EPRI Steam Generator Reference Book PDF

Leadership in Electrification
through Global Collaboration R E P 0 R T S U M M A R Y
Steam Generator Reference Book, Revision 1

Volume 1
The Steam Generator Reference Book documents the state of the art
in PWR steam generator technology, providing a comprehensive
source for operators, owners, and designers of PWR nuclear power
plants. The book summarizes pertinent steam generator operating
issues and provides recommendations to improve operational effi-
ciency. Information in the book represents 15 years of research and
development activity over the course of several hundred research
projects involving PWR steam generator issues.
BACKGROUND In 1977, EPRI established the Steam Generator Project Office in

INTEREST CATEGORIES the Nuclear Power Division to manage research and development (R&D) leading to
resolution of operational challenges associated with PWR steam generators. Under
the charter of the Steam Generator Owners Groups I and II followed by the Steam
Steam generator reliability Generator Reliability Project, EPRI staff worked on this project in conjunction with
. Nuclear component
reliability
uclear plant operations
participating utilities. The results of this work have been documented in several
hundred EPRI reports and papers. Additionally, research results from international
associates have been incorporated with EPRI and U.S. utility results. Altogether,
and maintenance these activities have resulted in a significant increase in steam generator avail-
Advanced light water ability, and in a reduction in lost capacity due to forced outages.
reactors
OBJECTIVE To integrate and summarize current PWR steam generator technol-
KEYWORDS ogy; to recommend design, operating, and maintenance improvements that will
increase plant availability and life.
Nuclear steam generators
Component reliability APPROACH EPRI editors arranged the R&D results by operational issue and
Corrosion assigned EPRI staff and key consultants/contractors to collect, organize, and write
Water chemistry chapters relating to each issue. Specifically, chapters in the Steam Generator
Maintenance Reference Book address each of the known and anticipated steam generator
challenges and the factors affecting them. Included are causes of steam generator
unavailability as well as possible actions to minimize problems. For ease of
retrieval, recommendations from all chapters are listed in Chapter 4. Each area/
issue has been critically reviewed by the editors, EPRI staff, and other experts in
the field.
RESULTS The Steam GeneratorReference Book summarizes options'and rec-

ommendations for improving the operating life of PWR steam generators. Options
will be utility specific and will depend on such factors as the age of the plant, design/
construction, plant siting, and utility policies. The book emphasizes a variety of
damage forms, addressing environmental factors likely to accelerate or inhibit/
control damage initiation and growth. Recommendations describe modifications to
water chemistry control, corrosion inhibitor use, alloy selection, and nondestructive
examination (NDE) methods. Bibliographies have been collected and organized for
each operational issue.
9703260284 970310
PDR TOPRP EXIEPRI
B CF
The guide includes discussions of the following specific areas: steam
generator performance history, steam generator design, operational guide-
lines to minimize corrosion, steam generator degradation, thermal and
hydrodynamic analysis of steam generators, primary water stress corro-
sion cracking (PWSCC), tube support and tubesheet corrosion, tube
wastage and phosphate secondary water chemistry, tube pitting, tube
supports and tubesheet corrosion (alkaline denting), secondary-side
intergranular attack (IGA) and stress corrosion cracking (SEC), intergran-
ular corrosion of alloy 600 from caustic compounds, intergranular corro-
sion from acidic compounds, lead cracking of alloy 600, tube corrosion
and wear, tube fretting and wear, tube fatigue, water chemistry control,
ionic impurity control, water chemistry options, corrosion product control,
material selection and alternative designs for steam generators, PWSCC
remedies, NDE methods, and major steam generator repairs/replacement.
EPRI PERSPECTIVE The Steam GeneratorReference Book, Revision 1

documents the state of the art at the time each chapter was written. The
format of this edition has been revised since publication of the original
document so that chapters can be updated as needed and inserted into
binders. EPRI has used the recommendations in this reference book to
help utility staff implement equipment modifications as well as changes in
operation and maintenance practices. As a result of this work and com-
plimentary work by others in the field, the availability of PWR steam
generators has risen dramatically.
PROJECT
RP2858; RP4004
Project Managers: J. P. N. Paine
Nuclear Power Group
EPRI Members: For ordering information about this report, call the
EPRI Distribution Center (510) 934-4212.
For membership information, call (415) 855-2514.
0
Steam GeneratorReference Book
Revision 1
Volume 1
Based on
work sponsored
by
The Steam Generator Owners Groups I and II

The Steam Generator Reliability Project
and
Electric Power Research Institute
J. Peter N. Paine, Editor

Ulla E. Gustafsson, Associate Editor
December 1994
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS REPORT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN
ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER
RESEARCH INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY
COSPONSOR, THE ORGANIZATION(S) NAMED BELOW, NOR ANY PERSON ACTING
ON BEHALF OF ANY OF THEM:
(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR
IMPLIED, (I)WITH RESPECTTOTHE USE OFANY INFORMATION, APPARATUS, METHOD,
PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS REPORT, INCLUDING
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (11)THAT SUCH
USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED RIGHTS,
INCLUDING ANY PARTY'S INTELLECTUAL PROPERTY, OR (111)THAT THIS REPORT IS
SUITABLE TO ANY PARTICULAR USER'S CIRCUMSTANCE; OR
(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY
WHATSOEVER (INCLUDING ANY CONSEQUENTIAL DAMAGES EVEN IF EPRI OR ANY
EPRI REPRESENTATIVE HAS BEEN ADVISED OFTHE POSSIBILITY OF SUCH DAMAGES)
RESULTING FROM YOUR SELECTION OR USE OFTHIS REPORT ORANY INFORMATION,
APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED INTHIS REPORT.
ORGANIZATION(S) THAT PREPARED THIS REPORT

Electric Power Research Institute
ORDERING INFORMATION
Requests for copies of this report should be directed to the EPRI Distribution Center,
207 Coggins Drive P.O. Box 23205, Pleasant Hill, CA 94523, (510) 934-4212. There
is no charge for reports requested by EPRI member utilities and affiliates.
Electric Power Research Institute and EPRI are registered service marks of Electric Power
Research. Institute, Inc.
Copyright © 1994 Electric Power Research Institute, Inc. All rights reserved.
In Dedication to
Stanley J. Green
for His Leadership and Service
to the Steam Generator Project and the Commercial Nuclear Industry
Stanley .J. Green served as Director of the Steam Generator Owners

Groups I and II and the Steam Generator Reliability Project from 1979 to
1991. Guided by his enthusiasm and leadership, the first Steam Generator
Reference Book was published in 1985.
As manager of steam generator research activities at EPRI, he helped
the nuclear industry increase the reliability of steam generators for
pressurized water reactor (PWR) plants. This formidable task involves
identifying the problems and their causes, providing technical and
management leadership to solve the problems, and transferring the resulting
technology to the electric utilities for implementation.
His contributions to steam generator technology cover a broad spectrum
including thermal-hydraulics, water chemistry, corrosion, and mechanical
wear. Stanley J. Green provided the leadership, technical advice, and
intuitive insight which allowed these diverse technologies to be addressed
in an integrated, well-defined manner resulting in significant advances in
steam generator design and operation.
Stanley J. Green has earned the highest respect of both the technical
and operational communities associated with PWR steam generator
technology. Through his efforts the reliability of these costly nuclear power
plant components have been significantly enhanced.
ii Steam Generator Reference Book, Revision I
ACKNOWLEDGMENTS
This book could not have been prepared without significant
professional, technical, and financial contribution from numerous
individuals and companies. In its several revisions, chapters of this book
have been critically reviewed. The contributions of the following individuals
0
and companies are gratefully acknowledged.
Authors and EPRI Steam Generator Project Office Staff

A. K. Agrawal, S. M. Laskowski,
Battelle Columbus Division Project Manager
M. J. Angwin, W. H. Layman,
Project Manager Director (1977-1979)
M. M. Behravesh, D. B. Lowenstein,
Manager Dominion Engineering, Inc.
W. J. Bilanin, R. G. Lucas,
Program Manager Project Manager
R. L. Coit, L. J. Martel,
Project Manager Technical Director (1977-1979)
G. W. DeYoung, A. R. McIlree,
Project Manager Project Manager
J. A. Gorman, P. Millett,
Dominion Engineering, Inc. Manager
S. J. Green, J. A. Mundis,
Director (1979-1991) Program Manager
G. Hetsroni, S. Oldberg,
Program Manager Project Manager
S. A. Hobart, J. P. N. Paine,
Adams & Hobart Manager
B. S. Holmes, M. J. Partridge,
Project Manager Dominion Engineering, Inc.
E. S. Hunt, W. L. Pearl,
Dominion Engineering, Inc. NWT
R. L. Jones, C. E. Shoemaker,
Director (1991-1993) Project Manager
J. E Lang, S. G. Sawochka,
Program Manager NWT
EPRI Licensed Material
Forward iii
G. Sirikantiah, A. P. L. Turner,
Manager Dominion Engineering, Inc.
Y. Solomon, R. G. Varsanik,
Senior Technical Advisor Project Manager
K. D. Stavropoulos, C. S. Welty, Jr.,
Dominion Engineering, Inc. Director (1993-)
D. A. Steininger, C. L. Williams,
Manager Project Manager
J. F Sykes, C. Wood,
TU Electric Manager
R. C. Thomas,
Project Manager
Financial Contributions
Alabama Power Company
Arizona Nuclear Power Project
Arizona Public Service Company
Arkansas Power & Light Company
Baltimore Gas & Electric Company
Boston Edison Company
Carolina Power & Light Company
Central Electricity Generating Board (Great Britain)
CRIEPI (Japan)
Commonwealth Edison Company
Consolidated Edison Company
Consumers Power Company
Duke Power Company
Duquesne Light Company
Electrabel
Electricit6 de France
Entergy Operations
Florida Power Corporation
Florida Power & Light Company
General Public Utilities
Georgia Power Company
GPU Corporation
Houston Lighting & Power Company
Long Island Lighting Company
Maine Yankee Atomic
iv Steam Generator Reference Book, Revision I
New York Power Authority

Northeast Utilities
Northern States Power Company
Omaha Public Power District
Ontario Hydro (Canada)
Pacific Gas & Electric Company
Portland General Electric Company
Power Authority of the State of New York
Public Service Electric & Gas Company
Rochester Gas & Electric Company
South Carolina Electric & Gas Company
Southern California Edison Company
Southern Nuclear Operating Company
South Texas Project
Spanish Utilities
Swedish State Power Board
Tennessee Valley Authority
Texas Utilities Electric
Virginia Power
Wisconsin Electric Power Company
Wisconsin Public Service Corporation
Wolf Creek Nuclear
Yankee Atomic
Steam Generator Owners Group I Executive Committee

W. E. Caldwell, Jr., Chairman (1977-1980), Senior Vice President,
Consolidated Edison Company of New York, Inc.
B. L. Dow, Jr., Manager, Nuclear Services,
W. F. Fee, Executive Vice President, Engineering & Construction,
Northeast Utilities Service Company
C. W. Fay, Assistant Vice President,
R. Garnsey, Head of Steam Water Chemistry Section,
Central Electricity Research Laboratories,
Central Electricity Generating Board (CEGB)
R. L. Haueter, Director of Special Projects,
J. G. Holmes, Jr., Assistant Manager of Power Administration,
Forward v
W. M. Menger, Chief Consulting Engineer,

J. P. Mercier, Director of Studies & Research,
J. B. Moore, Vice President,
D. M. Noble, Director of Operating Services,
W. L. Riedel, Principal Chemist,
L. Rylander, Department Manager,
P. F. Santoro, Chief, Generation Mechanical Systems Engineering,
A. D. Schmidt, Chairman (1980-1982), Vice Chairman (1977-1980),
Vice President, Power Resources,
G. J. Schnabel, Consulting Mechanical Engineer,
Public Service Electric & Gas Company of New Jersey
D. R. Sikes, Director, Fossil Operations,
R. A. Thomas, Vice President,
Southern Company Services, Inc.
L. D. White, Vice Chairman (1980-1982), Executive Vice President,
Operations & Engineering,
Rochester Gas & Electric Corporation
A. W. Wofford, Vice President,
Steam Generator Owners Group I Member Representatives

J. P. Bayne, Senior Vice President, Nuclear Generation,
G. E. Brobst, General Supervisor, Chemistry,
D. A. Bryant, Project Engineer,
Boston Edison Company
vi Steam Generator Reference Book, Revision 1
S. Burstein, Executive Vice President,

W. E. Caldwell, Jr., Senior Vice President,
W. G. Counsil, Senior Vice President, Nucl. Engineering & Operations,
J. D. Deress, Project Engineering Manager,
R. B. DeWitt, Vice President, Nuclear Operations,
R. Dietch, Vice President, Nuclear,
P. Fontaine, Adjoint au Chef,
Central Electricity Research Laboratories,
Central Electricity Generating Board
J. W. Griffin, Assistant Vice President,
R. L. Haueter, Director of Special Projects,
J. G. Holmes, Jr., Assistant Manager of Power Administration,
T. Kurosawa, Manager, Reactor Chemistry Department,
Energy & Environment Laboratory,
CRIEPI, Japan
R. L. Mansfield, Junior Plant Engineer,
E. R. McGrath, Senior Vice President, Power Generation,
Consolidated Edison Company of New York
W. M. Menger, Chief Consulting Engineer,
J. P. Mercier, Chef du Department Material,
Electricit: de France
J. B. Moore, Vice President,
Forward vii
B. Nilsson, Chief Engineer,

W. 0. Parker, Vice President, Steam Production,
Duke Power Company
W. L. Riedel, Principal Chemist,
A. D. Schmidt, Executive Vice President, Power Resources,
J. Schmieder, Executive Vice President and Chief Engineer,
D. R. Sikes, Director, Fossil Operations,
H. Takenouchi, Director, Energy & Environment Laboratory,
CRIEPI, Japan
R. A. Thomas, Vice President,
Southern Company Services
J. A. Tiernan, Manager, Nuclear Power Department,
E. E. Utley, Executive Vice President,
E. E. Van Brunt, Jr., Vice President, Construction Projects,
H. P. Walker, Assistant Plant Manager, Vogtle,
L. D. White, Jr., Executive Vice President,
B. D. Withers, Vice President, Nuclear,
A. W. Wofford, Vice President,
M. G. Zaalouk, Head of Nuclear Engineering Division,
viii Steam Generator Reference Book, Revision I
Steam Generator Owners Group I Technical Advisory Committee

R. J. Acosta, Plant Support Section Supervisor,
0. Batum, Vice Chairman (1977-1980), Manager,
Nuclear Safety & Licensing Department,
M. A. Bell, Supervising Mechanical Engineer,
J. P. Berge, Chef du Groupe des Laboratories,
J. C. Blomgren, Project Manager,
G. E. Brobst, General Supervisor, Chemistry,
D. J. Clowers, Licensing Engineer,
A. E. Curtis, III, Manager, Materials Engineering Division,
A. de Montardy, Head, Heat Transfer & Aerodynamics Department,
J. Engstr6m, Senior Mechanical Engineer,
J. M. Fackelmann, Supervisor, Nuclear Materials & Chemistry,
Northeast Utilities Services Company
G. D. Frieling, Senior Project Engineer,
Central Electricity Research Laboratory,
D. Goetcheus, Supervisor of Metallurgical Analysis,
R. L. Goyette, Manager, Systems Operations,
W. A. Haller, Manager, Technical & Environmental Services,
Duke Power Company
R. L. Haueter, Chairman (1977-1980), Director of Special Projects,
Forward ix
C. B. Hardee, Mechanical Equipment Specialist,

W. Harrison, Lead Engineer, Nuclear Engineering Department,
J. Kieffer, Service de la Production Thermique,
T. Kurosawa, Manager, Reactor Chemistry Department,
Energy & Environment Laboratory,
CRIEPI, Japan
R. Lane, Manager, Mechanical Engineering,
H. Mahlman, Principal Chemical Engineer,
R. E. Millard, Supervisor, Steam Generation Chemistry,
D. M. Noble, Chairman (1980-1982), Plant Superintendent,
H. N. Paduano, Section Supervisor,
A. B. Poole, Senior Nuclear Systems Engineer,
G. A. Reed, Manager, Nuclear Power Division,
A. C. Rogers, Nuclear Engineering Supervisor,
Arizona Nuclear Power Project
S. Rothstein, Consulting Engineer,
Consolidated Edison Company of New York
P. F. Santoro, Chief, Generation Systems Mechanical Engineering,
J. R. Schepers, Senior Engineer,
J. Schmieder, Executive Vice President & Chief Engineer,
D. G. Slear, Senior Engineer,
GPU Service Corporation
G. Slifer, Chemical Plant Engineer,
x Steam Generator Reference Book, Revision I
A. L. Sudduth, Design Engineer,

Duke Power Company
H. P. Walker, Assistant Plant Manager, Vogtle,
W. Way,
B. D. Withers, Vice President, Nuclear,
I. S. Woolsey, Central Electricity Research Laboratory,
A. W. Zeuthen, Metallurgical & Welding Engineer,
Steam Generator Owners Group IIExecutive Committee

J. C. Blomgren, Manager, Steam Generator Program
D. W. Cockfield, Vice President, Nuclear Division
Portland General Electric
B. L. Dow, Jr., Manager, Nuclear Services,
L. R. Eliason, General Manager, Nuclear Plants,
Northern States Power
C. W. Fay, Vice President, Nuclear Power,
R. Garnsey, Manager, R&D, Sizewell "B" Power Station,
Duke Power Company
W. Josiger, Manager, Special Projects,
R. P. McDonald, Senior Vice President, Nuclear Generation,
J.- P. Mercier, Chef, Department Materiel,
J. A. Mundis, Supervisor, Nuclear Services,
Forward xi

P. F. Santoro, Director, Generation Projects,
J. W. Williams, Jr., Group Vice President, Nuclear Energy,
B. D. Withers, President & Chief Executive Officer,
Wolf Creek Nuclear Operating Corporation
Steam Generator Owners Group IIMember Representatives

R. P. McDonald, Senior Vice President, Nuclear Generation,
S. H. Steinhart, Assistant Division Manager, Mechanical Engineering,
American Electric Power Services Corporation
R. M. Butler, Director, Technical Services,
J. Vorees, Manager, Nuclear Operations,
B. Dow, Manager, Nuclear Services,
S. McGregor, Arkansas Nuclear One,
J. A. Tiernan, Vice President, Nuclear Energy,
G. P. Beatty, Jr., Manager, Special Projects,
A. B. Cutter, Vice President, Nuclear Engineering & Licensing,
R. Garnsey, Manager, R&D, Sizewell B Power Station,
J. M. Zamarron, Jefe Ing. Y Suministros,
Central Nuclear de Almaraz, Spain
F. Alomar, Engineering Manager,
Central Nuclear de Asco, Spain
T. Mizuno, Director & General Manager, NPD, R&D Headquarters,
CRIEPI, Japan
xii Steam Generator Reference Book, Revision I
M. Yasuda, Deputy General Director,

CRIEPI, Japan
J. D. Deress, Project Engineering Manager,
E. R. McGrath, Executive Vice President, Power Generation,
M. Selman, Vice President, Nuclear Power,
Consolidated Edison Company of N.Y., Inc.
J. Schepers, Superintendent, Chemistry & Radwaste,
W. A. Hailer, Manager, Technical & Environmental Services,
Duke Power Company
J.- P. Mercier, Chef, Dept. Materiel,
F. Dierkens, Manager,
Electronucleaire
D. Borgese, Director,
ENEL-CRTN ENEL
J. W. Williams, Jr., Group Vice President, Nuclear Energy,
F. A. Cremades, Jefe de la Div. de Ingenieria y Equipos,
Fuerzas Electricas De Cataluna Fecsa
C. D. Pugh, Senior Plant Engineer,
D. K. Croneberger, Director, Engineering & Design,
GPU Nuclear Corporation
S. M. Dew, Deputy Project Manager,
J. Evans, Vice President, Power Supply Operations,
Kansas City Power & Light Company
D. T. McPhee, Senior Vice President, Systems Power Operations,
C. J. Ross, Senior Director, Nuclear Affairs,
G. L. Koester, Vice President, Nuclear,
Kansas Gas & Electric Company
Forward xiii
D. E. Adams, III, Utility Engineer,

Louisiana Power & Light Company
K. R. Iyengar, Nuclear Support Manager,
W. Josiger, Resident Manager,
G. D. Baston, Director, Nuclear Engineering & Operations Services,
P. F Santoro, Director, Generation Projects,
L. R. Eliason, General Manager, Nuclear Plants,
R. Andrews, Division Manager, Nuclear Production,
Omaha Public Power
R. L. Andrews, Division Manager, Nuclear Production,
T. J. McIvor, Supervisor, Technical,
J. D. Shiffer, Vice President, Nuclear Power Generation,
D. W. Cockfield, Vice President, Nuclear Division,
J. T. Boettger, Assistant Vice President, Nuclear Operations Support,
D. J. Jagt, Assistant General Manager, Project Engineering,
P. R. H. Landrieu, Vice President, Engineering & Construction,
T. J. Martin, Vice President, Nuclear Generation,
J. E. Maier, Senior Vice President, Operations & Engineering,
R. E. Smith, Chief Engineer,
R. J. Rodriguez, Assistant General Manager, Nuclear,
Sacramento Municipal Utility District
xiv Steam Generator Reference Book, Revision I
0. W. Dixon, Jr., Vice President, Nuclear Operations,

D. A. Nauman, Vice President, Nuclear Operations,
K. P. Baskin, Vice President,
J. A. Mundis, Supervisor, Nuclear Services,
L. B. Long, Manager, Nuclear Safety & Licensing,
Southern Company Services, Inc.
J. L. Hunyadi, Ing., Project Manager, Thermal Power,
N. A. Petrick, Executive Director,
SNUPPS
H. Abercombie, Nuclear Power Division,
T. Ziegler, Division of Nuclear Power,
J. C. Kuykendall, Vice President,
Texas Utilities Generating Company
D. R Schnell, Vice President, Nuclear,
Union Electric Company
P. Blanc,
Union Electrica, Fenosa, SA
R. M./Taylor, Nuclear Technical Services,
Virginia Electric & Power Company
J. W. Ogren, Director, Operations & Maintenance Support,
Virginia Power Company
W. A. Thornton, Supervisor, Systems Chemistry, Nuclear Operations,
C. W. Fay, Vice President, Nuclear Power,
B. D. Withers, President & CEO,
Wolf Creek Nuclear Operating Corporation.
Forward xv
Steam Generator Owners Group IITechnical Advisory Committee

R. J. Acosta, Supervisor, Nuclear Energy Section,
D. E. Adams, III, Chemistry Unit Coordinator,
G. P. Airey, Sizewell "B" Project,
B. Allen, Chemistry Department, Waterford Nuclear Plant,
J. A. Bailey, Manager, Mechanical Engineering,
Kansas Gas & Electric
P. A. Bauer, Project Engineer, Nuclear, Staff Support,
J. P. Berge, Chef du Groupe des Laboratoires,
Electricite de France
G. J. Bignold, Steam Water Chemistry Section,
P. Blanc,
Union Electrica, Fenosa, SA
J. C. Blomgren, Manager, Steam Generator Program,
D. Borgese, Director,
ENEL-CRTN
R. F. Brandt, Nuclear Plant Services Engineering,
D. W. Braswell, Superintendent, Engineering,
Comanche Peak Station,
E Brejani,
Ente Nazionale per l'Energia Elettrica
A. Buford,
G. R. Capodanno, Director, Fluid Systems,
W. C. Carr, Supervisor, Chemistry & Environment,
xvi Steam Generator Reference Book, Revision 1
J. 0. Cermak, Technical Director,

SNUPPS
S. B. Clay, Engineer, Nuclear Generation,
F. D. Crawford, Director, Nuclear Power,
F Alomar Cremades, Jefe de la Div. de Ingenieria y Equipos,
Fuerzas Electricas De Cataluna Feca
P. T. Crinigan, General Supervisor, Chemistry,
B. G. Croley, Group Manager, Technical Support,
South Carolina Electric & Gas
A. E. Curtis, III, Manager, Materials Engineering Division,
J. P. DeRoy, Engineer, Nuclear Operations,
P. M. Egebrecht, PVNGS Nuclear Chemist,
J. Engstr6m, Senior Engineer,
J. M. Fackelmann, Supervisor, Nuclear Materials & Chemistry,
J. J. Field, Superintendent, Technical Support,
G. Frederick, Manager, Nuclear Equipment,
Electronucleaire
G. D. Frieling, Superintendent, Systems Engineering,
J. A. Gabrielson, Engineering Staff Specialist, South Texas Project,
R. Garnsey, Manager, R&D, Sizewell "B "Power Station,
D. F Goetcheus, Supervisor, Metallurgical Analysis Section,
A. P. Graham, Nuclear Systems Engineering,
Forward xvii
D. R. Hafer, Assistant Manager, Heat Exchangers & Pumps Section,

American Electric Power Service Corporation
W. G. Hairston, III, Manager, Nuclear Engineering & Technical Support,
Duke Power Company
C. B. Hardee, Project Engineer, Special Projects,
J. L. Harness, Manager, Nuclear Staff Support,
Carolina Power & Light
C. T. Hillman, Palisades Nuclear Plant,
N. Hoadley, Lead Engineer, Wolf Creek Generating Station,
Kansas Gas & Electric Company
J. Johnston, Plant Maintenance,
R. D. Lane, Manager, Mechanical Engineering, Arkansas Nuclear One,
R. Lawson, Steam Generator Coordinator,
S. D. Lee, Plant Vogtle,
L. B. Long, Manager, Nuclear Safety & Licensing,
H. Mahlman, Staff Specialist, Engineering Design Division,
R. L. Mansfield, Junior Plant Engineer,
A. L. Matheny, Steam Generator Engineer,
S. McGregor, Arkansas Nuclear One,
T. Mclvor, Supervisor, Ft. Calhoun Unit 1,
T. Mizuno, Director, General Manager, NPD, R&D Headquarters,
CRIEPI, Japan
xviii Steam Generator Reference Book, Revision 1
J. E. Moaba, Section Supervisor, Materials, Codes & Inspection,

N. Nilsen, Engineer, Special Projects, Nuclear Engineering,
J. W. Ogren, Director, Operations & Maintenance Support,
W. S. Orser, General Manager, Technical Functions,
R. Pascali, Senior Research Engineer,
Ente Nazionale per L'Energia Electrica
A. C. Passwater, Superintendent, Licensing,
T. L. Patterson, Manager, Technical Support,
R. P. Pearson, Production Engineer, Prairie Island,
A. Bruce Poole, Senior Nuclear Systems Engineer,
L. Pope, Comanche Peak
C. David Pugh, Senior Plant Engineer, Vogtle Electric Generating Plant,
D. A. Rickett, Senior Engineer,
S. Rothstein, Chemical & Metallurgical Engineer,
G. C. Russell, Nuclear Generation Engineer,
P. Saint-Paul, Department Etudes de Materiaux,
E Schwoerer, Technical Director,
SNUPPS
V. J. Shanks, Superintendent, Chemistry,
Forward xix
A. L. Sudduth, Design Engineer, M&N Division,

Duke Power Company
R. M. Taylor, Nuclear Technical Services,
W. A. Thornton, Nuclear Operations Support,
P. E. Troy,
J. H. Uhl, Plant Mechanical Engineer,
J. J. Villa,
Central Nuclear de Trillo
H. P. Walker, Assistant Plant Manager, Vogtle Electric Generating Plant,
R. P. Wendling, Manager, Mechanical Engineering,
C. P. Yundt, General Manager, Technical Functions, Nuclear Division,
Steam Generator Reliability Project Executive Group

G. D. Baston, Director, Nuclear Engineering & Operations,
Northeast Utilities
W. Conway, Executive Vice President,
Arizona Public Service
D. K. Croneberger, Director, Engineering,
GPU Nuclear
R. P. McDonald, President,
S. Rosen, Vice President,
D. F. Schnell, Vice President, Nuclear,
M. Selman, Senior Vice President,
Consolidated Edison of NY, Inc.
J. L. Skolds, Vice President, Nuclear Operations,
South Carolina Electric & Gas
xx Steam Generator Reference Book, Revision 1
R. E. Smith, Senior Vice President, Production & Engineering,

M. Tuckman, Senior Vice President,
Duke Power Company
J. D. Woodard, Vice President, Nuclear,
J. J. Zach, Senior Manager, Nuclear Engineering,
Steam Generator Reliability Project Member Representatives
M. F. Hodge, Mechanical/Chemical Engineering Manager,

E. C. Sterling, Manager, Nuclear Engineering Production,
R. Lane,
J. Lozano,
Asociacion Nuclear De Asco, Spain
R. E. Denton, Manager,
R. B. Richey, Vice President,
G. Vaughn, Vice President,
J. M. Zamarron,
Central Nuclear de Almaraz, Spain
C. J. Durkin, Jr.,
Consolidated Edison Company of N.Y.,
R. P. Noberini, Manager, Technology Transfer R&D,
H. Hirano,
CRIEPI, Japan
Forward xxi
H. Takaku, Manager, Mechanical Engineering Section,

CRIEPI, Japan
W. A. Haller, Manager, Nuclear Technical Services,
Duke Power Company
J. D. Sieber, Vice President, Nuclear Group,
J. P. Berge, Controleur General Adjoint,
J.-P. Hutin, NSSS Division Manager,
G. Frederick, Manager, Nuclear Generation Department,
Electronucleaire
R. C. Widell, Director, Nuclear Operations,
W. H. Bohlke, Vice President, Nuclear Engineering & Licensing,
K. R. Craig, Nuclear Energy Division,
C. K. McCoy, Vice President,
B. D. Elam, Jr., Director, Components & Structures,
S. Rosen, Vice President
D. E. Baker, Manager, Nuclear Operations Support & Assessment,
D. Sturniolo, Principal Radiological Engineer,
Maine Yankee Atomic Power Company
W. A. Josiger, Resident Manager, Nuclear Operations,
G. D. Baston, Director, Nuclear Engineering & Operations,
E. Watzl, General Manager, Prairie Island,
G. P. Airey, Sizewell "B" Project Management Team,
Nuclear Electric
xxii Steam Generator Reference Book, Revision I
R. Garnsey, Manager, Safety & Technology,

Nuclear Electric
S. K. Gambhir,
J. Brown, Manager, Metallurgical Research Department
Ontario Hydro
J. A. Hayes, Supervisor, RECE,
L. F Womack, Manager, Engineering,
C. P. Yundt, Trojan Nuclear Plant,
H. Berrick,
R. E. Smith, Senior Vice President, Production & Engineering,
D. Moore,
J. L. Skolds, Vice President, Nuclear Operations,
R. P. McDonald, Jr., President,
J. D. Woodard, Vice President,
K. Persson, Manager, Ringhalsverket 3 & 4,
N. C. Kazanas, Vice President, Operations Services,
D. R. Keuter,
E. Matranga,
Toledo Edison Company
M. R. Blevins, Manager, Nuclear Operations Support,
TU Electric
D. F. Schnell, Vice President, Nuclear,
Forward xxiii
J. J. Zach, Senior Manager, Nuclear Engineering,

M. L. Marchi, Plant Manager,
Steam Generator Reliability Project Technical Advisory Group

J. D. Abramovici, H/X & PV Manager,
R. Affolter, Supervisor, Systems Engineering,
G. P. Airey, Sizewell "B" Project Management Team,
Nuclear Electric
R. L. Baker, Consulting Engineer,
G. J. Bollini, PISGV Project Manager,
Spanish Utilities
P. Brown,
Yankee Atomic
J. M. Cate, Engineer, Special Services,
K. R. Craig, Nuclear Energy Division,
P. T. Crinigan, Nuclear Energy Division,
M. R. Danak, Senior Staff Engineer,
Public Service Electric & Gas
R. J. Dolansky, Nuclear Generation Department,
R. W. Eaker, Technical Systems Manager II, Chemistry,
Duke Power Company
J. Engstr6m, Senior Engineer,
xxiv Steam Generator Reference Book, Revision I
J. M. Fackelmann, Principal Engineer,

Northeast Utilities
J. S. Frick, Supervisor, Special Projects & Performance Engineering,
G. D. Frieling, Manager, Nuclear Engineering,
J. K. Gillen, General Supervisor, Chemistry,
D. E Goetcheus, Maintenance Manager, NSS/SG,
D. J. Hampshire, Steam Generator Technical Coordinator,
D. F. Harrison, Senior Engineer,
Entergy Operations, Inc.
P. Hernalsteen, Director, Mechanical Division,
Laborelec
J. W. Hoss, Senior Engineer, Results Engineering,
TU Electric
L. D. Hughes, Supervisor, Primary Systems Engineering,
G. A. Kammerdeiner, Director, Materials & Standards Engineering,
M. L. Layton, Support Engineer, Chemistry/Radiochemistry,
R. Lewis, Supervisor, Results Engineering,
P. C. Lichtenberger, Supervising Scientist, Corr. & Tri. Technical Section,
Ontario Hydro
A. L. Matheny, Steam Generator Engineer,
E. Matranga,
J. A. "Buzz" Miller,
K. Neese, Steam Generator Project Engineer,
Forward xxv
N. Nilsen, Senior Engineer,

F. Nordmann, Associate Manager, Chemistry,
R. P. Pearson, Superintendent, Steam Generator Systems,
C. T. Polidoroff, Senior Nuclear Generation Engineer,
P. Skulte, Metallurgical Engineer,
J. F. Smith, Manager, Materials Engineering & Inspection Services,
C. S. Smoker, Supervisor, Plant Quality Programs,
P. Spekkens, Section Head, Corrosion Chemistry,
Ontario Hydro
D. Sturniolo, Principal Radiological Engineer,
Maine Yankee Atomic Power Company
K. Sweeney,
H. Takamatsu, Manager, SG Project, Nuclear Power Operations Dept.,
CRIEPI, Japan
R. H. Thompson, Nuclear Results Specialist,
G. Vaughn, Vice President,
I. S. Woolsey,
Nuclear Electric
The Electric Power Research Institute and my colleagues there have

provided administrative and intellectual support in bringinfg this effort to
completion. The support and guidance provided by John J. Taylor, Robin
L. Jones, and Charles S. Welty have been invaluable.
J. Peter N. Paine
Palo Alto, California
December 1994
PREFACE
Operation and maintenance of PWR Steam Generators with high

availability are among the most difficult challenges that face owners. In
response to corrosion-related difficulties causing increased O&M costs, a
group of utilities, owning or planning PWR nuclear power plants, formed
the Steam Generator Owners Group in 1977. The purpose of this group
was to sponsor a concerted effort to find solutions to these steam generator
operating concerns.
The Owners Group asked EPRI to manage the technical program. In
March 1977, EPRI established the Steam Generator Project Office in EPRI's
Nuclear Power Division. On July 26,1977, the Owners Group met in Chicago
to adopt a draft charter and to elect an executive committee. Initially, twenty-
one utilities joined the Steam Generator Owners Group and committed over
$30 million to support the program. Other utilities later became members
resulting in a total contribution of $36 million. In parallel, EPRI supported
steam generator studies at about $2.5 million per year. This work was
coordinated with the program sponsored by the Steam Generator Owners
Group.
Subsequently, because of continued challenges that emerged while this
program was underway, a new group, called Steam Generator Owners
Group II, was formed in 1983 to address these new issues over the period
1983-1986. Ultimately, the utility membership increased to 37, providing
$29 million in support of the program. In parallel, the total EPRI support of
the Owners Group programs over the period 1977 to 1986 was about
$26 million.
ii Steam Generator Reference Book, Revision I
The Steam Generator Owners Group II was successfully completed.

However, additional causes of steam generator unavailability continued to
arise and a new major project, called the Steam Generator Reliability Project,
completely funded by EPRI at a level of $30 million, was formed to operate
over the period 1987-1992. In addition to the U. S. members, seven
international utilities have participated in the Steam Generator Reliability
Project under technical exchange agreements.
The results of the programs have been transmitted to Steam Generator
Owners Groups I and II and to members of the Steam Generator Reliability
Project as described below.
Final Reports of Individual Contracts

Final reports of more than 175 individual contracts resulting from the
Steam Generator Owners Group I (SGOG I) efforts have been issued. These
reports were listed and summarized in one document entitled Steam
GeneratorProgramSummary dated March 31, 1983. Similarly, the results of
more than 225 individual contracts resulting from the Steam Generator
Owners Group II (SGOG II) effort have been issued. These reports are
summarized in the Steam GeneratorProgramSummary dated April 1989. The
Steam Generator Reliability Project (SGRP) effort is still underway and a
number of final reports have been issued. The progress of this effort is
reported periodically in Status of Steam GeneratorContracts. The last status
report is dated October 1991.
Technology Transfer Packages

In 1983, a set of technology transfer packages were prepared and
distributed to the members. Each information package contained four
videotapes, seven executive summaries, seven technical summaries, and
seven slide presentations. The objective was to transmit the results of the
program as effectively as possible both to the utility management and the
staff.
Design and Operation Guidelines to Minimize Steam Generator

Corrosion
Through the joint efforts of SGOG Advisory Committees and the Steam
Generator Project Office, 14 Design and Operation Guidelines were prepared
and issued during 1980-1983. These guidelines were based on the results
of SGOG research programs and utility, architect/engineer, and vendor
experiences as reported in the literature and at industry meetings. In
December 1986, they were updated and consolidated into a two volume set
Preface iii
of Plant Design and Modifications Guidelines and one volume of Plant and
Steam Generator Operation Guidelines. Also, two guidelines on the Design
and Operationfor Makeup Water Treatment System were issued at the same
time.
Two additional major guidelines have been issued and are being used
extensively by the industry. Revision 0 of the PWR Secondary Water Chemistry
Guidelines was published in 1982. Revision 1 was issued in 1985 and Revision
2 in 1988. The other, the PWR Steam GeneratorExamination Guidelines, was
issued informally for the utility community in early 1980. The original
version was issued in 1981 and revised in 1984. In 1988, Revision 2 of these
guidelines was issued as an EPRI report:
Steam Generator Reference Book

The original version of this book was based primarily on work
sponsored by the first Steam Generator Owners Group, but it also included
work funded by EPRI, pertinent literature, and experiences of the electric
utility industry. It documented the state-of-the-art in PWR steam generator
technology at that time and provided a ready reference for operators,
owners, and designers of PWR nuclear power plants. The Reference Book
was intended for use with the Design and Operation Guidelines described
above.
This Revision I of the Reference Book updates the original version by
including the results of research completed since the issuance of the book.
It includes work completed under Steam Generator Owners Group II and
much of the work completed under the Steam Generator Reliability Project.
The format of the book has been revised so that individual chapters can be
updated as necessary in the future.
The purpose of the original and Revision 1 of the Reference Book is to
integrate the results of research contracts and other available information
into a format that can be easily used by the utility industry. In addition, the
Reference Book summarizes options and recommendations for improving
the expected operating life of PWR steam generators. The options will be
utility specific and will depend on such factors as the age of the-plant,
design/construction, plant siting, utility policies, etc. (Most of these
recommendations are described in detail in the Design and Operation
Guidelines.)
Chapter 1 briefly describes the PWR steam generator and provides the
background that led to the formation of the original Steam Generator
Owners Group and the subsequent programs.
Chapter 2 reviews the history of PWR steam generator performance. It
includes a description of a typical PWR power plant and how the design
iv Steam Generator Reference Book, Revision 1
has evolved in 20 years. This chapter also contains a discussion of the

problems that have occurred in those steam generators.
Chapter 3 provides the bases and objectives of Steam Generator Owners
Groups I and II and the Steam Generator Reliability Project. The research
findings are summarized and several field applications are described.
Chapter 4 provides recommendations and options generated by Steam
Generator Owners Group programs and other work. The general areas
covered include chemistry control, plant operation, balance of plant design,
and steam generator design. The Design and Operation Guidelines are
summarized.
Chapters 5 through 19 address each of the known and anticipated steam
generator challenges and the factors affecting them. The relative importance
of each factor is included in the discussion, as well as, where possible, the
uncertainties of those judgments. The specific causes of steam generator
unavailability are identified so that corrective actions may be focused.
Chapters 20 through 27 discuss possible actions to minimize the
problems listed in Chapter 2. The benefits and risks of each of these actions
are documented. Where these actions have been implemented, the results
are discussed.
Appendix A includes a listing of final reports by contract number and
by final report number. Appendix B details plant characteristics of PWR
steam generators worldwide.
CONTENTS
I Introduction 1-1
PWR Steam Generators 1-2
Background 1-3
References 1-9
2 Steam Generator Performance History 2-1

Introduction 2-2
Pressurized Water Reactor (PWR) Power Plant Description 2-2
Steam Generators 2-3
Recirculating Steam Generators 2-4
Once-Through Steam Generators (OTSGs) 2-11
Steam Generator Fabrication 2-14
Balance-of-Plant Systems 2-15
Condenfsers 2-18
Condensate System 2-19
Condensate Polishing System 2-19
Feed System 2-19
Drain System 2-19
Blowdown Processing System 2-20
Makeup Water System 2-20
Moisture Separator Drain Routing 2-23
Operating Plant Performance Problems 2-23
References 2-34
ii Steam Generator Reference Book, Revision 1
3 Steam Generator Programs 3-1

Introduction 3-2
Steam Generator Owners Group I 3-2
Background 3-2
Technical Program 3-3
Subprogram 0801: Quality Nondestructive
Examination Techniques 3-5
Subprogram 0802: Evaluation of On-Order
Steam Generator Designs 3-6
Subprogram 0803: Perform Safety Analysis and
Consequence Evaluations 3-6
Subprogram 0804: Determination and Verification of
Required Water Chemistry Limits 3-6
Subprogram 0805: Evaluate Alternate Steam Generator
Materials and Processing 3-6
Subprogram 0807: Develop Neutralizer for Crevice Acids 3-6
Subprogram 0809: Thermal-Hydraulic Analysis
and Evaluation 3-7
Subprogram 0810: Evaluate Methods for Reducing
Carry-Over and Decreasing Pressure
Drop Through Steam Separators 3-7
Subprogram 0811: Development and Technical Support
of Initial Plant Application of
Chemical Cleaning 3-7
Subprogram 0812: Evaluate Secondary System
Layup Chemistry 3-7
Subprogram 0813: Evaluate Steam Generator
Layup and Transient 3-8
Subprogram 0814: Evaluate Methods for Reducing
Tube Vibration, Wear, and Fretting 3-8
Subprogram 0816: Evaluation of On-Line Chelant Additions 3-8
Subprogram 0818: Destructive Analysis of
Steam Generator Components 3-8
Subprogram 0820: Evaluate Condensate Polishers 3-8
Subprogram 0821: Develop High and Low
Temperature Filters 3-9
Table of Contents iii
Subprogram 0822: Evaluate and Test Boiler

Water Cleanup Systems 3-9
Subprogram 0823; Instrument Operating Units 3-9
Subprogram 0824: Evaluation of Hydrogen
Evolution Monitoring 3-9
Subprogram 0830: Improve Control of Condenser
Cooling Water Leakage 3-9
Subprogram 0836: Evaluate Secondary System Oxygen
Control Measures 3-9
Subprogram 0839: PWR Steam Generator
Technology Transfer 3-10
Steam Generator Owners Group II 3-10
Background 3-10
S301: Improved Steam Generator Nondestructive
Examination Techniques 3-10
S302: Arrest Steam Generator Intergranular Attack (IGA)
and Tube Cracking (SCC) in and Above the Tubesheet
Crevice and in Other Crevice and Dryout Locations 3-12
S303: Causes and Corrective Actions for
Primary Water Cracking of Steam Generator Tubing 3-12
S304: Destructive Analysis of Steam Generator Components 3-12
S305: PWR Steam Generator Chemical Cleaning Process
Development and Evaluation 3-12
S306: Evaluation and Improvement of Steam Generator
Performance and the Effectiveness of Preventive
and Corrective Measures 3-12
S307: PWR Steam Generator Technology Transfer 3-13
S308: Causes and Corrective Actions for Pitting
of Steam Generator Tubing 3-13
S309: Minimize the Effects of Sludge, Scale, and
Deposits on Corrosion in Steam Generators 3-13
S310: Tube Fretting and Fatigue in
. Preheat Steam Generators 3-13
S311: Tube Support Materials and Designs
and Crevice Corrosion 3-14
Steam Generator Reliability Project 3-14
iv Steam Generator Reference Book, Revision I
Background 3-14
S401: Improved Steam Generator Water Chemistry -
Field Studies and Water Chemistry Guidelines 3-14
S402: Improved Steam Generator Water Chemistry
Through Improved Balance-of-Plant Operation 3-15
S403: Improved Steam Generator Impurity
Removal Techniques 3-17
S404: Demonstrated Steam Generator In-Service Inspections 3-17
S405: Technology Transfer 3-17
S406: Steam Generator Primary Side Remedies 3-18
S407: PWR Steam Generator Secondary Side Remedies 3-18
S408: Improved Materials and Designs for Steam Generator
Repair and Replacement 3-19
S409: Alternate Secondary Water Chemistry 3-19
S410: Tube Vibration, Mechanical Damage,
Fretting, and Wear 3-19
S411: Thermal-Hydraulic Code Evaluation and Validation 3-20
S412: Steam Generator Strategic Planning Guidelines 3-20
Research Findings 3-20
Tube Wastage (Thinning) 3-20
Denting 3-21
Primary Side Stress Corrosion Cracking 3-22
Secondary Side Intergranular Attack and
Stress Corrosion Cracking 3-23
Pitting 3-25
Corrosion Fatigue 3-25
Tube Fretting and Wear 3-26
Steam Generator Performance 3-27
Steam Generator Replacement/Service Life 3-27
Lost Capacity 3-27
Forced Outages 3-28
References 3-31
Table of Contents v
4 Recommendations 4-1
Introduction 4-2
4.1 Steam Generator Design 4-3
4.1.1 Tube and U-Bend Support Structure Materials 4-3
4.1.2 Tube Support Structure Designs 4-4
4.1.3 Tube Materials 4-5
4.1.4 Tube Fabrication - U-Bends 4-6
4.1.5 Tube Fabrication - Tubesheet Expansion 4-6
4.1.6 U-Bend Support Structure 4-7
4.1.7 Steam Generator Tube Bundle Assembly 4-8
4.1.8 Mechanical Design - Penetrations 4-8
4.1.9 Mechanical Design -
Steam-Water Separation Equipment 4-9
4.1.10 Mechanical Design - Blowdown 4-10
4.1.11 Mechanical Design - Feed Ring 4-10
4.1.12 Mechanical Design - Shell 4-11
4.1.13 Shell Fabrication 4-12
4.1.14 Thermal Hydraulic Design 4-12
4.1.15 Design Against Water Hammer and Dynamic Flow
Instability 4-13
4.1.16 Archive Samples 4-13
4.1.17 Steam Generator Cleanliness 4-14
4.1.18 Steam Generator Repair and Replacement 4-15
4.1.18.1 Alternate Tube Repair Limits 4-15
4.1.18.2 Tube Repair Options 4-16
4.1.18.3 Tube Peening and In Situ Stress Relief 4-17
4.1.18.4 Girth Weld Cracking 4-19
4.1.18.5 Steam Generator Replacement 4-20
4.1.18.5 Repair/Replacement Options 4-20
4.2 Steam Generator Operations 4-22
4.2.1 Steam Generator Cleanliness and Cleanliness of
Interfacing Systems During Construction 4-22
4.2.2 Minimizing Corrosion Product and
Ionic Impurity Transport 4-24
vi Steam Generator Reference Book, Revision I
4.2.3 Enhancing Hideout Return Processes 4-24

