generators of nickel alloy and the inlet feeders
of CS; as its temperature rose in the fuel channels
the solubility at the high pH rose in concert (the
CANDU core contains no iron-bearing alloys, so it
cannot act as a source of dissolved iron).
Although the high-turbulence (and therefore highmass-transfer) regions are again the most affected in
primary coolant FAC, it is unlikely that the mechanism
in primary coolant is straightforward mass-transfer
control based on eqn [11]. First of all, the velocity
dependence is too high (the power of 1.5 rather than
0.60.9 as expected from correlations such as eqn [6]).
Second, measurements of the dissolution rate of magnetite under chemistry conditions close to those of
CANDU coolant29 have given a value of kd in eqn [4]
very much lower than the mass transfer coefficient h,
which would put the mechanism squarely under dissolution control with no velocity effect at all. The
alternative theory proposed is that dissolution of magnetite works in synergy with fluid shear stress at the

118

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

surface to degrade the oxide. Thus, the loosening of


the magnetite crystallites in the film makes them susceptible to removal by the fluid forces and as they are
eroded away the film becomes less protective. A mathematical model developed on this principle was able to
predict quite well the thinning of the walls of outlet
feeders at an operating plant in terms of the development of the oxide film, the pattern of attack around
representative bends, and the corrosion potential ECP
and velocity-dependence of FAC rate in individual
feeders.30 The model was adapted for predicting corrosion under conditions when the coolant is saturated in
dissolved iron and gave reasonable predictions of oxide
film growth and general corrosion in CANDU inlet
feeders, where corrosion rates are quite low.
It is probably more than a coincidence that FAC
under these primary coolant conditions, when magnetite solubility is low, seems not to be controlled directly
by mass transfer, while similar indications apply under
feedwater conditions at high pH, when solubilities
are also low. The parallel between the two situations
could be clarified if measurements of kd under feedwater conditions were available and the measurements
under primary coolant conditions were verified.
The high rate of general corrosion of CS caused
by aerated concentrated solutions of boric acid originating from leaking PWR coolant was described in
Section 5.06.2.1. Some of the studies that quantified
the attack were done with dynamic systems, such as
evaporating sprays, and it became clear that flow has
an effect.2 Of immediate concern is the corrosion of
RPV steel caused by borated coolant leaking through
cracked penetrations housing control rod drive
mechanisms. At the Davis Besse PWR in 2002, such
corrosion had threatened the integrity of the vessel.
The sequence of events that can lead to cavity formation next to a nozzle was postulated12 to be in three
phases: initially, slow seepage of coolant into the external annulus (crevice) in the head would be accompanied by low corrosion rates; next, when the crevice had
opened enough and the crack had lengthened to give
substantial leak rates, an evaporating coolant jet would
accelerate the attack through flow effects; finally, leakage into a cavity would create a turbulent evaporating
pool, extending the attack sideways.
An extensive testing program sponsored by the
Electric Power Research Institute (EPRI), Palo Alto,
California, investigated the phases of boric acid
attack at Davis Besse. The second phase, which experienced substantial flow effects, was simulated with
laboratory experiments in which a flashing jet of
borated coolant was directed onto a heated sample

of pressure-vessel steel and the damage assessed


in terms of system parameters notably, coolant
chemistry and flow rate.31 Volumetric (or massive)
metal loss was correlated with volumetric coolant
flow and seemed to behave differently from metal
penetration, which was correlated with jet velocity.
FAC was in evidence through miniature scallops in
the damage craters that formed around (but some
distance away from) the points of jet impact. Metal
loss rates attained about 3 cm3 year1 at a flow rate of
200 ml min1 with a boric acid concentration equivalent to 1500 ppm [B] and pH300  C of 6.9 adjusted with
lithium; the rate depended on the volumetric flow
in the jet raised to the power 0.84. Under the
same chemistry conditions, the penetration rate
reached 200 mm year1 at a jet velocity of 140 m s1
and the two were correlated via the velocity raised to
the power 4.3. It was notable that neither pH300  C nor
the boron concentration was the controlling chemistry parameter; rather, it was the ratio [B]/[Li].
5.06.2.2.3 Service experience and mitigating
actions

Many incidences of feedwater pipe thinning by FAC


from two-phase coolant were reported in the 1980s.
In 1985 March, a line downstream of a level control
valve for a feedwater heater at the Haddam Neck
PWR actually ruptured because of FAC induced by
flow-impingement. However, the first major incident
in a nuclear plant was the catastrophic pipe break at
the Surry Unit 2 PWR in December 1986, which led
to five deaths and several injuries. The 0.46 m diameter line thinned and ruptured at an elbow, 0.3 m from
a 0.6 m header, as a section of the pipe wall 0.6
1.2 m
was blown out. Until then, FAC by steamwater mixtures had been considered to be more serious than
FAC by single-phase coolant. Six months later, excessive thinning of a feedwater line was reported at the
Trojan plant and, in September 1988, Surry Unit 2
reported 20% wall loss in the suction line to a feedwater pump over a 1.2-year period.
Reports of serious thinning of feedwater piping
continued after the Surry incident, even though
plant inspections had generally become more rigorous
and chemistry control had tightened. In May 1990,
the Loviisa Unit 1 WWER (Eastern type PWR) in
Finland suffered a break in a 0.3 m diameter line in
the turbine hall, releasing about 50 m3 of steam and
water into the building, and in February 1993, a
similar incident occurred in Unit 2.
The latest major FAC incident in a nuclear plant
was the rupture of a feedwater line at the Mihama

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

Unit 3 PWR in 2004, which led to four deaths and


seven serious injuries.32 The thinning of the pipe wall
from 10 to 0.6 mm by FAC caused a large section to
peel back after rupture, allowing the coolant at

Figure 8 The ruptured feedwater line at Mihama-3.

140  C to flash to steam on release. The damaged


pipe is shown in Figure 8. Maximum thinning rates
were observed just downstream of an orifice plate,
where turbulence intensity was high. It should be
noted that chemistry had been maintained at high
pH with ethanolamine and that hydrazine was used to
scavenge oxygen. However, the location had not been
inspected since the plant start-up in 1976. The possibility of adding oxygen to the feedwater is being
considered and inspection procedures have been
revised extensively.
In 1997, at the Point Lepreau CANDU PHWR
in New Brunswick, Canada, the outlet feeder pipes
of CS that carry the heavy water coolant from the
core to the steam generators were found to be corroding excessively. The same problem has since then
been seen at all CANDUs in operation before 2000.
Figure 9 shows the arrangement of pipes at a reactor
face. The feeders are about 76 mm diameter and carry

119

2
5

8
10

1. Reactor outlet header


6. Calandria end shield face
2. Reactor inlet header
7. Tube spacers
3. Reactor outlet header
8. Support brackets
4. Reactor inlet header
9. Walkway
5. Feeder tube upper supports 10. End fittings

Figure 9 CANDU reactor face; cutaway view of feeder pipe arrangement, with permission from Atomic Energy of
Canada Limited.

120

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

coolant at velocities between about 12 and 22 m s1.


At the coolant temperature of 310  C and pH  10.6
(adjusted with lithium), the solubility of magnetite is
relatively low about 1.7 ppb and the FAC rates
accordingly attain only about 120 mm year1. The
attack is only a fraction of that observed under
feedwater conditions and has caused no safety issues,
but it means that feeder integrity may be lifelimiting and has necessitated replacements at some
reactors. Mitigating actions have been taken by
reducing the primary coolant alkalinity to the bottom of the recommended pH range, where the magnetite solubility is close to the minimum, while
plants in service since 2000 employ feeders with
a relatively high chromium content (0.3% in
contrast with the 0.02% in earlier reactors).
One trial of a coolant additive was made at the
Darlington Unit 3 900 MW CANDU in 2002,
when a titanium dioxide slurry was added to one
channel to give a concentration of ten or so microgram per kilogram at the outlet feeder33; significant
reductions in FAC rate were recorded, but it
was decided not to undertake the further development needed to proceed to the next stage of
full-plant addition. Meanwhile, feeder replacement
has become a manageable if costly operation at
severely affected plants.
The boric-acid corrosion at the Davis Besse PWR
in Ohio, which in 2002 was found to have a large
wastage cavity in the RPV head adjacent to a penetration housing a control rod drive mechanism, was
postulated to be partly due to FAC (see Figure 10).
About 2040 cm3 of the pressure-vessel steel had
corroded away and about 106 cm2 of stainless steel
cladding were exposed at the bottom. Substantial
quantities of solid boric acid had deposited close by.
Subsequent investigations12 determined that coolant
had leaked from a crack in the adjacent Alloy 600
nozzle into the surrounding annulus and in time
had widened it. The crack was an example of primary
water stress corrosion cracking (PWSCC), of which
numerous incidences have been recorded in PWRs.
No simple means of mitigation have been proposed
for existing plants, since the coolant boron level is
fixed by reactivity considerations. Long-term prevention entails avoiding operating with coolant
leaks (as is, in fact, the regulation in some jurisdictions), for example, through minimizing the possibility of PWSCC by using less-susceptible Alloy 690
material for the CRDM penetrations. In the meantime, more rigorous inspection regimes are being
implemented.

Figure 10 Cavity in the reactor pressure vessel head at the


Davis Besse pressurized water reactor.

5.06.3 Localized Corrosion and


Environmentally Assisted Cracking
5.06.3.1

Pitting

In CS & LAS, a shallow form of pitting can occur in


the complete absence of anionic water impurities as
the electrochemical corrosion potential (ECP) at the
steel surface is raised, for example, through oxygen
and other oxidizing species. Such corrosion pits often,
but not exclusively, initiate at MnS inclusions which
intersect the steel surface. Figure 11, originally compiled by Hickling, shows the critical boundaries
between uniform surface attack and shallow pitting
in high-purity water at low flow rates as a function of
temperature and dissolved oxygen level. The critical
oxygen concentration for pitting drops with decreasing temperature and is further reduced by a simultaneous mechanical straining of the surface, or by
increased sulfate and chloride impurity levels. Furthermore, pitting in CS & LAS is favored by high
steel sulfur contents and quasi-stagnant flow conditions. In the absence of impurities, increasing the flow
rate of water across the metal surface mitigates the
aforementioned form of pitting corrosion. Alkalization

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

shifts this boundary to much higher values, as does the


introduction of buffering and passivating species (e.g.,
on the secondary side of steam generators).3437
Some degree of pitting corrosion is inevitable
after long-term exposure of unclad CS & LAS surfaces to water in LWR systems and is not usually a
threat to either coolant purity or to structural integrity. Shallow pitting has been observed primarily in
specific piping systems with residual water because of

121

incomplete draining during nonoperational periods


(shutdown corrosion). This can be avoided by adequate wet or dry preservation techniques. If pitting
happens during normal plant operation at high temperatures, however, it indicates conditions under
which EAC may also occur (since this is controlled
by similar parameters, see Section 5.06.3.2) and can
even be directly implicated in the initiation of EAC
(Figure 12).

300

ck

ce

ta
at

ce

tin

fa

pit

ng

ur

tti

ls
ra
ne
Ge

Pi

ne

ra

ls

ur

fa

200
Ge

Temperature (C)

at

ta

ck

Pitting with
straining

Pitting
without
straining

100

SSRT

0
10

lzumiya, Tanno
Videm
Mizuno et al.
Ford

Coupon
specimens

1000

100

10 000

Dissolved oxygen content DO (ppb)


Figure 11 Boundaries between uniform corrosion and pitting attack in carbon and low-alloy steel in quasi-stagnant
high-temperature water. Compiled from The general and localized corrosion of carbon and low-alloy steels in
oxygenated high-temperature water. EPRI-NP-2853; EPRI: Palo Alto, CA, 1983; http://my.epri.com/; Electric Power
Research Institute. BWR environmental cracking margins for carbon steel piping. EPRI Report NP-2406; EPRI: Palo Alto, CA,
1982; http://my.epri.com/; Indig, M. et al. Rev Coat Corros 1982, 5, 173225; Videm, K. In Proceedings of the 7th
Scandinavian Corrosion Congress, Trondheim, Norway, May 2628, 1975; pp 444456.

(a)
20 mm

(b)

Acc.V Spot Magn


20.0 kV 5.9 1696x

20 mm
Det WD
SE 11.0 5/1;250C;HigN

Figure 12 Strain-induced corrosion cracks initiating from a corrosion pit or a (dissolved) MnS inclusion at the surface of a
low-alloy and carbon steel in high-temperature water in slow strain rate experiments. Adapted from Congleton, J. et al.
Corros. Sci. 1985, 25, 633650; Atkinson, J. D. et al. In Proceedings of the 13th International Conference on Environmental
Degradation of Materials in Nuclear Power Systems, Whistler, British Columbia, Canada, Aug 1923; King, P., Allen, T.,
Busby, J., Eds.; The Canadian Nuclear Society: Toronto, Canada, 2007; CD-ROM.

122

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

5.06.3.2 Environmentally Assisted


Cracking
5.06.3.2.1 Basic types of EAC and major
factors of influence

EAC is used here as a general term to cover the


full spectrum of corrosion cracking ranging from
stress corrosion cracking (SCC) under static load to
corrosion fatigue (CF) under cyclic loading conditions (Table 2).38,39 Strain-induced corrosion cracking (SICC) involving slow, dynamic straining with
localized plastic deformation of material, but where
obvious cyclic loading is either absent, or is restricted
to a limited number of infrequent events such plant
startup and shutdown, is increasingly used as an
appropriate term to describe the area of overlap
between SCC and CF.38,39
Under critical parameter combinations, EAC is
observed in all wrought and welded CS & LAS in
high-temperature water. The EAC crack path is
usually perpendicular to the direction of maximum
Table 2

tensile stress and transgranular in nature, with a


quasicleavage appearance showing a feathery morphology at high magnifications. The general fracture
appearance is similar for SCC, SICC, and even CF
(at least for strong environmental acceleration of
fatigue crack growth), thus confirming that EAC is
governed by the same basic process for all three
loading modes. In the case of cyclic loading at frequencies 103 Hz, the fracture surface also usually contains both ductile and brittle fatigue striations,
which are perpendicular to the local crack-growth
direction.38
EAC initiation and growth in CS & LAS are
governed by a complex interaction of environmental,
material, and loading parameters, and most influencing factors are both interrelated and synergistic. The
major parameters of influence, which have been
identified so far, are summarized in Table 3.38,39
The effect of these parameters on EAC initiation
and crack growth (including key thresholds) is

Basic types of EAC in CS & LAS and relevant nuclear codes

Mechanism

Environmentally assisted cracking (EAC)


CF
Corrosion fatigue

SICC
Strain-induced corrosion cracking

SCC
Stress corrosion cracking

Type of loading

Cyclic: low-cycle, high-cycle

Static

LWR operation
condition
Characterization of
crack growth
Characterization of
crack initiation

Thermal fatigue, thermal


stratification, . . .
ASME XI
Code Case N-643 (PWR)
ASME III
Fen-approach

Slow monotonically rising or very


low-cycle
Start-up/shut-down, thermal
stratification
High-sulfur line of F & A model as
upper bound
Susceptibility conditions:
ECPcrit, de/dtcrit, ecrit

Transient-free, steady-state
power operation
BWRVIP-60 disposition lines
scrit

Source: Seifert, H. P.; Ritter, S. Research and service experience with environmentally-assisted cracking of carbon & low-alloy steels in
high-temperature water. SKI-Report 2005:60; SKI: Stockholm, Sweden, 2005; ISSN 1104-1374. http://www.stralsakerhetsmyndigheten.
se/.
Hickling, J. et al. PowerPlant Chem. 2005, 7, 3142.

Table 3

Major influencing factors for EAC in C & LAS

Environmental parameters

Material parameters

Loading parameters

Corrosion potential, dissolved oxygen


content
Temperature

S-content, morphology, size, spatial distribution


and chemical composition of MnS

2

Cl, SO2
4 , S , HS

Susceptibility to dynamic strain ageing,


Concentration of interstitial C and N
Hardness/yield stress if >350HV5/800 MPa

Frequency, loading or strain


rate
Level of load, KI, stress,
strain, DK
Type of loading

Flow rate

Residual stress

Source: Seifert, H. P.; Ritter, S. Research and service experience with environmentally-assisted cracking of carbon & low-alloy steels in hightemperature water. SKI-Report 2005:60; SKI: Stockholm, Sweden, 2005; ISSN 1104-1374. http://www.stralsakerhetsmyndigheten.se/.
Hickling, J. et al. PowerPlant Chem. 2005, 7, 3142.

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

discussed in detail in Seifert and Ritter38 and an


interpretation of their synergism is given both there
and in Section 5.06.3.2.4.
5.06.3.2.2 Corrosion fatigue and straininduced corrosion cracking
5.06.3.2.2.1 Initiation and susceptibility
conditions

Slow strain rate (SSR)38,40 and low-cycle fatigue


(LCF) tests4042 with smooth specimens have clearly
shown that CF and SICC can occur in CS & LAS in
oxygenated, high-purity, high-temperature water if
the following conjoint threshold conditions are
simultaneously satisfied:
Water temperature >150  C. In LCF experiments, susceptibility then increases with temperature up to 320  C. SSR tests, on the other hand,
usually indicate a maximum of susceptibility
between 200 and 270  C, depending on strain rate.
Corrosion potential > ECPcrit 200 mVSHE or
dissolved oxygen content >30 ppb. Above this
threshold, EAC susceptibility then generally
increases with increasing ECP/oxygen content,
but saturates at very high levels.
Loading which leads to (local) macroscopic strains
at the water-wetted surface above the elastic limit.
The susceptibility then increases strongly with the
degree of plastic strain. SSR experiments with
tapered specimens, and LCF tests with different
waveforms, indicate a minimum critical strain of
0.150.2%, which is in a similar range to typical
oxide film rupture strains on CS & LAS (0.05
0.2%) in high-temperature water.
Positive strain rates below 103 s1. The EAC
susceptibility then increases with decreasing strain
rate de/dt. In most LCF investigations, saturation
of the decrease in fatigue life is observed below
a strain rate of 105 s1, but SSR tests indicate a
maximum of susceptibility between 105 and
107 s1, depending on ECP and temperature.
EAC susceptibility increases with increasing steel
sulfur content and a lower threshold is often quoted
at 0.003 wt%, but experimental evidence for the
latter is weak. Distinct material susceptibility
to dynamic strain aging (DSA) in the critical
temperature/strain-rate range, or a low yield
stress, may also favor crack initiation by SICC.
If one or other of these conjoint threshold conditions
is not satisfied, SICC initiation is extremely unlikely
and no, or only a minor, environmental reduction of

123

fatigue life is observed in high-temperature water.


Furthermore, in high-purity water, a high flow rate
may completely suppress SICC susceptibility and significantly retard CF crack initiation (in particular,
for small strain amplitudes or slow strain rates) compared to quasi-stagnant conditions, since the risk for
the formation of an aggressive, occluded water chemistry within small surface defects is significantly
reduced by convection. Note, however, that high levels
of chloride or sulfate may extend the range of susceptibility to less severe conditions (e.g., to lower ECP and
strain levels).
The range of system conditions where EAC crack
growth from incipient cracks may occur is significantly extended compared to the initiation susceptibility conditions specified earlier. For example,
CF crack growth has been observed in high-purity
PWR water at ECPs below 500 mVSHE under certain cyclic loading conditions (102 to 10 Hz).38
Apart from local stress raisers such as welding
defects, which may help overcome the strain threshold
in the field, the effect of initial surface condition (surface roughness, cold work, residual stress, oxide film,
and preoxidation) on SICC and CF initiation is much
less pronounced than for (high-cycle) fatigue in air, or
with SCC of stainless steels or Ni-base alloys. SICC
cracks usually, but not exclusively, initiate at MnS
inclusions or corrosion pits.38,40,43,44 Pitting, particularly if occurring actively, therefore facilitates SICC
initiation (Figure 12). CF cracks, on the other hand,
initiate mainly along slip bands, carbide particles, or at
the ferritepearlite phase boundary, and less frequently
at micropits or MnS inclusions.4042 The effect of
pitting and MnS inclusions on CF initiation is thus
moderate, but may become more pronounced in the
case of deep, high-aspect-ratio pits, mild environmental conditions, or at small strain amplitudes.40
5.06.3.2.2.2 SICC initiation and crack growth from
incipient cracks

In high-purity water in the absence of any significant


fatigue contribution, CS & LAS show distinct SICC
susceptibility only in highly oxidizing environments.
For example, it is almost impossible to initiate relevant
SICC crack growth in precracked specimens in slow
rising-load tests with constant load rate at KI values
<70 MPa m1/2 in high-purity water at an ECP of
<100 mVSHE. Even under highly oxidizing conditions (ECP  50 mVSHE), KI values of 25 MPa m1/2
have to be exceeded to initiate SICC in slow, risingload experiments in high-purity water. A maximum
in SICC initiation susceptibility (i.e., a minimum in

124

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

(a)

101

102

60

103

SA 533 B Cl. 1, 0.018% S


288 C, 8 ppm O2 65 ppbSO24

106

50

107

45
108

40
35
30
107

109
106

105

104

103

Crack opening displacement rate dCODLL/dt (mm s )

(b)

108

50

40

109

30
Weld filler material, 0.007% S
dCODLL/dt = 24 106 mm s1
20
100

102

8 ppm O2, 65 ppp SO24

150

200

250

300

SICC crack growth rate (ms1)

55

100

Stress intensity factor at onset of


SICC crack growth Kl,i (MPam1/2)

60

SICC crack growth rate (m s1)

Stress intensity factor at onset of


SICC crack growth Kl,i (MPa m1/2)

Stress intensity factor rate dKl/dt (MPa m h1)


101

350

Temperature (C)

Figure 13 Effect of loading rate (a) and temperature (b) on Strain-induced corrosion cracking initiation and crack growth
from incipient cracks in oxygenated high-temperature water in a low-alloy steel reactor pressure vessel steel. Reproduced
from Seifert, H. P.; Ritter, S. J. Nucl. Mater. 2008, 378, 312326.

the KI,i value needed to initiate SICC) occurs at intermediate temperatures around 200250  C and at slow
loading rates (Figure 13). Furthermore, the SICC
initiation susceptibility is affected by environmental
and material parameters in a very similar way as in
SSR tests with smooth specimens.38,45
Under highly oxidizing conditions (ECP 
50 mVSHE), SICC cracks can grow without any significant fatigue crack growth contribution and, in extreme
cases, SICC growth rates during plant transients with
severe thermal stratification can reach high values (up
to several hundreds of micrometers per event, or several millimeters per day). Under such circumstances,
SICC crack growth rates in both high-purity water and
water with increased sulfate or chloride levels are
comparable for all CS & LAS, independent of their
sulfur content. Once growth is initiated, crack velocities rise with increasing loading rates dKI/dt (and thus
crack-tip strain rates) and with increasing temperature
(Figure 13), at least up to 250  C.38,45
5.06.3.2.2.3 CF initiation and crack growth from
incipient cracks

Under cyclic loading, CS & LAS show a distinct


susceptibility to initiation of corrosion fatigue (CF)
from incipient cracks in all LWR environments.
However, environmental acceleration of fatigue
crack growth in these materials only occurs for
certain combinations of loading and environmental parameters. The cycle-based crack growth rate
da/dN then usually depends strongly on loading frequency and temperature, in contrast to fatigue in air,
and can achieve 10100 times higher values under
certain conditions (although the actual effects on

component integrity in the field are much less severe


than might appear at first glance from isothermal test
data). The combination of material, loading, and
environmental conditions, where strong environmental acceleration of fatigue crack growth (factor of
10) usually occurs, extends over a broad range in
the case of highly oxidizing BWR/NWC conditions,
but is restricted to a much narrower range at low
ECPs (Figures 14 and 15(a)), that is, in the case of
PWRs and of BWRs operating on HWC (with or
without NMCA (noble metal chemical addition)).
In contrast to SICC initiation, there is no threshold ECPcrit for CF in high-purity water, and material
parameters usually play a less pronounced role
(Figure 15(b)).46,47 The major parameters affecting
CF crack growth are shown in Table 4, and the
observed cracking behavior can be briefly summarized
as follows.38,47
Above an upper critical frequency ncrit,H of
1100 Hz and/or an upper DKCF,H threshold, environmental acceleration of fatigue crack growth disappears, because air fatigue crack growth rates are
higher than the maximum EAC rates for CS & LAS
in high-temperature water. These upper ncrit,H and
DKCF,H thresholds are shifted to lower values with
increasing DK and frequency, respectively, and do not
depend noticeably on environmental and material
parameters.
Strong environmental acceleration of fatigue
crack growth (factor of 10) in LWR environments
occurs with all CS & LAS below these upper thresholds for the combination of temperatures 100  C,
loading frequencies  ncrit,L f (ECP, DK, . . .), and
DK values  DKCF,L f (ECP, n, . . .). If one or more

125

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

102
0.4 or 8 ppm O2
+50 to + 200 mVSHE
0.2 ppm O2
0 mVSHE

ASME Xl air

104

103 102 101


Frequency n (Hz)

(a)

100

101

dK/dt , n

102

T = 288 C, R = 0.7 0.8, K = 14 22 MPa m1/2


SA 533 B Cl. 1 (0.018% S), 20 MnMoNi 5 5 (0.015% S)

105

101

we

< 0.005 ppm O2


500 mVSHE

100

Xl

ASME Xl Wet

101

102

101

ME

100

SA 533 B Cl. 1,0.0018% S


250 C, kout 0.07 S cm1
0.4 ppm O2

103

102
(b)

AS

101

da/dNCF (m cycle1)

da/dNCF (m cycle1)

102

ai

Xl

AS

Kl.max = 58 MPa m1/2


0.5 MPa m1/2h1
5 MPa m1/2h1
35 MPa m1/2h1
370 MPa m1/2h1

1
10
100
Stress intensity factor amplitude K (MPa m1/2)

(a)

106

SA 533 B Cl.1 (0.018 wt % S)


T = 274 288 C

108

1010
/d

t Air

1012

da

NWC, + 150 mVSHE


NWC, 100 mVSHE
HWC, 500 mVSHE

1012
1010
108
106
Fatigue crack growth rate in air da/dtAir (m s1)

Corrosion fatigue crack growth rate in


high-temperature water da/dtCF (m s1)

Corrosion fatigue crack growth rate


in high-temperature water da/dtCF (m s1)

Figure 14 Effect of electrochemical corrosion potential, loading frequency (a) and DK and loading rate (b) on corrosion
fatigue crack growth in low-alloy steel in high-temperature water and comparison with ASME XI Air and Wet, crack growth
rates. Reproduced from Seifert, H. P.; Ritter, S. Corros. Sci. 2008, 50, 18841899.

(b)

103
Superposition-model da/dtCF = da/dtENV + da/dtAir

105
107
109
1011
1013
1013

T = 240 288 C
0.4 8 ppm O2
<1 or 65 ppb SO42
n = 3E 6 to 8E 3 Hz
240 C
R = 0.20.8
288 C 250 C
K = 11 62 MPa m1/2
r
SA
533
B
Cl.1
(0.018% S)
i
t
20 MnMoNi 5 5 (0.004% S)
/d A
20 MnMoNi 5 5 (0.015% S)
da

1011

109

107

105

103

Fatigue crack growth rate in air da/dtAir (m s1)

Figure 15 Corrosion fatigue crack growth in the time-domain as a function of electrochemical corrosion potential (ECP) for
a high-sulfur low-alloy steel (a) and as a function of material and loading parameters at high ECPs >  50 mVSHE (b).
Reproduced from Seifert, H. P.; Ritter, S. Corros. Sci. 2008, 50, 18841899.

of these threshold conditions are not satisfied, the


environmental effect is moderate (factor of 5), or
even absent altogether. When these conjoint threshold
conditions are simultaneously satisfied, the environmental acceleration of fatigue crack growth increases
with increasing temperature (sometimes with a maximum around 250  C) and decreasing loading frequency and DK values, often showing maximum
enhancement close to the lower thresholds.
The lower thresholds ncrit,L and DKCF,L depend on
material, environmental, and loading parameters (in
particular, ECP and DK or frequency see Figure 14
and Table 4). They can reach very low values of
106 Hz and of 2 MPa m1/2 (for load ratio R!1)

or 5 MPa m1/2 (R! 0) under highly oxidizing conditions (ECP  100 mVSHE). Below these lower
thresholds, the environmental acceleration of fatigue
crack growths drops to moderate values (factor of 5)
and the cycle-based crack growth rate da/dN becomes
independent of frequency. The DK-dependence is the
same as in air and the effect of temperature is usually
moderate. Such behavior is sometimes designated
as true corrosion fatigue and corresponds to the
low-sulfur curve of the Ford and Andresen model
(see Section 5.06.3.2.4; Ford and Andresen48 and
Ford49). Once conditions exceed these lower thresholds, the material behavior corresponds to the highsulfur curve of the model.

126
Table 4

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels


Major parameter effects on CF crack growth with typical parameter values

Crack-tip environment

Low-sulfur, 0.1 ppm S2

High-sulfur, 15 ppm S2

Range
da/dtCF/dat/dtAir

n < ncrit,L
25
6 f(n)
m  3, that is, parallel to air curve
6f(n), that is, n 0
DKth, Air f(R)

ncrit,L n < ncrit,H  1 Hz


10100 (10 000)
n# ! da/dtCF/dat/dtAir"
m  12
n 0.50.6
n", R", ECP", SO2
4 ", S", DSA"! DKCF,L#
100 mVSHE: DKCF;L  25 MPa m1/2
500 mVSHE: DKCF,L  1030 MPa m1/2
ECP", SO2
4 ", S" DK" ! ncrit,L#
100 mVSHE: 1 E5 Hz to <1 E6 Hz
500 mVSHE: 1 E3 Hz to 1 E1 Hz
T > 100  C: T" ! da/dtCF"
EA  4050 kJ mol1
T# ! DKCF#
With minimum of DKCF at intermediate T?

da/dNCF BDKm
da/dNCF Ann
DKCF,L f(n, R, ECP, . . .)
ncrit,L f (DK, R, ECP, . . .)

No evidence for lower threshold for


low-sulfur behavior

Temperature

Moderate effect

Above 110 Hz, the environmental effects on crack growth disappear.


Source: Seifert, H. P.; Ritter, S. Research and service experience with environmentally-assisted cracking of carbon & low-alloy steels in
high-temperature water. SKI-Report 2005:60; SKI: Stockholm, Sweden, 2005; ISSN 1104-1374. http://www.stralsakerhetsmyndigheten.
se/; Seifert, H. P.; Ritter, S. Corros. Sci. 2008, 50, 18841899.

The water flow rate has little effect on CF crack


growth of long/deep cracks, in contrast to CF initiation and short-crack growth. In the latter cases, CF
effects can disappear entirely if flushing out of the
local crack-tip electrolyte occurs, but this is unrealistic for deep, semi-elliptical cracks.
5.06.3.2.2.4 Adequacy and conservatism of
fatigue design according to Section III of ASME
BPV Code in the context of environmental effects

Design against fatigue of CS & LAS primary pressureboundary components is often based on Section III of
the ASME BPV Code.50 It relies on the use of fatigue
curves and endurance limits, derived mainly from
strain-controlled LCF tests with small, smooth specimens in air at room temperature, which do not explicitly consider the possible effects of LWR environments.
The accumulated good service experience with
CS & LAS primary pressure-boundary components does not indicate any generic deficiencies in
the current fatigue design procedures arising from
lack of consideration of environmental effects (see
Section 5.06.3.2.5), even though laboratory investigations clearly reveal that the fatigue lives of CS &
LAS can be substantially reduced in LWR environments (Figure 16)38,41,42,51,52: Although the microstructures and cyclic-hardening behavior of CS &
LAS differ significantly, the effects of the environment on the fatigue life of all these steels are very
similar. The magnitude of the reduction in this

depends on temperature, strain rate, oxygen level in


the water (i.e., ECP), and sulfur content of the steel.
The decrease is significant only when four conditions
are satisfied simultaneously, that is, when the strain
amplitude, temperature, and dissolved oxygen content in the water are above certain threshold values
(0.15%, 150  C, 0.04 ppm) and the strain rate is
below a key threshold value (103 s1). Although
only a moderate decrease in life (by a factor of <2)
is observed if any one of the aforementioned threshold conditions is not satisfied, fatigue lives of CS &
LAS can be more than a factor of 20 lower in the
coolant environment than in air under certain environmental and loading conditions (Figure 16). Such
observations have thus raised some concern with
respect to the possibility of insufficient safety margins for the fatigue design of certain CS & LAS
pressure-boundary components.
Conservatism in the ASME Code fatigue evaluations may arise from (a) the fatigue evaluation procedures themselves and/or (b) the fatigue design
curves they use. Sources of conservatism in the procedures include the use of design transients that are
significantly more severe than those experienced in
service, conservative grouping of transients, and use
of simplified elasticplastic analyses that result in
higher stresses/strains. The design margins of 2 and
20 on stress and cycles, respectively, in the ASME III
design curve were intended to cover the effects of
some variables (e.g., surface finish, material

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

127

10.0

1.0

0.1
101
(a)

A533-Gr B low-alloy steel


288 C, ea 0.4%

150250 >250
0.050.2 >0.2
0.010.4 <0.01
0.006 0.006

Temp( C) :<150
DO (ppm) :0.05
Rate (%/s) :0.4
S (wt.%) :>0.006

Fatigue life (cycles)

Strain amplitude ea (%)

Carbon steel

Mean curve
RT air

ASME design curve


102

103

104

104

103

Air
Simulated PWR
0.7 ppm DO

102
105

105

106

Fatigue life (cycles)

(b)

104

103

102

101

100

Strain rate (%/s)

Figure 16 Comparison of fatigue initiation life of carbon steel in high-temperature water with ASME III mean and design
curves in air (a) and effect of strain rate and oxygen content on fatigue initiation life in low-alloy steel. Reproduced from US
NRC. Effect of LWR coolant environments on the fatigue life of reactor materials. NUREG/CR-6909; US NRC: Washington,
DC, 2007, http://www.nrc.gov/. Work performed by Argonne National Laboratory, managed and operated by UChicago
Argonne, LLC, for the US Department of Energy under Contract No. DE-AC02-06CH11357.

variability, load sequence, and size effects) that can


influence the fatigue life of components, but were not
actually investigated in the tests which provided the
original data for the curves. They were not intended to
cover the effects of LWR environments,38 but a recent,
detailed analysis41,42 revealed that the factors of 2 on
stress/strain and 20 on cycle number not only provide
appropriate margins for the intended factors but may
also contain excess conservatism that would partially
counteract reductions in fatigue life due to EAC.
Based on large research programs, methods have
been developed in the United States42 and in
Japan51,53 for incorporating the effects of LWR coolant environments into fatigue evaluations. Experimentally derived fatigue life correction factors Fen
(defined as the ratio of life in air at room temperature
to that in water at the service temperature) are used
to adjust component fatigue usage values for environmental effects. Such approaches have recently been
implemented in the Japanese JSME Code54 and the
new US NRC Regulatory Guide 1.207,55 which is to
apply to new plants.
The apparent discrepancy between isothermal
laboratory results (strong environmental effects) and
field experience (only a few CF incidents under very
specific circumstances, predominantly related to thermal transients) mainly arises from the large degree
of conservatism in the fatigue evaluation procedures
mentioned earlier. The significantly higher strains of
design transients may fully outweigh possible environmental effects for real plant transients. Furthermore,
one or more threshold conditions are often also not

satisfied for most transients. Even plant transients


with strong environmental effects (e.g., at slow strain
rates) are usually not very damaging with respect to
fatigue damage accumulation because of their small
strain amplitudes, together with the rather limited
number of cycles during the whole component lifetime. Finally, as discussed earlier, the turbulent flow
rate at most component surfaces significantly retards
corrosion fatigue crack initiation (in particular for
small strains and slow strain rates) compared to
the quasi-stagnant conditions used in most lab
investigations.38,41,42
5.06.3.2.2.5 Adequacy and conservatism of
fatigue flaw tolerance evaluations according to
Section XI of ASME BPV Code in the context of
environmental effects

Fatigue flaw tolerance evaluations in combination


with periodic in-service inspections are based on
Section XI of the ASME BPV Code. Article A-4300
in Appendix A of Section XI contains a set of reference fatigue crack growth rate da/dN curves for CS
& LAS in air (air curves) or in LWR coolant environments (wet curves).56 The current ASME XI wet
curves (Figure 17) are based on lab data obtained
prior to 1980. They depend explicitly on DK and load
ratio R, but not on other variables that are known to
be important in CF, such as loading frequency, ECP,
or steel sulfur content. The same curves are used for
different types of CS & LAS and for BWR/NWC,
BWR/HWC/NMCA, and PWR primary or secondary side conditions. System conditions (or thresholds)

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

100
101
102

101

trise = 14 s
R = 0.95
0.018% S

100
101
102

43 s
6 0
N 20
e
=
s
ca t ise
e r
od .65
C 0
=
R

HWC, 274/288 C

3
64 s
N- 14
e =
s
e
ca t ris
de 95
Co = 0.
R

rv

103

ME
AS X I W
et,
ME
R
XI
0
W
.65
et,
R

0.2
5

101

N-643, EAC-curve for S > 0.013 % S, R = 0.65, tR = 100 s


N-643, EAC-curve for S > 0.013 % S, R = 0.65, tR > 1 s
N-643, non-EAC curve, R = 0.65
ASME XI Wet, R = 0.65
ASME XI Air, R = 0.65

AS

da/dN (m per cycle)

102

da/dNCF (m per cycle)

128

ir

XI

AS

trise = 2002000 s
R = 0.30.7
0.0040.018% S

cu

AC

-E

on

N
103
104
1
10
100
1
10
100
(a)
Stress intensity factor amplitude K (MPa m1/2)
(b)
Stress intensity factor amplitude K (MPa m1/2)

Figure 17 ASME XI Air and Wet curves and Code Case N-643 for high and low-sulfur steels (a) and comparison of cyclic
corrosion fatigue crack growth rates under hydrogen water chemistry conditions for different loading conditions with the
corresponding curves (b). Reproduced from Seifert, H. P.; Ritter, S. Corros. Sci. 2008, 50, 18841899.

where environmental effects on fatigue crack growth


can be neglected or excluded are not defined in the
present ASME Section XI Code.38,47
A more specific Code Case N-64357 for fatigue
crack growth in CS & LAS exposed to PWR primary
environments has been developed since 2000 and
may be used as an alternative to the ASME XI wet
reference fatigue crack growth rate curves for this
specific environment (Figure 17). Depending on system conditions, the Code Case N-643 procedure
predicts either lower or higher crack growth rates
than the general Section XI approach. The main
advantage of this newer Code Case is that it contains
criteria for the onset/cessation of EAC and that it
better reflects the experimentally observed cracking
behavior in PWR environments, since it considers
steel sulfur and frequencyloading-rate effects to a
certain extent. However, this approach has not (yet)
found general acceptance, primarily because of the
difficulty of defining appropriate rise times for actual
plant transients and the complications involved in
including these in component analyses.
In a similar way to that discussed earlier for
fatigue life design, conservatism in fatigue flaw tolerance evaluations may arise from (a) the fatigue flaw
tolerance evaluation procedures themselves (e.g., by
the use of design transients) and/or (b) the fatigue
crack growth rate curves. As discussed in Seifert and
Ritter,38,47 the current ASME XI wet curves conservatively cover the CF crack growth rate laboratory data
under most combinations of loading, environmental,
and material parameters. Even under highly oxidizing
BWR/NWC conditions, they are only significantly
exceeded under some very specific (but plantrelevant) circumstances (Figures 14 and 17), which

have caused some isolated CF cracking incidents in


the past. The Section XI curves might therefore be
regarded as an adequate, general bounding approach
under most system conditions, but they do not
realistically describe and reflect the experimentally
observed CF crack growth behavior of CS & LAS in
oxygenated high-temperature water. The curves predict crack growth rates which are either significantly
too high (e.g., n 102 Hz and ECP < 200 mVSHE)
or too low (e.g., n 102 Hz and ECP > 0 mVSHE).
Furthermore, system conditions or thresholds (e.g.,
n > 1 Hz), where environmental effects on fatigue
crack growth can be neglected, or even excluded, are
not defined in ASME XI.
5.06.3.2.3 Stress corrosion cracking
5.06.3.2.3.1
conditions

Initiation and susceptibility

Table 5 shows an assessment scheme according to


Hickling,58 based on both laboratory and field experience, for SCC initiation susceptibility and crack
growth in CS & LAS at normal strength levels
under BWR/NWC conditions.
Initiation of propagating SCC cracks from
smooth, defect-free surfaces under static load in
high-purity water is only observed for the following
conjoint conditions: stresses at the water-wetted surface above the high-temperature yield strength,
quasi-stagnant flow conditions, and dissolved oxygen
contents 0.2 ppm. Furthermore, if complete exhaustion of low-temperature creep is allowed to occur
before the specimens are exposed to high-purity,
high-temperature water, no SCC is observed, thus
indicating nonclassical SCC behavior and confirming the importance of slow dynamic surface straining

Assessment scheme for SCC susceptibility of CS & LAS

O2 (ppm)

Operating medium: HT water or steam condensate with T > 170  C


Flow conditions

k (mS cm1)

Crack initiation by SCC?

Derivation

Crack growth by SCC?

Derivation

<0.2
<0.2
<0.2

Typical for reactor


Quasi-stagnant
Quasi-stagnant

Typical, that is, 0.1


0.2
Raised (e.g., by impurities)

1
1
2

Typical for reactor


Quasi-stagnant

Typical, that is, 0.2


0.2

0.20.4

Quasi-stagnant

Raised (e.g., by impurities)

0.4

Typical for reactor

Typical, that is, 0.2

0.4

Quasi-stagnant

<1 often (0.2)

0.4

Quasi-stagnant

Raised (e.g., by impurities)

No susceptibility
No susceptibility
Possibility cannot be excluded,
perhaps after incubation time
No susceptibility
Possibility cannot be excluded,
perhaps after incubation time
Possibility cannot be excluded,
perhaps after incubation time
Susceptibility is suppressed
through flow
Possibility cannot be excluded,
perhaps after incubation time
Possibility cannot be excluded,
perhaps after incubation time

1
2
3

0.20.4
0.20.4

No susceptibility
No susceptibility
For stress levels at the water-wetted surface in
the region of the HT yield pointa
No susceptibility
For stress levels at the water-wetted
surface > HT yield pointa
For stress levels at the water-wetted surface in
the region of the HT yield pointa
In general, no susceptibility (? at stress
levelsHT yield point)
For stress levels at the water-wetted surface 
HT yield pointa
For stress levels at the water-wetted in the
region of the HT yield pointa

1
2
2
1, 3
2
2

a
Possibility cannot be excluded, perhaps after long incubation time.
1 from experiments in more aggressive environments; 2 from appropriate autoclave experiments; 3 no direct experimental evidence.
Source: Hickling, J.; Reitzner, U. VGB Kraftwerkstech. 1992, 72, 359367.

1
2
2
2
2
2

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

Table 5

129

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

under such conditions. Initiation of small nonpropagating SCC surface microcracks (surface
length < 15
grain diameter) at (MnS) inclusions
or corrosion pits may occur under less severe conditions than specified earlier. Furthermore, the possibility of SCC at stresses around the high-temperature
yield strength range cannot be excluded for very unfavorable parameter combinations.38,40,58
Since the primary design stresses are limited to
values below the yield strength and turbulent flow conditions prevail at most component locations, SCC initiation appears to be extremely unlikely under stationary,
transient-free, LWR power operation conditions, but
growth of a preexisting crack could occur under certain
conditions, as discussed in the next section.
5.06.3.2.3.2 SCC crack growth

Stress corrosion crack growth rate


da/dtscc (m s1)

In contrast to SICC and CF crack growth (and to the


behavior of austenitic stainless steels and nickel-base
alloys), CS & LAS reveal very low susceptibility to
sustained SCC crack growth up to high KI levels
in LWR environments at operating temperatures.
Such high KI-levels would correspond to rather
deep cracks in pressure-boundary components. In
laboratory tests with relatively small specimens,
no sustained SCC crack growth (or only very
slow (<0.6 mm year1), often localized propagation)
is observed up to KI values of 5060 MPa m1/2 in a
chloride-free BWR/NWC environment and up to
80100 MPa m1/2 in PWR or BWR/HWC/NMCA
environments. Above these KI thresholds, SCC growth
rates increase rapidly within a small KI-interval of
1020 MPa m1/2 and reach a plateau at high growth
rates of up to several meters per year.38,5961

(a)

106
T = 288 C, 0.004 0.018 % S
O2 = 200600 ppb, < 1 165 ppb SO42
2

108

1010

O2 = 8 ppm, <165 ppb SO4

Down-pointing arrows indicate continuous decay of SCC


with subsequent crack arrest within 1000 h

BWRVIP-60 SCC DL 1

1012
40
60
80
100
20
Stress intensity factor Kl (MPa m1/2)

The very low SCC crack growth susceptibility


at KI values below 50 MPa m1/2 is also confirmed by
long-term exposure of precracked specimens in
actual reactors,61,62 as well as by the field experience
with both unclad LAS components and with cracks
in claddings or dissimilar metal welds, where the
crack-tip extends to LAS at the fusion boundary
(see Section 5.06.3.2.5).
On the other hand, critical conditions for water
chemistry (e.g., chloride content > 5 ppb high ECP
> 0 mVSHE), for loading (e.g., ripple loading), or for
the material itself (e.g., hardness > 350 HV5 or high
DSA susceptibility at intermediate temperatures) can
result in sustained and relatively fast (>10 mm
year1) SCC for KI-values well below 60 MPa m1/2
(Figures 18 and 19). Such conditions, identified in
laboratory experiments, generally appear atypical
for current, optimized BWR power operation practice
with modern, properly fabricated CS & LAS pressureboundary components. Changing to HWC (i.e., low
ECP) always results some few hours later in a significant reduction of laboratory SCC crack growth rates
(by more than one order of magnitude even under
otherwise critical conditions).38,60
5.06.3.2.3.3 Adequacy and conservatism of
BWRVIP-60 SCC disposition lines

With regard to structural integrity, the possible


occurrence of SCC is usually more critical than CF,
since fatigue was considered in the original component design, whereas even very slow SCC growth
rates might result in relevant crack advance over
long operational periods. Within the framework
of the BWRVIP program of the Electric Power
Stress corrosion crack growth rate
da/dtscc (m s1)

130

(b)

106

108

T = 288 C, Weld HAZ,


350 HV5 + 0.022% S
T = 288 C, 460 HV5
T = 288 C, KI KIJ

1010
BWRVIP-60 SCC DL 1

1012

180 C < T < 270 C + high DSA


Base metal
Weld filler
Weld HAZ

20
40
60
80
100
Stress intensity factor Kl (MPa m1/2)

Figure 18 (a) Experimental confirmation of the conservative character of the BWRVIP-60 stress corrosion cracking
disposition line 1 for stationary, transient-free boiling water reactor power operation at 274288  C. (b) Constant load test
results with stress corrosion cracking growth rates above the BWRVIP-60 stress corrosion cracking disposition lines 1.
Reproduced from Seifert, H. P.; Ritter, S. J. Nucl. Mater. 2008, 372, 114131.

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

131

8.0 ppm DO

109

0.4 ppm DO

108

0.2 ppm DO

line
r SCC
sulfu
h
g
Hi

line
CC
S
r
fu
sul
wo
L

100 ppb Cl
50 ppb Cl
20 ppb Cl
15 ppb Cl, 250 C
10 ppb Cl
5 ppb Cl

on curv
e

+200 mVSHE
8 ppm DO
0.02 wt % S

1010

Transiti

SCC crack growth rate da/dtSCC (m s1)

107

1011

BWRVIP-60 SCC DL 2 for water chemistry transients


BWRVIP-60 SCC DL 1 for stationary power operation

101210

20

(a)

30
40
50
60
70
80
Stress intensity factor KI (MPa m)

90

100

Accelerated SCC crack growth (>10 mm year1) at KI < 60 MPa m

T = 274/288 C
NWC

200

0
Continuous operation
allowed
Prompt
shut-down

EPRI action
Level limit 3

400

EPRI action
Level limit 1

200
HWC

Corrosion potential ECP (mVSHE)

No SCC (<0.6 mm year1) at KI < 60 MPa m

600
1
(b)

10
Chloride content (ppb)

100

Figure 19 (a) Effect of chloride content on stress corrosion cracking crack growth in low-alloy steel in simulated boiling
water reactor/normal water chemistry environment and comparison with the BWRVIP-60 stress corrosion cracking
disposition lines and the F & A model. (b) Synergistic effect of electrochemical corrosion potential and chloride content on the
onset of fast stress corrosion cracking in low-alloy steel under simulated boiling water reactor conditions and comparison
with the typical conditions during stationary boiling water reactor power operation and Action Levels 1 and 3 of the EPRI BWR
water chemistry guidelines. Reproduced from Seifert, H. P.; Ritter, S. J. Nucl. Mater. 2008, 372, 114131.

Research Institute (EPRI), SCC disposition lines for


flaw tolerance evaluations (Figure 18) were developed in 1999 based on lab investigations and service
experience61 and later accepted by the US NRC as an
interim position. The BWRVIP-60 SCC disposition
line 1 applies to crack growth in LAS under static
loading and transient-free, stationary BWR/NWC or
HWC power operation conditions. On the other
hand, disposition line 2 (low-sulfur SCC line of
the Ford and Andresen model63) may be used for

estimating SCC crack growth during and 100 h after


transients in water chemistry (>EPRI action level 1
limit of EPRI BWR water chemistry guidelines64) or
load transients not covered by fatigue evaluation
procedures. These curves are currently undergoing
minor revisions.
The conservative character of the BWRVIP-60
disposition line 1 for SCC crack growth in CS &
LAS under NWC conditions has been confirmed
for temperatures in the range of 274288  C and

132

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

both base and weld filler/heat-affected zone (HAZ)


materials (Vickers hardness < 350 HV5, 0.02 wt% S)
if the water chemistry is maintained within current
BWR/NWC operational practice (<EPRI action
level 1 limit).38,60
In the case of NWC conditions, several results
clearly indicate, however, that line 1 may be slightly
exceeded at intermediate temperatures (180270  C)
in CS & LAS which show distinct susceptibility to
DSA. Not unexpectedly, sustained SCC with crack
growth rates significantly above line 1 is observed at
288  C if excessive hardness (350 HV5) is present in
the steel, for example, in bad weld HAZs. Furthermore, BWRVIP-60 disposition line 2 is significantly
exceeded for the case of ripple loading (R > 0.95) or
during chloride transients ( EPRI action level 1
limit), even at fairly low stress intensity values
around 30 MPa m1/2 (see Figure 19). On the other
hand, line 2 seems to cover even very severe sulfate
transients above the EPRI action level 3 limit.38,60
At low ECP, that is, in the case of BWR/HWC or
PWR conditions, line 1 appears conservatively to
cover the SCC crack growth in LAS, even under
the otherwise critical conditions mentioned earlier
for more oxidizing, environments.38,60

FRAD-EAC-mechanism
1

5.06.3.2.4 EAC mechanisms and models


5.06.3.2.4.1

EAC crack growth mechanism

For the case of EAC in CS & LAS in hightemperature water, the film rupture/anodic dissolution (FRAD), also referred to as slip dissolution/
oxidation, and hydrogen-assisted EAC (HAEAC)
cracking mechanisms have been proposed in the
literature (see Figure 20). Either of these may be
superimposed on pure mechanical fatigue.
In the FRAD mechanism, the protective oxide film
formed on CS & LAS in high-temperature water is
ruptured by plastic straining at the crack-tip. The
crack-tip then advances by anodic dissolution of the
bare metal matrix. Anodic dissolution is slowed down
and finally stopped by the nucleation and reformation
of the oxide film (repassivation). Thus, continued
crack advance will depend on a further oxide rupture
process due to the action of a strain rate at the cracktip. The crack propagation rate is controlled by both
anodic dissolution/repassivation kinetics and the frequency of oxide film rupture at the strained crack-tip.
The first part is governed by the chemical composition
of the local crack-tip electrolyte and the material. The
second part is determined by the fracture strain of
the oxide film and the crack-tip strain rate.63

Hydrogen-assisted EAC mechanism


syy

Crack surface covered by


an oxide film

Bare surface of metal


generated by oxide film rupture and
anodic dissolution of the metal

Elasticplastic
stress distribution

Acidic oxygen free water


Oxide layer
1

Void
8

8
3

MnS

MnS

Fe2+
4

1. Local anodic reaction

Repassivation: Oxide
nucleation and growth

ad

Fe Fe2+ + 2e
2. Hydrolysis and generation of H+
Fe2+ + H2O FeOH+ + +
3. Liquid phase transport
4. Local cathodic reaction
H+ + e H
MnS + 2H+ H2S + Mn2+

7
5. Hydrogen absorption
6. Hydrogen transport in lattice
7. Hydrogen trapping on inclusions
8. Hydrogen-induced cracking
9. Linkage of microcracks to
main fracture

Figure 20 Schematic illustrations of film rupture/anodic dissolution (left) and of hydrogen-assisted environmentally
assisted cracking mechanisms for carbon and low-alloy steel in high-temperature water. Adapted from Ford, F. P. J. Pressure
Vessel Technol. 1988, 110, 113128; Hanninen, H. et al. Corros. Sci. 1983, 23, 663679.

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

Oxide film rupturerepassivation events at the


strained crack-tip are also involved in the HAEAC
model, but here, hydrogen-induced microcrack formation ahead of the crack-tip and linkage of
these microcracks to the main crack are the prime
sources of EAC crack growth, and result in discontinuous crack propagation. The hydrolysis of metal
cations from anodic dissolution is an important
source of hydrogen, but in contrast to the FRADmechanism anodic dissolution does not contribute
significantly to crack advance. Hydrogen transport in
the electrolyte and in the metal lattice is believed to
be fast. Therefore, the generation of bare metal surface by film rupture and film reformation may be the
rate-controlling steps and explain the strain rate
dependence of EAC.65,66
The FRAD and HAEAC mechanisms may be
simultaneously active and controlled by the same
rate-limiting steps (e.g., oxide film rupture rate or
repassivation kinetics). Both mechanisms are able to
explain the observed, dominant effect of strain rate
and of MnS inclusions on EAC. The EAC behavior of
CS & LAS in high-temperature water can best
be rationalized by a superposition/combination of
the FRAD and HAEAC mechanisms. At lower temperatures (<100  C) and/or high yield strength/
hardness levels (>800 MPa/>350 HV5) and high
strain rates (>103 s1), hydrogen effects are more

HCl H2SO4

NaCl

Role of MnS inclusions

Sulfur-anions as HS, S2, and SO2


may signifi4
cantly retard repassivation after oxide film rupture
and therefore increase crack advance by anodic dissolution in the FRAD model.38,63 Retarded repassivation of the film-free surface and adsorbed HS, S2,
or H2S also increase hydrogen absorption into the
metal lattice and therefore favor HAEAC.38,65,66 Furthermore, the dissolution of MnS is a further potential source of hydrogen, and MnS inclusions in the
region of maximum hydrostatic stress ahead of the
crack-tip may act as strong hydrogen traps and thus
HAEAC microcrack initiation sites.38,65,66
The effects of steel sulfur content are synergistic
with environmental variables, such as (sulfur-) anionic
impurities in the bulk environment, ECP (dissolved
oxygen content), and flow rate (Figure 21).38,67 This is
believed to be due to the creation of a sulfur-rich
crack-tip environment responsible for EAC, which
arises both from the dissolution of MnS intersected
by the growing crack and by the transport of sulfuranions by migration/diffusion/convection within the
crack enclave.38

SSRT, 1106s1
hpw, 288 C
Low-flow autoclave

100
Corrosion potential ECP (mVSHE)

Corrosion potential ECP (mVSHE)

5.06.3.2.4.2

Filled symbols TGSCC


Open symbols no SCC

100
200

pronounced. At high temperatures (150  C) and/or


lower yield strength/hardness levels and slow strain
rates (<103 s1), anodic dissolution appears to
dominate.38

Pure water
Cl
SO42

Na2SO4

300
400
500

SCC
200

300
No SCC

400
A533-B A508 A533-B

0
(a)

133

100

200

300

400

Anion concentration (ppb)

500

A533-B

0.005 0.010 0.015 0.020 0.025

600
(b)

Sulfur content (wt.%)

Figure 21 Synergistic effect of electrochemical corrosion potential and sulfate anion concentration (a) and steel sulfur
content (b) on strain-induced corrosion cracking susceptibility of low-alloy steel in high-temperature water. Data obtained
from Sund, G.; Rosborg, B. The influence of impurities on the tendency to stress corrosion cracking of pressure vessel steel
A533-B in water at 288  C; Personal communication; 1991.

134

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

(which affect repassivation/pH/oxide film stability)


in the crack-tip environment. The onset and extent of
EAC is crucially dependent on simultaneously maintaining a slow, positive crack-tip strain rate and a
critical, local sulfur-anion activity of about 15 ppm
S2. If these two conjoint requirements are not met,
no SCC and SICC and only minor environmental
acceleration of fatigue crack growth are generally
observed. If the critical anion concentration is
exceeded, the EAC crack growth rate depends primarily on the crack-tip strain rate and increases with
increasing external strain rate up to an upper critical
limit of around 103102 s1 (Figure 23). Most
EAC thresholds, as well as phenomena involving
cessation or local retardation of crack growth, may
be directly attributed to this conjoint requirement
and thus to crack-tip environment chemistry and
crack growth rate/crack-tip strain rate, rather than
to overall loading parameters or bulk environmental
conditions per se. For example, suitable combinations
of different parameters may help exceed the critical
crack-tip sulfur-anion concentration, provided that
there is at least one source of sulfur (MnS inclusions
or sulfur anions in the bulk environment).38,63,69
These two local factors are governed by a system
of interrelated and synergistic corrosion parameters:
In general, the crack-tip strain rate increases with
increasing loading rate/frequency/level, crack growth
rate, yield strength, and DSA susceptibility. The
sulfur anion activity in the crack-tip environment
increases with increasing steel sulfur content, concentration of sulfur anions in the bulk environment outside the crack, crack growth rate (exposure of fresh,
dissolvable MnS by the growing crack), and corrosion

Under certain combinations of temperature (100


350  C) and strain rate (108102 s1), DSA may
synergistically interact with both mechanisms to
increase EAC susceptibility and provide an additional contribution to the crack growth process
(Figure 22). DSA may result in the occurrence of a
higher crack-tip strain/strain rate than for loading
outside the DSA range, or than in a material which
is not susceptible to DSA. The localization of plastic
deformation and increase in planar deformation from
DSA probably support mechanical rupture of the
oxide film and result in a reduction of the local
fracture toughness, thus favoring quasi-brittle crack
extension.38,68
The concentrations of free, interstitial nitrogen
and carbon (which mainly govern the DSA susceptibility in CS & LAS) might therefore be just as relevant for EAC susceptibility as the steel sulfur content,
and some data suggest that DSA may even overwhelm sulfur effects under certain conditions. DSA
generally results in an extension of the susceptibility
region and can affect the temperature and strain-rate
dependence of EAC. The most pronounced effects
of DSA on EAC38 are typically observed close to
crack growth thresholds (e.g., for SCC under static
load,60 or for CF close to critical frequencies under
cyclic load47).
5.06.3.2.4.4 Controlling factors for EAC crack
growth

As shown by lab testing, EAC growth from incipient


cracks is essentially governed by the crack-tip strain
rate and the activity of sulfur or chloride anions

108

80

90
20 MnMoNi 5 5, A, 0.004 % S
DO = 8 ppm, 65 ppb SO42

20 MnMoNi 5 5, A, 0.004 % S
SRL tests (8 ppm DO, 65 ppb SO42)
Tensile tests in air (DIN 50125)

70

SCC crack growth da/dtSCC (m s )

80

Open symbols: no SICC detected by DCPD (Kmax


)
I

75

60

70

50
65

40

Reduction of area Z (%)

Stress intensity factor at the onset of


SICC crack growth KI,i (MPa m1/2)

90

85

109 BWR VIP 60 DL K = 6075 MPa m1/2


I

80
1010
75
1011
70
1012

Reduction of area Z (%)

5.06.3.2.4.3 Role of dynamic strain ageing

65

30
100

(a)

150

200

250

Temperature ( C)

300

1013

60
350

(b)

50

100

150

200

250

300

60
350

Temperature ( C)

Figure 22 Coincidence of maximum of strain-induced corrosion cracking susceptibility in slow rising load tests (a) and
maximum in stress corrosion cracking growth rates in constant load experiments (b) with maximum in DSA susceptibility in
terms of temperature (and strain rate). (a) Reproduced from Seifert, H. P.; Ritter, S. J. Nucl. Mater. 2008, 378, 312326;
(b) reproduced from Seifert, H. P.; Ritter, S. J. Nucl. Mater. 2008, 372, 114131.

SA 533 B Cl. 1, C, 0.018% S, SICC


20 MnMoNi 5 5, A, 0.004% S, no SICC
20 MnMoNi 5 5, A, 0.004% S, SICC

107

108
6

Scatter band for dCODLL/dt =2 410

mm s1

8 ppm DO, + 150 mvSHE,288 C


dCODLL/dt = 24 106 mm s1

109

1010

135

DO = 8 ppm, <1or 65 ppb SO42, T = 288 C, 250 C, 200 C

106

SCC or SICC crack growth rate


da/dt scc or da/dtSICC (ms1)

SICC crack growth rate da/dtSICC (m s1)

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

100
200
300
Sulfate content (ppb)

106

Constant load
Very low-frequency cyclic load

107

Slow rising load


Linear fit of
95% prediction interval

108
109

1010

400

da/dtEAC (dCODLL/dt)0.8
Model: da/dtEAC = A . de/dtn

1011

109 108 107 106 105 104 103


Crack-opening displacement rate dCODLL/dt (mm s1)

Environmental component, +200 mvSHE


Mechanical component, air fatigue

106

Max. environm.
component

e
ur lin

ig

ue

-sulf

High

Shifted to
higher de/dt
10
10
with ECP , S
SO42 or Cl

108

rf

at

108

1012 10
10

(a)

DSA, YS de/dt for given loading conditions

fu

ul

-s

Ai

EAC crack growth rate (ms1)

104

lin

105

107

DO

DH [ppm]

21 0.79
8
0
0.4 0
HWC
+150 mVSHE

DO

DH

8.4 1.26
0 0.095
0 0.150

104

Crack-tip strain rate de/dt

102

100

(s1)

(b)

: NMCA

HWC
500 mVSHE T = 27428 C
low-alloy steel

109

F & A-model for NWC


F & A-model for HWC
PSI NWC curve
da/dtAir

1011

w
Lo

106

Corrosion fatigue crack growth rate in


high-temperature water da/dtCF (m s1)

Figure 23 Effect of sulfate content on strain-induced corrosion cracking crack growth in a low- and high-sulfur low-alloy
steel (left) and of crack mouth opening rate (as a measure for the crack-tip strain rate) on environmentally assisted cracking
crack growth rate under different loading conditions (right). Evidence for the described conjoint conditions and the important
role of crack-tip strain rate. Reproduced from Seifert, H. P.; Ritter, S. J. Nucl. Mater. 2008, 378, 312326.

High-sulfur line
Low-sulfur line

t Air

/d
da

1011

109

107

105

Fatigue crack growth rate in air da/dtAir (m s1)

Figure 24 Basic environmentally assisted cracking crack growth curves of the F & A model (a) and good correlation of the
observed corrosion fatigue crack growth behavior under boiling water reactor/normal water chemistry and hydrogen water
chemistry or noble metal chemical addition conditions with the predictions of the F & A model (b). (a) Adapted from Ford, F. P.
J. Pressure Vessel Technol. 1988, 110, 113128 and (b) reproduced from Seifert, H. P.; Ritter, S. Corros. Sci. 2008, 50,
18841899.

potential/dissolved oxygen content (enhanced migration of sulfur anions toward the crack-tip).38
5.06.3.2.4.5 Ford & Andresen EAC model

The mixed mechanisticphenomenological Ford and


Andresen model (Figure 24) is the sole quantitative
approach for CS & LAS and is based on the FRAD
mechanism.48,49,63,69,70 In the case of corrosion
fatigue, a linear superposition of fatigue and EAC
crack growth is assumed. In this model, EAC crack
growth is controlled by the crack-tip strain rate
de/dCT and the sulfur-anion activity cCT in the
crack-tip electrolyte. The EAC crack growth rate
da/dtEAC increases with increasing crack-tip strain

rate de/dCT according to a power law and saturates


above a critical strain rate of 103102 s1 (continuous dissolution at a crack tip free of protective
oxide).
The exponent of this power law relationship
is dependent on the sulfide-anion activity in the
crack-tip environment. Above a sulfide concentration
of 0.02 ppm (low-sulfur threshold), the formation of
a new oxide layer is increasingly delayed by increasing sulfide content, which thus leads to a larger
increment of crack advance by anodic dissolution
per oxide-rupture event. Above a sulfide content of
2 ppm (high-sulfur threshold), the sulfide effect on
repassivation saturates. Based on this experimentally

136

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

derived relationship, lower and upper limiting EAC


crack growth curves for CS & LAS (Figure 24) were
defined (so-called low- and high-sulfur EAC lines).
A critical, high sulfur-anion content of  2 ppm S2
has to be maintained in the crack-tip electrolyte to
sustain fast, high-sulfur EAC crack growth rates, otherwise the crack growth rates quickly drop down to
low-sulfur crack growth rates.48,49,63,69,70
The sulfur-anion concentration in the crack-tip
environment from dissolution of MnS inclusions is
assumed to be proportional to the steel sulfur content
and the crack growth rate itself (intersection of the
crack with new, dissolvable MnS inclusions) and to
increase exponentially with the potential gradient/
ECP (based on crevice experiments and modeling).
Similarly, the degree to which sulfur anions are
enriched in the crack-tip environment depends on
how much they tend to migrate there from the bulk
environment. It is thus proportional to their concentration and increases exponentially with potential
gradient/ECP.48,49,63,69,70
The Ford and Andresen model correctly predicts
most experimentally observed EAC data trends for
CS & LAS over a wide range of conditions.38 Furthermore, it has also been successfully applied to
predict LCF initiation.70 It therefore has good potential for data analysis and the definition of disposition
lines. The high-sulfur line of the model conservatively covers almost all EAC crack growth rate data.
Recent results indicate, however, that the transition
curve between the low- and high-sulfur CF lines
might be nonconservative under highly oxidizing
BWR/NWC conditions (ECP > 0 mVSHE) and very
low loading frequencies (<104 Hz),47 possibly as a
result of DSA (which is not considered in the model).
Furthermore, the model predicts much too high
SCC crack growth rates at KI levels < 60 MPa m1/2
in high-purity water.38 This is related to the cracktip strain rate formulation for static loading conditions,
which was originally derived for stainless steels and is
generally unsatisfactory when applied at these lower
KI levels.38 In addition, the model does not explain the
unexpectedly strong effect of relatively low chloride
levels on SCC crack growth rates (see Figure 19).
5.06.3.2.5 Service experience and
mitigation actions
5.06.3.2.5.1 Service experience

The accumulated operating experience with CS &


LAS pressure-boundary components in the primary
and secondary coolant circuits of LWRs is very good

worldwide. Use of the current fatigue design and


evaluation codes (ASME III and XI) has been generally successful in preventing fatigue cracks and failures in CS & LAS components and therefore the
limited (and nonspecific) consideration of EAC
in these codes would at first sight seem to be
adequate under most operating circumstances, or
even (fortuitously) conservative. However, isolated
instances of EAC have occurred under specific circumstances and these always involved several unfavorable factors. EAC has occurred particularly in
BWR service, most often in CS & LAS piping, less
frequently in vessels, and very rarely in the clad RPV.
EAC cracking has been observed in wrought, weld
filler, and weld HAZ materials and has been transgranular in nature. In BWRs, steam and feedwater
piping as well as condensate systems and RPV
feedwater nozzles have been affected (Figure 25).
In the secondary circuit of PWRs, cracking has
been observed in feedwater piping/tanks and heat
exchangers, feedwater nozzles of PWR steam generators, and steam generator girth welds. EAC damage was usually detected during in-service inspection
and seldom led to through-wall penetrations with
leakage.1,38,39,46,63,7174
These EAC incidents have been clearly associated
with an oxidizing environment (usually due to dissolved oxygen), severe dynamic straining (due to
global and local thermal stratification/striping, or
due to thermal and pressurization cycles during plant
transients), and either high local stresses (around or
above the high-temperature yield strength) or high
secondary/residual stress (due, e.g., to welding defects,
pipe bends in conjunction with inadequate pipe supports or restraints, and localized post-weld treatment).
Most PWR cracking incidents involved aerated feedwater under specific start-up operating conditions (no
application of nitrogen blankets in condensate storage
tanks) or at special locations (e.g., locations with air
present during shutdown), where the normally low
contents of oxygen and low ECP during operation
cannot be assumed upon startup. The fact that cracking frequently occurred in BWR lines with stagnant
steam or nondegassed condensate, but not in comparable lines carrying flowing steam, is another indication that oxygen concentration and ECP are crucial
parameters with respect to the occurrence of EAC,
since such conditions are known to encourage
the formation of a condensate film rich in dissolved
oxygen. Thermal hydraulics (thermal stratification/
striping phenomena) and local stress raisers played
a key role and were often more important than water

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

5 mm

(a)

(c)

(b)

Feed water pipe

Weld root

5 mm

137

5 mm

RPVnozzle

(d)

200 mm

Figure 25 (a) Circumferential strain-induced corrosion cracking crack at the feedwater pipe to reactor pressure vessel
nozzle weld of a German boiling water reactor originating form a weld root notch due to severe thermal stratification during
plant start-up and hot stand-by at low feedwater flow rates. Reproduced from Hickling, J.; Blind, D. Nucl. Eng. Des. 1986, 91,
305330. (b) Crack interruptions markings on a strain-induced corrosion cracking fracture surface indicating intermittent
crack growth during plant transients. Reproduced from Hickling, J.; Blind, D. Nucl. Eng. Des. 1986, 91, 305330. (c) Formation
of multiple, semielliptical strain-induced corrosion cracking cracks at a circumferential weld between a straight seamless pipe
and a valve casing in a boiling water reactor. Reproduced from Hickling, J.; Blind, D. Nucl. Eng. Des. 1986, 91, 305330.
(d) Corrosion fatigue cracks in carbon steel instrumentation line at the main steam pipe of a pressurized water reactor
originating from corrosion pits with intermittent corrosion during shutdown and subsequent reinitiation of cracking during
plant transients. Reproduced from Roth, A. et al. In Proceedings of the 12th International Conference on Environmental
Degradation of Materials in Nuclear Power System Water Reactors, Snowbird, Salt Lake City, UT, Aug 1418; Todd, R. A.,
King, P. J., Nelson, L., Eds.; The Minerals, Metals and Materials Society: Warrendale, PA, 2005; pp 795802.

chemistry or material aspects. These cases were


attributed either to SICC or CF. Cracking incidents
with a major or relevant contribution of SCC to the
total crack advance in properly manufactured and
heat-treated CS & LAS primary pressure-boundary
components have not been observed. In fact, longterm exposure and reinspection of unclad components, such as BWR feedwater nozzle radii or vessel
heads, and several cracking incidents in dissimilar
metal welds (e.g., in core shroud support welds,
where crack-tips in the susceptible Alloy 182 weld
metal arrested in the interface region to the adjacent
LAS) confirm the generally excellent resistance of
CS & LAS to SCC in reactor coolant.38,39,72,73
Cracks often initiated from water-wetted geometric discontinuities (e.g., welding defects) and/or in
regions with stagnant flow or creviced geometry.

In several cases, EAC cracks preferentially initiated


from corrosion pits (Figure 25) and corrosion during
shutdown periods (pitting and sometimes corrosion
products (rust) close to the water line as a possible
source of anionic impurities during startup) can be
a further contributing factor. In general, crack initiation
has been more affected by high-frequency, high-cycle
fatigue, for example, due to local thermal stratification
or thermal striping loads which were limited to nearsurface regions. Crack propagation was often dominated by LCF from slower and less frequent transients,
for example, due to more global thermal stratification
or to operational power transients.38,39,72,73
Several incidents were related to unanticipated
sources of thermal stress cycles in critical locations.
In many cases, fabrication or design deficiencies
(e.g., welding defects) favored local plastification of

138

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

the water-wetted material surface and sometimes


resulted in an increased loading level (from residual
stress) or increased EAC susceptibility of the material
(e.g., excessive hardness of weld HAZs and distinct
DSA susceptibility of weld metal). Reoccurrence of
this kind of cracking can be avoided by improved
design of the components and/or fabrication procedures, as well as by better quality control during
the manufacturing process, and in some cases by
optimized operational procedures.38,39,7274

5.06.3.2.5.2 Critical components and operational


conditions

Special attention should be paid to components which


are likely to undergo significant (localized) mechanical
loading, or which are exposed to increased oxygen
levels in the coolant, higher conductivities, and/or
quasi-stagnant flow conditions during operation.
Critical components, for example, are piping carrying nearly stagnant steam, or nondegassed condensate, during normal reactor operation (e.g., in systems
which are used only intermittently), feedwater nozzles and adjacent sections of horizontal piping (if
thermal stratification can occur) and thin-walled piping, and pipe bends in conjunction with inadequate
pipe supports or restraints. Operational transients
associated with thermal stratification and thus high
secondary strains (e.g., start-up, hot stand-by, or slow,
but relevant pressure and temperature changes)
should receive attention, in particular, if they occur
with a sufficiently high frequency.38
5.06.3.2.5.3 Possible mitigation actions and
countermeasures

Apart from good design, the common EAC mitigation


strategy is, firstly, to exclude large preexisting defects
by nondestructive examination and quality assurance
measures during fabrication and installation of components and, secondly, to avoid during operation
those water chemistry and stress or strain combinations which could lead to either EAC crack initiation
or to accelerated crack growth. This procedure
should be complemented by periodic, nondestructive, in-service inspection of critical component locations. In summary, EAC risks can be minimized by38:
Selection of suitable materials, for example, lowsulfur steel (<0.003 wt% S) with low EAC and
DSA susceptibility and optimized high toughness
(! larger critical crack size).
Selection of suitable manufacturing and fabrication practices to avoid welding defects and HAZs

with high hardness (350 HV5) and to reduce residual stress (e.g., stress relief heat treatment and
narrow gap welding).
Improved design to reduce regions of high local
stresses (by increased wall thickness, by internal
flush grinding of joints and optimization of welding
technology, avoidance of discontinuities and constraints, optimized pipe supports, etc.).
Reduction of the number and severity of thermal
and pressurization cycles (thermal-stratification
during hot stand-by, startup/shutdown), for example, by optimized operating procedures and/or by
improved design of the affected component (thermal sleeves for feedwater nozzles, feedwater spargers, etc.).
Avoidance of near-stagnant operating conditions
(including in crevices or dead-legs) and reduction
in dissolved oxygen levels (e.g., by inert gas shielding of make-up water, modification of startup procedures to allow venting of piping, improvement of
drainage in horizontal lines, etc.).
Maintenance of adequate water chemistry control
with regard to anionic impurities (EPRI or VGB
guidelines) and application of HWC/NMCA to
reduce the ECP in BWRs.