4.2.4 Crevice Flushing 4-25
4.2.5 Sludge Lancing 4-26
4.2.6 Chemical Cleaning 4-26
4.2.7 Thot Reduction 4-27
4.3 Plant Design, Modification, and Operation 4-28
4.3.1 Steam Generator and Balance-of-Plant Layup 4-29
4.3.2 Steam Generator Inspection and
Nondestructive Testing 4-30
4.3.3 Steam Generator Blowdown 4-30
4.3.4 Condensate Polishing 4-31
4.3.5 Condensers 4-33
4.3.6 Water Chemistry 4-34
4.3.7 Secondary Systems Design 4-35
4.3.8 Makeup Water Treatment System 4-36
4.4 Design and Operation Guidelines
to Minimize Steam Generator Corrosion 4-39
4.4.1 Guidelines for PWR Steam Generator Tubing
Specifications and Repair 4-40
4.4.2 Plant Design for PWR Steam Generator Inspection
and Nondestructive Testing 4-40
4.4.3 Plant Design Guidelines for Layup and Cleanup of
Steam, Feedwater, and Condensate Systems 4-41
4.4.4 Guidelines for Design of PWR Steam Generator
Chemical Cleaning Systems 4-42
4.4.5 Guidelines for Design of Steam Generator
Blowdown Systems 4-42
4.4.6 Condensate Polisher Design, Operation, and
Maintenance Guidelines 4-43
4.4.7 Design Guidelines for Plant Secondary Systems 4-44
4.4.8 Plant Design for PWR
Steam Generator Replaceability 4-44
4.4.9 Design Guidelines for Layup of PWR Steam Generator 4-45
4.4.10 Design and Operating Guidelines for Plant Condensers 4-45
4.4.11 Guidelines for Impurity Removal and Control 4-46
Table of Contents vii
4.4.12 Preoperational Flushing, Cleaning, and Layup of PWR

Steam, Feedwater, and Condensate Systems 4-47
4.4.13 Guidelines to Minimize Contamination of PWR Steam
Generators During Plant Construction 4-47
4.4.14 Steam Generator Inspection Guidelines 4-48
4.4.17 PWR Primary Water Chemistry Guidelines 4-49
4.4.18 PWR Secondary Water Chemistry Guidelines 4-49
4.4.19 Guidelines for the Design and Operation of
Makeup Water Treatment Systems 4-49
References 4-50
5 Steam Generator Degradation (Technical Description

of Causes, Consequences, and Prevention) 5-1
6 Thermal and Hydrodynamic Analysis

of Steam Generators 6-1
7 Primary Water Stress Corrosion Cracking 7-1

Introduction 7-2
Babcock & Wilcox 7-3
Combustion Engineering 7-4
Kraftwerk Union 7-4
Types of PWSCC Degradation 7-4
Extent of PWSCC Degradation 7-7
Description of PWSCC Degradation 7-13
Cracking at Expanded Regions Within Tubesheet 7-13
Cracking at Expansion Transitions 7-14
Depth of Expansion Transition Within the Tubesheet 7-15
Expansion Method 7-15
Longitudinal Cracks at Expansion Transitions 7-17
Circumferential Cracks at Expansion Transitions 7-24
Cracking at Dented Tube Support Plate Intersections 7-27
Cracking at U-Bends 7-27
Cracking at U-Bend Apex 7-27
Cracking at U-Bend Tangent 7-31
Causes of PWSCC 7-32
viii Steam Generator Reference Book, Revision I
General Discussion 7-32

Material Susceptibility 7-34
Final Mill Annealing Temperature 7-37
Carbon 7-39
Sensitization and Thermal Treatment 7-43
Impurities 7-45
Stress 7-45
Expansion Transition Stresses 7-45
Dented Tube Support Plate Intersections 7-47
U-Bend Stresses 7-47
Stress Threshold for LTMA Alloy 600 Material 7-47
Influence of Stress on Time to PWSCC 7-48
Environment 7-50
Temperature 7-50
Hydrogen 7-53
Lithium 7-53
Electrochemical Potential 7-54
Rate of PWSCC Progression 7-54
Expansion Transition PWSCC 7-56
Inner Row U-Bend PWSCC 7-56
Detection of PWSCC 7-58
Bobbin Coil Eddy Current Testing 7-58
Rotating Pancake Coil Eddy Current Testing 7-58
Ultrasonic Testing 7-59
Leak Testing 7-59
Dented Region Inspection 7-60
References 7-61
8 Tube Support and Tubesheet Corrosion

(Acidic Crevices-Tube Denting) 8-1
Introduction: Description and Causes of Tube Denting 8-2
PWR Steam Generator History 8-2
Steam Generator Design 8-2
Steam Generator Materials 8-2
Table of Contents ix
Denting Experience 8-3

Failure Analysis of Support Plate Sections 8-3
Effects on Tubes, Tube Supports, and Tubesheet 8-6
Causes of Denting 8-9
Corrosion Processes 8-9
Nature of the Oxide/Corrosion Product Layer 8-11
Crevice Chemistry 8-14
Effect of Dissolved Oxygen 8-17
Laboratory Denting Experiments 8-18
Heat Flux and Concentration Effects 8-18
Isothermal Capsule Tests 8-20
Capsule Design 8-20
Reference Chemistry 8-20
Corrosion Process 8-21
Corrosion in Various Chloride Solutions 8-22
Effect of Temperature 8-30
Alternate Support Plate Materials 8-32
Effects of Other Anions 8-34
Heated Crevice Tests 8-36
Model Boiler Tests 8-40
Effect of Superheat 8-45
Summary of Corrosion Rate Test Data 8-47
Prediction of Dent Initiation 8-48
Conclusions 8-52
Improved Operating and Maintenance Practices 8-53
Improved Steam Generator Designs 8-54
References 8-54
9 Tube Wastage and Phosphate

Secondary Water Chemistry 9-1
Introduction 9-2
PWR Plant Experience 9-3
Nature of Phosphate Wastage 9-3
x Steam Generator Reference Book, Revision I
Field Experience 9-4

Laboratory Studies 9-12
Sodium Phosphate Solution Chemistry 9-12
Potassium Phosphate Solution Chemistry 9-16
Sodium Phosphate Corrosion Tests 9-17
Interactions of Phosphate With Magnetite 9-22
Discussion 9-23
Conclusion 9-24
References 9-26
10 TUBE PITTING 10-1

Introduction 10-2
Scope of Problem 10-2
Impact of Tube Pitting on Plant Performance 10-2
NDE Detection 10-3
Location of Pit Indications 10-3
Detection and Early Indications of Tube Pitting 10-5
PWR Field Data 10-9
Introduction 10-9
Major Pitting 10-10
Indian Point 3 10-10
Millstone 2 10-15
Connecticut Yankee 10-20
Trojan 10-20
Retired Surry Steam Generator 10-20
Calvert Cliffs 10-23
PWR Plant Experience - Minor Pitting 10-24
Summary of PWR Field Data 10-24
Laboratory Tests 10-24
Introduction 10-24
Summary of SGOG I Work 10-26
Ohio State University Test Results 10-27
Combustion Engineering Test Results 10-30
Table of Contents xi
Brookhaven National Laboratory 10-40

Battelle Columbus Division's Test Results 10-44
University of Leuven Test Results 10-50
Causes of Pitting 10-52
Sludge and Tube Scale 10-53
Oxidizing Conditions 10-54
Acid Chloride 10-54
Operating Temperature 10-57
Theory of Steam Generator Tube Pitting 10-57
Location of Pitting 10-58
Pit Initiation 10-59
Pit Propagation 10-59
Characteristics of Units Most Likely to Develop Pitting 10-63
Utility-Implemented Corrective Actions 10-64
Recommendations to Prevent/Mitigate
Steam Generator Tube Pitting 10-64
Sludge and Scale 10-68
Acid Chloride 10-68
Oxidizing Conditions 10-69
Temperature 10-69
Miscellaneous Recommended Actions 10-69
Layup Chemistry 10-69
Eddy Current Inspection 10-70
Sleeving 10-70
Tube Examination 10-71
References 10-71
11 Tube Supports and Tubesheet Corrosion -

Alkaline Denting 11-1
Introduction 11-2
Plant Experience 11-2
Denting in PWR Plants 11-2
Fort Calhoun 11-2
D.C. Cook Units l and 2 11-2
xii Steam Generator Reference Book, Revision I
Point Beach Units 1 and 2 11-3

San Onofre Unit 1 11-3
Summary of Plant Experience 11-3
Possible Interaction Between Alkaline Denting and IGA/SCC 11-4
Causes and Mechanism of Alkaline Denting 11-5
Laboratory Test Results 11-9
Improved Operating and Maintenance Practices 11-14
Improved Steam Generator Designs 11-15
Summary 11-16
References 11-16
12 Secondary Side Intergranular Attack and Stress

Corrosion Cracking - Plant Experience on AVT
Water Chemistry 12-1
Introduction 12-2
Tubesheet Crevices 12-2
Tube Support Crevices 12-3
Sludge Pile Crevices 12-3
Progression of Intergranular Corrosion 12-4
Definition of Terms 12-9
Summary of Plant Experience 12-12
Locations 12-12
Alloys 12-13
Contaminants 12-16
Temperature and Superheat 12-16
Stresses 12-17
Cold Work 12-17
Electrochemical Potential and Presence of Oxidants 12-17
Focus 12-18
Plant Experience: Definition of Problem 12-18
Operating History 12-18
Temperature (Hot Leg Versus Cold Leg Location,
and Hot Leg Temperature) 12-19
Seawater Site Versus Freshwater Site 12-22
Table of Contents xiii
IGA/SCC at Tubesheet Crevices of Part-Depth

Rolled Units With LTMA Tubing 12-23
IGA/SCC in Hot Legs of Combustion Engineering Units 12-26
IGA/SCC at Hot Leg, Drilled Hole Tube
Support Plate Crevices of Units With LTMA Tubing 12-26
Secondary Plant Materials 12-28
Impurity Ingress 12-29
Water Treatment History 12-32
Period of Operation Before First Occurrence -
Aging of Material 12-34
Presence/Use of Full Flow Condensate Polishers, Problems
With Other Ion Exchangers 12-35
Use of Boric Acid 12-36
Use of Morpholine 12-38
Elevation in Tube Bundle 12-38
Summary for Units With Mill Annealed Alloy 600 Tubing 12-39
Failure Analysis of Samples 12-41
Tubesheet Crevices - Recirculating Steam Generators 12-41
Freshwater Plants 12-42
Seawater Plants on AVT 12-47
Top of Tubesheet of Full-Depth Expanded Plants
(OD of Expansion Transition at Top of Tubesheet) -
Recirculating Steam Generators 12-50
Combustion Engineering Units 12-50
Westinghouse-Type Units 12-52
Hot Leg Sludge Piles - Recirculating Steam Generators 12-53
Freshwater Plants on AVT 12-53
Seawater Plants on AVT 12-53
Hot Leg Tube Supports - Recirculating Steam Generators 12-54
Freshwater Plants on AVT 12-54
Seawater Units on AVT 12-56
Hot Leg Free Spans in Recirculating Steam Generators 12-57
U-Bend Support Region in Recirculating Steam Generators 12-57
Cold Legs in Recirculating Steam Generators 12-58
Free Spans 12-58
xiv Steam Generator Reference Book, Revision I
Supports 12-59
Top of Tubesheet 12-59
Tubesheet Crevices 12-59
Upper Tubesheet Crevice and Upper Tube Supports in
Once-Through Steam Generators 12-60
Lower Tubesheet Crevice and Sludge Pile in
Once-Through Steam Generators 12-60
Summary of Failure Analysis Results 12-60
Effects of Intergranular Corrosion on Tubing Materials 12-63
Remedial Measures 12-65
Stress 12-66
Microstructure 12-66
Environment 12-67
Minimization of Tubesheet Crevices 12-68
Lower Temperature 12-68
Adding pH Neutralizers 12-68
Flushing or Soaking 12-69
Inleakage and Particulate Control 12-69
Control of the Concentrating Capability of the Sludge Pile 12-69
Ratio Control 12-70
Avoidance of Oxidizing Conditions 12-70
Protection of Sensitized Material 12-70
Inhibitors 12-70
Recommendations 12-70
References 12-71
13 Intergranular Corrosion of Alloy 600 From

Caustic Compounds 13-1
Introduction 13-2
Investigation of Caustic Conditions That Cause
Intergranular Corrosion 13-5
Behavior in Pure Caustic 13-6
1% Caustic 13-6
10% Caustic 13-8
50% Caustic 13-10
Table of Contents xv
Effects of Other Dissolved Species 13-12

Effects of Presence of Solid Oxides 13-16
Effects of Concentration 13-18
Effects of Temperature 13-19
Effects of Stress and Prestrain 13-23
Effect of Electrochemical Potential 13-27
Effects of Heat Treatment and Material Microstructure 13-30
Model Boiler Tests 13-33
Mechanisms of IGA/IGSCC in Caustic 13-36
Oxide Film Rupture 13-37
Grain Boundary Properties 13-38
Plasticity at Grain Boundaries 13-39
Grain Boundary Energy 13-39
Chemical Segregation at Grain Boundaries 13-40
Grain Boundary Carbides 13-41
Dealloying 13-42
References 13-44
14 Intergranular Corrosion From Acidic Compounds 14-1
15 Lead Cracking of Alloy 600 15-1

Introduction 15-2
Sources of Lead 15-2
Typical Lead Levels 15-3
Cracking in Steam Generator Tubes 15-5
Eddy Current Detection 15-6
Laboratory Experience 15-7
Lead Chemistry 15-13
Discrimination Between Cracking by
Lead and Other Corrodents 15-15
Possible Remedies for Lead Cracking 15-16
References 15-18
xvi Steam Generator Reference Book, Revision I
16 Tube Corrosion and Wear: U-Bend Supports (AVBs)

and TSP Intersections 16-1
Introduction 16-2
AVB Wear 16-6
Examination of Sample Tubes 16-8
Modeling of Flow-Induced Vibrations 16-11
Repairs and Remedial Measures 16-12
Conclusions 16-13
Cold Leg Thinning 16-14
Examination of Sample Tubes 16-16
Measurements of Tube Motions 16-18
Mechanisms . 16-18
Laboratory Corrosion Studies 16-21
Impact on Steam Generator Operations
and Potential Remedies 16-26
Non-Destructive Detection and
Measurement of Cold Leg Thinning 16-27
Preheater Wear 16-27
Fretting in OTSG Tubes 16-29
Summary of Concerns With Corrosion and
Wear at Support Structures in Nuclear Steam Generators 16-29
References 16-30
17 Tube Fretting and Wear 17-1
18 Tube Fatigue 18-1
19 Other Problems 19-1
20 Water Chemistry Control 20-1
21 Ionic Impurity Control 21-1
22 Water Chemistry Options 22-1
23 Corrosion Product Control 23-1

Table of Contents xvii
24 Material Selection and Alternative Designs for

Steam Generators 24-1
Material Alternatives for Tubes and Support Structures 24-2
Tube Materials 24-2
Stress Corrosion Cracking 24-4
Sulfates 24-33
Sulfur 24-35
Lead 24-36
Intergranular Corrosion 24-39
Pitting 24-41
Wastage 24-43
Discussion of Tube Alloys 24-47
Tube Support and Tubesheet Materials 24-51
Corrosion of Support Materials in Chlorides 24-53
Model Steam Generator Test in Sulfates 24-62
Electrochemical Tests 24-63
Corrosivity of Representative Environments 24-65
Effects of Support Hole Design on Corrosion 24-67
Matching Service Environments to
Support Alloy Composition 24-69
References 24-70
25 PWSCC Remedies 25-1

Introduction 25-2
Regulatory Requirements 25-2
PWSCC in Expanded Region 25-5
P* Criterion 25-5
F* Criterion 25-5
PWSCC at Expansion Transitions 25-6
PWSCC at Dented Tube Support Plate Intersections 25-7
PWSCC at U-Bends 25-7
Remedial Measure Strategies by Plant/Material Type 25-8
Early Steam Generators With Lower Strength Tubing 25-8
Steam Generators With Higher Strength, Non-Thermally
Treated, Alloy 600 Tubing 25-8
xviii Steam Generator Reference Book, Revision 1
Expanded Zone Within the Tubesheet 25-9

Part-Depth Roll Expansion 25-9
Full-Depth Expansion 25-9
Recent Steam Generators With Thermally Treated
Alloy 600 or Alloy 690 Tubing 25-11
Assessing Remedial Measure Effectiveness 25-11
Stainless Steel Tubes in Boiling Magnesium Chloride 25-11
Polythionic Acid and Sodium Tetrathionate ,25-12
10% Sodium Hydroxide at 550'F (288°C) 25-12
10% Sodium Hydroxide at 660'F (349°C) 25-13
Elevated Temperature Steam Tests 25-13
Primary Water Tests at Elevated Temperature 25-14
Primary Water Tests at Operating Conditions 25-14
Summary 25-15
Rotopeening 25-15
Process Description 25-15
Qualification Testing 25-17
Belgian/Westinghouse Development 25-18
French Rotopeening Development 25-18
Summary 25-19
Shot Peening 25-19
Testing of Uncracked Roll Transition Specimens 25-22
Testing of Pre-Cracked Roll Transition Specimens 25-22
Summary 25-23
Thermal Stress Relief and
Heat Treatment 25-24
Required Time and Temperature for Stress Relief 25-24
Tubing Stress Relief 25-24
Table of Contents xix
Tubing Sensitization 25-24

Tubing Recrystalization and Grain Growth 25-25
Tubesheet Transformation and
Fabrication Stress Relief Temperatures 25-25
Summary of Stress Relief Time and
Temperature Guidelines 25-27
Required Time and Temperature for
Metallurgical Improvement 25-27
In-Situ Stress Relief and Heat Treatment Methods 25-27
U-Bend Stress Relief 25-27
Local Expansion Transition Stress Relief 25-29
Global Tubesheet Stress Relief 25-32
Dented Tube Support Plate Intersection Stress Relief 25-34
Summary 25-34
Electroplating 25-34
Summary 25-36
Re-Expansion 25-36
Plugging 25-37
Welded Plugs 25-38
Explosive Plugs 25-38
Mechanical Plugs 25-38
Memory Metal Plugs 25-39
Summary 25-39
Sleeving 25-39
Sleeve Designs 25-41
Inadequate Heating of Brazed Joints 25-48
Dissolution at Brazed Joints 25-48
Poor Brazed Joints Due to Axial Restraint 25-49
Leakage at Hydraulically Expanded Joints 25-49
Primary Side IGSCC of Deformed Areas 25-49
xx Steam Generator Reference Book, Revision 1
Comparative Sleeving Tests at Ringhals 2 25-49

Summary 25-50
Temperature Reduction 25-50
Laboratory Basis 25-50
Hydrogen Concentration Reduction 25-52
Laboratory Basis 25-52
Secondary Side Water Chemistry Control 25-52
Denting Control 25-52
Secondary Side Chemical Cleaning 25-53
Arrest Denting 25-53
References 25-53
26 Nondestructive Examination 26-1
27 Major Steam Generator Repairs/Replacement 27-1
Appendix A A-1
List of Steam Generator Publications
Appendix B B-1
Steam Generator Designs
Glossary G-i
Glossary of Terms
Index I-1
1
INTRODUCTION
Contributing Author/Editor
S. J. Green, EPRI
1993
1-2 Steam Generator Reference Book, Revision 1
PWR STEAM GENERATORS

Steam generators in pressurized water reactor (PWR) nuclear power
plants are heat exchangers (Figure 1-1). They transfer heat from a primary
coolant system (pressurized water) to a secondary coolant system. Like a
fossil fuel plant, a nuclear power reactor heats water to produce steam.
The steam then drives a turbine which turns an electric generator. The
PWR shown in Figure 1-1 is fueled by slightly enriched uranium in the
FIGURE 1.1
Pressurized Water Reactor (PWR) Power Plant
M Water
Steam
Containment Structure Steam Steam Line
Steam
Generator
Control
Rods
Turbine
Generator
Presue, <•i '" "

Vessel
Cooling
Water
form of uranium oxide pellets held in zirconium alloy tubes in the core.
Water is pumped through the core and heated by the fission process. It is
kept under pressure in the core to prevent boiling. This primary coolant
water completes its cycle by passing through the steam generator,
transferring its heat to the secondary coolant water to make steam. Three
American companies have manufactured PWRs: Westinghouse Electric
Corp.; Combustion Engineering, Inc.; and the Babcock & Wilcox Co. (B&W).
The steam generators in these PWRs, as well as those supplied by Kraftwerk
Union, Framatome, and Mitsubishi are the subject of this document.
(Kraftwerk Union has had a technical interchange agreement with
Combustion Engineering, Inc., and Framatome and Mitsubishi with
Introduction 1-3
Westinghouse Electric Corp.) Horizontal steam generators used by the

Russians, and CANDU steam generators are not explicitly covered here.
However, many of the later chapters contain much pertinent information.
Two types of steam generators are in operation: recirculating and once-
through steam generators. B&W supplied a system which employs a once-
through steam generator while all the other vendors manufactured
recirculating steam generators. Babcock & Wilcox Co., Canada, fabricates
vertical recirculating steam generators, also.
Figure 1-2 shows the significant features of a recirculating-type steam
generator (RSG). A more detailed description of an RSG is given in
Chapter 2. In a variation of this design, a section of the secondary side
serves as a feedwater preheater to achieve greater thermal efficiency. In
this case, a section of tubes at the outlet or cold leg side of the U-bends is
partitioned off from the rest of the secondary side to form a preheat section.
Water is then fed to the preheat section rather than into the downcomer.
Figures 1-3a and 1-3b illustrate the split flow and counter flow in
Westinghouse units. Figure 1-3c illustrates the Combustion Engineering
preheat unit, in particular, the lower part of the bundle.
Figure 1-4 shows a once-through steam generator (OTSG). The primary
coolant flows from top to bottom. The feedwater enters into a feed annulus
between the 9th and 10th tube support plates, where it is mixed with steam
and preheated to saturation. A more detailed description of an OTSG" is
given in Chapter 2.
BACKGROUND
Steam generators in PWRs were designed for a 30-40 year operating
life. However, in the 15-25 years that PWRs have operated commercially,
they have experienced reduced reliability. Steam generator tubing and other
components have prematurely degraded from such mechanisms as
vibration, fretting, high-cycle fatigue, water hammer, stress corrosion
cracking, wastage (wall thinning), pitting, intergranular attack, denting,
and erosion-corrosion. These problems are widespread. By the end of 1992,
approximately 36 of the units in commercial operation had experienced
tube denting, 39 units had showed signs of wastage, 9 had inside-diameter
tube cracking within the tube sheet crevice, 16 had tube cracking in the
U-bend, 40 had experienced outside-diameter stress corrosion cracking and
intergranular attack, and at least 20 experienced water hammer, etc. In
some units, some of these problems are arrested or slowed by equipment,
design, or operating changes. Without close attention to maintenance and
operation, steam generator operators can expect to experience one or more
of the above problems.
FIGURE 1-2
PWR Recirculating Steam Generator
Steam Outlet
Secondary Separator~S
Primary Separators
0 Feedwater Inlet
Downcomer Annulus
Tube Bundle
Tube Supports
it'
Tube Sheet
Primary Inlet Primary Outlet

Introduction 1-5
FIGURE 1-3
Preheat Units
Tube
Sheet
(b) Counter Flow
(c) Axial Flow (cold leg side)

FIGURE 1-4
Once-Through Steam Generator (OTSG)
Primary Inlet Upper

Tubes heet
t" Auxiliary
Feedwater
Inlet
Steam
Outlet
Feedwater
Inlet
Feed
Annulus
Aspirating
Steam
1st Supporl
Plate
Lower
Tubesheet
Introduction 1-7
Degradation of tubing or support structures often requires unscheduled

or extended outages for maintenance. These outages are costly in terms of
inconvenience, equipment repair, replacement power, and personnel
radiation exposure. Continuing problems also increase the probability of
steam generator replacement before the end of the power plant design
lifetime. Two utilities have replaced steam generators in four PWRs after
less than ten years of commercial operation. Steam generators in seven
other units were replaced after only 10 to 13 years of service. A summary
of steam generator replacements through 1993 is presented in Table 1-1.
These replacements are expensive and require extensive outages. Estimates
of steam generator replacement costs approach $100 million, which does
not include replacement power costs.
TABLE 1-1
Steam Generator Replacements (1.2)
Year of Years of
Completion Unit Cause Service
1980 Surry 2 Denting 7
1981 Surry 1 Denting 8
1982 Turkey Point 3 Denting 9
1983 Turkey Point 4 Denting 10
1983 Obrigheim SCC 14
1984 Point Beach 1 SCC 13
1984 H. B. Robinson Wastage
1989 D. C. Cook 2 SCC 11
1989 Indian Point 3 Pitting 12
1989 Ringhals 2 SCC 14
1990 Dampierre 1 SCC 10
1991 Palisades Wastage 19
1993 Millstone 2 SCC 17
1993 North Anna 1 SCC 15
1993 Beznau 1 SCC 24
1993 Doel 3 SCC 11
1-8 Steam Generator Reference Book, Revision I
By late 1976, many operating steam generators were known to have

experienced wastage, denting, stress corrosion cracking, apparent fatigue
cracking, water hammer, and carryover. It was suspected that tube corrosion
was due, in part, to condenser inleakage. However, there was very little
quantitative data available to define the magnitude of the various damage
forms, and without laboratory simulation of the damage, there was little
understanding of the causes. For instance, one vendor had postulated that
prior phosphate operation was required for denting to occur (1.). Most
steam generators in the United States used phosphate chemistry control
originally, but have converted to all-volatile treatment (AVT). Almost all
were outfitted with alloy 600 tubes and carbon steel drilled hole tube
supports. The steam generator designs evolved from smaller units; most
were scaled-up versions of units from the 1950s and 1960s. Thermal-
hydraulic analyses of these early designs were limited. Plant designers
focused on increased performance and cost reduction per unit of capacity
($/kW). Without quantitative data, utilities had little incentive to maintain
tight condensers, isolate air leaks, or shut the plant down in response to
high levels of contaminants in the steam generators. Nor was there a strong
incentive to install condensate polishing and demineralizers, replace feed
train materials, reroute/add piping to allow removal of contaminants, or
to install improved instrumentation for measuring lower levels of
contaminants and/or corrosion rates. Chemical cleaning and neutralization
were practices common to the fossil fuel industry, but these concepts were
never demonstrated for nuclear plant application. The only methods for
detecting damage in the steam generator tubes were single-frequency eddy
current testing and radiation monitors that were used to detect tube leaks.
By 1976, the utility industry recognized that steam generators required
special attention. Corrective actions were needed. The costs to implement
these corrective actions were expected to be very high. In early 1977, the
industry began a major effort to develop methods for improving this
situation. They organized the Steam Generator Owners Group and funded
a 5-year program managed by the Steam Generator Project Office at EPRI
at a cost of about $40 million. Twenty-eight utilities supported the Owners
Group, including utilities from three European countries and Japan. The
24 U. S. utilities represented 54 of the 101 PWRs that were planned to be in
operation in the United States by 1986.
The program involved a complex interaction of many systems and
disciplines, such as mechanical, thermal, and hydraulic design of the steam
generators, water chemistry, control of this chemistry, the design and
materials for construction of secondary plant components and systems, plant
operation, and steam generator inspection techniques. The objective of this
program was to provide improved technology alternatives to the utilities,
Introduction 1-9
vendors, and architect/engineers so that each utility could select the most
appropriate options, as well as develop a specific plan for improving steam
generator performance.
This five-year program was completed successfully However, new
challenges emerged while this program was under way, and a new group
called Steam Generator Owners Group II was formed in 1983 to address
these new problems over the period 1983-1986. The approach and objectives
of the new program were similar to those of the first Owners Group. Its
budget was about.$29 million. Thirty-seven utilities, including utilities from
six European countries and Japan supported this activity.
This program was also successfully completed. Again, new steam
generator challenges continued to arise. Therefore, a new EPRI-funded
effort, the Steam Generator Reliability Project (SGRP), was formed to operate
at a funding level of $30 million over the period 1987-1992. Five European
countries, one Canadian utility, and one Japanese utility are participating
in this program under technical exchange agreements.
REFERENCES
1.1 "Steam Generator Update-1976." Steam Generator Symposium,
Westinghouse Electric Corp., Pittsburgh, PA, July 1976.
1.2 "Steam Generator Progress Report, Revision 10." EPRI Steam
Generator Management Project. November 1994.
2
STEAM GENERATOR
PERFORMANCE HISTORY
Contributing Authors/Editors
S. J. Green, EPRI
J. A. Mundis, Southern CaliforniaEdison Co.
T. Oldberg, EPRI
C. S. Welty, EPRI
J. P.N. Paine,EPRI
INTRODUCTION
This section provides a description of the pressurized water reactor
(PWR) for those unfamiliar with this type of power plant. It also discusses
the design features that impact steam generator operating performance and
reliability.
The technology that evolved into the modern PWR steam generator.
and its related systems was greatly influenced by the fossil-fired drum
boilers which were first widely used to generate electricity commercially
50 years ago, and by the power plants used in the U.S. navy nuclear power
program.
PRESSURIZED WATER REACTOR (PWR) POWER

PLANT DESCRIPTION
In general, PWR power plants generate heat by the fission of slightly
enriched uranium in a reactor core (Chapter 1, Figure 1-1). Pressurized
water is used to transfer this heat to steam generators, where the heat is
used to produce steam to drive the turbine generator. The steam generators
are shell and tube heat exchangers, with the reactor coolant (the primary
side) inside the tubes and the feedwater/steam on the shell side (the
secondary side). The PWR steam generators must be designed to avoid
leakage of the radioactive water, which is used for core cooling, into the
non-radioactive water in the steam system.
The PWR power plant has four water systems that influence steam
generator performance: primary system, secondary system or steam cycle,
condenser cooling water system, and makeup water system.
The primary system is a closed cycle cooling circuit that removes heat
generated by the fission process in the reactor core and transfers it to the
secondary system. The principal components are the reactor vessel and
core, the primary side of the steam generator, the pressurizer, and the reactor
coolant pumps. The primary coolant is pure water, with additions of lithium
and boric acid, which is maintained at a pressure of about 2,250 psia to
minimize boiling in the core. It is pumped by the reactor coolant pumps
through the reactor core, through the steam generator tubes, and back to
the reactor coolant pumps.
0
The secondary system is another closed cycle cooling circuit where heat
is removed from the primary coolant to generate steam on the shell side of
the steam generators at design pressures of 770-1,050 psi. The steam flows
through the high pressure turbine, the moisture separator/reheater (MSR),
the low pressure turbine, and then to the condenser. The condensate is
Steam Generator Performance History 2.3
then pumped forward through a series of components such as the

condensate polishing system (if so equipped), the low pressure feedwater
heaters, the deaerating feedwater heater and storage tank (if so equipped),
the high pressure feedwater heaters, and then to the steam generator to
repeat the cycle. The low pressure heater drains are normally routed to the
condenser where they return to the cycle with the initial condensed steam
from the low pressure turbine. The high pressure heater drains and MSR
drains, which consist of about 30% of the total flow to the steam generators,
are generally pumped forward to rejoin the cycle prior to the last, high-
pressure stages of feedwater heating.
The condenser cooling water system is an open or semi-open cooling
cycle which functions as the heat sink for the steam cycle. As will be
discussed in more detail below, the water in this system can introduce a
significant quantity of harmful impurities to the secondary system through
condenser leakage.
The makeup water system is used to purify site water to provide high
purity water to the primary and secondary systems to compensate for losses
from these systems. Depending on its design and operation, the makeup
water system can also be a significant source of impurities to both the
primary and secondary systems.
With the exception of the steam generator, the components of the PWR
steam cycle are similar to those of other steam-generating electric power
plants. However, the steam generator's unique design features, materials
of construction, and operating parameters (i.e., heat flux, temperature, and
pressure) combine to make it extremely sensitive to the design and operation
of the remaining steam cycle components.
Steam Generators
Commercial light-water PWR power plants have one to four steam
generators depending on the supplier of the nuclear steam supply system
(NSSS) and the size of the power plant. The steam generator is a shell and
tube heat exchanger that transfers heat from the primary system to the steam
cycle, and, as noted above, functions as the boundary between the
radioactive primary coolant and the non-radioactive steam cycle water.
One unique feature of the PWR steam generator, that sets it apart from
conventional fossil plant steam generators, is that boiling takes place on
the shell side of the heat exchanger rather than inside tubes. Because the
primary coolant is at high pressure (about 2,250 psia) it is desirable for the
primary coolant to be on the inside of the tubes. Then, the shell side can be
at relatively lower pressure, which minimizes the required shell wall
thickness. By keeping the boiling on the shell side, high primary fluid flow
velocity is maintained which enhances heat transfer. This design, however,
results in numerous crevices on the boiling or secondary side of PWR steam

generators, which makes the steam generator very susceptible to the
corrosion problems described in this book and different from fossil-fueled
drum boilers.
A PWR steam generator differs significantly from a conventional
fossil-fired recirculating drum boiler in another respect; it does not have a
"mud" drum where corrosion products or sludge can deposit from the heat
flux region, and from which corrosion products can be readily removed by
blowdown. Rather, in a standard PWR recirculating steam generator,
corrosion products build up on the tubesheet, where removal by blowdown
is ineffective and where high heat flux exists. As a result, PWR steam
generators are susceptible to problems caused by a buildup of sludge piles
and by dryout and impurity concentrations in these sludge piles.
There are two types of PWR steam generators in use today, recirculating
steam generators (RSGs) and once-through steam generators (OTSGs).
These two types of steam generators and the principal problems associated
with each are described below.
Recirculating Steam Generators. In a recirculating steam generator, water

is fed into the downcomer where it is mixed with two to four volumes of
recirculating water, which drains from the moisture separators. The
downcomer water flows to the bottom of the steam generator, across the
tubesheet, and then upward through the tube bundle where steam is
generated. The thermodynamic quality of the water-steam mixture at the
top of the bundle as it enters the steam separators is about 20% to 33%,
corresponding to a circulation ratio in the range of 5:1 to 3:1. The controlled
pressure on the secondary side is between 770 to 1,050 psia. The primary
coolant flows through U-tubes at a pressure of about 2,250 psia. It enters
the steam generator at about 590 to 6257F (310'-330'C) and leaves at about
550°F (288°C). At the primary inlet, there is about a 65 to 70 F0 (36'-39C°)
temperature difference across the tube wall, corresponding to a heat flux of
between 100,000 to 140,000 Btu/hr-ft 2 . At the primary outlet or cold side,
the temperature difference between the primary and secondary sides is
about 20 to 25F° (11 to 14WC), corresponding to a heat flux of about 30,000
Btu/hr-ft 2 .
A knowledge of the materials, water chemistry, and tube support
arrangements used in RSGs is required to understand the problems that
have occurred.
Material. The heat transfer tubes in domestic units were made of Ni-
Cr-Fe alloy 600. Early RSGs were tubed with mill-annealed material,
while in subsequent RSGs thermally treated materials were used to
improve resistance to stress corrosion cracking. All OTSGs have
Steam Generator Performance History 2-5
sensitized alloy 600 tubes. Kraftwerk Union (KWU) has used alloy 800
(modified) tubes while, most recently, utilities have chosen alloy 690
tube material for their RSGs. The tube support plates or grids in early
RSGs were made of carbon steel while current RSGs use 405/409
stainless steels to improve resistance to denting. KWU uses 347 stainless
steel grids. The tubesheet and shell are made of a low alloy carbon
steel.
Water Treatment. Early RSGs operated with phosphate water
treatment, which is the conventional water treatment method used for
fossil-fired boilers with similar steam pressures. When corrosion
problems emerged, domestic utilities switched to all volatile water
treatment (AVT). Most recently, morpholine instead of or in addition
to ammonia has been applied for pH control. Hydrazine is added to
control oxygen.
Tube Support Arrangements. Tube support structures in a majority
of the early RSGs had a drilled round hole configuration. Due to
corrosion concerns, the tube support plate hole geometry was either
changed to a non-round broached hole with lands, or the support plate
took the shape of an eggcrate structure.
Figure 2-1 indicates the main areas where significant degradation has
been experienced. A brief description of the main types of degradation
and the plant and steam generator design features associated with each
of them is discussed below.
Thinning/Wastage. Wall thinning or wastage has occurred in the

sludge pile region above the tubesheet of RSGs operating on coordinated
phosphate water treatment, and also within the lower tube support
plate of some RSGs on AVT. Wastage was caused by aggressive
conditions developing as a result of chemical concentration processes
in dryout regions in the sludge pile. The main features leading to this
problem are:
* secondary system materials and design which allow the
introduction of corrosion products, and inleakage of impurities,
into the steam generator, generating sludge piles, and
* a design that allows these corrosion products to settle in a high
heat flux area where they cannot be effectiyely removed by
blowdown.
Domestically, this situation was corrected mainly by switching to
AVT. In Germany, where a different phosphate treatment regime was
practiced, PWR operators also focused on reducing impurities and
sludge pile removal.
FIGURE 2-1
Sketch of Recirculating Steam Generator With Indicated Problems
Feedwater (ingress
of impurities)
Tube Support Crackinof

Plate (top view) Tube U-Tend
Flow lotsDeformation of
of
FwSlots,
Hourgiassing ff [" :.
Tube U-Bend
Section of
Tube Suppr Denting
P late (enlarged'1) 0 . . " . : .Local Local
0 0Concentration of
Tube Support Impurities/Wastage
Hole __p'riToB awdown
Flow Hole To Blowdown
-- - VemrllOvIl ol
impurities)
I • -- Tube
Intergranular
•.:..,, -A. Attack
Pitting P
Primary Inlet Primary Outlet
Pitting. Pitting has occurred in the tubesheet sludge pile regions of

several PWRs. This form of degradation appears to be an acid attack
associated with the introduction of chlorides, sulfur anions, and copper
oxides into the steam generator. The design features involved are those
leading to formation of sludge piles as described above, combined with
inleakage of impurities due to condenser leaks, exposure to air during
layup (due to lack of appropriate layup procedures), and use of copper
in secondary system heat exchangers. The main corrective approaches
to pitting are:
* eliminate inleakage of impurities into the condenser,
* eliminate copper from the secondary system,
* minimize the introduction of corrosion products by a variety of
methods,
* install and utilize layup control methods,
* perform sludge lancing and/or chemical cleaning to remove the
sludge pile, and
* perform crevice flushing to clean out impurities.
Denting. Denting is a process where rapid corrosion of carbon steel

tube support plates or egg crate structures occurs at the tube-to-tube
support annulus. It is the result of boiling in tube-to-support crevices
causing concentration of impurities (especially chlorides introduced
from condenser leakage at seawater locations) which, in turn, causes
rapid corrosion of the carbon steel tube support structures. The
corrosion products fill the annulus and squeeze the tube, causing dents.
This process can ultimately lead to cracking from the primary or
secondary side at tube support dents, deformation and cracking at
U-bends, cracking of tube support plate ligaments, and tube support
plate flow slot distortion. The design features contributing to denting
include:
* non-leak-tight condenser designs leading to inleakage of impurities,
* plant design features leading to oxygen or copper oxide intrusion
into the steam generators,
* use of copper alloys for secondary system heat exchangers,
* tube support plate hole designs (drilled holes) in which flow
blockage and impurity concentration are likely,
* high transport into the steam generator of particulates from
secondary system corrosion, and
• use of carbon steel tube support material that is subject to rapid
corrosion.
The following approaches are used to prevent or minimize denting:

* eliminate inleakage of impurities to the condenser,
* reduce the introduction of oxygen and metal oxides into the steam
generator to as close to zero as practical,
" use corrosion-resistant tube support materials,
* eliminate copper materials in the secondary system, and
* use tube support geometries which eliminate or minimize dryout.
Intergranular Attack and Stress Corrosion Cracking. Intergranular

attack and secondary-side-initiated stress corrosion cracking have
occurred in tubesheet crevices, at tube support intersections, in sludge
piles, and recently in the free span between the sludge pile and supports.
The attack in the tubesheet is mostly caused by a concentration of caustic
in the crevices. The attack in the tube supports and in the sludge piles
appears to be caused primarily by caustic. However, acid sulfate and
lead are implicated in several plants. The causes of this type of corrosion
appears to be the use of a tube-to-tubesheet joint and tube support plate
design with a built-in crevice along with the introduction of impurities
into the secondary system. However, this form of corrosion is on-going
and is not yet under control. The main approaches identified to date
are as follows:
• eliminate tubesheet crevices by expanding tubes for the full height
of the tubesheet,
* eliminate inleakage of impurities,
'perform crevice flushing or chemical cleaning operations during
shutdown periods, and
* add boric acid to counteract the caustic corrosion.
Primary Side Cracking. Primary side cracking first occurred in the

roll transition region in the tubesheet, apparently as a result of residual
stresses, and at U-bends and tube support plate intersections due to
denting. A combination of susceptible microstructure and high residual
stresses generated during the fabrication process and/or high stresses
and strains caused by service-induced deformation are apparently
required for cracking to occur. The following approaches have been
identified to minimize the effects:
" perform in situ stress relief or peening of areas with high residual
stresses,
" sleeving over cracked areas,
* use thermally treated material with increased resistance for new
installations,
• use materials such as alloy 800 or alloy 690 which are not susceptible
to primary side stress corrosion cracking, and
" minimize denting using the measures described earlier.
Clad Separation. Clad separation was first discovered in connection

with a primary-to-secondary leak through the tube-to-tubesheet weld
in an operating U-tube steam generator. Leaks were caused by a
separation of the tubesheet cladding. All 11 units that were affected
plugged all of the row one tubes.
The units in question, including some which had not been in
operation, were repaired. This included removal of the channel divider
plate and re-cladding the tube lane by a weld deposit process. Tubes
on either side of the divider plate were plugged. To prevent the
occurrence of cladding separation on future steam generators, the
cladding application process was changed to a plasma arc-weld overlay
process. Since then, none of the units using this technique has
experienced tubesheet cladding separation.
Tube Fretting and Wear. Several units have experienced fretting and
wear of tubes at the antivibration bars as a result of tube vibration caused
by high cross flows in the U-bend region. The antivibration bars are
installed in the upper U-bend region to reduce the unsupported length
of the tubes and to reduce vibration. Tolerance clearance between
U-bends and their respective antivibration bars leads to a lack of support
allowing the tube(s) to vibrate. In order to minimize this situation,
new antivibration bars with a broader region for tube contact are
installed and the clearance between the bars and the tubes is reduced.
Significant wear has also been observed in split-flow and counter-
flow preheat steam generators (see Chapter 1). The wear in the split-
flow units (Figure 1-3a) was caused by the high inlet cross velocity and
a relatively small impingement baffle. The situation was corrected by
reducing the inlet flow and by increasing the size of the impingement
baffle to reduce the local transverse velocities. Similarly, in the counter-
flow units (Figure 1-3b), the wear was caused by the vibration induced
by the high cross flow velocities and relatively large clearance between
the tubes and the holes in the tube support plates. This was corrected
by expanding the tubes at the tube support plate levels and thus
reducing the tube gap clearance.
Tube wear has been observed at the bat wing supports of several
axial preheat units (Figure 1-3c) in the periphery of the central cavity
region. The wear was a result of flow-induced vibrations resulting from
long unsupported spaces between the first tube contact point and the
center of the bundle. The affected tubes were plugged. No additional

corrective actions were taken since it was judged that the damage was
limited to one region of the steam generator.
High Cycle Fatigue. In three cases, double-ended tube ruptures and
large primary-to-secondary leaks caused by high cycle fatigue failures
have occurred. These failures at the top tube support plate level resulted
from flow-induced vibration caused by cross flow in the U-bend region.
The failures are probably due to a combination of factors such as:
* denting at the top support plate which imposes a high mean stress,
* denting or packing of the tube-to-tube-support crevice which
reduces the damping,
* no antivibration support or a large clearance between the tube and
the antivibration support which permits larger amplitude vibration
and higher cyclic stress, and
* a possible corrosive environment at the tube-to-tube-support-plate
crevice which could lead to a significant decrease in the tube's
fatigue strength (corrosion fatigue).
Corrective action would include soaking or chemical cleaning to
reduce the corrosion environment, redesign of the antivibration bars to
minimize the tube-to-bar clearance, and plugging of the susceptible
tubes.
This degradation mechanism has also been encountered in Candu
steam generators.
Steam Pressure Decrease. A number of units have observed a steam
pressure decrease. If the decrease is large enough, it can no longer be
compensated by opening the main steam throttle valve. Once this valve
is fully open, any further pressure decrease leads to a reduction in power
output. It is judged that the pressure decrease is caused by a buildup
of corrosive products and scales on the tubes (fouling), which imposes
an additional thermal resistance and results in a decrease in the overall
heat transfer coefficient, which, in turn, decreases the performance of
the steam generator. Corrective actions involve chemical cleaning to
reduce the deposits as well as close chemistry control (morpholine
addition or high pH) to reduce the sludge input to the steam generator.
Shell Cracking. Several units have observed pitting and cracking of
the steam generator shell at the upper shell to transition cone weld.
The cracks initiate at pits in the weld metal at the heat-affected zone
and propagate by environmentally assisted cracking. The pits probably
are formed as a result of poor chemistry control during layup (presence
of oxygen, etc.). Corrective action includes grinding out the cracks and,
Steam Generator Performance History 2-1 1
in extreme cases, deposition of new weld metal. Maintenance of good

chemistry, particularly during layup, to avoid oxidizing and pitting
conditions would reduce the risk of cracks.
Once-Through Steam Generators (OTSGs). In the OTSG, water enters a

feed annulus above the 9th tube support plate level. There it is mixed with
steam aspirated from the tube bundle area and preheated to saturation.
The saturated water flows down the annulus, across the lower tubesheet,
and upward into the tube bundle where it becomes steam. It reaches 100%
quality (on the average) in the 9th and 10th support plate region and achieves
about 40 to 60'F (22 to 33QC) of superheat at about 925 psia at the top of the
unit. The superheated steam flows radially outward and then down the
annulus to the steam outlet connection. The primary coolant flow is from
top to bottom. It enters at about 600 to 620'F (316 to 327°C) and leaves at
about 555 to 560'F (291 to 293°C). The temperature difference between the
primary and secondary sides at the bottom of the steam generator is similar
to that on the cold leg of a recirculating steam generator.
OTSG materials are similar to those used in RSGs. Tubes are Ni-Cr-Fe
alloy 600 in a mill annealed plus sensitized condition (due to stress relief of
the OTSG). Tube support plates are made of carbon steel with a broached
holes configuration and three lands supporting the tubes. Tubesheets are
low alloy carbon steel. OTSGs started out on AVT water chemistry as
opposed to RSGs which began operating on sodium phosphate. The OTSG
supplier, Babcock & Wilcox, has always recommended secondary systems
that include deaerators, condensate polishers, and feedwater heaters
without copper alloys.
Figure 2-2 indicates areas where OTSGs have experienced degradation.
A brief description of the main types of degradation and the plant and
steam generator design features associated with each are given below.
Upper Span Circumferential Cracking. Cracks have been observed
at the top tube support plate and at the bottom of the upper tubesheet
in tubes along the inspection lane. These cracks seem to be caused by a
corrosion fatigue mechanism. The underlying causes appear to be high
steam flow velocities along the inspection lane and unvaporized water
with concentrated impurities reaching this region by streaming up the
inspection lane. The main corrective measure used to date has been to
plug affected tubes, which have been limited in number. Devices for
blocking the flow of water up the inspection lane have also been
designed for installation at selected tube support plates. To date these
have not yet been installed.
FIGURE 2-2
Sketch of Once-Through Steam Generator With Indicated Problem
Areas
Upper
Tubesheet
Crevice Cracks rimary Inlet Upper
Tubesheet
Upper Span
Lane Region
Circumferential
Cracks Auxiliary
Feedwater
Inlet
15th Support
Plate Wear
14th Tube
Support Plate
Periphery Steam
Tube Erosion Outlet
Feedwater
Inlet
Feed
Annulus
Aspirating
Dings at 9th Steam
Support Plate
Lower
Tubesheet
Fretting and Wear at Upper Tube Support Plate. Fretting and wear
of tubes along the inspection lane has been observed at the upper tube
support plate. The basic cause of this problem appears to be high-
steam cross flow velocities in the inspection lane. The main corrective
measure used to date has been to plug affected tubes, which have been
limited in number.
Erosion-Corrosion. Large numbers of erosion-corrosion type grooves
or pits have been observed in one plant. They occur near the top of the
steam generator around the periphery. The exact cause of this
phenomenon has not been determined. However, it appears to be
related to unvaporized water with concentrated impurities reaching
this region through gaps at the tube bundle periphery between the tube
support plates and the wrapper sheet, and also to the presence of large
amounts of solid corrosion products in the generator. Firm, corrective
measures have not been identified, except to plug the affected tubes.
However, some consideration has been given to chemically cleaning
the steam generator to remove solid corrosion products present on the
tube support plates.
Cracking in Upper Tubesheet Crevices. Cracks, apparently caused
by sulfur attack, have occurred in upper tubesheet crevices at two plants.
The source of the sulfur is probably leakage of either condensate polisher
regeneration chemicals or resin fines. The tubing is susceptible to sulfur
attack because of its sensitized condition. To prevent the occurrence of
cracking, the water chemistry should be controlled as well as possible
both during operation and layup, with the minimum possible
introduction of resin fines or other impurities. For new installations,
use of thermally treated tubing should provide resistance to sulfur
attack.
Primary Side Sulfur Cracking. One plant experienced extensive
cracking of tubes from the primary side during a single incident. The
cracks were located mainly in the upper tubesheet region, most
frequently in the roll transition area. The causes of the cracks appear to
have been as follows:
* the sensitized condition of the alloy 600 tubing which makes it
susceptible to sulfur type attack,
• sulfur contaminants in the primary system, and
* series of events including water levels at the upper tubesheet
elevation and oxygen inleakage which resulted in the sulfur
contaminants reaching aggressive concentrations and oxidation
states in local areas.
The main preventive measure for this problem is to prevent sulfur

contamination of the primary system. In new installations, use of
thermally treated tubing should provide resistance to sulfur attack.
Pressure Drop Increases. In most of the once-through units, there
has been an increase in pressure drop across the lower part of the bundle
due to solids deposition and plugging of the tube support hole crevices.
This increase causes the liquid level in the annulus to rise. The pressure
drop increases can lead to flooding of the feedwater inlet, which, in
many cases, leads to a power decrease to avoid flooding. The corrective
actions have been to switch to morpholine chemistry to reduce the iron
sludge input and to chemically clean to reduce the sludge inventory
within the unit.
Steam Generator Fabrication

Fabrication is subdivided into two major parallel operations: (1) lower
shell assembly and (2) upper shell assembly. The first shipment includes
the tubesheet, the channel head, and the lower shell barrel plate along with
all material code inspections and records. After receipt, the tubesheet is
machined and cladded on the primary side, machined again and then
drilled. The cladding is inspected for bond and thickness prior to and after
drilling. In parallel, the lower shell barrels and the transition cone are formed
from plate ( some may be forged). The shell barrels and cone are then welded
together. These barrels are next attached to the tubesheet by welding and
the lower shell assembly is ready for internals installation. Handholes can
be constructed by adding weld build-up to the shell and then machining to
final dimensions. All pressure boundary welds of low alloy steel are stress
relieved and inspected by several methods as required by code (ultrasonic,
radiography, magnetic particle, or dye penetrant inspection).
The wrapper is formed from thin plate material, inserted into the lower
shell assembly, and attached to its supports. Centering and alignment of
the wrapper with the tubesheet drilling pattern is achieved using jacking
screws which are located circumferentially at each tube support elevation.
The wrapper is supported by attachments to the lower shell which control
vertical and rotational motion. Next, the tube supports are installed and
aligned with the tubesheet drilling pattern. Alignment of the wrapper and
tube supports is now achieved using lasers which are mounted on the
tubesheet. Stayrods, spacer pipes, and peripheral support blocks are used
to place the tube supports at the proper elevations. Tube support plate
manufacture involves drilling, broaching, buffing, and several dimensional
inspections. Finally, an intensive inspection for the assembled structure is
performed to confirm alignment. After tubesheet hole cleaning the tubes
are inserted. As each tube reaches the primary face of the tubesheet it is
lightly expanded and then welded to the primary side clad. The welds are
inspected using a helium leak test and a dye penetrant test.
After completion of the weld inspections, the tube is expanded the full
depth of the tubesheet hole. This expansion process is carefully controlled
by length adjustments, the initial tubesheet thickness machining, and
dimensional mapping. The remaining depth of the tubesheet crevice at the
secondary side of the tubesheet is minimized. Upon completion of this
process, the U-bend antivibration bar supports are installed. The U-bend
clearances between antivibration bars and tubes are controlled by attaching
the bars to a retaining bar structure. The inspection of the assembled U-bend
completes the lower assembly tube bundle structuring operation.
The channel head is supplied in the cast, forged, or formed plate form,
and then machined and cladded. Machining includes manway openings
and nozzle safe end weld preps. The divider plate is installed in the head
and the head is attached to the tubesheet, which is part of the lower shell
assembly. The channel-head-to-tubesheet weld is then stress relieved and
inspected. The divider plate is welded to a stub runner which is part of the
tubesheet assembly The stub runner helps with divider plate alignment
and provides clearance from the row one tube ends. This operation
completes the lower shell assembly manufacturing phase.
In parallel with the lower shell construction, the upper shell is formed
from plate and welded to the upper head. The steam nozzle forging and
the flow limiting venturis, the two secondary manway forgings, the
feedwater nozzle forging, and the feedring support plugs are welded into
the upper shell. The primary and secondary moisture separator assemblies
and the feedring are manufactured as subassemblies. The upper portion of
the moisture separator assembly is attached by welding to the upper head.
Next, the feedwater distribution ring is welded to the feed nozzle and
attached to its supports. Then, the remaining portion of the primary
moisture separator assembly is installed.
The final step in the manufacture of the steam generator is to join the
upper and lower shell assemblies. After the pressure boundary weld stress
relief and inspection have been completed, the internal wrapper assembly
is joined to the primary separator assembly. Upon completion of this
process, the unit is cleaned, sealed, air and moisture extracted, and then
filled with nitrogen gas. The steam generator is then ready for shipment.
Balance-of-Plant Systems
Balance of plant systems are shown in Figure 2-3 and 2-4. Figure 2-3 is
typical of early PWR designs while Figure 2-4 is of the currently
recommended type. The main parts of the balance of plant systems and
their impact on steam generator degradation are described in the sections
which follow.
FIGURE 2.3
Typical Early Design of PWR Balance-of-Plant Systems
FMoisture
Reheater- Separator
" 2 3 Seals
'[
[• II• ,Glandsa
Reg.
Generator
S Drain MoistureCod
Separator I nReheaterd
LL.P. Turine - eirto
GnrtrMain.Steamn Tanks Drain Tank ~v T[u"
' b•'"
StamAi • Condenser -
Ejeto Gland rAir
FSea J
ILI
Cond ond. Cond,
H.P. Heaters Drain

.Heater
DasReun Pum Pondnisher p.-
Recirculation Line
Condenser Recirculation
Pum 0
FIGURE 2.4
Recommended Design of PWR Balance-of-Plant Systems
High
I Pressure - --.
Steam Mo ture Separator/ TH'"
• • ,
Reheater
... j 1ITurb
' - . - "P. LR
Steam Turb
Generator (Reheate --- Mosture
i Auxiliary T Separator
1ILoop Short ;< Steam •. Tý Drain Tank
II • •
- --- JCondenser Air Remova
Chiller System 0"
.,
I 'Deaerator ' " " I"ll
J I I I --
-' -tem forBubbler Devce I
•1 IlI Storage Tank Jr--
,1- - 1.. .• J •
r(0
l
r- -- I condenýsate
tPump Sucionr
i3 eJa" I-r_ Header SideStam

-- No. 3 HeaterIICondensate
" -
Air Ejector Condenser Dmineralzer
eryI beo
I W
No.2Conden
Fed~te I Legen
N.pump
5 Booster Extraction Steam
Pm -.... Drains and Blowdown
TramnCondensate Feedwater
System -- Cleanup Recycle Water Linend
Cleanup Reycle Line (Long Loop) - CDirection ot Flow ,ri

Condensers. Condensers are large shell and tube heat exchangers which
condense steam exiting from the low pressure turbine. The removed heat
is transferred to the circulating cooling water on the inside of the tubes.
The steam (or shell) side of condensers typically has a vacuum
corresponding to an absolute pressure of I or 2 inches of mercury, while
the cooling water (or tube) side is at a positive pressure.
Leakage of circulating cooling water into the condenser has been the
major source of impurities introduced into steam generators, and, thus,
has been a major factor in most steam generator corrosion. A variety of
condenser problems has caused inleakage, including
* use of tube materials, such as admiralty brass, which are susceptible to
erosion-corrosion;
* use of rolled tube joints without seal welds and without double
tubesheet protection;
" poor design leading to tube failures caused by steam impingement or
by other parts loosened by steam impingement; and
* improperly supported tubes such that vibration failures occur.
As a result of the numerous problems experienced with condensers,
and because maintaining leak-tight condensers has been found to be
essential to steam generator integrity, design and maintenance
recommendations have been developed to provide the required integrity.
These include use of titanium tubes for seawater- or brackish-water-cooled
condensers, use of titanium or stainless steel tubes for freshwater-cooled
condensers, performing periodic eddy current tests of tubes, and use of
sensitive leak location equipment.
Condensers typically have been designed to deaerate the condensate
during full power operation, but have not been designed to effectively
deaerate at low or zero power. This has resulted in oxygen ingress during
low-power operation and power startup. Design changes have been
identified which can correct this deficiency, including the use of an air-
removal chiller system.
The condenser-to-turbine expansion joint has been a frequent source of
air inleakage. This air inleakage has been a significant contributor to oxygen
ingress into steam generators and resulting corrosion. Improved expansion
joint designs, such as water-sealed rubber expansion joints, are available
and are recommended.
Condensate System. The condensate system takes water from the

condenser hotwell and pumps it through a series of components to the
suction of the feed pump. These components include plant exhaust and
steam air ejector condensers, condensate polishers (if installed), low-pressure
feedwater heaters, and deaerator (if installed). Some main features of the
condensate system are known to contribute to steam generator corrosion.

(1) The parts of the system on the suction side of the condensate pumps are
under vacuum and are a frequent source of air inleakage. (2) Copper alloy
feedwater heater tubes have been a source of copper and copper oxide which
have aggravated corrosion attack in the steam generators. (3) The lack of
condensate polishers has resulted in corrosion products and ionic impurities
not being effectively removed from the condensate. (4) The lack of
deaerators in the system has resulted in less than optimum deaeration of
the condensate. These problems can be addressed by use of features such
as bellows or diaphragmn-sealed valves, stainless steel tubes in feedwater
heaters, and installation of condensate polishers and deaerators.
Condensate Polishing System. The condensate polishing system installed

downstream of the hotwell but upstream of the first stage of low-pressure
feedwater heaters, processes approximately 70% of the final feedwater flow
(but 100% of condensate flow), and functions both as an ion exchanger and
as a corrosion product filter. Plants with well designed and operated
polisher systems generally operate with higher purity feedwater, based on
the currently established control parameters, than plants without such
systems. However, plants with poorly designed and/or operated systems
can, in fact, degrade water purity, based on ionic control parameters, by
release of regenerant chemicals from deep-bed system, and by the release
and subsequent thermal breakdown of resin from both deep-bed and
powdered-resin systems. These ionic impurities, primarily sodium, sulfates,
and chlorides, are implicated or known to be involved in several forms of
steam generator corrosion damage.
Feed System. The feed system serves to pump condensate to a high

pressure, pass it through high pressure feedwater heaters, and introduce it
into the steam generators. The main feature of feed systems which has
contributed to steam generator corrosion is the use of copper alloys for
feedwater heater tubing. Copper from this and other sources (low pressure
heaters, moisture separator/ reheaters, and condensers) has aggravated
denting and pitting problems. Replacement of copper alloy. tubing with
stainless steel tubing and then chemically removing copper buildup in the
feed system is recommended.
Drain System. Condensate and feedwater is pumped through low-pressure

and high-pressure feedwater tube sides in order to heat the condensate/
feedwater and improve cycle thermal efficiency. The heater shell sides are
heated by condensing steam taken from various turbine extraction stages.
The condensed steam is handled by the drain system. A typical balance-of-
plant-system arrangement is shown on Figure 2-3. The drain system consists

of the high-pressure and low-pressure feedwater heater drains, MSR drains,
gland seal and air ejector condenser drains, drain tanks, and heater drain
pumps. Typically, high-pressure drains are pumped forward into the feed
system, while low-pressure drains are cascaded back to the condenser; in
some cases, low-pressure drains are also pumped forward.
The main contributing factor to steam generator reliability is that high-
pressure drains pumped forward introduce significant amounts of corrosion
products into the steam generators, including copper and copper oxides if
the feedwater heaters or MSRs use copper alloy tubes. In addition, if low-
pressure drains are pumped forward, they can introduce oxygen as well as
corrosion products. Recommended solutions to these probleins are to direct
drains to the condenser and then through condensate polishers if available;
as a result, the drains become deaerated and filtered and replace susceptible
components with corrosion-resistant materials. In addition, copper alloys
should be eliminated from MSRs, especially if their drains are pumped
forward or if condensate polishers are not installed.
Blowdown Processing System. A typical blowdown system for an early

PWR is shown on Figure 2-5. It is used to withdraw water from the top of
the steam generator tubesheet at a rate equal to about 0.2 to 0.5% of feed
flow from the steam generators and discharge this water, after
depressurization and cooling by flushing, to industrial or radioactive waste.
Because neither the heat nor the water is recovered, there is an incentive
with this system to keep flow rates low. As a result, the blowdown is
relatively ineffective at controlling impurity levels in the steam generator.
Recommended systems now in use at many operating PWRs (Figure 2-6)
handle 1% of the feedwater flow and recover the heat and water so that
maintaining the full design flow is practical.
Makeup Water System. The makeup water system provides high purity
water to makeup for blowdown and other secondary cycle losses during
normal operation. It also provides water to place steam generators and
other secondary plant components in wet layup during shutdown. In
addition, this system provides water for flushing and cleanup of secondary
systems following shutdown. The system consists of a series of components
and subsystems which convert the site water source, which spans the
impurity range from salt water and treated sewage to relatively pure fresh
water, to the extremely high purity water required for reliable steam
generator operation. A poorly operated or under-designed system can be
a significant source of impurities for the secondary cycle. The significance
of normal and transient contaminant levels in makeup water depends on
FIGURE 2-5
Typical Blowdown System for Early PWRs
Steam Discharge
to Atmosphere
CD
<- Cooling Water
~1
~1
- -=-*To Radwaste
FIGURE 2-6
Recommended PWR Blowdown System
Element
CdFlow
TO
t'• •V•• •Feedwater vSteam

/P • P ressure Heater
Steam
•Generator) - Valve oControl
Flow Condenser
-CoControl
Flo Valve
Elements-• •,,IC"
PoleLevel
(Ot l Control
STank
Steam Non-regenerative Valve
Generator Heat Exchanger
Regenerative
Heat Exchanger- Fite ...
Filter,..,/
To Heater , Demnineralizer .
Drain Tank -"I Service Condensate
To Low Pressure " Water IPump
Feedwater Heater -
Condensate
Polisher
(Optional)
how the water is used and the magnitude of other sources of contaminants.
For a plant without blowdown recovery and without condensate polishing,
makeup water and condenser leakage are the main sources of contaminants.
Moisture Separator Drain Routing. Since most of the soluble feedwater

impurities in an OTSG and some in an RSG are transported to the high-
pressure turbine with the steam, the turbine can be exposed to the
contaminants present in the feedwater. These contaminants can deposit in
the turbine in the liquid phase if their solubility drop below that which is
thermodynamically possible in the steam phase. For example, a salt of the
form xy, with x representing the cationic and y the anionic species, may
lead to formation on turbine surfaces of acidic species derived from anion
y or caustic solutions derived from cation x.
With initial condensation in the high pressure turbine, species such as
sodium, chloride or iron are transferred preferentially to the liquid phase
and are removed partially by turbine drains and by the moisture separator.
Since moisture separator and high-pressure feedwater heater drains often
are pumped forward into the feedwater, contaminant concentrations in the
feedwater generally are greater than at the effluent of the condensate
polisher systems. By routing some or all of these drains back to the
condensate polisher inlet, significant cleanup can be achieved.
OPERATING PLANT PERFORMANCE PROBLEMS

Table 2-1 presents a listing of operating plants, indicating date of
commercial operation, nuclear steam system supplier, and status of damage
as of 1993. Table 2-2 summarizes the number of units impacted by the
various problems. These tables have been prepared based on input from
NSSS vendors, from SGOG/Steam Generator Management Project (SGMP)
member utilities, and on information available in Reference 2.1.
TABLE 2-1
Status of Units Reporting Problems, December 1993
Denting Denting High

Date Support Tube- SCC sCC IGA IGA Cycle Water- Moisture
Plant ComI NSSS Wastaae Plate sheet Pittina (OD) (ID) (OD) (ID) Frettino Fatilue Erosion hammer Carryover
Almaraz 1 Dec-81 W 0 0 0
Almaraz 2 Sep-83 W 0 0 0
Angra 1 Dec-84 W 0 0 0 0
ANO 1 Dec-74 B 0 0 0 0
ANO 2 Nov-80 B 0 0 0 0 0
Asco 1 Dec-84 W 0 0 0 0
Asco 2 Mar-86 W 0 0 0
Atucha 1 Mar-74 KVNU 0 0
Beaver Valley 1 Oct-76 W 0 0 0
Beaver Valley 2 Nov-87 W 0 0 0
Belleville 1 Jun-88 F 0
Belleville 2 Jan-89 F 0
Beznau 1 (Org) Dec-69 W 0 0 0
Beznau 1 (Rpl) Apr-93 F
Beznau 2 Mar-72 W 0 0 0 0
Biblis A Aug-74 IKVU 0 0
Biblis B Apr-76 KWU 0 0 0 0
Blayais 1 Dec-81 F 0 0 0
Blayais 2 Feb-83 F 0 0 0
Blayais 3 Nov-83 F 0 0 0
Blayais 4 Oct-83 F 0 0 0
Borssele Jul-73 KWU 0 0
Braidwood 1 Jul-88 W o
Braidwood 2 Oct-88 W 0I
o L
0
TABLE 2-1 cont'd


Date Support Tube- SCC SCC IGA IGA Cycle Water- Moisture
Plant ComI NSSS Wastaae Plate sheet Pittino (OD) (ID) (OD) (ID) Frettina Fatiaue Erosion hammer Carryover
Brokdorf Oct-86 KWU
Bruce-A 1 Sep-77 AECL 0
Bruce-A 2 Sep-77 AECL 0 0 0
Bruce-A 3 Feb-78 AECL 0
Bruce-A 4 Jan-79 AECL 0
Bruce-B 5 Mar-85 AECL
Bruce-B 6 Sep-84 AECL 0
Bruce-B 7 Apr-86 AECL 0 U)
Bruce-B 8 May-87 AECL 0
Bugey 2 Feb-79 F 0 0 0 0
(D
Bugey 3 Feb-79 F 0 0 0 0
Bugey 4 Jul-79 F 0 0 0 0
(D
Bugey 5 (Org) Jan-80 F 0 0 0 0
Bugey 5 (Rpl) Feb-94 F
Byron 1 Sep-85 W 0 0 0 0 z
Byron 2 Aug-87 W 0
Callaway Apr-85 W 0
CD
Calvert Cliffs 1 May-75 C 0 0 0 0 0 0
Calvert Cliffs 2 Apr-77 C 0 0 0 0 0 0
Catawba 1 Jun-85 W 0 0 0 0
Catawba 2 Aug-86 W 0 r)
(D
Cattenom 1 Apr-87 F 0
Cattenom 2 Feb-88 F 0
Cattenom 3 Jul-90 F
TABLE 2-1 cont'd

Date Support Tube- SCC SCC IGA IGA Cycle Water- Moisture CD
0)
Plant Com'I NSSS Wastage Plate sheet Pitting (OD/ Frettina Fatiaue Erosion hammer Carryover
Cattenom 4 May-91 F
Chinon B1 Feb-84 F 0 CD
Chinon B2 Aug-84 F 0 0 0)
0
Chinon B3 Mar-87 F
Ct
Chinon B4 Apr-88 F 0 CD
Comanche Peak 1 Aug-90 W CD
CI
CD
Comanche Peak 2 Aug-93 W
CD
Conn. Yankee Jan-68 W 0 0 0 0 0 0
Cooki1 Aug-75 W 0 0 0 0 0 0 0 C
Cook 2 (Org) Jul-78 W 0 0 0 0 0 0 0 C

Cook 2 (Rpl) Mar-89 W CD
Cruas 1 Apr-84 F 0
Cruas 2 Apr-85 F 0 0
Cruas 3 Sep-84 F
Cruas 4 Feb-85 F
Crystal River 3 Mar-77 B 0 0 0
Dampierre 1 (Org) Sep-80 F 0 0
Dampierre 1 (Rpl) Jun-90 F
Dampierre 2 Feb-81 F 0 0 0
Damplerre 3 May-81 F 0 0 0 .0
Dampierre 4 Nov-81 F 0 0 0 0
Darlington 1 Nov-92 AECL
Darlington 2 Oct-90 AECL
Darlinqton 3 Feb-93 AECL
TABLE 2-1 cont'd
Plant Com'l NSSS Wastaae Plate sheet Pittina (OD) (ID) (OD) (ID) Frettina Fatiaue Erosion hammer Carryover
Darlington 4 Jul-93 AECL
Davis Besse 1 Jul-78 B 0 0
Diablo Canyon 1 May-85 W 0 0 0
Diablo Canyon 2 Mar-86 W 0 0 0 0
Doel 1 Feb-75 W 0 0 0 0 0 0
Doel 2 Nov-75 W 0 0 0 0 0 0
Doel 3 (Org) Oct-82 FC 0 0 0 0 0
Doel 3 (Rpl) Jun-93 FC
Doel 4 Jul-86 ACW 0 0 0 0 0 0
Emsland Apr-88 KWU
Farley 1 Dec-77 W 0 0 0
Farley 2 Jul-81 W 0 0 0
Fessenheim 1 Jan-78 F 0 0 0 0 0 CD
Fessenheim 2 Apr-78 F 0 0 0 0 0 0
Flamanville 1 Dec-86 F 0
Flamanville 2 Mar-87 F 0 0
Ft. Calhoun Sep-73 C 0 0 0
Genkai 1 Oct-75 M 0 0
Genkai 2 Mar-81 M rD
Ginna Jul-70 W 0 0 0 0 0 0
Golfech 1 Jun-90 F 0
Gosgen Feb-79 KWU 0 0
Grafenrheinfeld Dec-81 KWU
Gravelines 1 Nov-80 F 0
OLO OO
TABLE 2-1 cont'd M
¢D
Plant Com'l NSSS Wastaue Plate sheet Pittina (ODI, (ID1 (OD) I(ID1 Frettina Faticaue Erosion hammer Carryover
Gravelines 2 Dec-80 F 0 0 0 0
Gravelines 3 Jun-81 F 0 0 0 0
Gravelines 4 Oct-81 F 0 0 0 0
Gravelines 5 Jan-85 F 0 0 0
Gravelines 6 Nov-85 F 0 0
Grohnde Aug-84 KWU 0
Ikata 1 Sep-77 M 0 0
Ikata 2 Mar-82 M 0
Indian Point 2 Aug-74 W 0 0 0
Indian Point 3 (Org) Aug-76 W 0 0 0 0 0
Indian Point 3 (Rpl) Jun-89 W
Isar 2 Jan-88 KWU
Jose Cabrera Aug-69 W 0 0 0 0
Kewaunee Jun-74 W 0 0 0 0 0
Kori 1 Jun-77 W 0 0 0
Kori 2 Jul-83 W 0 0 0
Kori 3 Sep-85 W 0
Kori 4 Apr-86 W 0
Lovilsa 1 May-77 AEE
Lovilsa 2 Jan-81 AEE
Maine Yankee Dec-72 C 00 0 0
McGuire 1 Aug-80 W 0 0 0 0
McGuire 2 Mar-84 W 0 0 0 0
Mihama 1 Nov-70 W 0 0
0
TABLE 2-1
Plant Com'l NSSS Wastage Plate sheet Pittinq (OD) (ID) (OD) (ID) Fretting Fatigue Erosion hammer Carryover
Mihama 2 (Org) Jul-72 M 0 0 0 0 0 0
Mihama 2 (Rpi) Apr-94 M
Mihama 3 Dec-76 M 0 0 0
Millstone 2 (Org) Dec-75 C 0 0 0 0 0 0
Millstone 2 (Rpl) Jan-93 BWC
Millstone 3 Apr-86 W - 0,
Neckarwestheim 1 Jun-76 KWU 0 0
Neckarwestheim 2 Jan-89 KWU
Nogent/Seine 1 Feb-88 F 0 0 0
Nogent/Seine 2 May-89 F 0 0
North Anna 1 (Org) Jun-78 W 0 0 0 0 0 0 C~)
North Anna 1 (Rpl) Apr-93 W
North Anna 2 Dec-80 W 0 0 0 0 0 0
Obrigheim (Org) Oct-68 KWU 0 0 ~17
Obrigheim (Rpl) Aug-83 KWU
Oconee 1 Jul-73 B 0 0
Oconee 2 Sep-74 B 0 0
Oconee 3 Dec-74 B 0 0
Ohi 1 Mar-79 W 0 0
Ohi 2 Dec-79 M 0 0
Ohi 3 Dec-91 M
Ohi 4 Feb-93 M
Palisades (Org) Dec-71 C 0 0 0 0 0
Palisades (Rpl) Mar-91 C I I _II_
TABLE 2-1 cont'd
Plant Com'l NSSS Wastage Plate sheet Pitting (OD) (ID) (OD) (ID) Fretting Fatigue Erosion hammer Carryover
Palo Verde 1 Jan-86 C 0 0 0 0
Palo Verde 2 Sep-86 C 0 0 0 0
Palo Verde 3 Jan-88 C 0 0 0 0
Paluel 1 Dec-85 F 0 0 0
Paluel 2 Dec-85 F 0 0 0
Paluel 3 Feb-86 F 0 0 0 rD
*1
Paluel 4 Jun-86 F 0 0 0
Penly 1 Dec-90 F CJ
n
Penly 2 Feb-92 F
Philippsburg 2 Dec-84 KWU
Pickering-A 1 Jul-71 AECL 0
Pickerinq-A 2 Dec-71 AECL 0
Pickering-A 3 Jun-72 AECL
Pickering-A 4 Jun-73 AECL
Pickering-B 5 May-83 AECL 0
Pickering-B 6 Feb-84 AECL 0
Pickering-B 7 Jan-85 AECL
Pickering-B 8 Feb-86 AECL
Point Beach 1 (Org) Dec-70 W 0 0 0 0 0
Point Beach 1 (Rpl) Apr-84 W 0
Point Beach 2 Oct-72 W 0 0 0 0
Prairie Island 1 Dec-73 W 0 0 0 0 0
Prairie Island 2 Dec-74 W 0 0 0
Rancho Seco Apr-75 B 0 0 0 0
TABLE 2-1 cont'd
Date Support Tube- sCC SCC IGA IGA Cycle Water- Moisture
Plant ComI NSSS Wastage Plate sheet Pittinq (OD) (ID) (OD) (ID) Fretting Fatigue Erosion hammer Carryover
Ringhals 2 (Org) May-75 W 0 0 0 0 0 0
Ringhals 2 (Rpl) Aug-89 W
Ringhals 3 Apr-81 W 0 0 0 0 0
Ringhals 4 Oct-83 W 0 0
Robinson 2 (Org) Mar-71 W 0 0 0 0
Robinson 2 (Rpl)
Oct-84 W
Saint-Alban 1 Aug-85 F 0
Saint-Alban 2 Jul-86 F 0 0 0
Salemi1 Jun-77 W 0 0 0 0
Salem 2 Oct-81 W 0 0 0 c-I

San Onofre 1 Jan-68 W 0 0 0 0 0 0 0 0 0
Mt
San Onofre 2 Aug-83 C 0 0
San Onofre 3 Apr-84 C 0 0
Seabrook Jul-90 W 0
Sendai 1 Jul-84 M 0
Sendai 2 Nov-85 M 0
0 0 0 0 0 (D
Sequoyah 21
Sequoyah Jul-81 W
Jun-82 W 0 0 0 0 0 -t
Shearon Harris May-87 W 0 0 0 0 0 0

South Texas Proj.1 Aug-88 W 0 0 0 1 1 0 1
South Texas Proj. 2 Jun-89 W 0
St. Lucie 1 Dec-76 C0 0 0 0
St. Lucie 2 Aug-83 C0 0 0
St.-Laur.-des-Eaux B1 Aug-83 F O0 0 0 0O
TABLE 2-1 cont'd
Plant Com'l NSSS Wastaae Plate sheet Pitting (OD) (ID) (OD) (ID) Frettina Fatiaue Erosion hammer Carryover
St.-Laur.-des-Eaux B2 Aug-83 F 0 0 0
Stade Jan-72 KWU 0
Summer 1 Jan-84 W 0 0 0
Surry 1 (Org) Dec-72 W 0 0 0 0 0
Surry 1 (Rpl) Jul-81 W 0
Surry 2 (Org) May-73 W 0 0 0 0
Surry 2 (Rpl) Sep-80 W 0
Takahama 1 Nov-74 W 0 0 0 0
Takahama 2 Nov-75 M 0 0
Takahama 3 Jan-85 M 0
Takahama 4 Jun-85 M 0
Three Mile Island 1 Sep-74 B 0 0 0
T~ihange 1 Sep-75 W 0O 0 0 0 0 0 0
Tihange 2 Feb-83 FC 0 0 0
Tihange 3 Nov-85 FC 0 0 0 0
Tomari 1 Jun-89 M
Tomari 2 Apr-91 M
Tricastin 1 Dec-80 F 0 0 0 0
Tricastin 2 Dec-80 F 0 0 0 0
Tricastin 3 May-81 F 0 0
Tricastin 4 Nov-81 F 0 0 0 0
Trillo 1 Aug-88 K
Trojan May-76 W 0 0 0 0
Tsuruga 2 Feb-87 M I I 1 0
TABLE 2-1 cont'd
Status of Units.Reporting Problems, December 1993
Date Support Tube- SCC SCI IGA IGA Cycle Water- Moisture
Plant Com'lI NSSS Wastaae Plate sheet Pittina (ODI (ID) (OD) (ID) Frettina Fatiaue Erosion hammer Carryover
Turkey Point 3 (Org) Dec-72 W 0 0 0 0
Turkey Point 3 (Rpl) Apr-82 W 0
Turkey Point 4 (Org) Sep-73 W 0 0 0 0 0 0
Turkey Point 4 (Rpl) May-83 W 0 0
Uljin 1 Sep-88 F
Uljin 2 Sep-89 F
Unterwesser Oct-78 KWU 0
Vandellos 2 Dec-87 W 0
Vogtle 1 May-87 W 0
Vogtle 2 May-89 W
0 rfD
Waterford 3 Sep-85 C 0
Watts Bar 1 W
Watts Bar 2 W
-t
Wolf Creek Sep-85 W 0 C-3
0t
Wolsung 1 Apr-83 AECL
Yankee Rowe Jul-61 W 0 0
Younggwang 1 Aug-86 W 0
Younggwang 2 Jun-87 W 0
Zion 1 Jan-73 W 0 0 0 0 0 01
Zion 2 Sep-74 W 0 0 0 0 0
WI
TABLE 2-2
Number of Units Affected by Steam Generator Problems
No. of PWR Units

As of 3/77 As of 8/82 As of 12/93
Reported Problems 52 99 235
Denting
* Tube support corrosion 15 30 36
" Tubesheet Corrosion 6 12 50
Tubing corrosion
" Wastage 19 28 39
" Pitting 0 3 16
* ID Cracking 1 22 102
" OD SCC/IGA 6 22 85
Mechanical Damage
" Fretting 9 15 128
" Fatigue Cracking 3 4 15
" Impingement 0 2 10
No problems 26 32 48
No problems >5 years 1/14 4/57 20/205
operation (no. of units/no.
>5 years operation)
REFERENCES
2.1 "Steam Generator Progress Report, Revision 10." EPRI Steam
Generator Strategic Management Project. November 1994.
3
STEAM GENERATOR PROGRAMS
S. J. Green, EPRI
J. P N. Paine, EPRI
INTRODUCTION
Since 1977 there has been a major effort funded by the electric utility
industry and EPRI to address steam generator availability issues, as
discussed in Chapters 1 and 2. From 1977 to 1982, Steam Generator Owners
Group I (SGOG I) conducted the initial effort. From 1983 to 1986 a follow-
on effort was conducted under the guidance of Steam Generator Owners
Group II (SGOG II). A major EPRI-funded effort, called the Steam Generator
Reliability Project (SGRP), has addressed the continuing steam generator
problems over the period 1987 to 1992. The three programs have been
managed by the Steam Generator Project Office at EPRI. It is the purpose
of this chapter to summarize the three programs conducted under these
efforts.
STEAM GENERATOR OWNERS GROUP I
Background
Steam Generator Owners Group I was established for the specific
purpose of determining the best solutions to steam generator problems.
The Owners Group was conceived when it became evident that a concerted
effort on the part of interested utilities was needed to bring about solutions
to the growing number of malfunctions in steam generators.
The objective of the Steam Generator Owners Group effort was to
provide options such that each pressurized water reactor (PWR) owner
could select the options most appropriate to his situation and develop his
company's individual plan. The vendors were unable to fund the level of
effort that the Owners Group determined necessary to best serve the interests
of the utility industry. The major steam generator vendors did, however,
support the Owners Group program by performing work on a no-fee and
cost-shared basis.
In May of 1977, at a meeting in Chicago, the Interim Executive
Committee of the Steam Generator Owners Group presented an appraisal
of what was to be accomplished. A formula to allocate costs among owners
and prospective owners of pressurized water reactor (PWR) units was also
presented to the Owners Group.
At the July 1977 meeting, the Owners Group elected a ten-member
Executive Committee. Concurrent with this meeting, the Executive
Committee approved the budget for the highest priority portion of the
technical program. By the end of the program on December 31, 1982, 28
utilities had contributed $36 million to support the Steam Generator
Steam Generator Programs 3-3
Program. With interest, the total available funds amounted to $43.8 million.
The program was completed for $42.4 million including program
management costs. Essentially all work, excluding the transfer of
technology, was completed in 1982.
Technical Program
The overall objective of the Steam Generator Owners Group I technical
program was to provide the utilities, vendors, and architect/engineers with
technology-in the form of recommendations and guidelines-that would
minimize the likelihood of steam generator degradation over the life of the
plant. For owners of operating units, this included methods to minimize
inleakage of impurities into the secondary system, methods for cleaning
up feedwater and condensate that may become contaminated with these
impurities, steam generator chemistry control options that minimize the
effects of those impurities, periodic maintenance procedures for their
removal or neutralization, changes in the mode of operation, and some
feasible design modifications. The program also evaluated alternate
materials, systems, and component designs which would be more tolerant
to the potential ingress of impurities.
The needs of PWR steam generator owners were addressed in the overall
steam generator program plan. This plan recognized that any corrective
action which is developed by plant and laboratory studies must ultimately
be tried and proven in the plant. Thus, there was an iterative flow of
information and recommended actions from the plant to the laboratory and
back to the plant. The resulting program approach, depicted in Figure 3-1,
addresses the needs under the following major categories:
* definition of factors which affect steam generator reliability,
" actions in the steam generator to minimize corrosion and mechanical
damage,
" actions in the balance of plant to minimize corrosion and mechanical
damage, and
* improved steam generator designs.
A summary of the program is shown in Figure 3-2. This program
involved a complex interaction of many systems and disciplines including
the mechanical, thermal and hydraulic design of the steam generators; the
water chemistry; water chemistry control; design and construction materials
of secondary plant components and systems and plant operations. The
following paragraphs provide the objectives for the subprograms within
the program.
FIGURE 3-1
Steam Generator Program Approach
FIGURE 3-2
Steam Generator Owners Group I Program
Inspection-destructive and nondestructive

-Surveillance of plant chemistry and steam
generator damage
-Safety analysis, water hammer-
-Define problems tube plugging criteria
-Causes of corrosion and damage
-Thermal and hydraulic analyses of
steam generator
-Vibration, wear, and fatigue studies
ondensers-avoid inleakage of cooling

water and oxygen
ondensate polishing-reduce soluble
Actions in balance-of-
and insoluble impurities
Solve -plant to reduce ingress Filters-remove insolubles
steam and corrosion Feed train materials, chemistry and routing
generator Secondary system layup-reduce corrosion
problems Oxygen control measures-reduce
oxidant ingress
-Water soaks-remove solubles

-Neutralization-neutralize corrodants
Actions in steam -Steam generator layup-reduce corrosion
generator to minimize -Chemical cleaning-remove corrosion
corrosion/damage products
-Alternate chemistries-reduce corrosion
LSludge lance
New materials, tubing and

tube support plates
4mprove new designs -- New support plate designs
-New steam generator designs
Modify tube support plate and
tubesheet design-reduce concentration
of corrosive chemicals
Subprogram 0801: Qualify Nondestructive Examination Techniques.

The purpose of this subprogram was to further develop NDE capabilities
for damage and diagnostic assessments in current and future steam
generators. The primary objectives of this subprogram were to develop
NDE capabilities to permit the following:
* Evaluation of conditions in steam generators leading to any of the
existing or projected mechanisms for damage, emphasizing denting
and related crevice blockage.
* Determination of essential information on initiation, type, location,

azimuthal orientation, depth, shape, and other important parameters
for characterization of defects.
" Evaluation of corrective action measures for denting corrosion and other
damage mechanisms.
" Minimization of downtime and personnel radiation exposure required
for steam generator inspections through use of advanced techniques.
Subprogram 0802: Evaluation of On-Order Steam Generator Designs.

The objective of this subprogram was to conduct comprehensive reviews
of thermal and hydraulic aspects of the designs for steam generators on
order. The ability to calculate overall and local thermal/hydraulic conditions
was augmented and verified. Based on the relationship between thermal/
hydraulic conditions and steam generator problems, the susceptibility of
design features to operating malfunctions was assessed.
Subprogram 0803: Perform Safety Analysis and Consequence

Evaluations. The objective of this subprogram was to provide the analytic
capability to demonstrate that continued operation of steam generators
which are degraded and steam generators which have undergone
subsequent corrective action does not pose an unacceptable risk to the health
and safety of the public.
Subprogram 0804: Determination and Verification of Required Water

Chemistry Limits. The objective of this subprogram was to determine the
maximum concentrations of impurities such as sodium, potassium, chloride,
sulfate, copper, iron, and oxygen, etc., that can be tolerated in conjunction
with various chemical additive systems (e.g., ammonia, hydrazine,
morpholine, sodium phosphates, etc.) while maintaining steam generator
corrosion rates at acceptable levels.
Subprogram 0805: Evaluate Alternate Steam Generator Materials and

Processing. The objective of the subprogram was to identify, evaluate,
and test steam generator materials, materials processing and designs for
new steam generators, and replacement tube bundles.
Subprogram 0807: Develop Neutralizer for Crevice Acids. The objective

of this subprogram was to collect and analyze data to identify the causes of
problems in operating steam generators and establish priorities for corrective
action. These causes were related to the site characteristics, the steam
generator or unit design, and/or utility operating practices. Specific goals
of this subprogram were:
" To relate steam generator chemistry variations during all phases of

operation to observed corrosion degradation in the steam generator.
* To develop a detailed understanding of corrosion product and ionic
impurity transport to steam generators and identify techniques for
reducing such impurity ingress.
" To develop a data bank on steam generator chemistry variations in
PWRs during all phases of operation in order to demonstrate achievable
chemistry control levels.
* To identify and quantify, to the extent possible, the value of techniques
(e.g., condensate polishers, flushing, etc.) for reducing impurity levels
in steam generators and thus corrosion attack.
* To provide additional data as required to support other subprograms.
Subprogram 0809: Thermal-Hydraulic Analysis and Evaluation. The

objective of this subprogram was to determine the thermal and hydraulic
conditions which exist in local regions of steam generators, and to develop
theoretical models to explain them. Such understanding would allow
analytical evaluation of the role of thermal and hydraulic conditions in steam
generator degradation, assessment of corrective actions, and design of new
units to avoid the problems.
Subprogram 0810: Evaluate Methods for Reducing Carry-Over and

Decreasing Pressure Drop Through Steam Separators. The objective
of this subprogram was to develop analytical models to describe the
operation of moisture separators in steam generators and to predict their
performance. This ability would permit separators to be designed to
perform satisfactorily and eliminate the need for costly modifications in
the field. Low moisture carry-over would also minimize adverse effects on
turbines. Moreover, such analytical techniques would support the design
of lower pressure drop separators, the use of which would yield a greater
circulation ratio in steam generators.
Subprogram 0811: Development and Technical Support of Initial Plant

Application of Chemical Cleaning. The objective of this subprogram was
to develop a chemical cleaning process for removal of corrosion products
from the secondary side of PWR steam generators. The selected chemical
cleaning agent would be able to dissolve the wide variety of corrosion
products present in steam generators and still be compatible with the
materials of the steam generator.
Subprogram 0812: Evaluate Secondary System Layup Chemistry

Options. The objective of this subprogram was to evaluate secondary
system designs and operating practices that were being or would be used
in operating plants and in plants on order to control layup or transient
chemical environments in secondary systems, and to develop a method of
minimizing secondary system corrosion during layup.
Subprogram 0813: Evaluate Steam Generator Layup and Transient

Chemistry Control Options. The objective of this subprogram was to
evaluate steam generator and supporting system designs and operating
practices that were being or would be used in operating plants and plants
on order to control the layup and transient chemical environment in the
steam generator. Optional methods were to be developed for preventing,
alleviating, or minimizing steam generator corrosion during layup and
transients.
Subprogram 0814: Evaluate Methods for Reducing Tube Vibration,

Wear, and Fretting. The objectives of this subprogram were to identify,
develop, evaluate, and recommend methods which can be used to evaluate
and reduce tube vibration and fretting in currently operating units and in
new steam generator designs both in once-through units and U-tube units
where crevice clearances may be increased by chemical cleaning or where
circulation ratios are increased.
Subprogram 0816: Evaluation of On-Line Chelant Additions. The

objective of this subprogram was to determine the viability of using chelating
agents for on-line additions to steam generators in order to remove and
prevent the buildup of corrosion products, particularly on the tubesheet
and in the tube-to-tube-support plate crevice region to reduce/eliminate
one of the contributing factors to. steam generator corrosion.
Subprogram 0818: Destructive Analysis of Steam Generator

Components. The objective of this subprogram was to facilitate the
destructive examination of components removed from operating steam
generators. This includes promoting the acquisition of field samples, the
destructive examination of these components, and the analysis of the
resulting data. The intent was to provide accurate diagnosis of the cause
and effect of steam generator component degradation, promoting better
understanding of steam generator issues, and providing information to
better plan corrective actions.
Subprogram 0820: Evaluate Condensate Polishers. The objective of

this subprogram was to evaluate the effectiveness and side effects of
condensate polishers. Specific steps to achieve this objective included
evaluation of the effect (e. g., type and extent of corrosion) that condensate
polisher impurities can cause under normal and faulted operating conditions
with and without condenser leakage as a consequence of their continuous
and/or intermittent operation.
Subprogram 0821: Develop High and Low Temperature Filters. The

objective of this subprogram was to develop, test, and demonstrate the
effectiveness and operational reliability of a filtration system to substantially
remove insoluble impurities from the condensate and feed trains of
pressurized water reactor plants. A further objective was to develop a
parametric analysis procedure which will aid in determining filter system
location for any given power plant.
Subprogram 0822: Evaluate and Test Boiler Water Cleanup Systems.

The objective of this program was to develop improved methods for
removing aggressive impurities (soluble chemicals) from the bulk fluid and
insoluble corrosion products (sludge and scale) from internal surfaces
(tubesheet, tube support plates, tubes, separator deck, etc.) of recirculating
steam generators.
Subprogram 0823: Instrument Operating Units. The objective of this

subprogram was to instrument representative operating steam generator
units to an extent that will allow recorded data (1) to be used in the
verification of existing and planned steam generator thermal-hydraulic
codes, (2) substantiate design and performance predictions associated with
the instrumented unit and, (3) form the basis from which design and
operational criteria can be formulated in order to enhance steam generator
performance.
Subprogram 0824: Evaluation of Hydrogen Evolution Monitoring. The

objective of this subprogram was to develop And operationally test an on-
line hydrogen monitoring system which can accurately determine the rate
of steam generator corrosion that results in hydrogen formation.
Subprogram 0830: Improve Control of Condenser Cooling Water

Leakage. The objective of this subprogram was to identify the sources of
condenser cooling water inleakage and to identify and develop methods to
detect, isolate, and to obtain essentially zero inleakage.
Subprogram 0836: Evaluate'Secondary System Oxygen Control

Measures. The objective of this subprogram was to recommend balance-
of-plant design features and operating/maintenance procedures which are
aimed at effective elimination of dissolved oxygen-induced corrosion from

the secondary plant of PWR nuclear power plants.
Subprogram 0839: PWR Steam Generator Technology Transfer. The

objective of this subprogram was to transfer the results of the Steam
Generator Owners Group program to the electric utility industry in.a form
that is most likely to be applied. Information packages were to be developed
for use by utilities, steam generator vendors, architect/engineers, and
component manufacturers. The packages will facilitate the implementation
of the latest steam generator technology with minimum effort.
STEAM GENERATOR OWNERS GROUP II
Background
The SGOG I was completed successfully. However, additional
challenges arose while this program was under way, and the utilities decided
to form a new group and program to address the newer issues. The new
group comprised more than 37 utilities and provided $29 million to support
the program. As discussed below, the program was formulated to focus on
specific problems and remedial actions.
Technical Program
This program had comparable objectives to the SGOG I and naturally
built on the results of that program. However, the program was designed
to address the causes and remedies for specific damage forms. A listing of
the major components (or projects) of the program is presented in Table 3-1
and a summary of each project is described in the following paragraphs.
S301: Improved Steam Generator Nondestructive Examination

Techniques. Extending the results of the SGOG I program, the objective
of this project was to further improve nondestructive examination (NDE)
capabilities for damage and diagnostic assessment in current and future
steam generators. The primary objectives were to develop new or improved
NDE capabilities in the following areas:
* Improved detection and characterization of known defect conditions
such as tube cracking, pitting, wastage, magnetite build-up, support
plate and lattice bar cracking, U-bend stress corrosion cracking (SCC),
and IGA/SCC in the tubesheet area, etc. Detection and characterization
of previously unidentified tubing flaws/anomalies were also included.
TABLE 3-1
Steam Generator Owners Group IIProgram Summary
Project
Number Project Title
S301 Improved Steam Generator Nondestructive Examination
Techniques
S302 Arrest Steam Generator Intergranular Attack and Tube Cracking in
and Above the Tubesheet Crevice and in Other Crevice and
Dryout Locations
S303 Destructive Analysis of Steam Generator Components
S304 Destructive Analysis of Steam Generator Components
S305 PWR Steam Generator Chemical Cleaning Process Development
and Evaluation
S306 Evaluation and Improvement of Steam Generator Performance
and the Effectiveness of Preventive and Corrective Measures
S307 PWR Steam Generator Technology Transfer
S308 Causes and Corrective Actions for Pitting of Steam Generator
Tubing
S309 Minimize the Effects of Sludge, Scale, and Deposits on Corrosion
in Steam Generators
S310 Tube Fretting and Fatigue in Preheat Steam Generators
S311 Tube Support Materials and Designs and Crevice Corrosion
o Testing of selected systems in the retired Surry Unit 2 steam generator

located at the Hanford, Washington, site as well as at other retired steam
generators. This includes both newly developed systems and existing
systems.
o Field qualification of newly developed inspection systems, including
systems developed under SGOG I and through EPRI work in
operational steam generators.
* Assembly of an archive of tubing samples containing realistic defects
whose sizes have been determined nondestructively. The samples were
used in the evaluation of inspection systems and in the training and
qualification of inspectors.
S302: Arrest Steam Generator Intergranular Attack (IGA) and Tube

Cracking (SCC) in and Above the Tubesheet Crevice and in Other
Crevice and Dryout Locations. The objectives of this project were to
determine the causes and mechanisms of SCC and IGA in and above the
tubesheet crevice and in other crevices or dryout locations and qualify
remedial measures. Subobjectives included: determine the material
conditions which have been shown to be susceptible in the field units; define
the environmental conditions which cause corrosion; define susceptible
materials and required stress levels (including evaluation of surface residual
stresses) in laboratory testing; evaluate the thermal/hydraulic and chemical
conditions which lead to corrosion in laboratory testing; and develop and
qualify chemical, operational, or mechanical solutions to the problem. The
output of this project was an understanding of the degree of susceptibility
and alternative solutions for the extension of steam generator life.
S303: Causes and Corrective Actions for Primary Water Cracking of

Steam Generator Tubing. The objectives were to evaluate the causes of
primary water cracking occurring in steam generators, to determine the
matrix of tubes likely susceptible to cracking; and to identify, verify, and
follow the implementation of practical solutions. Emphasis was on
identifying those conditions in steam generators (e.g., loading, environment,
and results of material processing) which have caused primary water
cracking of steam generator tubes. The project addresses mill-annealed,
stress-relieved, and thermally treated alloy 600 as well as other corrective
heat treatments proposed. In addition, alloys 800 and 690 were tested as
comparison materials.
S304: Destructive Analysis of Steam Generator Components. The

objective of this project was to facilitate the examination and evaluation of
components removed from operating steam generators.
S305: PWR Steam Generator Chemical Cleaning Process Development

and Evaluation. The objective of this project was to supplement the work
done as part of the Steam Generator Owners Group I program in the
development of a generic process for the chemical cleaning of nuclear steam
generators, and to follow, monitor, and report on the first application of
that process in an operating PWR.
S306: Evaluation and Improvement of Steam Generator Performance

and the Effectiveness of Preventive and Corrective Measures. The
objective was to evaluate the improvement in performance of steam
generators following implementation of the various options being applied
by utilities as remedies for steam generator corrosion.
S307: PWR Steam Generator Technology Transfer. The objective of

this project was to transfer the results of other steam generator projects, as
they were obtained, to the electric power industry in a form that is most
likely to be applied. Information packages were developed for use by
utilities, and, if appropriate, for architect/engineers, steam generator
vendors, and manufacturers of other components. The packages were
designed to advise executives of key findings and corrective actions in a
persuasive manner. They also provided technical supervisors with the
materials necessary, including technical details, to easily and effectively
provide appropriate members of their staffs with training in the latest steam
generator technology.
S308: Causes and Corrective Actions for Pitting of Steam Generator

Tubing. The objective of this project was to define the water chemistry and
other environmental conditions under which pitting of steam generator
tubes may occur and qualify remedial measures for pitting corrosion.
S309: Minimize the Effects of Sludge, Scale, and Deposits on Corrosion

in Steam Generators. The objectives of this project were to summarize
and document past research and utility experience; to identify the problems
and the best practices for sludge and deposit control and removal; to develop
improved and optimized processes for sludge and deposit control and
removal with AVT operation; and to evaluate the role of sludge, scale, and
deposits on corrosion processes.
S310: Tube Fretting and Fatigue in Preheat Steam Generators. The

objectives of this project were to document the causes of and corrective
actions for tube vibration and wear in preheat steam generators; model the
phenomena analytically; and establish criteria for preventing and correcting
flow-induced vibration in steam generators, particularly those with
preheaters. There are three different preheater types: split flow and counter
flow types with baffles to induce cross flow in the tube bundle and a counter
flow type in which the flow is parallel to the tube axes. For those with cross
flow baffles, there are several different variations for each type. It was
desirable to establish criteria for preventing or correcting flow-induced tube
vibration that can be applied to all types and variations of preheaters in
steam generators.
A parallel objective was to identify the thermal and hydraulic conditions
in preheat steam generators which are associated with the denting corrosion
reported in one plant and which could lead to tube corrosion (e.g., areas of
flow stagnation). Corrective actions can then be formulated to eliminate or
avoid such conditions.
S311: Tube Support Materials and Designs and Crevice Corrosion.

The objectives were to identify and test tube/tube support materials and
designs for susceptibility to crevice corrosion and to qualify them for use in
steam generators. Qualification tests were performed under prototypical
conditions when possible.
STEAM GENERATOR RELIABILITY PROJECT
Background
Although the SGOG II program was completed successfully, steam
generator corrosion and mechanical degradation continued to occur, and it
was the judgment of the utility management familiar with steam generator
operations that a focused steam generator industry effort should continue
but at a somewhat reduced effort. The EPRI management considered that
the new effort should be supported primarily under the EPRI base program,
without major support from the utilities. It was judged that steam generator'
degradation was still very serious but not of "crisis" proportions as had
been the case during the previous Owners Group programs. Accordingly,
a major EPRI project was formed, called the Steam Generator Reliability
Project, at a funding level of $30 million over the period 1987 through 1992.
Technical Program
Again, this program was focused on addressing the causes and remedies
for the corrosion and mechanical damage forms that persisted. A listing of
the projects comprising the program is presented in Table 3-2. To further
guide the program, a steam generator .R&D program strategy was
developed. This strategy is depicted in Figure 3-3. It breaks down the
program into three major categories: mechanism management
(understanding the causes of damage observed and providing qualified
long-term remedies), defect management (near term corrective
maintenance), and life extension/replacement (long-term corrective
maintenance). The various elements in these three categories are shown.
Methods of evaluation and measuring performance are also listed. The
objectives of each project in Table 3-2 are discussed below.
S401: Improved Steam Generator Water Chemistry-Field Studies and

Water Chemistry Guidelines. The objective of this project was to develop
improved guidelines on water chemistry control through field studies on
hideout/hideout return, and ionic and corrosion product transport
throughout the secondary cycle for plants with various generic-type design
features (e.g., powdered resin or deep-bed condensate polishers, makeup
water treatment plants with and without filtration, feed trains with and
without copper components, plants with drains returned to the condenser
or pumped forward, etc.) This project built on work conducted in the
SGOG I and II programs, and provided for periodic review and revision of
the PWR Secondary Water Chemistry Guidelines. Field-related tasks were
co-funded by the host utility
TABLE 3-2
Steam Generator Reliability Project Program Summary
Project
Number Project Title
S401 Improved Steam Generator Water Chemistry-
Field Studies and. Water Chemistry Guidelines
S402 Improved Steam Generator Water Chemistry
Through Improved Balance-of-Plant Operations
S403 Improved Steam Generator Impurity Removal Techniques
S404 Demonstrated Steam Generator In-Service Inspections
S405 Technology Transfer
S406 Steam Generator Primary Side Remedies
3407 Steam Generator Secondary Side Remedies
S408 Improved Materials and Designs for Steam Generator Repair and
Replacement
S409 Alternate Secondary Water Chemistry
S410 Tube Vibration, Mechanical Damage, Fretting and Wear
S411 Thermal-Hydraulic Code Development and Validation
S412 Steam Generator Strategic Planning Guidelines
S402: Improved Steam Generator Water Chemistry Through Improved

Balance-of-Plant Operation. The objective is to conduct studies to improve
the design, operation, and maintenance of balance-of-plant components
and systems as they affect steam generator corrosion-related environmental
conditions. Close coordination is maintained with other EPRI projects
FIGURE 3-3 --L
Steam Generator R&D Program Strategy
STEAM GENERATOR MANAGEMENT
1. Identify problem/issues
Performance Evaluation: 2. Track progress
1. Lost capacity due to steam 3. Transfer technology/package material
generator problems 4. Establish priorities
2. Average annual maintenance 5. Integrate sub-element
costs
Mechanism Management Defect Management Life Extension/Replacement
(Preventive Maintenance) (Near-Term Corrective Maintenance) (Long-Term Corrective
Maintenance)
Damage Forms 1. Define cause 1. Remove only those tubes from 1. Long range planning
1. Wastage 2. Develop remedies service that are required to be a. repair/replace decision
2. Denting a. W/C G/L removed (for reliability and/or 2. Improved design
3. Wear (1) supporting G/L safety) and no others. a. materials
4. Pitting b. flushing, soaks, layup a. ISI/NDE b. configuration
5. ODSCC c. alternate W/C b. plugging limit 3. Replacement activities
6. ODIGA d. sludge control (1) criteria
7. IDSCC (1) chemical cleaning c. leak detection
8. Fatigue e. stress relief d. leak location
f. AVB replacement e. progression rate
g. FIV models 2. Assess consequences
h. T-H codes (rupture effects analyses)
peformance 1. Annual tubes plugged 1. Annual tube leak outages 1. Duration of replacement outages
Measures: 2. Tube rupture events (4/6) 2. Tube rupture events (2/6) 2. Average cost of replacement
3. Excess tube plugging (including replacement power)
4. Regulatory concern 3. Future S/G problems
directed at research to improve component and system reliability for its

own sake. Wherever possible, steam generator work is factored into these
EPRI projects to decrease the number of research efforts in S402. This task
focuses on revisions of the Guidelines for Design and Operation to Minimize
Steam Generator Corrosion, which were developed in the SGOG I and II
programs. This task incorporates and builds on related work in other
programs. Field-related tasks are cofunded by the host utility.
S403: Improved Steam Generator Impurity Removal Techniques. This

project builds on prior SGOG and EPRI work and provides for issuance of
qualified procedures for improved steam generator impurity removal
techniques, including sludge lancing, chemical cleaning, soaking, and
crevice flushing. This project incorporates work in previously cofunded
research by Consolidated Edison Co. and ESEERCO. Field-related tasks
are cofunded by the host utility. Specific goals include:
o To improve steam generator sludge lancing capabilities.
* To improve steam generator soaking and flushing processes.
" To improve once-through steam generator sludge removal capabilities.
" To improve steam generator chemical cleaning processes.
S404: Demonstrated Steam Generator In-Service Inspections. The

objective of this work was to extend the results of SGOG programs to further
improve NDE capabilities for damage and diagnostic assessments for steam
generator tube conditions. The final result of the task will be consistent
and accurate steam generator NDE and effective use of utility resources.
The information obtained was incorporated into revisions of the Steam
Generator In-Service Inspection Guidelines. This will minimize missed calls,
false calls, unnecessary tube plugging, and forced outages due to tube leaks.
S405: Technology Transfer. The objective of this task was to transfer

SGRP results on a continuous basis to the electric power industry in a form
that is most likely to be applied. Information was developed for use by
utilities, architect/engineers, steam generator vendors, and manufacturers
of other components. This information was designed to advise executives
of key findings and corrective actions in an effective manner. It also provided
technical supervisors with the materials, including technical details,
necessary to easily and effectively provide appropriate members of their
staffs with training in the latest steam generator technology. Major items
included:
* Product technology transfer packaging and distribution.

* Periodic (annual) assessment of steam generator problems and
resolution progress.
" Periodic newsletters and bulletins.
S406: Steam Generator Primary Side Remedies. The objective of this

project was to establish the susceptibility of installed alloy 600 tubing to
primary side cracking and corrosion and to provide verified remedial
measures. Another objective was to provide for evaluation of field remedies
instituted by utilities:
• To evaluate effects of stress on crack initiation (longer-term objective).
* To evaluate effects of environment on crack initiation.
" To develop methods of determining material susceptibility to IGSCC
in situ.
* To develop and apply a model for prediction of IGSCC (longer-term
objective).
" To develop and evaluate field remedies for U-bend and transition
cracking.
* To analyze samples removed from steam generators in service.
S407: PWR Steam Generator Secondary Side Remedies. The objectives

for this project were as follows:
* To establish the causes of steam generator tubing secondary side
corrosion and to develop and evaluate remedies for each of the several
damage forms; also, to develop predictive capabilities allowing early
implementation of selected remedies.
* To determine causes and evaluate remedies for alloy 600 (IGA/IGSCC)
in tube support structure crevices.
• To determine causes and evaluate remedies for other alloy 600 corrosion
*damage forms at tube support structure crevices and at other flow-
restricted areas (e.g., sludge piles and scales) in the steam generator
bundle.
* To continue evaluation of effectiveness of previously proposed remedial
actions.
* To analyze samples removed from steam generators in service.
* To extend and verify a water concentration model (MULTEQ) for
predicting crevice environments. This included measuring hideout
rates, speciation and pH in laboratory testing, model expansion,
database measurements of thermodynamic functions for database
improvements, and comparison to field measurements.
* To predict steam generator corrosion rates.
* To follow aging of in-service alloy 600 tubing and focus on signs of

low-temperature aging phenomena that may affect future lifetimes.
" To evaluate the effect of proposed alternate secondary water chemistry
treatments on steam generator construction materials and various forms
of corrosion.
S408: Improved Materials and Designs for Steam Generator Repair

and Replacement. This project included the following objectives:
* To specify improved steam generator materials and designs for major
repairs and replacement.
e To develop and life-cycle-qualify advanced sleeving and other repair
materials and procedures.
* To qualify thermally treated alloy 690.
o To develop advanced PWR steam generator design improvements.
S409: Alternate Secondary Water Chemistry. The objective of this project

was to develop an alternative or several alternatives for chemistry control
of the secondary cycle. The longer-term goal of such a program will be a
chemistry control system that overcomes the weaknesses of the present AVT
system and is compatible with all-secondary system construction materials
and methods of operation. Work items include:
* Development of improved water chemistry control schemes for:
(1) more effective secondary cycle pH control, (2) improved corrosion
product management, (3) improved oxygen scavengers, (4) improved
IGA inhibitors, and (5) improved layup additives.
* Performance of materials qualification testing for candidate alternatives.
o Effectiveness testing and breakdown product analyses for candidate
alternatives.
S410: Tube Vibration, Mechanical Damage, Fretting, and Wear. A steam

generator tube wear predictive methodology for straight, multi-span tubes
at the entrance region of the tube bundle was developed and partially
verified experimentally during SGOG II. It has been successfully applied
to the so-called "cold-side wastage" problem of the Westinghouse Model
51 steam generator. However, steam generators were suffering significant
tube wear in the U-bend region. The objective of this program was to extend
the methodology for application in the steam generator U-bend region.
Additionally, it could be verified for continued application in the straight
leg portions of the steam generator. Upon completion of this program,
tube wear rates can be predicted over a range of tube-to-tube support plate
and AVB clearances.
S411: Thermal-Hydraulic Code Evaluation and Validation. The objective

was to develop, verify, and validate an advanced method for three-
dimensional simulation of two-phase flow and thermal hydraulics in all
types of steam generators for accurate calculations of non-homogeneous,
thermal, non-equilibrium flow fields. This methodology takes into account
the detailed geometry configuration inside the steam generators for utility
application in the areas of thermal-hydraulic performance evaluations, flow-
induced vibration analysis, and detailed steady state and transient flow
field calculations.
S412: Steam Generator Strategic Planning Guidelines. The objective

of this program was to develop guidance for utilities in all aspects of steam
generator management and the associated strategic planning required to
provide each PWR operator with the required options to optimize steam
generator reliability. This work included documentation of past repair/
replacement decisions, existing long-term steam generator management
approaches, and the capabilities, limitations, and potential benefits of and
problems with degradation progression models.
RESEARCH FINDINGS
The various steam generator programs cover a variety of topics that
are interdisciplinary in nature addressing the causes of damage forms
(principally tube damage), their avoidance and mitigation, as well as
damage mechanism management. The research results have been
documented in technical publications, the first edition of the Steam
Generator Reference Book, and in a number of guidelines. Appendix A in
the Reference Book contains a listing of the various SGOG I and II and SGRP
technical publications, cross-referenced by report numbers (A. 1) and project
numbers (A.2). Appendix B contains a listing of plant characteristics. The
guidelines are described in Chapter 4. The following discussion presents a
brief summary and overview of the results categorized by steam generator
tube damage form.
Tube Wastage (Thinning)

Tube wastage or thinning was one of the first types of corrosion
occurring in recirculating steam generators that operated on sodium
phosphate as a secondary water treatment.
As recognition grew that this new phenomenon was widespread and
of a generic nature, the two U. S. vendors of recirculating-type steam
generators launched major laboratory investigations to better understand
this phenomenon. They demonstrated in pot boiler tests that the location
of attack was related to the concentration of aggressive species at steam/
water interfaces.
As a result of these findings, the vendors recommended that the
operators in the affected plants adopt an all-volatile water treatment (AVT)
based on the use of ammonia and hydrazine. This recommendation was
universally followed in the United States except for two plants.
However, this change led to another corrosion problem, namely denting.
Denting
Denting was discovered in 1975 when eddy current probes were
prevented from passing through tube/tube support plate intersections by
tube diameter restrictions. By 1977 denting had become a widespread
problem and resulted in the formation of the Steam Generator Owners
Group as a concerted effort to address the problem.
Sample intersections of tubes and support plates removed from dented
steam generators have shown local chloride concentration of over 4,000
ppm in the dented region. The high local chloride concentration is caused
by local thermal-hydraulic conditions within the crevice between the tube
and the tube support plate. The source of chloride is generally condenser
leakage, particularly at plants cooled by sea water.
To correct this situation it is recommended that plants modify the water
chemistry environment to make it less aggressive. To help guide utilities in
this area, secondary water chemistry guidelines were established. The
guidelines which are based on laboratory and field data, recommends
reduced levels of impurities in the steam generator and methods for
achieving these greatly lowered impurity levels. This includes reducing
condenser inleakage, producing purer makeup water, and using condensate
polishers to purify the water. The environment can also be modified by
adding boric acid, which inhibits the acid chloride attack.
For new plants, one vendor minimized the potential for formation of
an aggressive environment by switching from drilled to broached support
plates with flat lands. Another vendor switched to a lattice bar support
design. With these designs, the concentration of chemicals within the tube/
tube support crevice is greatly reduced. For new plants, the corrosion
potential was further reduced by installing more corrosion-resistant
materials.
In summary, denting, caused by the corrosion of carbon steel, is
primarily resulting from a concentration of impurities in the tube/tube
support crevices. The corrective actions include:
" modifying the environment by reducing, inhibiting, or neutralizing the

impurities in the steam generator,
* using a tube support design that reduces the concentrating mechanism,
and
" using more corrosion-resistant materials.
Primary Side Stress Corrosion Cracking

Primary side stress corrosion cracking (PWSCC) of alloy 600 steam
generator tubing evolved from a laboratory prediction during the 1950s
and 1960s to a major degradation mechanism of the 1970s and 1980s in
operating steam generators.
The first confirmed primary side cracking of hot leg roll transition
regions at the tubesheet and suspected primary cracking in U-bends were
confirmed in 1971. U-bend leakage was experienced in Obrigheim steam
generators, manufactured by Kraftwerk Union, after only two years of
operation.
In recent years, cracking of alloy 600 tubes from the primary side is of
increasing importance. Cracking in the U-bend has occurred mainly on the
inner row at the apex and at the tangent points. Cracks have also occurred
in the tubesheet in transition-expansion and roll expansion regions,
particularly when tubes have been mechanically expanded into the
tubesheet. Primary side cracking is a form of intergranular stress corrosion
cracking (IGSCC). This type of IGSCC, like other forms of stress corrosion
cracking, occurs when tensile stress, certain environmental, and material
susceptibility factors are sufficiently severe. Alloy 600 tubing with a low
final mill anneal temperature is particularly susceptible. A reduction of
temperature in the inner row U-bends is considered to be an ameliorative
measure (lengthening initiation time) against stress corrosion cracking. For
operating plants and for plants already built but not operating, reduction
of tensile stresses by stress relief heat treatment and peening of the inner
diameter (ID) surface are the most practical approaches.
Experience has shown that shot peening and stress relief are particularly
effective if applied before operation. If applied after startup, already initiated
cracks will continue to grow.
Sleeving is a repair method. Sleeves up to 44 inches long have been
installed in the region between the lower face of the tubesheet and the first
support plate as a corrective measure against pitting and intergranular attack
at the outer surface. In some cases, even longer sleeves have been installed.
The sleeves bridge the damaged area and are attached to an unaffected
material beyond either end of the damage. The sleeve ends are expanded
hydraulically or explosively and are in most cases sealed by rolling, welding,
or brazing.
In more recent plants, thermally treated alloy 600 tubing has been
installed. Extensive tests have shown this material to be superior to mill
annealed alloy 600 with respect to PWSCC, but in some cases it is still
susceptible to cracking. More recent replacement steam generators are being
equipped with thermally treated alloy 690 and alloy 800 (modified). These
alloys have been suggested to be superior to thermally treated alloy 600.
Secondary Side Intergranular Attack and Stress Corrosion Cracking

Many steam generators operating with seawater or freshwater cooling
have experienced one or several forms of intergranular corrosion. In some
cases, progression has been sufficiently rapid torrequire mid-cycle
inspections and unscheduled outages to plug or repair leaking tubes, which
has caused significant economic loss to the PWR operator. Fortunately,
most of the corrosion has been confined to crevice locations, so leaks have
been small and without risk of a large rupture. Intergranular corrosion has
occurred as intergranular stress corrosion cracking (IGSCC) (single or
multiple cracks), intergranular or general attack (IGA), and intergranular
penetration (IGP), which is a combination of the other two forms.
Within the steam generator, intergranular corrosion has been found in
the tube/tubesheet crevice (an annular gap remaining after steam generator
manufacture), and the tube/tube support plate crevice. Intergranular
corrosion at the latter location has been found at several Japanese sites and
has been occurring with increasing frequency in the United States and
Europe. Corrosion has also been observed in the sludge region above the
tubesheet.
Tests have shown how aggressive chemicals can concentrate in these
crevices and in the sludge under heat transfer conditions.-Alternate
wetting and drying is a particu.larly effective concentrating mechanism.-
Six classes of environmental contaminants have been postulated to explain
the occurrence of intergranular corrosion. These include:
" high concentration of sodium and/or potassium hydroxide,
" products generated by the reduction of sulfate ions with hydrazine or
hydrogen (reactive sulfur-bearing species are postulated),
o products of thermal decomposition of ion exchange resins and organic
inleakage chemicals (sulfates, chlorides, etc.),
o highly concentrated salt solutions at neutral or nearly neutral pH (These
salt solutions are the natural consequences of condenser leakage
concentrated to high levels of salt by the boiling processes in the steam
generator.),
o alkaline carbonates and sulfates and/or their reaction or hydrolysis
products (believed to affect the nature of the passive film on the alloy
surface), and
* lead and lead compounds (Tests have shown that concentrations of

caustic and lead can crack all tubing alloys.).
It has been shown with reasonable certainty that increased service stress,
dynamic strain, and a high residual stress level can all be major factors in
accelerating initiation or propagation of IGSCC. A similar statement for
IGA can not be made since experimental results have shown that IGA can
develop even with compressive stress.
For new plants, modification of the material by thermal treatment, to
improve the microstructure can reduce the susceptibility to caustic-induced
IGSCC. For operating plants, sleeving can be applied to mitigate the effect
of intergranular corrosion.
Modification of the crevice environment appears to be the most straight-
forward method for preventing or arresting IGA and may apply equally to
IGSCC. Modification can include several factors such as lowering
temperature, adding a pH neutralizer, removing the corrodent by flushing
or soaking, changing the concentration and/or ratio of bulk water
contaminants, etc. Addition of boric acid has been shown to be effective in
the laboratory in preventing caustic-induced IGSCC, but the results of its
use in a number of operating steam generators has not been clearly effective.
Several plants have applied flushing or soaking off-line maintenance
procedures with varying success. The flushing procedures are based on
laboratory tests and involve a depressurization that causes boiling within
the tube/tubesheet crevice and injection of concentrated solutions from the
crevice. It should be recognized that flushing or adding a pH neutralizer
may be difficult if denting has occurred at the top of the tubesheet or the
annulus is fouled with corrosion products, thus blocking access to the
crevice. The concentration of contaminants accessible to the crevice may
be controlled by eliminating or reducing ingress of contaminants to the
steam generator and by controlling the concentrating capability of the sludge
pile above the crevice. Control of the sludge pile, which is an effective
concentrating mechanism, suggests several courses of action:
* effective periodic sludge lancing;
" minimizing particulate transport by preventing air ingress and/or
providing for feedwater filtration such as by powdered resin condensate
polishers;
" addition of chemical products (e.g., amines) which raise the pH of water
flowing in single and two phase flow areas of the secondary systems,
thereby reducing general corrosion rates; and
* preventing the ingress of chemical species that tend to promote
agglomeration.
Chemical cleaning has been employed to remove the sludge on the

tubesheet. One group of utilities has attempted to modify the aggressive
environment in the tube/tubesheet crevices by fully expanding the tubes
in the tubesheet, thus eliminating the crevice.
Pitting
Extensive pitting on the outer surfaces of tubes has been observed in
several units. The pitting. occurred primarily on the cold leg between the
tubesheet and the first support plate in regions where sludge or tube scale
was present. It is judged that the pits are caused by chlorides, low pH, and
an oxidant such as CuC12 or oxygen. Temperatures >150'C (300'F) are
required to form pits like those observed in operating units. It is further
believed that sludge and scale act as a medium where bulk impurities can
concentrate to higher levels by the boiling action.
For existing plants, the tubing is repaired and retained in service by the
use of sleeves. The principal corrective action is to modify the environment
to make it less corrosive. Sludge and scale can be reduced by minimizing
the ingress of solids (reduced air inleakage), by sludge lancing, and by
chemical cleaning. All of these methods are now being employed. Chemical
cleaning has been effective in arresting pitting in one plant. In addition,
the environment will become less aggressive by minimizing the ingress of
soluble contaminants (principally chlorides and oxidants).
Corrosion Fatigue
Some tubes have developed cracks on the outer surface in the upper
region of several OTSGs. It is believed that these cracks are caused by
corrosion fatigue resulting from small-amplitude vibration combined with
the transport of impurities into the upper regions of the OTSG units,
particularly in the open lane (a row without tubes, to provide for inspection).
Laboratory tests have shown a decrease in alloy 600 fatigue strength in the
presence of chemicals that are judged to be present in these upper regions.
In these tests, a substantial decrease in fatigue resistance of alloy 600 was
observed in acid sulfate/silicate solution.
The environment must be modified and made less aggressive by
minimizing the ingress of impurities into the steam generator. Also,
mechanical flow-diverting lane blockers have been tested and shown to be
a useful steam generator modification; they reduce the flow of liquid
droplets to the upper regions via the open lane. In new plants the open
lane has been eliminated.
High-cycle fatigue cracking has also been observed at the top support
plates in at least three recirculating steam generators and at similar locations
3.26 Steam Generator Reference Book, Revision I
in CANDU steam generators. This type of cracking is due to the vibration

of the U-tubes. (The cross flow caused the tubes to vibrate.) In several
cases, the tendency to crack was increased by denting at the top support
plate. Denting increased the mean stress and reduced the tube damping.
This, in turn led to increased vibration due to the absence of AVB support
and an increase in alternating stress.
Thermal hydraulic and structural codes have been developed to predict
tube motion and resulting stresses. With the aide of these codes the
following corrective measures can be recommended: reduced flow velocities
and redesign of the antivibration bars to provide increased support and
reduced gap between the tube and antivibration bars. Chemical cleaning
or soaking to reduce the corrosive environment around the tube and top
tube support plate crevice is another potential corrective action.
Tube Fretting and Wear

Tube fretting and wear has been observed in the U-bend region of
recirculating units. This is attributed to the design of the antivibration bars
which were in point contact with the tubes and also to out-of-tolerance
clearance between the bars and tubes. The corrective action has been to
use wider bars designed to reduce the clearance between the bars and the
tubes.
Excessive tube wear was found in split-flow preheat units after short
periods of operation. The wear was caused by tube vibration induced by
high flows and unduly small impingement baffles. The situation was
corrected by reducing the flow to the main nozzle and by installing a new
impingement baffle, which distributed the flow more uniformly around
the bundle.
Wear was also found in the first two to three rows of several counter-
flow preheat units in the entrance region. The affected tubes in the periphery
were expanded at several of the tube support locations to reduce the
vibration amplitude and resultant wear by reducing the annular clearance
between the tube and the inside diameter of the tube support plate hole.
Analytical codes have been developed which can predict tube motion.
Furthermore, extensive tests have been completed which measure the wear
rates under various contacting conditions between the tube and tube support
plate (sliding and impact). With the use of these analytical techniques and
wear data, wear rates can be predicted with reasonable accuracy, provided
operating conditions and tube-to-antivibration bar clearances are known.
Steam Generator Performance

Steam generator performance has been monitored and documented in
annual reports by the Steam Generator Owners Groups and by the Steam
Generator Reliability Project (3.1, 3.2 3 3.44 3 3 3 39 3.10).
Topics such as lost generation, forced outages, lists of plants reporting
problems, tube sleeving, and steam generator replacement are covered.
Some of this information is included below.
Steam Generator Replacement/Service Life. One of the principal benefits

derived from the research is the extension of steam generator life. When
the SGOG first started, denting was widespread and steam generators were
being replaced after 7-10 years. Hence, it was envisioned that several steam
generator replacements would be required over the approximate life of the
plant, which was estimated to be 40 years. However, as the results of the
research and lessons learned were applied, steam generator degradation,
while still serious and continuing, slowed down and the life of subsequent
replacements gradually increased (Table 3-3). As of February 1991
(Table 3-4), more than 35 plants have operated for more than 15 years (8
plants more than 20 years). Thus, it now appears likely that no more than
one steam generator replacement will be required over the life of the plant.
Lost Capacity
Figure 3-4 shows the capacity factor loss by year in domestic plants
due to steam generator degradation. The average loss over the period 1980-
1984 was 5.7%. Assuming a conservative replacement power cost of 3-40/
kW/hr, this represents an average annual loss of $186 million. The target
for the SGRP was to reduce this loss from 5.7% average to 3%. The figure
shows that the steam generator performance in 1989 was slightly behind
the target (3.2%). Still this resulted in a conservatively estimated annual
savings of $85 million.
Forced Outages
Forced outages is another measure of performance. Utilities are
experiencing a constantly decreasing rate of forced outages, which is due
to factors such as better understanding of the problems, improved corrective
actions, and more extensive and reliable in-service inspection.
TABLE 3-3
Steam Generator Replacements
Unit Year of Completion Years of Service

Surry 2 1980 7
Surry 1 1981 8
Turkey Point 3 1982 10
Turkey Point 4 1983 9
Obrigheim 1983 14
Point Beach 1 1984 13
H B Robinson 1984 13
D C Cook 2 1989 11
Indian Point 3 1989 13
Ringhals 2 1989 14
Dampierre 1 1990 10
Palisades 1991 19
Millstone 2 1993 17
Bugey 5 1994 13
Doel 3 1993 11
North Anna 1 1993 15
Beznau 1 1993 24
Mihama 2 1994 18
TABLE 3-4
Steam Generator Performance
As of 2/91, 35 plants have operated 15 calendar years or more.
Calendar Years of
Unit Mo/Yr Commercial Service (tO 2/91)
Yankee Rowe 7/61 29.6
Chooz A 4/67 23.8
Connecticut Yankee 1/68 23.1
San Onofree 1 1/68 23.1
Jose Cabrera 2/69 22.0
Beznau 1 9/69 21.4
R. E. Ginna 7/70 20.6
Mihama 1 11/70 20.2
Beznau 2 12/71 19.2
Stade 5/72 18.7
Mihama 2 7/72 18.6
Point Beach 2 10/72 18.3
Maine Yankee 12/72 18.2
Borssele 6/73 17.7
Oconee 1 7/73 17.6
Prairie Island 1 12/73 17.2
Zion 1 12/73 17.2
Ft. Calhoun 6/74 16.7
Kewaunee 6/74 16.7
Indian Point 2 8/74 16.5
Oconee 2 9/74 16.4
Three Mile Island 1 9/74 16.4
Zion 2 9/74 16.4
Takahama 1 11/74 16.2
Nuclear One 1 12/74 16.2
Oconee 3 12/74 16.2
Prairie Island 2 12/74 16.2
Biblis A 2/75 16.0
Doel 1 2/75 16.0
Calvert Cliffs 1 5/75 15.7
D.C. Cook 1 8/75 15.5
Tihange 1 10/75 15.3
Genkai 1 10/75 15.3
Takahama 2 11/75 15.2
Millstone 2 12/75 15.2
FIGURE 3-4
Capacity Factor Loss in the U. S. Due to Steam Generator Problems
0
Oi
C)
8 7.2
A 7 673
S6 5.152
L - 3.5 3.4I 3.2 3.229 3.3
, *i.I>
.
,__i
a. 37 I9 2 2iIE
2 1.9 0"
0 0
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93
Year
REFERENCES
3.1 "Steam Generator Progress Report, Revision 1," September 1985
3.2 "Steam Generator Progress Report, Revision 2," December 1986
3.10 "SteamGenerator Progress Report, Revision 10," December 1994
4
RECOMMENDATIONS
Contributing Authors/Editors:
J. P. N. Paine,EPRI
A. R. Mcllree, EPRI
C. S. Welty, EPRI
D. A. Steininger,EPRI
C. L. Williams, EPRI
S. A. Hobart,Adams & Hobart
INTRODUCTION
This chapter provides the recommendations developed as a result of
Steam Generator Owners Group (SGOG) and Electric Power Research
Institute (EPRI) programs. Supporting information is presented in other
parts of this document. The recommendations are presented as a set of
options that a utility should consider when undertaking a program to
optimize PWR steam generator reliability. Only those options that meet
the needs of a particular power plant should be adopted. It should be
emphasized that a particular plant's needs are based on many considerations
such as: plant size and equipment; plant age and operating history; site
characteristics, such as cooling and makeup water sources; and
administrative factors, such as utility management structure, resource
allocation priorities, cost-benefit methodology, support staff size, and public
utilities commission (PUC) interactions.
This chapter is divided into four sections: 4.1, "Steam Generator Design,
Modifications, and Repairs"; 4.2, "Steam Generator Operations"; 4.3, "Plant
Design, Modification, and Operation"; and 4.4, "Abstracts of SGOG
Published Guidelines". Sections 4.1, 4.2, and 4.3 integrate and summarize
by topic the primary recommendations from all sources, i.e., workshop
conclusions, research results, and published guidelines. Section 4.4 provides
a brief abstract of each of the published SGOG guidelines documents
(Guidelines for Tubing Specifications and Repair, Nuclear Plant Designand
Modification Guidelines for PWR Steam Generator Reliability, Preoperational
Practicesfor Steam Generators and Secondary System Components, Design and
OperatingGuidelinesfor NuclearPowerPlantCondensers,Design and Operating
Guidelinesfor CondensatePolishersin NuclearPower Plants.Guidelinesfor Design
of PWR Steam Generator Chemical Cleaning Systems, PWR Steam Generator
Inspection Guidelines, PWR Secondary Water Chemistry Guidelines, Guidelines
for the Designand Operationof Makeup Water Treatment Systems, and Guidelines
for Impurity Removal and Control) (4.1-4.10). An abstract of the PWR Primary
Water Chemistry Guidelines (issued by the Materials and Chemistry
Department of EPRI) is also included for completeness (4.11). The available
guidelines documents were issued to SGOG members at the end of the
SGOG program. Later revisions are available as EPRI reports. They provide
a significantly greater level of detail than this chapter, as well as technical
justification for recommendations, and are intended to be used in
conjunction with the Steam GeneratorReference Book. As an aid to identifying
specific recommendations in the various publications, a computerized
database of recommendations has been prepared. Copies of this database
and the user manuals are available from the Steam Generator Project Office.
Recommendations 4-3
4.1 STEAM GENERATOR DESIGN

This section summarizes recommendations for steam generator design
features (e.g., configuration and materials). Information presented can be
used as non-plant-specific criteria in procurement documents for new or
replacement steam generators. Selected recommendations may also be
implemented as modifications for existing steam generators. Research
results and field experience, where available, indicate that adoption of these
recommendations will decrease susceptibility to corrosion and mechanical
damage. Thus, implementation of the recommendations should produce
significant increases in steam generator reliability and operating life.
The information in this section is the only comprehensive compilation
of SGOG recommendations regarding steam generator design. Some have
been extracted from research reports, guidelines documents, and
workshops. However, many have been developed specifically for this
publication from evaluation of data produced by SGOG and EPRI research
projects (both published and to be published). Information supporting this
second type of recommendation is presented in Chapters 5, 17, 24, 26, and
27. References to specific paragraphs in Section 4.4 (abstracts of published
SGOG guidelines) and later chapters of this book are located after each
subsection title. Table 4-1 gives a listing of subject areas from this section
and the reference documents that provide related guidance.
It should be noted that all recommendations related to heat treatment
of tubing refer implicitly to the use of alloy 690. Although alloy 800 is
mentioned as a possible alternative tubing material, its consideration is not
widespread enough in the United States at this time to merit parallel
discussions in these recommendations.
4.1.1 Tube and U-Bend Support Structure Materials

(Chapter 24)
Recommendations
* The choice of support material should be based on corrosion data for
site-specific, off-normal chemistries, multiple chemical cleanings, and
wear data. As a minimum, location at seawater versus freshwater sites
should be considered in evaluating appropriate corrosion data and
selecting materials.
* The tube support structure in the straight leg portion of the bundle
should be constructed of ferritic or austenitic stainless steel or of
austenitic nickel alloy. Experience included in a substantial database
suggests that austenitic stainless steel and 12% chromium ferritic steel
be given first consideration. A much more limited data base suggests
that alloys 600,690, and 800 should be considered, especially at seawater
sites.
For U-bend supports (i.e., anti-vibration bars [AVBs]), consideration

may be given to a low-alloy steel. Oxides from low-alloy steels have
been found to be more lubricating than those from stainless steels.
Laboratory tests of this type material resulted in lower tube wear rates
than stainless steels. However, the potential for support structure
corrosion should be weighed against the potential for wear when a
choice of materials is made. A possible compromise may be a ferritic
stainless steel (e.g., 405/409), which allows a thicker stable oxide layer
to form than does an austenitic steel.
4.1.2 Tube Support Structure Designs

(Chapter 24)
Recommendations
" Tube supports should be designed to ensure that continuous dryout
patterns do not occur at tube-to-tube-support intersections,
susceptibility to fretting and wear is minimized, and fatigue is
precluded. Computer programs such as ATHOS, PORTHOS, and
THIRST should be used for these evaluations.
-Straight or convex support surfaces, such as those in grid or broached-
hole designs, can minimize the crevice areas that are subject to dryout.
However, convex surfaces have the potential for accelerating tube wear,
and broached holes may be susceptible to fouling and plugging.
-The designer should demonstrate (e.g., by modeling or experiment)
minimal susceptibility to corrosion, fretting, and wear, and consider
potential concentration factors for corrodents of interest in both clean
and fouled conditions.
-Thermal hydraulic and structural analyses should be performed to
predict wear and tube stability at the end of steam generator life.
" The following design criteria should be considered, as a minimum, to
accomplish the recommended corrosion, fretting, and wear design
objectives. Other considerations should include seismic and design
objectives.
Support Plates:
-Tube holes should be configured to maximize axial flow in the area
around the tube circumference, especially near points of possible tube/
tube support plate contact. Drilled tube support plate holes are not
acceptable; grids or broached holes should be used.
0
-Tube support plate hole clearances should be small enough for the
plate to act as a true support structure for a tube undergoing vibration.
The support plate should represent a structural node for the tube after
all tolerances of design, manufacture, mechanical and chemical
Recommendations 4-5
cleanings, and operational conditions are considered in determining

the relative position of tube and support plate. These considerations
are especially important for tubes that experience cross flow, e.g.,
locations near the entrance, steam outlet, or U-bend region of a steam
generator.
Support Grids:
-- Grids should present a flat surface to the tube.
-Grid tolerances should be small enough so that the intersection is a
structural node in all conditions (see "Support Plates" above).
Flow Distribution Plates:
-Holes should be configured to maximize axial flow around the tube
circumference, especially near points of possible tube/flow plate
contact, yet maintain the desired pressure drop.
-Drilled holes are not acceptable; IGA/SCC has been observed on
tubing at flow distribution plate holes at several sites.
4.1.3 Tube Materials