5.06.3.2.5.4 Service experience vs. experimental


and theoretical background knowledge

Plant-operating experience fits well to the accumulated


experimental/mechanistic background knowledge on
EAC and pitting behavior for CS & LAS components
exposed to high-temperature water. Both show the
same qualitative parameter trends, for example, for
dissolved oxygen (ECP), flow rate, and strain. They
confirm the high SCC resistance of CS & LAS under
static loading conditions (steady-state power operation) and clearly reveal that slow, positive (tensile)
straining with associated (localized) plastic yielding
and sufficiently oxidizing conditions are essential for
EAC initiation in high-purity, high-temperature water.
Since primary design stresses are generally limited to
values below the elastic limit, thermal-hydraulics (e.g.,
thermal stratification) and local stress raisers have
therefore played a key role for SICC/CF in the field
and were generally more important than material or
water chemistry aspects. EAC cracks often initiated
from corrosion pits and pitting of LAS is strongly
favored under oxidizing conditions, especially at low
and intermediate temperatures, and by slow, dynamic
straining. Quasi-stagnant or low-flow conditions promote EAC initiation by the formation of an aggressive, occluded water chemistry at the base of the pit

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

(or within small surface defects), as well as by the


enrichment of oxygen in steam condensate.38,39,72,73
By taking the exact boundary conditions in laboratory tests and in the field into account, no major
discrepancy between operating experience and experimental trends is found. The few EAC cracking incidents in the field can be rationalized by the identified
susceptibility conditions and by EAC flaw tolerance
evaluations. The apparently higher cracking frequency in lab tests may be explained by the beneficial
effect of a turbulent flow rate in the field (which is
characteristic for most component locations) upon
EAC initiation, the large degree of conservatism
in the fatigue evaluation procedures, and the conjoint
threshold conditions for the onset of EAC with
regard to ECP (or dissolved oxygen), strain rate,
strain, and temperature. In general, for most transients/component locations, one or more threshold
conditions are not satisfied. The threshold strain
and stress for SICC/CF and SCC initiation, in particular, are seldom exceeded in the case of good
operational practice and the absence of design or
fabrication deficiencies.38,39,7173

5.06.4 Conclusions
5.06.4.1 Uniform and Flow-Accelerated
Corrosion
Uniform corrosion of CS and LAS poses no significant threat to the integrity of coolant systems under
normal operating conditions; the typical corrosion
rates of less than 1 mm year1 or so are manageable.
A significant drawback of the material is the rather
thick corrosion-product layer that develops in hightemperature water double-layered magnetite in
alkaline, reducing conditions (such as PWR or
CANDU primary coolants) but with an admixture
of oxidized species such as maghemite or hematite in
neutral, oxidizing conditions (such as BWR coolant).
The fouling and contamination of reactor circuits
from the release of such oxides by spalling in high
flows, as well as by dissolution as a result of solubility
gradients, is minimized by careful chemistry control.
Similarly, although the material in contact with PWR
primary coolant is mainly stainless steel or nickelbased alloy, the transport of nickel-ferrite corrosion
products is minimized by continuous control of alkalinity. The reactor circuit in BWRs exposes mainly
stainless steel to the coolant, but depending upon
the chosen operating chemistry (NWC, HWC,
HWC/NMCA) the chief source of corrosion

139

products (CS/LAS in the feedtrain) can be controlled


by the addition of metal ions to the feedwater so as to
give an optimum ratio of Fe/Ni and minimize radiation field buildup. CANDUs have unclad CS in the
primary circuit and operate at constant high pH;
although this controls activity transport, general fouling of the circuit from corrosion products released
from FAC can create thermohydraulic problems.
Some CANDUs have had to clean components in
the primary system in order to restore optimum
operating conditions.
The stainless-steel cladding in PWR primary circuits prevents corrosion of the LAS base material,
especially under shutdown conditions. However,
leaks at high temperature can allow the boric acid
in the coolant to concentrate on unclad external surfaces and, together with oxygen from the atmosphere,
can provoke highly corrosive situations. Flow effects
can exacerbate such attack. Rigorous surveillance
regimes are required to avoid damage to components
such as flanges, valve fittings and especially, the
pressure-vessel head in the vicinity of leaking penetrations. Corrosion of unclad CS & LAS in PWR
secondary systems is prevented by chemistry control
during HT-operation, but additional measures (e.g.,
nitrogen blanketing) are required during shutdown.
FAC of CS in reactor systems is most damaging in
feedwater circuits, where the solubility of normally
protective oxide films is high, dissolved iron is low,
and turbulence is intense. Chronic FAC of feeders in
the CANDU primary circuit occurs through similar
mechanisms, although the oxide solubility is low at
the constant high pH and temperature of the coolant.
Depending upon the circumstances, oxide solubility
can be controlled to some extent by adjusting the
chemistry with pH additives (such as ammonia)
and/or adding oxygen (as is done in various feedwater circuits) or even by replacing components with
higher-Cr material. Commercial software is available
to pinpoint likely vulnerable areas, give some indications of possible rates of attack, and manage databases. However, rigorous surveillance of vulnerable
components, for example, where high coolant turbulence is induced, is still necessary to avoid catastrophic pipe failures.
5.06.4.2 Localized Corrosion and
Environmentally Assisted Cracking
In CS & LAS, a shallow form of pitting can occur in
the complete absence of anionic water impurities as
the corrosion potential (ECP) is raised, for example,

140

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

through oxygen and other oxidizing species. Some


degree of pitting corrosion is inevitable after longterm exposure of unclad CS & LAS surfaces to water
in LWR systems and is not usually a threat to either
coolant purity or to structural integrity. Shallow
pitting has been observed primarily in specific piping
systems with residual water because of incomplete
draining during nonoperational periods (shutdown
corrosion). This can be avoided by adequate wet or
dry preservation techniques.
Laboratory investigations have revealed significant effects of simulated reactor environments on
fatigue crack initiation/growth, as well as the possibility of SCC crack growth under certain critical
combinations of environmental, material, and loading
parameters. During the last three decades, the major
factors of influence and EAC susceptibility conditions have been clearly identified. The effects of
most parameters on EAC initiation and growth are
adequately known with acceptable reproducibility
and reasonably understood by mechanistic models.
Tools for incorporating environmental effects into
ASME III fatigue design curves have been developed/qualified, but not yet widely adopted. The
BWRVIP-60 SCC disposition lines and ASME XI
reference fatigue crack growth curves are usually
conservative and adequate under most PWR and
BWR operational circumstances, but there is nevertheless valuable potential for the reduction of undue
conservatism and the need for improvements to eliminate some specific inadequacies.
The operating experience of CS & LAS primary
pressure-boundary components in LWRs is very good
worldwide. However, isolated instances of EAC have
occurred, particularly in BWR service, most often in
piping and more rarely in the RPV itself. Oxidizing
conditions (usually dissolved oxygen) and relevant plastic straining were always involved. These cases were
attributed either to SICC or CF and can be rationalized
by the experimental/mechanistic background knowledge. Both service experience and appropriate laboratory experiments confirm the high resistance of CS
& LAS to SCC under static loading conditions (representative of steady-state power operation) and
clearly reveal that slow, positive (tensile) straining,
with associated (localized) plastic yielding and sufficiently oxidizing conditions, are usually necessary
for EAC initiation in high-temperature water. Based
on the above, different remedial and mitigation
actions have been qualified and successfully applied
to further reduce the low EAC cracking frequency
in the field.

In spite of the absence of SCC in operating reactors, several unfavorable, critical parameter combinations have been identified in the laboratory, which
can lead to sustained, fast SCC, with crack growth
rates well above the BWRVIP-60 SCC disposition
lines. Most of them appear atypical for current
BWR plant operation with properly manufactured
CS & LAS components, but some further clarification is required, in particular, with regard to
improved quantification of the boundaries/thresholds for the transition from very low to high SCC
crack growth rates. In this context, research should be
focused on the effects of chloride transients and DSA/
yield strength on the SCC crack growth behavior of
CS & LAS (including weld and HAZ materials) under
BWR/NWC conditions, as well as on the behavior of
Alloy 182-LAS dissimilar metal weld joints. Additionally, the mitigation effect of HWC/NMCA should be
confirmed under these critical conditions.

References
1.
2.

3.
4.
5.
6.
7.
8.
9.
10.

11.

12.

13.

Electric Power Research Institute. Materials handbook for


nulcear plant pressure-boundary components, EPRI
1002792; EPRI: Palo Alto, CA, 2002. http://my.epri.com/.
Electric Power Research Institute. Boric acid corrosion
guidebook, revision 1; managing boric acid corrosion
issues at PWR power stations, EPRI 1000975; EPRI:
Palo Alto, CA, 2001http://my.epri.com/.
Potter, E. C.; Mann, G. M. W. In Proceedings of the 1st
International Congress on Metallic Corrosion,
Butterworths: London, 1961; p 417.
Berge, P.; et al. Corrosion 1977, 33(5), 173178.
Bloom, M. C.; et al. Corrosion 1957, 13, 297302.
Mabuchi, K.; et al. Corrosion 1991, 47(7), 500508.
Hickling, J. Private communication; EPRI: Palo Alto, CA,
2006.
Beverskog, B.; Puigdomenech, I. Corros. Sci. 1996,
38(12), 21212135.
Dooley, R. B. Power Plant Chem 2008, 10(2), 6889.
Uchida, S.; et al. In Proceedings of the 13th International
Conference on Environmental Degradation of Materials in
Nuclear Power Systems, CD-ROM. Whistler, British
Columbia, Canada, Aug 1923; King, P., Allen, T.,
Busby, J., Eds.; The Canadian Nuclear Society: Toronto,
ON, 2007.
Slade, J.; Gendron, T. In Proceedings of the 12th
International Conference on Environmental Degradation of
Materials in Nuclear Power System Water Reactors,
Snowbird, Salt Lake City, UT, Aug 1418; Todd, R. A.,
King, P. J., Nelson, L., Eds.; The Minerals, Metals and
Materials Society: Warrendale, PA, 2005.
Marks, C.; et al. In Proceedings of the 6th International
Symposium on Contribution of Materials Investigations to
Improve the Safety and Performance of LWRs (Fontevraud 6),
CD-ROM, Fontevraud, France, Sept 1822. The French
Nuclear Energy Society (SFEN): Paris, France, 2006.
Woolsey, I.; Quirk, G. In International Conference on Flow
Accelerated Corrosion (FAC2008), DVD-ROM, Lyon,
France, Mar 1720; EDF: France, 2008.

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels


14. Bouchacourt, M. EPRI Workshop on Erosion-Corrosion of
Carbon Steel Piping, Washington, DC, Apr 1415, EPRI:
Palo Alto, CA, 1987.
15. Berge, P.; et al. In Proceedings of the Second International
Conference on Water Chemistry of Nuclear Reactor
Systems, Bournemouth, UK, October, British Nuclear
Energy Society (BNES): London, UK, 1980; Vol. 1.
16. Lister, D. H.; et al. In Proceedings of the 16th Pacific Basin
Nuclear Conference (16PBNC), CD-ROM, Aomori, Japan,
Oct 1318 Japanese Industrial Forum (JAIF): Japan, 2008.
17. Liu, L.; Lister, D. H. In Proceedings of the International
Conference on Water Chemistry of Nuclear Reactor
Systems (NPC08), Berlin, Germany, Sept 1518, VGB:
Essen, Germany, 2008.
18. Bignold, G. J.; et al. In Proceedings of the International
Specialists Meeting on ErosionCorrosion of Steels in
High-Temperature Water and Wet Steam, Les
Renardieres, France, May 1112 EDF: France, 1982.
19. Tremaine, P.; LeBlanc, J. C. J. Solution Chem. 1980, 9(6),
415442.
20. Heitmann, H. G.; Schub, P. In Proceedings of the Third
Meeting on Water Chemistry of Nuclear Reactors,
Bournemouth, UK, October British Nuclear Engineering
Society (BNES): London, UK, 1983. pp 243252.
21. Berger, F. P.; Hau, K. F. F. L. Int. J. Heat Mass Transf.
1977, 20, 1185.
22. Sydberger, T.; Lotz, U. J. Electrochem. Soc. 1982, 129(2),
276283.
23. Lister, D. H.; et al. In Proceedings of the 13th International
Conference on Environmental Degradation of Materials in
Nuclear Power Systems, CD-ROM, Whistler, British
Columbia, Canada, Aug 1923; King, P., Allen, T.,
Busby, J., Eds.; The Canadian Nuclear Society: Toronto,
ON, 2007.
24. Lister, D. H.; et al. Power Plant Chem. 2008, 10(11),
659667.
25. Henzel, N.; et al. Erosion corrosion in power plants under
single- and two-phase flow conditions-updated
experience and proven counteractions. In Proceedings of
the American Power Conference, Chicago, IL, USA, 1820
April 1988; Vol. 50. pp 9921000.
26. Lee, S.-H.; et al. J. Korean Nucl. Soc. 1999, 31(4),
375384.
27. Lee, S. H.; et al. J. Korean Nucl. Soc. 1999, 31(4), 375384.
28. Lister, D. H. In Proceedings of the 1998 JAIF International
Conference on Water Chemistry in Nuclear Power Plants,
Kashiwazaki, Japan, 1316 Oct, Japan Atomic Industrial
Forum (JAIF): Japan, 1988; pp 442-1442-8.
29. Balakrishnan, P. V. Can. J. Chem. Eng. 1977, 55, 357.
30. Lister, D. H.; Lang, L. C. In Proceedings of the International
Conference on Water Chemistry in Nuclear Reactor
Systems (Chimie 2002), CD-ROM, Avignon, France, Apr
2226; The French Nuclear Energy Society (SFEN): Paris,
France, 2002.
31. Ponguak, J.; et al. Boric acid corrosion of reactor pressure
vessel steel caused by an impinging jet of simulated PWR
coolant. In Proceedings of the International Conference on
Water Chemistry of Nuclear Reactor Systems, CD-ROM,
Jeju Island, Korea, 2326 Oct, 2006.
32. The Nuclear and Industrial Safety Agency. Secondary
piping rupture accident at Mihama power station, unit 3 of
the Kansai Electric Power Co., Inc. (Final Report),
Revision 1 (translated by Japan Nuclear Energy Safety
Organization (JNES)); The Nuclear and Industrial Safety
Agency: Japan, 2005; http://www2.jnes.go.jp/.
33. McGrath, M. A.; et al. In Proceedings of the International
Conference on Water Chemistry of Nuclear Reactor
Systems (NPC08), Berlin, Germany, Sept 1518 VGB:
Essen, Germany, 2008.

34.

35.
36.
37.

38.

39.
40.

41.
42.
43.
44.

45.
46.

47.
48.

49.
50.

51.
52.
53.

54.

141

Electric Power Research Institute. The general and


localized corrosion of carbon and low-alloy steels in
oxygenated high-temperature water, EPRI-NP-2853;
EPRI: Palo Alto, CA, 1983; http://my.epri.com/.
Electric Power Research Institute. BWR environmental
cracking margins for carbon steel piping. EPRI Report
NP-2406; EPRI: Palo Alto, CA, 1982; http://my.epri.com/.
Indig, M.; et al. Rev. Coat. Corros. 1982, 5, 173225.
Videm, K. Corrosion of steel in high-temperature
water influence of oxygen in water of performed oxide
coatings. In Proceedings of the 7th Scandinavian
Corrosion Congress, Trondheim, Norway, May 2628,
1975; pp 444456.
Seifert, H. P.; Ritter, S. Research and service experience
with environmentally-assisted cracking of carbon & lowalloy steels in high-temperature water. SKI-Report
2005:60; SKI: Stockholm, Sweden, 2005; ISSN 11041374; http://www.stralsakerhetsmyndigheten.se/.
Hickling, J.; et al. PowerPlant Chem. 2005, 7, 3142.
Seifert, H. P.; et al. Crack initiation due to environmentallyassisted cracking in carbon and low-alloy steels exposed
to high-temperature water-Part 1: Overview of results from
laboratory tests. Workshop on Detection, Avoidance,
Mechanisms, Modeling, and Prediction of SCC Initiation in
Water-Cooled Nuclear Reactor Plants (CD-ROM),
Beaune, Burgundy, Sept 712, 2008.
Chopra, O. K.; Shack, W. J. J. Pressure Vessel Technol.
2009, 131: 021409-1021409-21.
US NRC. Effect of LWR coolant environments on the
fatigue life of reactor materials, NUREG/CR-6909;
US NRC: Washington, DC, 2007; http://www.nrc.gov/.
Congleton, J.; et al. Corros. Sci. 1985, 25, 633650.
Atkinson, J. D.; et al. In Proceedings of the 13th
International Conference on Environmental Degradation of
Materials in Nuclear Power Systems, CD-ROM, Whistler,
British Columbia, Canada, Aug 1923; King, P., Allen, T.,
Busby, J., Eds.; The Canadian Nuclear Society: Toronto,
Canada, 2007.
Seifert, H. P.; Ritter, S. J. Nucl. Mater. 2008, 378, 312326.
Electric Power Research Institute. Corrosion fatigue of
water-touched pressure retaining components in power
plants, EPRI TR-106696; EPRI: Palo Alto, CA, 1997;
http://my.epri.com/.
Seifert, H. P.; Ritter, S. Corros. Sci. 2008, 50, 18841899.
Ford, F. P.; Andresen, P. L. In Proceedings of the 3rd
International Atomic Energy Agency Specialists Meeting
on Sub-critical Crack Growth, Moscow, USSR, May
1417; Cullen, W., Ed.; US NRC: Washington, DC, 1990.
Vol. 1; 105124, NUREG/CP-0112.
Ford, F. P. J. Pressure Vessel Technol. 1988, 110, 113128.
The American Society of Mechanical Engineers. The 2004
ASME Boiler and Pressure Vessel Code, Section III, Rules
for Construction of Nuclear Power Plant Components;
The American Society of Mechanical Engineers:
New York, 2004.
Higuchi, M.; Iida, K. Nucl. Eng. Des. 1991, 129, 293306.
Van Der Sluys, A. PVRCs position on environmental
effects on fatigue life in LWR applications, WRC Bulletin
487; Welding Research Council: New York, 2003.
Higuchi, M. In Proceedings of the ASME 2008 Pressure
Vessel and Piping Conference (PVP 2008), CD-ROM,
Chicago, IL, 2731 July; The American
Society of Mechanical Engineers: New York, 2008; ISBN
0-7918-3828-5.
The Japanese Society of Mechanical Engineers. Codes
for nuclear power generation facilities: Environmental
fatigue evaluation method for nuclear power plants,
JSME S NF12006; The Japanese Society of Mechanical
Engineers: Tokyo, Japan, 2006.

142

Corrosion and Environmentally-Assisted Cracking of Carbon and Low-Alloy Steels

55. US NRC. Guidelines for evaluating fatigue analyses


incorporating the life reduction of metal components due
to the effect of the light-water reactor environment for new
reactors, NRC Regulatory Guide 1.207; US NRC:
Washington DC, USA, 2007; http://www.nrc.gov/.
56. The American Society of Mechanical Engineers. The 2004
ASME Boiler and Pressure Vessel Code, Section XI, Rules
for Inservice Inspection of Nuclear Power Plant
Components; The American Society of Mechanical
Engineers: New York, 2004.
57. The American Society of Mechanical Engineers. The
2004 ASME Boiler and Pressure Vessel Code, Code
Cases for Nuclear Components, Code Case N-643;
The American Society of Mechanical Engineers:
New York, 2004.
58. Hickling, J.; Reitzner, U. VGB Kraftwerkstech. 1992, 72,
359367.
59. Kussmaul, K.; et al. Nucl. Eng. Des. 1997, 168, 5375.
60. Seifert, H. P.; Ritter, S. J. Nucl. Mater. 2008, 372, 114131.
61. Ford, F. P.; et al. In Proceedings of the 9th International
Conference on Environmental Degradation of Materials in
Nuclear Power Systems-Water Reactors, Newport Beach,
CA, Aug 15; Bruemmer, S., Ford, P., Was, G., Eds.; The
Minerals, Metals and Materials Society: Warrendale, PA,
1999; pp 855863.
62. Seifert, H. P.; Ritter, S. Review and assessment of SCC
experiments with RPV steels in Oskarshamn 2 and 3
(ABB Report SBR 99020), SKI Report 2005:61; The
Swedish Nuclear Power Inspectorate: Stockholm,
Sweden, 2005; ISSN 11041374; http://www.
stralsakerhetsmyndigheten.se/.
63. Electric Power Research Institute. Environmentally
assisted cracking of low-alloy steels, EPRI NP-7473-L;
EPRI: Palo Alto, CA, 1992; http://my.epri.com/.
64. Electric Power Research Institute. BWR water chemistry
guidelines 2004 revision, BWRVIP-130, 1008192; EPRI:
Palo Alto, CA, 2004; http://my.epri.com/.
65. Hanninen, H.; et al. Corros. Sci. 1983, 23, 663679.
66. Hanninen, H.; et al. In Proceedings of the 2nd International
Atomic Energy Agency Specialists Meeting on SubCritical Crack Growth, Sendai, Japan, 1517 May;

Cullen, W. H., Ed.; US NRC: Washington, DC, 1985,


Vol. 2; 7397, NUREG/CP-0067.
67. Sund, G.; Rosborg, B. The influence of impurities on the
tendency to stress corrosion cracking of pressure vessel
steel A533-B in water at 288 C. Personal communication.
1991.
68. Hanninen, H.; et al. In Proceedings of the 10th International
Conference on Environmental Degradation of Materials in
Nuclear Power Systems Water Reactors, CD-ROM, Lake
Tahoe, NV, Aug 610; The National Association of
Corrosion Engineers: Houston, TX, 2001; Paper No. 47.
69. Andresen, P. L. In Proceedings of the Third International
Conference on Fatigue and Fatigue Thresholds,
Charlottesville, VA, June 28July 3; Ritchie, R. O.,
Starke, E. A., Eds.; EMAS Publishing: Warrington, UK,
1987; 11891200, Vol. III. pp 11891200.
70. Electric Power Research Institute. Environmentally
assisted fatigue crack initiation in low-alloy steels
A review of the literature and the ASME Code
requirements, EPRI TR-102765; EPRI: Palo Alto, CA,
1993; http://my.epri.com/.
71. Hickling, J.; Blind, D. Nucl. Eng. Des. 1986, 91, 305330.
72. Roth, A.; et al. Crack initiation due to environmentallyassisted cracking in carbon and low-alloy steels exposed
to high-temperature water Part 2: Overview and
assessment of operating experience. Workshop on
Detection, Avoidance, Mechanisms, Modeling, and
Prediction of SCC Initiation in Water-Cooled Nuclear
Reactor Plants, CD-ROM, Beaune, Burgundy, France,
Sept 712, 2008.
73. Roth, A.; Hickling, J. In Proceedings of the 13th
International Conference on Environmental Degradation of
Materials in Nuclear Power Systems, CD-ROM, Whistler,
British Columbia, Canada, Aug 1923; The Canadian
Nuclear Society: Toronto, Canada, 2007.
74. Roth, A.; et al. In Proceedings of the 12th International
Conference on Environmental Degradation of Materials
in Nuclear Power System Water Reactors, Snowbird,
Salt Lake City, UT, Aug 1418; Todd, R. A., King, P. J.,
Nelson, L., Eds.; The Minerals, Metals and Materials
Society: Warrendale, PA, 2005; pp 795802.

5.07

Performance of Aluminum in Research Reactors

K. Farrell
Formerly of Oak Ridge National Laboratory, Oak Ridge, TN, USA

2012 Elsevier Ltd. All rights reserved.

5.07.1
5.07.2
5.07.2.1
5.07.3
5.07.3.1
5.07.3.2
5.07.4
5.07.5
5.07.6
5.07.6.1
5.07.6.2
5.07.6.2.1
5.07.6.2.2
5.07.6.2.3
5.07.7
5.07.7.1
5.07.7.2
5.07.7.3
5.07.7.4
5.07.8
References

Introduction
Typical Applications
History of Aluminum Applications in Research Reactors
Properties of Aluminum
Practical Characteristics
Alloy Types, Temper Designations, and Tensile Properties
Fuel Elements
Corrosion
Radiation Effects
Basics
Microstructures
Fluence
Temperature
Transmutation products
Property Changes
Swelling
Mechanical Properties
Effects of Neutron Spectrum
Radiation Softening, Creep, and Stress Relaxation
Conclusion

Abbreviations
AIME
ANL
ANSI
ASM
ASTM
ATR
CRC
CTE
EBR-II
Emod
ETR
GR
HEU
HFIR
HPRR
IAEA
INL
IRV-M2

American Institute of Mining,


Metallurgical, and Petroleum Engineers
Argonne National Laboratory
American National Standards Institute
American Society for Metals
American Society for Testing Materials
Advanced Test Reactor
Chemical Rubber Company
Coefficient of thermal expansion
Experimental Breeder Reactor-II
Modulus of elasticity
Experimental test reactor
Graphite Reactor
Highly enriched uranium
High Flux Isotope Reactor
High performance research reactor
International Atomic Energy Authority
Idaho National Laboratory
Acronym for a recent Russian research
reactor

LANL
LEU
MTR

OPAL
ORNL
ORR
PIE
PIREX
RERTR
RR
SNF
STP
TRIGA
TEM
UTS
VPH
YS

144
144
144
145
146
147
149
153
158
158
159
160
161
161
166
166
166
169
170
173
173

Los Alamos National Laboratory


Low enriched uranium
Specifically, MTR is the Materials Testing
Reactor at Idaho National Laboratory.
Also used generically for materials test
reactors
Open Pool Australian Light water reactor
Oak Ridge National Laboratory
Oak Ridge Research Reactor
Post irradiation examination
Proton Irradiation Experiment facility
Reduced enrichment for research and
test reactors
Research reactor
Spent nuclear fuel
Special Technical Publication
Test, research, isotopes, general atomic
Transmission electron microscopy
Ultimate tensile stress
Vickers pyramid hardness
Yield stress

143

144

Performance of Aluminum in Research Reactors

5.07.1 Introduction
Aluminum alloys are generally too weak or have
temperature limitations that preclude their use in
reactors built to produce electricity, high-temperature
process heat, or marine propulsion. But in the milder
conditions in most research reactors (RRs) where
bulk water coolant temperatures are usually <100  C,
aluminum alloys are quite comfortable and are universally employed and have greatly contributed to
the success and longevity of the reactors. RRs are
those whose principal function is to generate neutrons for purposes of nuclear education and training,
production of medical and industrial isotopes, neutron activation analyses, neutron scattering studies,
and even semiconductor doping, neutron radiography, and food preservation treatments. RRs are also
employed to study basic radiation effects in materials
and as test beds for evaluating candidate structural materials and fuels/assemblies for power reactors. RRs come in many shapes, sizes, and types.
For descriptions of the various classes of RRs, see
http://www.world-nuclear.org/ and West.1 They are
generally low power, typically about a few kilowatts,
thermal, but range up to about 250 MW. According to
the recently updated list2 of worldwide RRs published
by the IAEA, a total of 674 RRs have been built in
57 countries, of which 234 are still operational, and
7 are planned or under construction. Two new ones
are OPAL, the 20 MW Open Pool Australian Light
water-cooled reactor, which opened at Lucas Heights,
Sydney, in April 2007, and the Russian 4 MW pooltype IRV-M2 commissioned in 2008.
This chapter is a review, more a tutorial, of the
behavior of aluminum alloys in water-cooled RRs.
It is a somewhat personal view, based on American
experience in the area. Because that experience
has been adopted in many countries and is still
influencing the current state of the art, this chapter
should be of interest outside the borders of the
United States.

5.07.2 Typical Applications


Aluminum is the material of choice for construction of many components in low-temperature
water-cooled-and-moderated RRs. Typical applications are the reactor tanks in open-pool reactors;
containment vessels in some sealed reactors; core
grids; pedestals; neutron beam tubes; cold neutron
source moderator vessels; shrouds to direct and

separate water flows; shuttles (rabbits) and aluminum filler powder used to convey isotope target
materials and test materials rapidly in and out of
the reactor via aluminum hydraulic and pneumatic
tubes; sheaths and finned tubing for stationary longterm isotope target rods; cladding for control plates/
rods; cladding and liners for reflector materials; cladding and thermal conduction filler for fuel rods/
plates; and temporary plugs for closing idle irradiation
facilities in and around the core. Applications outside
the reactor per se are in-pool tool extension arms;
transfer gates between pool sections; restraint baskets
in some shipping casks; support beams for pool covers;
and hot cells manipulator arms.
5.07.2.1 History of Aluminum
Applications in Research Reactors
Aluminum was at the forefront of the development
of nuclear technology. It has the distinction of being
the first nonfissile, non-neutron absorber class metal
used in the worlds first continuously operating
nuclear reactor, the X-10 Graphite Reactor at Oak
Ridge, TN. The Graphite Reactor became critical
on 4 November 1943, <1 year after Fermis demonstration of a self-sustaining nuclear fission chain in
the graphite pile at the University of Chicago on
2 December 1942. In Fermis experiment, the only
metals in the pile were natural uranium and the
cadmium-coated control rods. The pieces of natural
uranium (238U containing about 0.7 at.% 235U) and
uranium oxide were bare, placed in shallow depressions carved into the upper faces of the graphite slabs,
and cooled by convection of ambient air. The power
level was about 2 kW. The X-10 Graphite Reactor
pile3 was much bigger than the Chicago pile and
was designed to operate at 1 MW thermal power,
later upgraded to 4 MW. It was built to produce
pilot plant quantities of plutonium isotopes. The
Chicago pile had no shielding; the Graphite Reactor
was shielded by a 2.2-m thickness of high-density
concrete. Aluminum made its debut in the Graphite Reactor as fuel cladding to protect the highly
chemically reactive uranium from contamination
by air and graphite during the higher power and
longer fissioning periods and to safeguard it from
attack by water during subsequent radioactive decay
in underwater storage. In addition, it trapped the
more copious volatile radiation products resulting
from the longer irradiation exposures. These
aluminumclad pieces of natural uranium, called
slugs, were the forerunners of metalclad fuel

Performance of Aluminum in Research Reactors

elements. A slug was made by placing a solid cylinder


of uranium in a thimble-shaped aluminum can
25 mm diameter  100 mm long with a 0.75 mm
wall. A flat Al end cap was added, and the assembly
was passed through a die to force the can walls tightly
around the fuel. Surplus wall material was cut off
above the cap, and the cap was welded all around its
edge. These slugs were pushed end to end into the
reactor via round horizontal holes through the concrete face, which were aligned with 44 mm square
holes cut through the full 7.3 m width of the cubic
array of graphite blocks. The square holes were oriented on edge such that the slugs occupied the lower
corner, allowing cooling spaces around the slugs.
Cooling was simple: two large fans at the rear of the
pile sucked ambient air through the holes around the
slugs and discharged it up a tall chimney. The slugs
exited the pile at the rear face and were channeled into
a deep water canal where they were held until shipped
to hot cells for processing to extract the plutonium.
Some early problems4 were encountered in the slugs,
including faulty welds and blisters and formation of
an intermetallic UAl3 phase by interdiffusion at the
UAl interface, especially in the high-temperature
regions in the center of the reactor. The blistering
was attributed to fast-growing gas bubbles in the UAl3
phase. These problems were overcome by better
welding practice and the development of bonded
slugs as described next.
The next phase of exploitation of aluminum was
in the B reactor at the Hanford site in Washington
State, which went critical on 27 September 1944.
The B reactor was a scaled-up production model
of the Graphite Reactor designed to operate at
250 MW. At such power, forced air cooling would
have been inadequate. So the horizontal holes
were replaced with aluminum tubes in which
aluminumclad uranium slugs were cooled with
flowing water from the Columbia River. To improve
the transfer of heat from the uranium to the cladding, the spaces between them were filled with a low
melting Al12% Si eutectic alloy by melting the
eutectic in situ. A bonus of this treatment was that
it killed the formation of the UAl3 phase and associated blistering, presumably due to an inhibiting
effect of the silicon. The successes of these upgrades
established aluminum as a suitable material for use
in combined conditions of intensive irradiation and
a flowing aqueous environment. Aluminum became
more firmly entrenched in RRs with the development of advanced fuel elements, as described in
Section 5.07.4.

145

Before concluding the present subsection, another


lesser-known first for aluminum deserves mention.
It has particular relevance to the nuclear power
industry. It is not widely known that aluminum was
involved in the earliest demonstration of electricity
produced from steam made by boiling water in a
nuclear reactor. Normally, the heat from nuclear
fission in RRs is discarded, not used to produce
electricity. However, in August 1948, two staff members at the X-10 Graphite Reactor, Mansel Ramsey
and Charles Cagle, placed an aluminum can containing ten aluminumclad uranium slugs and some
water in a normally unused side channel of the reactor where it was exposed to reactor neutrons. The
trapped heat generated in the slugs boiled the water.
Steam from the process was piped to a small model
steam engine, rated at 1/1000 hp (0.75 W), which
rotated an armature mounted between the poles
of a permanent magnet. Sufficient electricity was
generated to light a flashlight bulb. The thermal
efficiency was estimated to be 2%. The Graphite
Reactor is now a National Historic Landmark and is
open to the public. A commemorative plaque and a
replica of the steam engine and coupled dynamo
from Ramsey and Cagles pioneering boiling water
power reactor are displayed in a small showcase
in the reactor lobby. The official first production
of nuclear electricity is credited to the lighting of
another bulb on December 1951 at the Experimental
Breeder Reactor-I, Arco, Idaho, now the Idaho
National Laboratory.

5.07.3 Properties of Aluminum


Heat removal and reduced generation of heat
are major considerations in the popularity of aluminum in RRs. Most of the energy released from controlled nuclear fission appears as heat. Much of the
heat, >80%, arises in the fuel from nuclear fissions.
However, a significant portion, 520%, is produced
in the nonfissile materials in the core and its surroundings by bombardment with particles emanating
from the fission reactions and from decay of fission
products. For power reactors, the heat is essential to
generate the electrical output. In the case of RRs, the
heat is a nuisance product; and the goals are to
minimize heat generation from the nonfissile materials in the system and to get rid of it from those
materials and from the fuel as fast as possible.
Hence, structural materials that create the least heat
and/or conduct it away the fastest are the most

146

Performance of Aluminum in Research Reactors

favored for RRs. In this regard, aluminum is outstanding. Generally, heat production is greater with
increasing material density and with decreasing
specific heat. It is increased by high cross-sections
for neutron absorption and scattering, which also
reduce reactor efficiency by stealing neutrons from
participation in fission processes. Heat removal rate
is larger with higher thermal conductivity. Therefore, construction materials with low density, high
specific heat, high thermal conductivity, and low
nuclear cross-sections offer the best prospects for
minimizing heat generation and maximizing heat
removal. In Table 1, the relevant properties for
aluminum are compared with those of other cladding and structural materials used in power reactors
and for uranium. All values are for room temperature or 100  C. The scatter in values for a given
parameter and material is due in part to sensitivity
to chemical composition and heat treatment, etc.
These variations do not mask the large differences
between Al and the other materials. The density of
Al is 1/21/3 of those of the other cladding materials, and only 1/7 that of U. Its specific heat capacity
is twice as high as the other materials. And its
thermal conductivity is 510 times greater than
the values for the other materials. Additionally,
its neutron capture and scattering cross-sections are
much smaller than those of the other materials, except
for nuclear-grade Zr. In that respect, it should
be remembered that in the early days when Al was
establishing its foothold in nuclear technology commercial Zr was contaminated with up to 3% of the
strong neutron absorber Hf. It was also inordinately
expensive.

Table 1

5.07.3.1

Practical Characteristics

Having attractive physical properties for reactor use


is of no merit if those properties cannot be exploited
in a practical manner. The commercial and economic
attributes of aluminum that encourage its deployment in RRs are: It is ductile, plentiful, cheap, and
light weight. It is castable, machineable, and weldable, and it can be shaped readily by conventional
processes of rolling, forging, extrusion, drawing, and
cupping. It has good aqueous corrosion resistance due
to near-insolubility in water and formation of a passive, self-restoring surface film of hydrated aluminum
oxide. It is nonmagnetic and nonsparking. Although
aluminum is inherently weak, it can be strengthened
by cold work, solid solution hardening, and precipitation treatments. It has an fcc crystal structure and
no crystallographic phase changes. Its crystal structure is near isotropic, ensuring that it will not suffer
damaging directional thermal expansion and radiation growth like those exhibited by graphite and the
hexagonal metals Mg and Zr. It does not form stable
embrittling hydride phase(s) as Ti and Zr do. At low
temperatures, it has no ductile-to-brittle transition.
On the contrary, it is somewhat special in that at
cryogenic temperatures, where it gains strength, it
often gains ductility too. This combination of no
hydride phase, outstanding low temperature properties, and low neutron cross-sections make aluminum
the prime material for building cold neutron sources.
Another attractive feature is that pure aluminum has
no long-lived radioisotopes. The major source of
immediate radioactivity is from decay of 24Na produced via 27Al(n,a)24Na, decaying by g-emission

Relevant properties of reactor materials

Material

Aluminum
Zirconium
Austenitic steel
Ferritic steel
Uranium

Density
(kg m3)

2700
6490
8000
7900
1900

Specific heat
(J kg1 K1)

887963
254285
377565
440494
111167

Thermal
conductivity
(W m1 K1)

160230
840
1121
1742
1128

Melting
point ( C)

660
1852
1425
1525
1132

Emod (GPa)

70
8898
190201
200210
176208

CTE, lin.
(106 K1)

23
5.7
16
12
13.9

Nuclear
cross-section
(barns)
sabs

ss

0.23
0.19
3.0
2.5
7.6

1.5
6.4
10
11
8.9

Sources: Matos, J. E.; Snelgrove J. L. Selected Thermal Properties and Uranium Density Relations for Alloy, Aluminide, Oxide, and Silicide
Fuels; IAEA- TECDOC-643, International Atomic Energy Agency, Vienna, 1992; pp 119, article Appendix I-1.1 in Research reactor core
conversion guidebook Volume 4; Fuels (Appendices IK). Lide, D. R., Ed. CRC Handbook of Chemistry and Physics, 86th ed.;
Taylor & Francis: Boca Raton, FL, 20052006; Gale, W. F.; Totemeier, T. C., Eds. Smithells Metals Reference Book, 8th ed.; Elsevier and
ASM International, Amsterdam and Materials Park, OH, 2004; Cverna, F., Ed. ASM Ready Reference: Thermal Properties of Metals;
ASM International: Materials Park, OH, 2002.

Performance of Aluminum in Research Reactors

with a half-life of 15 h. In alloys, long-lived radioactivity arises from decay of isotopes produced from
alloying elements and residual impurity elements
present in the aluminum, primarily 65Zn, 51Cr, 59Fe,
with half-lives of 250, 28, and 45 days respectively.
So if low residual radioactivity is an objective it can
be met to a large extent by avoiding alloys containing
significant quantities of Zn, Cr, and Fe.
Aluminum is not without its shortcomings. It has a
low elastic modulus and low melting temperature.
The former means that in their annealed conditions
aluminum alloys have low strengths compared with
annealed austenitic steels, Zr, and bcc metals. However, aluminum can be hardened by various treatments
as described in Section 5.07.3.2. However, the low
melting temperature of 660  C imposes operating
temperature limits of 100150  C, which are homologous temperatures of 0.40.45Tm where lattice
vacancies are mobile and can invoke susceptibility
to creep and stress relaxation. Even without imposed
stresses, the strength condition of prehardened alloys
can become compromised at temperatures above
150  C because of the possibility of thermal overaging
as described in Section 5.07.3.2 Aluminum has poor
abrasion resistance. It can be sensitive to localized
galvanic and pitting corrosion. It is prone to liquid
metal embrittlement, particularly Hg. Difficulties
may be encountered in obtaining leak-tight fusion
welded joints for hi-tech applications, mainly due
to porosities resulting from solidification shrinkage
(volumetric change) and dissolved gases, in particular,
hydrogen.5 In addition, aluminum does not undergo a
color change on heating, and during manual welding
may melt abruptly without warning, allowing overheating that can cause excessive sagging and dropthrough of the weld bead. The advent of a solid-state
joining process, namely friction-stir welding,6 has
largely overcome those welding troubles.
5.07.3.2 Alloy Types, Temper
Designations, and Tensile Properties
There is no universally embraced international standard system for defining the types and conditions
of aluminum alloys. The International Standards
Organization does have classifications for aluminum
and its alloys, but most countries adhere to their
own systems. The system followed in the United
States of America is ANSI H35.1-1990, instituted
by the American National Standards Institute and
supported by the Aluminum Association and ASM
International. The ANSI system and the US alloys

147

covered by it are described in reference,7 which is an


excellent source of aluminum data; it includes a short
list of alloys for other nations and their national
designations. The ANSI system is used herein. In its
entirety, it is a morass. Here, it is outlined just to the
extent that is necessary to provide an uninformed
reader with enough details to understand the nomenclature and the various processing treatments and the
upper service temperature limits those treatments
impose for maintaining stability of the processed
materials.
The system has two classifications, one for
wrought alloys and one for cast alloys. Only the
wrought alloy classification is described here. Briefly,
it is an eight-character code consisting of two groups
of four characters separated by a hyphen. The first
four characters are all numerals and they identify the
alloy group by chemical composition. There are eight
aluminum alloy groups. The first digit of the first alloy
group is 1, which represents alloys with a minimum of
99.00 wt% aluminum. In this group, the major foreign
elements are Fe and Si, which are really residues from
the aluminum extraction process and will be found to
various degrees in all aluminum alloys. The next three
digits in the group identify specific alloys in the same
series, and the group as a whole is denoted the 1xxx
series, often vocalized as the one-thousand series.
The other seven alloy series are 2xxx (major alloying
element, Cu), 3xxx (Mn), 4xxx (Si), 5xxx (Mg), 6xxx
(Mg Si), 7xxx (Zn), and 8xxx (other element).
An upper case X preceding the series identifier
numeral indicates an experimental alloy.
The second group of four characters in the eightcharacter designation represents the temper condition, that is, the heat treatment or degree of cold
work. The first character of the four-character temper condition is an upper case letter representing a
type of treatment. The other three characters are digits
indicating variations within the treatment. There are
many temper treatments. Only the three treatments
most likely to be encountered in RR materials are
described here. They are O for the fully annealed
condition, H for a strain-hardened condition, and
T for a precipitation-hardened condition. The
O condition is attained by annealing the material at
about 340  C then slowly cooling it. There are no
specified variations of the O condition. The H temper
is more complex. The first digit after the H is a 1, 2,
or 3. H1 signifies strain-hardened only. H2 is strainhardened and partially annealed. H3 is strain-hardened
and stabilized by a low temperature heat treatment.
The second digit, that is, the one following the H1,

148

Performance of Aluminum in Research Reactors

H2, or H3 designation is a number between 1 and


8 and is the degree of reduction in thickness or crosssectional area given to the alloy in its fully annealed
condition to bring it to the desired strength level.
Level 8 corresponds to a maximum reduction of
about 75%. Level 1 represents approximately oneeighth of 75%, 2 is two-eighths, and so on. The third
digit, if used, implies a variation of the two-digit
temper. Partial annealing for the H2 condition is
applied to products that are strained beyond the
desired final amounts and are then brought back
to the needed strength level by the partial anneal.
Stabilization heat treatment for the H3 condition
is applied to those products that, unless stabilized,
would gradually age-soften at room temperature.
Partial annealing also inhibits age softening. This
tendency for softening of some cold-worked aluminum alloys at room temperature is important
because such recovery requires the involvement of
mobile lattice vacancies and/or self-interstitial
atoms that promote climb and rearrangement of the
cold work dislocations. It indicates the occurrence of
atomic movement at room temperature, which, as we
shall see shortly, is a factor affecting the development of radiation damage in aluminum.
In addition to hardening by cold work, aluminum
can be strengthened by solid solution treatment
and by precipitation hardening. Only two alloying
elements, Mg and Li, have sufficient solubility
(several %) at room temperature to provide significant solid solution strengthening. AlLi alloys are
not recommended for reactor use because natural
Li contains about 7.5% 6Li, which has a large crosssection for transmutation to 3H and 4He, both of
which can be highly detrimental to aluminum.
So the only solid solution-hardened alloys available
for reactor use are the 5xxx (AlMg) series. Other
metallic elements, principally Cu, Si, and Zn, have
little or no solubility in aluminum at room temperature but are modestly soluble at higher temperatures
near the melting point. This latitude permits considerable strengthening of such alloys by quenching-andaging, also known as precipitation hardening. The
ANSI designations for the precipitation-hardened
T conditions comprise ten subdivisions, T1T10.
For all T treatments, the alloy is heated to a temperature of 500540  C to dissolve segregated alloying
elements, followed by a rapid quench into cold water,
which gives an unstable supersaturated solid solution.
Precipitation is achieved by allowing the material to
sit at room temperature for periods of weeks called
natural aging (tempers T1T4) or by artificial

aging at temperatures of 160190  C for times of


624 h (tempers T5T10). Flattening or straightening treatments may be applied before or after the
aging treatment and are indicated by numbers in
the third and fourth character positions. The temper
conditions for aluminum alloys most often encountered in RRs are T4, T6, and T651. A T651
condition indicates a material that has been artificially aged then subjected to a light stretching
operation insufficient to change its mechanical
properties from those of the T6 condition. Of the
precipitation-hardened alloys, the 6xxx series hardened by precipitates of Mg2Si is by far the most
popular for RR service. The 6061 alloy in its T6
and T651 conditions has been approved for service as a class 1 nuclear components material under
the Boiler and Pressure Vessel Code of the American Society of Mechanical Engineers, Case N-519.8
Two types of precipitation-hardenable wrought
aluminum alloys, the 2xxx series (AlCu) and the
7xxx series (AlZn), both of which can be heat
treated to greater strengths than the 6xxx alloys,
are not usually found in nuclear reactors. Some
2xxx alloys are prone to aqueous pitting corrosion
or may release Cu ions to the coolant that could
be deleterious to other materials in the reactor
such as stainless steel. The 7xxx series alloys
have too low ductility and are the most difficult to
weld. Their high zinc contents will cause high
radioactivity.
Unlike the cold-worked 1xxx alloys that can
undergo recovery at room temperature, the
precipitation-hardened alloys are thermally stable
at temperatures up to about 150  C provided they
have been given appropriate natural or artificial
aging treatments. However, exposure to higher temperatures will cause overaging and associated reduction in mechanical strength. This softening is
illustrated in Figure 1 for 6061-T6 alloy after heating
to various temperatures for various times and testing
at room temperature.9 It can be seen that softening is
promoted by both time and temperature. For times
up to 1 h, softening commences at about 200  C and
is substantial but incomplete at about 370  C. For
a longer exposure of 1000 h, the softening begins
around the aging temperature, indicated by the
down-pointing arrow, and is essentially complete at
temperatures between 260 and 300  C. The data in
Figure 1 are for specimens reheated without load.
If reheating occurs under loads sufficient to induce
creep and stress relaxation, the softening temperatures are pushed downward.

Performance of Aluminum in Research Reactors

149

350
Softening effects of reheating temperature and time on room temperature
properties of 6061-T6Al (originally aged 18 h at 160 C)

300

200

1000 h

30 min

6 min

150
50
100
Elong., 1000 h

25

Elongation

50

Elong., 6 and 30 min

100

200
Reheat temperature (C)

% Elongation

0.2% yield stress (MPa)

YS
250

0
400

300

Figure 1 Softening effects of reheating temperature and time on room temperature properties of 6061-T6 aluminum
(originally aged 18 h at 160  C). Data from Structural Alloys Handbook, 1989 ed., Vol. 3, Battelle Memorial Institute, Columbus,
OH, 1989; p. 14.

Table 2

Example alloys and their room temperature tensile properties

Alloy

Composition (wt%)

YS (MPa)

UTS (MPa)

Elongation (%)

1100-O
4032-T6
5052-H34
6061-T651

<1 (Fe Si)


12.2Si, 1Ni
2.5Mg, 0.25Cr
1Mg, 0.6Si, 0.28Cu, 0.2Cr

35
320
210
280

90
380
260
310

40
9
16
17

Source: Aluminum Standards and Data, 10th ed.; The Aluminum Association: Washington, DC, 1990.

Table 2 gives typical tensile properties of various


Al alloys employed in RRs. The weak 1100-O alloy
is simply annealed commercial purity aluminum with
no deliberate alloy additions; it is hardenable to an
H condition by cold work if so desired. The 4032
alloy is a eutectic composition of Si in aluminum that
has been solution-treated and aged to create finely
divided precipitates of Si; this alloy is used principally as a filler wire to improve the weldability of
aluminum alloys. The 5052 alloy is a solid solution
alloy of 2.5% Mg with a small amount of Cr added to
control grain size and strengthen the grain boundaries. The particular 5052 alloy in the table has been
work hardened to a 4/8, or half-hard, condition
before stabilization. The 6061-T651 alloy has been
solution treated and artificially hardened by precipitates of Mg2Si phase and its precursors, then given a
mild stretching treatment.

5.07.4 Fuel Elements


The most crucial and demanding applications of
aluminum in RRs are in the fuel elements. There it
is used inside the fuel element as a thermal conduction matrix in which a dispersion of fuel particles is
embedded and as a cladding material that protects
the fuel from corrosive attack by the cooling water,
retains fission products, and transfers heat from the
fuel and matrix to the coolant. As RRs matured,
considerable improvements were made in the fuel
elements. Most fuel in RRs is no longer unalloyed,
metallic a-phase uranium whose orthorhombic crystal structure is prone to severe radiation growth and
swelling leading to distortion and cracking. It has
been ousted by more stable and more isotropic uranium compounds that can also better accommodate
fission gases with minimum swelling. The most

150

Performance of Aluminum in Research Reactors

common ones are U3O8, UAlx, where x can be 2, 3, or 4


but is usually considered10 to be a mixture of 3 and 4,
and U3Si2. There is also a hydride fuel, UZrH1.6,
which is used exclusively in the open-pool TRIGA
(test, research, isotopes, General Atomic) types where
the fuel is in the form of slugs comprised of particles of
U dispersed in the ZrH1.6 phase (see Chapter 3.12,
UraniumZirconium Hydride Fuel). Originally, the
TRIGA slugs were sheathed in aluminum, which has
now been replaced with stainless steel or nickel alloy.
However, TRIGA reactors still contain other aluminum components.
There is no outstandingly superior aluminum
cladding alloy. The most common aluminum cladding
alloys are 1100 and the stronger 6061. Other alloys
have been investigated in neutron irradiations,11
namely 5052; X800N, where N is 1, 2, or 3 and
whose compositions are Al1Ni1Fe; and two sintered aluminum powder alloys, M257 and M470,
which were fabricated by ball milling flake powder
of 1100Al in air until it contained a dispersion of 6%
and 10% Al2O3, respectively, then consolidating by
pressing, sintering, and hot rolling. The Mxxx alloys
were deemed to be no better than 1100 and 6061
types. They are more difficult and expensive to
make and harder to weld than regular melted-type
alloys. In Europe, particularly in France, two preferred
alloys are AlFeNi, a relative of X8001 with the composition 1Fe1Ni1Mg, and AG3-NET, a 5xxx-type
with 2.53.0Mg and low residuals. The greatest
concern for cladding is its corrosion behavior (see
Section 5.07.5).
A feature of RR fuels is that they are much more
highly enriched in 235U than those in power reactors:
1293% versus about 2.5%. Drivers for raising the
235
U levels were extended fuel cycles; the growing
demands for industrial and medical isotopes, particularly 99Mo the parent of the all-important medical
diagnostics tool 99mTc; and the need for higher neutron fluxes for increased production of the heavy,
transuranic isotopes. The use of highly enriched
uranium (HEU) meant higher heat generation and
required improved means of removing the heat. The
solution was the development of dispersion fuels
in which particles of the enriched fuel were distributed
in a matrix of thermal conductor material, all compressed together in sealed aluminum cans. The
thermal conductor is aluminum powder, usually a
1xxx-type, often atomized powder of better than
99.5% purity and particle size <100 mesh (150 mm
maximum, 2348 mm mean). Atomized powder particles are denser, pour more easily than milled flake

powders, and have less low conductivity surface oxide


per unit volume. The aluminum matrix may occupy
more than 50 vol% of the fuel/aluminum mixture.
A huge advance in fuel element morphology and
heat removal efficiency took place when Eugene
Wigner designed his thin, curved fuel plates for the
high flux Materials Testing Reactor (MTR) built at
Arco, Idaho. A thin plate has a number of advantages
over cylindrical slugs. The rolling treatment used to
produce the plates from a fuel slab, or from a dispersion of fuel particles in aluminum matrix powder,
sandwiched between two aluminum cladding sheets
gives superior mutual contact of cladding, matrix, and
fuel for improved heat transfer paths to the cladding.
The much larger surface-to-volume ratio of plates
provides more efficient heat transfer to the coolant,
thus permitting higher fuel loadings per unit volume.
The benefit of a curved fuel plate is that any buckling
and bowing in the plate due to irradiation will be
focused in the direction of the radius of curvature.
Thus, in a fuel element comprised of a stack of
curved plates restrained at their edges and separated
from each other by cooling channels of the same
width as the thickness of the plates, any such distortions will be accommodated cooperatively in the
radial direction without unacceptable narrowing of
the cooling channels. An MTR fuel element contained
18 plates each about 72 mm wide and about 727 mm
long bent to a curvature of 140 mm radius in the width
direction. The plate thickness was 1.27 mm including a
minimum cladding thickness of 0.25 mm on each face.
The plate edges were brazed into sturdy side panels
to seal the plate edges and impart rigidity to the
assembly. The water gap was 1.27 mm. The cladding
and side panels were made from 1100Al; the Al
brazing alloy contained about 13% Si.12 This assembly was then enclosed in a long, rectangular aluminum box fitted with end fixtures for remote handling.
The end fixtures were castings of Al7% Si. The
reactor core was built from groups of such elements
assembled upright in rectangular arrays held together
by aluminum grid plates. Refueling was done from
the top, and any element could be replaced by a box
of the same size containing a reactor experiment or
materials for isotope production, or a beryllium
reflector or a control rod. These MTR-type boxed
fuel elements in open grid core arrangements performed very well and became very common for RRs.
To satisfy demands for higher power densities and
more sophisticated tailoring of local neutron fluxes,
the next advancement in aluminumclad fuel elements
was the development of upright, annular

Performance of Aluminum in Research Reactors

elements using curved fuel plates in which the fuel


particles may be required to be graded in concentration across the thickness and width. Beryllium reflectors surrounding the annulus direct neutrons from
the fuel back to the hollow center, or trap, of the
annulus where reactor experiments and isotope targets are placed. The Be also creates additional neutrons from (n, 2n) reactions. Vertical holes
bored through the reflector allow passage of cooling
water and house irradiation experiments. Two highperformance beryllium-reflected reactors using annular fuel elements are the High Flux Isotope Reactor
(HFIR) at Oak Ridge National Laboratory (ORNL),
rated at 100 MW thermal and currently running at
85 MW, and the Advanced Test Reactor (ATR) at
Idaho National Laboratory, rated at 250 MW but
lately operating at 100125 MW. The cores of these
reactors are of uncommon designs and deserve comment. The ATR core13 is 1.22 m diameter and 1.22 m
high. It contains a continuous serpentine-like wall of
fuel elements looped around nine flux traps each
about 120 mm diameter arrayed in a square 3  3
grid. In plan view, the wall forms the shape of a
four-leaf clover. It fully embraces the central flux
trap and the four corner ones. The other four traps
lie just outside the wall; each is tucked in between the
junctions of two leaves and is about half wrapped by
the wall. At each corner lobe, there are four shim
control cylinders just outside the wall and six shim
rods at the neck of the wall inside the cloverleaf.
These controls allow each of the four lobes to be
run at different power levels simultaneously, as
needed by the experiments in the traps. The remainder of the space in the core is occupied by blocks of
Be reflector containing numerous experiment holes.
The wall is built14,15 from 40 individual wedgeshaped fuel elements, each containing 19 curved
fuel plates. The cross-sectional area of an element is
a 45 sector of a circular annulus. Its outer arc, plate
#19, has a radius of 137 mm and an arc length of
100.9 mm. Its inner arc, plate #1, has a radius of
77 mm and an arc length of 54.1 mm. The 19 fuel
plates are attached by roll-swaging to 6061-T6Al side
panels 64.6 mm wide  1257 mm long. Within the
elements, the curved plates are concentric with the
circumferences of the traps. The plates are 1.27 mm
thick except for #1 and #19, which are thicker. The
water gap is 1.98 mm. The ATR fuel is UAlx enriched
with 235U to 93%, dispersed in a matrix of Al powder
and clad with 0.38 mm thick 6061-OAl.
The HFIR core16 is more compact, about the size
of a small trash can, into which are packed 540 fuel

151

plates in quite a different arrangement than in the


ATR. The core diameter is 435 mm and it is 791 mm
tall. It has a single central flux trap, 129 mm diameter.
The fuel is granules of U3O8 enriched with 235U to
93% and embedded in Al powder. The cladding is
6061Al. The core consists of two concentric annular
arrays of involute-curved fuel plates, as shown in the
sketch of a radial segment in Figure 2. The black
region in the fuel plates is the fuel dispersed in its
Al matrix; the white area is Al filler. There are 369
plates in the outer annulus and 171 in the inner
annulus. The plates are 610 mm high with widths for
the inner and outer annulus plates of 94 and 81 mm,
respectively, before bending. The plate thickness and
coolant gaps are 1.27 mm, as in the MTR-type elements. The two annuli are fabricated separately and
are united when loaded into the reactor. In addition to
the unique radial-like orientation of the fuel plates, the
fuel particles are uniquely distributed in the plates. To
minimize the radial peak-to-average power density
ratio, the thickness of the compacted fuel mix is varied
along the arc of the involute curve as seen in Figure 2.
This shaped region is backed by filler Al containing
no fuel particles. For the inner annulus, the filler
powder backing the shaped fuel region contains

Al filler
Al + 41 wt% U3O8
1.27 mm
Coolant
channel

Outer
annulus
sidewalls

Al + 30 wt% U3O8

Al filler +
B4C poison

1.27 mm

Figure 2 Horizontal section through a small segment


of the HFIR core showing fuel plate curvatures and
fuel distributions in the plates. Modified from
Binford, F.T.; Cramer, E. N. The High Flux Isotope
Reactor; A Functional Description, Vol 1B, Illustrations;
ORNL-3572 (Rev.2); Oak Ridge National Laboratory:
Oak Ridge, TN, 1968.

152

Performance of Aluminum in Research Reactors

particles of B4C burnable poison. Two concentric


cylindrical control plates clad in Al are located immediately surrounding the core. Outside the control
plates are four concentric cylindrical Be reflectors.
Because beryllium generates copious quantities of
helium and tritium from neutron irradiation, it tends
to swell and crack, particularly at the faces of its high
neutron flux regions. To retain chips spalled from
these surfaces, the reflector and any penetrations in it
are clad with aluminum. Four horizontal 6061Al beam
tubes and numerous vertical holes penetrate the
reflector.
Like most dispersion-type fuel plates, the HFIR
and ATR plates are fabricated by what is called
a picture frame technique. This utilizes powder metallurgy methods to disperse the fuel particles uniformly in the Al matrix and press the mixture into
a hard rectangular compact. The rigid compact is
placed in a window of the same size cut in an Al
slab or frame, which is usually the same alloy as the
cladding. Sheets of cladding material are welded to
the top and bottom faces of the filled frame and the
assembly is hot rolled through a large reduction in
thickness to ensure that the cladding is fully bonded
to the fuel charge and the frame. After verifying the
location of the fuel charge, the rolled plate is cold
rolled to flatten it and bring it to the specification
thickness. It is then given a final anneal at 500  C
to reveal any blisters and rolling defects in the
cladding surfaces. After verifying the location of the
fuel region, the plates are blanked to finished size in
a press.
Of course, it is not as simple as that. Strict quality
assurance standards have to be met, and at every stage
in the operation, there are numerous inspections
and rigorous sizing and confirmation tests. To reproducibly obtain the graded fuel distributions in the
HIFR plates, a special procedure was developed.17,18
A custom-designed contoured auxiliary die plate
is mounted over the cavity of the powder press to
facilitate mounding of the fuel/matrix powder mix in
a semicylindrical hump. Another auxiliary die plate is
added to allow filler powder to be leveled on top
of the humped fuel charge. This duplex charge is
withdrawn into the press cavity, the auxiliary dies
are removed, the rectangular punch is inserted into
the die mouth and pressure applied, and the charge
is consolidated in a single cold pressing operation.
The HFIR fuel plates are bent to the desired involute
shape in an elastomer-faced punch and die press.
They are welded into the cylindrical inner and outer
sidewalls of the fuel elements. The sidewalls are

machined from extruded-type 6061 aluminum tubing


in the T6511 temper. Twenty-seven equally spaced
circumferential weld grooves are turned on one face
of each sidewall, and slots are milled at prescribed
depths and angles on the other face of the wall. The
weld grooves intrude a short way into the slots.
The fuel plates are slid into the slots and properly
spaced with the aid of temporary Teflon separators.
The plates are machine welded in place through the
grooves. A 4043Al weld filler wire and an argon shield
gas are used. End fixtures machined from 6061Al
tubing are welded to the ends of the elements, and
final machining and inspection are conducted.
These multiplate fuel elements are a testimonial
to designer ingenuity and superb fabrication skills,
and the versatility of aluminum. Manufacturing these
fuel elements is not only painstaking but also expensive. In year 2007, each HFIR element cost $1 M.19
It is replaced after its regular lifetime operating
cycle of 26 days. With so much effort and cost
invested in it, a rejected element is a severe financial
loss. The specifications and acceptance standards are
so high that the chances of producing a fuel element
completely free of specification violations are very
low. The first 30 000 fuel plates suffered a rejection
rate of 12%, and of the first 45 fuel assemblies, only
4 inner elements passed the final inspection.20 However, the degrees of severity of the violations were
all minor or were correctable. With waivers, all 45
elements were accepted and gave exemplary service.
After operation of the first 60 fuel cores at the
full design power level of 100 MW, 4 of them were
autopsied.21 No significant faults were found. The
in-reactor performance of these complex aluminumbased fuel elements has been incredible, surpassing all
expectations.
Development of RR fuels and fuel plates is
continuing. Concerns over the possibilities of nuclear
weapons proliferation and terrorism led to establishment of the Reduced Enrichment for Research and
Test Reactor (RERTR) program at Argonne National
Laboratory.22 The goal of RERTR is to eliminate
the use of highly enriched uranium (HEU) in RRs
by converting to the use of low enriched uranium
(LEU). HEU is defined as uranium that has the
fraction of the fissile isotope 235U greater than 20%,
LEU is less than 20%. Historically, RRs have used
enrichment levels of 235U up to 93%. RERTR
is intended to be achieved without impairing the
safety and performance of the reactors and/or jeopardizing the production of important isotopes,
and at minimum cost for changes in fuel elements.

Performance of Aluminum in Research Reactors

In some RRs with modest uranium enrichment and


low power levels the RERTR LEU goal was met by
diluting the fuel with natural uranium. For many of
the high performance RRs (HPRRs) that must retain
their 235U levels and cannot tolerate the burden of
added 238U without excessive operational penalty, the
RERTR dilution can be achieved by replacing the
HEU fuel with LEU compounds or alloys containing
higher fractions of U. To that end, the initial focus of
RERTR was on the development of uranium silicide
fuels, U3Si and U3Si2, dispersed in aluminum and
clad with aluminum.23,24 While this move has been
successful for many RRs it is not sufficient for the
most demanding HPRRs. For them, attention has
turned away from dispersion fuels to monolithic alloy
fuels where higher U densities are attainable. The goal
is to develop fuel plates built from foils of LEU alloy,
250500 mm thick, clad with aluminum.19,2527 In
order to prevent buckling and cracking of the foil
during multiple rolling and recrystallization treatments and to inhibit radiation growth and warping,
there must be just enough alloying metal in the
U to stabilize it in its isotropic g-phase. Several
alloying metals are suitable, but the field of contenders has been reduced to the UMo system. A 90%
LEU-10% Mo alloy currently holds the best prospects. Some serious hurdles are recognized. Interdiffusion between the cladding and the fuel foil during
annealing and in-reactor exposure encourages the
formation of reaction layers of uraniumaluminum
compound(s) with low thermal conductivity and low
resistance to growth of fission gas bubbles. Such
layers threaten the integrity of the fuel/cladding
interface. Development of these layers is retarded
by additions of Zr or Ti to the fuel, or Si in the
cladding. When Si is incorporated in the cladding, it
is found to segregate at the fuel/cladding interface,
acting like a diffusion barrier. Thin film diffusion
barriers of Si, Zr, and ZrN applied directly to the
surfaces of the fuel foil by co-rolling and thermal
spraying have done well in reactor tests. The current
hot roll bonding processes used for attaching cladding to dispersion fuel plates may not be fully
adaptable to barrier-coated foil fuels. Other bonding
methods such as hot isostatic pressing are under
investigation. For HFIR plates, where the foils must
be tapered in both width and length and have involute shapes, fitting and bonding diffusion films and
cladding to the fuel foil on a mass production scale
is a challenge. Hot roll bonding will not work
because the foil and the cladding will not deform
to the same extent and will result in nonuniformly

153

thick cladding, and shear deformation during rolling


may damage the diffusion barrier. It is recommended19 that the tapered foil, bent to its involute
shape and with an adherent diffusion barrier, should
be prepared separately then sandwiched in shaped
recesses in two full-length clamshells of cladding
of appropriate thickness and bonded over all mating
surfaces. Alternatively, if the clamshells can be made
from a two-ply Al sheet, like the commercial OneSide Alclad, the inner layer of, say, 1100Al, could
contain the ingredients for a diffusion barrier. The
hot isostatic pressing route may then allow bonding
and barrier filming in a single operation and in batch
mode. If burnable poison cannot be incorporated in
the fuel foil, it may be possible to accommodate it
in the inner cladding layer with the diffusion barrier
components.
The corrosion behavior of the Al cladding on alloy
foil fuel elements will need to be explored thoroughly. A penetration of the cladding will probably
be more serious than one in current dispersion fuel
plates because the alloy fuel will likely be more reactive and soluble in water than the dispersant-type
intermetallic and refractory fuels.

5.07.5 Corrosion
Metallic corrosion, the removal of metal atoms from
the metal surface by the electrochemical action of
the environment, has many forms: uniform, galvanic,
pitting, grain boundary, crevice, etc. Uniform corrosion and pitting are the types of most interest to RRs.
The greatest worry is the aluminum fuel cladding
where the environmental conditions are most aggressive and where an unexpectedly high corrosion rate
might breach the cladding and allow release of highly
radioactive fission products throughout the water
system. Pitting corrosion is the major form of attack
on the cladding of spent fuel elements during longterm storage in water basins.28 Herein, the focus is on
uniform corrosion of cladding.
Aluminum is a very reactive metal. In dry air,
it combines with oxygen to form an adhesive,
self-healing Al2O3 film that retards further oxidation
at the metal surface. Such films are usually quite thin,
tens of nanometers, usually described as amorphous.
Films formed in moist air and water are much thicker,
1 mm or more. The water-formed reaction films developed on aluminum cladding are variously described
as hydrated oxides and hydroxides, and oxidehydrates, and they are generically referred to as

154

Performance of Aluminum in Research Reactors

oxide films. In HPRRs, the films grown on the fuel


cladding may be 2050 mm thick.29,30 The most common corrosion products28,30 reported on aluminum
cladding are boehmite, a crystalline monohydrated
aluminum oxide, Al2O3H2O, and bayerite, a crystalline trihydrated oxide, Al2O33H2O. At temperatures
below about 77  C, the boehmite phase is formed
preferentially but may transform to bayerite with
continued immersion. At temperatures above 77  C
and below 100  C, a pseudoboehmite structure
grows, which may age to other hydrated oxide forms
or retain its pseudoboehmite structure. Between 100
and 400  C, crystalline boehmite will form. A gelatinous boehmite is the chemical precursor of both of the
crystalline hydroxides.30 The mature hydroxides are
normally white color but other hues have been
reported and may stem from absorption of Fe, Cr, Ni,
or other metal ions leached from steels in the reactors
or in the corrosion test loops.
The corrosion film is both the reaction product
and the medium through which the corrosion process
occurs. Whether corrosion is governed by ingress of
O and OH ions through the film to the metal surface
or by egress of Al ions to the film/water interface,
it is expected to be diffusion controlled. Thus, all else
being equal, an increase in film thickness should
lower the corrosion rate by increasing the diffusion
length, and vice versa. Therefore, the corrosion rate
should be parabolic with time and have an Arrheniustype dependence on temperature. Moreover, ideally,
if all the corroded metal was retained in the corrosion film, if the chemical composition and physical
structure of the film were constant throughout
the thickness, and if all of the film was retained
on the metal, the film thickness would be proportional
to the amount of metal corroded. Alas, such ideality
does not prevail. The corrosion process is confounded by a number of interacting factors, including
the following: there is a one-sided heat flux on the
cladding; the corrosion film is a thermal insulator
compared with the Al cladding, so the temperature
of the film will increase with thickness; the film may
not be of uniform composition and/or structure;
the film is soluble to some extent in water, and its
solubility is strongly susceptible to the pH of the water,
which is related to water composition; the film is
subject to erosion in flowing water and to spontaneous
spallation above some uncertain thickness, about
50 mm in one case.31 And to further complicate the
situation, there is wide variation in the ways the corrosion tests are conducted and in the parameters that
are measured.

The tests may be carried out in open cups, closed


autoclaves, vented autoclaves, closed loops, bypass
loops, or on used fuel plates. Evaporation or consumption of the water may require that it will need
to be periodically replaced or its volume readjusted.
Except in in-reactor tests and loop test systems with
bypass monitoring and adjustment of the water, the
chemistry of the water may change substantially during the test. Few corrosion rates for cladding materials are measured directly. They are usually derived
from measurements of the thickness of the corrosion
film. A thickness measurement gives the thickness of
the film adhering to the substrate at the time of the
measurement. It will not include film that has been
dissolved and/or eroded away. On a spent fuel element, it may include film that has formed in a storage
pool over time periods much longer than it experienced in-reactor, and with no forced cooling. During
preparation for post irradiation examination (PIE) in
a hot cell, the spent element is no longer fully
immersed. It gets hot and has to be periodically
sprayed with water to cool it. It has been opined21
that the resultant steaming and thermal cycling may
cause more corrosion than in-reactor operation and
underwater storage. There is no guarantee that the
density and the composition of the film will be invariant through the film thickness. On the contrary, multilayer films are more common than not. Almost all
films have a thin, monolithic base in contact with the
Al surface, presumably associated with the ubiquitous
air-formed Al2O3 film. On top of this base, there may
be one to three distinct layers. Some films contain
pores or are cracked. Only the films on irradiated fuel
elements have been exposed to the effects of neutron
irradiation and radiolysis of the water. The way in
which the film thickness is measured may be questionable, too. At least six different methods are used,
viz.: (1) Scaled measurements by optical or scanning
electron microscopy of metallographically polished
and etched cross-sections of the corroded test piece;
(2) micrometer measurements of the thickness of the
test piece before corrosion and after the corrosion
product is removed by electrolytic polishing until the
shiny metal is seen; (3) weight gains of coupons with
film in place; (4) weight losses of coupons after
removal of the film; (5) acoustic and eddy current
measurements with instruments calibrated against
accepted standard films; and (6) temperature increases
measured with thermocouples attached to the noncorroding back surface of the test piece during the test,
and related to spot film thicknesses measured metallographically after the test.

Performance of Aluminum in Research Reactors

A neglected aspect of film measurements is that


almost all of the measurements have been made on
specimens that, deliberately or unavoidably, were
dried at room temperature or at 100  C32 before the
measurement was attempted, or before the measuring
instrument was calibrated. Until recently, nobody
seems to have determined whether such drying treatments will shrink, spall, crack, or otherwise alter the
bulk film. The gelatinous surface layer that precedes
the crystalline corrosion films will almost certainly
be altered during dehydration. It is not uncommon
for test coupons to be dried, weighed, and placed
back in the test for the next exposure period, and
so on until the termination of the campaign. That
was the method used in one seminal laboratory test
study.33 The first periods in the full exposure sequence were the shortest ones, 1 or 2 days, and they
always showed the largest weight gains, usually
6090% of the total weight gained during the full
duration of the test, which was about 22 days. Weight
gains after the first period were linear with time and
were relatively minor. That is not parabolic corrosion
behavior. The abrupt change in weight gain indicates
that something happened during the first interruption of the test that set the scene for a sudden switch
from an initial rapid corrosion rate to a subsequent
constant low rate. Likely, the first drying treatment
irreversibly altered the structure and permeability of
the hydrated film. Recent autoclave tests34 on AlFeNi
alloy reinforce that suspicion. It was demonstrated
that during a 34-day test, interruptions made every
7 days to remove, dry, weigh, descale, dry, reweigh,
and replace the test piece in the autoclave with
refreshed water for the next exposure period had
serious consequences to the corrosion kinetics. Without interruptions, the inner and outer oxide layers
were twice as thick, the weight gain was 26% higher,
and the amount of metal removed from the substrate
was 23% higher.
Some efforts have been made to correlate film
thicknesses with corrosion rates.3133 Tests made
under controlled conditions in a corrosion loop31
found that the thickness of the boehmite film on
1100, 6061, and X8001 alloys was about 1.4 times
the depth of penetration into the aluminum regardless of changes in test parameters that changed the
film thickness, as long as there was no stripping or
spallation of the film. Using a literature value for the
density of boehmite, it was estimated that about 70%
of the corroded Al remained in the adherent film and
about 30% was lost to the coolant. When spallation
did occur, which was usually above a film thickness of

155

50 mm, the 1100 and 6061 alloys always showed localized attack of the aluminum under the spalled area,
whereas the X8001 alloy showed only uniform attack
under all conditions. This correlation was for a
closed, single set of data. It should not be considered
representative of all data and situations. Other data
by some of the same authors,32 where the principal
variables were temperature and flow rate, showed
that the ratio of corrosion product retained to the
weight of metal corroded ranged from a high of 0.54
at a low temperature of 170  C and flow rate of
6.19.5 m s1 to a low of 0.08 at 290  C and 29 at
32.6 m s1. Another source29 quotes a retention level
of 5080% of the oxide on the cladding surface, but it
may be citing Griess et al.31 In general, the relationship between film thickness and corrosion rate is not
well established.
Film thicknesses from laboratory tests31,3538 display power law growth with exposure time, but the
time exponents, preexponential factors, and activation energies differ from one experimenter to another
and may be applicable only to the particular set of
data from which they were determined. Nevertheless,
the laboratory tests have established that the corrosion films are sensitive to a number of interacting
factors. They include the temperature and surface
condition of the cladding; the heat flux density on
the cladding; and the temperature, pH, flow rate, and
purity of the water. In RRs, water purity is controlled
by filtration and ion exchange systems; it is also
linked to pH. With regard to pH, the films will
dissolve if the water is strongly acidic (pH < 4.5)
or strongly alkaline (pH > 8.5); films are most stable in the range 5.06.5, the closer to 5.0 the better.
The pH of reactor water and spent fuel storage pool
water tends to converge toward the desired range
by carbonic and nitric acids formed from CO2 and
N absorbed from air. It can be maintained close
to 5.0 by controlled additions of nitric acid. The
strongest increase in film growth is from increase in
temperature, and the controlling temperature is that
at the hydroxide/water interface.31 To lesser extents,
increased heat flux density and water flow rate
will raise the film growth rate. For the alloys 1100,
6061, and X8001, which all corroded alike until spallation occurred,31 the rate of oxide formation at a
heat flux of 1.58 MW m2 was about half of that at
3.136.31 MW m2, other conditions being the same.
At coolant flow rates in the range 7.613.7 m s1, the
rate of accumulation of the corrosion product was
the same for all three alloys. Corrosion rates measured
on the insides of 1100Al production tubes39 were

156

Performance of Aluminum in Research Reactors

found to be unchanged by water velocities in the range


0.3055.58 m s1. Reduced water temperatures will
reduce the film growth rate.
Despite differences in strengths, compositions,
and microstructures, the alloys 1100, 6061, and X8001
all seem to have similar corrosion behavior under
similar conditions.31,32 Spalling tends to introduce
local attack in 1100 and 6061 but not in X8001.31
The AlFeNi alloy shows good performance at
temperatures up to 250  C in autoclave tests35 and
in-reactor exposures40 at temperatures below 120  C,
but it has not been tested under high heat fluxes.
AG3-NET cladding on U3Si2 dispersion fuel plates
undergoing in-reactor tests failed41 at a heat flux of
5.5 MW m2. The cladding was swollen and breeched
by a combination of a very thick corrosion film and
subfilm intergranular corrosion. Cross-section X-ray
spectroscopy analyses showed that oxygen had penetrated intergranularly all the way through the cladding to the meat. The corroded cladding was
interesting in other ways. The outer oxide layer was
monolithic and was exceptionally thick, 100 mm.
Directly beneath it was a region about 80 mm thick
containing many round 30 mm size pores. Below the
porous region, the grain boundaries were enriched in
Mg and oxygen. The plates were intended to reach a
temperature of about 180200  C at the exterior surface of the cladding and 220240  C in the fuel.
Temperatures estimated from the thick corrosion
layers were >300  C for the water/corrosion film
interface and >400  C for the fuel meat. The AG3NET alloy has a history of intergranular cracking in
beam tubes and other structures in the Reacteur Haut
Flux at the Institut Laue-Langevin in France.
Although that cracking occurred at high fluences,
the irradiation temperatures were low. Such low temperature intergranular cracking is a sign of pending
weakness in the alloy and does not bode well for
applications at higher temperatures as in fuel
cladding.
The influences of neutron flux and radiolysis of
water are unclear. These parameters are omnipresent
in RRs and we might imagine them to strongly affect
aqueous corrosion of fuel cladding by damaging
the cladding and its corrosion film and by altering
the activity of the water. One researcher42 writes that
reports of neutron flux effects on the hydroxide films
are few and there is disagreement; he claims that
the opinion of most (Russian) researchers is that
neutron irradiation decreases, rather than increases,
the corrosion rate. Effects of radiolysis are uncertain.
According to Golosov,42 one Russian authority argues

that radiolysis may either accelerate corrosion by


facilitating cathodic processes or reduce corrosion
by promoting anodic passivation. Data from laboratory
corrosion loop tests without radiation fields seem to be
fairly compatible with data from irradiated fuel elements in terms of oxide thicknesses, compositions,
and pH effects. There are no outlandish differences
that would immediately draw attention to radiation
effects. At least, none that has been strong enough to
insist that loop tests should be repeated in irradiation
fields. A similar conclusion was reached for aqueous
corrosion of aluminum process tubes in production
reactors.39 Therefore, irradiation effects must be
modest at worst. However, there are some troubling
reports that seem to indicate large effects of irradiation fields in nonreactor conditions. Sindelar et al.43
studied 6061Al coupons exposed to moist air at 150
and 200  C, with and without exposure to a 60Co g
source at 1.8  106 R h1. Weight gains and film
thicknesses were measured. The corrosion product
was patches of loosely aggregated, randomly oriented
1 mm size boehmite crystals sitting on a thin monolithic base layer, even at 100% relative humidity
where the product was permanently under a film of
water. g-Irradiation seemed to double the weight
gains and increase the film thicknesses by a factor of
10. There was substantial surface blistering of the
base layer, attributed to hydrogen gas. The paper
provided no details of the experimental conditions.
Enquiries to the authors produced a lengthier publication44 with the missing details. Those details cast
grave doubt on the conclusions drawn from Sindelar
et al.43 In particular, the experiments with the g-field
were made under radically different conditions
than those without the field. Specimens for the
g-irradiations were sealed in small stainless steel
cans of just 78 ml and each can represented an uninterrupted test for a given exposure period of 1, 4, 8,
and 12 weeks. The tests without the g-field were made
in stainless steel autoclaves of volume 37 850 ml for
15 unequal exposure periods totaling about 30 weeks.
At the end of each period, the specimens were
removed, dried, weighed, and replaced in the autoclave with a new charge of water. In light of the
effects of interruptions described in Wintergerst
et al.,34 the effects of g-irradiation described in
Sindelar et al.43 and Lam et al.44 are inconclusive.
In the other work,45 Al coupons of undeclared composition and condition were exposed to static brackish water of pH 89, at undisclosed temperature
for periods of up to 30 days, with and without low
dose irradiations with neutrons from a 252Cf source