(Sections 4.4.1 and Chapter 24)
Recommendations
* Tube material and its manufacturing method should be selected to
maximize resistance to corrosion.
-Techniques of heat treatment should be selected to provide a grain
boundary structure resistant to SCC. Specifically, grain boundaries
should have a continuous carbide boundary.
-Techniques of heat treatment, bending, and expanding should be
utilized to minimize geometric discontinuities and residual tensile
stresses on the tube ID and OD surfaces. The tubes should have minimal
eddy current noise so that in-service defect detection and sizing is
maximized.
* Alloy 600 and austenitic stainless steels are not recommended for steam
generator tubing (4.,1 Volume 1).
* Alloy 690 tubing in the thermally treated condition is the preferred
material. It has been proven equal to or superior to alloy 600 in all
laboratory tests to date. However, steam generator operating experience
with alloy 690 is minimal. A material specification is provided in EPRI
final report NP-6743-L, Volume 2 (4.1).
" Titanium-stabilized (Ti/C>12) alloy 800 tubing may be considered for
use. It has been successfully used in both German and Canadian nuclear
steam generators. Siemens (Germany) has specified a tight control of
composition of alloy 800 (specifically, Ti/C>1 2) to stabilize the material
for resistance against grain boundary chromium depletion
(sensitization). They call this material "modified" alloy 800. In addition,

the tubing is cold-drawn to increase the yield strength and can be OD
shot peened to-increase resistance to secondary side SCC. German
(KWU) steam generators have been tubed with alloy 800 since the 1969 0
startup of the Stade plant. As of 1978, tubing with a shot peened OD
surface has been used. Alloy 800 has superior corrosion resistance to
alloy 600, with the exception of exposure to concentrated caustic and
to chloride under oxidizing conditions. Based on the most current
laboratory data available at the time of tube specification, each utility
should select the best alloy for their particular plant chemistry history.
4.1.4 Tube Fabrication-U-Bends

(Chapter 24)
Recommendations
* U-bend fabrication and heat treatment techniques should yield finished
tubing that provides maximum resistance to SCC. The preferred
fabrication and heat treatment techniques, including methods for
minimizing residual stresses, are described in NP-6743-L, Volume 2 (4.1),
which specifies a post-bending thermal treatment to reduce residual
stresses. The applicable codes (ASME, ASTM, ANSI, etc.) should be
followed.
* Cross-sectional shape in the bend region should be maintained within
strict limits of ovality; the bend, including transition to straight sections,
should be smooth and variations in wall thickness should be minimized.
-- Ovalized and deformed tubing, especially in the transition to straight
sections, is subject to large local tensile stresses during operation and
has been found to contribute significantly to susceptibility to SCC in
alloy 600 (4.1 Volume 1).
-Ovality can lead to unacceptable tube and support structure
clearances (Section 4.1.6).
4.1.5 Tube Fabrication-Tubesheet Expansion

(Chapter 24)
Recommendations
* Tube-to-tubesheet intersections should be designed to provide
maximum resistance to primary and secondary SCC and IGA, The
applicable codes (ASME, ASTM, ANSI, etc.) should be followed.
* All crevices between the tube and the tubesheet should be eliminated.
The only option currently provided by steam generator suppliers is to
expand the tubes to contact the tubesheet. Care should be taken to
prevent expanding the tube beyond the secondary face of the tubesheet,
Recommendations 4-7
in order to minimize geometric discontinuities in the tube wall which

can result in high residual stresses. Field experience suggests that
hydraulic and explosion methods are preferred methods. Care should
be taken to properly apply any tube expansion process used.
-Tubesheet holes should be drilled to the minimum oversize
requirement for fabrication to minimize expansion stresses. Tubesheet
hole profile irregularities should be minimized to preclude associated
stresses.
-The expansion method should be qualified to demonstrate acceptably
low residual tensile stresses, hence improved resistance to SCC. The
expansion technique should be evaluated to ensure that it will not
degrade the material's resistance to SCC. The expansion technique
should be qualified in both production-sized and maximum-allowed
oversized tubesheet holes.
-Hydraulic expansion should be followed by a hard roll well below
the top of the tubesheet and the expansion transition, to eliminate the
possibility of residual crevices.
-After fabrication of the tube bundle, 100% of the tubes should be
inspected by profilometry and eddy current testing (ECT) in the
expansion zone. ECT should be performed according to the NDE
guidelines (4.8) for baseline inspections.
4.1.6 U-Bend Support Structure

(Chapters 5 and 12)
Recommendations
o Tube supports should be designed to ensure that continuous dryout
patterns do not occur at tube-to-tube-support intersections and that
susceptibility to fretting and wear is minimized. The following design
criteria should be considered to accomplish the recommended design
objectives.
-Supports should present a flat surface to the tubes at the point of
contact.
-Supports should not impede water or steam flow around points of
contact.
o Wear qualification tests between tube and support material should be
carried out with suitable design geometries and in suitable, faulted
water conditions. For these tests, net clearances between tube and
support should be designed to maximize wear.
4.1.7 Steam Generator Tube Bundle Assembly

(Chapter 24)
Recommendations
* Tube alignment and fitup should be tightly controlled to ensure that
unacceptable stresses are not induced and vibration will not be a factor
introducing wear or fatigue.
* Tubesheet hole alignment with the first tube support elevation should
be verified to eliminate bending stresses on the tubes.
* Performance of a full-unit stress relief at the completion of all assembly
and welding is highly desirable to ensure that residual stresses have
been reduced to acceptable levels. If a full unit stress relief is not
performed, then a program of local stress relief (e.g., of U-bends) and
fabrication qualification tests (e.g., for expansions and expansion
transitions) is needed to ensure that residual stresses will not aggravate
primary or secondary side SCC.
" A program should be designed to demonstrate tubing and tubesheet
cleanliness prior to installing tubes within the tubesheet (Section 4.2.1),
especially the removal of shot or other debris from shell ID surface
preparation activities.
" Care should be taken during assembly to prevent scratching the tube
surfaces. Scratches have been shown capable of acting as initiation
sites for SCC.
4.1.8 Mechanical Design-Penetrations

(Sections 4.4.2, 4.4.8, Chapters 24, 26, and 27)
Recommendations
* Appropriate design features should be included in new steam
generators, both OTSGs and RSGs, to facilitate inservice inspection and
testing. Items to be considered include number and location of access
ports and covers, tube and tubesheet fabrication criteria, and insulation
design.
" Suggested secondary side penetration considerations are presented
below.
-The steam generator should include sufficient and appropriately
placed shell penetrations to allow water recirculation and nitrogen
sparging during wet layup.
-The steam generator design should include provisions for complete
and rapid shell side drainage and simultaneous nitrogen over
pressurization at temperatures of up to 250'F (121'C).
-A sufficient number of personnel access ports should be incorporated
so that access is available to any untubed region.
Recommendations 4-9
-Handholes should be present to allow access to tubed regions at each

support elevation under inner row U-bends, and to any untubed lane
in the tube bundle.
-The shell should contain openings that would support full bundle
chemical cleaning and easy and effective sludge lancing of each support,
if found necessary, and optical (photographic or television) inspection
of the bundle.
-Access ports of appropriate number and size should be provided
near the top of the tubesheet to permit inspection and retrieval of loose
parts.
-- Containment and near-containment equipment and facility design
should incorporate features to support the service connections and
storage/office requirements of steam generator inspection and testing
(4.3).
" Steam generator design should consider in-place primary side
decontamination provisions (4.3).
* Insulation around access ports and penetrations should be designed so
that it is easily removable for ECT or sludge lancing. Enough insulation
should be removable so that the possibility of insulation damage or
contamination is prevented.
4.1.9 Mechanical Design-Steam-Water Separation Equipment

(Chapter 24)
Recommendations
* Minimum moisture carry-over is desirable to minimize corrodent
transport to the turbine. Current vendor specifications for moisture
carry-over in the steam are "not to exceed" 0.25% of the steam flow. It
is not clear from available data that 0.25% is low enough to minimize
turbine corrosion (0.10% maximum is achievable with state-of-the-art
designs).
* Moisture separators should have sufficient capacity margins to permit
modification of the recirculation ratio and power rating upgrade at a
future date.
* The primary separators should be designed to drain away water
immediately after separation occurs within the separator and return it
through low-pressure drop paths to the downcomer. The interstage
height between the primary separator outlet and the dryer entrance
should be sufficient to take advantage of gravitational separation and
therefore reduce water loading on the dryers.
" The dryers should be designed to present a clear flow approach for the
incoming steam/water mixture and provide efficient drainage of
separated water to the downcomer through low-pressure drop paths.
The vapor flow rate to the dryers should be uniform across the dryer
entry face in order to avoid preferential loading of some areas, which
would result in decreased separation efficiency. i
4.1.10 Mechanical Design-Blowdown

(Chapter 24)
Recommendations
* OTSGs should be designed to permit blowdown in the low power
recirculating mode (<15% power).
* Blowdown provisions within the steam generator, both OTSGs and
RSGs, and the supporting external system should include the following:
-sufficient capacity (1 wt% main steaming rate [MSR], minimum) such
that rapid steam generator cleanup can be performed;
-features such as flow distribution plates to maximize the fraction of
the particulates carried to the blowdown header, and
-blowdown header hole locations and header and nozzle sizes and
materials such that effective high capacity blowdowns for particulate
removal can be performed.
* Blowdown provisions for RSGs should be able to accommodate a
periodic blowdown of 3-7 wt% MSR for short durations (two to five
minutes), for removal of tubesheet sludge (see operating
recommendations in Section 4.3.3) (4.3).
* The blowdown nozzles should be sized to accommodate the blowdown
flow rates and system design described in Section 4.3.3.
* Particulate erosion may occur in the blowdown header if a proper choice
of materials is not made. Materials high in chromium content and with
known erosion corrosion resistance should be chosen for the design.
4.1.11 MechanicalDesign-Feed Ring (Chapter 24)

Recommendations
* The feed ring should be vented at the top with sufficient flow area to
relieve differential pressures that might result if vapor were condensed
internally.
• J-tubes should be considered in order to prevent the feed ring from
draining if the steam generator water level drops below the feed ring
level.
0
* The four damage mechanisms that should be considered in feed ring
design and fabrication are erosion corrosion of the feed ring, water
hammer damage of the feed ring, thermal fatigue cracking of the steam
generator feed nozzle, and thermal fatigue of the steam generator shell
at the feed ring elevation.
-Erosion corrosion may occur in the feed ring and J-tubes if a proper
choice of materials is not made. Steels high in chromium content and
with known erosion corrosion resistance should be chosen for the
design. Suitable materials include: 2.25% Cr-1 % Mo, 9% Cr-1 % Mo, and
12% Cr stainless steel.
-The feedwater supply line, internal to the steam generator, should
be fabricated of erosion corrosion-resistant steels and be designed to
preclude feedwater pipe and ring drainage resulting from low water
level in the steam generator. This will reduce the likelihood of water
hammer.
-The feed ring piping should be level or sloped downward to the steam
generator shell.
-There are varieties of feedwater supply line alloys available, all
displaying acceptable resistance to erosion corrosion. Research indicates
that 9% Cr 1% Mo is a superior alloy for this supply line. Alloy 600
should not be used in a steam generator.
-The tradeoff between erosion corrosion resistance and cost should
be evaluated on a case-by-case basis. Computer codes, such as CHEC
and CHECMATE, should be used to determine the optimum balance
of design, material, and cost.
-The feed line thermal sleeves should maintain leak-tight connections
with the steam generator nozzle in order to preclude thermal cracking.
Possible leak-tight designs include welded joints and seal mechanisms.
4.1.12 Mechanical Design-Shell

Recommendations
* Steam generator design should consider in-place decontamination
provisions and cut lines and lifting bails needed for steam generator
replacement. Connecting lines and supports should be designed for
ease of disengagement (4.3).
* To reduce the probability of girth weld cracking, avoid steels for
fabrication of new or replacement steam generator shells that have an
unfavorable distribution of sulfides; i.e., sulfur content should be
<0.01%.
" The sulfur content of weld metal also should be minimized to prevent
cracking.
* Steam generator shell areas exposed to primary coolant (e.g., channel
heads) should have low cobalt-high chromium weldments (including
weld metal).
4.1.13 Shell Fabrication

Recommendations
* The steam generator pressure shell should be fabricated to standard
boiler code requirements. During the post-girth-weld stress relief anneal
at the tubesheet, the tube bundle should not be heated to a temperature
that induces sensitization or excessive oxidation of the tubing. In
addition, bundle distortion due to local thermal stresses should be
avoided.
* The upper transition area girth weld should be post-weld heat treated
to reduce residual stresses and limit susceptibility to cracking (ASME
code requirement).
* Generally, non-slag-producing weld techniques such as TIG welding
are preferable to slag-producing techniques such as stick welding, due
to the potential impact on operating water chemistry.
* The inside surface of the steam generator shell should be cleaned after
welding to minimize the impact on operating water chemistry.
" Measures should be taken during shell fabrication to prevent debris
from entering the tube bundle region.
4.1.14 Thermal Hydraulic Design (Chapter 24)

Recommendations
* The steam generator design should minimize the potential adverse
effects of the thermal hydraulic processes that take place on the
secondary side including: flow induced vibration with resulting tube
wear, high cycle fatigue with resulting tube failure, boiling dryout with
local corrodent concentration processes, and stagnant flow with
resulting deposition of sludge and potential corrosion of the tubes.
Thermal and hydraulic codes, such as ATHOS or PORTHOS, should
be used to verify flow velocities.
* Flow distribution plates should be designed to ensure that downcomer
flow sweeps across the face of the tubesheet. This will help lower the
possibility of early boiling in the vicinity of the tubesheet secondary
face and sweep corrosion products toward the blowdown line suction
for removal. Boiling on tube surfaces at the face of the tubesheet has
been correlated with sludge deposition.
* Entrance flow velocities into the tube bundle and steam cross flow
velocities at the exit of the steam generator should not exceed values
that result in large amplitude, flow-induced tube vibration. Tube
excitation mechanisms may involve turbulence, vortex shedding, and
fluid elastic phenomena.
4.1.15 Design Against Water Hammer and Dynamic Flow

Instability
Recommendations
Once Through Steam Generators (OTSGs)
* Keep the horizontal run of feedwater pipe entering the steam generator
as short as practical. The optimum configuration is a 90' downward
elbow just outside the steam generator shell.
Recirculating, Non-Preheater Steam Generators
" Incorporate top discharge J-tube nozzles on the feedwater header and
ensure a good seal at the thermal sleeve to minimize water draining
from the header.
as short as possible to minimize the potential size of a trapped steam
pocket. The optimum configuration is a 900 downward elbow just
outboard of the steam generator shell.
" Position the auxiliary feedwater line(s) to enter the main feedwater line
at a point close to the wall of the steam generator.
Recirculating Preheater Steam Generators
o Incorporate positive measures to prevent draining the horizontal
auxiliary feedwater line. These can include:
-upturned auxiliary feedwater discharge line inside the steam
generator to prevent forward draining to header (Westinghouse);
-trap-type water seal (Combustion Engineering); and
-check valves in auxiliary feedwater line.
* Size the auxiliary feedwater piping to be compatible with required flow
rates so that the pipe remains full during water injection.
as short as practical to minimize the potential size of a trapped steam
pocket. The optimum configuration is a 900 downward elbow just
outside the steam generator shell.
* For Westinghouse plants, install temperature sensors on the auxiliary
feedwater piping adjacent to the steam generator to detect steam in
this piping.
* Vent the economizer box (C-E plants) to minimize any steam pockets.
4.1.16 Archive Samples

Recommendations
* Records of material heat number versus tube location should be
provided to the utility.
" Archive samples of tube material, support structure material and all
weldments should be obtained and stored together with detailed records
pertaining to tubing material and steam generator fabrication

procedures and materials. Experience has shown the advisability of
complete records and archive materials to facilitate investigation of
corrosion and mechanical damage that may occur during operation.
" Archive samples of tubing should be supplied and maintained.
Suggested tubing archive samples are as follows:
-both row one and row two U-bends;
-two meters of each heat following thermal treatment, separate from
U-bend pieces; and
-one production sample containing ten tubes expanded in a simulated
tubesheet.
" All archive tubing samples should receive the same baseline inspections
as performed on the tube bundle: ovality and shape of the bends, ECT,
profilometry of the expansion, etc.
• Archive samples of all heats of materials used in steam generator
manufacture (including tube plugs) and an adequate supply of
weldments should be maintained to support, if necessary, chemical
cleaning or investigation of abnormal corrosion or mechanical damage.
4.1.17 Steam Generator Cleanliness (Sections 4.2, 4.4.11,4.4.12, and

4.4.13, Chapters 21 and 23)
Recommendations
* Detailed attention to cleanliness is required to prevent steam generator
contamination that may result in corrosion or mechanical damage when
the steam generator is in operation or in layup. This concern should
include cleanliness in the manufacturing shop, during shipping and
storage, during testing and during various phases of operation.
Section 4.2 presents details of steam generator cleanliness activities
recommended for construction and operation.
* Procedures should be developed for controlling surface contamination
on steam generator interior surfaces.
* Procedures should be developed for cleaning the steam generators at
various periods in construction and operation. The procedures should
include requirements for measuring chemical contamination (S, Pb, Na,
Cl, etc.) by analysis of wipe samples or taking of other samples at
selected locations.
-Sludge lancing or visual inspection of the tubesheet secondary surface
is recommended prior to operation.
-Consideration should be given to pre-operational chemical cleaning
in order to retard the corrosion product buildup in the support plates
and on tube surfaces.
-Sludge lancing after a short time during startup operation with

steaming should be considered to remove any debris carried into the
steam generator from the feedwater train.
4.1.18 Steam Generator Repair and Replacement

(Section 4.4.8, Chapter 27)
4.1.18.1 Alternate Tube Repair Limits

It should be noted that NRC steam generator tube plugging limits may
be developed concurrently with or subsequent to recommendations in this
document. The NRC plugging limits must take preference to the
recommendations in the Steam GeneratorReference Book.
Recommendations
* Site-specific, imperfection-specific revisions of Technical Specification
tube plugging limits should be submitted to the NRC in order to
minimize unnecessary plugging and sleeving of tubes. These submittals
address specific types of imperfections, where the imperfection type
can be identified reliably and leak and burst data for the specific
materials and application are available.
* Approval for alternate tube repair criteria should be sought well in
advance of need, in order to prevent the potentially lengthy NRC review
process from adversely impacting operations.
* Plants that have demonstrated the presence of tubesheet expansion zone
(up to and including the tubesheet roll transition) primary water stress
corrosion cracking (EZ PWSCC) should use information provided in a
generic industry support document (4.12).
-Submittals may propose the use of crack length, rather than crack
depth, for plugging limits (allows some cracks to' be through-wall).
-A safety-related "defense in depth" program should be implemented
and include: using a reduced primary-to-secondary leak rate limit to
enhance leak-before-break conformance; performing enhanced ECT
(rotating pancake coil or equivalent) of 100% of roll transition zones;
plugging tubes with axial imperfections longer than a conservative
plugging limit; and limiting the total number of cracked tubes in service
based on crack size distribution, in order to limit leakage during
postulated accidents.
* Plants that have demonstrated the presence of outside diameter stress
corrosion cracking (ODSCC) at tube support plates should use
information provided in a generic industry support document (4.13).
-Submittals may propose the use of bobbin coil signal amplitude
(voltage), rather than crack depth, for repair limits (may allow some
through-wall cracks).
-A safety-related "defense in depth" program should be implemented

and include: using enhanced in-service inspection (perform eddy-
current inspections of 100% of the tube support plate regions where
ODSCC has been detected); performing selected inspections using eddy-
current to confirm the ODSCC damage mechanism; using a reduced
primary-to-secondary leak rate limit to enhance leak-before-break
conformance; and limiting the number of cracked tubes remaining in
service (determined by leakage limits at faulted loads).
4.1.18.2 Tube Repair Options

Recommendations
" Sleeving should be included as one option in the overall imperfection
management program. Sleeving can be used in place of plugging when
the NRC approves a license amendment for sleeving, based on the
justification that the sleeved tubes can be returned to service with no
loss of strength. It would be prudent to perform a cost-benefit analysis
(including factors such as present value of projected costs and
cumulative radiation exposure to employees and contractors) to
determine whether steam generator repair or replacement is the
preferable option, and whether plugging or sleeving is the most cost-
effective repair.
* Tube plugging (4_1 Volume 3) is typically less expensive than sleeving
and should be considered if
-the plant has a large plugging margin and a slow degradation rate;
-future degradation is anticipated in the tube bundle above the
planned sleeving elevation; or
-fully qualified sleeves are not available.
* Sleeves should not be installed until the qualification testing work is
completed. An exception can be made when sleeves are installed for
evaluation purposes or when the tube plugging margin has been
reached and no other option exists (4.1_, Volume 3).
" Sleeves and their joints (4,1 Volume 3) should have:
-adequate strength to compensate for structural degradation of the
original tube;
-adequate corrosion resistance to survive for the desired remaining
life; and
-adequate leak tightness to meet technical specifications and ALARA
requirements.
* If the original tube material is susceptible to SCC, caution should be
exercised to ensure that the sleeve joints (at the top and bottom of the
sleeves) do not produce unacceptable residual stresses. It may be
necessary to stress relieve them (4_1 Volume 3).
* Sleeves should be long enough to bridge the degraded portion of the

original tube so that joints can be made in sound material (4.1
Volume 3).
* Sleeve material should be resistant to both primary and secondary
environments. Thermally treated alloy 690 has been demonstrated to
be an acceptable material (4., Volume 3).
o The addition of welding filler metal or brazing materials in joints should
be compatible with the original tube and sleeve materials and the
operating environment (41 Volume 3). Autogenous welding processes,
e.g., laser welds, should be considered.
* Inspection procedures should be qualified to ensure that they are
capable of identifying flaws in the sleeve joint and original tube behind
the sleeve (4.1, Volume 3).
o Nickel plating is a repair that can be used to reduce leakage; it imparts
no strength to a cracked tube. The application of nickel plating should
be limited to relatively small cracks. Sometimes this treatment is used
after re-expansion of tubes in the tubesheet in order to protect against
PWSCC.
" Plug material should be resistant to PWSCC. Alloy 600 is not acceptable.
Alloy 690 TT is recommended for all kinds of plugs.
" For plugged tubes with circumferential cracks, the use of a stabilizer
should be considered.
o Utilities with PWR steam generators should institute a tube plug defect
management plan.
-Plug types and locations should be identified.
-Heat pedigree should be identified for all installed plugs.
-The utility and plug manufacturer and/or steam generator vendor
should develop tube plug lifetime predictions.
-Plugs should be inspected by non-destructive examination, or
selective plug removal with destructive examination, according to a
sample plan.
-A plan should be developed to visually inspect plugs for signs of
leaking due to poor installation or cracks.
-Plugs that do not meet pre-established acceptance criteria or are
leaking as noted by visual examination should be repaired or replaced.
4.1.18.3 Tube Peening and In Situ Stress Relief

Recommendations
* Shot peening should be used to inhibit primary side stress corrosion
cracking due to tensile stresses in tubing at the expansion transition
areas. The following precautions should be taken when performing
shot peening (4.14):
-Small-diameter shot should be used to prevent inducing high residual

tensile stresses on the tubing outside diameter (OD), which could
adversely affect resistance to secondary side SCC at the roll transition
and above.
- -Recommended maximum diameters for shot are 0.16 mm for
glass, 0.21-0.30 mm for ceramic, and 0.6 mm for stainless steel. The
mean diameter for stainless alloy powder is considered proprietary
by the vendor, but it is believed to be in the range of 0.2 mm.
- -- Outside diameter (OD) tensile stresses induced by shot peening
should be no more than 20-30 MPa in either circumferential or axial
direction.
- -Inside diameter (ID) stresses induced by shot peening should
be no more than 600 MPa, with the compressive layer extending to a
depth of 0.1-0.12 mm.
-Shot-peened materials should be evaluated with respect to potential
embedment in the tube wall and compatibility of residues with primary
water chemistry and equipment, e.g., bearings and seals.
-Surface indicator chemicals should be evaluated with respect to
compatibility with primary water chemistry and effect on corrosion of
primary system materials. Surface cleanliness criteria should be
established to ensure that no residual chemicals will significantly
contaminate the primary coolant or contribute to tubing ID corrosion.
-The shot peening process utilized should be qualified (by testing)
prior to field application. The inability to measure stress distribution
in treated tubing necessitates critical control over all the process
application parameters, i.e., bead nature and size, exposure time, air or
turbine speeds, nozzle diameter, distance from nozzle to tube, and angle
between bead stream and tube surface.
-During the field peening process, consideration should be given to
performing peening on control samples with subsequent evaluations
to assess the effectiveness of the process.
-Provisions should be made for quantitatively removing shots from
the steam generator.
-Recovered shot should be cleaned to remove contaminating materials
and broken beads prior to being reintroduced to the peening equipment.
In situ stress relief can be used to reduce residual stresses and, therefore,
primary side stress corrosion cracking of roll transition regions and
inner-row U-bends of mill-annealed alloy 600 tubing. The process used
should be qualified by testing prior to field application. The following
precautions should be taken when performing this type of stress relief
(4.15):
Recommendations 4.19
-Control should be maintained over the actual process application

parameters; that is, temperature should be held at 1,300-1,450'F (705-
780'C) for 5 to 15 minutes.
-Provision should be made for verifying that the temperature and
strains of the tube support plate regions adjacent to the heat-treated
U-bends remain acceptable during the treatment.
4.1.18.4 Girth Weld Cracking

Recommendations
* Avoid secondary chemistry conditions that promote corrosion pits,
which can be the points of initiation for shell and weld cracking. Pitting
may occur in high oxygen, low temperature conditions (such as layup).
See Sections 4.2.1,4.3.1, and 4.3.6 for information on layup and operating
chemistry. As a minimum, consider the following recommendations:
-Minimize time spent at hot standby conditions while using aerated
makeup water.
-Use nitrogen blanketing during layup when possible.
* Avoid steels for fabrication of new or replacement steam generator shells
that have an unfavorable distribution of sulfides, i.e., sulfur content
should be <0.01%.
• The sulfur content of weld metal should be minimized to prevent
cracking.
* Either verify that post-fabrication stress relief processes were effective
or implement a post-installation heat treatment of susceptible joints to
relieve residual stresses.
* Evaluate the potential effect of any in-generator activities such as
inspection, repair, or grinding. Cracking susceptibility is greater at
locations where the protective black oxide (magnetite) has been removed
from the shell or weld surface.
* Weld and weld repair techniques must conform to requirements of the
American Society of Mechanical Engineers (ASME) Boiler and Pressure
Vessel Code, or be covered by an exception procedure such as a code
case or NRC variance.
-- Generally, non-slag-producing weld techniques such as TIG welding
are preferable to slag-producing techniques such as stick welding, due
to the potential impact on water chemistry.
-Although temper bead repair welds are permitted where heat
treatment for stress relief is impracticable, consideration should be given
to whether the areas having base metal residual stresses produced by
this technique would be exposed to stress-corrosion-cracking-producing
environments.
* The sulfur content of weld metal should be minimized to prevent

cracking.
4.1.18.5 Steam Generator Replacement

Recommendations
* Chemical decontamination prior to replacement should be considered
in order to reduce radiation levels.
* Replacement planning should incorporate "lessons learned" from other
plants' experiences.
* Replacement activities should address recommendations in published
reports on planning and implementation (4.16).
4.1.18.6 Repair/Replacement Options

Recommendations
* Steam generator replacement should be kept as a viable option to
preserve plant design life.
-Areas outside the equipment hatch should be clear. Polar and
auxiliary cranes should be located so that they can be used to assist
with steam generator replacement (4.3).
-Connections should be provided inside containment for installation
of remote tools and equipment needed to disassemble and remove the
steam generator, and for installation of remote viewing cameras (4.3).
-Provision should be made on site for storage of replacement
components and equipment needed for the replacement operation (43).
* Although site-specific, the repair/replacement decision should include
the following steps (4.16).
-Identify site-specific variables, such as degradation modes,
degradation locations, projected future degradation rates, and impact
on steam generators.
-Define alternatives available, to include repair (plugging tubes,
sleeving, Thot reduction, chemical cleaning, new water chemistry, new
condenser/feedwater heater tubes, and chemical additives to the
secondary side) and replacement (replace entire steam generator, modify
internals, and replace only the lower assembly).
-Selectalternatives that provide the highest probability of successful
implementation and/or offer an operating benefit (better margins, more
power, etc.).
-Evaluate selected alternatives, with respect to operational, impacts
(heat transfer/flow margins, when derating required, safety concerns),
costs (direct, personnel radiation exposures, replacement power interest,
inflation, and future outage durations), event trees of scenarios for each
alternative (continued corrosion rates, effectiveness of alternatives, and

material life), and net present value of alternatives.
-Decide on strategy.
-Implement decision.
Minimization of lifetime costs should be considered as well.
TABLE 4-1
Reference Documents for Steam Generator Design Topics
Reference
1
Steam Generator Tube Supports PWR Steam Generator Reference Book
2
Steam Generator Tube Materials Guidelines for PWR Steam Generator
Tubing Specifications and Repair (4j)
Tube Fabrication Guidelines for PWR Steam Generator
Tubing Specifications and Repair (4.1)
PWR Steam Generator Inspection
Guidelines (4.8)
Steam Generator Tube Bundle PWR Steam Generator Reference Book
Assembly
Mechanical Design Plant Design for PWR Steam Generator
Inspection and Nondestructive Testing
(4-3)
Plant Design for PWR Steam Generator
Replaceability (4.3)
Shell Fabrication PWR Steam Generator Reference Book
3
Thermal Hydraulic Design PWR Steam Generator Reference Book
Design Against Water Hammer PWR Steam Generator Reference Book
and Dynamic Flow Instability
Archive Samples PWR Steam Generator Reference Book
Steam Generator Cleanliness PWR Steam Generator Reference Book
Steam Generator Guidelines for PWR Steam Generator
Repair/Replacement Tubing Specifications and Repair (4.1)
Plant Design for Steam Generator
Replaceability (4.3)
PWR Steam Generator Tube Plugging
Limits -Technical Support Document for
Expansion Zone PWSCC in Roll
Transitions (4.12)
TABLE 4-1 (cont'd)

Reference Documents for Steam Generator Design Topics
Reference
Steam Generator PWR Steam Generator Tube Repair
Repair/Replacement Limits-Technical Support for Outside
Diameter Stress Corrosion Cracking at
Tube Support Plates (4.13)
Qualification of Remedial Methods to
Prevent Primary-Side Stress Corrosion
Cracking of Steam Generator Tubing,
Vol. 1 (4.14)
In Situ Heat Treatment of U-Bends
(4.15)
Proceedings of the Steam Generator
Repair Replacement Workshop-July
1985 (4.16)
1
Additional information can be found in EPRI reports NP-3045, NP-4647,
NP-2046 (4.25-4.27)
2
Additional information can be found in EPRI reports NP-5072, NP-4665-SR,
NP-2838, and -SD, and NP-6719-M (4.25-4.31).
3
NP-2838, and NP-3068 (4.27, 4.32-4.34)
4.2 STEAM GENERATOR OPERATIONS

This section summarizes recommendations for operations that should
minimize steam generator corrosion. References to paragraphs in Section 4.4
(abstracts of published SGOG guidelines) and later chapters of this book
are located after each subsection title. Table 4-2 presents a listing of subject
areas from this section and the reference documents that give related
guidance.
4.2.1 Steam Generator Cleanliness and Cleanliness of

Interfacing Systems During Construction
(Sections 4.4.12 and 4.4.13)
Recommendations
* Prior to shipment from the manufacturer, the internal primary side
surfaces of the steam generator should be cleaned to meet Class B
cleanliness criteria and the internal secondary side surfaces should be
cleaned to meet Class C criteria, as described in ANSI N45.2.1, 1973.

Prior to on-site storage, the exterior of the steam generator should meet
Class D cleanliness criteria, as described in ANSI N45.2.1, 1973 (4.4).
" For shipment and on-site storage, the steam generator should be
pressurized with dry nitrogen. The stored steam generator and other
components should be inspected at least once every 30 days. Before
installation, the nitrogen should be released from the steam generator
and openings sealed against entry of dirt and moisture. Air circulated
through the secondary side should be filtered and dehumidified. The
primary side should be sealed against weld smoke and dirt.
* During installation, the steam generator should be periodically
inspected against its cleanliness requirements. Precautions should be
taken to prevent intrusion of weld splatter and minimize spread of weld
flux on internal surfaces. After installation, the steam generator should
receive a final inspection against cleanliness criteria (4.4).
* Internals of main steam, turbine, condensate, condensate demineralizer,
condensate transfer and storage, and condensate and feedwater
chemical control piping and equipment should be designated as
requiring Class C cleanliness, as described in ANSI N45.2.1, 1973 (4.4).
* Cleaning of component internals to meet Class C requirements should
be in accordance with Appendix E of the SGOG Guidelines for
Preoperational Flushing, Cleaning, and Layup of PWR Steam,
Feedwater, and Condensate Systems. Flushing procedures should
consist, as a minimum, of those steps described in Appendix A and
Appendix B of that Guidelines document. In addition, flushing should
be continued until the concentrations of sodium, potassium, and
fluoride are less than 10 ppb. Cleanliness verification techniques and
acceptance criteria outlined in ANSI N45.2.1, 1973, should be
incorporated into plant procedures (4.4).
* After installation, the steam generator should be placed in wet layup
with chemically treated, deoxygenated, deionized water with a nitrogen
blanket (4A.). The water quality should be verified three times per week
as being within specifications until chemical concentrations are stable,
then weekly. However, dissolved oxygen in recirculating steam
generators should be verified three times per week until stable and
then once weekly (4.2).
* If the inerted steam generators must be opened for any reason and
refilled with nitrogen, precautions should be taken to prevent oxygen
blanketing in static regions.
4.24 Steam Generator Reference Book, Revision 1
4.2.2 Minimizing Corrosion Product and Ionic Impurity Transport

(Sections 4.4.6, 4.4.11, 4.4.14, and 4.4.16, Chapters 21 and 23)
Recommendations
During operation, the primary efforts to prevent steam generator
corrosion are those that will prevent the transport and accumulation of
impurities. Water chemistry should be maintained in accordance with
the PWR Secondary Water Chemistry Guidelines (4.2).
-To minimize corrosion product transport and sludge accumulation,
preventive measures include: reducing erosion corrosion by operating
at high pH or employing a pH-control agent with a more favorable
liquid/vap6r distribution coefficient than ammonia and removing
corrosion products before their entry to the steam generator by full-
flow condensate polishing or, possibly, filtration (4.10).
-To minimize transport and accumulation of ionic species, preventive
measures include ensuring leaktight condenser operation, minimizing
oxygen in leakage, and minimizing contaminant throw from condensate
polishers (4.10).
-Makeup water treatment systems should be operated to control the
types and amounts of impurities entering the secondary cycle.
-Condensate polisher and blowdown demineralizer systems should
be operated to control the types and amounts of impurities recirculating
in the secondary cycle.
4.2.3 Enhancing Hideout Return Processes

(Sections 4.4.11 and 4.4.16, Chapter 21)
Recommendations
* Hideout return of soluble and precipitated species can be used to
facilitate removal of chemical contaminants from hideout areas.
Hideout return occurs during cooldown and heatup, and during outage
shutdown and layup. Each plant should make measurements of hideout
return rates during those operating conditions to determine what plant-
specific procedures will optimize removal.
* At a minimum, the following species should be measured: sodium,
potassium, magnesium, calcium, chloride, sulfate, and silica. In
addition, aluminum, manganese, phosphate, fluoride, acetate, formate,
propionate, as well as other organics that may be introduced by the
secondary cycle chemical treatment, such as glycolate, oxylate and other
stable organics should be measured periodically. They should be
monitored more frequently if they are present in significant quantities.
Hideout return rates should be determined to evaluate when removal
would be most effective. Each temperature hold should be maintained
until the return rate has reached and passed its maximum.
* Hideout return quantities from each stage of each outage should be

computed and trended. Calculation of cation-to-anion ratios and
MULTEQ analyses should be performed at 0% power/operating
temperature in order to infer crevice chemistry. The EPRI mass balance
software, MASSBAL, can be used to assist with data reduction. Hideout
return data should be analyzed to deduce impurity sources that can
then be minimized (4.10).
o During hideout return procedures, the concentration of impurities in
bulk water should remain within water quality limits specified by the
plant or in Reference 4.2.
o The steam generator should be drained and refilled before layup and
subsequent startup, in order to remove the dissolved hideout return
species (4,2).
* Steam generator wet layup time should be maximized during refueling
outages to encourage additional contaminant diffusion from crevices
and other occluded areas (4.10), and to provide maximum protection
of the steam generator from corrosion.
" Dry layup should not be utilized.
" The use of mid-cycle shutdowns to remove contaminants from crevices
should be considered (4.10), if it can be shown by material balance and
engineering analysis that the benefits of removing sufficient quantities
of hideout impurities outweighs the loss of power generation.
4.2.4 Crevice Flushing

(Section 4.4.11, Chapters 21 and 23)
Recommendations
o Plants should consider performing tubesheet crevice flushing each
refueling outage to remove hideout contaminants or to introduce boric
acid for mitigation of crevice corrosion. Sludge lancing should be
performed before crevice flushing, and deoxygenated, deionized water
should be used for both lancing and flushing operations (4.17-4.18).
' A mixture of reactor coolant pump circulation and natural circulation
procedures should be used, unless it is clearly evident from contaminant
removal rates that one is more effective than the other (4.1, 4.17).
o Heatup and cooldown cycles from 800 to 200'C should be considered
as an alternative to depressurization procedures, and implemented on
a regular basis if results are positive (4..10 4.17).
* Selection of temperature for depressurization (90'C or 150'C) should
be based on a plant-specific determination of which increases impurity
removal (4.10 4.17).
o Nitrogen overpressure of at least 50 psig should be provided between
depressurization cycles (4..10 4.17).
4.2.5 Sludge Lancing

(Section 4.4.11, Chapter 23)
Recommendations
* A sludge-lancing program should be considered each outage. The
sludge should be characterized in terms of consistency, amount,
distribution, and chemical composition. Visual (optical) and EC
inspection should be considered after each lancing to determine the
amount of sludge remaining. Sludge accumulation data from each
outage should be trended (4.10).
* Consider using CECIL or an equivalent inspection device to perform
periodic bundle inspections (4.19-4.21).
* Lance water quality at the point of discharge to the steam generator
should meet plant water quality specifications (4.10).
* Consider sampling the filtered sludge lance water prior to ion exchange
for comparison with hideout return results (Section 4.2.3).
• Minimize the time the steam generator remains open and exposed to
air (oxygen). Consideration should be given to lancing in a nitrogen
atmosphere (4.10
4.2.6 Chemical Cleaning

Recommendations
" If lancing is ineffective for sludge removal or if hard deposits have
formed on heat transfer surfaces, consideration should be given to
lancing with CECIL and/or chemical cleaning.
* Chemical cleaning should be considered for removal of deposits and
corrosion products from crevices and removal of copper and copper
oxides from surfaces.
* Before a cleaning process is selected, corrosion products should be
quantified and chemically characterized. Acceptable amounts of
corrosion for all important construction materials in the steam generator
should be established. The cleaning process selected should be subjected
to a plant-specific test to demonstrate that it is effective in corrosion
product removal and that it minimizes materials corrosion. The plan
for administering chemical cleaning should incorporate the objectives
concerning the areas to be cleaned and the thoroughness desired (4.10).
* Each plant should perform a site-specific engineering review of the
chemical cleaning process to be used, in order to characterize the waste
and to choose the optimum waste processing method. Ultimate disposal
of chemical cleaning wastes should be considered before a process is
selected (47,7 4.10).
" Any system for administering steam generator chemical cleaning

solutions should be designed to protect existing plant equipment and
interconnected systems. The system should be designed specifically
for the chemical cleaning purpose intended, i.e., tubesheet cleaning,
full-height cleaning, or crevice cleaning. Nozzle dams or inflatable plugs
should be installed to isolate the steam generator channel head from
the rest of the reactor coolant flow path, as a safeguard against leakage
of chemical solutions through the tubesheet and into the channel heads
(4.7,4.10).
" Procedures should include requirements for measuring surface chemical
contamination (sulfur, lead, sodium, chloride, etc.) by analysis of wipe
samples or taking of other samples at selected locations.
4.2.7 Thot Reduction

* Consider reducing Thot to minimize the potential for IGA/SCC.
* Evaluate the need for turbine modifications and the effects from altered
heater drain flows due to Thot reduction.
TABLE 4-2
Reference Documents for Steam Generator Operations Topics
TOpic Reference
Steam Generator and Balance of Preoperational Flushing, Cleaning, and
Plant Cleanliness Layup of PWR Steam, Feedwater, and
Condensate Systems (4.4)
Guidelines to Minimize Contamination of
PWR Steam Generators During Plant
Construction (4.4)
PWR Secondary Water Chemistry
Guidelines (4.2)
Impurity Transport Design and Operation of Condensate
Polishers (4.10)
Operating Guidelines for Condensate
Polishers (4.10)
Guidelines for Impurity Removal and
Control (4.10)
Guidelines (4.2)
TABLE 4.2 (cont'd)

Reference Documents for Steam Generator Operations Topics
Reference
Hideout Return Guidelines for Impurity Removal and
Control (4.10)
Guidelines (4.2)
Crevice Flushing Guidelines for Impurity Removal and
Control (4.10)
EPRI Final Report NP-4989 (4.17)
Sludge Lancing Guidelines for Impurity Removal and
Control (4.10)
Nuclear Engineering International Paper
(4.20)
Joint Power Generation Conference Paper
(4.21)
Chemical Cleaning Guidelines for Design of PWR Chemical
Cleaning Systems (4.7)
Guidelines for Impurity Removal and
Control (4.10)
Thot Reduction PWR Steam Generator Reference Book,
Rev. 0
4.3 PLANT DESIGN, MODIFICATION, AND

OPERATION
This section provides a brief summary of the principal recommendations
resulting from the Steam Generator Owners Group (SGOG) Programs and
the subsequent Steam Generator Reliability Project (SGRP), various research
reports, guidelines documents, and workshops. Only the major
recommendations are summarized here. Detailed operating and design
recommendations can be found in the referenced documents. Table 4-3
presents a listing of subject areas addressed in this section and the reference
documents that give related guidance. Abstracts of published SGOG
guidelines are presented in Section 4.4. References to specific paragraph
numbers in Section 4.4 are located after each title within this section. Also
referenced, where appropriate, are sections in Parts II or III of this book
where additional technical support for the recommendations is provided.
4.3.1 Steam Generator and Balance-of-Plant Layup

(Sections 4.4.3 and 4.4.9, Chapters 23 and 26)
Recommendations
" During cooldown, as soon as possible after reaching <200'F (<93 0C),
the steam generators should be filled with deoxygenated (<100 ppb
02), chemically treated water (4.2) throughout the full height of the
tube bundle in order to protect it against corrosion. The solution should
be continuously mixed and the deoxygenation preserved beneath a
nitrogen blanket (43 4.22). The chemistry of the solution should be
determined and maintained in accordance with Reference 4.2.
-Whenever the steam generator is required to be drained but can be
sealed, a positive-pressure nitrogen blanket should be established and
maintained (4.3). Time in this condition should be minimized.
-Whenever the steam generator must be opened to air, a positive-
pressure nitrogen blanket should be reestablished and the steam
generator should be refilled with deoxygenated water as soon as
conditions permit (4.3).
-Pre-planning should be used to minimize the time that steam
generators are drained.
* Controlled layup should be applied to all balance of plant secondary
systems during outages exceeding six months, whenever possible, in
order to reduce corrosion of balance of plant components and corrosion
product transport to the steam generators during startup. Consideration
should be given to applying controlled layup for any outage expected
to exceed two months, to ensure that equipment is properly protected
in the event of unanticipated delays (4.3). For outages of less than two
months, layup conditions should conform to the recommendations in
the PWR Secondary Water Chemistry Guidelines (4.2).
-Nitrogen blanketing can be utilized for layup of all systems and/or
equipment that can be kept closed or that can accommodate a
continuous purge (4.23).
-Dry air layup can be utilized for layup of all secondary systems (4.23).
Dry air layup should not be applied to steam generators.
-Wet layup (using deoxygenated, chemically treated water with a
nitrogen blanket (4.2) can be utilized for layup of the condensate-
feedwater train and the feedwater heater shells. However, wet layup
is not suitable for systems that cannot retain water for long periods of
time due to leaky valves (4.23) or systems that will be drained for
maintenance. Layup water quality should be determined and

maintained in accordance with Reference 4.2. Design provisions should
provide for maintaining the chemically treated water in a well mixed
condition (4_3 4.23).
* Shutdown and layup procedures should preclude cooling water lying
stagnant in condenser tubes for any significant period of time. ECT
should be performed on condenser tubes each outage. Debris buildup
in waterboxes should be removed during each outage (4.5).
* Prior to startup after a layup period, the condensate and feedwater
systems should be recirculated through condensate demineralizers until
water quality meets the requirements of Reference 4.2.
* During startup after a layup period, heater drains should be routed
back to the condenser hotwell so that corrosion product and ionic
impurities can be removed by condensate polishers (4.6).
4.3.2 Steam Generator Inspection and Nondestructive Testing

(Sections 4.4.2 and 4.4.16, Chapters 23 and 26)
Recommendations
* A baseline inspection of all steam generator tubes should be performed
prior to operation of the steam generator.
• Tapes, strip charts, and films containing NDE data and analyses should
be stored in a protected environment (4.8).
* Steam generators should be inspected during each planned steam
generator outage. The sample of tubes to be inspected should be chosen
on the basis of NDE guidelines (4.8), unit operating history, and that of
similar units.
* All data collected should be analyzed. Degradation and anomalous
signals should be documented. Each utility should develop a basis for
decisions to plug tubes, particularly those with low signal-to-noise ratio
indications.
* NDE inspection should be performed using the NDE guidelines (4i.8)
and qualified analysts and equipment.
4.3.3 Steam Generator Blowdown