Performance of Aluminum in Research Reactors

(1010 n m2 s1) and, separately, g-rays from a 60Co


source at 15 Sv h1 (1.6  104 R h1). Corrosion was
determined from weight losses. It was not stated
whether the specimens were recycled from one period
to the next. The neutrons and the g-rays had the same
effects and to the same degree; they promoted formation of a grayish layer on the specimen surfaces; they
reduced the weight losses by 2530%; and they almost
eliminated severe pitting corrosion displayed by the
unirradiated specimens. None of these three reports
mentioned whether radiation heating was a factor.
The laboratory loop tests have verified the expectation that the corrosion film is a thermal insulator compared with the Al cladding, and they have provided31
a thermal conductivity value of 2.25 W m1 K for
boehmite, which is a factor of 70100 less than Al.
However, it is not always ascertained whether a particular film is boehmite or bayerite or a mix of both.
No thermal conductivity value is available for bayerite. When insulating films build on the Al cladding
of heat sources like the fuel and long-term heavy
isotope targets, the temperatures of the sources and
their claddings or containers will rise. This temperature rise will increase the corrosion rate and
the growth rates and dissolution rates of the corrosion
films. In HPRRs, a side effect of an increase in cladding temperature by the adherent corrosion product
is the threat of plate buckling.31 As described earlier,
the strengths of the cladding and Al fuel matrix can
be decreased significantly by tens of degrees increases
in temperature, and creep rates will increase. If an
insulating corrosion film increases the temperature
gradients between the center thickness of the fuel
plate and the surface of the film, and between the
fuel-loaded portions of the fuel plates and their cooler
frames, the plates may distort. If the distortion is not in
phase from one plate to the next, it might perturb the
coolant flow and accelerate the temperature changes.
Griess et al.31 envisaged that the insulation provided by
the corrosion-product film might be more of a limitation on the use of aluminumclad fuel elements in
high flux reactors than is corrosion damage per se and,
in the worst case, may lead to burnout of parts of the
fuel plates. Fortunately, that prophecy has not been
fulfilled. Serious plate distortion has not been a widespread issue. One case of plate distortion is described
in Shaber and Hofman.30 Plate buckling found in some
MTR elements12 was blamed on new design changes.
It is recommended30 that new fuel elements
should be prefilmed with a hydroxide film to reduce
the rate of in-reactor buildup of the corrosion layer.
Tests32 with 1100, 5154, 6061, and X8001 alloys at flow

157

rates in the range 6.120.4 m s1 found that preexposure of the test pieces to water at 250300  C for 24 h
in an autoclave caused a significant improvement in
corrosion resistance, but not at higher flow rates. The
ATR elements are pretreated30 by immersing them in
water for 48 h at 180  C and pH 5.0. In the early days
of HFIR operation, the new fuel elements were often
stored in the reactor pool water for up to 3 months
before being placed into service. This immersion
resulted in the formation of a rather thick, gelatinous,
corrosion product film on the element surfaces.21 In
an attempt to avoid that condition, some of the elements were pretreated by boiling them in deionized
water for 24 h to produce a thin, boehmite film on the
surfaces of the elements before they were placed into
service. When the pretreated elements were used, the
coolant flow rate was found to gradually decrease
and the pressure drop across the elements gradually
increased during the reactor fuel cycle. No significant
damage was caused. Changes in coolant flow rate
and pressure drop were not observed when the reactor was operated with non-pretreated fuel elements.
Metallographic examinations of cross-sections of
the spent fuel plates revealed much thicker corrosion
films on the pretreated plates. Pretreatment of the
HFIR fuel elements was discontinued. Most RRs do
not practice pretreatment of their fuel elements. It is
proposed here that because of the seemingly large
effects of dehydration on retarding subsequent film
growth as discussed earlier, at least one in-reactor
trial should be made of a prefilmed fuel plate with
a dehydration step or a low temperature baking
treatment added. A drying treatment might also be
worthwhile for a newly spent fuel element before it
enters pond storage.
What we really need to learn from corrosion measurements and film thickness data is the thickness of
uncorroded Al cladding remaining on the fuel element at the end of reactor service, and whether that
thickness will be sufficient to continue to seal the
spent fuel through further corrosion expected during
cool-down storage in water basins. That is, we need
reliable corrosion rates pertinent to the particular
application. Corrosion product thickness data are
invaluable in identifying and characterizing the
major factors governing corrosion and the interplay
between them, but they are meaningless to corrosion
rates if a reproducible relationship between film
thickness and corrosion rate is not established. We
need predictability. To that end, efforts are underway
to derive predictive models for film thicknesses40,46
and corrosion rates.42 These models are in their

158

Performance of Aluminum in Research Reactors

infancy. Because they lack a large body of consistent


data to draw on, the authors must make many assumptions, fittings, and correlations to derive constants,
correction factors, adjustment factors, and augmentation factors. With so much flexibility built into
the prediction equations, it is not surprising that the
authors can find good correlations with selected data
from measurements made on spent fuel cladding.
This is not intended as a criticism of the modelers;
it is a reflection of the paucity of input data. Reliable
modeling is essential. But it needs reliable input data.
Data obtained from recycled test coupons should
either be excluded from the models or modeled as a
separate category. To be generically applicable, film
thickness models and corrosion rate models should
attempt to merge in a complementary manner.
In low power RRs where convective flow is
sufficient to take care of cooling and water quality
is adequately controlled, problems from corrosion
films formed on the aluminum cladding and on other
aluminum components elsewhere in the reactor are
uncommon. In HPRRs, the most prominent corrosion
problems were those in the early days of operation that
caused a milky turbidity of the coolant and a white
deposit and increase in surface radioactivity on all
surfaces exposed to the coolant. The turbidity was
identified as a fine suspension of boehmite, and the
g-radioactivity was consistent with decay of 24Na, both
effects attributable to corrosion/erosion of the fuel
cladding. The turbidity is created by increase in the
cladding temperature due to the warming effects of
the hydroxide film. In turn, the temperature of the
coolant in immediate contact with the film is raised.
This increases the solubility of aluminum oxide in the
immediate volume of coolant. When this small volume
moves on and merges with the cooler bulk coolant,
the solubility falls and much of the dissolved film is
released as a particulate suspension. Particles of film
washed directly into the coolant by erosion of the
cladding due to the high coolant flow rate contribute
to the turbidity. Since turbidity ensues when the concentration of aluminum in the bulk water exceeds the
solubility of the aluminum oxide, turbidity problems
are brought under control by tuning demineralization
treatments to remove dissolved aluminum from the
bulk water and by reducing the degree of dissolution
through adjustments in pH to between 5.1 and 5.4
where aluminum oxides have minimum solubility.
In-reactor pitting corrosion and galvanic corrosion have not been serious problems. Pitting of Al,
which is encouraged by the presence of ions of Cu,
halides, and bicarbonates, is more serious in storage

pools where poorer water chemistry and nearly stagnant water conditions may exist, but diligent monitoring and control of water chemistry can mitigate
these concerns. Intergranular corrosion has not been
a problem in RRs, but it could become an issue at
high irradiation temperatures as evidenced by the
AG3-NET cladding described earlier.
Overall, aluminum cladding has given very good
service in water-cooled RRs and continues to do so.
The major variables influencing the corrosion process
(es) and corrosion products are fairly well identified
except for effects of irradiation. More data from spent
fuel elements are needed to guide and refine models
for predicting film thicknesses and corrosion rates.

5.07.6 Radiation Effects


5.07.6.1

Basics

As in other metals, irradiation of Al with neutrons or


charged particles introduces lattice vacancies, selfinterstitial atoms, and transmutation products that
evolve into radiation damage microstructure, which
causes swelling, radiation hardening, and loss of ductility. Radiation damage effects in aluminum differ
from those in most other metals in two respects. One
is that the radiation damage is affected strongly by a
solid transmutation product, silicon, discussed more
in Section 5.07.6.2.3. The other is that Al is much
more tolerant of radiation effects than most other
metals. At least, it is for irradiations conducted at
ambient temperatures. Neutron irradiation of Al
at temperatures between 25 and 100  C does not
induce detectable radiation hardening until the fast
neutron fluence exceeds about 1  1024 n m2, whereas
in Fe and Zr, radiation hardening is detectable at
fluences two to three orders of magnitude less than
that.47 Moreover, even when Al is radiation hardened
at 25100  C, it still retains significant ductility when
compared with considerably reduced ductilities in Fe
and Zr. This delayed display of radiation hardening
exists despite the fact that the number of atomic
displacements per atom in Al are about twice as
many as in other metals at the same fast fluence,
which is brought about by the lower displacement
threshold energy for Al. The larger part of Als better
tolerance of radiation damage is owed to its low
melting temperature, which makes its homologous
temperature high compared with those for Fe and Zr.
At room temperature, the homologous temperature
of aluminum is 0.32Tm, versus 0.175 for austenitic
steel, 0.17 for ferritic steel, and 0.26 for a-Zr if

Performance of Aluminum in Research Reactors

referred to the a ! b-transition temperature of about


860  C, or 0.14 if referred to the m.p. of b-Zr, 1852  C.
In general, noticeable thermally induced movement
of lattice vacancies will occur in metals at homologous temperatures above about 0.3Tm. Because of that
movement at room temperature in Al, there will be a
greater loss of radiation-produced vacancies and of
interstitials to mutual recombination, resulting in less
nucleation of the point defect clusters that are the
seeds of damage microstructure, hence less radiation
hardening.
A feature of radiation damage in polycrystals is
the absence of point defect damage microstructure at
grain boundaries and other incoherent interfaces.
Point defect clusters and voids do not develop on
the boundaries, and damage-free zones are formed
on each side of a grain boundary. This denuding may
be difficult to see in many high melting point metals
irradiated at temperatures below 100  C because
the zones are narrow. In Al, the high Tm allows
development of wide and conspicuous denuded
zones. Incoherent interfaces are comprised of structural dislocation networks and high equilibrium
concentrations of vacancies that make the boundaries deep sinks for absorption and recombination of
freely migrating point defects. They are pulled in
from the near regions of the butting grains, leaving a
volume of matrix straddling the boundary that is
diminished in radiation-produced point defects.
Due to the greater mobility of the interstitials and
the bias of dislocations for absorption of interstitials,
the zone deprived of interstitials is wider than that
for the vacancies. This creates an unbalanced concentration of vacancies at the rim of the denuded
zone. Therefore, vacancy clusters are encouraged to
form in that rim and they become more numerous
and/or larger than those in the grain matrix. Impurities and transmutation products are also drawn
into the grain boundary, but they are not annihilated
there; they accumulate. If they are largely insoluble,
as H, He, and Si are in Al, they will precipitate and
grow there, the gases as bubbles and the Si as particles
or films. Within the grains, the gases will stabilize
embryo clusters of vacancies and facilitate their
growth into voids as long as there is an excess of
vacancies. Some of the Si will attach itself to the
voids. Grown-in dislocations in the grain interiors
are also sinks for point defects.
Diffusion and binding of freely migrating point
defects and solute atoms are important for understanding and analyzing radiation effects. Some useful
parameters for pure Al are:48,49

159

 The threshold energy for atomic displacements is


25 eV compared with 40 eV or more for most other
metals.
 The self-diffusion rate, Dsd A(Qsd/kT), where A is a
constant and the activation energy Qsd is the sum
of the formation energy of a vacancy, Evf , and the
migration energy of the vacancy, Evm .
 For temperatures between 298 and 580 K, A is
1.6E  5 m2 s1 and Qsd is 1.3 eV.
 For temperatures between 570 and 923 K, A is
2.0E  4 m2 s1 and Qsd is 1.48 eV.
 Evf 0.6 eV.
 Evm 0.7 to 0.88 eV, deduced from Evm Qsd  Evf .
 Eif the formation energy of an aluminum interstitial atom, >3 eV.
 Eim the migration energy of an aluminum interstitial atom, 0.1 eV.
 Evb-s the binding energy between a vacancy and a
solute atom and is considered to be <0.1 eV for Ag,
Cu, Mg, Zn, and Si.
 Eib-s the binding energy between an aluminum interstitial atom and a solute atom. No
values are available, but for the solute to reduce
b
needs to
radiation damage microstructure Eis
m
m
b
E

E
.
be >Ev i
vs
Solutes that seem to reduce radiation damage structure most strongly at concentrations of 100 appm
are Cr, Cu, Mn, Ti, V which have the largest negative
lattice misfits, defined as (a-a0)/fa0, where a0 is the
lattice parameter of pure Al, a is the lattice parameter
of the alloy, and f is the atomic fraction of solute.
No relationship is found between degree of radiation damage and thermal diffusion rates of solutes.
5.07.6.2

Microstructures

Examples of the damage microstructure in highpurity Al irradiated to a fast neutron dose of


3.5  1024 n m2 at 50  C are presented in Figure 3.
The two photographs are not the same field. The
one on the left shows typical radiation-produced
dislocation loops. Some of the smaller spots are
particles of radiation-produced Si. The other photograph is tilted to put the dislocations out of contrast and reveal the voids more clearly; they are
facetted. The loops and voids are of order 30 nm
diameter. In ferritic steel, austenitic stainless steel,
and Zircaloy-4 alloys irradiated under similar
temperature and neutron fluence conditions as in
Figure 3, the radiation damage microstructure
is resolvable as 12 nm black dots.47 The loops in
Figure 3 are not faulted. Nobody has reported

160

Performance of Aluminum in Research Reactors

(a)

0.1 mm

(b)

0.1 mm

Figure 3 Dislocation loops (a) and voids (b) in high-purity aluminum after irradiation at 50  C to a fluence of
3.5  1024 n m2 (E > 0.1 MeV).

faulted loops or stacking fault tetrahedra in neutron-irradiated Al, presumably because the
very high stacking fault energy (SFE) of Al,50,51
about 160200 mJ m2, would inhibit faulting. Yet
faulted loops, and multilayered loops, are formed in
aluminum during electron bombardment in a highvoltage electron microscope52 and in thin foils that
are water-quenched from temperatures near the
melting point then aged.53 The occurrence of faulting in these cases is increased with increasing aging
temperature and impurity level. It is possible that
the SFE may have been reduced by contamination
occurring through the foil surfaces.
5.07.6.2.1 Fluence

Experiments54 made to establish the minimum fast


neutron fluence for the onset of visible radiation
damage microstructure in high-purity Al foils irradiated at <60  C indicate a threshold in the vicinity
of 3  1023 n m2, E > 1 MeV. The specimens were in
two conditions, annealed and cold worked, and were
of three thicknesses, 100 nm, 12.7 mm, and 76.2 mm.
None of the 100 nm foils showed damage structure
for exposures up to 5  1022 n m2. At a fluence of
3  1023 n m2, the 12.7 and 76.2 mm foils, both
annealed and cold worked, showed many small
loops, about 25 nm size in low concentration and
spotty distribution. After a dose of 1.6  1024 n m2,
the loop concentration was higher, but there was little
or no change in size. At both of these doses, the
annealed specimens contained loosely tangled
dislocation lines at higher density than is characteristic of well-annealed high-purity Al. These dislocations were kinked, and some of them that moved
while under observation in transmission electron
microscopy (TEM) examination were seen to have

been pinned at the loops. In annealed specimens, the


grain boundaries had well-defined denuded regions
about 0.35 mm wide on each side at a dose of
1.3  1024 n m2, and in the regions next to the
denuded zones many of the loops were large and
there were dislocation segments among them. Some
of these segments spanned the denuded zones. These
denuded regions are wider than the 100 nm thick foils
in which no radiation damage was found, suggesting
that the radiation point defects had migrated from
those foils. For cold-worked specimens, nothing
was said about the cold work dislocations. It is quite
possible that considerable recovery of the coldworked dislocation structure occurred before and
during irradiation. At grain boundaries in the coldworked material after irradiation to 6  1023 n m2,
a few loops and some dislocation segments are present within a distance of about 0.35 mm from the
boundary. In the adjoining regions, loops and dislocation segments are present in high concentrations, and
many of the loops are large, >0.1 mm, and they encircle smaller loops or are kinked by smaller loops.
Evidently, the dislocation segments are portions of
growing loops. The tangled dislocation lines in the
annealed specimens at the lower doses probably arose
from growth of the earliest loops. No radiation voids
were seen in these experiments. It was speculated that
the loops were vacancy-type, growing from collapse of
vacancy clusters produced by the irradiation. French
studies55,56 have corroborated and enlarged on the
heterogeneous nature of evolution of early damage
microstructure and the roles of dislocations in Al.
With increasing fluence at constant irradiation
temperature, the loops evolve into dislocation lines,
and voids and Si precipitates arise, which increase in
concentrations and sizes. The voids are larger and

Performance of Aluminum in Research Reactors

161

0.1 mm

Figure 4 Denuded grain boundary and associated void


enhanced regions in 49 purity aluminum after irradiation at
50  C to a fluence of 3.4  1026 n m2 (E > 0.1 MeV).

less numerous than the Si precipitates. The voids are


facetted and so are the larger particles of Si. Most,
if not all, of the voids have a facetted Si particle
attached to the outside of one facet of the void.
The majority of the Si particles are not attached to
voids. The grain boundary denuded regions are not
enlarged, but the voids at their rims are exaggerated,
as illustrated in Figure 4. An unpleasant potential
consequence of these sheets of large voids is that
in the event of an unexpected overload, they may
provide paths for premature failure by a tearingalong-the-dotted-line-type separation.
5.07.6.2.2 Temperature

Raising the irradiation temperature coarsens the


damage microstructure and decreases the degree
of radiation hardening for a given dose. At 150  C,
Figure 5, the dislocation structure almost disappears;
there are fewer voids, but they are larger than
at 50  C and are strongly facetted, and many of
them are very much elongated.57 Particles of
radiation-produced Si are attached to one face of a
void, usually at the narrow end of elongated voids.
Freestanding Si particles are likewise facetted and
elongated, some in ribbon shapes. The denuded
regions straddling grain boundaries are wider, about
1 mm each side. For annealed materials irradiated at
temperatures above 150  C, certainly above 200  C,
no dislocation-type radiation damage microstructure
or voids are produced; coarse Si particles are seen. In
cold rolled pure Al, some large cavities remained58
after irradiation at 220  C. During postirradiation
annealing59 removal of damage microstructure was
slowed by impurity content and by higher doses. For
1 h anneals of 1100Al, void swelling began recovery at
200  C and was almost complete at 300  C where gas

1 mm

Figure 5 Voids in high-purity aluminum irradiated at


150  C to a dose of 2  1025 n m2 (E > 0.1 MeV). Right:
Preinjected with 3 appm He. Reproduced from Farrell, K.;
Wolfenden, A.; King, R. T. Radiat. Eff. 1971, 8, 107114, with
permission from Taylor and Francis.

bubble swelling intervened. Void ripening preceded


void elimination. Si precipitates ripened and were
prevalent on grain boundaries.
5.07.6.2.3 Transmutation products

The gases helium and hydrogen produced from (n, a)


and (n, p) reactions of fast neutrons with lattice atoms
considerably affect the development and effects of
radiation damage structure by encouraging the nucleation of voids, dislocation loops, and bubbles.57,60,61
Helium is insoluble in Al. It binds strongly with
vacancies and has very limited mobility. Hydrogen
is almost insoluble at ambient temperatures, becoming more soluble with increasing temperature. In the
lattice, it is mobile even at room temperature. The
degree of promotion of voids and loops by the gases
decreases with increasing irradiation temperature,
and the promotion of bubbles increases. In the righthand micrograph of Figure 5, it is evident that even
for a high irradiation temperature of 150  C, the
presence of just 3 appm He implanted at room temperature has boosted nucleation of voids. The string
of voids in the field is not on a grain boundary. It
probably marks the position of a grown-in dislocation
present during implantation that has climbed away
during irradiation. At higher irradiation temperatures58 or during postirradiation anneals,59 the gases
form bubbles. The levels of helium and hydrogen
produced in Al are not widely different from those
in other metals. For the purpose of comparison we
can use the production rate tables of Gabriel et al.62

162

Performance of Aluminum in Research Reactors

to estimate the levels of He and H produced in Al,


Fe, and Zr during a 1-year exposure in the high flux
peripheral target positions of the HFIR core. There,
on the horizontal mid-plane at current operating
conditions, the annual fast neutron fluence will be
2.4  1026 n m2, E > 0.11 MeV, and the thermal fluence will be 4.5  1026 n m2. In Al, this fast fluence
will generate gas concentrations of 11 appm He and
63 appm H, and in Fe there will be 5 appm He
and 95 appm H. In Zr, which is noted for its low
cross-sections, only about 0.25 appm He and 5 appm H
would be produced. The atomic displacement levels
for the three metals can also be calculated. They are
36 dpa for Al, 18 dpa for Fe, and 17 dpa for Zr.
The larger dpa level in Al is due primarily to its low
effective displacement energy, 25 eV versus 40 eV for
Fe and Zr.
There are three interesting reactions with thermal
neutrons that produce gases from the foreign elements Li, B, and Ni, which may be present in some
Al alloys. The spatial distributions of gases from these
sources are each different. Lithium has high solubility in Al and generates a uniform distribution of
gases. Boron and Ni are insoluble and they produce
localized concentrations of gas. Lithium is not a common impurity in Al, but there is a commercial series
of AlLi alloys developed for their lightweight highstrength properties. Laboratory-made AlLi alloys
have been used to study radiation hardening, helium
embrittlement, and swelling.63,64 Natural Li contains
7.5% of the 6Li isotope, which has a capture crosssection of about 950b and decomposes to He and
tritium via the reactions 6Li nth ! 7Li ! 4He 3H.
The present writer65 made an Al0.052 wt% Li
alloy using 69 high-purity Al and Li enriched to
98% with 6Li, and irradiated it to a dose of
5.5  1025 n m2, E < 0.0253 eV and 2.2  1025 n m2,
E > 0.1 MeV at 55  C. About 9599% of the 6Li was
burnt up to produce about 2200 appm each of He and
tritium. The atomic displacement level was about
3 dpa, not including any displacements from the
recoiled gases. The effects of these high levels of
gases were striking, see Figure 6. The insert is an
enlarged view of the matrix cavities. Compared with
irradiated pure Al control specimens, the concentrations of matrix cavities were increased 1000-fold,
and their sizes decreased tenfold; dislocation densities
were increased tenfold. Most grain boundaries were
crammed with large bubbles, many so interconnected
that it was difficult to obtain thinned foils for TEM
examination because the grain boundaries were eaten
away before much thinning of the grain interiors

0.1 mm

1 mm

Figure 6 Modification of void structure by very high


helium and tritium levels from burnup of 6Li. Reproduced
from Farrell K.; Houston J. T. Combined Effects of
Displacement Damage and High Gas Content in Aluminum,
ORNL-TM-5395; Oak Ridge National Laboratory: Oak
Ridge, TN, May 1976. Also available in Proceedings of
International Conference on Radiation Effects and Tritium
Technology for Fusion Reactors, Gatlinburg, TN, Oct. 13,
1975, U.S. Department of Commerce CONF-750989,
Mar 1976; pp. II-209II-233.

occurred. The grain boundary in Figure 6 is one


with a low concentration of bubbles. Hardness measurements gave a Vickers pyramid hardness (VPH) of
137 MPa for the annealed, unirradiated Al and the
alloy, and 382 and 902 MPa for the irradiated specimens. In bend tests made in air and liquid nitrogen
(LN), the unirradiated materials and the irradiated
pure Al were bent through full circles without rupture. The irradiated alloy broke with an audible crack
and with no detectable plastic strain. Fracture was
accompanied by release of tritium. The fracture surfaces displayed 100% intergranular failure. These
are incredible hardening and embrittling effects of
the gases. Electron microcopy examination of carbon
replicas taken from the fracture surfaces showed huge
irregular interconnected bubble cavities. Failure
occurred by plastic tearing of the small areas of intact
grain boundaries between the cavities. Postirradiation
annealing treatments caused the appearance of a coarse
distribution of large facetted matrix cavities superimposed on the small matrix cavities, and with
no denuding of the surrounding small cavities. These
enlarged cavities were frequently associated with large
silicon particles that grew concurrently during the
anneals. An anneal at 500  C showed incipient disintegration of the specimens and TEM foils could not

Performance of Aluminum in Research Reactors

be obtained. It was postulated that the large cavities


grown during annealing were tritium bubbles.
Al often contains trace quantities of B in the
form of small B4C inclusions. Natural B contains
19.8% 10B, which has a large neutron capture
cross-section of 3835b, producing Li and He via
10
B nth ! 11B ! 7Li 4He. The range of the
recoiled He atoms is about 5 mm, and the He is
segregated in a well-defined band in a halo around
the parent inclusion. The larger Li atom has a smaller
range and is soluble; it forms a diffuse halo. At low
irradiation temperatures and low doses, the halos are
very prominent because of heavy decoration with
dislocation loops. As the dose increases, the loops
grow and move off leaving dense halos of voids, especially for the He halo. The writer has seen hundreds of
these halos. Most of them were circular or near circular, with an occasional cigar shape, depending on the
shape of the mother particle. Most were isolated
randomly, but some were in groups or were strung
in chains on a grain boundary. One is illustrated in
Figure 7. This particular halo is slightly squashed,
following the elliptical contour of the central particle.
The dark region of the outer halo is actually filled
with small cavities, resolvable at higher magnification. Where a halo intercepts a grain boundary the
voids seem to disappear, but during annealing they
become visible as bubbles on the boundary that grow

1 mm

Figure 7 Damage halos around a suspected B4C particle


in 1100-OAl irradiated to 2.9  1026 n m2 at about 55  C.

163

faster than those elsewhere in the halo. Such highly


heterogeneous distributions of transmutant gas have
been perceived more as a novelty than as a possible
threat to the integrity of the host material. This
attitude may be unwarranted. A highly localized concentration of helium in a patch on a grain boundary
could be a prime site for premature helium embrittlement at stresses and temperatures below the ranges
for normal helium embrittlement elsewhere in the
specimen. A spongy helium halo that intercepts the
surface of Al cladding may provide a potential site
for initiation of local corrosion. For these reasons, it
might not be a good idea to consider placing particles
of B4C burnable poison in single-layer cladding on
monolithic LEU fuel plates; a better location would
be in the inner layer of a two-layer cladding.
Some of those considerations apply to He produced from Ni in Al. It comes from the 59Ni isotope,
which is not found in natural Ni. The 59Ni must
first be created from 58Ni that comprises 68.1% of
natural Ni. The two-step process66 to yield the He is
58
Ni nth ! 59Ni; 59Ni nth ! 56Fe 4He. Helium
generation via this route does not scale linearly with
time. It is slow to start while the 59Ni accrues, then
it increases as the square of the fluence. It is favored
by long-term exposures or strongly thermalized neutron spectra. Only trace quantities of Ni are found
in most Al alloys except the X8001 and AlFeNi-type
alloys, which contain a nominal 1 wt% Ni. These
alloys were developed for cladding because early
laboratory corrosion tests indicated they might have
better corrosion resistance than existing cladding
alloys. Trials of the X8001 have not shown superior
performance. The alloying elements in X8001 are
insoluble in the solid alloy and form intermetallic
inclusions that are malleable and become deformed
and extended into stringers during unidirectional
rolling and extrusion processing. The helium atoms
formed from the Ni in the stringers are recoilimplanted into the near-matrix regions surrounding
the stringers where they accelerate local formation of
voids and dislocations. The recoiled 56Fe atoms cause
extra dpa locally.67 These He-enriched regions are
not as obvious as those around B4C inclusions. They
are indicated by higher concentrations of voids, and
the emergence of more numerous He bubbles during
postirradiation annealing. This localized damage
offers an explanation of a hitherto inexplicable puzzle found in the corrosion response of X8001 alloy.
A characteristic of extruded X8001 tubes undergoing
aqueous corrosion in reactors is that smooth shallow
troughs or discontinuous ruts lying in the direction of

164

Performance of Aluminum in Research Reactors

the tube axis are created in the corroded surfaces.68


Such troughs have not been reported in laboratory
corrosion tests of X8001. It is suggested here that the
troughs are stringer beds left when the Ni-rich stringers are eased out of the surface by selective corrosion/erosion of the more highly damaged He-rich
matrix at the stringers.
Localized enhancement of He at Ni-rich stringers
is also believed to play a major role in the occurrence
of axial cracking of the X8001 cladding on HFIR
long-term isotope target rods.69,70 The target material is a cylindrical compact of actinide oxides in
an Al powder matrix, 6.3 mm diameter  14.5 mm
long, each jacketed in 1100Al. The meat contains
about 10% porosity to accommodate fission gases.
A target rod consists of 35 jacketed capsules stacked
in a tube of X8001 alloy that is hydrostatically compressed around them to form the outer cladding. The
tubes are made by extrusion and have six equally
spaced longitudinal fins. Before the target slugs are
loaded into the tubes, most of the fins are machined
off, leaving short lengths of fins at several locations
along the active length of the rod. The loaded target
rod is slid into an X8001 tubular sheath with hexagonal ends, known as a hex can. The groups of remnant
fins along the length of the rod act as spacers that
centralize the rod in the hex can and maintain
an annular water-cooling channel around the rod.
A bundle of 31 sheathed target rods just fills the
vertical HFIR trap. Cooling water flows inside and
outside the hex cans. Cracks were found in the mid
length, high flux regions of the target rods during a
search for the source of a-contamination detected in
the exiting coolant. Investigation showed that the
cracks were intergranular and were oriented in the
length of the rods at locations where the fins had been
removed. Lengths were up to 66 mm. The cracks
originated in the target rod cladding, but some of
the larger ones had penetrated the 1100Al jackets of
the target slugs. There was no evidence that corrosion
was involved. The hex cans were not cracked. The
exposure history of the rods is that they were first
irradiated for about 1 year in the D2O environment of
the C reactor at Savannah River Nuclear Laboratory
at a temperature of 20  C. They were returned to
ORNL and inserted in the HFIR at 46  C where,
during their fifth fuel cycle, the a-leak was detected.
The summed fluences were 6.9  1026 n m2, thermal, and 1.2  1026 n m2, E > 0.82 MeV. The conclusion from the investigation was that gas swelling of
the target meat had imposed a hoop stress on the
radiation-damaged cladding that had become too stiff

to undergo plastic flow and had cracked instead.


Rupture tests on unirradiated lengths of the cladding
tubes by internal pressure caused failure along the
machined-off fin lines, indicating the lines were weak
regions in the tubes. A possible solution to the cracking problem was to decrease the swelling-related
hoop stresses on the cladding by raising the pore
volume in the target meat from the then current
10% level to 20% and 25%. Trials were successful,
and the cracking no longer occurs. The questions of
why the cracks were located only on the fin lines and
why it was intergranular were not answered, but were
pursued.71 The intergranular nature of the fractures
made helium embrittlement a suspect. However, calculations of helium levels from the 27Al(n, a) reaction
with fast neutrons had given 7 appm, which was
considered inadequate for helium embrittlement at
the low irradiation temperatures experienced by the
rods. But supposing it was occurring, why did it favor
the fin lines? Metallographic and TEM examinations
of pieces of unirradiated cladding tubing showed that
the extrusion process had stretched the inclusion
particles into stringers and forced sheets of them
into the fins from which they extended back into
the tube wall. Removal of the fins left behind an
aggregation of stringers protruding into the tube
wall. Ergo the weakened regions along the fin lines
seen in the rupture tests of the unirradiated tubing.
When the enhanced production of helium from Ni by
thermal neutron capture was announced,66 a connection between aggregates of Ni-rich stringers and
helium embrittlement was discerned. Irradiated pieces
of high-purity Al, 1100Al, and a X8001 hex can were
sent to a specialist laboratory for helium analyses. For
a common thermal fluence of 1  1026 n m2, the
results were 1.8, 4.8, and 9.5 appm, respectively. After
a fluence of 3  1026 n m2, the corresponding
values were 7.2, 18.1, and 145. A piece of hex can
irradiated to 5.8  1026 n m2 yielded 220 appm.
A piece of 6061Al at 13.8  1026 n m2 gave 47 appm,
much less than the hex can at 5.8  1026 n m2. These
results leave no doubt that the presence of Ni in Al
irradiated to high thermal neutron fluences greatly
boosts the helium levels. And since the He will
remain close to the Ni particles, there must be very
large concentrations around the stringers. Any grain
boundaries overlapped by those local helium clouds
will be prime candidates for helium embrittlement
cracking under the influence of hoop stresses. Hence,
the intergranular cracking at the fin lines.
In most metals, the gaseous transmutation products
play a larger role in the development and effects of

165

Performance of Aluminum in Research Reactors

radiation damage structure than do the nongaseous


transmutants, one reason being that most construction
metals do not produce much nongaseous transmutants. Al is different. Depending on the degree of
thermalization of the neutron spectrum, Al can produce large quantities of silicon from the two-step reaction 27Al nth ! 28Al g; 28Al ! 28Si b. A rough
guide to the quantity expected annually in the HFIR
PTP spectrum can be obtained by multiplying the
thermal neutron fluence by 230mb, the standard thermal neutron (0.0235 eV, 2200 m s1) absorption crosssection for Al. The result is 1.035 at.% Si (1.073 wt%).
The Si is insoluble in Al at temperatures below about
350  C and is usually manifest as a precipitate of
elemental Si.72 This precipitate makes a substantial
contribution to radiation damage in Al, and is the
dominant hardening agent at high thermal neutron
fluences. There is one outstanding qualifier to that
generalization. In the 5xxx-type AlMg solid solution
series, the free Mg atoms dissolved in the Al will react
with the atoms of transmutant Si to form a precipitate
of Mg2Si.73 Thus, a 5xxx series alloy will be converted
to a 6xxx-like alloy.7376 Figure 8 shows the Mg2Si
microstructure formed in irradiated 5052-O alloy.
Because this precipitate occurred at a temperature
below the usual 160  C aging temperature used to
obtain the T6 tempered condition in 6061Al, the
Mg2Si precipitate developed in the 5000 alloy is finer
than in the 6061-T6 alloy. The microstructure of
heavily irradiated 6061-T6 alloy is illustrated in
Figure 9. Since there are usually larger quantities of
Mg in the 5xxx alloys than in the 6xxx alloys, a greater
volume of Mg2Si can be created in the former 5xxx
alloys. Hence, irradiated 5xxx alloys will undergo radiation hardening and precipitation hardening simultaneously, and their overall hardening rate will be larger
than in other Al alloys exposed to the same neutron
fluence. Note that there are no voids in Figure 8.
Sparsely distributed voids are found75 at a higher
fluence of 1.8  1027 n m2. At half of that dose, the
6061-T6 alloy contains many more voids, Figure 9.
The association of the transmutant Si with voids is
interesting. We saw earlier that voids and particles of
Si become visible in the microstructure at about the
same dose. The voids are larger and fewer than the Si
particles. A Si particle is usually attached to one facet
of a void, and that particle is larger than its unattached brethren in the matrix. It is also facetted.
As irradiation continues, a change occurs in the
void-Si relationship. The voids lose their facetted
shape and become rounded.77 They are completely
covered with a thin coating consisting of mostly Si

100 nm

200

000

220

020

Figure 8 Precipitates of Mg2Si and excess Si in formerly


5052-OAl irradiated to 5.7  1026 n m2 at 55  C.
Reproduced from Farrell, K. J. Nucl. Mater. 1981, 97, 3343,
with permission from Elsevier.

0.1 mm

Figure 9 Voids with background precipitate of


radiation-produced silicon and silicon-decorated original
Mg2Si precipitates in 6061-T6Al irradiated to 1027 n m2
at 55  C.

166

Performance of Aluminum in Research Reactors

0.1 mm
Figure 10 Si particles and Si-coated voids on a
carbon extraction replica from 1100-OAl irradiated to
1.4  1027 n m2 (E > 0.1 MeV) and 2.3  1027 n m2
(E < 0.025 eV) at 55  C. Reproduced from Farrell, K.;
Bentley, J.; Braski, D. N. Scripta Metall. 1977, 11,
243248.

with some Al. The coating is noncrystalline and flexible. The unattached Si particles in the matrix are also
rounded but are crystalline. The coated voids jut out of
the thinned edge of the hole in TEM foils and can be
lifted from the matrix on carbon extraction replicas.
Figure 10 is an example. The larger features with the
dark rims are the coated voids. Four of them have
partially collapsed without breaking, indicating a
highly ductile coating. Many of the Si particles seem
to have a layered structure. The AlSi system is a
simple eutectic; there are no compounds. It is suspected
that the small amount of Al found in the void coatings
may be from the Al matrix that was not completely
dissolved from the voids during the electrolytic
extraction process. Silicon is obviously involved in
void formation and growth but its specific role is
unclear.

5.07.7 Property Changes


5.07.7.1

Swelling

Radiation swelling is the increase in volume arising


by accumulation of voids from excess vacancies and
by formation of gas bubbles. For Al, there are also
small swelling contributions from build-up of particles of transmuted Si and, in 5xxx alloys, Mg2Si,
which have densities of 2329 and 1990 kg m3,
respectively.78 Gas bubble swelling is not an issue
for Al in RRs because the temperature is too low,
except perhaps in fuel cladding where some pores

found in the cladding may have been formed by the


accumulation of hydrogen. Swelling can be measured
from dimensional changes. More often it is determined from changes in immersion density values.
Swelling in various Al alloys is shown in Figure 11.
These alloys were all irradiated in the core of the
HFIR and they make the most comprehensive and
consistent set of swelling data.79 For reference, the
dotted line is estimated for the swelling from Si alone.
It is evident that the unirradiated chemical compositions and microstructures have major effects on the
degree of radiation swelling. The purest grades, sixnines and four-nines, show swelling earliest in dose
and swell at the highest rates with dose. The rates
decrease above a dose of about 1  1025 n m2.
Swelling in the two-nines grade (1100-O) requires
significantly higher doses, but the swelling rate is
unchanged. The 6061-T6 alloy, with its inherent
Mg2Si phase, starts swelling appreciatively later in
dose than the 1100-O. This is traceable to reduced
nucleation of voids, but its swelling rate is about the
same as the other alloys. The greatest resistance is
in the 5052-O alloy. There, the swelling is less than
for the Si alone. In this alloy, much of the early
swelling is not due to voids; it is caused by the silicon
and the new Mg2Si phase and by the increase in the
original density of the matrix, r0, as Mg is drawn
from solution to create the Mg2Si.
The effects of prior cold work on swelling in
Al agree in general that the presence of cold work
dislocation structure decreases the overall void
swelling but the reduction is not massive; concurrence
of dislocation recovery confuses the details.8083
5.07.7.2

Mechanical Properties

The major consequences of radiation damage structures on the mechanical properties of Al alloys are
radiation hardening and associated loss in ductility.
There are too many data from too many sources
to be described in detail here. A good source of
compiled data, including the sparse information on
fracture toughness and weldments, is Marchbanks.84
For 6061Al in particular, see Farrell.85 Strengthening and loss of ductility are demonstrated best
in tensile properties. In Figure 12 we can directly
compare the changes in strength and ductility of
different alloys irradiated and tested under the
same conditions.79 The most striking feature is the
relatively rapid hardening displayed by the 5052-O
alloy. As explained earlier, this is caused by the
combined effects of radiation damage and in-reactor

Performance of Aluminum in Research Reactors

wt % Si 0.1

dpa 0.1

10

10

10

100

6-9
4-9
2-9 (1100-O)
6061-T6
5052-O

( ri )

Ti = 328 K (0.35 Tm)


1
6061-T6

Swelling

rori

167

1100-O
5052-O
0.1
Pure aluminum
28

Si

0.01
1024

1025
1026
Fluence (n m-2 > 0.1 MeV)

1027

Figure 11 Radiation-induced swelling in various Al alloys as a function of fast fluence. Reproduced from Farrell, K.
In Proceedings of the Conference on Dimensional Stability and Mechanical Behaviour of Irradiated Metals and Alloys,
Brighton, Apr 1113, 1983; British Nuclear Energy Society: London, 1983; Vol. 1, pp 7376, with permission from British
Nuclear Energy Society (now The Nuclear Institute).

formation of a fine precipitate of Mg2Si. In contrast,


the 6061-T6 alloy, which contains Mg2Si before
irradiation, begins radiation hardening at about the
same fluence as the 1100-OAl, and hardens thereafter at the same rate. The 1100-O alloy contains no
Mg2Si before or after irradiation. From which we
deduce that preexisting Mg2Si precipitates play no
role in radiation hardening. This is an interesting
conclusion. It contradicts the expectation that the
precipitates would diminish the degree of radiation
hardening in 6061-T6Al by promoting the recombination of freely migrating vacancies and interstitials.
Perhaps that expectation is wrong. But it seems
satisfactory for explaining the delayed swelling in
the 6061-T6 alloy, where nucleation of voids is
retarded, perhaps until the transmutant gases enable
achievement of critical size cavity nuclei. Alternatively, maybe the radiation-produced Si dominates
the hardening process. It is a mystery. The fluence
for the onset of radiation hardening in the weak
49Al is about one order of magnitude less than in

the other alloys, and the subsequent rate of hardening is less than for the others. Here, again, we invoke
the recombination argument. This higher purity
material contains less solutes and inclusions. Thus
there is less trapping and annihilation of freely
migrating point defects, hence more point defect
clusters are formed in the early stages of irradiation.
It is suspected that the reduced rate of hardening is
connected with dynamic recovery of deformation
during the tensile test. It was pointed out in Section
5.07.3.2 that recovery from cold work occurs readily
in high-purity Al at room temperature. Loss in
uniform elongation in all of the alloys is concomitant with increase in strength . . . to a point. At a
fluence of about 1026 n m2 the ductility reaches a
plateau of 35% even though the strength continues
to rise. The 1100-OAl has the least ductility in the
plateau region and displays an intergranular-like
fracture mode that may be caused by tearing and
void interconnection in the void-rich regions lying
alongside the grain boundaries.

Performance of Aluminum in Research Reactors

600

Stress (MPa)

500

Tirr = 328 K
Ttest = 323 K
.
~ 10-4 S-1
'

168

UTS
0.2% FS

400

300

6061-T6

200
5052-O
100
1100-O
4-9AI
0

Uniform elongation (%)

40

30

20

4-9AI
5052-O
1100-O

10

6061-T6

0
1023

1024

1025

1026

1027

-2

Fluence (n m > 0.1 MeV)


Figure 12 Radiation-induced changes in room temperature tensile properties of various Al alloys. Reproduced from
Farrell, K. In Proceedings of the Conference on Dimensional Stability and Mechanical Behaviour of Irradiated Metals and
Alloys, Brighton, Apr 1113, 1983; British Nuclear Energy Society: London, 1983; Vol. 1, pp 7376, with permission from
British Nuclear Energy Society (now The Nuclear Institute).

Irradiated metals undergoing plastic deformation


are prone to strain localization, seen as dislocation
channels swept free of point defect clusters. Such
channeling coincides with changes in the tensile test
curve, notably introduction of a sharp yield point
and a reduction in the slope of the strain-hardening
portion of the curve. In severe cases, a yield point
drop occurs that leads directly into prompt necking.
Dislocation channeling has not been reported for
neutron-irradiated Al. Nor do the tensile curves for
neutron-irradiated Al display sharp yield points and
pronounced decreases in work hardening rate. The
reason for this apparent resistance to channeling in

Al is not known. Al is not immune to channeling.


It occurs in unirradiated quenched-and-aged Al.
When high-purity Al crystals are quenched from
near the melting point into water at room temperature or into iced brine, then aged at room temperature for 14 days or at 60  C for 2 h, there is
a considerable increase in yield stress (YS) and large
decrease in strain hardening rate.86,87 Such quenchaging treatment produces many small vacancy clusters. Yielding occurs abruptly, and large widely
spaced slip bands appear on the specimen surfaces,
forming mirror images from one surface to the
opposite surface. TEM examination87 reveals that

Performance of Aluminum in Research Reactors

A5 AI, ~35% CW, 77 K (0.08 Tm)

400

~1 1023 n m-2, E > 1 MeV

Stress (MPa)

300

~5 1021 n m-2

7.5 1021 n m-2, then


held 60 h at room
temperature before
test at 77 K

200
Unirradiated

100

Unirradiated,
tested 293 K

0
0

10

20

30

40

Elongation (%)

Figure 13 Tensile curves of pure Al irradiated and tested


at cryogenic temperatures, showing recovery at room
temperature. Reproduced from Bochirol, L.; Brauns, P.;
Claudet, G. Prog. Refrigeration Sci. Technol. 1973, 1,
643650, with permission from International Institute of
Refrigeration, AVI Publishing Company.

the slip bands are dislocation channels prominently


displayed against the background of undisturbed
vacancy clusters remaining in the matrix between
the channels. Another way to induce a sharp yield
point and reduced strain hardening rate in Al is to
irradiate and tensile test at cryogenic temperature.
Figure 13 gives tensile curves for an A5 alloy
(>99.95% Al) in the half-hard, H24 (35% cold
worked), condition after neutron irradiation and testing at 77 K in LN.88 A curve for an unirradiated
specimen tested at room temperature is included as
a reference condition. Note that this unirradiated
specimen has only 14% elongation, in keeping with
its cold-worked condition. When tested in LN the
unirradiated material is not only stronger it is much
more ductile, with an elongation of 44%. This is an
example of the enhancement of ductility at low temperature mentioned in Section 5.07.3.1. Irradiation
to 5  1021 n m2 in LN further increases the YS,
and although the elongation falls a little, it is still
larger than at room temperature; moreover, the initial
part of the strain hardening curve to about 8% elongation is steeper than that for the unirradiated specimen, indicating no irradiation-induced dislocation
channeling. However, the reduced slope after 8%
elongation and prior to onset of necking in the irradiated specimen signals possible late entry of channeling. Note that for this dose at room temperature
there would be no discernible radiation damage
structure and no change in tensile properties. In
the LN specimen irradiated to the higher fluence

169

of 1  1023 n m2 intervention of channeling is


more likely. There, very steep initial strain hardening
passes though a peak at 1% elongation and plunges
into prompt necking. Even greater strengthening and
more exaggerated prompt necking is found for both
cold-worked and annealed specimens when the irradiation and testing are done at 27 K in liquid neon.88
The deformation modes were not determined for
these specimens. Identification of the deformation
mode would have required TEM examinations of
specimens cut from the gauge sections of the specimens and would probably not have revealed the
channeling because the specimens would undergo
rapid recovery from radiation hardening when they
are brought up to room temperature. Such recovery
is illustrated by the curve for the specimen irradiated
to 1  1022 n m2 then held at room temperature
for 60 h before testing in LN. Almost complete recovery has occurred, indicating elimination of the point
defect clusters needed to provide background contrast for detection of dislocation channels. The good
result of this recovery is that it relieves the stored
energy built up by accumulation of the radiation
damage microstructure at the cryogenic temperature.
Advantage is taken of this recovery process in Al cold
neutron sources by performing periodic in situ room
temperature anneals on them. These anneals do not
erase the transmutation products. Residual hardening
from the gaseous products will be negligible but
the accumulating Si, whose levels will be high due
to the highly thermalized neutron spectrum at the
cold source, will give increasing retained hardening as it forms precipitates at room temperature.
For cold source vessels constructed from 5xxx alloys,
the transmutation-produced Si will react with dissolved Mg at room temperature to create naturally
aged fine precipitates of Mg2Si, with greater hardening effects than from Si alone.
5.07.7.3

Effects of Neutron Spectrum

When tensile data for irradiated Al from different


reactors or from different regions of a single reactor
are compiled in a single traditional plot of property
versus fast neutron fluence, there tends to be a large
scatter in the data, particularly at the higher fluences.
It was noted89 that the scatter was reduced somewhat
if the plot was made in terms of thermal fluence.
This improvement was attributed to a spectral effect
involving the production of Si precipitates by the
thermal neutrons and their modification by fast neutrons. That explanation has since been taken a little

170

Performance of Aluminum in Research Reactors

wt% silicon

0.1

0.01

10

Tensile strength (MPa)

800
6061-T6 aluminum . grouping by fth/ft ratio

700

20

Unirradiated
HFIR target
HFBR surveillance
HFBR V15
HFBR CRDF

600
500

21

21

21

8 5-81-3
22

400
5

1.3

.57
53

0.5

300
200
0

1023

1024

1025
1026
Thermal fluence (n m- 2)

1027

1028

Tensile strength (MPa)

800
20

700

21

600

21

21
5-8
8

500
5

22

400
5

1-3
1.3
.57
.53

0.5

300
200
0

1023

1024

1025

1026

1027

1028

Fast fluence, E > 0.1 MeV (n m- 2)

Figure 14 Radiation-induced strengthening of 6061-T6Al discriminated by thermal and fast fluence. Reproduced from
Farrell, K. In Proceedings of Materials Research Society Symposium on Microstructure of Irradiated Materials, 1995; Vol. 373,
pp 165170, with permission from Materials Research Society.

farther.90 Figure 14 presents tensile strength data


for the 6061-T6 alloy irradiated in two reactors,
the HFIR and the High Flux Beam Reactor (HFBR),
at 5065  C. The same strength data are plotted
against thermal fluence in the upper box and against
fast fluence in the lower box. Ignoring for the
moment the lines in the plots, it is evident that at
fluences above 1  1025 n m2 the data points are
more scattered when plotted against fast fluence. The
HFIR specimens were all irradiated at the same
location where the ratio of thermal-to-fast flux, th/
f, was narrow, 1.72.3. The HFBR was cooled and
moderated by heavy water, and the HFBR specimens
were taken from regions with th/f values between
0.5 and 22. When the data are discriminated by th/
f as indicated by the lines labeled with the large
numerals, a pattern of dependence on th/f
emerges. The trend is the same in both boxes but
the separation is clearer in the fast fluence plot. It is
evident that when irradiation is conducted in a hard
spectrum, th/f 0.5, there is little increase in
strength, even at high fluence. As the spectrum softens, th/f ! 21, the specimens harden. This
response is obviously due to greater generation of
silicon precipitates in the more thermalized spectra.

TEM examinations of the more heavily irradiated


specimens showed that the dominant microstructural
feature introduced by the irradiation in both reactors
was particles of Si. They were more numerous and
smaller in the softer spectrum, in agreement with
the strengthening trend. After eliminating possible
contributions from differences in damage rates, irradiation temperatures, and exposure times, it was concluded that the spectrum controls the ripening
behavior of the Si precipitates, which determines the
degree of strengthening. A hard spectrum causes more
sputtering of the precipitates and more transport of
the sputtered atoms, resulting in greater ripening of
the precipitates and less strengthening per unit Si
level or unit fluence.
5.07.7.4 Radiation Softening, Creep, and
Stress Relaxation
There are some highly contentious reports of radiation softening of Al alloys that demand some discussion. These are not claims of minor softening such as
a decline in the strain-hardening region of a tensile
curve. They are full-blown changes from an agehardened or cold-worked condition to a dead soft,

Performance of Aluminum in Research Reactors

annealed state. And they occur in specimens that are


not carrying a load. The softening is usually interpreted as being due to destruction of dislocation
tangles and precipitate particles through dynamic
atomic displacements created by fast neutrons. It is
the present writers opinion that they are actually
unintended instances of thermal softening by radiation heating, which can happen very easily and
quickly in Al if adequate precautions are not taken
to ensure adequate cooling during irradiation. In each
of these reports, the specimens were irradiated in
sealed cans with external water cooling and no
means of measuring the internal temperature. Using
sealed cans is an invitation to radiation heating of the
contents unless very strict actions are taken to ensure
satisfactory heat removal. In the first91 of these
reports, disks of 2024-O, 2024-T3, 1100, and sintered
Al powder of undeclared Al2O3 content were placed
in aluminum cans evacuated and back filled with
helium. The cans were irradiated in the MTR for
60 days to fast neutron doses of 3  1022, 2  1023, and
8.5  1023 n m2 at estimated temperatures of 100,
119, and 138  C, respectively. Hardness tests on the
2024-T3 disks showed: unirradiated control, Hv 170;
3  1022 178; 2  1023 68; 8.5  1023 60, indicating a pronounced change from the age-hardened-T3
condition to a fully softened condition. The hardnesses
of the 1100 and the sintered aluminum powder alloy
were unchanged by the irradiations. This softening of
2024-T3 alloy conflicts with tensile and hardness
data92 for a 2024-T3 or -T4 alloy irradiated to doses
between 2  1023and 1.3  1026 n m2 that show no
change in properties to a dose of 1.2  1024 n m2,
above which there is hardening. These latter data
were for specimens that were irradiated in direct
contact with flowing water to minimize nuclear heating. The writer has also tested tensile specimens
of 2024-T351Al that were irradiated in contact
with flowing water; there was no change in properties
for fluences between 4  1021 and 1  1024 n m2,
beyond which hardening was displayed.93
The second softening claim94 was made for tensile
specimens of Al3Mg in a cold-worked condition
(degree of work not given), and 6061-T6 bombarded
with 600800 MeV protons in the Los Alamos Meson
Physics Facility (LAMPF). Annealed specimens of
the two alloys were included for reference. Assemblies of seven specimens laid side-by-side were sealed
in envelopes made from 0.13 mm thick foil of 1100Al.
Sixteen packages were stacked with gaps between
them to allow cooling water to flow though at a
rate of 1 l s1 and pressure of 17 bar (1700 kPa) to

171

ensure good contact of the foil with the specimens.


The beam spot size was 50 mm, the average current
was 574 mA, and the exposure period was 750 h.
Displacement damage up to about 0.2 dpa was created by the direct impact of the protons and by highenergy neutrons spalled from surrounding materials
and the nearby beam stop. No Si was generated.
Helium and hydrogen contents were 67 and
275 appm. From the beam parameters, it can be seen
that the power input was about 200 MW m2, which
is about 20 times larger than the heat fluxes reached
on fuel cladding. The irradiation temperature for the
LAMPF specimens was estimated to be 4050  C,
certainly not exceeding 100  C. Postirradiation tests
revealed that the yield strengths of the cold-worked
Al3Mg and 6061-T6 specimens were reduced
by 7080% and the ultimate strengths by 4065%,
which placed them at the same strength levels as
those for unirradiated annealed specimens of the
alloys. The irradiated annealed alloys did not
undergo irradiation hardening; they actually softened
by 1020%. Metallographic studies and TEM examinations of the formerly cold-worked Al3Mg alloy94,95
showed that recrystallization had occurred. At a dose
of only 0.1 dpa, the tightly packed dislocation cell
structure was gone, replaced by very loosely tangled
dislocations, and the grains were now equiaxed.
In the former 6061-T6 alloy, there were no longer
any Mg2Si precipitates. In the irradiated annealed
Al3Mg specimens, there were gas bubbles on the
grain boundaries. These microstructural changes are
very much like those found for thermal anneals or
high temperature irradiations; they require sustained,
long range, high-intensity diffusion of vacancies and
solutes. Usually, irradiation with energetic particles
does not create or support such condition unless
radiation heating occurs. Under adequate temperature control, radiation enhanced diffusion is highly
local and sporadic and is rapidly quenched. Normally, radiation attrition of precipitates results in
precipitate atoms being recoiled into the adjacent
matrix and there forming new satellite particles of
the mother phase or a new phase. A situation of total
elimination, with no reappearance of the precipitate
by aging, is more typical of a high-temperature solution anneal and a slow cool. Nevertheless, the authors
attributed these extreme changes in microstructure
to the absence of transmutation-produced Si which
stabilizes radiation damage microstructure in regular
neutron irradiations, and to the different nature
of the damage produced by high-energy protons,
this despite efforts they and others were making

172

Performance of Aluminum in Research Reactors

at the time to promote such irradiations as a substitute


for slower neutron irradiations. They were unable to
account for the disagreement with an earlier 600 MeV
proton irradiation of a solution-treated AlMgSi
alloy similar to 6061 at 150  C in the Swiss PIREX
facility, which induced precipitation of a high density
of Mg2Si precipitates.96 Strangely, they gave no consideration to the possibility that they might have
overestimated the cooling efficiency of their irradiation experiment in LAMPF. It should have been the
prime consideration because at that time LAMPF
was undergoing changes to make it suitable for materials irradiations. None had been performed there
previously. The changes were incomplete but the
experiment was made anyway. It was a pseudo atest. Since then, improvements have been made to
the facility and many materials irradiations have been
conducted that simulate reactor irradiations. Later
irradiations of 6061-T6 in LAMPF97,98 have not
encountered the softening found in the inaugural
proton irradiation.
The third softening claim99,100 was again for
a cold-worked, nominal Al3Mg alloy and for a
precipitation hardened AlMgSi (6061-type) alloy.
Tensile specimens and annealed companion specimens were irradiated to fast fluences of 8.7  1021
and 2.5  1022 n m2 in the wet channels of the
Egyptian Nuclear Research Reactor. The responses
of the materials to both of these fluences were almost
identical and are confusing. The AlMgSi alloy in two
different age-hardened conditions and its as-received
cold rolled condition displayed large decreases
in hardness, YS, and ultimate tensile stress (UTS).
The Al3Mg alloy in its as-received cold rolled condition showed large reduction in hardness and relatively small reductions in YS and UTS. In a partially
annealed condition (1 h at 240  C) its hardness was
reduced 30% but its YS and UTS were increased by
55% and 27%. In the fully annealed state (1 h at
400  C) the hardness was decreased by 5%, the YS
was increased by a whopping 380%, and the UTS rose
by 97%. Although no microstructural studies were
made, the authors concluded that the observed softening was due to removal of cold work dislocations
and dissolution of Mg2Si precipitates as seen in
Lohmann et al.94 and Singh et al.95 The conclusion
that Mg2Si precipitates are unstable in a neutron flux
is at odds with the well-established fact that the precipitates persist to very high fluences in 6061-T6
specimens that are properly cooled during irradiation,
and they are actually generated by the irradiation in
AlMg solid solution alloys. The unusually large

hardening caused in the partially and fully annealed


Al3Mg specimens was not discussed. It was ascertained from the lead author that the specimens had
been irradiated in sealed cans. It was conveyed to the
author that ongoing experiments at ORNL with
6061-T6 were not finding any signs of radiation
softening at fluences similar to those in his experiments. Some years later, he published a third paper101
on the AlMgSi alloy in four conditions: Cold rolled;
annealed; naturally age hardened; and artificially age
hardened. The specimens were wrapped in Al foil
and were packed in aluminum powder in sealed Al
cans. The fast neutron fluence was 3.7  1022 n m2,
somewhat higher than his previous irradiations.
Hardness and tensile tests revealed no softening of
the cold-worked and age-hardened specimens. The
annealed specimens displayed a 23% increase in
Vickers hardness and a 111% increase in yield
strength. This radical departure from the previous
softening results was not credited to better cooling
of the specimens during irradiation. Rather, the newfound nonsoftening of the prehardened materials was
attributed to the longer irradiation exposure. It was
explained that the prehardened alloys actually did go
through a full softening process in the region of
1  1022 n m2 as before, by the mechanism of
radiation-induced dissolution of the cold work dislocations and Mg2Si precipitates. This removal of
these strong point defect sinks allowed irradiation
damage microstructure to build up as irradiation
continued. At the exposure of 3.7  1022 n m2 sufficient damage microstructure had formed to restore
the strengths of the softened materials to their former
values.
These claims of radiation softening, controversial
as they are, do serve a useful purpose. They highlight
the previously discussed temperature sensitivity of
cold-worked and age-hardened aluminum. Moreover,
they send a message that in the event of an unplanned
temperature excursion on such materials they should
be examined carefully to ensure that they still meet
the strength requirements for the application.
Radiation creep and stress relaxation are timedependent plastic deformation processes driven by
applied external stresses or by stored internal stress.
They can be accelerated by point defects from atomic
displacement events. The few data that exist for
radiation creep in aluminum are confusing and inconclusive. Details are available in Farrell.85 Creep can
lead to failure if the load is continuously applied.
Strain from stress relaxation ceases once the load is
relaxed. There are at least two cases81,102 of stress

Performance of Aluminum in Research Reactors

relaxation in Al reactor components. Both were


found in retired reactor components that had entered
service in cold-worked conditions, one 1100Al and
the other precipitation-hardened 6063Al. During service they had been continuously in direct contact
with the coolant water, so thermal interference was
not involved. Some softening occurred due to partial
recovery of the cold work structure at low doses,
followed by radiation hardening at higher doses.
Complete softening to an annealed level did not
occur during irradiation.

5.07.8 Conclusion
Aluminum and its alloys have contributed immensely
to development of water-cooled RRs and to our
understanding of radiation effects in metals. They
continue to do so.

10.

11.

12.

13.

14.

15.

16.

Acknowledgments
During preparation of this chapter, interactions with
present and former Oak Ridge National Laboratory
personnel, S. A. David, R. D. Godfrey, S. J. Pawel, J. D.
Sease, and R. E. Stoller were sincerely appreciated.
Opinions and conclusions in this chapter are solely
the responsibility of the writer.

17.

18.

19.

References
1.
2.
3.

4.

5.
6.

7.
8.

9.

West, C. D. Nuclear News, Oct 1997; pp 5056.


http://www.iaea.org/worldatom/.
Rosenthall, M. W. An Account of Oak Ridge National
Laboratorys Thirteen Nuclear Reactors; ORNL/
TM-2009/181; Oak Ridge National Laboratory: Oak
Ridge, TN, Aug 2009, Revised Mar 2010.
Williams, R. O. Terminal Report on ORNL Slug Problems:
Causes and Prevention; ORNL-CF-50-7-160; Oak Ridge
National Laboratory: Oak Ridge, TN, period covered
Sept 1948 to July 1950.
Flemings, M. C. Solidification Processing; McGraw-Hill:
New York, 1974.
Thomas, W. M.; Nicholas, E. D.; Needham, J. C.;
Murch, M. G.; Temple-Smith, P.; Dawes, C. J.
Friction-Stir Butt Welding. GB Patent No. 9125978.8.
International Patent Application No. PCT/GB92/02203;
Welding Research Institute: Great Britain, 1991.
Aluminum Standards and Data, 10th ed. The Aluminum
Association: Washington, DC, 1990.
Use of 6061-T6 and 6061-T651 Aluminum for Class 1
Nuclear Components, Section III, Division 1, Cases of
Boiler and Pressure Vessel Code, Supplement 10-NC,
Case N-519; American Society of Mechanical Engineers:
New York, 1994.
Structural Alloys Handbook, 1989 ed., Vol. 3; Battelle
Memorial Institute: Columbus, OH, 1989; p 14.

20.
21.

22.
23.

24.
25.

26.

173

Matos, J. E.; Snelgrove J. L. Selected Thermal Properties


and Uranium Density Relations for Alloy, Aluminide,
Oxide, and Silicide Fuels; IAEA-TECDOC-643,
International Atomic Energy Agency, Vienna, 1992;
pp 119, Appendix I-1.1 in Research reactor core
conversion guidebook Volume 4; Fuels (Appendices IK).
Gibson, G. W.; Shupe, O. K. Annual Progress Report on
Fuel Element Development for Fiscal Year 1961; IDO-167
27, TID-4500 (17th ed.), Phillips Petroleum Company:
Idaho Operations Office, 1961.
Bartz, M. H. In Properties of Reactor Materials,
Proceedings of the 2nd United Nations International
Conference on the Peaceful Uses of Atomic Energy,
Geneva, Sept 113, 1958; Vol. 5, pp 466474.
Marshall, F. M. Advanced Test Reactor Capabilities and
Future Irradiation Plans; INL/CON-06-01211; Oct 2006;
www.inl.gov/technicalpublications/Documents/
3489341.pdf.
Beaver, R. J.; Adamson, G. M., Jr.; Patriarca, P.
Procedures for Fabricating Aluminum-Base ATR Fuel
Elements; ORNL-3632; Oak Ridge National Laboratory:
Oak Ridge, TN, Jun 1964.
Kim, S. S.; Pope, C.; Taylor, L. L. Criticality Analysis for
Proposed Maximum Fuel Loading in a Standardized SNF
Canister with Type 1a Baskets; INL/EXT-07-12326; Idaho
National Laboratory: Idaho Falls, ID, Feb 2007.
Binford, F. T.; Cramer, E. N. The High Flux Isotope
Reactor; A Functional Description, Vol 1B, Illustrations;
ORNL-3572 (Revision 2); Oak Ridge National Laboratory:
Oak Ridge, TN, Jun 1968.
Adamson, G. M., Jr Fabrication Procedures for the Initial
High Flux Isotope Reactor Fuel Elements; ORNLTM-2196; Oak Ridge National Laboratory: Oak Ridge,
TN, Feb 1969.
Knight, R. W.; Morin, R. A. Fabrication Procedures
for Manufacturing High Flux Isotope Reactor Fuel
Elements II; ORNL/6852; Oak Ridge National
Laboratory: Oak Ridge, TN, Dec 1999.
Sease, J. D.; Primm, R. T., III; Miller, J. H. Conceptual
Process for the Manufacture of Low-Enriched Uranium/
Molybdenum Fuel for the High Flux Isotope Reactor;
ORNL/TM-2007/39; Oak Ridge National Laboratory: Oak
Ridge, TN, Sept 2007.
Adamson, G. M., Jr.; Knight, R. W. HFIR Fuel Element
Production and Operation; ORNL-TM-2196; Oak Ridge
National Laboratory: Oak Ridge, TN, Jun 1968.
Richt, A. E.; Knight, R. W.; Adamson, G. M., Jr.
Postirradiation Examination of the Performance of HFIR
Fuel Elements; ORNL-4714; Oak Ridge National
Laboratory: Oak Ridge, TN, Dec 1971.
http://www.rertr.anl.gov/.
Hofman, G. L.; Rest, J.; Snelgrove, J. L. Comparison of
Irradiation Behavior of Different Uranium Silicide
Dispersion Fuel Element Designs. CONF-9409107-6,
ANL/TD/CP-85108; Presented at the 1994
International Meeting on Reduced Enrichment for
Research and Test Reactors, Williamsburg, VA,
Sept 1823, 1994.
Sears, D. F.; Conlon, K. T. Development of LEU Fuel to
Convert Research Reactors: NRU, MAPLE, and SLOPOKE;
http://www.7ni.mfa.no/NR/rdonlyres/EC9F5....
Clark, C. R.; Knighton, G. C.; Meyer, M. K.; Hofman, G. L.
Monolithic Fuel Plate Development at Argonne National
Laboratory, 2003; http://www.rertr.anl/RERTR25/PDF/
Clark.pdf.
Clark, C. R.; Wight, J. M.; Knighton, G. C.; Moore, G. A.;
Jue, J. F. Update on Monolithic Fuel Fabrication
Development; INL/CON-05-00855; Idaho National
Laboratory: Idaho Falls, ID, Nov 2005.

174
27.

28.
29.

30.
31.

32.

33.
34.
35.

36.

37.
38.

39.

40.
41.
42.

43.
44.

45.

Performance of Aluminum in Research Reactors


Moore, G. A.; et al. Monolithic Fuel Fabrication Process
Development at the Idaho National Laboratory, 30th
International RERTR Meeting, Washington, DC, Oct 59,
2008; INL/CON-08-14550.
Corrosion of Research Reactor Spent Fuel in Water,
2003; http://www-pub.iaea.org/MTCD/publications/
PDF/TRS418_web.pdf.
Griebenow, M. L.; Hanson, G. H.; Larrick, A. P. TRA
Oxide Film Control and Surveillance; RE&C Report
RE-A-77-059; Idaho National Laboratory: Idaho Falls, ID,
Oct 1977.
Shaber, E.; Hofman, G. Corrosion Minimization for
Research Reactor Fuel; INL/EXT-05-00256; Idaho
National Laboratory: Idaho Falls, ID, Jun 2005.
Griess, J. C.; Savage, H. C.; Rainwater, J. G.;
Mauny, T. H.; English, J. L. Effect of Heat Flux on the
Corrosion of Aluminum by Water. Part III. Final Report
on Tests Relative to the High Flux Isotope Reactor;
ORNL-3230; Oak Ridge National Laboratory: Oak Ridge,
TN, Dec 1961.
English, J. L.; Rice, L.; Griess, J. C. The Corrosion of
Aluminum Alloys in High Velocity Water at 170 to 290 C;
ORNL-3063; Oak Ridge National Laboratory: Oak Ridge,
TN, Jun 1961.
Draley, J. E.; Ruther, W. E. Aqueous Corrosion of 2S
Aluminum at Elevated Temperatures; ANL-5001;
Argonne National Laboratory: Lemont, IL, Feb 1, 1953.
Wintergerst, M.; Dacheux, N.; Datcharry, F.; Herms, E.;
Kapusta, B. J. Nucl. Mater. 2009, 393, 369380.
Griess, J. C.; Savage, H. C.; Mauny, T. H.; English, J. L.
Effect of Heat Flux on the Corrosion of Aluminum by
Water. Part I. Experimental Equipment and Preliminary
Test Results; ORNL-2939; Oak Ridge National
Laboratory: Oak Ridge, TN, Apr 1960.
Griess, J. C.; Savage, H. C.; English, J. L. Effects of Heat
Flux on the Corrosion of Aluminum by Water. Part IV.
Tests Relative to the Advanced Test Reactor and
Correlation with Previous Results; ORNL-3541; Oak
Ridge National Laboratory: Oak Ridge, TN, Feb 1, 1964.
Pawel, R. E.; Yoder, G. L.; Felde, D. K.; Montgomery, B. H.;
McFee, M. T. Oxid. Met. 1991, 36(1/2), 175194.
Pawel, S. J.; Felde, D. K.; Pawel, R. E. Influence of
Coolant pH on Corrosion of 6061 Aluminum Under
Reactor Heat Transfer Conditions; ORNL/TM-13083;
Oak Ridge National Laboratory: Oak Ridge, TN,
Oct 1995.
Carlson, P. A.; Curtiss, D. H.; Miller, N. R.; Van
Wormer, F. W. Summary Status Report: Internal
Corrosion of Ribbed Aluminum Process Tubes;
HW-72728; Hanford Atomic Products Operation:
Richland, WA, Feb 19, 1962.
Kim, Y. S.; Hofman, G. L.; Robinson, A. B.; Snelgrove, J.
L.; Hanan, N. J. Nucl. Mater. 2008, 378, 220228.
Leenaers, A.; et al. J. Nucl. Mater. 2004, 327,
121129.
Golosov, O. A. Corrosion of aluminum alloys in water at
temperatures up to 100 C. Paper S6-P18 in RERTR 2009
I 31st International Meeting on Reduced Enrichment for
Research and Test Reactors, Beijing, China, Nov 15,
2009.
Sindelar, R. L.; Lam, P. S.; Louthan, M. R., Jr.; Iyer, N. C.
Mater. Char. 1999, 43, 147157.
Lam, P. S.; Sindelar, R. L.; Peacock, H. B., Jr. Vapor
Corrosion of Aluminum Cladding Alloys and
AluminumUranium Fuel Materials in Storage
Environments (U); WSRC-TR-97-0120; Westinghouse
Savannah River Company: Aiken, SC, Apr 1997.
Kumar, R.; Baheti, G. L.; Chacharkar, M. P.; Khatri, P. K.
J. Radioanal. Nucl. Chem. 1989, 132(1), 39.

46.

47.

48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.

60.
61.
62.

63.

64.
65.

66.
67.
68.

69.
70.

Kim, Y. S.; Hofman, G. L.; Hanan, N. A.; Snelgrove, J. L.


Prediction model for oxide thickness on aluminum alloy
cladding during irradiation. In International RERTR
Meeting, Chicago, IL, Oct 510, 2003.
Farrell, K.; Byun, T. S.; Hashimoto, N. Mapping Flow
Localization Processes in Deformation of Irradiated
Reactor Structural Alloys Final Report; ORNL/TM-2003/
63; Oak Ridge National Laboratory: Oak Ridge, TN, Sept
2003.
Farrell, K.; Houston, J. T. J. Nucl. Mater. 1979, 83, 5766.
Bishop, M.; Fletcher, K. E. Int. Metall. Rev. 1972, 17,
203225.
Smallman, R. E. Modern Physical Metallurgy;
Butterworths: London, 1962; p 188.
Hammer, B.; Jacobsen, K. W.; Milman, V.; Payne, M. C.
J. Phys. Condens. Matter 1992, 4(50), 1045310460.
Wolfenden, A. Micron 1973, 4, 295305.
Peck, R. L.; Westmacott, K. H. Met. Sci. J. 1971, 5, 155.
Bierlein, T. K.; Mastel, B. J. Appl. Phys. 1962, 3339,
28732875.
Risbet, A.; Levy, V. Philos. Mag. 1975, 31, 975983.
Risbet, A.; Brebec, G.; Lanore, J.-M. K.; Levy, V. J. Nucl.
Mater. 1975, 56, 348354.
Farrell, K.; Wolfenden, A.; King, R. T. Radiat. Eff. 1971, 8,
107114.
Risbet, A.; Levy, V. J. Nucl. Mater. 1973, 46(3), 341352.
Jostsons, A.; Long, E. J., Jr.; Stiegler, J. O.; Farrell, K.;
Braski, D. N. Annealing of Voids in Aluminum; ORNL-TM3494; Oak Ridge National Laboratory: Oak Ridge, TN,
Oct 1971.
Ells, C. E.; Evans, W. Trans. AIME 1963, 227, 437.
Farrell, K.; Chickering, R. W.; Mansur, L. K. Philos. Mag.
1986, 53, 126.
Gabriel, T. A.; Bishop, B. L.; Wiffen, F. W. Calculated
Irradiation Response of Materials Using Fission Reactor
(HFIR, ORR, and EBR-II) Neutron Spectra; ORNL/
TM-6361; Oak Ridge National Laboratory: Oak Ridge,
TN, Aug 1979.
Shiraishi, K.; Nagasaki, R. J. Nucl. Sci. Technol. 1965,
212, 499505; Shiraishi, K.; Murata, T. J. Nucl. Sci.
Technol. 1966, 311, 466472; Shiraishi, K. J. Nucl.
Sci. Technol. 1967, 43, 136142; Shiraishi, K.
J. Nucl. Sci. Technol. 1971, 85, 250255.
Smith, I. O.; Russell, B. J. Nucl. Mater. 1970, 35, 137;
1970, 37, 96; 1971, 38, 1.
Farrell, K.; Houston, J. T. Combined Effects of
Displacement Damage and High Gas Content in
Aluminum, ORNL-TM-5395; Oak Ridge National
Laboratory: Oak Ridge, TN, May 1976. Also available in
Proceedings of International Conference on Radiation
Effects and Tritium Technology for Fusion Reactors,
Gatlinburg, TN, Oct. 13, 1975, U.S. Department of
Commerce CONF-750989, Mar 1976; pp. II-209II-233.
Bauer, A. A.; Kangilaski, M. J. Nucl. Mater. 1972, 42,
9195.
Greenwood, L. R.; Kneff, D. W.; Skowronski, R. P.;
Mann, F. M. J. Nucl. Mater. 1984, 123, 10021010.
Hodgson, W. H. Final Report on Program for Using
X-8001 Aluminum Alloy Cladding Material for Hanford
Fuel Elements PT-IP-43-A-84-MT, IP-80-A-91-FP and
IP-2-I-99-FP, HW-65754; General Electric Company:
Richland, WA, July 22, 1960.
Cunningham, J. E. Severe Radiation Damage to
Aluminum Alloys; ORNL-TM-2138; Oak Ridge National
Laboratory: Oak Ridge, TN, Mar 1968.
Lotts, A. L.; et al. Analysis of Failure of HFIR Target
Elements Irradiated in SRL and in HFIR An Interim
Status Report; ORNL-TM-2236; Oak Ridge National
Laboratory: Oak Ridge, TN, Feb 1972.