Recommendations
* A maximum continuous blowdown rate of 1 wt% MSR (main steaming
rate) should be the design criteria for blowdown systems for RSGs
during normal operation, and for OTSGs at <15% power (see design
recommendations in section 4.1.10). For RSGs the system should be
able to accommodate a periodic blowdown of 3-7% MSR for short
durations (2-5 minutes) for removal of tubesheet sludge (4.3).
* During normal operation, RSG blowdown should be filtered, ion

exchanged, and recycled to the condenser.
* For OTSGs, blowdown during periods of operation <15% power should
be routed to the condenser, upstream of the condensate demineralizer,
or discarded to waste. This will minimize corrosion product transport
to the steam generators and reduce the sludge burden (43).
* During times of normal plant operation when open cycle blowdown
would result in better secondary cycle water chemistry (e.g., better
cation-anion balance, minimized regenerant leakage), its use should
be considered.
4.3.4 Condensate Polishing

(Sections 4.4.6, 4.4.7, and 4.4.14, Chapters 21 and 23)
Recommendations
* A full-flow condensate polisher system should be installed, and
operated only when it proves to be a net impurity sink. The actual
operating scheme used (full-flow versus partial-flow versus operation
only during startup or condenser leakage) will be site- or utility-specific
and determined after consideration of the impact on steam generator
crevice chemistry. Operating schemes selected should be based on
makeup water quality and characteristics and balance-of-plant design
and materials (4.6). The system design and operation should provide
effluent water quality that meets or exceeds the feedwater and steam
generator blowdown water chemistry requirements of Reference 4.2
during all modes of normal and abnormal plant operation.
* Utility and plant management should be committed to invest in
significant operator training and routine maintenance for a condensate
demineralization system (4.6).
* Design of the condensate demineralizers should provide for cleanup
before and during plant startup. The system should incorporate
equipment sizes and configurations capable of accommodating
treatment of the high-pressure drains during startup and full-flow
recirculation of feedwater prior to startup (4.6).
* Operation of the polishers should address impurity removal before and
during plant startup. Before plant startup, the feedwater and
condensate systems should be recirculated through the condensate
polishers to achieve water quality specified in Reference 4.2. The
condensate polishers should be used during startup to remove hideout
return contaminants that solubilize at temperatures lower than
operating temperature. During startup, high-pressure drains should
be routed to the condenser to reduce corrosion product transport to the
steam generator (4.6).
* Condensate polisher system design should address minimization of

resin fine leakage.
-For deep-bed demineralizers, resin strainers should be installed
downstream of demineralizers; they should be capable of preventing
the passage of resin (bead or powdered) in the event of support screen
failure. To prevent resin fine leakage, either strainers with fine screens
should be utilized or cartridge filters should be placed downstream of
each polisher vessel. System design for use of cartridge filters should
include consideration of pressure drop and pressure limitations of
piping and components (4.6),
-For powdered resin filter demineralizers, the septa design should
prevent precoat resin leakage from the polisher during cut-in, transient,
and normal operation (4.6).
* Condensate polisher operation should address minimization of resin
fine leakage. Flow should be increased gradually during vessel cut in.
Consideration should be given to returning polisher effluent to the
hotwell during the first few minutes after cut-in. Resin traps should be
cleaned and inspected on a time and differential pressure basis. For
deep-bed polishers, backwash flow rates and times should be increased
when necessary to remove resin fines (4.6).
* Design consideration should be given to sidestream installation of the
demineralizers to provide constant flow through the demineralizers
and to provide uninterrupted flow to the feedwater pumps in the event
of filter or strainer plugging with resins (4.6).
• Condensate polisher design should address the minimization of ionic
leakage.
-Consideration should be given to providing a cation-mixed bed
configuration and limiting design flow rates to less than 200 volumes
of anion resin per hour (4.6).
-For deep bed polishers utilizing external regeneration, the
regeneration system should provide for complete transfer of resin from
the service vessel to the regeneration vessel and back (4.6). The
regeneration vessel should be designed to provide a distinct, verifiable
interface between resin layers following separation, and should provide
for uniform regenerant chemical flow throughout the length of the bed
being regenerated (4.6). The use of an inert resin layer should be
considered (4.24), and a resin sampling penetration at the interface level
should be provided (4.6).
-System design should permit recirculation of water through each bed
at near normal operating flow rates (4.23).
-Deep-bed vessel design should incorporate the following features:
inlet header and underdrain designs such that plug flow is maintained
for the full length of the bed and water does not fluidize the top of the
bed; and sight glasses to observe internal operations.
-Powdered resin systems should be designed such that a vessel may
be precoated with demineralized, deaerated water and maintained
using the same type of water. Inlet flow distributors should be designed
to prevent flow impingement on the precoat (4.6).
4.3.5 Condensers
Recommendations
Condensers should be designed and operated, and materials should
be selected to provide essentially zero cooling water inleakage over the
life of the plant (4.4). This can be achieved by implementing the
following:
-Condensers should not utilize copper-bearing materials, in order to
avoid steam generator tubing corrosion associated with copper. In
addition, condenser tubing should be of corrosion- and erosion-resistant
materials such as titanium. Various grades of stainless steel are also
acceptable, where proven compatible with site conditions. For new
installations, tubesheet materials should be selected to permit use of
welded tube-to-tubesheet joints (4.3, 4.5).
-Expanded and welded tube-to-tubesheet joints should be used for
new installations. A non-welded double (or integral groove) tubesheet
with expanded joints may be an acceptable alternative. For existing
installations where welding or use of double tubesheets is not practical,
reliance should be placed on detailed control of new tube joint
installations and on rigorous leak detection and correction methods.
-For new installations, design, design review, and fabrication control
methods should be followed in order to protect against the problems
of steam impingement and tube vibration. The recommended methods
include use of special design details, performance of calculational
checks, and invocation of quality standards to provide assurance against
failures. Such methods should also be used for repairs or modifications
to existing condensers (4.5).
-New condensers should be designed to provide sufficient access such
that inspection and repair of critical areas can be readily performed.
This includes, for example, providing a waterbox geometry that permits
sufficient clearance for ECT and plugging of all tubes, including
peripheral tubes (4.5).
* Use of bellows-type expansion joints is recommended for high
temperature steam connections to the condenser in order to avoid air
inleakage at low power (4.5).
-The flexible joint between the turbine and condenser should have
minimum air leakage. Alternatives to achieve this goal include use of
a conventional rubber seal provided with a water seal, or a welded
metal seal (4.5).
" The necessity for cathodic protection should be evaluated with respect
to materials of construction and cooling water quality (4.5).
" Condenser operation should minimize air and circulating water
inleakage (4.3) and include monitoring for corrosion damage and air
inleakage. Air inleakage should be monitored daily (Section 4.2.2,4.2).
Shutdown and layup procedures should preclude cooling water lying
stagnant in tubes for any significant time period. ECT should be
performed on tubes each outage. Debris buildup in waterboxes should
be removed during each outage (4.5).
4.3.6 Water Chemistry

(Sections 4.4.15 and 4.4.16, Chapters 20, 21, 22, and 23)
Recommendations
* A primary coolant water chemistry program should be established and
supported by utility and plant management to optimize steam generator
and fuel crud buildup considerations. The program should be based,
as a minimum, on guidelines described in Reference 4.11. The program
should provide parameter values that initiate various responses from
chemistry and operations personnel. It should specify the framework
for data evaluation and data management, and should delineate all
continuous and grab sampling analyses. The quality control program
that ensures data validity should be described (4.11).
-Consideration should be given to operating at high pH, in order to
minimize transport and deposition of radioactivity that would hinder
steam generator maintenance and inspection activities. High pH
operation should be undertaken only after evaluation of potential plant-
specific effects, such as accelerated fuel cladding and steam generator
tube corrosion.
* A secondary water chemistry program should be established and
supported by management that limits the steam generator water
impurity concentrations. The program should be based, as a minimum,
on guidelines described in Reference 4.2. The program should provide
operating chemistry levels that initiate various responses (including
plant shutdown) from chemistry and operations personnel. It should
specify the framework for data evaluation and data management, and
should delineate all continuous and grab sampling analyses. The quality
control program that ensures data validity should be described (4.2).
-Consideration should be given to operating at high pH in order to
minimize erosion corrosion in the wet-steam piping, and thus minimize

sludge buildup in the steam generators. High pH operation should be
undertaken only after evaluation of potential plant-specific effects, such
as accelerated corrosion of copper-bearing alloys (4.10).
-Consideration should be given to operating with organic amines (such
as morpholine) as the pH control agent, rather than ammonia, in order
to minimize erosion corrosion in the wet/steam piping and minimize
sludge buildup in the steam generators. High pH operation should be
undertaken only after evaluation of potential plant-specific effects, such
as accelerated morpholine decomposition to organic acids, which is
due to the presence of organic impurities (e.g., greases, linseed oil), or
elevated dissolved oxygen concentrations.
-- Consideration should be given to operating with on-line boric acid
addition for AVT chemistry, or with coordinated phosphate chemistry
in some situations, in order to mitigate crevice corrosion mechanisms
such as denting and intergranular attack/intergranular stress corrosion
cracking (IGA/SCC). Boric acid treatment should be implemented only
-after evaluation of potential plant-specific effects, such as acceleration
of erosion corrosion in wet-steam portions of the secondary cycle. Boric
acid pH reducing effects can be offset by coincident implementation of
high pH or morpholine chemistry (4.2).
4.3.7 Secondary Systems Design

Recommendations
" Design features to minimize steam generator corrosion should be
incorporated into new plants and modifications to existing plants.
Features designed to minimize dissolved oxygen and transport of
corrosion products to the steam generator include: deaerating feedwater
heaters; full-flow condensate demineralizers; deaerated makeup water,
stored condensate, and auxiliary feedwater, increased capacity
blowdown treatment and recovery systems; cascade of all heat
exchanger drains (including moisture separator and reheater drains)
to the condenser; secondary system cleanup recycle lines to the
condenser, and condenser modifications to increase deaeration (e.g.,
improved air ejection and steam bubbler in hotwell) (4.3).
" Water piping materials in moisture-separator-reheater (MSR) drains and
wet/steam piping (including extraction lines and cross-over piping)
should be constructed of chromium-molybdenum low alloy steel. MSR
integral fins and tubes should be constructed of Type 439 ferritic stainless
steel, or equivalent, to minimize two-phase erosion corrosion (4.3) and
transport of corrosion products to the steam generators.
* Condensate sub-atmospheric piping components (valve stems,

expansion joints) should be designed and selected with concern for
prevention of oxygen inleakage (4.3).
* Copper alloys should be eliminated from secondary systems and
condensers. After this is done, the secondary cycle pH can be raised to
provide additional protection to ferrous alloys (4_,2 4.3).
" Consideration should be given to the use of high-Cr-containing alloys
for feedwater piping in high-turbulence, high-flow areas to minimize
corrosion rates, 2.25% Cr 1% Mo should be considered, as a minimum.
4.3.8 Makeup Water Treatment System

(Section 4.4.17)
Recommendations
* The makeup system should produce a high-purity effluent water (4.9).
* Consideration should be given to providing a dedicated operating staff
for the Makeup Water Treatment System (4.9).
" Plant procedures should describe critical operating parameters,
monitoring parameters, and corrective actions. Operating criteria in
the SGOG Guidelines for the Design and Operation of Makeup Water
Treatment Systems should be incorporated into those procedures (4.9).
" Makeup water storage tank designs and materials should minimize
contamination (e.g. nitrogen blanketing and linings that will not leach
impurities).
* The throughput capacity of the system should be sufficient to support
routine plant operations, outage, and startup activities, such as steam
generator drain and fill (4,3 4.9).
TABLE 4-3
Reference Documents for Plant Design, Modification, and Operation
Topics
Reference
Steam Generator and Balance PWR Secondary Water Chemistry
of Plant Layup1 Guidelines (4.2)
Design for Layup of Steam Generators
(4.3)
Design for Layup and Cleanup of Steam,
Feedwater, and Condensate Systems
(4.3)
Preoperational Flushing, Cleaning, and
Layup of PWR Steam/Feedwater/Condensate
Systems (4.4)
Design and Operating Guidelines for Plant
Condensers (4.5)
Design and Operation of Condensate
Polishers (4.6)
Steam Generator Inspection Steam Generator Inspection Guidelines (4.8)
and NDE (including Appendices G and H)
Steam Generator Blowdown 2 Guidelines for Design of Steam Generator
Blowdown Systems (4.3)
Condensate Polisher System 3 PWR Secondary Water Chemistry Guidelines
(4.2)
Design and Operation of Condensate
Polishers (4.6)
Design Guidelines for Plant Secondary
Systems (4ý3)
4
Condensers PWR Secondary Water Chemistry Guidelines
(4.2)
Design and Operating Guidelines for Plant
Condensers (4.5)
Design Guidelines for Plant Secondary
Systems (4.3)
TABLE 4.3 (cont'd)

Reference Documents for Plant Design, Modification,
and Operation Topics
LReference
Condensers 4 EPRI Final Report NP-7381 (4.4)
Water Chemistry Guidelines for Impurity
Removal and Control (4.10)
PWR Primary Water Chemistry Guidelines
(4.11)
PWR Secondary Water Chemistry Guidelines
(4.2)
Secondary Systems Design Design Guidelines for Plant Secondary
Systems (4.3)
PWR Secondary Water Chemistry Guidelines
(4.2)
Water Treatment System Plant Design Guidelines for Layup and
Cleanup of Steam, Feedwater, and
Condensate Systems (4.3)
Makeup Water Treatment Systems Design
and Operating Guidelines (4.9)
1
NP-4705,
2
and NP-4754 (4.35-4.38).
Additional information can be found in EPRI reports NP-2656 and NP-2977
(423 4.35).
itional information can be found in EPRI reports NP-2656, NP-2977,
NP-2978, NP-2981, NP-3037, NP-3046, NP-4550, NP-4553, NP-5074, and
NP-2975 (4.23, 4.35, 4.39-4.46).
4
Additional information can be found in EPRI reports NP-0263,
NP-0481,NP-0962, NP-0912, NP-1468, NP-1467, NP-2062, NP-2371, NP-2448,
NP-2492, NP-2597, NP-3536-SR, NP-3905, NP-4551, NP-4705, and numerous
reports published by EPRI divisions other than Nuclear Power (report prefixes of
EA-, CS-, FP-, and GS-) (4.47-4.61)
4.4 DESIGN AND OPERATING GUIDELINES TO

MINIMIZE STEAM GENERATOR CORROSION
Early in the course of the SGOG program (1979), it was recognized that
the industry could use some of the results of the program in the form of
design and operating guidelines. Thus, through joint efforts (SGOG
Architect/Engineer and Technical Advisory Committees, the EPRI Steam
Generator Project Office, the SGOG PWR Secondary Water Chemistry
Guidelines Committee, and various contractors) 15 such guidelines have
been prepared, reviewed, and issued for use by members of the EPRI Steam
Generator Program. The major recommendations from these guidelines
have been summarized and presented in Sections 4.1, 4.2, and 4.3. This
section presents an abstract of each of the guideline documents, detailing
the scope and purpose.
The design and operating guidelines cover a variety of topics and are
designed for use by design engineers from utilities and architect/
engineering firms, as well as plant construction, startup, operation, and
maintenance personnel. They should assist in preventing some of the
problems that have been experienced previously. In the design area, the
guidelines cover:
* steam generator tubing materials,
* balance-of-plant design,
* design for layup,
* condensate polishers,
* condenser design,
* blowdown system design, and
o makeup water treatment system design.
The following design guidelines provide information in anticipation
of problems and encourage the need to address them:
* design for nondestructive examination,
* design for chemical cleaning, and
" design for replaceability.
Finally, there are guidelines for use by plant personnel (during
construction, startup, operation, and maintenance) to prevent steam
generator corrosion:
* cleanliness during construction,
* preoperational cleaning,
* water chemistry control,
* impurity removal, and
• nondestructive examination.
The guidelines on layup, blowdown, condensers, and condensate,

polishers also have information of interest to plant personnel.
While by no means all inclusive, these guidelines should provide much
useful information for minimizing steam generator corrosion. The following
sections present descriptions of the individual guidelines.
4.4.1 Guidelines for PWR Steam Generator Tubing

Specifications and Repair (4.1)
The objective of these four volumes is to formulate guidelines
representing the best industrial practices for procuring tubing material,
sleeving existing tubing, and examining tubes removed from steam
generators. Information was obtained from surveys of industry practices
and operating experience.
These guidelines are applicable to both RSGs and OTSGs:
* Vol. 1: Alloy 600 ProcurementGuidelines;
* Vol. 2: Alloy 690 ProcurementGuidelines;
" Vol. 3: Tube Sleeve Design Specifications and Procurementand Installation
Guidelines; and
* Vol. 4: Tube Section Removal and Examination Guidelines.
4.4.2 Plant Design for PWR Steam Generator Inspection and

Nondestructive Testing (4.2)
This document provides guidance on the periodic nondestructive
examination for determining the integrity of PWR steam generator tubes
and is based on NRC regulatory requirements, plant operating experience,
and SGOG- and EPRI-sponsored research. The guidelines are applicable to
Westinghouse, Combustion Engineering, and Framatome RSGs; and to
Babcock & Wilcox OTSGs.
The guidelines document provides:
* an overview of steam generator nondestructive examination (NDE)
efforts;
* NDE guidelines designed to help utilities implement a steam generator
in-service inspection (ISI) program;
* a survey of available remote positioning, data acquisition, and data
analysis equipment for determining tube integrity;
* a survey of available technology for identifying and monitoring tube
damage precursors; and
" a summary of adverse operating experience that documents damage
forms, vulnerable locations within the steam generator, and NDE
experience.
The NDE guidelines provide specific recommendations for:

* NDE equipment for detecting and monitoring tube integrity and.
damage precursors;
• probe designs, instrument settings, and signal interpretation methods
appropriate for various forms of degradation at various locations within
the generator; and
• increased surveillance of regions within the generator that
historically have been prone to degradation.
Finally, the guidelines document provides recommendations (and the
bases) for implementing a steam generator ISI program. Recommendations
address areas of the generator for increased inspection surveillance and
NDE methods for monitoring tube integrity and damage precursors.
4.4.3 Plant Design Guidelines for Layup and Cleanup of Steam,

Feedwater, and Condensate Systems (4.3)
This is a design guidelines document that provides requirements for
certain aspects of the design of PWR nuclear power plant steam, feedwater,
condensate systems, and for interfacing support systems. The guidelines
are directed at ensuring that neither corrosion during shutdown periods
nor release of corrosion products during plant startups following shutdowns
result in excessive amounts of corrosion products being introduced into
the steam generators. To this end, the guidelines cover:
* design of systems and components to minimize corrosion during the
shutdown periods; and
* design of systems to permit effective cleanup following shutdown
periods, so that corrosion products are removed from the feedwater
before they are introduced into the steam generators.
These guidelines apply to the steam, feedwater, and condensate systems
and to interfacing systems necessary for layup and cleanup, such as makeup
water, auxiliary steam, chemical addition, nitrogen blanketing, and air
drying systems. Only superficial coverage is given to those components or
systems that are addressed in separate guideline documents, such as
makeup water s'stems.
Protection of secondary systems during layup will minimize system
corrosion and corrosion product transport to the steam generator. Equally
important is the decreased startup time associated with cleanup of
condensate and feedwater. This latter concern is of greater significance in
units without condensate filters or demineralizers, since cleanup of corrosion
products generated during layup and subsequently released during heatup
is more difficult than when demineralizers are installed.
4.4.4 Guidelines for Design of PWR Steam Generator Chemical

Cleaning Systems (4.7)
The recommendations contained in this guideline apply to the SGOG
chemical cleaning process. This process is designed for removal of iron
oxide sludges, copper, and copper oxide sludges, and iron oxides in crevice
regions for plants operation on all-volatile treatment (AVT). The guidelines
are written to be applicable to both recirculating and once-through-type
steam generators. Any recommendations specific to a particular steam
generator design are appropriately noted. The steam generator chemical
cleaning operations covered include:
* concentrated chemicals storage;
* mixed chemical solutions;
• - heatup and cooldown operations;
* solvent and rinse water injection, recirculation, and draining;
* spent solvent waste, storage; and
* rinse solution waste storage.
There are two major plant issues that are not addressed by this guideline
format: NRC approval for performance of the chemical cleaning process,
and the disposal method for the large quantity of hazardous or radioactively
contaminated solutions that are generated. The NRC approval is site-specific
and generally should assure the post-cleaning structural and pressure
boundary integrity of the steam generator.
These guidelines address tubesheet cleaning, full-height steam
generator cleaning, and tube support plate (TSP) cleaning.
4.4.5 Guidelines for Design of Steam Generator Blowdown

Systems (4.3)
These guidelines are applicable to new plant designs and to retrofit
designs for units under construction or modification. Some design features
discussed in these guidelines cannot be backfit without major system or
steam generator modifications. The following assumptions are made
regarding new PWR plants:
• the condenser is designed to be tight with no cooling water leaks;
* the condenser is designed to have low air inleakage and high air ejector
capacity; and
* the steam generator and its blowdown header are designed to
enhance the removal of particulates via the blowdown system.
The guidelines are mainly directed at plants with RSGs supplied by
Combustion Engineering or Westinghouse, and should be useful for other
vendor designs. Chapter 6 covers the blowdown systems being provided
for Babcock & Wilcox OTSG plants for use at low power (<15% full power),
when they operate in a recirculation mode. No portion of the blowdown

system within the steam generator is addressed in this document.
The guidelines provide guidance for the following system features:
" capacity,
" system configuration,
" equipment types,
* instrumentation,
* materials of construction, and
* control.
The guidelines do not cover features necessary to meet safety
requirements of NRC documents, since the requirements are primarily
directed at assurance of structural and pressure boundary integrity (not
with system performance) which is the main focus of these guidelines.
4.4.6 Condensate Polisher Design, Operation, and Maintenance

Guidelines (4.6)
Design Guidelines. A wide variety of condensate polishing system designs

and resin types has been used successfully in condensate polishing
applications. The goal of this guideline is to provide recommendations to
ensure that the polisher system effectively controls steam generator impurity
ingress, consistent with industry water purity goals. The recommendations
include the following system parameters:
* design and layout of vessel internals,
* operating parameters,
* resin performance,
* system monitoring instrumentation, and
* equipment and instrument maintenance.
The guidelines do not cover design aspects such as structures, pressure
boundaries, detailed system layout and material selection. These features
do not significantly affect system performance, which is the main focus of
these guidelines.
Operating Guidelines. The goal of this guideline is to provide operating

recommendations applicable to the wide variety of installed systems, in
order to ensure that the polisher system effectively controls steam generator
impurity ingress. The recommendations cover the following topics:
• management commitments,
" deep-bed polishers,
* powdered resin polishers,
* resins, and
* maintenance.
In addition, descriptions of various regeneration schemes are presented,

and procedures for testing ion exchange resins are included.
4.4.7 Design Guidelines for Plant Secondary Systems (4.3)

These design guidelines are applicable to PWR secondary systems,
which include:
* condensate, and condensate storage systems;
* feedwater system;
* feedwater heater drains;
• moisture separator and reheater, including drains;
" steam system, including extractions; and
" auxiliary feedwater system
The intent of the design guidelines is to reduce the ingress of corrodents
into the secondary system from external sources; to minimize corrosion of
the secondary system components; and to minimize the subsequent
transport of corrosion products to the steam generator. Because there is a
direct relationship between the corrodents transported to the steam
generators and those transported to the turbines, the impact on turbines is
also assessed.
These guidelines are essentially recommendations for new PWR
designs. At existing PWR stations with RSGs and OTSGS, it would be
desirable to incorporate all of the features proposed for new plants.
However, not all of the proposed features are technically and/or
economically feasible as retrofits. The decision to adopt or reject a
recommendation must be strictly utility specific, based on current plant
performance and utility resource allocation.
4.4.8 Plant Design for PWR Steam Generator Replaceability (4.3)

The purpose of this guideline is to identify those aspects of nuclear
power plant design that can affect the replaceability of a steam generator.
The intent of the guidelines is to ensure that steam replacement operations
can be performed with a minimum of time and system modifications. All
replacement operations are considered in order to include design facets
that could potentially facilitate or hamper steam generator replacement.
The following specific areas of plant design are discussed:
" containment structure,
* polar and auxiliary cranes,
" steam generator and steam generator cavity,
* other containment systems,
* service connections,
" areas outside of containment, and
" record keeping.
4.4.9 Design Guidelines for Layup of PWR Steam Generator (4.3)

The recommendations contained in this document apply to the
generators and to related support systems, such as
* steam generator recirculation system,
* steam generator drain and blowdown systems,
• makeup water requirements, and
* nitrogen systems.
In addition, the guidelines also cover the parts of the steam and
feedwater systems that establish the layup boundary of the steam generator.
These guidelines are directed at the steam generator secondary side
and do not address primary side requirements. Corrosion attack can also
occur on the primary side during shutdown periods, and protection against
such attack is required. However, protection against primary side corrosion
involves control of overall reactor coolant system chemistry and conditions,
which is outside the scope of this document.
4.4.10 Design and Operating Guidelines for Plant Condensers (4.5)

These guidelines cover design features and analyses recommended for
new condensers and for replacement of tube bundles. Maintenance
procedures to minimize air inleakage and to locate and plug cooling water
leaks are also covered, as well as recommended operating practices for use
during startup, shutdown, and layup.
The discussions include detailed descriptions of calculations
recommended for evaluating the adequacy of condenser design features.
Sample calculations and listings of computer programs are included, where
appropriate.
These guidelines have been prepared to supplement, not replace, users'
specification and design manuals. Information is included that can be used
as an aid to understanding the design features that are of major importance
when specifying steam condensers. While intended primarily for nuclear
plant applications, these guidelines can be used by engineers who design,
specify, or evaluate large steam surface condensers for all types of power
plants. The guidelines can also be used by plant operations and maintenance
personnel responsible for condenser operation and repair.
Design Guidelines. Design topics include:

* materials of construction,
* cathodic protection,
* heat transfer,
* instrumentation,
* removal of air and noncondensable gases,
* expansion joints, and
* tube inspection and plugging.
Operating Guidelines. Operations topics include:

" startup operation,
" air inleakage determination,
• heat transfer performance,
• tube/shell/waterbox inspection and repair, and
• shutdown and layup procedures.
4.4.11 Guidelines for Impurity Removal and Control (4.10)

The purpose of these guidelines is to provide information for utility
personnel, particularly plant operations personnel, about the best means
for removing impurities from steam generators and for controlling their
ingress. These impurities may be dissolved species, particulates, or
combinations thereof. The following factors are discussed:
" costs (dollars, radiation exposure, plant downtime);
• risks (to the steam generators, personnel, other equipment); and
* options available.
Each method described in these guidelines has been used for at least
one operating PWR. The guidelines do not cover features that may be
necessary to meet NRC safety requirements, since those requirements are
primarily concerned with assurance of structural and pressure boundary
integrity, not with system performance, which is the main focus of these
guidelines.
Methods for impurity removal and control discussed include:
* sludge lancing,
* hideout return,
" tubesheet crevice flushing,
" boric acid addition, and
• chemical cleaning.
4.4.12 Preoperational Flushing, Cleaning, and Layup of PWR

Steam, Feedwater, and Condensate Systems (4.4)
These guidelines describe desirable field practices for steam, feedwater,
and condensate systems in the following areas:
* defined levels of cleanliness and acceptance criteria,
* verification of component cleanliness,
* cleanliness control to minimize contamination, and
* acceptable cleaning methods and materials for removal of deleterious
materials.
This guideline addresses neither operational requirements nor methods
necessary to achieve operational requirements. Rather, cleanliness control
during erection of the secondary systems is discussed, so that only water
flushing or rinsing of an installed system is required to render it ready for
service. When more than a water flush or rinse is needed to produce the
specified cleanliness, additional cleaning in accordance with this guideline
may be necessary.
The guideline recommends chemical cleaning be performed only if the
attempt to build and maintain a clean system is unsuccessful.
4.4.13 Guidelines to Minimize Contamination of PWR Steam

Generators During Plant Construction (4.4)
These guidelines provide the measure required to preserve steam
generator cleanliness during power plant construction. Preparation for
shipment, receipt, handling, storage, and installation of steam generators
are covered. The following topics are addressed:
* cleanliness classes;
* protective measures;
* cleanliness inspection criteria;
o inspection procedures for steam generator receipt, storage, and
installation;
* cleanliness requirements for the storage site; and
o handling procedures for receipt and installation.
The guidelines do not provide instruction on implementing detailed
recommendations. Instead, they do outline general methods and criteria
to be used for minimizing steam generator contamination during shipment
and installation. The guidelines consider all major activities that may affect
steam generator cleanliness during these time periods.
4.4.14 Steam Generator Inspection Guidelines (4.8)

The purpose of these guidelines is to provide recommendations, based
on research results and plant experience, that will allow utilities to achieve
the maximum benefit from periodic steam generator examination. They
will help minimize the likelihood of forced outages due to tube leaks during
each operating cycle, assure that established regulatory requirements are
met, and minimize unnecessary tube plugging and sleeving. The guidelines
are comprised of the following subject areas:
" steam generator nondestructive examination (NDE) objectives;
* general recommendations for implementing a steam generator
examination program;
* recommended practices for steam generator examination;
" summary of steam generator operating experience, including damage
mechanisms, vulnerable damage locations, and NDE experience; and
* industry-wide survey of leaker-caused outages, including specific
causes.
Specific recommendations are provided for:
" random tube sampling strategies for general steam generator
surveillance, with augmented sampling for units with active damage
mechanisms;
* methods for monitoring tube integrity and damage precursors; and
o establishment of an ECT data analyst performance demonstration
program.
Guidelines for analyzing data and preparing NDE bid specifications
are appended to the document.
4.4.15 PWR Primary Water Chemistry Guidelines (4.11)

These guidelines were developed in order to standardize chemistry
programs to. ensure fuel and material integrity, as well as minimize plant
radiation fields. Information that is applicable to the operation and integrity
of steam generators includes:
* chemistry parameters to reduce primary side cracking of alloy 600 steam
generator tubes; and
• primary chemistry regime(s) that will reduce steam generator
channel head radiation fields.
The second topic affects the worker-rem exposure required to perform
0
an NDE of steam generator tubing integrity.
4.4.16 PWR Secondary Water Chemistry Guidelines (4.2)

These guidelines provide the details and recommendations for a
secondary water chemistry control program that is necessary in order to
minimize corrosion damage to PWR steam generators. The document
specifies the water chemistry parameters that should be monitored and
controlled. It also provides suggested action level values for parameters,
as well as permissible out-of-specification time limits and operational
responses. The document provides guidance on management functions
related to implementing and maintaining a water chemistry program, i.e.,
establishing plant-specific action levels, determining techniques for
monitoring and measuring parameters, and developing procedures for data
handling, program surveillance, and data review. The document is
comprised of the following subject areas:
o management functions and responsibilities;
* control parameters for RSGS;
* control parameters for OTSGS;
* analytical and monitoring methods; and
o data management, evaluation and surveillance.
Guidelines include information on the use of the MULTEQ computer
code, interpretation of chemistry data in the presence of boric acid and
morpholine, recommendations for plants using boric acid as a secondary
coolant additive for mitigation of IGA/SCC, and recommendations for
plants using morpholine as a secondary coolant additive for minimizing
corrosion product transport to the steam generators.
4.4.17 Guidelines for the Design and Operation of Makeup Water

Treatment Systems
These guidelines present design and operating criteria for 17 unit
processes that can be used in a makeup water treatment system. The
guidelines are intended for use in designing new systems and modifications,
and improving water quality by modifying operating practices.
Performance criteria for effluent quantity and quality are given, with
guidance for evaluating existing systems.
The design guidelines discuss how to develop a complete system design
by first selecting the desirable unit processes, then the types of equipment
best suited for each process. Information is included to assist with economic
evaluation of alternative designs, preparation of equipment specifications,
and evaluation of vendor bids.
Operating guidelines discuss how to evaluate system and component

operation in order to determine needed improvements or set up a
monitoring system and data management program. Trouble shooting
guidelines for the entire system and individual components are included,
with critical operating and monitoring parameters identified. Guidance
for selection and purchase of media and chemicals are also provided.
REFERENCES
4.1 "Guidelines for PWR Steam Generator Tubing Specifications and
Repair." EPRI NP-6743-L. February 1990.
4.2 "PWR Secondary Water Chemistry Guidelines, Rev. 2. EPRI
NP-6239." December 1988.
4.3 "Nuclear Plant Design and Modification Guidelines for PWR Steam
Generator Reliability." EPRI NP-7380. September 1991.'
4.4 "Preoperational Practices for Steam Generators and Secondary-
System Components." EPRI NP-7381. September 1991.
4.5 "Design and Operating Guidelines for Nuclear Power Plant
Condensers." EPRI NP-7382. September 1991.
4.6 "Design and Operating Guidelines for Condensate Polishers in
Nuclear Power Plants." EPRI NP-7383. September 1991.
4.7 "Guidelines for Design of PWR Steam Generator Chemical Cleaning
Systems." EPRI NP-7384. September 1991.
4.8 "PWR Steam Generator Inspection Guidelines, Rev. 2." EPRI
NP-6201. December 1988.
4.9 "Guidelines for the Design and Operation of Makeup Water
Treatment Systems. "EPRI NP-6377. June 1989.
4.10 "Guidelines for Impurity Removal & Control, Rev. 1. Plant and
Steam Generator Operating Guidelines." EPRI Steam Generator
Project Office, December 1986.
4.11 "PWR Primary Water Chemistry Guidelines, Rev. 2." EPRI NP-
7077. November, 1990.
4.12 PWR Steam Generator Tube Plugging Limits Technical Support
Documentfor Expansion Zone PWSCC in Roll Transitions. EPRI
NP-6864-L. December 1990.
4.13 "PWR Steam Generator Tube Repair Limits-Technical Support for
Outside Diameter Stress Corrosion Cracking at Tube Support
Plates." EPRI TR-100407. March 1992.
4.14 Frederick, G., P. Hernalsteen, and J. Stubbe. "Qualification of
Remedial Methods to Prevent Primary Side Stress Corrosion
Cracking of Steam Generator Tubing," Vol. 1. EPRI NP-5249. June
1987.
4.15 Pement, E W., G. Economy, and R. G. Aspden. "In Situ Heat

Treatment of U-Bends." EPRI NP-5496. November 1987.
4.16 Proceedings of the Steam Generator Repair Replacement Workshop,
July 25-26, 1985, Seattle, Washington.
4.17 "Evaluation of PWR Tubesheet Crevice-Flushing Approaches."
EPRI NP-4989. November 1986.
4.18 Baum, A. J., and J. B. Prestegiacomo. "Chemical Enhancement of
Crevice Flushing. " EPRI NP-5199. June 1987.
4.19 Lee, R. A. S. "CECIL-A Robot for Secondary-Side Maintenance
of PWR Steam Generators." EPRI NP-5929. February 1989.
4.20 Reinhardt, M. "CECIL Grows Up." Nuclear Engineering
International. (December 1991): 44-45.
4.21 Trovato, S. A., et al. "Design, Development, and Field Testing of
CECIL: A Robot for Secondary Side Steam Generator Maintenance."
Paper No. 90JPGC/NE-5, ASME/IEEE Joint Power Generation
Conference, Boston, MA, October 21-25, 1990.
4.22 MacArthur, A. D. "Evaluation of Steam Generator Fluid Mixing
During Layup." EPRI NP-2993. May 1983.
4.23 "Evaluation and Categorization of Secondary System Layup and
Cleanup Practices for PWR Plants." EPRI NP- 2656. December
1982.
4.24 Kibler, J. M., M. D. Moretti, C. E. Hamilton, S. L. Harper, and G. T.
Upperman. "Resin and Ionics Leakage From Condensate Polishers
With and Without Inert Resin." EPRI NP-4521. April 1986.
4.25 McKubre, M. C. H., S. C. Leach, and L. E. Eiselstein. "Crevice
Corrosion of Lattice Support Alloys in Secondary Environments of
Nuclear Steam Generators." EPRI NP-3045. July 1983.
4.26 "Oxide Growth Mechanisms on Chromium Alloy Steels." EPRI
NP-4647. July 1986.
4.27 Curlee, N. J., Jr., and A. J. Baum. "Single-Tube Thermal and
Hydraulic Tube Support Test." EPRI NP-2046. September 1981.
4.28 Monter, J. V. "Specially Prepared Alloy 600 Tubing." EPRI NP-5072.
February 1987.
4.29 "Proceedings: Workshop on Thermally Treated Alloy 690 Tubes for
Nuclear Steam Generators." EPRI NP-4665M-SR. July 1986.
4.30 "Proceedings: Alloy 690 Workshop." EPRI NP-6750-M and -SD.
April 1990.
4.31 "Proceedings: Primary Water Stress Corrosion Cracking (PWSCC)
Remedies Workshop." EPRI NP-6719-M. April 1990.
4.32 Baum, A. J., and P. K. Greaney. "Steam Generator Sludge Pile Model
Boiler Testing." EPRI NP-1941. July 1981.
4.33 Cassell D. S., and D. W. Vroom. "Thermal-Hydraulic Tests of Steam

Generator Tube Support Plate Crevices." EPRI NP-2838. January
1983.
4.34 Patel, B. M., and J. K. Hayes. "Investigation of Steam Generator
Corrosion Products Under Typical PWR Operating Conditions."
EPRI NP-3068. May 1984.
4.35 Cleary, W. F "Evaluation of Secondary-System Layup and Cleanup
Practices and Processes." EPRI NP-2977. April 1983.
4.36 Whyte, D. D. "Laboratory Program to Examine Effects of Layup
Conditions on Pitting of Alloy 600." EPRI NP-3012. April 1983.
4.37 Bogaerts, W. E, and C. Bettendorf. "Electrochemistry and Corrosion
of Alloys in High-Temperature Water." EPRI NP-4705. July 1986.
4.38 Isaacs, H.S., 0. Oyeleye, M. Davidson, D. Dudek, T. A. Hatton, J.
W. Tester, R. K. Helling, and J. C. Erickson. "Electrochemical
Corrosion of Nickel-Based Alloys." EPRI NP-4754. September 1986.
4.39 Bell, M. J., R. 0. Carey, L. S. Lamanna, N. J. Mravich, D. E Stanionik,
and R. R. Zeller. "State-of-the-Art Evaluation of Condensate
Polisher Performance." EPRI NP-2978. April 1983.
4.40 Cleary, W. F and G. B. Cockley. "Condensate Polisher Resin Leakage
Qualification and Resin Transport Study." EPRI NP-2981. April
1983.
4.41 "Condensate Polishing at Surry Nuclear Power Station." EPRI
NP-3037. June 1983.
4.42 "Evaluation of Condensate Polishers." EPRI NP-3046. June 1983.
4.43 "Condensate Polishers for Brackish Water-Cooled PWRS." EPRI
NP-4550. July 1986.
4.44 "Sidestream Condensate Polishing for PWRs." EPRI NP-4553. June
1986.
4.45 Siegwarth, D. P. "Evaluation of Changing PWR Polisher Operation
From Regenerated to Throwaway Resins." EPRI NP-5074. March
1986.
4.46 Siegwarth, D. P., K. A. Friedman, and D. A. McNea. "A Review of
Condensate Demineralizer Regeneration Procedures." EPRI
NP-2975. April 1983.
4.47 "A Summary of Nuclear Power Plant Operating Experience for
1975." EPRI NP-0263. October 1976.
4.48 "Steam Plant Surface Condenser Leakage Study." EPRI NP-0481.
March 1977.
4.49 "Analysis of Steam Chugging Phenomena." EPRI NP-0962.
January 1979.
4.50 "Location of Condenser Leaks at Steam Power Plants." EPRI
NP-0912. 'October 1978.
4.51' Beavers, J. A., A. K. Agrawal, W. E. Berry. "Corrosion-Related

Failures in Power Plant Condensers." EPRI NP-1468. August 1980.
4.52 "Assessment of Condenser Leakage Problems." EPRI NP-1467.
August 1980.
4.53 "Steam Plant Surface Condenser Leakage Study Update." EPRI
NP-2062. May 1982.
4.54 "'Condenser Retubing Criteria Manual." EPRI NP-2371. May 1982.
4.55 Oiker, I., and D. Katsman. "Evaluation of Secondary System
Oxygen Control in PWR Power Plants in the USSR." EPRI NP-2448.
June 1982.
4.56 Beavers, J. A., G. A. Breeze, and W. E. Berry. "Inhibition of Steam
Condensate Corrosion of Copper-Based Alloys by Hydrazine."
EPRI NP-2492. July 1982.
4.57 Kassen, W. R., T. A. Putkey, S. G. Sawochka, W. L. Pearl, and M. E.
Clouse. "Condenser Inleakage Monitoring System Development."
EPRI NP-2597. October 1982.
4.58 "EPRI Research on Power Plant Mechanical Auxiliaries." EPRI
NP-3536-SR. May 1984.
4.59 "Pitting Corrosion of Alloy 600 Stream Generator Tubing Results
of a Laboratory Scoping Study" EPRI NP-3905. February 1985.
4.60 "Condenser Improvements for PWRs." EPRI NP-4551. August
1986.
4.61 Bogaerts, W. F. and C. Bettendorf. "Electrochemistry and Corrosion
5
STEAM GENERATOR
DEGRADATION
(TECHNICAL DESCRIPTION OF
CAUSES, CONSEQUENCES, AND
PREVENTION)
6
THERMAL AND HYDRODYNAMIC

ANALYSIS OF
STEAM GENERATORS
7
PRIMARY WATER
STRESS CORROSION CRACKING
A. R. Mcllree, EPRI
E. S. Hunt, Dominion Engineering,Inc.
J. A. Gorman, Dominion Engineering,Inc.
INTRODUCTION
Stress corrosion cracking (SCC) is the cracking-type failure of susceptible
material under the combined action of a corrosive environment and
sustained tensile stress. Stress corrosion cracking of nickel-chromium-iron
alloy 600 is typically intergranular and proceeds in a direction roughly
perpendicular to the stress field as shown in Figure 7-1. The environment
and stress levels leading to stress corrosion cracking can be sufficiently mild
that neither the applied stress nor the corrosive environment by itself would
cause cracking. Stress corrosion cracking can initiate at pre-existing cracks,
surface discontinuities, or at locations of submicroscopic differences in
material composition, protective film thickness, or corrodent concentration.
Prior to 1959, alloy 600 was considered to be immune to stress corrosion
cracking in pure water. In 1959, Coriou reported on laboratory test work
which showed that alloy 600 can be susceptible to stress corrosion cracking
in pure water at high temperature (662°F [350°C1) (7.1).
Most PWR steam generators have been fabricated using alloy 600 tubing
with a few exceptions: (1) A few of the earliest steam generators in small,
low-temperature plants have stainless steel tubes; (2) KWU steam generators
have alloy 800 tubing (the original Obrigheim steam generators supplied
by KWU had alloy 600 tubing); and (3) New and replacement steam
generators installed starting in the late 1980s have alloy 690 tubing.
Over the past 20 years, primary water stress corrosion cracking
(PWSCC) of alloy 600 tubing has become a significant concern for PWR
steam generators. PWSCC of steam generator tubing has led to the need
for increased inspections, remedial measures to prevent PWSCC, remedial
measures to permit degraded tubes to remain in service, tube plugging,
and, in extreme cases, steam generator replacement.
The purpose of this chapter is to summarize the types, extent and causes
of steam generator PWSCC and methods currently used to detect and
monitor PWSCC in operating plants. Further details are available in a
number of survey reports (7.2-Z7) and proceedings of EPRI workshops (7.8-
7.14). PWSCC remedial measures are discussed in Chapter 25.
With few exceptions, PWSCC problems have been limited to
Westinghouse-type steam generators fabricated by Westinghouse or its
licensees (W/L). The main reasons that PWSCC has not occurred
extensively in Babcock & Wilcox (B&W), Combustion Engineering (C-E),
and KWU steam generators are documented below.
Primary Water Stress Corrosion Cracking 7-3
FIGURE 7-1
Stress Corrosion Cracking in Alloy 600 Tubing Material
Babcock & Wilcox

Babcock & Wilcox once-through steam generators were subjected to a
full bundle and shell stress relief heat treatment after installation of the
tubes. This process developed a microstructure in the tubes which is more
resistant to PWSCC, and reduced fabrication-induced residual stresses. The
improvements relative to PWSCC were achieved at the expense of
sensitizing the tubing which makes it more susceptible to attack under
abnormal operating conditions (acidic and oxidizing).
TMI Unit 1 steam generators, which were supplied by B&W,
experienced some primary side cracking after an incident when sodium
thiosulfate was injected into the steam generator during a functional test.
The event at TMI I is probably unrelated to the PWSCC problems discussed
in this chapter except to the extent that the stress relief operation, which
provides improved resistance to PWSCC, increases the risk of cracking
under acidic and oxidizing conditions.
Combustion Engineering
Two main factors are believed to account for the nearly complete absence
of PWSCC in C-E steam generator tubing: (1) The tubing was mill annealed
at high enough temperatures to develop a more PWSCC-tolerant
microstructure. (2) The fabrication procedures used to expand the tubing
in the tubesheet and to fabricate U-bends are believed to have resulted in
lower residual stresses than in many earlier Westinghouse-design plants.
At one time, several tubes in Combustion Engineering steam generators
at San Onofre 3 experienced PWSCC in service. It was determined that
these tubes had not been mill annealed after cold drawing. These tubes
were removed from service, and there has been no further problem of this
kind.
Kraftwerk Union
Kraftwerk Union concluded in the late 1960s that alloy 600 material
can be susceptible to PWSCC in primary system water. Accordingly, they
changed from alloy 600 to alloy 800 tubing material. Alloy 800 has about
25% more chrome and about half the nickel of alloy 600. This material has
proven to be resistant to PWSCC. However, it has experienced wastage in
sludge piles of steam generators which operate on phosphate water
chemistry.
TYPES OF PWSCC DEGRADATION