Performance of Aluminum in Research Reactors


71.

72.
73.
74.
75.
76.

77.
78.
79.

80.
81.

82.
83.
84.
85.
86.
87.
88.
89.

Farrell, K. ORNL Studies of Radiation Damage in


Aluminum and Its Alloys: A Status Report (December
1973); Intra-Laboratory Correspondence UCN-430
(3 5-61); Oak Ridge National Laboratory: Oak Ridge, TN,
Jan 9, 1974.
Farrell, K.; Stiegler, J. O.; Gehlbach, R. E. Metallography
1970, 3, 275284.
Jostsons, A.; King, R. T. Scripta Metall. 1972, 6, 447452.
King, R. T.; Jostsons, A. Metall. Trans. A 1975, 6A,
863868.
Farrell, K. J. Nucl. Mater. 1981, 97, 3343.
Lyjbrink, E.; van Grol, H. J.; Dekker, F.;
van Witzenburg, W. In Effects of Radiation on Materials:
Eleventh International Symposium, Philadelphia, PA
1982; ASTM STP 782, pp 765778.
Farrell, K.; Bentley, J.; Braski, D. N. Scripta Metall. 1977,
11, 243248.
Lide, D. R. Ed. CRC Handbook of Chemistry and Physics,
86th ed.; Taylor & Francis: Boca Raton, FL, 20052006.
Farrell, K. In Proceedings of the Conference on
Dimensional Stability and Mechanical Behaviour of
Irradiated Metals and Alloys, Brighton, Apr 1113;
British Nuclear Energy Society: London, Vol. 1 1983;
7376.
Risbet, A.; Levy, V. J. Nucl. Mater. 1973, 46(3), 341352.
Yoshida, H.; Kozuka, T.; Sagane, T. Reactor irradiation
effects on Al 1100. In Proceedings of the 24th Japan
Congress on Materials Research Metallic Materials,
Kyoto University, Japan 1981; pp 16.
van Witzenburg, W.; Mastenbroek, A. J. Nucl. Mater.
1985, 133&134, 553557.
Munitz, A. J. Nucl. Mater. 1989, 165, 305312.
Marchbanks, M. F. Advanced Neutron Source Materials
Handbook; ORNL/M-4582; Oak Ridge National
Laboratory: Oak Ridge, TN, Aug 1995.
Farrell, K. Materials Selection for the HFIR Cold Neutron
Source; ORNL/TM-99-208; Oak Ridge National
Laboratory: Oak Ridge, TN, Aug 2001.
Maddin, R.; Cottrell, A. H. Philos. Mag. Ser. 7 1955,
46(38), 735743.
Mori, T.; Meshii, M. Acta Metall. 1969, 17, 167175.
Bochirol, L.; Brauns, P.; Claudet, G. Prog. Refrigeration
Sci. Technol. 1973, 1, 643650.
Weeks, J. R.; Czajkowski, C. J.; Tichler, P. R. In Effects of
Radiation on Materials: 14th International Symposium;

90.

91.

92.

93.
94.
95.

96.
97.

98.

99.
100.
101.
102.

175

American Society for Testing and Materials: West


Conshohocken, PA, 1990; Vol. II, ASTM STP 1046,
pp 441452.
Farrell, K. A spectral effect on phase evolution in
neutron-irradiated aluminum? In Microstructure of
Irradiated Materials, The Materials Research Society
Symposium Proceedings; 1995; Vol. 373, pp 165170.
Wallack, S. The Effect of Radiation on the Physical and
Mechanical Properties of Metals and Alloys; WADC
Technical Report 58-605, ASTIA Document No. AD
215540; Feb 1959.
Gronbeck, H. D. ETR Radiation Damage Surveillance
Programs, Progress Report II; IN-1036, Radiation Effects
on Materials TID-4500; Idaho Nuclear Corporation
Report; Feb 1967.
Farrell, K.; Mahmood, S. T. Tensile properties of neutron
irradiated aluminum alloys 2024 and 7075. Paper in
preparation.
Lohmann, W.; Ribbens, A.; Sommer, W. E.; Singh, B. N.
Radiat. Eff. 1986, 101, 283299.
Singh, B. N.; Lohmann, W.; Ribbens, A.; Sommer, W. F.
In Radiation-Induced Changes in Microstructure: 13th
International Symposium (Part1); American Society for
Testing and Materials: Philadelphia, PA, 1987; ASTM
STP 955, pp 508519.
Singh, B. N.; et al. J. Nucl. Mater. 1986, 141143,
743747.
Sommer, W. F.; Stubbins, J. F. Los Alamos National
Laboratory Accelerator Production of Tritium Project
B&R GB0508302: Topical Report Materials Safety
Experiment; LANL Report LA-UR-93-2850;
Aug 10, 1993.
Dunlap, J. A.; Borden, M. J.; Sommer, W. F.; Stubbins, J. F.
In Effects of Radiation on Materials: 17th International
Symposium; American Society for Testing and Materials:
West Conshohocken, PA, 1996; ASTM STP 1270,
pp 10471056.
Ismail, Z. H.; Mohammed, H. G. Scripta Metall. 1989, 23,
20672072.
Ismail, Z. H. Radiat. Eff. Defects Solids 1990, 112,
105110.
Ismail, Z. H.; Birt, B. J. Nucl. Mater. 1995, 218,
289292.
Munitz, A.; Shtechman, A.; Cotler, C.; Talianker, M.;
Dahan, S. J. Nucl. Mater. 1998, 252, 7988.

5.08

Irradiation Assisted Stress Corrosion Cracking

P. L. Andresen
GE Global Research Center, Schenectady, NY, USA

G. S. Was
University of Michigan, Ann Arbor, MI, USA

2012 Elsevier Ltd. All rights reserved.

5.08.1

Introduction

177

5.08.2
5.08.2.1
5.08.2.2
5.08.3
5.08.3.1
5.08.3.2
5.08.4
5.08.4.1
5.08.4.2
5.08.4.2.1
5.08.4.2.2
5.08.4.2.3
5.08.4.3
5.08.5
References

Irradiation Effects on SCC: Laboratory and Plant Data


Individual Effects of Radiation on IASCC
Service Experience
Irradiation Effects on Water Chemistry
Radiolysis and Its Effect on Corrosion Potential
Effects of Corrosion Potential on IASCC
Irradiation Effects on Microchemistry and Microstructure
Radiation-Induced Segregation
Microstructure, Radiation Hardening, and Deformation
Irradiated microstructure
Radiation hardening
Deformation mode
Radiation Creep and Stress Relaxation
Summary

180
180
183
187
187
189
190
190
194
194
196
198
201
202
202

Abbreviations
AES
AGR
BWR
CT
CW
DPA
FEGSTEM
FWHM
HWC
HWR
IASCC
IGSCC
NWC
PWR
RIS
RH
SCC
SFE
SGHWR
SS
SSRT
STEM

Auger electron spectroscopy


Advanced gas cooled-reactor
Boiling water reactor
Compact type (specimen)
Cold work
Displacements per atom
Field emission gun TEM
Full-width half-max (for profiles)
Hydrogen water chemistry (in BWRs)
Heavy water reactor
Irradiation-assisted stress corrosion
cracking
Intergranular stress corrosion cracking
Normal water chemistry (in BWRs)
Pressurized water reactor
Radiation-induced segregation
Radiation hardening
Stress corrosion cracking
Stacking fault energy
Steam generating heavy water reactor
Stainless steel
Slow strain rate test
Scanning transmission electron
microscopy

TEM
TG

Transmission electron microscopy


Transgranular

5.08.1 Introduction
Nuclear power accounts for about 17% of the worlds
electricity production, and the rapid expansion in
nuclear power throughout the world will necessitate
that they operate with high reliability and safety.
Stress corrosion cracking (SCC) has occurred in all
water cooled reactors, including boiling-water reactors (BWRs) and pressurized-water reactors (PWRs),
with a greater incidence in unirradiated, out-of-core
components, especially between 1970 and 1990. As
these materials, component designs, and water chemistries have improved, an increasing percentage of
cracking problems has occurred in irradiated components. While irradiation-assisted stress corrosion
cracking (IASCC) has been observed since early
plant operation, increasing operating time and fluence has led to an increased incidence of cracking.
Setting aside zircaloy fuel cladding and pressure
vessel steels, most irradiated core components consist
of austenitic stainless steels and nickel-base alloys
177

178

Irradiation Assisted Stress Corrosion Cracking

exposed to environments that span oxygenated to


hydrogenated water at 270340  C. The core of a
nuclear reactor is an extreme environment consisting
of high-temperature water, imposed stresses and
strains, and an intense radiation field that affects
the water chemistry, stress, and microstructure of
the core materials (Figure 1). For background, the
reader is referred to Chapter 1.03, RadiationInduced Effects on Microstructure; Chapter 1.05,
Radiation-Induced Effects on Material Properties
of Ceramics (Mechanical and Dimensional), and
Chapter 1.07, Radiation Damage Using Ion Beams
for more detailed treatments of radiation effects on
materials. Initially, the affected components were primarily small components (bolts, springs, etc.) or components designed for replacement (fuel rods, control
blades, or instrumentation tubes). However, in the
last 20 years, IASCC has been observed in structural
components (e.g., PWR baffle bolts and BWR core
shrouds and top guides).
Extensive literature exists for SCC under unirradiated conditions, and the basic factors and dependencies are well defined and reasonably well
modeled for austenitic stainless steel and nickel alloys
(e.g., Alloys 600 and its weld metals).19 A complete
consensus on the underlying mechanism of cracking
has not emerged although the well-behaved continuum in crack growth rate response versus material/
composition (including from stainless steels to nickel
Solution renewal
rate to crack-tip
Stress

f Anionic
transport

Oxide rupture
rate at
crack-tip

Environment

Microstructure

g-field
Crack tip f [A], pH
Passivation rate
at crack-tip
Grain boundary denudation

Hardening
Relaxation

N-fluence
Segregation

Figure 1 Schematic of the primary engineering


parameters that effect stress corrosion cracking stress,
microstructure, and environment and the underlying crack
tip processes that control stress corrosion cracking. The
primary ways in which radiation affects stress corrosion
cracking is also shown: segregation, hardening, relaxation,
and radiolysis. Radiolysis can increase the corrosion
potential, which in turn increases the potential gradient
() and the crack tip potential , anion concentration
[A], and pH.

alloys), water chemistry, temperature, and radiation


suggests that a common crack growth mechanism is
operative.813
Our understanding has evolved from the view that
SCC occurs under very specific and unique conditions
to the view that a continuum in response exists.813 With
steady improvement in laboratory and plant detection
of SCC, it is clear that SCC occurs under a wide range
of conditions and also at a wide range of growth rates.
Figures 2 and 3 show examples of the effect of environment (corrosion potential and water purity), material
condition (sensitized vs. cold-worked), and stress (stress
intensity factor) on SCC growth rates; the solid curves
are the predicted response.8,9,12,13 SCC occurs even at
low corrosion potential (Figure 2), and thus the behavior in BWRs and PWRs is linked, with the primary
differences being dissolved H2, temperature, and the
dissolved ion chemistry (B and Li are added to PWR
primary water; see Chapter 5.02, Water Chemistry
Control in LWRs).810,14 Of these, temperature has a
universal effect, variations in dissolved H2 are particularly important in nickel alloys, and B/Li has little or
no effect in deaerated water.9,14
Early plant (Figure 4) and laboratory (Figure 5)
observations showed that the same basic dependencies
existed for unirradiated and irradiated stainless steels,
and that increasing fluence produces a well-behaved
increase in SCC susceptibility (Figure 6). Figure 4
shows a strong effect of water purity for both unirradiated and irradiated BWR components, and Figure 5
shows a very similar response to corrosion potential to
that in Figure 2. Thus, it was proposed that radiation
enhances SCC primarily in four ways: segregation,
hardening, relaxation, and radiolysis (Figure 1). The
neutron fluence where these processes have an effect is
shown in Figure 7, along with the current end-of-life
fluence for various BWR and PWR components. The
primary radiation effects on materials operate in a
similar range of fluences, and thus their individual
contributions can be difficult to distinguish. An example of their interaction in altering SCC growth rate is
shown in the prediction of cracking of a weld in a BWR
core shroud (Figure 8) in which the individual effects
are plotted along with the resulting crack length versus
time. While many of the enhancements in SCC susceptibility from irradiation dose (neutron fluence)
have been well established, it remains possible that
additional factors will emerge at high fluences (e.g.,
>30 displacements per atom (dpa)).
Intergranular (IG) SCC is promoted in austenitic stainless steels above a threshold fluence

179

Irradiation Assisted Stress Corrosion Cracking

1.0E05
1

25 mm CT specimen
Furnace sensitized; 15 C cm2
288 C water
; 0.10.3 mS cm1
Constant load
; 25 Ksi in1/2

Crack growth rate (mm s1)

42.5 in h1

107

10

1.0E06

11

Crack growth rate (mm s1)

106

Sensitized 304 stainless steel


30 MPa m1/2, 288 C water
0.060.4 S cm1, 025 ppb SO4
filled triangle = constant load
open squares = gentle cyclic

14

14.2 in h1
Theoretical
curves

S cm1

0.3
0.2

0.1

108

200 ppb O2
500 ppb O2
2000 ppb O2

304 Stainless steel

3 7
4

Screened round robin data


- highest quality data
- corrected corr. potential
- growth rates corrected
to 30 MPa m1/2

42.5
28.3
14.2
1
in h

1.0E07
GE pledge
predictions
30 MPa m1/2

0.5
2000 ppb O2
Ann. 304SS
200 ppb O2

0.25
1.0E08
0.1

12

0.06 S cm1

Hydrogen water
chemistry

Normal water
chemistry
(ex-core)

600

400
200
0
+200
Corrosion potential (mVshe)

+400

30 MPa m1/2
1.0E09
0.6 0.5 0.4 0.3 0.2 0.1 0 0.1
Corrosion potential (Vshe)

1.0E05
Sensitized 304 stainless steel
30 MPa m1/2, 288 C water
0.060.4 S cm1, 025 ppb SO4
SKI round robin data
filled triangle = constant load
open squares = gentle cyclic

200 ppb O2
500 ppb O2
2000 ppb O2

109

0.06 S cm
Industry mean

0.2

0.3

0.4

4 dpa
304SS

Crack growth rate (mm s1)

1.0E06
316L (A14128, square)
304L (Grand gulf, circle)
nonsensitized SS
50% RA 140 C (black)
10% RA 140 C (gray)
1.0E07

20% CW
A600
42.5
28.3
14.2
in h1

20% CW A600
GE pledge
predictions
30 MPa m1/2
Sens SS

0.5
2000 ppb O2
Ann. 304SS
200 ppb O2

0.25

1.0E08

0.1

0.1 S cm
Means from analysis of
120 lit. sens SS data

0.06 S cm1

0.06 S cm1

GE pledge predictions for


Unsens. SS (upper curve for 20% CW)

1.0E09
0.6 0.5 0.4 0.3 0.2 0.1 0 0.1
Corrosion potential (Vshe)

0.2

0.3

0.4

Figure 2 Stress corrosion cracking growth rate versus corrosion potential for stainless steels tested in high-purity water
at 288  C containing 2000 ppb O2 and 953000 ppb H2. Dissolved O2 strongly influences corrosion potential, which in turn
affects crack chemistry and growth rate of sensitized stainless steels (two graphs at left) as well as cold-worked stainless
steels and Alloy 600 (large rectangular symbols on right graph) and irradiated stainless steel (large triangular symbols).
Cold-worked or irradiated materials have an elevated yield strength, which causes an increase in growth rate at both low
and high potential. RA, Reduction in area; CW, Cold work.

180

Irradiation Assisted Stress Corrosion Cracking

105

Stress intensity (ksi in1/2)


6 8 10
20 30 40 60 80

103
Sens. 304 stainless steel
288 C water

104

107

NRC disposition
line

105

*
108

Theory
15 C cm2, 50 mVshe
0.5 ms cm1
Theory
2
15 C cm , 50 mVshe
0.2 ms cm1

109

106

Theory
15 C cm2, 200 500 mVshe
0.2 ms cm1

1010

Crack growth rate (in h1)

Crack growth rate (mm s1)

106

107

8 10
20 30 40 60 80
Stress intensity (MPa m1/2)

Figure 3 Effect of stress intensity factor on stress


corrosion cracking growth rate for sensitized stainless steel
exposed in various water chemistries at 288  C.

(Figures 5 and 6). This occurs in oxygenated (e.g.,


BWR) water above 25  1020 n cm2 (E > 1 MeV),
which corresponds to about 0.30.7 dpa, and depends
on the stress, water chemistry (especially, sulfate and
chloride), and other factors. Attempts to reproduce the
same level of IG cracking in inert environments have
been unsuccessful, confirming that it is an environmental cracking phenomenon, not simply a change in
the mechanical properties and response of the material
in an inert environment.
Cracking in hydrogenated water (i.e., BWR hydrogen water chemistry (HWC) or PWR water) is typically observed at roughly a 4 higher fluence than in
oxidizing water, with IASCC enhanced at elevated
temperature (Figure 9). For both BWR and PWR
conditions, the same basic dependencies exist for
unirradiated and irradiated materials.
It is important to distinguish the results of different
kinds of SCC testing. Crack-growth testing typically
uses fracture mechanics specimens, commonly a compact type (CT) specimen. It has the significant

advantage of continuous, online monitoring of crack


length versus time, usually by employing an electric
current, potential drop technique. It can provide a
resolution of about 1 mm and can accurately characterize the inherent resistance to crack advance (i.e., beyond
the ill-defined initiation stage) as well as the dependencies on corrosion potential, stress intensity factor,
etc. Smooth specimen tests, whether by constant load,
constant displacement, or slow strain rate (SSR), are
simpler tests to perform, but represent some combination of initiation and growth. SSR tests impose failure
and can overstate or understate the SCC susceptibility.
Constant load or displacement tests generally require
periodic interruption for examination, and initiation
can be microscopic cracks or complete failure.
Work over the last 25 years has enabled many aspects
of IASCC phenomenology to be explained and predicted on the basis of the experience with intergranular
stress corrosion cracking (IGSCC) of nonirradiated
stainless steel in high-temperature water environments.
This continuum approach has successfully accounted
for radiation effects on water chemistry and its influence on electrochemical corrosion potential. However,
all radiation-induced microstructural and microchemical changes that promote IASCC are neither fully
known nor fully reproducible in similar materials.
Well-controlled data from well-characterized irradiated
materials remain insufficient due to the inherent experimental difficulties and financial limitations. Many of
the important metallurgical, mechanical, and environmental aspects that are believed to play a role in the
cracking process are illustrated in Figure 1. Only persistent material changes are required for IASCC to
occur, but in-core processes such as radiation creep
and radiolysis also have an important effect on IASCC.

5.08.2 Irradiation Effects on SCC:


Laboratory and Plant Data
5.08.2.1 Individual Effects of Radiation
on IASCC
IASCC can be categorized into radiation effects on the
water chemistry (radiolysis) and on the material/stress,
and the accepted definition of IASCC encompasses
cases where either factor is dominant (low-fluence
materials tested in water undergoing radiolysis, or preirradiated materials tested without an active radiation
flux). Radiation dose rate in rads/h is often used in
radiolysis, and in neutrons per square centimeter
(n cm2) or displacements per atom in materials.
In light water reactors (LWRs), 1 dpa corresponds

Frequency of SCC initiation


increases dramatically with
increasing conductivity

0.2

Best fit

1.2
1.0
0.8
0.6

Threshold
conductivity for
SCC initiation
increases as level
of sensitization
decreases
Low carbon SS
High carbon SS

Nonsensitized
low carbon SS

0.2
0
0

0.2

(b)

1.0
0.9

Sensitized
high carbon SS

0.4

0.6

Plant average conductivity (ms cm1)

0.4

0.6

Plant average conductivity (ms cm1)

0.8

Frequency of IGSCC
initiation increases
with plant conductivity

0.7
0.6
0.5

0.4

* Experienced
substantial high
conductivity
excursions not
reflected in
average value

0.3
0.2

0.1
0

(c)

181

1.4
Upper bound

0.4

% with IGSCC/on-line months

(a)

1.5
1.4
1.3
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0

% with IGSCC/on-line months

% with IGSCC/on-line month

Irradiation Assisted Stress Corrosion Cracking

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Plant average conductivity (ms cm1)

Figure 4 The effects of average plant water purity shown in field correlations of the core component cracking behavior for
(a) stainless steel intermediate and source range monitor dry tubes, (b) creviced stainless steel safe ends, and (c) creviced
Inconel 600 shroud head bolts, which also shows the predicted response versus conductivity. Adapted from Brown, K. S.;
Gordon, G. M. In Proceedings of 3rd Environmental Degradation of Materials in Nuclear Power Systems Water Reactors;
The American Institute of Mining, Metallurgical, and Petroleum Engineers (AIME): New York, NY, 1988; pp 243248.

to 7  1020 n cm2 when counting neutrons with


E > 1 MeV, or 1.5  1021 n cm2 for E > 0.1 MeV.
The primary effects of radiation on materials8,1522
include microcompositional effects (grain boundary
chemistry) and microstructural changes (formation of
dislocation loops, voids, precipitates, and the resulting changes hardening and deformation mode). In
terms of their effect on IASCC, the primary effects
of radiation are the following:
 Radiolysis of water, in which a variety of short- and
long-lived radicals and species are produced.
There is no evidence that the specific species
formed are important, and indeed their effect on
cracking appears to be fully captured by their overall effect on the corrosion potential of the material.
 Radiation-induced segregation (RIS), which produces an enrichment in some species (e.g., Ni and
Si) at grain boundaries and other defect sinks, and a
depletion in other species (e.g., Cr). Even though the
distance over which RIS occurs is very limited (a few

nanometers), studies of unirradiated materials have


shown that the narrow profiles can affect SCC.23,24
 Radiation hardening (RH), which results from radiation damage and the creation of vacancy and interstitial loops, which impede dislocation motion. Once
a few dislocations move along a given slip plane, they
clear the channel of most of these barriers, and
subsequent dislocation occurs primarily in these
channels.
 Radiation creep relaxation, which reduces constant displacement stresses such as in bolts or associated with weld residual stress. During active
irradiation, radiation creep can promote dynamic
strain, and thereby SCC.
 Swelling, which occurs to a limited extent at temperatures above 300  C, but can be sufficient to
produce reloading of components such as PWR
baffle former bolts. Swelling occurs differently in
different materials, and is delayed in cold-worked
materials. Stresses due to swelling are balanced by

182

Irradiation Assisted Stress Corrosion Cracking

Data of Jacobs( ) and Kodama( )


Postirrad. SSRT 23 107 s1 288 C
Comm. purity 304SS( ) and 316SS( )
42 ppm O2-satd vs.
0.02 ppm O2

100

80

100

40

42

ppm O2

60
%IGSCC

% IGSCC fracture

Data shifted right by Init. grain boundary Cr enrichment

0.2
0.02
0.05

fC, Vshe

20
0.4

0
1019

0.2

0.2

1020
1021
Neutron fluence (n cm2 ) (E > 1 MeV)

1022

Percentage of spot welds inspected with IGSCC

Figure 5 Dependence of irradiation-assisted stress corrosion cracking on fast neutron fluence as measured in slow strain
rate tests at 3.7  107 s1 on preirradiated type 304 stainless steel in 288  C water. The effect of corrosion potential via
changes in dissolved oxygen is shown at a fluence of 2  1021 n cm2. Reproduced from Jacobs, A. J.; Hale, D. A.; Siegler, M.
Unpublished Data on SCC of Irradiated SS in 288  C Water and Inert Gas; GE Nuclear Energy: San Jose, CA, 1986.
SSRT, Slow strain rate test.
PWR control
BWR core BWR end rod failures
(IASCC)
component
of life
failures
(IASCC)

100

80

1020

BWR creviced control blade sheath


60
Threshold fluence for IGSCC
5 1020 n cm2

40

0.1

0
0

0.2

0.4

0.6

0.8

1.2

1.4

Neutron fluence (n cm2 1021) (E > 1 MeV)

Figure 6 Dependence of irradiation-assisted stress


corrosion cracking on fast neutron fluence for creviced control
blade sheath in high-conductivity boiling-water reactors.
Reproduced from Gordon, G. M.; Brown, K. S. In Proceedings
of 4th International Conference on Environmental Degradation
of Materials in Nuclear Power Systems Water Reactors;
NACE: Houston, TX, 1990; pp 14-4614-62.

radiation creep relaxation, but the resulting stress


can be sufficient to cause IASCC.
 Other microstructural changes, such as precipitation or dissolution of phases in materials. While
there is no clear evidence that such changes affect
IASCC response, this may only reflect the limited

PWR end
of life
PWR life
extension

1021
1022
1023
Neutron fluence (n cm2) (E >1 MeV)
Irradiation dose (dpa)
1
10
Significant changes
in grain boundary
composition, alloy
strength, and ductility

20

PWR baffle
bolt failures
(IASCC)

100
Onset of significant
void swelling
and possible
embrittlement

Figure 7 Neutron fluence effects on irradiation-assisted


stress corrosion cracking susceptibility of type 304SS
in boiling-water reactor environments. Reproduced
from Bruemmer, S. M.; Simonen, E. P.; Scott, P. M.;
Andresen, P. L.; Was, G. S.; Nelson, J. L. J. Nucl. Mater.
1999, 274, 299314.

characterization and IASCC studies that has been


performed on high fluence materials.
The individual (segregation, hardening, and creep,
Figure 10) and composite (SCC, Figures 59) effects
of radiation increase with dose in much the same
manner, which makes the isolation of, and attribution
to, individual contributions difficult. The dislocation
loop microstructure is closely tied to radiation

Irradiation Assisted Stress Corrosion Cracking

30

20

30
25

Effect of
rad segregation

20

Effect of rad hardening

15
10

Depth
10

Stress intensity (K)

Effect of stress relaxation


0
0

100

200
300
Time (month)

Fraction of IG cracking area

40

In PWR primary water


CW316, 340 C
CW316, 325 C
CW316, 290 C
Type 304, 325 C

*
*
1.0E + 20

20
Loop line
length
SCC

15

10

1.0E + 21
1.0E + 22
Fluence (n cm2) (E > 0.1 MeV)

CP-304SS
3.2 MeV protons
360 C

0
0

3
Dose (dpa)

Figure 10 Schematic diagram showing the dose


dependence of key irradiated microstructure features
(radiation-induced segregation and dislocation
microstructure) and radiation hardening along with stress
corrosion cracking susceptibility. Reproduced from
Was, G. S. In Proceedings of 11th International
Conference on Environmental Degradation of Materials
in Nuclear Power Systems Water Reactors; American
Nuclear Society (ANS): La Grange Park, IL, 2004;
pp 965985.

irradiation. Subsequent sections focus on the possible


mechanisms by which water chemistry, RIS, microstructure, hardening, deformation mode, and irradiation creep individually or on concert may affect
IASCC.

* Specimen broke at the pin hole

20

Cr
Cr depletion

100

60

25

0
500

400

Figure 8 Predicted effect of radiation segregation,


radiation hardening, and radiation creep relaxation on a
boiling-water reactor core shroud, where the through-wall
weld residual stress profile is the primary source of
stress. Less aggressive water chemistry (corrosion
potential and water purity) would result in less crack
advance early in life, which would give a greater
opportunity for radiation creep relaxation. The leak depth
is the wall thickness of the shroud. While radiation
hardening continues to increase the yield strength, its effect
on crack growth is reduced (see Figure 21(a)). EPR,
Electrochemical potentiokinetic repassivation (a test for
sensitization).

80

Hardness

Arbitrary units

Crack depth (mm)

Leak depth

38.1-mm thick 304SS, two-sided weld


0.75 Vshe, 0.15 S cm1
EPR0 = 10.8 C cm2 (0.050% C)
Flux = 3 1019 n cm2 year

K, segregation, hardening, or relaxation

40

30

183

1.0E + 23

Figure 9 Percentage intergranular stress corrosion


cracking versus fluence for cold-worked type 316 stainless
steel tested at various temperatures. Despite the low
potential environment of pressurized-water reactor primary
water, at high fluence (especially at higher temperature)
there is significant susceptibility to stress corrosion
cracking.

hardening and both increase with dose until saturation occurs by 5 dpa. RIS also increases with dose
and tends to saturate by 5 dpa. Although dependent
on both metallurgical and environmental parameters,
IASCC generally occurs at doses between 0.5 dpa (for
BWRs) and 23 dpa (for PWRs), which encompasses
the steeply rising portion of the curves in Figure 4
that describe the changes in materials properties with

5.08.2.2

Service Experience

The IASCC service experience extends over 50 years,


and the early observations projected an accurate view
of the important characteristics and dependencies,
and pointed to the assisted nature of radiation in
enhancing SCC. As with other forms of SCC, early
observations suggested that a growing incidence with
time and neutron fluence should be expected.
IASCC was first reported in the early
1960s7,8,1522,2536 and involved intergranular cracking of stainless steel fuel cladding, with the initiation of multiple cracks occurring from the water side.
By contrast, mostly ductile, transgranular cracking
was observed in postirradiation mechanical tests
performed in inert environments at various strain
rates and temperatures. Grain boundary carbide

184

Irradiation Assisted Stress Corrosion Cracking

precipitation was rarely observed although preexisting thermal sensitization was present in some cases.
A correlation between time-to-failure and stress level
was reported, with failure occurring first in thinwalled rods with small fuel-to-cladding gaps, where
fuel swelling strains were the largest. The highest
incidence of cracking occurred in peak heat flux
regions, corresponding to the highest fluence and
the greatest fuelcladding interaction (highest stresses and strains.) Similar stainless steel cladding in
PWR service exhibited fewer instances of intergranular failure. At that time, off-chemistry conditions or
stress rupture were often considered to be the cause
of PWR failures.
In the last 30 years, a growing number of other
stainless steel (and nickel alloy) core components
have exhibited IASCC, including neutron source
holders in 1976 and control rod absorber tubes in
1978. Instrument dry tubes and control blade handles
and sheaths, Figure 5, which are subject to very low
stresses, are also cracked, generally in creviced locations and at higher fluences.15,18,20,37,38 These initial
failures in the most susceptible components were

Table 1

followed by more numerous incidents of IASCC in


the past 20 years, perhaps most notably in BWR core
shrouds8,15,16,39 and PWR baffle bolts.19,40,41
A summary of reported failures of reactor internal
components is listed in Table 1 and demonstrates
that IASCC is not confined to a particular reactor
design, material, component, or water chemistry. For
example, stainless steel fuel cladding failures were
reported years ago in commercial PWRs and in PWR
test reactors.15,22,25,2936,42 At the West Milton PWR
test loop, intergranular failure of vacuum-annealed
type 304 stainless steel fuel cladding was observed31
in 316  C ammoniated water (pH 10) when the cladding was stressed above yield. Similarly, IASCC was
observed in creviced stainless steel fuel element ferrules in the Winfrith steam generating heavy water
reactor (SGHWR),43,44 a 100 MWe plant in which
light water is boiled in pressure tubes, where the coolant chemistry is similar to other boiling-water reactor
designs. The 20% Cr25% NiNb stainless steel differs from type 304 primarily in Ni and Nb content, as
well as in its lower sulfur (0.006%) and phosphorus
(0.005%) contents. The ferrules were designed for

Irradiation-assisted stress corrosion cracking service experience

Component

Material

Reactor type

Possible sources of stress

Fuel cladding
Fuel cladding
Fuel claddinga
Fuel cladding ferrules
Neutron source holders
Instrument dry tubes
Control rod absorber tubes
Fuel bundle cap screws
Control rod follower rivets
Control blade handle
Control blade sheath
Control blades
Plate type control blade
Various boltsb
Steam separator dryer boltsb
Shroud head boltsb
Various bolts
Guide tube support pins
Jet pump beams
Various springs
Various springs
Baffle former bolts
Core shroud
Top guide

304SS
304SS
20% Cr25% NiNb
20% Cr25% NiNb
304SS
304SS
304/304L/316L SS
304SS
304SS
304SS
304SS
304SS
304SS
A-286
A-286
600
X-750
X-750
X-750
X-750
718
316SS cold work
304/316/347/L SS
304SS

BWR
PWR
AGR
SGHWR
BWR
BWR
BWR
BWR
BWR
BWR
BWR
PWR
BWR
PWR and BWR
BWR
BWR
BWR and PWR
PWR
BWR
BWR and PWR
PWR
PWR
BWR
BWR

Fuel swelling
Fuel swelling
Fuel swelling
Fabrication
Welding and Be swelling
Fabrication
B4C swelling
Fabrication
Fabrication
Low stress
Low stress
Low stress
Low stress
Service
Service
Service
Service
Service
Service
Service
Service
Torque, differential swelling
Weld residual stress
Low stress (bending)

Cracking in AGR fuel occurred during storage in spent fuel pond.


Cracking of core internal occurs away from high neutron and gamma fluxes.
AGR, Advanced gas-cooled reactor
b

Irradiation Assisted Stress Corrosion Cracking

a 5-year exposure during which the peak fast neutron


flux is 23  1013 n cm2 s (E > 1.5 MeV), yielding a
peak fluence over 5 years of 35  1021 n cm2.
The similarity of IASCC in BWRs and PWRs was
also noted in swelling tube tests performed in the
core44,45 on a variety of commercial and high-purity
heats of types 304, 316, and 348 stainless steel and
Alloys X-750, 718, and 625. Swelling was controlled
by varying the mix of Al2O3 and B4C within the
tubes; the latter swells as neutrons transmute B to
He. Nominally identical strings of specimens were
inserted into the core in place of fuel rods.
Historically, the oxidizing potential in a PWR
core is lower than in BWRs, but in the last decade,
most BWRs employ an electrocatalytic technology
called NobleChem(4648) to create a low corrosion
potential. Some early investigators attributed PWR
cracking to low ductility stress rupture (of course,
this mechanism would apply equally to BWRs).
A few laboratory studies reported small amounts of
intergranular cracking of irradiated stainless steels in
SSR tests in 300  C inert environments49 although
in many related experiments50,51 no intergranular
failure was found. Small amounts of intergranular
cracking in inert tensile tests are not surprising, and
since the early 1990s, the plant and laboratory
IASCC data show that cracking is environmentally
assisted and follows a well-behaved continuum in
response over ranges in fluence, corrosion potential,
temperature, stress, etc.8,1517
Factors that distinguish PWRs from BWRs
include their higher operating temperature, 10
higher maximum neutron fluence in core structural
components, higher hydrogen fugacity, and borated
lithiated water chemistry (including the possibility of
localized boiling and thermal concentration cells in
crevices from gamma heating, which could lead to
aggressive local chemistries). The possible role of RIS
of Si may be especially important in accounting for
the limited difference in SCC response at high
potential (BWR) versus low potential (PWR) at
high fluence.52,53
Brown and Gordon37,38 (Figure 4) accumulated
and analyzed data for cracking in Alloy 600 shroud
head bolts (first observed in 1986) as well as stainless
steel safe ends (first observed in 1984) and in-core
instrumentation tubes (first observed in 1984) with a
focus on components that were creviced, a factor
known to exacerbate cracking.7,37,38 The highest radiation exposure occurred for the intermediate range
and source range monitor (IRM/SRM) dry tubes,
which contain flux monitors housed in thin-walled,

185

annealed stainless steel tubes. Cracking initiated


in the crevice between the spring housing tube
and the guide plug at fluences between 0.5 and
1.0  1022 n cm2 (E > 1 MeV). Wedging stresses
from the thick oxide observed in the crevice were
implicated, since other (applied and residual) stresses
were negligible. The primary variable from plant
to plant is the average coolant conductivity, which
correlates strongly with cracking incidence
(Figure 4(a)4(c)). Each point in Figure 4 represents
inspection results for one BWR plant, and data are
normalized using reactor operating time (i.e., percentage of components with intergranular cracks
divided by the online exposure time). The scatter in
Figure 4(a) was attributed to variations in fluence
and specific ion chemistry, as well as limitations in
the resolution of underwater visual inspection. Scatter can also result from short-term excursions in
conductivity, which is not adequately reflected in
the average, as identified in Figure 4(c).
Correlations between IASCC and conductivity
were also reported for cracking in shroud head bolts
(Figure 4(c)) and creviced safe ends (Figure 4(b)).
The strong influence of conductivity on cracking of
stainless steel has also been shown in laboratory tests
and plant recirculation piping, where predictive
modeling8,12,13,15,54,55 has been compared to field
data on the operational time required to achieve a
detectable crack (typically, 10% of the wall thickness). Preliminary prediction of the shroud head bolt
cracking response8 also provides reasonable agreement with observation (Figure 4(c)).
High-strength, nickel-base alloy components15,22
have also experienced IASCC (Table 1), with many
incidents in lower radiation flux regions (e.g., where
the end-of-life fluence is below 5  1019 n cm2)
such as cracking of Inconel X-750 jet pump beams
in BWRs. Inconel X-750 cracking has also occurred
extensively in PWR fuel hold-down springs, which
attain an end-of-life fluence of 110  1021 n cm2; it
is proposed that cracking has been aggravated by
vibrational stresses (corrosion fatigue). The effects of
irradiation on IASCC in high-strength, precipitationhardened nickel-base alloy components as well as in
stainless steels have not been characterized.
BWR core shrouds8,15,16,39 and PWR baffle
bolts40,41 are the two most common examples of
IASCC although susceptibility clearly exists in
other areas, such as control blade components, fuel
components, the BWR top guide, etc. SCC in the
BWR core shroud occurs almost exclusively near the
welds (both circumferential and vertical), and

186

Irradiation Assisted Stress Corrosion Cracking

initiation is observed from both the inside (ID) and


outside (OD) surfaces (the shroud separates the
upward core flow from the downward recirculation
flow that occurs in the annulus between the shroud
and the pressure vessel). This large-diameter welded
pipe has little active (DP) loading, and its susceptibility arises primarily from weld residual stresses and
weld shrinkage strains.5658 Cracking is observed in
both low fluence and moderate fluence areas, and the
extent of the enhancement in SCC susceptibility by
irradiation is limited because, while RH and RIS
occur, radiation creep also relaxes the weld residual
stress. SCC predictions for a BWR core shroud that
account for the damaging effect of RIS and RH and
the beneficial effect of radiation creep relaxation are
shown in Figure 8 and illustrate the complexity of
the interactions of these phenomena in the evolution
of cracking. Predictions also indicate that, if SCC
does not nucleate early in life (e.g., below 0.5 dpa),
for example, from high coolant impurity levels or
severe surface grinding, susceptibility tends to
decrease with fluence in the shroud welds because
of radiation-induced creep relaxation (although
many shroud welds are in very low flux areas).
The last decade has also seen extensive failures of
PWR baffle bolts40,41 although large plant-to-plant
and heat-to-heat differences are observed. Most baffle bolts are fabricated from type 316 stainless steel
cold-worked to 15% to increase their yield
strength. The complex baffle former structure exists
in a PWR because their fuel does not have a surrounding channel, so the baffle former structure
must conform closely to the geometry of the fuel to
provide well-distributed water flow. The baffle former plates are usually made from annealed material,
typically type 304 stainless steel. Because of their
proximity to the fuel, very high fluences can develop
up to 80 by the end of the original design life. The
high gamma flux produces significant heating in the
components, in some instances estimated at 40  C,
especially in designs where the PWR coolant does
not have good access to the bolt shank. While the
heat-to-heat variations are not understood, it is clear
that plants that load-follow (and therefore undergo
power level changes and thermal cycles) are much
more prone to baffle bolt cracking. Another aggravant
is the thermal gradient and possible boiling (resulting
in altered water chemistry) in the shank area of the
bolt if the coolant access is restricted. However, primary factors must be the very sizeable stress relaxation that occurs early in life (e.g., during the first

5 dpa), followed by preferential radiation swelling of


the annealed baffle plates over the cold-worked baffle
bolts, which will cause reloading. The dynamic equilibrium between swelling and radiation creep, which
determines the reloading stress in the bolt, is likely a
complex function of many parameters, including
local neutron flux, temperature, baffle plate geometry, and composition.
The number of IASCC incidents continues to
grow, and there can be no question that many LWR
components are susceptible. Strategies to mitigate
IASCC (e.g., NobleChem(4648)) and manage
IASCC (e.g., by showing some IASCC could be tolerated, installing mechanical restraints to mitigate
the impact of IASCC in BWR shrouds, or selectively
inspect and replace baffle bolts) have been successful.
IASCC field experience has led to the following
trends and correlations:
 Intergranular cracks associated with radiation
effects on solution-annealed stainless steel were
once thought to occur only at fluences above
0.3  1021 n cm2. But significant intergranular
cracking in BWR core shrouds (which do not have
thermal sensitization) occurs over a broad range
of fluences, showing that a true fluence threshold
does not exist.15,16 The observations of SCC in unirradiated, unsensitized stainless steel (with or without
cold work) also undermine the concept of a threshold fluence below which no SCC occurs. This also
holds for thresholds in corrosion potential, water
impurities, temperature, etc.8,54,59,60
 SCC susceptibility is affected by fluence in a complex fashion. SCC in BWR shrouds and PWR
baffle bolts does not always correlate strongly
with fluence, and one important reason for this is
that radiation creep produces relaxation of the
stresses from welding and in bolts. The need to
account for many changing factors is necessary in
interpreting and predicting SCC.
 Most early incidents involved high stresses or
dynamic strains, but cracking has since been
observed at quite low stresses at high fluences
and longer operating exposure. Laboratory and
field data indicate that IASCC occurs at stresses
below 20% of the irradiated yield stress, and at
stress intensities below 10 MPa m1/2.
 Extensive laboratory and field data show that corrosion potential is a very important parameter, with
its effect being consistent from low to high fluence,
except in some high fluence materials and/or

Irradiation Assisted Stress Corrosion Cracking





under high stress intensity factor conditions. Materials prone to high radiation-induced changes in
grain boundary Si level may exhibit a very limited
effect of corrosion potential.52,53 There is no evidence of threshold potential, and indeed irradiated
materials exhibit IASCC in deaerated water.
Impurities, especially chloride and sulfate, strongly
affect IASCC in BWR water (Figure 4). As noted
by Brown and Gordon,37 this correlation applies
equally to low and high flux regions and to stainless
steels (Figure 4(a) and 4(b)) and nickel-base alloys
(Figure 4(b)). Indeed, the correlation closely parallels that from out of core.8,12,13,15,38,54,55 At higher
levels, the same impurities can affect SCC in
PWRs. Similarly, if high corrosion potential conditions form in the PWR primary (where B and Li are
present), high growth rates can result.9,14
Crevices exacerbate cracking primarily due to
their ability to create a more aggressive crevice
chemistry from the gradient in corrosion potential
(in BWRs) or in temperature (most relevant to
PWRs). Crevices can also produce stress and strain
concentration.
Cold-working often exacerbates cracking (especially abusive surface grinding), although it can
also delay the onset of some radiation effects.
Temperature has an important effect on IASCC,
enhancing both crack initiation and growth rate.
Preexisting grain-boundary carbides or chromium
depletion is not required for susceptibility
although furnace-sensitized stainless steels are
clearly highly susceptible to cracking in-core. Cr
depletion will develop or be magnified by irradiation, and increase IASCC susceptibility, although
its effect is most pronounced in pH-shifted environments, as can develop when potential or thermal
gradients exist. The role of N, S, P, and other grainboundary segregants is less clear.
IASCC is enhanced at a fluence that is dependent
on applied stress and strain, corrosion potential,
solution conductivity, crevice geometry, etc. At
sufficiently high conductivities, cracking has been
observed in solution-annealed stainless steel in
the field (Figure 6(a) and 6(b))37 and in the laboratory.8,12,13,15,54,55 Thus, while convenient in
a practical engineering sense, the concept of a
threshold fluence (or stress, corrosion potential,
etc.) is scientifically misleading8,52,59,60; cracking
susceptibility and morphology are properly considered an interdependent continuum over many
relevant parameters.

187

5.08.3 Irradiation Effects on Water


Chemistry
5.08.3.1 Radiolysis and Its Effect on
Corrosion Potential
SCC susceptibility is fundamentally influenced by
corrosion potential, not by the oxidant and reductant
concentrations per se.8,15,61 The lower H2 concentration in BWRs distinguishes BWRs from PWRs
because it permits the radiolytic formation of oxidants. Above 500 ppb (5.6 cm3 kg1) H2, radiolytic
formation of oxidants is effectively suppressed, and
the corrosion potential remains close to its thermodynamic minimum (which is a function of temperature, H2 fugacity, and pH). BWRs cannot achieve this
H2 level in the core because H2 partitions to the
steam phase, which begins to form about a quarter
of the way up the fuel rods. Thus, this section is
primarily relevant to BWRs.
Radiolysis and the presence of oxidizing species
require that many sequential and nonlinear dependencies that must be considered, for example, radiation flux produces oxidizing and reducing species,
the corrosion potential is controlled by multiple
reactions of these species, the crack chemistry is
nonlinearly dependent on the corrosion potential
and the dissolved ions present, and the SCC growth
rate is a nonlinear function of the crack chemistry.8,61 The relationship between radiation flux and
SCC cannot easily be determined empirically, but
rather requires a fundamental understanding of each
subprocess.
Water is decomposed by ionizing radiation into
various primary species6265 including both radicals
(e.g., e
aq, H, OH, HO2) and molecules (e.g., H2O2, H2,
O2), which can be oxidizing (e.g., O2, H2O2, HO2) or
reducing (e.g., e
aq, H, H2). The predominant species
that are stable after a few seconds are H2O2 and H2,
with O2 forming primarily from the decomposition of
H2O2. Because H2 partitions to the steam phase and
H2O2 is not volatile, 87% of the water that is
recirculated in a BWR (1114% of the core
flow becomes steam) is oxidant rich. H2 is introduced
in the feed water, which mixes with the recirculated
water near the top of the annulus (the region of
downflow between the core shroud and pressure
vessel).
The concentrations of radiolytic species are
roughly proportional to the square root of the radiation flux in pure water. The radiation energy versus
intensity spectrum influences the concentration of

Irradiation Assisted Stress Corrosion Cracking

0.4
0.3
Corrosion potential (VSHE)

each radiolytic species, which is described in terms of


a yield, or G value (molecules produced per 100 eV
absorbed by water). In LWRs, the G values for most
species are within a factor of 3 for fast neutron
versus gamma radiation. Despite this similarity, the
influence of fast neutron radiation is much stronger
than gamma radiation primarily because the energy
deposition rate, or mean linear energy transfer
(LET), is greater (40 eV nm1 for fast neutrons versus
0.01 eV nm1 for gamma radiation63). Also, the neutron flux in LWRs (e.g., 1.03  109 Rad h1 core
average and 1.68  109 Rad h1 peak in a BWR4 of
51 W cm3 power density) is also higher than the
gamma flux (0.34  109 Rad h1). Indeed, the moderate gamma levels present in the downcomer in the
outside annulus of a BWR core actually promote
recombination of hydrogen and various oxidants.62,65
This is a key element in the effectiveness of HWC,
and BWRs that have a wider annulus and lower
gamma flux near the pressure vessel respond less to
a given H2 addition. The contribution of thermal
neutrons and beta particles to radiolysis is small
in LWRs.
As in many electrochemical processes, the
integrated effects of various oxidants and reductants
on environmental cracking is best described via
changes in corrosion potential, which controls the
thermodynamics and influences the kinetics of most
reactions. Since electrochemical potentials are logarithmically dependent on local oxidant, reductant,
and ionic concentrations (via the Nernst relationship,
o RT/nF ln [products/reactants], where is
electrochemical potential, R is the gas constant, T is
temperature in kelvin, n is the number of moles, and F
is Faradays constant), radiation-induced increases in
concentration of various species by many orders of
magnitude may have comparatively small effects on
the corrosion potential in hot water. Further, corrosion potentials are mixed potentials involving a balance of anodic and cathodic reactions on the metal
surface. At low oxidant concentrations, the rapid drop
in corrosion potential to 0.5 Vshe (Figure 11(a))
results from mass-transport-limited kinetics of oxidants to the metal surface.
The relationship between dissolved oxygen and
corrosion potential in hot water as a function of
radiation type and flux is shown in Figure 11(a),
in which the connected points represent data
obtained in controlled radiation on/off experiments.
The data from these latter experiments are shown in
Figure 11(b) in terms of a radiation-induced shift in
potential. The curves in Figure 11(a) represent the

2 points, 200 ppb H2O2


1 point, 31ppb Cu2+

0.2
0.1

0
0.1

200 ppb H2

H2Wc
BWR top core n + g
Proton rad. = BWR peak
1
30200 MGy h g

0.2
0.3

1
0.32 MGy h g
Pt, various rad levels

0.4
0.5

Open symbol = no radiation


Lines indicate paired data (e.g., g on/off)
Band of data for unirradiated conditions

0.6
0.7

100

(a)

101

102

103

104

Dissolved oxygen (ppb)

0.3
0.2
Dfc with radiation (VSHE)

188

0.1

200 ppb H2O2

0
0.1

200 ppb H2
2 points

0.2

Proton, ECP in crevice


BWR top core n + g
Proton rad. = BWR peak
30200 MGy h1g
0.32 MGy h1g
Pt, various rad levels

0.3
<100
(b)

102
101
103
Dissolved oxygen (ppb)

104

Figure 11 (a) Effect of radiation on the corrosion


potential of type 304 stainless steel in water at 288  C.
The curves denote the range of typical values in the
unirradiated corrosion potential data (reproduced from
Andresen, P. L.; Ford, F. P.; Higgins, J. P.; et al.
In Proceedings of ICONE-4 Conference; ASME
International: New York, NY, 1996). (b) Effect of radiation
on the shift in corrosion potential from the value under
unirradiated conditions for type 304 stainless steel in water
at 288  C. With the exception of the boiling-water reactor
measurements, all data were obtained in controlled
radiation on/off experiments (reproduced from
Andresen, P. L.; Ford, F. P.; Higgins, J. P.; et al.
In Proceedings of ICONE-4 Conference; ASME
International: New York, NY, 1996). Curves in (b) show the
trends in the proton-irradiated data, where the effects of
radiation (on/off and over a range of fluxes) were evaluated
for a variety of dissolved O2 and H2 concentrations under
otherwise identical conditions. H2Wc, Hydrogen Water
chemistry.

scatter band for the data obtained under unirradiated


conditions. Similar scatter also exists in the irradiated
corrosion potential data in Figure 11(a) and comprises contributions from both real effects and experimental error. High radiation flux experiments were

Irradiation Assisted Stress Corrosion Cracking

performed by Taylor66 and Andresen et al.67 using


multiple, fundamentally different types of radiationhardened reference electrodes. In-reactor measurements have also been performed by Indig68,69 using
multiple, radiation-qualified silver chloride reference
electrodes.
Figure 11(a) and 11(b) show that little, if any,
elevation in corrosion potential results from irradiation sources that do not include neutrons or simulate
their contribution (e.g., using high-energy protons67).
Some studies using gamma radiation15,66 showed a
significant decrease in corrosion potential, especially
in the intermediate (e.g., 10200 ppb) range of dissolved oxygen. This is consistent with enhanced
recombination of oxidizing and reducing species,
which occurs in the downcomer region of BWRs.65
In instances where neutrons or protons have been
used, a consistent, significant elevation in corrosion
potential is observed. This is more pronounced at low
dissolved oxygen concentrations with no dissolved
hydrogen (Figure 11(b)), where increases of over
0.25 V occur. At higher inlet oxygen concentrations
(e.g., 200 ppb), the data still show a significant
shift (typically 0.1 to 0.15 V) in corrosion potential for radiation conditions representative of peak
LWR core fluxes (Figure 11(b)); less increase is
observed for inlet oxygen concentrations associated,
for example, with air saturation (8.8 ppm O2) or
oxygen saturation (42 ppm O2 at STP). A similar
elevation in corrosion potential is observed for
additions of hydrogen peroxide (200 ppb H2O2,
Figure 11(a)), which suggests that H2O2 may be a
major factor in increasing the corrosion potential
under irradiated conditions.
In-core, in situ measurements in BWRs show that
the corrosion potential, which is 0.2 to 0.25 Vshe
in normal water chemistry (NWC), can be decreased
by >0.5 V by sufficient additions of dissolved hydrogen in a BWR.69 This is corroborated by other measurements67 (Figure 11(b)), which show very little
radiation-induced elevation in corrosion potential
when the fully deaerated inlet water contains moderate dissolved hydrogen (>200 ppb H2, 0 ppb O2). At
high H2 levels, the core becomes reducing, and the
small concentration of 16N (transmuted from 16O)
changes from soluble NO
3 to volatile NOx and
NH3, causing a large increase in radiation level in
the steam lines and turbine.
The effect of radiation on the corrosion potential
within a crack or crevice has also been of interest,
with the possibility that a net oxidizing environment
in the crack could be created that could elevate the

189

corrosion potential above the potential at the crack


mouth. In the absence of radiation, measurements in
high-temperature water in artificial crevices (e.g.,
tubing),70,71 at the tip of growing cracks,72 and of
short crack growth behavior73 show that the corrosion potential remains low (i.e., 0.5  0.1 Vshe in
pure water at 288  C) for all bulk oxygen concentrations, indicating that complete oxygen consumption
occurs within the crack. Measurements of radiation
effects in crevices67 show that the elevation in corrosion potential is limited to <0.05 V (c < 0.45 Vshe)
in-core; this is consistent with interpretation of available corrosion potential data on free surfaces.8,15,16
These small changes will not significantly affect the
0.75 V (0.25 Vshe (near mouth) minus 0.5 Vshe (in
crack)) potential difference in the crack under irradiated normal BWR water chemistry conditions. The
potential difference, along with other factors, controls the enhancement mechanism that can lead to
an increased anion activity and altered pH at the
crack tip.12,13,54,55
5.08.3.2 Effects of Corrosion Potential
on IASCC
Preirradiated stainless steels were evaluated in
SSR tests50,74 in hot water using additions of oxygen
and/or hydrogen peroxide to elevate the corrosion
potential to simulate the effect of radiation. Jacobs
et al.50 showed that IASCC in stainless steel irradiated
to 3  1021 n cm2 was strongly affected by dissolved oxygen (and, by inference, corrosion potential)
(Figure 5(b)). Ljungberg74 also evaluated preirradiated materials in SSR tests and observed decreasing
average crack growth rates with decreasing corrosion
potential. Less IASCC occurs at low corrosion potential, but crack growth rate tests (discussed later) and
other tests confirm that IASCC does not vanish at
low corrosion potential (Figures 5 and 9).
Continuously monitored fracture mechanics
specimens were installed in the recirculation piping
and core of the Nine Mile Point Unit 1 BWR.
Furnace-sensitized type 304 stainless steel specimens
in both locations showed higher growth rates in
the core, where the corrosion potential was higher
(Figure 12). Specimens of annealed type 304 and
304L stainless steel showed growth after different
fluences but, once growth initiated, the growth rates
at a given fluence were identical. The delay time
(fluence) and differences in delay time were attributed to the need to transition from a transgranular
fatigue precrack to IG SCC.

190

Irradiation Assisted Stress Corrosion Cracking

104

100
Furnace sensitized 304 st. st.
27.5 MPa m1/2

Crack length (mm)

100

Nine Mile Point 1 BWR


DCB specimens for monitoring SCC

96
94

Incore

Observed
Theory

Nine Mile Point 1 BWR


DCB specimens for
monitoring SCC
C304SEN, thermally
27.5 MPa m1/2
sensitized 304
C304ST, solution
treated 304

102

92
90

Crack length (mm)

98

98
96
94

C304L, low
carbon 304

92
90

88
88
86

Recirculation line

86

84

84
0

(a)

500

1000 1500 2000 2500 3000 3500 4000


Time (h)

2000

4000

6000

8000

10 000

12 000

Time (h)

(b)

Figure 12 Data for fracture mechanics specimens of type 304 stainless steel exposed in the high flux region of the
core and in the recirculation line of the Nine Mile Point Unit 1 BWR. All specimens were precracked and
wedge loaded to an initial stress intensity factor of 27.5 MPa m1/2. (a) Comparison of predicted and observed crack length
versus time for furnace-sensitized type 304 stainless steel specimens in the core and recirculation line. (b) Crack
length versus time for one furnace-sensitized and two annealed specimens of type 304 stainless steel in the core. Adapted
from Andresen, P. L.; Ford, F. P.; Murphy, S. M.; Perks, J. M. In Proceedings of 4th International Symposium on
Environmental Degradation of Materials in Nuclear Power Systems Water Reactors; Cubicciotti, D., Theus, G. J., Eds.;
NACE: Houston, TX, 1990; pp 183; Andresen, P. L.; Ford, F. P. In Proceedings of 7th International Symposium on
Environmental Degradation of Materials in Nuclear Power Systems Water Reactors; NACE: Houston, TX, 1995;
pp 893908; Was, G. S.; Andresen, P. L. Corrosion 2007, 63(1), 1945; Andresen, P. L.; Ford, F. P. In Corrosion/89; NACE:
Houston, TX, 1989; Paper no. 497; Andresen, P. L.; Ford, F. P. International Cooperative Group on Irradiation Assisted
Stress Corrosion Cracking (ICG-IASCC) Minutes, Apr 1989.

Figures 2 and 13 present the first laboratory SCC


growth rate testing ever performed on irradiated
(4 dpa) type 304SS, which was well behaved at high
corrosion potential and showed a large change in
growth rate at lower corrosion potential. One evolving concern for high-fluence stainless steels is the
prospect of a major loss of the effect of corrosion
potential and stress intensity factor, as has been
observed in unirradiated alloys possessing high
Si.52,53 The detrimental effect of Si is likely associated
with its ability to oxidize at all LWR-relevant potentials, coupled with its relatively high solubility in hot
water (i.e., it does not form a protective oxide like
Cr2O3 or Fe/Ni oxides/spinels).
These data are compared with other irradiated
and unirradiated data in Figure 2 based on simultaneous measurements of corrosion potential and crack
growth rate in fracture mechanics specimens; the
accompanying curves represent model predictions.8,13,15,16,54,55,75,76 Clearly, the in situ data compare favorably with the spectrum of unirradiated
data, and data obtained on a fracture mechanics
specimen of furnace-sensitized type 304 stainless
steel using high-energy proton irradiation to simulate the mix of neutron and gamma radiation present
in power reactors.

5.08.4 Irradiation Effects on


Microchemistry and Microstructure
5.08.4.1

Radiation-Induced Segregation

RIS describes the redistribution of major alloying


elements and impurity elements at point defect
sinks.7787 Because IASCC is intergranular, the sinks
of greatest interest are the grain boundaries. RIS is
driven by the flux of radiation-produced defects to
sinks and is therefore fundamentally different from
thermal segregation or elemental depletion due to
grain boundary precipitation. Vacancies and interstitials are the basic defects produced by irradiation and
can reach concentrations that are orders of magnitude higher than those at thermal equilibrium, resulting in large increases in diffusion rates. If the relative
participation of alloying elements in the defect fluxes
is not the same as their relative concentration in the
alloy, then a net transport of the constituents to or
from the grain boundary will occur (Figure 14). This
unequal participation of solutes in the vacancy and/
or interstitial fluxes to sinks results in either enrichment or depletion of an alloying element at the grain
boundary. The species that diffuse more slowly by the
vacancy diffusion mechanism are enriched, and the
faster diffusers become depleted.

Irradiation Assisted Stress Corrosion Cracking

191

250
12.59
2.7 108 mm s1

12.53

150

To 0 ppb O2@1508 h

Crack length (mm)

12.55
1.3 106
mm s1

12.51
12.49

100

Constant K

Dissolved oxygen (ppb)

200

12.57

50

CT#2 - irradiated 304SS, 4 dpa


21 MPa m1/2, 220 ppb O2, 288 C, 20 ppb SO4

12.47
12.45
1475

0
1495

1515

1535

1555

1575

1595

Time (h)
Figure 13 Crack length versus time for Type 304 stainless steel irradiated to 4 dpa and tested in at 21 MPa m1/2 in water at
288  C. A large reduction in crack growth rate is observed as the dissolved O2 and corrosion potential are decreased.
Reproduced from reproduced from Andresen, P. L.; Ford, F. P.; Higgins, J. P.; et al. In Proceedings of ICONE-4 Conference;
ASME International: New York, NY, 1996.

Cv

Ci

Jv
JA

Ji
JB

JB

(a)

JA

(b)

CB
CA0,B
CA

(c)
Figure 14 Schematic illustration of radiation-induced
segregation in a binary 50A50B system showing (a) the
development of the vacancy concentration profile by the flow
of vacancies to the grain boundary balanced by a equal and
opposite flow of A and B atoms, but not necessarily in equal
numbers, (b) the development of the interstitial concentration
profile by the flow of interstitials to the grain boundary
balanced by a equal and flow of A and B atoms migrating as
interstitials, but not necessarily in equal numbers, and (c) the
resulting concentration profiles for A and B. Reproduced
from Was, G. S. Fundamentals of Radiation Materials
Science: Metals and Alloys; Springer: Berlin, 2007.

Enrichment and depletion can also occur by association of the solute with the interstitial flux. The undersized species will enrich, and the oversized species will
deplete.80 The magnitude of the buildup/depletion
is dependent upon several factors such as whether
a constituent migrates more rapidly by one defect
mechanism or another, the binding energy between
solutes and defects, the dose, dose rate, and the temperature. RIS profiles are also characterized by their
narrowness, often confined to within 510 nm of the
grain boundary, as shown in Figure 15 for an irradiated stainless steel.
Segregation is a strong function of irradiation
temperature, dose, and dose rate (Figure 16). Segregation peaks at intermediate temperatures since a
lack of mobility suppresses the process at low temperatures, and back-diffusion of segregants minimizes
segregation at high temperature (where defect concentrations approach their thermal equilibrium
values). For a given dose, a lower dose rate results in
a greater amount of segregation. At high dose rates, the
high defect population results in increased recombination which reduces the number of defects that are able
to diffuse to the grain boundary. Figure 16 shows the
interplay between temperature and dose rate for an
austenitic stainless steel. RIS occurs in the intermediate temperature range, and this range rises along the

192

Irradiation Assisted Stress Corrosion Cracking

18

4
JEOL 2010F
0.75 nm probe

16

14
12
10
8

Ni

Si

P
0

6
20

15 10 5
0
5
10 15
Distance from grain boundary (nm)

20

Figure 15 Compositional profiles across grain


boundaries obtained by dedicated scanning transmission
electron microscopy from a low-strain, high-purity 348
stainless steel swelling tube specimen irradiated to
3.4  1021 n cm2 in water at 288  C in a boiling-water
reactor. Composition profiles were measured using
a field-emission gun scanning transmission electron
microscope. Reproduced from Jacobs, A. J.;
Clausing, R. E.; Miller, M. K.; Shepherd, C. M. In
Proceedings of 4th International Symposium on
Environmental Degradation of Materials in Nuclear Power
Systems Water Reactors; Cubicciotti, D., Ed.; NACE:
Houston, TX, 1990; pp 1421.

temperature scale with increasing dose rate to compensate for the higher recombination rate.
In FeCrNi alloys, the vacancy exchange
(inverse Kirkendall) mechanism successfully explains
the observed major element segregation.84,85 Studies
have shown that nickel segregates to grain boundaries
while chromium and iron deplete. The directions of
segregation are consistent with an atomic volume
effect in which the subsized solute migrates preferentially with the interstitial flux, and the oversized
solute participates preferentially in the vacancy flux.
The results are also consistent with the diffusivity of
the solutes in FeCrNi, in which Ni is the slow
diffuser, Cr is the fast diffuser, and Fe is intermediate.
In commercial austenitic stainless steels, chromium
depletes at grain boundaries and nickel enriches,
while iron can either deplete or enrich according to
the magnitude of the diffusion coefficient relative to
the other solutes.88
RIS increases with neutron dose in LWRs and
saturates after several (5) displacements per atom
in the 300  C temperature range. Figure 17 shows
grain boundary chromium depletion for austenitic

Homologous temperature (T/Tm)

5
Cr

Si or P concentration (wt%)

Cr or Ni concentration (wt%)

20

1105

0.8
0.7

Back diffusion
of vacancies

933

0.6

761
589

0.5
0.4

Radiation-induced
segregation

416

LWR
0.3 peak flux region
0.2
0.1
108

244

Recombination of
vacancies and interstitials
107
106
105
104
1
Radiation flux (dpa s )

72

Temperature for g stainless steel (C)

22

101
103

Figure 16 Dependence of radiation-induced segregation


on homologous temperature and dose rate for austenitic
stainless steels.

stainless steels as a function of dose.8997 As the slowest diffusing element, nickel becomes enriched at the
grain boundary. Since iron depletes in 304 and 316
stainless steels, the nickel enrichment makes up for
both chromium and iron depletion and can reach
very high levels up to 30 wt%.
Minor alloying elements and impurities also segregate and have been implicated in the IASCC process.
Mn and Mo strongly deplete at the grain boundary
under irradiation,98 but neither is believed to be a
significant factor in IASCC. Minor alloying or impurity
elements such as Si and P also segregate under irradiation. Silicon strongly enriches at the grain boundary
to as much as ten times the bulk (0.72.0 at.%) composition in the alloy99 and can be important in IASCC.
Phosphorus is present at much lower concentrations
and is only modestly enriched at the grain boundary
because of irradiation.83,98 Phosphorus tends to
segregate to the grain boundary following thermal
treatment, which reduces the amount of additional
segregation to the grain boundary during irradiation,
making the contribution due to irradiation difficult to
detect.98 Undersized solutes such as C, B, and N should
also segregate, but there is little evidence of RIS,
due in part to the difficulty of measurement. Another
potential segregant is helium, produced by the transmutation of 10B. The mobility of He is low at LWR core
temperatures, but the opportunity for accumulation
at the grain boundary is increased by segregation of
B to the boundary. Overall, the behavior of these minor
elements under irradiation is not well understood.

Irradiation Assisted Stress Corrosion Cracking

Fast neutron fluence (E > 1 MeV) 1025 n m2


Grain boundary Cr concentration (wt%)

26

8
304 (82)
304 (13)
304 (91)
304 (92)
304 (93)
316 (82)
316 (94)
348 (91)

24
22
20

10

12

14

HP 304 (95)
CP 304 (95)
CP 316 (95)
CP 304 (17)
CP 316 (17)
HP 316 (17)
CP 304 Protons 96
CP 316 Protons 96

18
16
14
12
10
0

10
Dose (dpa)

15

20

Figure 17 Dose dependence of grain boundary


chromium concentration for several 300-series austenitic
stainless steels irradiated at a temperature of about 300  C.
Adapted from Asano, K.; Fukuya, K.; Nakata, K.; Kodama, M.
In Proceedings of 5th International Symposium on
Environmental Degradation of Materials in Nuclear Power
Systems Water Reactors; Cubicciotti, D., Simonen, E. P.,
Gold, R. E. Eds.; American Nuclear Society (ANS):
LaGrange, IL, 1992; p 838; Jacobs, A. In Proceedings of 7th
International Conference on Environmental Degradation of
Materials in Nuclear Power Systems Water Reactors;
NACE: Houston, TX, 1995; p 1021; Jacobs, A. J.;
Wozadlo, G. P.; Nakata, K.; et al. In Proceedings of 6th
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems Water Reactors;
Gold, R. E., Simonen, E. P., Eds.; The Minerals, Metals, and
Materials Society (TMS): Warrendale, PA, 1993; p 597;
Kenik, E. A. J. Nucl. Mater. 1992, 187, 239; Nakahigashi, S.;
Kodama, M.; Fukuya, K.; et al. J. Nucl. Mater. 1992,
179181, 1061; Jacobs, A. J.; Clausing, R. E.; Miller, M. K.;
Shepherd, C. M. In Proceedings of 4th International
Symposium on Environmental Degradation of Materials in
Nuclear Power Systems Water Reactors; Cubicciotti, D.,
Ed.; NACE: Houston, TX, 1990; pp 1421; Walmsley, J.;
Spellward, P.; Fisher, S.; Jenssen, A. In Proceedings of 7th
International Symposium on Environment Degradation of
Materials in Nuclear Power SystemWater Reactors;
NACE: Houston, TX, 1997; p 985; Was, G. S.; Busby, J. T.;
Gan, J.; et al. J. Nucl. Mater. 2002, 300, 198.

Oversize solutes can affect the microchemistry or


microstructure of the alloy, thereby altering the
IASCC susceptibility. They are believed to affect
RIS by acting as vacancy traps, thereby increasing
the effective recombination of vacancies and interstitials and thus reducing RIS. Kato et al.100 conducted
electron irradiations of several stainless steels at temperatures of 400500  C up to 10 dpa. Results showed
that some solutes (Zr and Hf) consistently produced a
large suppression of radiation-induced chromium

193

depletion, while others resulted in less suppression


or suppression at only certain temperatures. Fournier
et al.101 conducted irradiation of 316 containing Hf or
Pt using 3-MeV protons (400  C) and 5-MeV Ni ions
(500  C). Ni irradiations showed little effect of the
oversize impurity in reducing grain boundary chromium depletion (Cr depletion increased in the case
of Hf), but proton irradiation showed a significant
suppression of RIS of chromium at low dose (2.5 dpa)
with the effect diminishing at higher (5.0 dpa) dose.
Pt had a smaller effect on Cr. Ti and Nb similarly
produced little change in the grain boundary chromium concentration after irradiation with 3.2-MeV
protons to 5.5 dpa at 360  C. In Zr-doped 304SS,
there were no consistent results of suppression of
grain boundary chromium after 3.2-MeV proton irradiation to 1.0 dpa at 400  C.102 Neutron irradiation at
very low dose (0.5 dpa) shows a small effect of Ti and
Nb on grain boundary Cr.103 In all, the data on the
effect of oversize solutes on RIS of chromium are
inconsistent.
RIS is understandably implicated in IASCC of
stainless steels, especially in oxidizing environments,
in part because of the effect of thermal sensitization
in extensive data from lab and plant operational
experience.1520,37,38,104,105 As shown in Figure 17,
grain boundary chromium depletion during irradiation can be severe. Figure 18(a) shows a correlation
between grain boundary chromium level and IGSCC
susceptibility in stainless steels where the grain
boundary depletion is due to thermal sensitization.106
Much data have been accumulated to support the
role of chromium depletion as an agent in IGSCC
of austenitic alloys in oxidizing conditions. Numerous studies show that, as the grain boundary chromium level decreases, intergranular SCC increases.
Typical chromium-depleted zone widths are of the
order 100300 nm full width at half-maximum
(FWHM), providing a significant volume of depleted
material adjacent to the grain boundary.
Figure 18(b) shows a similar correlation between
grain boundary chromium level and IASCC susceptibility as measured by the percentage IG cracking
on the fracture surface during SSR experiments.
A major difference between Cr depletion profiles
resulting from RIS and those due to precipitation
reactions is that the width of the RIS profiles can
be as much as 2 orders of magnitude smaller, typically 510 nm. There is large scatter in the data
that makes a direct correlation difficult to support,
and differences in testing conditions undoubtedly
contribute.

194

Irradiation Assisted Stress Corrosion Cracking

%IGSCC in slow strain rate test

100
80
60

e = 2 107 S1

= 1 106 S1
40
Alloy 600
SSR tests,
23 C sulfuric
acid

20

Type 304SS
SSR tests, 288 C
8 ppm O2 water

0
4
(a)

6
8
10
12
14
16
18
20
Minimum grain boundary chromium concentration (wt%)

100

HP 316 SS with single


solute additions108
3xx SS17

80

%IG

60
40
20
0
10
(b)

12

14
16
18
20
Grain boundary Cr content (wt%)

irradiation. However, it should be noted that these


alloys were not irradiated, and this difference may be
important in the relevance of such experiments to
IASCC. Using 1.55% Si stainless steels of both
standard (e.g., 304L) base composition and synthetic
irradiated grain boundary composition, Andresen has
observed significantly increased growth rates, reduction in the benefit of lowering corrosion potential,
and very little effect of stress intensity factor between
27 and 13 ksi in1/2.
The data on impurity segregation effects on
IASCC remain inconclusive. Extensive experiments
have been conducted to isolate the effect of particular
impurities such as S, P, C, N, and B in IASCC, but
none have yielded unambiguous results. Sulfur has not
been found to segregate under irradiation, and, while
P thermally segregates to a significant extent, irradiation-induced P segregation is small in comparison. C,
N, and B cannot be measured in STEM; N and B are
very difficult to identify in AES; and C is a common
contaminant. Overall, it has been a challenge to establish a link between impurity element segregation and
IASCC in austenitic stainless steels.

22

Figure 18 Effect of grain boundary Cr content on


intergranular stress corrosion cracking for (a) sensitized
stainless steel and Alloy 600 and (b) irradiated stainless
steels. Slow-strain-rate tests are tests in which the
specimen is monotonically strained versus time.
Reproduced from Bruemmer, S. M.; Was, G. S. J. Nucl.
Mater. 1994, 216, 348.

Among the minor alloy elements, only Si is known


to segregate to high levels, and Si segregation is
correlated with IASCC. Experiments by Busby
et al.107,108 on a high-purity 316 base alloy doped
with 1 wt% Si showed severe IASCC in NWC and
in primary water after irradiation to 5.5 dpa at 360  C.
STEM measurements of grain boundary Si confirm
levels up to 6 wt%. Past studies comparing Auger
electron spectroscopy (AES) and scanning transmission electron microscopy (STEM) results have shown
that the actual concentration of Si at the grain boundary plane may be as high as 1520 wt%. Though the
electron beam probe in STEM is very small, the
measurement underpredicts the concentration at
the grain boundary by a factor of 2 to 3. Yonezawa
et al.109112 and Li et al.113 have provided extensive
evidence to show that increased Si in stainless steel
results in increased IGSCC in alloys tailored to imitate the composition of grain boundaries under

5.08.4.2 Microstructure, Radiation


Hardening, and Deformation
5.08.4.2.1 Irradiated microstructure

The microstructure of austenitic stainless steels


under irradiation changes rapidly at LWR service
temperatures. Point defect clusters (called black dot
damage when electron optics could not resolve the
details) begin to form at very low dose, dislocation
loops and network dislocation densities evolve with
dose over several displacements per atom, and the
possibility exists for the formation and growth of
He-filled bubbles, voids, and precipitates in core
components in locations exposed to higher dose and
temperatures.114120 Below 300  C, the microstructure is dominated by small clusters and dislocation
loops. Near 300  C, the microstructure contains
larger faulted loops plus network dislocations from
loop unfaulting and cavities at higher doses.
The primary defect structures in LWRs are
vacancy and interstitial clusters and Frank dislocation
loops. The clusters are formed during the collapse of
the damage cascade associated with primary and secondary atom collisions after an interaction with a
high-energy particle. The larger, faulted dislocation
loops nucleate and grow as a result of the high mobility of interstitials. The loop population grows in size
and number density until absorption of vacancies and

Irradiation Assisted Stress Corrosion Cracking

interstitials equalize, at which point the population


has saturated. Figure 19 shows the evolution of loop
density and loop size as a function of irradiation dose
during LWR irradiation at 280  C. Note that saturation of loop number density occurs very quickly,
by 1 dpa, while loop size continues to evolve up
to 5 dpa. The specific number density and size are
dependent on irradiation conditions and alloying elements, but the loop size rarely exceeds 20 nm and
densities are of the order of 1  1023 m3.
The hardening process and IASCC susceptibility
are influenced by small defects. The traditional view
that small defect clusters are predominantly faulted
interstitial loops and vacancy clusters121 may be inaccurate. Analysis of recent postirradiation annealing
experiments by Busby et al.122 and Simonen et al.123
suggests that there are at least two types of defects
with different annealing characteristics: vacancy and
interstitial faulted loops, each with different annealing
kinetics. The step change in hardness as a function
of annealing time suggests that the density of vacancy
loops is perhaps much higher than previously believed,
and higher than the density of interstitial loops.122
Above 300  C, voids and bubbles may begin to
form, aided by the increased mobility of point defects
at the higher temperature. The dislocation structure
will evolve into a network structure as larger Frank
loops unfault. The reduction in the sink strength
of the dislocation loops aids in the growth of voids
and bubbles. While their size and number density
increase with temperature, the dislocation microstructure continues to be t