Thus far, PWSCC has mainly occurred at the four locations in
recirculating steam generator tube bundles shown in Figure 7-2. These
locations are: (1) expansion irregularities within the tubesheet, (2) tubesheet
expansion transitions, (3) dented tube support plate intersections, and (4) the
transition and apex regions of row I and 2 U-bends. In each case, PWSCC
has been confirmed in tubing specimens removed from operating plants.
Below is a brief chronology of significant events:
* In 1959, Coriou reported on the cracking of high nickel alloys in "high
purity" water at 662°F (350'C) (a.).
* In 1971, after two years of operation, cracks occurred in the U-Bend
region of alloy 600 tubes in the original Obrigheim steam generators.
This was followed by cracks at roll transitions in 1972 (7.15-7.17). These
were the earliest reported occurrences of PWSCC in an operating PWR
steam generator. PWSCC at Obrigheim was attributed to an unusual
primary water chemistry which contained no lithium. The Obrigheim
steam generators were replaced in 1983 as a result of PWSCC at
tubesheet roll transitions.
FIGURE 7-2
PWSCC Locations in Typical Westinghouse-Design
Steam Generators
t Transition
Cracks
•//•' .... Opposite
Row 1 and 2 U-Bends
Dented Tube Support Plate

Intersections
Most Cracking Axial

But Some Circumferential
Hard Rolled
Areas
" In 1976, PWSCC was identified at tube support plate intersections and
U-bends in several plants affected by severe denting at tube support
plates. The plants most seriously affected were Surry I and 2 and Turkey
Point 3 and 4 (7_.8 7.18). This PWSCC occurred in material which is
moderately resistant to PWSCC, but the severe denting produced strains
of several percent in the tubing. Rapid PWSCC would probably not
have occurred at these plants in the absence of denting. Steam
generators were replaced at all four of these plants between 1980 and
1983.
* In 1976, a leak occurred in a row 1 U-bend at Takahama 1 without the
occurrence of tube support plate denting (7.19). This was followed in
1978 by a similar leak at Trojan (7.20). The Takahama and Trojan PWSCC
was traced to localized plastic deformations of the tube at the "opposite
transition." The opposite transition is a deformed area at the transition
between the U-bend and straight tube where the ball mandrel, used by
Westinghouse to fabricate row I and 2 U-bends, stops up upon
completion of the bending operation. This area may be characterized
by severe changes in tube ovality (7.20). A row 1 U-bend rupture in the
apex region at Doel 2 in 1979 was attributed to excess ovality (7.21).
Since these occurrences, many plants have experienced leaks at row 1
U-bends. Most row I U-bends made by Westinghouse have now been
plugged or stress relieved as remedial measures.
" In 1978, axial PWSCC was identified at a roll transition in a part-depth
rolled tube at Doel 2 as a result of a primary to secondary leak (7.22).
Since 1978, more than half of the tubes in the Doel 2 steam generators
have developed PWSCC in the roll transition region. However, there
has been no PWSCC in the Doel 1 steam generators which have tubing
supplied by the same manufacturer, were manufactured by the same
vendor, started up the same year and operated at the same hot leg
temperature as Doel 2. The major difference between the two plants
appears to be the tube material microstructure. Since 1978, PWSCC
has been detected in over half of the W/L design plants with roll
transitions. In several cases, over half of the tubes in a steam generator
are affected.
* Since 1981, circumferential stress corrosion cracks have been detected
in expansion transitions of several plants under a variety of conditions.
Circumferential cracks are of special concern since they may have
increased potential for sudden tube rupture without the warning
normally provided by low level leakage from axial cracks, and may
have more potential for whipping and causing damage to adjacent
tubes. While some of these instances involve cracking at intermittent
rolls within the tubesheet, others involve cracking at the upper

transition. Leading examples of circumferential cracking at the upper
transition include:
-circumferential cracks of explosively expanded tubes in the sludge
pile region. at Fessenheim 1 (7.23)
-circumferential cracks in improperly fabricated kiss roll expansions
at Dampierre 1 (7.23).
-short circumferential cracks associated with axial cracks in roll
transitions at Doel 2 and Ringhals 3 and 4 (7.23 7.24).
-circumferential cracks associated with denting at the top of the
tubesheet in several EdF plants (7.25).
-circumferential cracks at explosive expansions at North Anna 1, North
Anna 2, and Trojan (7.26).
PWSCC has also occurred more recently in alloy 600 components (e.g.,
nozzles, sleeves, housings, etc.) elsewhere in the primary circuit.
EXTENT OF PWSCC DEGRADATION

Table 7-1 shows typical plants which have been affected by each of the
main types of PWSCC degradation. This data shows that some degradation
mechanisms have occurred at only a few plants and that other mechanisms
are widespread.
Table 7-2 shows the number of plants with W/L-fabricated or -designed
steam generators with reported PWSCC in the expansion transition region
as of August 1988. This data shows that over 40 percent of the plants had
reported some expansion transition PWSCC as of that time. Figure 7-3
shows the number of tubes plugged for expansion transition and U-bend
PWSCC per year since 1977. This data suggests that the rate of tube plugging
for PWSCC is increasing. Table 7-3 shows that PWSCC has affected over
half of the steam generator tubes in worst-case plants after only ten years
of operation.
Figure 7-4 shows a statistical (Weibull) plot of the fraction of tubes in
EdF steam generators with eddy current indications of PWSCC in kiss roll
expansion transitions (7.28). This data shows that the extent of PWSCC
can vary widely in plants with similar tubing, similar fabrication practices,
similar operating conditions and similar inspection techniques. However,
the data also suggests that a significant percentage of the tubes in all of
these plants will ultimately be affected by expansion transition PWSCC.
TABLE 7-1
Types of Steam Generator PWSCC and Typical Plants Affected
Location Expansion Transitions-Longitudinal PWSCC

Full Depth Full Depth Part Depth Intermittent
w/o DAM With DAM
Sketch
Typical Aimaraz 1 Bugey 5 Doel 2 Obrigheim

Plants V.C. Summer Dampierre 1 Mihama 1 Doel 2*
Affected Bugey 3, 4 Doel 3 Ringhals 2 Zorita**
Ohi 1 Tihange 2 Takahama 1 Conn. Yankee
Many others Ringhals 3 Conn. Yankee
Many others Many others
*Attempted repair using hybrid mechanical-hydraulic expansion at mid-height.
**Top-expanded area 2-4 inches above top of tubesheet.
Location Expansion Transitions-Circumferential PWSCC

Full Depth Full Depth Part Depth Intermittent
w/o DAM With DAM
Sketch
Typical Explosive Dampierre 1V Doel 2 Obrigheim**

Plants Expansion Tricastin 3 Ringhals 2 Zorita**
Affected -Fessenheim 1 Several others
-North Anna 1
-North Anna 2
-Trojan
Roll Expansion
-Ringhals 4
*Improper rolling at Dampierre 1.
**Cracks at bottom roll transitions.
TABLE 7-1 (cont'd)

Types of Steam Generator PWSCC and Typical Plants Affected
Expanded Areas
Location Longitudinal Circumferential
IGSCC IGSCC
Sketch
F1 i
Typical Almaraz 1 Doel 3

Plants V.C. Summer
Affected Dampierre 1,2
Doel 2,3
Ikata 1
McGuire 1
Ohi 1
Ringhals 2
Tihange 2
TABLE 7-2
Percentage of Westinghouse-Design Steam Generators Affected by
PWSCC (7.27)
Tube Expansion Mfg./ Plants With Total % Plants

Method Location PWSCC Plants Affected
Partial Depth USA 7 19 37
Expansion Japan 2 4 50
Europe 4 7 57
Full Depth Westinghouse 14 44 32
Expansion Mitsubishi 3 9 33
Framatome 20 48 42
Cockerill 4 4 100
Total of All Plants 58 139 42
FIGURE 7-3
Number of Tubes Plugged for PWSCC in Westinghouse-Design
0
Steam Generators (7.27)
3000
2500
Number of
2000 ll i_
Tubes Plugged
per Year 1500
for PWSCC
1000
500
0
1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989
Year
Primary Water Stress Corrosion Cracking 7.1 1
TABLE 7-3
Extent of Roll Transitions With PWSCC Indications at Plants With
>10% of Tubes Affected (Z.70)(1)
Years of
%Tubes Year Operation
With Data When
Plant S/G PWSCC( 2 ) Reported Reported
Bugey 5 1,2 51,52 1987 7
Dampierre 1 1,2,3 22, 79, 37 1988 8
Dampierre 2 2 12 1987 6
Dampierre 3 2,3 18,11 1988 7
Doel 2 A, B =80, 74(3), >60 1989 14
Doel 3 G, R, B 35, 40, 57(4) 1989 7
Gravelines 1 1,2,3 23, 37, 30 1988 8
Gravelines 2 1,2 10,14 1988/89 8/9
Gravelines 4 1,2 15, 15 1988 7
Ringhais 2 1,2,3 70, 20, 18 1988 14
24,12, 18(4)
Ringhals 3 1,2,3 1989 8
1 11 (4)
Ringhals 4 1989 6
V.C. Summer 1 A 13 1990 6
Tihange 2 1,2,3 38, 33, 42(4) 1989 6
Tricastin 1 3 31 1987 7
Tricastin 3 1,2,3 20, 22, 23 1988 8
Tricastin 3 1 19 1988 7
1
2
Table published inApril 1990.
Based on bobbin coil inspectionunless otherwise noted.
3
Value in parentheses based on rotating pancake coil (RPC) inspection of more
than
4
100 tubes.
Based on 100% RPC inspections.
It should also be noted that several plants with W/L-design steam

generators have not experienced significant PWSCC to date. Two of the
most significant examples of plants which remain essentially free of PWSCC
are Prairie Island 2 (14 years of operation) and Doel 1 (15 years of operation).
Prairie Island 2 has both Westinghouse and Huntington Alloys tubing and
operates at a relatively low temperature, 599°F (315'C). The first indications
of roll transition cracking at Prairie Island 2 were reported in 1989 after 14
years of plant operation. Doel 1 and Doel 2 provide an interesting
comparison. These plants are essentially identical units, each having two
steam generators fabricated using mill annealed Mannesman tubing, and
both plants are operated by the same crew. Doel I has been free of PWSCC
for 15 years, while Doel 2 has experienced serious PWSCC problems which
started a few years after the plant went into service. The most significant
difference between the two units is the grain size of the alloy 600 tubing.
Doel 1 steam generator tubing has larger grain size which may have resulted
from a higher final annealing temperature. Doel 2 steam generator "A"
tubing has the smallest grain size and the greatest PWSCC problems. Doel 2
steam generator "B" tubing, on the other hand, has intermediate grain size
and intermediate PWSCC behavior.
FIGURE 7-4
Fraction of Tubes in EdF Steam Generators With Reported Kiss Roll
Expansion PWSCC (7.28)
French Steam Generators

.90 Veibull Plot
0- - -. 0.63
.......... Dampierre SGG2 .......... . ......... .50
- . .--
-..- ---- -"
. . .. .•y . . . . 0.2
.10
Fraction .... ... ... ... ....
-- ... .. . ... .05
of Tubes
With
Indications A
.01 1
.001
Service lime (EFPY)

DESCRIPTION OF PWSCC DEGRADATION

As previously noted, PWSCC has been reported at: (1) expansion
irregularities within the tubesheet, (2) tubesheet expansion transitions,
(3) dented tube support plate intersections, and (4) transition and apex
regions of row I and 2 U-bends. Below is a general description of the
degradation phenomenon and its location.
Cracking at Expanded Regions Within Tubesheet

A number of plants have reported cracks within the expanded region
for tubes expanded by a multiple-step rolling procedure. One of the better
documented examples, is Doel 3 (7.10). After about 10,000 hours of
operation, leaks were discovered at three tubes. It was determined that the
tubes at these locations were installed in oversize holes and had not been
fully expanded to provide proper contact between the tube and tubesheet.
Subsequent examinations showed cracks at roll overlaps, skip rolls, and
non-overlap areas in tubes installed in normal size holes (7.10). Most of the
reported cracks were short and longitudinal; however, some of the cracks
have a circumferential component.
Further evaluation of tubes in holes of specified nominal tolerance at
Doel 3 and Tihange 2 showed ID profile "waves" approaching the
manufacturing limit. While the cause of the "waves" was not determined,
the evidence suggests that the rollers were not parallel. This could possibly
result from metal pickup on rollers or some other similar phenomenon.
The dimensions of the "waves" were such that the peaks of the "waves"
may not have been placed in compression and thereby could have residual
tensile stresses.
Longitudinal and circumferential cracks have also occurred in the
transition regions of intermittent roll expansions at Obrigheim as shown in
Figure 7-5 (7.16). This cracking resulted in leaks in 227 tubes in two steam
generators and the preventive plugging of an additional 180 tubes. A related
problem at Obrigheim was local buckling of tubes between the intermittent
rolls as shown in Figure 7-5. The buckling was attributed to leakage of
water into the annulus between the tube and tubesheet through small cracks
followed by expansion of the water during plant heat-up cycles. It has
been hypothesized that the small cracks closed up during the heat up cycle
thereby trapping high pressure water on the outside of the tube. High
local stresses at the bulges resulted in cracks at these locations.
FIGURE 7-5
Bulge and Cracks at Intermittent Rolls in Obrigheim
Steam Generator Tubes (7.16)
Cracking at Expansion Transitions

Steam generator tubes in all PWR steam generators are expanded into
the tubesheet and then seal-welded to the cladding on the primary side of
the tubesheet. Two key factors should be considered regarding PWSCC at
the transition between the expanded and unexpanded tube: (1) depth of
the expansion transition within the tubesheet and (2) the expansion method.
Depth of Expansion Transition Within the Tubesheet. In the earliest steam

generators, tubes were expanded over a short length of 2-4 inches
(50-100 mm) at the bottom of the tubesheet. This left a crevice between the
tube and tubesheet which proved to be a source of secondary side IGA/
IGSCC. During the early '70s, designs were changed to provide an
expansion over the full tubesheet depth. Some plants, which were originally
given a part-depth roll expansion, were re-expanded to essentially full depth.
Some of the re-expansions were performed in the field. In the case of several
Japanese plants, tubes were re-expanded to within 2 inches (50 mm) of the
top of the tubesheet.
The expansion depth has a significant effect on safety evaluations and
the need for remedial measures. Specifically, cracks at the expansion
transition in tubes with full-depth expansion pose a more serious safety
concern than similar cracks in tubes with part-depth expansion. This is
because the tubes in part-depth expansions are captured within the
tubesheet such that a worst-case circumferential rupture of a single tube at
the expansion transition should not lead to a high leakage rate or to tube
whipping.
Expansion Method. The main expansion methods used in PWR steam

generators are:
Roll Expansion. Roll expansion involves inserting a tool into the tube
and rotating and pressing on an inner tapered mandrel of the tool, which
in turn moves upward and pushes three to five tapered rollers into
contact with the tube wall. The tube is expanded into contact with the
tubesheet until a specified torque level is reached. For the case of full-
depth roll expansions, high stresses can exist at the top most expansion,
at elevations where rolling was inadvertently missed (skip rolls) and at
irregularities such as roll overlaps.
Part Length Roll Followed by Re-Expansion to Full Depth. Many
plants which were originally fabricated with part-depth roll expansions
were subsequently re-expanded over the full depth to eliminate the
crevice. This re-expansion was often performed in the field. Methods
of re-expanding the tube over the full tubesheet depth included
explosive expansion, mechanical roll expansion, hydraulic expansion,
and elastomeric expansion.
Mechanical Rolling Followed by DAM or "Kiss Roll". Some European
utilities were concerned that the mechanical rolling process would
produce high residual stresses on the outside surface of the tube at the
expansion transition. Therefore, a short length of intermediate
expansion was made above the fully expanded region at most European
plants to reduce the OD stresses. This process is called "dudgeonnage

ameliore mecaniquement" (DAM) or "kiss roll" and is illustrated in
Figure 7-6. This process was applied to steam generators fabricated by
Cockerill and Framatome starting in the late 1970s, and to the Ringhals 3
steam generators which were fabricated by Westinghouse.
FIGURE 7-6
DAM or "Kiss" Rolling (from TRACTEBEL)
DAM Rolling (19 mm)

Explosive Expansion. Explosive expansion involves inserting an

explosive charge into the tube and detonating it. The resulting
detonation acts to expand the tube into contact with the hole in the-
tubesheet. Combustion Engineering steam generators have had their
tubes expanded over the full depth of the tubesheet using an explosive
process called "explansion". The C-E "explansion" process was
performed in the shop during fabrication.
In the mid 1970s, Westinghouse developed an explosive expansion
process to increase the original part-depth roll expansions to full-depth
expansions. This process was called the "Wextex" process and was
applied in the field to steam generators at 11 U.S. plants and one French
plant. In some cases, the Wextex expansion was performed with the
steam generators in the horizontal position prior to installation. In other
cases, the Wextex expansion was performed with the steam generators
in the vertical position after installation.
Hydraulic Expansion. Starting in the late 1970s, tubes in Westinghouse-
fabricated steam generators have been hydraulically expanded over
the full tubesheet depth. The hydraulic expansion process was done
using commercially available equipment and utilized a nominal water
pressure of 207 mPa (30 psi). The hydraulic expansion mandrel was
designed to keep the expansion 0.25 inch (6 mm) below the top of the
tubesheet.
As will be discussed later, all of these expansion processes involve
deforming the tube plastically. The residual stresses resulting from the
deformation differ in both orientation and magnitude between the various
methods. In general, rolling processes, which produce surface burnishing
(cold work) and non-axisymmetric deflections, are considered to produce
higher inside surface residual stresses than hydraulic or explosive expansion
processes for the same tubing properties.
Longitudinal Cracks at Expansion Transitions. Longitudinal cracks at

hot leg expansion transition were first detected at Doel 2 in 1978 after only
three years of operation. Doel 2 tubes were part-depth roll expanded in the
tubesheet. Similar cracks have subsequently been detected in other plants
with part-depth roll expansion, full-depth roll expansion, and full-depth
roll expansion with DAM treatment.
Perhaps the best documented evaluation of removed tubes with axial
PWSCC defects was performed at Ringhals 2 (7.30-Z.32). Ringhals 2 tubes
were fabricated by Westinghouse using mill anneal conditions which,
experience has shown, result in PWSCC susceptibility. The tubes were roll-
expanded over the bottom 2.0-2.4 inches (50-60 mm) of the tubesheet depth.
About six years after the plant went into operation, IGSCC was detected on
the inside surface of the tubing in the expansion transition zone. A total of
18 tubes were pulled from the steam generators during the period '81-'83
to investigate the cause of the IGSCC (7.31). Primary side cracks were found
on 12 of 15 tubes examined. Typical cracking patterns and the corresponding
tube expansion profiles are shown in Figure 7-7. As indicated in the figure,
the cracks are axially oriented and are about equal in length to the transition
zone or less than 0.4 inches (10 mm). As shown in Figure 7-7, there tend to
be many parallel cracks around the circumference. On some tubes, the
multiple, parallel axial cracks have a slight spiral orientation possibly
suggesting that some non-hoop residual stress is present from the rolling
operation. The cracks appear to be slow-growing, suggesting that crack
growth slows once the crack propagates into a zone of lower stress.
Experience has shown that the cracks grow through the wall thickness and
generally cause low volume leakage. Similar cracks have appeared in
Ringhals 4 as shown in Figure 7-8.
Figure 7-7 also shows that the number of cracks around the tube
circumference is a function of the roll transition profile. Specifically, the
gradient (r/l) of the transition defined by the radial expansion (r) and the
transition length (1)has a decisive influence on the extent of cracking (.32).
Tubes with a severe geometry (high r/l ratio) in the transition have more
cracks.
Plants with kiss roll transitions also tend to develop many short parallel
axial cracks. Figure 7-9 from Ringhals 3 shows that the cracks tend to
develop first at the lower transition. Note in this figure that there is a short
circumferential crack in addition to the many axial cracks.
EdF has performed SCC tests of sensitized alloy 600 roll expansion
specimens in sodium tetrathionate to assess PWSCC at the expansion
transitions. Figure 7-10 suggests that, while the cracks start out longitudinal,
they can develop a circumferential component with time. However, there
have only been isolated reports of circumferential cracking in roll transitions
in the field to this date. The EdF tests also suggest that for tubes given the
DAM treatment, cracking initially starts at the lower transition, then starts
at the upper transition, and finally the cracks at the two transition elevations
can join together making a longer crack. This testing also suggests that the
DAM process may delay the initiation of PWSCC, but does not prevent
PWSCC in susceptible material and may ultimately lead to longer cracks
than for a roll expansion without DAM treatment or "kiss roll."
Figure 7-11 shows the cracking patterns which developed during high
temperature steam tests of pressurized roll transitions at Westinghouse
plants (7.34). The cracks are primarily axial, but circumferential components
develop at the transition between the tapered region and the unexpanded
tube.
FIGURE 7-7
Expansion Transition PWSCC at Ringhals 2 (7.31)
30 300 240 1800 0 .2 .4 mm

Location of Cracks vs. Rolling Profile
07.5 mm •
180'
Crack Thickness
Iqo • •qud•lll aml
0 010 020 030 040 050

Cracks vs. Transition (0i) Ratio
FIGURE 7.8
Ringhals 4 SG1 R17-C81
Flattened Tube Profile EC-Testing
T' 565 550 ' 565

E 560 545E 560
555I 540 0 555
550
Z0545 , , I 5530lii
5 ... I 550
5545
4
I
540 - 50 "o 0 "
3
SS5450 ....1,....
90 180
..... ....
270 360 525
0
L .i
0.1 0.2 0.3 •
540
0 90 180 270 360
Circumferential Position [degree] r [mm] Circumferential Positiorn [degree]
Stereo Microscope Penetrant Flaw De tection
T" 565 565

•_56o E 560
E555
0 555
550 , I i
545 111 .I I . .'I3 E550
50I
a545 I i
- 540 - 540
5351 ....... 9, , ........ . .,,.... ........
'5535• S90 180 270 360
90 180 270 360
Circumferential Position [degree]
Circumferential Position [degree]
SEM Ultrasonic Testing
T 565 Ef 565
2 560 560
555
550
545
I I II II 555
550
L 545
540 540
5350 535,
90 180 270 360 360
Circumferential Posittion [degree] Circumferential Position [degree]
0
FIGURE 7-9
Flattened Tube Ringhals 3 SG 2 R9-C13 EC-Testing

Profile
T 565 .-.. . . .-... . . . . 565 565
,E..5560
5551 5551-
560- C 560
555
550 550- (n
550
n 545 I I I, ' I - ], I 5451- I a0'h
540 5401-1 I 7-a 545
535 (0'.......
L9 .......1........ . ......... 535 540
5350 .. ..... I ........ I .I
90 180 270 360 0 0.1 0.2 0.3 90 180 270 360
Circumferential Position [degree] r [mm] Circumferential Position [degree]
Stereo Microscope Penetrant Flaw Detection
W 565 Flattened Part E 560 CtL

. 560 of Tube C 560
555 0 555
a-
550
545
- ~ &
550
550
Metallographic (. 545 0'
to
.- 540 Sections a 540
535
S535 '360< 180 270 360
90 180 276 ..... 0 90 -M
Circumferential Position [degree] Circumferential Position [degree] (n
SEM Ultrasonic Testing
T 565 565 -
E 560 560 ML
. 555
555
• 550 No Indications -•550
0 0
d- 545 (L545
Fz 540 I,
a 540 'X5350 I I... . I. I. . i...
. , . I. ,. , I, I __
,. .
,,,I, ,,,,,,I,,,,
535L 90 180 270 360
180 270 360
Circumfe rential Position [degree] (Circumferential Position [degree]
FIGURE 7-10
Results of SCC Tests of Sensitized Alloy 600 in Tetrathionate for Roll
Transitions With and Without DAM Treatment (from EdF)
2.a. Without Dam 2.b. With Dam

Roll Expansion-3 to 4% Roll Expansion plus
Wall thickness reduction DAMA 0- 0.12 to 0.16mm
V
FIGURE 7-11
Surface View of ID Cracks in Alloy 600 Tube at Both Transitions of
Specimen DE-492 (3/4 in. OD, Low MA, AE) After Splitting and
Flattening the Tube. Specimen exposure was 763 hours in 750°F
(7980 C), 3,000 psig steam with 11 psia hydrogen inside the tube and
1,500 psig steam outside the tube.
8r~a
iKAI-F1~
The current status is that axial PWSCC has been reported in most types
of tubes with roll expansion, including hot leg expansion transitions in part-
depth expansions, full-depth expansions, full-depth expansions with DAM
treatment, and transitions which were rotopeened or shot peened after the
plant went into service. PWSCC has also been reported in small numbers
of cold leg roll expansion transitions. In addition, this type of PWSCC has
been observed in material with a variety of microstructures, most severely
in low temperature mill annealed material, but also in higher temperature
mill annealed and thermally treated material.
To date, longitudinal PWSCC has not been confirmed (by examination
of pulled tubes) in tubes which were explosively or hydraulically expanded,
or roll expanded and shot peened prior to operation. However, there has
been limited operating experience with these configurations. One plant

with explosively expanded tubes (Trojan) has reported that axial PWSCC
has been detected by RPC eddy current examination.
Circumferential Cracks at Expansion Transitions. Circumferential cracks

are of special concern since they have the potential to grow to 3600 of the
tube circumference and thereby increase the risk of sudden tube rupture.
In addition, tubes which fail due to circumferential rupture could potentially
damage adjacent tubes by whipping. Fortunately, circumferential cracking
at hot leg expansion transitions has occurred much less frequently than
longitudinal cracking and most circumferential cracking has been non-
uniform around the tube circumference such that it should exhibit leak
before break behavior.
Circumferential PWSCC at the upper expansion transition (as
distinguished from expansion transitions within the tubesheet) was first
reported at Fessenheim I (7.23). Fessenheim I steam generator tubes were
originally rolled over the bottom 3.15 inches (8 cm) and were then
explosively expanded over the full depth using the Wextex process.
Circumferential cracks were first detected on the ID of tubes in one generator
about four years after the plant went into service. As of the summer of
1989, about 85 tubes in this steam generator have been plugged due to these
cracks. The cracks shown in Figure 7-12 have been isolated to one small
area in the hot leg sludge pile region. The cracks range from about 0.5 inches
(1.2 cm) above the top of the tubesheet to 0.75 inches (1.9 cm) below the top
of the tubesheet and have resulted in a leakage rate of 12 gph (48 liters/hr)
or less. It was reported that there was some bending of the removed tubes.
However, this was not attributed to the tube removal process. While no
specific cause of the circumferential cracking at Fessenheim I has been firmly
identified, it is speculated that it may be associated with denting at the top
of the tubesheet.
Circumferential PWSCC has been reported at several plants with roll
expanded tubing. The first case was at Dampierre 1 (7.23). After three
years of operation, a significant circumferential defect was found in the roll
transition of a tube removed due to a leak. The circumferential defect, shown
in Figure 7-13, was found to extend around the tube and penetrate 75-90%
through wall. The defect was located between the DAM roll and the last
main roll. EdF determined that this defect was the combined result of the
top main roll being too high above the top of the tubesheet secondary face
9
and the DAM roll having too largean expansion. EdF preventively plugged
the remaining tubes which had this configuration.
FIGURE 7-12
Circumferential PWSCC in Explosive Transitions at Fessenheim 1
~ s-...,............................................... 0
...................
....... ...
>. ........................... 0
t0 o.,...............f ......................
o .,,,,..,,,,.,..,,,,,,.4 ................ ......................... t
co
.~ ~-. t
O.............. ~ .
0. to tttt
t................ .
o2 .... .. ... .. .. ..
. . . . 0.........0to .. .. .. ... 0 o 0 o
0 0 0 00 00 0
Tubes With Circumferential Cracking (11981 -1989 Inspections)

Figure 7-13
Circumferential PWSCC in Kiss Roll Transition at Dampierre 1 (7.23)
Dam Roll-Over
Expanded (0.4mm
vs. Design of 0,15mm)
Circumferential
3600, 75-90% Top Main Roll

Top of Thru Wall Extended Above
Tubesheet Top of Tubesheet;
Should Stop Below
Top of Tubesheet
Subsequent to the Dampierre I case, circumferential PWSCC has been

confirmed in a number of plants with roll transitions: Ringhals 3 and 4
(7.24), V. C. Summer (Z.35), and several EdF plants (.25). The Ringhals
experience (Figures 7-8 and 7-9) suggests that small circumferential cracks
can develop along with the large number of short axial cracks in a roll
transition, and that, under certain circumstances, circumferential cracks can
develop without the presence of axial cracks.
Circumferential cracks have recently been reported in explosive
transitions at North Anna 1, North Anna 2, and Trojan (7.26).
There has been only one reported case of circumferential cracking in
plants with hydraulic expansions. The single exception is at Kori 2 where
the utility has reported ECT indications which could represent either
circumferential, or short axial, cracks caused by denting (7.36). However,
there is less operating experience with hydraulic expansions than with
explosive expansions. It is noted that there have been reports of PWSCC

associated with hydraulic re-expansions of original part-depth roll
expansions at Doel 2 and Takahama 1 (7.10).
Cracking at Dented Tube Support Plate Intersections

PWSCC has occurred at tubes dented by corrosion products building
up in the crevice between the tube and tube support plates. The greatest
amount of denting-related PWSCC occurred at Surry I and 2 and at Turkey
Point 3 and 4. Cracks occurred within the dented region in the first, second,
and third support plates at Surry 1 and 2 and leaks were reported. All four
Surry and Turkey Point steam generators were replaced due to denting
related problems, including, but not limited to, PWSCC.
Denting-related PWSCC has also been reported in about 5% of the tubes
at North Anna 1 and in a few tubes at North Anna 2. Two tubes were pulled
from North Anna Unit 1 and examined (7.37). Measurements taken on the
tube before removal from the steam generator showed ovalization of up to
7%. As shown in Figure 7-14, the ovalization resulted in two groups of
axial cracks located 1800 apart circumferentially, which extend axially
through the tube support plate region and for a short distance above and
below the tube support plate. Axial PWSCC with a depth of 50% through
thickness was observed in tubes with as little as 0.005 in (0.12 mm) difference
between the maximum and minimum diameters.
Cracking at U-Bends
PWSCC at U-Bends has led to several tube ruptures. However, this
problem can be prevented by plugging or stress relieving inner row U-bends
and avoiding denting at tube support plates.
Cracking at U-Bend Apex. PWSCC has occurred in the U-bend apex region
at Surry I and 2, Turkey Point 1 and 2, Obrigheim, Doel 2, and Beznau 2.
These problems have been traced to two different sources, denting at tube
support plate holes and excessive ovality.
Apex PWSCC Due to Tube Support Plate Denting. Severe denting
occurred at tube to tube support plate intersections at the Surry and
Turkey Point plants due to secondary side water chemistry problems
.(7. 18). As indicated in Figure 7-15, severe denting at large numbers
of tubes caused: (1) deformation of the tube support plate, especially
in the region of flow slots; (2) pinching of the U-bend legs; (3) high
bending stresses in the tube wall at the apex of the U-bend; (4) PWSCC
on the inside of the tube wall; and (5) ultimate tube rupture. This
situation can be avoided by preventing denting.
FIGURE 7-14
Axial PWSCC at Dented Tube Support Plate Intersection at
North Anna Unit 1
High Tensile
Licaoof
~and Axial
Cracksk
Tub Shoot
-------------------------
Bottom-------------
SP 2
30' 2100
190- 90. 10. 10. 270 ' 190.

ID-SCC Location in Flattened Tube R3 C41 at SP #2
PrimaryWater Stress Corrosion Cracking 7-29
FIGURE 7-15
Source of High Bending Stress Leading to U-Bend Apex PWSCC at
Surry
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0
2. Deformation of flow
slot causes ovalization
, nof U-Bend
Apex PWSCC Due to Excessive Tube Ovality. Non-denting-related

PWSCC at the U-bend apex has been confirmed at Obrigheim, Doel 2,
and Beznau 2. The best documented of these cases is the 1979 incident
at Doel 2 in which a tube ruptured suddenly at operating pressure and
temperature with the plant at zero power (7.21). The leak rate was
estimated to be about 150 gpm. Dobbeni has described this crack as
longitudinal, located within the U-bend apex, and with a length of
2.75 inches (7 cm) (Figure 7-16). The crack was attributed to high
stresses resulting from excessive ovality (>10%). Corrective action
consisted of plugging all row 1 U-bends with high ovality. Since that
time, additional cracks have been detected and all row 1 U-bends are
now plugged (7.38).
FIGURE 7-16
U-Bend Apex PWSCC at Doel 2 (Based on Description in 7.21)
Hot Leg Cold Leg

Cracking at U-Bend Tangent. PWSCC has been reported at the tangent

area between the straight run of tube and row 1 and 2 U-bends at several
plants. The best documented cases are Takahama 1 (7.19), and Trojan (7.20).
These bends were made by Westinghouse using an internal metal ball
mandrel in contrast to a cylindrical plastic mandrel or no mandrel used by
other fabricators (.7.19 7.39). Experiments have shown that a ball mandrel
produces an irregular transition or bump at the end of the bending operation
as shown in Figure 7-17. In addition to the bump, the tube is thinned locally
by as much as 7 mils (0.18 mm). This bump is sometimes referred to as an
"opposite transition."
FIGURE 7.17
U-Bend Opposite Transition Cracks at Takahama 1 and Trojan
a. U-Bend Crack at Takahama 1
1800
b. U-Bend Crack at Trojan

The Trojan row 1 U-bends had multiple cracks oriented essentially

parallel to the tube axis on the extrados surface. The experience to date is
that cracks of this type increase in length slowly. In the case of Trojan, the
leakage rate at one shutdown was 3-6 gph (12-24 liters/hr) and was
increasing slowly.
0
In addition to the above reported leaks, PWSCC has been detected in
row 1 U-bends at many other plants; it has also started to be detected in
row 2 U-bends at numerous plants. PWSCC of row I and row 2 U-bends
can be prevented by preventive plugging or stress relieving. The metal ball
mandrel was not used on row 3 and higher rows of tubes, and the operating
stresses are lower. It is possible that row 3 and higher U-bends could
ultimately develop PWSCC as a result of high stresses, or ovality, but there
have been no reports of this problem to date.
CAUSES OF PWSCC
Stress corrosion cracking was previously defined as the cracking type
failure of susceptible material under the combined action of a corrosive
environment and sustained tensile stress. It was also noted that stress
corrosion cracking is typically intergranular and proceeds in a direction
roughly perpendicular to the maximum tensile stress. This section provides
a general discussion of how stress corrosion cracking occurs and a more
detailed discussion of the three factors necessary for stress corrosion
cracking: (1) material susceptibility, (2) tensile stress, and (3) aggressive
environment.
General Discussion
Metals, such as alloy 600 steam generator tubing, are crystalline solids
in which the atoms line up in regular- lattice arrays upon cooling from the
molten state. Crystal growth initiates at many points simultaneously, and
the individual crystallites grow until they come into contact with adjacent
growing crystallites at irregularly shaped boundaries. Individual crystallites
do not tend to merge into a single larger crystallite at the boundaries because
their lattice axes do not match. The resultant individual crystallites are
called "grains." A key factor from the standpoint of PWSCC is that
impurities do not fall easily into place in the lattice as the grain is forming
and therefore tend to concentrate at the grain boundaries. The concentration
of impurities at grain boundaries can be 1,000 to 10,000 times that in the
center of the grains. Therefore, the resultant metal consists of individual
grains bound together by atomic forces at the boundaries and with a higher
concentration of impurities at the grain boundaries than in the centers of
the grains.
Impurities at grain boundaries generally decrease the corrosion

resistance of a material. If a material is sufficiently susceptible, and the
environment sufficiently aggressive, intergranular corrosion can occur in
which the grain boundaries are selectively corroded away leaving the grains
intact but not joined. This is analogous to removing the mortar from the
joints in a brick wall and is illustrated in Figure 7-18b. Intergranular
corrosion is a concern in secondary side crevices, but has not been a problem
on the primary side of steam generator tubes.
FIGURE 7-18
Schematic Representation of Intergranular Corrosion and
Intergranular Stress Corrosion Cracking
a. Intact
b. IGA L
=H15
C. SCC
Under ideal conditions, a protective film forms over the metal surface
which retards corrosion of the grains or grain boundaries. When a tensile
stress is imposed on a material, the resulting strain may break the protective
film, allowing the corrosive environment to come into contact with the metal.
This in turn can lead to accelerated dissolution of the more susceptible grain
boundary material as illustrated in Figure 7-18c. Once a crack is initiated,
the applied tensile stress acts to keep the crack open and allows the corrosive
environment access to the inside of the crack. Further, the geometry results
in an increased stress intensity at the crack tip. Accordingly, the crack tends
to follow the original grain boundaries into the material. Stress corrosion
cracking tends to follow a single path from the surface since the cracking
relieves the tensile stress on grain boundaries on either side of the one which
is cracked. Stress corrosion cracking can initiate at pre-existing cracks,
surface discontinuities, or at locations of submicroscopic differences in
material composition, protective film thickness, or corrodent concentration.
The effect of stress corrosion cracking is to reduce the effective strength
of the material. In the absence of conditions which lead to PWSCC, material
can be stressed to the ultimate strength statically without failing. Under
conditions which lead to PWSCC, material may fail at a much lower stress.
However, there appears to be a practical stress threshold of about 60% of
the engineering yield strength below which PWSCC does not become a
practical problem in PWR steam generator tubing. Finally, if the applied
tensile stress changes periodically, the crack can grow by fatigue as well as
corrosion. This is called "corrosion fatigue," and it has a lower stress
threshold than for stress corrosion under static load conditions.
Material Susceptibility
During the cooling which occurs after annealing, carbon combines with
chromium to form chromium carbides. The carbides precipitate within the
grains or at grain boundaries depending largely upon the material carbon
content, the temperature reached during the annealing process, the time at
temperature, the presence of prior precipitates, and the cool down rate.
In 1973, Coriou published the results of tests which indicate that heat
treatments which produce material with chromium carbides concentrated
primarily at the grain boundaries (intergranular) provide better resistance
to SCC in high purity water environments than heat treatments which
produce material with chromium carbides dispersed within the grains
(intragranular) (7.2). Subsequently, other researchers have confirmed
Coriou's conclusion (7.40, 7.44). The best resistance to PWSCC appears to
be associated with a microstructure characterized by copious quantities of
grain boundary chromium carbides (intergranular carbides) and a few
chromium carbides within the matrix of the grains as illustrated
schematically in Figure 7-19.
FIGURE 7-19
Schematic Representation of Susceptible and Resistant
Alloy 600 Microstructure
•. . *
a. Susceptible to Primary Side IGSCC b. Resistant to Primary Side IGSCC

- Few Grain Boundary Carbides - Copious Grain Boundary Carbides
- Copious Intragranular Carbides - Few Intragranular Carbides
- Small Grain Size - Large Grain Size
A good example of the correlation between carbide morphology and

PWSCC is the metallurgical examination of 15 tubes pulled from Ringhals 2
in 1983 (7.31). As shown in Figure 7-20, none of the three tubes with
significant amounts of intergranular carbides evidenced PWSCC, while all
but one of the 11 tubes which had small amounts of intergranular carbides
and significant amounts of intragranular carbides also evidence PWSCC.
The most widely accepted hypothesis for the beneficial effect of
intergranular carbides on PWSCC resistance has been proposed by
Bruemmer (7.45) and is illustrated by Figure 7-21. In essence, grain
boundary carbides promote crack blunting due to their effectiveness as
dislocation sources. The greater the number of carbides at the grain
boundaries, the more dislocation sources there will be, and the farther
dislocations will run within the grain. Another explanation proposed by
Smialowska (7.46) is that alloy 600 material passivates more readily in the
presence of grain boundary carbides.
Regardless of the particular 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. The two factors which most strongly affect the grain boundary
microstructure of alloy 600 appear to be the final annealing temperature
and the carbon content.
FIGURE 7-20
Correlation Between Carbide Morphology and PWSCC at Ringhals 2
(7.31)
No. of Tubes
F.
0
EM Intragranular Carbides
r-D Grain Boundary Carbides
4 N
0
0 1-4 5-9 10-14 15-19 20-24
No. of ID-cracks
FIGURE 21
Proposed Effect of Grain Boundary Carbides on Crack Tip Shape
Sharp-Tipped Crack:
Few Low Energy
D islocation Sources
Final Mill Annealing Temperature. For a given heat of alloy 600 material,
the main fabrication variable which controls the carbide morphology is the
final mill annealing temperature after cold working to the final tube size.
Significant factors in the tube forming and mill annealing process, which
affect material susceptibility, are as follows:
* Chromium carbides are present prior to final tube forming and final
mill annealing as a result of previous forming and heat treating
operations.
* Material is cold worked during the final tube forming operations.
* Material recrystalizes during the final mill annealing treatment and
forms new grain boundaries upon cooling.
" Dissolution of carbides during final mill annealing is a strong function
of the mill anneal temperature. At a low mill anneal temperature, many
carbides do not go fully into solution. These carbides tend to slow
down grain growth and to remain at intragranular locations after
completion of the annealing. In addition, relatively sparse decoration
of grain boundaries by carbides occurs because of the limited dissolution
of carbides and the large grain boundary surface area. At a high mill
anneal temperature, carbides are dissolved more freely, the grains grow
in size which reduces the grain boundary surface area and the carbides
re-precipitate at the new grain boundaries upon cooling. Therefore, a
high mill annealing temperature results in large grains, copious grain
boundary (intergranular) carbides and few intragranular carbides.
The effect of final mill anneal temperature on PWSCC susceptibility
has been confirmed by several researchers. For example, Norring tested
alloy 600 reverse U-bends (RUBs) in high purity water containing hydrogen
(7.47). The time to crack initiation increased by a factor of four when the
mill anneal metal temperature was increased from 1,697°F (9250C) to 1,877 0F
(1,025 0 C).
A possible explanation why some plants have experienced severe
PWSCC and others have been relatively free of PWSCC is that the final
tube mill anneal conditions have changed over time.
' Prior to the early 1970s, alloy 600 tubing used in most Westinghouse
and Combustion Engineering plants was processed using a high enough
final mill anneal metal temperature to produce recrystalization, large
grain size (typically <ASTM No. 6), and good dissolution of carbides.
At this mill anneal temperature, carbides were free to precipitate at the
new grain boundaries during cooling. This heat treatment tended to
produce a material resistant to PWSCC but with relatively low yield
strength.
" In the early 1970s, the mill anneal metal temperature used for tubing in
W/L type steam generators was lowered to about 1,750'F (955'C). This
temperature was selected to obtain a higher yield strength in the range
of 50-60 ksi (345-450 MPa). This mill anneal temperature is high enough
to cause recrystalization, but it is not high enough to cause much grain
growth or carbide dissolution. Thus, it results in smaller grain size
(typically ASTM No. 8-11). This leads to a higher concentration of
intragranular carbides and a lower concentration of desirable
intergranular carbides. This material has good resistance to acid attack,
but is more susceptible to PWSCC than material annealed at higher
temperatures.
* During the late 1970s, W/L started to use a "special thermal treatment"
primarily to increase the resistance to secondary side caustic attack.
The process involves holding the tubing at a temperature of 1,300'F
(705'C) for about 15 hours after final mill annealing. This process was
intended to produce the desirable condition of continuous grain
boundary carbides and provide sufficient time for the chromium
concentration at the grain boundaries to return to a level which will
resist attack under acidic and oxidizing conditions (i.e., not sensitized).
The special thermal treatment also serves to reduce residual stresses of
the base tubing material. The thermal treatment does not reduce the
residual stresses developed in the material during subsequent tube
bending or expansion operations, except by having reduced the yield
strength and reducing the work hardening characteristics of the
material.
" About 1980, research was completed by Airey which showed that the
desired carbide microstructure could be reliably obtained by the thermal
treatment only if a sufficiently high final mill anneal temperature is
used (Z.41). The current EPRI specifications for alloy 600 and 690 tubing
call for a high enough final mill anneal temperature to assure that a
good microstructure is obtained (7.48).
* An alternative heat treatment approach used on all B&W once-through
steam generators is to stress relieve the entire steam generator for about
10 hours at 1,1507F (620'C) after final assembly. This process results in
a high concentration of desirable intergranular carbides and relief of
the fabrication induced residual stresses. As a result, this tubing has
been free from PWSCC. However, the stress relief results in chromium
depletion at the grain boundaries (i.e., a "sensitized" condition) and
poor resistance to IGSCC under acidic oxidizing conditions. Acidic
attack has proven to be a problem with TMI-1 providing the best
example.
Carbon. The mill anneal temperature required to develop the desired

PWSCC resistant microstructure (copious intergranular carbides with few
intragranular carbides) is a function of the tubing carbon content. The higher
the carbon content, the higher the final mill anneal temperature required to
lobtain the desirable microstructure. This did not pose a significant problem
for typical high mill anneal temperatures (1,800-1,950'F [980-1,065°C]) since
these temperatures were high enough to develop the desired microstructure
for the typical ranges of carbon content supplied. However, the combination
of carbon content near the high end of the typical mill range (0.07%) and
typical low mill anneal temperatures (1,750'F [955 0C]) results in increased
PWSCC susceptibility.
EdF conducted a series of tests on reverse U-bends (RUBs) from 22
heats of production alloy 600 tubing material processed using a low mill
anneal temperature and with carbon contents ranging from 0.01 to 0.07%
(Z.43), Figure 7-22. The tests were conducted in elevated temperature pure
water and primary water with hydrogen overpressure. After final mill
anneal heat treatment, the tube materials were characterized by
microstructure where Structure I is the desirable microstructure consisting
of large grains and intergranular carbides, Structure III is the undesirable
microstructure consisting of small grains and essentially uniform carbide
distribution and Structure 11 lies between the two extremes. Three plots
were prepared of the data.
* Figure 7-22a shows the correlation between microstructure and PWSCC
susceptibility. As expected, the desirable microstructure (Structure I)
results in low PWSCC susceptibility and the undesirable microstructure
(Structure III) results in high PWSCC susceptibility.
* Figure 7-22b shows the correlation between ultimate tensile strength
and microstructure. As expected, the material microstructure which
correlates with low PWSCC susceptibility (Structure I) has the lowest
tensile strength (<717 MPa 1<104 ksi]).
" Figure 7-22c shows the correlation between ASTM grain size, carbon
content and microstructure. As expected, material with low carbon
(<0.018%) and/or large grains (<ASTM 8) has low PWSCC
susceptibility.
The EdF data suggests that not all low temperature mill anneal alloy
600 tubing will be highly susceptible to PWSCC. Rather, PWSCC susceptible
material is that which is mill annealed at a low temperature and has a carbon
content greater than about 0.018%, or small grain size (>ASTM 8) (.43).
FIGURE 22a
Relationship Between Alloy 600 Steam Generator Tubing
Microstructure and PWSCC Susceptibility (7.43)
/.-A = •V, . ,. . , _ . . ,
AA
, rI , . .•,
* #. *., . I. ,.
• I . ..-, -* ", ,-. --
a) Structure I
~ A-
4' L - :. +. - , •- .•
• 4,''•
, "- 1•i "
,. .4 a :, :- -A•, •' .: : . ,
b) Structure II
.'•" :'. , . '•'.•'

• •' • •' "÷ 'v •
•"•" ..'}
.• ' .', .; ' ." ., . • d .," ,. . . ,k
" •1 l . ' " : :. . ' .' : . . ' . . r ¢ " ,: ,
':[:-.+ •: , •,'i• i : 7 . ... :3:
.•. .'•,:!
•;:,,.- c);.Structur .I"I -' .
FIGURE 7-22a (cont'd)
* Structure I
* Structure II
* Structure IIt BC C54 C52
a'- + NNUI-L-C79
C68
6J1 6L I**
+ J'1
C62 C64 C59
6P 6J'2
S
6R C61
6E 6S BB 6J3
S,
6F 6D 6A
Ipd tuu
Type de Structure
FIGURE 7-22b
Relationship Between Alloy 600 Steam Generator Tubing Ultimate
Tensile Strength and PWSCC Susceptibility (7.43)
Sensibilit6 A la Corrosion
650 700 750

Rm 20'C (MPa)
FIGURE 7-22c
Relationship Between Alloy 600 Steam Generator Tubing Ultimate
Tensile Strength and PWSCC Susceptibility (7.43)
Teneur en Carbone (%)

(3£I]•.
L, -
I0~
Structure II
6D*
0,04 . 6R
+6 J'3 6J2 A 6J'1A A
* 6J1 AC 6 4
6E 6L
6S * C62
0,02 -
6F A C61,
6P
Structure I
tA
V i i I I I
5 6 7 8 9 10 11
G(AFNOR)
Teneur en Carbone (%)

0,i[]X
J~.
wv
C54
N
0,061.
BC 05
C59U
a
I C79
0,04, Structure I
C68
BB
* Structure III
0,021
BA
0 I I I i
5 6 7 8 9 10 11
G(AFNOR)
The effect of carbon on PWSCC susceptibility suggests that fabrication

of alloy 600 tubing for steam generator applications must be controlled in
one of three ways: (1) use a high enough mill anneal temperature to ensure
a desirable microstructure for all anticipated carbon contents, (2) control
the carbon content range to develop desired properties using a fixed mill
anneal temperature, or (3)adjust the mill anneal temperature based on the
measured carbon content of each heat of material. The second option is
incorporated in the EPRI guidelines for alloy 600 and 690 tubing (7.48).
Sensitization and Thermal Treatment. Upon forming desirable chromium

carbides at grain boundaries, the chromium concentration is depleted locally
near the grain boundaries, a condition known as "sensitization." The
sensitized zone of alloy 600 material is subject to intergranular corrosion in
environments which are both acidic and oxidizing. While the sensitized
region has poorer resistance to intergranular corrosion under acidic
oxidizing conditions, operating experience has shown this condition to have
good resistance to PWSCC and caustic-induced secondary side IGSCC, and
adequate resistance to intergranular corrosion for normal PWR water
chemistry conditions.
The resistance of the material to acidic oxidizing attack can be improved
by a thermal treatment in which the material is taken up to a temperature
below the recrystalization range and held there for about 15 hours (7,40
7.41). This process results in little change in carbide distribution but allows
time for chromium to diffuse into the depleted zones and thereby improve
the resistance to intergranular corrosion. Essentially all alloy 600 material
used for steam generator tubes, plugs and sleeves over the past few years,
and the improved alloy 690 material currently being used for these
applications, has received a thermal treatment.
The effect of thermal treatment on the resistance of alloy 600 to stress
corrosion cracking in pure or primary water environments has been studied
by several investigators. Norring (7.67) tested reverse U-bends (RUBs) of
mill annealed (MA) and thermally treated (TT) tubes in pure water with
50 ml H2/kg at 690'F (365°C). No direct comparisons of MA and T[ material
from the same heat were included. Both MA and TT samples experienced
cracking, but, as shown in Figure 7-23, the TT samples took significantly
longer to crack than the MA samples. Jacko (7.68) made direct comparisons
of MA and TT samples from several heats of material by testing RUBs in
simulated primary water at 680'F (360'C). Jacko also found that thermal
treatment had a significant beneficial effect on PWSCC resistance. Many
TT samples did not crack in 13,000 hours while the majority of the MA
samples cracked in a shorter time. Rouillon and others tested RUBs from
19 tubes from different heats of alloy 600 in the MA and TT conditions
(.69). The test environment was pure water with hydrogen at 617 or 680'F
(325 or 360'C). Results of the tests were as follows:
" RUBs from 4 tubes did not crack in either the MA or TT conditions.
" RUBs from 5 tubes cracked in both the MA and TT conditions.
* RUBs from 9 tubes showed that TT had a beneficial effect on time to
PWSCC.
• The RUB from I tube showed that TT had a detrimental effect on time
to PWSCC.
Further evaluation showed that the material which showed a
detrimental effect, and two of the materials which showed no effect of TT,
had compositions or processing considered not representative of tubing in
actual steam generators. Thus, of the twelve materials considered to be
representative, nine exhibited a beneficial effect from thermal treatment
and three showed no effect.
FIGURE 7-23
Effect of Thermal Treatment on Time to PWSCC (7.67)
99 t11rFTTi i 1TTI
I I I J
99
90 ............. ....... ... ... . ........
50 ................................................................. ............... .. .............

MA TT
S25
10 ...
. . . ... . . ..... . ...V . . .. . .. ..
I. . . .. .
•3 ,/i
ci)
] ......... ................
....... "................... ....
5 .................................. .............. ......
100 2
10O0 200 500 1000 2000 5000 10000 20000
Exposure Time [h]

Impurities. Impurities concentrate at the grain boundaries of any metal

during cooling. However, there is no conclusive evidence at this time that
normal levels of impurities permitted by tubing material specifications have
a detrimental effect on resistance of alloy 600 tubing to PWSCC.
Stress
Operating experience suggests that significant PWSCC should not occur
in alloy 600 tubing at operating condition stresses less than about 35 ksi
(240 MPa) for temperatures up to 615'F (324°C). Since design stresses
permitted by the ASME Boiler and Pressure Vessel Code are much less than
35 ksi (240 MPa), PWSCC failure would not be expected to occur due to
applied pressure and/or thermal loadings. This is supported by the fact
that essentially all PWSCC indications have occurred at locations where
(1) high residual stresses are produced during fabrication (expansion
transitions and inner row U-bends); (2) high stresses are produced as a result
of strains induced during operation (dented tube support plate intersections
or U-bends deformed by tube support plate denting); or (3) high stresses
are produced as a result of geometric abnormalities such as excessive tube
ovality, oversize tubesheet holes or improper kiss rolling. There has been
only one report of PWSCC in a straight run of tubing remote from
discontinuities and this was attributed to high local residual stresses at
fabrication induced scratches (7.71).
The following summarizes the applied operating stresses, the residual
stresses induced by various fabrication processes, the threshold stress below
which PWSCC should not be a concern over the plant life, and the influence
of stress on time to PWSCC.
Expansion Transition Stresses. Normal operating stresses remote from

expansion transitions consist primarily of pressure and thermal stresses
due to the through-wall thermal gradient. The normal operating stresses
for a typical steam generator tube are given in Table 7-4. This table shows
that tensile operating hoop stresses will range from a high of about +11 ksi
(75 MPa) for the case of part-depth expansions or full-depth expansions
with a sludge pile which reduces heat flux, to a low of about +4 ksi (3 MPa)
for a full-depth hot leg expansion without sludge pile to restrict heat flux.
This data suggests that PWSCC should be more severe in plants with part-
depth expansions or full-depth expansions with a sludge pile since these
conditions result in higher tensile stresses and higher metal temperatures.
The operating stresses can be increased as a result of tube bending, locking
at tube support plates due to denting, etc. Operating stresses at the
expansion transition are somewhat less than in the free tube due to the
effect of the tubesheet in restraining pressure expansion.
TABLE 7-4
Operating and Residual Stresses in ID Expansion Transitions
OPERATING STRESSES
Hoop Axial
Stress Stress
Parameter fki (kSul
Differential Pressure 11.0 5.5
Thermal-Gradient -7.0 -7.0
Total Pressure + Thermal 4.0 -1.5
RESIDUAL STRESSES
Max. Max.
Tensile Tensile
Hoop Axial
Stress Stress
Type of Expansion Ref. (ksi) k Sul
Normal Roll Expansion
- EdF corrosion/x-ray data 7.25 54 49
- Penn State Univ. x-ray 7.49 62 45
Kiss Roll Expansion
- EdF corrosion/x-ray data 7.25 49 38
Hydraulic Expansion
- Framatome x-ray measurements 7.50 28 35
- Westinghouse (NSD) FE analysis 7.51 30 38
- Dominion Engineering FE analysis 7.52 38 41
Residual stresses in expansion transitions have been determined by

several organizations using calibrated corrosion tests, x-ray diffraction
measurements, or elastic-plastic finite element analyses. It is recognized
that each of these methods of determining stresses has some uncertainty:
(1) corrosion tests have uncertainty as a result of difficulties in accurately
developing desired stress levels in calibration samples and errors in
determining the time to cracking as a function of stress, (2) x-ray diffraction
requires disassembly of the joint and splitting of the tube to obtain inside
surface stresses and averages stresses over a relatively large area, and
(3) finite element analyses to date have not modeled surface cold work
effects and non-axisymmetric loading associated with rolling and explosive
expansion. With these uncertainties in mind, the latest predictions of

residual stresses in expansion transitions are also given in Table 7-4. It must
be recognized that there are differences in material properties, fabrication
variables, and modeling techniques between these sets of data such that
the results are not entirely consistent. Nevertheless, these data suggest that
maximum residual stresses in roll expansions are about 20 ksi (140 MPa)
higher than in hydraulic expansions. It also suggests that hoop stresses are
higher than axial in roll expansions and lower than axial stresses in hydraulic
expansions. No analyses have been performed for the case of explosive
expansions but it is anticipated that the stresses will be closer to those for
hydraulic expansions than for roll expansions.
Dented Tube Support Plate Intersections. Stresses in dented tube support

plate intersections can be computed using classical equations for line contact
on opposite sides of a circular tube as shown in Figure 7-14. These
calculations show that yield-strength-level hoop stresses of about 50 ksi
(345 MPa) are developed on the inside surface of tubes dented inward by
only 0.004 inches (0.10 mm) on the diameter.
U-Bend Stresses. Operating and residual stresses in row 1 U-bends of

0.875 inch (22.2 mm) outside diameter tubing have been determined by a
combination of analysis, calibrated corrosion testing, and x-ray diffraction
measurements. The stresses for the case of U-bends formed using the
Westinghouse ball mandrel process are given in Table 7-5.
A finite element analysis by Cloud (7.53) also included calculation of
stresses for tubes with initial ovality. Figure 7-24 shows that tensile hoop
stresses on the inside surface of a tube can increase significantly in the
presence of ovality produced during fabrication or as a result of operating
deflections such as occurred during the tube support plate denting at the
Surry and Turkey Point plants.
Stress Threshold for LTMA Alloy 600 Material. Field experience implies
that a practical stress threshold for PWSCC of susceptible LTMA alloy 600
tubing under hot leg operating conditions is about 35 ksi (240 MPa). This
practical stress threshold is based on the fact that there has been relatively
little expansion transition PWSCC in plants with LTMA tubing and
explosive expansions after about ten years of operation. While there have
been no analyses of stresses in explosive transitions, they are considered to
be close to the stresses in hydraulic transitions. The preceding results show
maximum tensile residual stresses in hydraulic transitions to be about 35-
40 ksi (240-275 MPa) with the operating stresses increasing the total by about
4 ksi (28 MPa).
TABLE 7-5
Operating and Residual Stresses in Row 1 U-Bends Formed With
Westinghouse Ball Mandrel
Max. Max.
Tensile Tensile
Hoop Axial
Stress Stress
Type of Expansion Ref. tkSuL Sul
Normal Operating Stresses
- R. L. Cloud FE analysis 7.53 31 -
- Mitsubishi FE analysis 7.19 17 -
- EdF/Framatome analysis 7.54 36 -
Residual Stress
- Mitsubishi strain gage tests 7.19 50 -
- Penn State Univ. x-ray measurements 7.55 68 90
EdF Operating + Penn State Univ. Residual - =100 -
Mitsubishi Operating + Residual 7.19 50-60 -
The practical stress threshold of about 35 ksi is close to the elastic limit
for LTMA alloy 600 tubing and about 60% of the engineering yield strength
(0.2% offset). This is consistent with test work reported by Yonezawa (7.56)
which shows that the time to PWSCC for LTMA alloy 600 tubing increases
greatly when the applied stresses are less than 80% of the 6ngineering yield
strength.
Influence of Stress on Time to PWSCC. Test data by Bandy and van

Rooyen shown in Figure 7-25 suggests that the time to PWSCC in elevated
temperature water varies inversely with the fourth power of the applied
stress (7.58). Test data by Yonezawa in Figure 7-25 suggests that the time to
PWSCC varies inversely with the sixth or seventh power of the applied
stress (7.56). Figure 7-26 shows data by Yonezawa which suggests 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
time to PWSCC at a given level of applied stress should increase as the
material yield strength is increased by cold working. Conversely, for cases
where the residual stress is a fixed percentage of the material yield strength,
such as in an expansion transition, increasing the yield strength can decrease
the time to PWSCC.
FIGURE 7-24
Effect of U-Bend Ovality on Stress (7.53)
50
a Perfect Tube
40 o 5% Ovality
o 10% Ovality
30
• 20
10
Minor Axis
- Major Axis
0 View A-A
-10
0 30 60 90 120 150 180
0 (Degrees)
Inside Surface Hoop Stress Along Flank, Temperature & Pressure Loading
0 = Angular Distance Around U-bend From Tangent to Tangent.

Ovality = Dmajor -Dminor
Doriginal
A strain rate damage model has been proposed by Garud (7.57) which
postulates that cracking will occur when the integral of a function of the
strain rate reaches a certain value. The dependence of strain rate on stress,
temperature and the environment is modeled using constituative equations
for elastic-viscoplastic strain-hardening materials. These equations lead to
calculations with a large number of factors which must be fit to experimental
data. An alternative approach would be to fit experimental data to a curve

of the form shown in Figure 7-27 which shows the effect of short times to
failure at ultimate strength, long times to failure for stresses below a stress
threshold, and a transition region between the two extremes. The equation
for the curve can be adjusted to provide a good fit to experimental data,
and may be of the form.'m
t.[In ( -](7-1)
where,
x (Yu--Grth
%&
7 = applied stress
cGth = threshold stress
au = ultimate strength
Environment
The major environmental factors affecting PWSCC of alloy 600 steam
generator tubing are temperature; hydrogen and lithium concentrations;
and electrochemical potential.
Temperature. Temperature is a significant environmental factor influencing

the initiation of PWSCC in PWR environments. This is evidenced by the
fact that the vast majority of PWSCC in steam generator expansion
transitions has occurred on the hot leg side of the tubesheet. Since the hot
and cold leg transitions are essentially identical in all other respects, the
50-70°F (30-40°C) temperature difference between the hot and cold legs is
enough to account for the significant difference in time to PWSCC. For
example, over 50% of the hot leg expansion transitions developed PWSCC
indications in the Doel 2 steam generators after five effective full power
years of operation, yet there was essentially no PWSCC at cold leg expansion
transitions.
Temperature is believed to affect the rate of PWSCC attack in accordance
with the activation energy model for thermally controlled processes,
e-Q/RT, where Q is the apparent activation energy, R is the gas constant
and T is the absolute temperature. Figure 7-28 from Bandy and van Rooyen
(7.58) shows that SCC of alloy 600 RUBs in elevated temperature pure water
follows the activation energy model and suggests an apparent activation
energy of 33 kcal/mole. The current consensus, based on additional testing
and correlation with field experience, is that the apparent activation energy
falls in the range of 40-60 kcal/mole and many current predictions are based
on 50 kcal/mole.
FIGURE 7-25
Correlation Between Stress and Time-to-PWSCC for Alloy 600 Steam
Generator Tubing (7.56, 7.57)
10 .
As-received Alloy 600 Tubing. 3650C. Pure H20

Slope of (TF- k Stress') - -4
0r 4
C.W A R _0_345-C (AR.)
U3
o
1 10 100 1000
Failure, Days
100
90
< 80
x-. 70
a-c a -7
A 60
E.2 50
0.2% Offset Proof Stress at R.T 3.3kgf/mmn

0<0 40
102 10' 101
Stress Corrosion Cracking Testing Time. (h)
I Crack No Crack I
6
a"
70
60
U)
50
40
C,)
CL 30
20 0.2% Offset Proof Sfress
10 at 380'C
0
102 101 101
Stress Corrosion Cracking Testing Time (h)
FIGURE 7.26
Effect of Applied Stress to Yield Stress Ratio on Time-to-PWSCC for
Alloy 600 Steam Generator Tubing (7.56)
AppliedAple
Material Stress Stress Corrosi on Cracking en Stess Stress Corrosion Cracking ,
Heat ICl (kgf/mrr?) Testing Time(h) -• (kgf/mn) Testing Time(h)
o
IC (9Q Treat-WCorkin
Working 052 .o00o0 oIooooIoo o . 0.e50 ,Ptg *o*in
,= - -o--
mnf.n - i.
, ment " e IQ•
-- 24 eso
GSCc
i Tensile ti%813704'
45 0 6,98 2
Strained 01121 IGSCC .
Tensile
Strained 57 2 793) 4 ,,aGSCC
CCSCC
0.07 lAnnea. Tensile 72 57

ling Strained (3.231 .:'r, IGscc!
209/.o 7IGC
Cold 8 tzIGC --
RolledI
I
r no crack I crack
FIGURE 7-27
Probable Relationship Between Applied Stress and Time to PWSCC
Log (Time)
FIGURE 7-28
Correlation Between Temperature and Time-to-SCC of Alloy 600
Tubing (7.58)
Temperature, °C 365
290 315 325 345 360
1000 I I I -
SCC in Pure Water. U-Bends of Commercial Tubing.
\N\
100 4
Points o
a) a
0-2
33 kcal/mole 8
r 0.94
10 Projected a
Failure at 0
290°C.330 Weeks
1.80 1.76 1.72 1.68 1.64 1.60 1.56
Inverse Temperature, 1 x 1000
Hydrogen. Bandy and Van Rooyen (7.58) and Airey (7.41 7.42) have shown
that dissolved hydrogen increases the rate of PWSCC at hot leg
temperatures. More recently, Norring has reported on tests to determine
the effect of hydrogen overpressure in 626°F (330'C) water (7.59). Results
of the tests, shown in Figure 7-29, suggest that the rate of PWSCC increases
with increasing hydrogen overpressure. Recent tests by EdF confirm this
effect with concentrations from 0 to 150 ppm dissolved hydrogen. However,
with increased concentration (i.e., 757 ppm) susceptibility to cracking was
decreased (7.72).
Lithium. Norring and Westinghouse have been conducting tests to

determine the IGSCC resistance of alloy 600 as a function of the lithium
concentration of primary water (7.,59 7.60). Results of these tests suggest
that the addition of lithium tends to increase the SCC aggressiveness at

typical hot leg temperatures. Norring's results are shown in Figure 7-29.
Electrochemical Potential. PWSCC can be significantly affected by the

electrochemical potential (ECP). As shown by Smialowska in Figure 7-30,
small changes in the ECP can have a large effect on PWSCC (7.46). However,
the electrochemical potential at specific locations in the primary loop is not
well known, and no practical methods for controlling it have been identified.
RATE OF PWSCC PROGRESSION

Field experience has shown that early plants with relatively low PWSCC
susceptibility mill annealed alloy 600 tubing and lower operating
temperatures have not been seriously affected by PWSCC. However, there
are reports of some minor PWSCC in the earliest plants, such as Zorita and
Connecticut Yankee, after 20 years of operation.
Field experience has also shown that many plants with low temperature
mill annealed alloy 600 tubing, which operate at high temperatures, are
highly susceptible to PWSCC. However, the time to first PWSCC detection,
and the rate of PWSCC degradation, vary considerably from plant to plant
and generator to generator. The best example of the inherent variability is
Figure 7-4 which represents the time to a given percentage of tubes with
PWSCC in EdF steam generators with kiss roll expansion transitions. There
is a large scatter in the data despite the fact that the plants all have similar
tubing, similar fabrication procedures, similar operating procedures, the
same hot leg temperature and the same inspection sensitivities, etc.
Nevertheless, the conclusion from this data is that all of these generators
are likely to develop significant hot leg expansion transition PWSCC over
the plant life time.
Figures 7-31 and 7-32 show the number of tubes affected by roll
expansion and U-bend PWSCC as a function of effective full power years
of operation for several well documented plants with low temperature mill
anneal tubing.
FIGURE 7-29
Effect of Hydrogen Overpressure and Lithium on Time-to-IGSCC of
Alloy 600 Steam Generator Tubing (7.59)
B&W 170OF 3/4"
Temperoture ........... 329.9 C 329.6 C E ... !..... .... !..

2 ... ...... . .......
50 . . .. .. . . . . .
Boron ...................... 1471 ppm 1441 ppm ....:...... i..... . . . . ! . .

Lithium ................... 2.42 ppm 2.38 ppm ; ...... .... . .... .-
..
Hydrogen content... 13.1 mt/kg 25.2 mi/kg
activity,.. 7.0 kPo 13.6 kPo

7.351 0 00
pH .......................... 7.354 100 200 5 1000 20 S Q1000
B&W 170OF 7/8" (6) 6kw 1700F 7/8" (6A)
23 ..... :....... ..... :/. ..... *;. .
010 . .. . ..... . .
100 200 500 1000 2000 5M000OOO 100 Ted l'o SC

1300 0000
Expa-ýe time (h] E.- tin- (h)
(a) The Influence of Hydrogen on the Tendency to PWSCC
B&W 170OF 3/4"

gg i • , .. .... .. .,....
Temperature ........... 328.9 C 329.6 C
o .... ...... ....... ..... ... .?.....
Boron ...................... 1241 ppm 1441 ppm E
Lithium .................. 3.54 ppm 2.38 ppm
Hydrogen content... 24.7 ml/kg 25.2 ml/kg

•
.. ...... .... . . ...
13.6 kPa
octvity... 13.6 kPo
pH .......................... 7.575 7.351 100

•..:. -.. . . ...... i. . .
S.-. 200 500 1000 =0 500010000
twoe~we time (h])
B&W 1700F 7/8" (6) B&W 170oF 7/B" (SA)
.::
at
3.....
...
30 . .
.......
10 .. .. . . . . .
. . .. . . . .
.. .
30 ....
I. 10..
.......
...... ....
100 200 50 1000 50 0 100 2 500 1000 2W 000010w0

E.0sue - [h]
"time
Eq [h
(b) The Influence of Lithium on the Tendency to PWSCC

FIGURE 7-30
Effect of Electrochemical Potential on the Rate of PWSCC in
Alloy 600 Material (7.46)
100I I
o PH2=0.005MPa
* PH2 =0.1MPa
80
40°
60
20 Alloy 600
7350°C
O01rm H 3BO3+0.00lm LiOH Soln
II I I
-1000 -300 0 300
(Eco)
Potential, mnV
Expansion Transition PWSCC e
Figure 7-31 shows the number of tubes with hot leg expansion transition
PWSCC indications as a function of effective full power years for several
plants with low mill anneal temperature tubing. The plants include
Ringhals 2 which has part-depth roll expansions, EdF plants which have
kiss roll expansions, and North Anna I which has full-depth explosive
expansions.
Inner Row U-Bend PWSCC

Figure 7-32 shows the number of tubes with inner row U-bend PWSCC
indications as a function of effective full power years for a number of French
plants with low mill anneal temperature tubing. This data suggests that,
without stress relief, PWSCC can affect 1%of the row 1 Westinghouse tubes
formed with a metal ball mandrel in as little as three effective full power
years. The data also shows that the time to 1% PWSCC initiation in row 1
U-bends with Vallourec tubing, which is formed with a cylindrical plastic
mandrel, and Westinghouse row 2 U-bends is eight to nine years. It should
be kept in mind that this data is representative of PWSCC behavior without
stress relieving the U-bends. If the U-bends are stress relieved before
operation, it is likely that PWSCC will be largely prevented.
The above data demonstrate that PWSCC has the potential to be a
significant problem in plants with low temperature mill annealed tubing.
FIGURE 7-31
Typical Rate of Expansion Transition PWSCC in Several Well
Documented Plants With PWSCC-Susceptible Material
100.(
C
0
Cu
C-)
C
10.00%
0~ 1.00%
Cfl
0)
.0
0.10%
I-
C
0)
0)
0~ 0.01% 1
1.00 10.00 100.00
Effective Full Power Years
* Doel 2 (SGA) Ringhals 2

-- EdF Kiss Roll Plants North Anna 1
FIGURE 7-32
Typical Rate of U-Bend PWSCC in EdF Plants
02 100.00% _______ ____

.0
10.00%
00
.-=• 10.00%
C,)
- 0.10%
I-
0.01%
1u 100
Effective Full Power Years
It WRowi1 VRow2 -*" W Row 2
DETECTION OF PWSCC
Knowledge of the location and size of PWSCC defects is necessary to
demonstrate that the steam generator tubing meets licensing requirements
regarding tube integrity, and to avoid forced outages resulting from
excessive primary-to-secondary side leakage. Accordingly, it is necessary
to identify and characterize PWSCC in row I and row 2 U-bends, at severely
dented tube support plate intersections and in the expansion transition
regions. There are several methods available to perform inspections for
PWSCC.
Bobbin Coil Eddy Current Testing

Bobbin coil eddy current testing was originally the primary method of
inspecting the full length of tubes including U-bends, tube support plate
intersections and expansion transitions. The method was used primarily
because it is rapid. However, laboratory testing and evaluation of pulled
tubes shows that this method does not detect tight, low volume, PWSCC
type defects as well as other methods do. The bobbin coil method is
particularly weak in detecting circumferential cracking. For example,
bobbin coil inspection did not locate a circumferential crack at Ohi 1 which
extended about 2700 around the roll transition and was through wall over
about 30% of its length (7.61). U-bend inspections at Ringhals by bobbin
coil, rotating pancake coil (RPC), and helium leak testing showed that the
bobbin coil method only detected three of 43 leaking tubes (7.62). RPC
detected all but seven of the 43 leaking tubes. In spite of these weaknesses,
bobbin coil ECT is still used by many utilities to perform a 100% inspection
of all tubes. Tubes which show indications greater than a specified size are
then inspected using more accurate, but slower, methods such as RPC eddy
current testing.
Rotating Pancake Coil Eddy Current Testing

Rotating pancake coil (RPC) eddy current testing is commonly used to
perform more detailed examinations for cases where bobbin coil testing
indicates potential problems. The RPC technology is also being optimized
and is currently being used by some utilities to perform inspections of 100%
of the tubes in limited areas such as expansion transitions. Laborelec reports
that the hot leg expansion transitions of 100% of the tubes in a steam
0
generator can be inspected in about 2.5 days time by the RPC method (7.63).
An evaluation of inspection methods at Ringhals 3 and 4 has shown
that RPC is capable of detecting all major PWSCC in the roll expansion
region. The only exceptions noted were small axial cracks which were not
through wall (7.62) and small circumferential cracks which were bridged
between axial cracks (7.24). Figures 7-8 and 7-9 show the typical correlation
between RPC and detailed laboratory inspections of tubes at Ringhals. EdF
has also reported that the .RPC technique may not detect small
circumferential cracks in the expansion transition region. As noted in the
preceding section, RPC was able to detect all but seven of 43 leaking U-bends
in Ringhals 4.
Ultrasonic Testing
Work is underway to apply the ultrasonic inspection method to the
detection of PWSCC (.761 7.64 7.65). EdF has reported excellent
characterization of circumferential PWSCC in three pulled tubes (7.64).
However, the Ringhals test results (Figures 7-8 and 7-9) show that both the
ultrasonic and RPC inspections missed the same short circumferential crack
bridging between two axial cracks as the RPC method (7.24). In summary,
the ultrasonic inspection method shows promise but is not yet to the same
state of development as the RPC method.
Leak Testing
Through wall PWSCC in U-bends, tube support plate intersections, and
expansion transitions can sometimes be located by leak testing. Helium
leak testing is the most common method and involves drying the steam
generator secondary side and pressurizing with a low pressure, 6 bar
(90 psi), mixture of about 90% air and 10% helium. A sniffer is programmed
to pass below each tube and check for helium (7,63 7.66).
Another method of leak testing is to use fluorescein dye when
performing a hydrostatic test on the secondary side of the steam generator.
This type of leak testing involves filling the secondary side with water
containing 50 ppm fluorescein and pressurizing the secondary side to 25-40
bar (360-600 psi). After 8-20 hours, the underside of the tubesheet is
inspected using an ultraviolet lamp to identify leaking tubes (7.63).
Testing at Doel 3 showed that fluorescein dye was capable of detecting
long (0.5-0.9 inches [12-22 mmi) cracks. However, smaller cracks down to
0.08 inches (2 mm) length, which were detectable by RPC, were not detected
by fluorescein dye (7.63). The tests at Doel 3 also showed problems with
the helium leak test which were attributed to humidity or incomplete drying
of the secondary side leading to moisture plugging the tight PWSCC cracks.
The overall conclusion was that many leaking tubes were detected by one
or the other of the two methods, but only a few were detected by both
methods.
EdF found that RPC inspection was capable of locating 98% of the
defects identified by helium leak testing (7.66). On the other hand, EdF has
found that PWSCC detected by eddy current testing may not be detected
by helium leak testing. Since helium leak testing is capable of detecting all
significant size through-wall cracks, EdF uses the technique to quickly
identify tubes which should be inspected in greater detail by RPC. As shown
in Figure 7-33, EdF considers this procedure to have reduced the number
of forced outages without the need for 100% RPC inspection.
FIGURE 7-33
Reduction in Forced Outages in EdF Plants After Start of
Helium Leak Testing (7.66)
Number of Unscheduled Outages
Use of Helium Test
0 2 2 2 Year
82 83 84 85 86 87 88 89
Dented Region Inspection

Dents large enough to cause PWSCC create such a large "noise" signal
using conventional bobbin coil eddy current techniques that cracks are often
difficult to detect. Nevertheless, methods of evaluating bobbin coil signals
for dented tube support plate intersections have been developed, reviewed
with the NRC and are being used at North Anna.
REFERENCES
7.1 Coriou, H., L. Grall, Y. Le Gall, and S. Vettier. "Corrosion Fissurante
Sous Contrainte De L'Inconel Dans UEau A Haute Temperature."
Third Collogue de Metallurgie Corrosion. Centre d'Etudes Nucleaires
de Saclay, France. Amsterdam: North Holland Publishing Co.,
1959, p. 161.
7.2 Coriou, H., et. al. "Historical Review of the Principal Research
Concerning the Phenomena of Cracking of Nickel-Base Austenitic
Alloys." In Proceedingsof Conference on FundamentalAspects of Stress
CorrosionCracking and Hydrogen Embrittlement of Iron Based Alloys.
Unieux-Firminy France. NACE-5. June 12-16, 1973.
7.3 Van Rooyen, D. "Review of the Stress Corrosion Cracking of Inconel
600." Corrosion 31 (1975): 327.
7.4 Serra, E. "Stress Corrosion Cracking of Alloy 600." EPRI NP-2114-
SR. November 1981.
7.5 Garud, Y. S., and T. L. Gerber. "Intergranular Stress Corrosion
Cracking of Ni-Cr-Fe Alloy 600 Tubes in PWR Primary Water-
Review and Assessment for Model Development." EPRI NP-3057.
May 1983.
7.6 Szklarska-Smialowska, S. "Factors Influencing IGSCC of Alloy 600
in Primary and Secondary Waters of PWR Steam Generators."
Paper presented at Fourth International Symposium on
Environmental Degradation of Materials in Nuclear Power Systems,
Water Reactors, Jekyll Island, Georgia, August, 1989.
7.7 Hunt, E. S., and J. A. Gorman. "Status and Suggested Course of
Action for Nondenting-Related Primary Side Cracking of
Westinghouse-Type Steam Generators." EPRI NP-4594-LD. May
1986.
7.8 "U-Bend Tube Cracking in Steam Generators." EPRI WS-80- 136.
June 1981.
7.9 "Proceedings: 1983 Workshop on Primary-Side Stress Corrosion
Cracking of PWR Steam Generator Tubing." EPRI NP-5498.
November 1987.
7.10 "Proceedings: 1985 Workshop on Primary-Side Stress Corrosion
Cracking of PWR Steam Generator Tubing." EPRI NP-5158. June
1987.
7.11 1985 SGOG/EPRI Workshop on Management of Steam Generators
Susceptible to Primary Side Cracking: Remedial Actions and NDE
Methods. St. Petersburg Beach, FL. December 10-12, 1985.
7.12 "Proceedings: 1987 EPRI Workshop on Mechanisms of Primary

Water Intergranular Stress Corrosion Cracking." EPRI NP-5987SP.
September 1988.
7.13 "Proceedings: 1989 EPRI PWSCC Remedial Measures Workshop."
EPRI NP-6719. April 1990.
7.14 "Proceedings: 1990 EPRI Circumferential Cracking Workshop."
NP-7198-S. March 1991.
7.15 Schenk, H. J. "Investigation of Tube Failures in Inconel 600 Steam
Generator Tubing at KWU Obrigheim." Materials Performance
(March 1976): 25.
7.16 Schenk, H., E. Pickel, and A. Hummeler. "The Replacement of the
Steam Generators in the Nuclear Power Plant at Obrigheim." ASME
Paper 84-NE-18.
7.17 Schuecktanz, G. "Obrigheim: Experience With Primary Side
Cracking." EPRI NP-7198-S. March 1991.
7.18 "Stress Analysis of Pressurized Water Reactor Steam Generator Tube
Denting Phenomenon." EPRI NP-828. July 1978.
7.19 Yashima, S., K. Uragami, H. Utsumi, H. Ikenaga, K. Nakamura,
and 0. Takaba. "Stresses of Steam Generator U-Tubes Affecting
Stress Corrosion Cracking." ASME Paper 82-NE-5.
7.20 Aspden, R. G., and P. J. Kuchirka. "Evaluation of Steam Generator
U-Bend Tubes From the Trojan Nuclear Power Plant." EPRI NP-
2629-LD, September 1982. Johnson, A. B., Jr., R. L. Bickford, R. L.
Clark, T. J. Davis, and F. A. Simonen. NP-3571-LD, July 1984.
7.21 Dobbeni, D. P., J. R. Stubbe-Leroy, and C. E. van Melsen. "U-Bend
Cracking at Doel 2." In Workshop Proceedings: U-Bend Tube Cracking
in Steam Generators. EPRI WS-80-136. June 1981.
7.22 Hernalsteen, P, and R. Houben. "Preventive and Corrective Actions
for Doel 2 Steam Generators." Paper presented at October 1984
Specialist Meeting on Steam Generators in Stockholm, Sweden.
7.23 Gorman, J. A., and E. S. Hunt. "Status of Cracking and Remedial
Measures for PWR Steam Generators With Full-Depth Expanded
Tubing." EPRI NP-4459-LD. February 1986.
7.24 Wilson, B., and J. Engstr6m. "Experience of Circumferential
Cracking in Ringhals Units 3 and 4." In Proceedings: 1990 EPRI
CircumferentialCracking Workshop. EPRI NP-7198-S. March 1991.
7.25 Comby, R., B. Flesch, and F. de Keroulas. "Circumferential Cracking
of French Reactor Steam Generator Tubes in the Roll Transition
Zones Under the Effect of Stress Corrosion." In Proceedings:1990
EPRI Circumferential Cracking Workshop. EPRI NP-7198-S. March
1991.
7.26 Brown, S. D. "Wextex Expansion Cracking Experience." In

Proceedings: 1990 EPRI Circumferential Cracking Workshop. EPRI
NP-7198-S. March 1991.
7.27 Malinowski, D. D. "Operating Plant Status of Westinghouse Steam
Generators." Eighth Annual Steam Generator NDE Workshop.
August 9-11, 1989, Williamsburg, VA.
7.28 Gorman, J. A., R. W. Staehle, and K. D. Stavropoulos. "Statistical
Analysis of Steam Generator Tube Degradation." EPRI NP-7493.
September 1991.
7.29 Cattant, F "Examen de Deux Tubes du Generateur de Vapeur B."
EdF report D.5004/CTT/R.84.37. April 20, 1984.
7.30 Engstr6m, J., and K. Norring. "Primary and Secondary Cracking
at Ringhals 2." In Proceedings:1983 Workshop on Primary-Side Stress
Corrosion Cracking of PWR Steam Generator Tubing. EPRI NP-5498.
November 1987.
7.31 Norring, K. "Examination of Fifteen Tubes From Steam Generator
1 at Ringhals 2." Studsvik report EI-84/56. March 26, 1984.
7.32 Norring, K. "Influence of Roll Transition Shape on Primary Side
IGSCC of Steam Generator Tubes. " Paper presented at October
1984 Specialist Meeting on Steam Generators, Stockholm, Sweden.
7.33 K. Norring. "Examination of Twelve Steam Generator Tubes Pulled
From Ringhals 3 and 4 in 1988." Studsvik report EX-89/20.
7.34 Pement, F. W., G. Economy, and R. J. Jacko. "Tubesheet Expansion
Improvements." EPRI NP-5547. December 1987.
7.35 Frye, C. R., P. A. Sherbourne, and M. E. Browne. "Cracking in the
Expanded Region of Two Tubes From the V. C. Summer Nuclear
Station." In Proceedings: 1990 EPRI Circumferential Cracking
Workshop. EPRI NP-7198-S. March 1991.
7.36 Lee, C. Y., M. K. Lim, J. S. Lim, and T. E. Chung. "Steam Generator
Maintenance Experience of Korean PWR Nuclear Units." Third
International Topical Meeting on Nuclear Power Plant Thermal
Hydraulics and Operations, Seoul, Korea, November, 1988.
7.37 Albertin, L., and P. Kuchirka. "Examination of Tubes R3C41HL
and R9C58HL of Steam Generator C, North Anna Unit 1." EPRI
NP-5420-LD. October 1987.
7.38 Houben, R., and P. Hernalsteen. "Status of the Doel 2 Unit." In
Proceedings:1989 EPRI PWSCC Remedial Measures Workshop. EPRI
NP-6719. April 1990.
7.39 Berge, P., H. D. Bui, J. R. Donati, and D. Villard. "Residual Stresses
in Bent Tubes for Nuclear Generators." Corrosion32, no. 9 (1976):
357.
7.40 Domain, H., R. H. Emanuelson, L. Katz, L. W. Sarver, and G. J. Theus.

"Effect of Microstructure on Stress Corrosion Cracking of Alloy 600
in HighPurity Water." Corrosion33 (1977).
7.41 Airey, G. P. "Optimization of Metallurgical Variables to Improve
Corrosion Resistance of Inconel Alloy 600." EPRI NP-1354, March
1980. EPRI NP-3051, March 1980.
7.42 Airey, G. P. "The Stress Corrosion Cracking (SCC) Performance of
Inconel Alloy 600 in Pure and Primary Water Environments." In
Proceedings of the International Symposium on Environmental
Degradationof Materials in Nuclear Power Systems -Water Reactors.
Myrtle Beach, South Carolina, August 22-25, 1983. NACE, 1984.
7.43 Garriga Majo, D. "Prediction of the In-Service Behavior of Alloy
600 Tubes Used in Steam Generators of Pressurized Water Reactors."
Colloque International Fontevraud II, September 10-14, 1990.
7.44 Stein, A. "Stress Corrosion Cracking of Alloy 600 in Primary Water."
In Proceedings: 1983 Workshop on Primary-Side Stress Corrosion
Crackingof PWR Steam GeneratorTubing. EPRI NP-5498. November
1987.
7.45 Bruemmer, S. M., L. A. Charlot, and C. H. Henager. "Microstructure
and Microdeformation Effects on IGSCC of Alloy 600 Steam
Generator Tubing." In Proceedings: 1987 EPRI Workshop on
Mechanisms of Primary Water IntergranularStress CorrosionCracking.
EPRI NP-5987SP. September 1988.
7.46 Smialowska, S. "Hydrogen-Induced IGSCC of Alloy 600 in High
Temperature Aqueous Environments." In Proceedings:1987 EPRI
Workshop on Mechanisms of Primary Water IntergranularStress
CorrosionCracking. EPRI NP-5987SP. September 1988.
7.47 Norring, K., J. Engstr6m, and P. Norberg. "Intergranular Stress
Corrosion Cracking in Steam Generator Tubing, Testing of Alloy
690 and Alloy 600 Tubes." In Proceedings of the Third International
Symposium on EnvironmentalDegradationof Materials in Nuclear Power
Systems. The Metallurgical Society, 1987.
7.48 "Guidelines for PWR Steam Generator Tubing Specification and
Repair." EPRI NP-6743-L.
7.49 Rudd, C., and M. Jacobs. "Roll Transition Stresses-Measured
Residual Stress Patterns." Paper presented at EPRI meeting at
Pennsylvania State University, May 14-15, 1990.
7.50 Data provided to EPRI by Framatome.
7.51 Data presented by J. Begley of Westinghouse NSD at EPRI meeting
at Pennsylvania State University, May 14-15, 1990.
7.52 Hunt, E. S. "Elastic-Plastic Finite Element Analysis of Hydraulic
Expansion." Paper presented at May 14-15, 1990 EPRI meeting at
Pennsylvania State University.
7.53 Vanger, J. "Elastic Stress Analysis of Small-Radius U-Bend Steam

Generator Tubes." EPRI NP-294. "Inservice Stresses in 900 MWe
PWR Steam Generator U-Bends." 10th SMIRT, Division D, 1989.
7.55 Rudd, C. 0. "Residual and Applied Stress Analysis of Inconel Alloy
600 7/8 Inch U-Bend Tubing." In Proceedings:1987 EPRI Workshop
on Mechanisms of Primary Water IntergranularStress Corrosion
Cracking. EPRI NP-5987-SP. September 1988.
7.56 Yonezawa, T., N. Sasaguri, K. Inimura. "Metallurgical Factors on
Stress Corrosion Cracking of Ni-Base Alloys in High Temperature
Water." JAIF International Conference on Water Chemistry in
Nuclear Power Plants, Tokyo, Japan, 1988.
7.57 Garud, Y. S. "Development of a Model for Predicting Intergranular
Stress Corrosion Cracking of Alloy 600 Tubes in PWR Primary
Water." EPRI NP-3791. January 1985.
7.58 Bandy, R., and D. van Rooyen. "Quantitative Examination of Stress
Corrosion Cracking of Alloy 600 in High Temperature Water-Work
During 1983." In Proceedings:1983 Workshop on Primary-SideStress
CorrosionCracking of PWR Steam GeneratorTubing. EPRI NP-5498.
November 1987.
7.59 Norring, K., et al. "Influence of Li OH and H 2 on Primary Side
IGSCC of Alloy 600 Steam Generator Tubes." Colloque
International Fontevraud II. September 10-14, 1990.
7.60 Jacko, R. J., and C. J. Wood. "The Influence of Primary Water Li
Content on PWSCC of Alloy 600." Colloque International
Fontevraud II, September 10-14, 1990.
7.61 Fukui, S. "Inspection Techniques for Circumferential Cracking of
Steam Generator Tubes." In Proceedings:1990 EPRI Circumferential
Cracking Workshop. EPRI NP-7198-S. March 1991.
7.62 Engstr6m, J. "Experience of PWSCC in Ringhals Units 3 and 4." In
Proceedings:1989 EPRI PWSCC Remedial Measures Workshop. EPRI
NP-6719. April 1990.
7.63 Houben, R., P. Hernalsteen, and C. Leblois. "Comparison of
Hydrostatic and Helium Leak Test Results With RPC Crack Length
Measurements." In Proceedings: 1989 EPRI PWSCC Remedial
Measures Workshop. EPRI NP-6719. April 1990.
7.64 Comby, R. "Circumferential Cracking in Steam Generator Tubes
-In-Service Inspection Techniques." In Proceedings: 1990 EPRI
Circumferential Cracking Workshop. EPRI NP-7198- S. March 1991.
7.65 Dobbeni, D., and D. Degreve. "Field Implementation of Ultrasonic
Inspection Technique." In Proceedings: 1990 EPRI Circumferential
Cracking Workshop. EPRI NP-7198-S. March 1991.
7.66 Gouhier E., and J. L. Germain. "Application of Helium Leak Testing

for Crack Detection." In Proceedings: 1989 EPRI PWSCC Remedial
Measures Workshop. EPRI NP-6719. April 1990.
7.67 Norring, K. "Intergranular Stress Corrosion Cracking in Steam
Generator Tubing." Studsvik Energy. EX-90/9. January 1990.
7.68 "Corrosion Evaluation of Thermally Treated Alloy 600 Tubing in
Primary and Faulted Secondary Water Environments." EPRI
NP-6721-SD. June 1990.
7.69 Rouillon, Y., et al. "PWSCC Behavior of Thermally Treated Alloy
600." In Proceedings: 1987 EPRI Workshop on Mechanisms of Primary
Water IntergranularStress Corrosion Cracking. EPRI NP-5987-SP.
September 1988.
7.70 "PWR Steam Generator Tube Plugging Limits-Technical Support
Document for Expansion Zone Primary Water Stress Corrosion
Cracking in Roll Transitions," Rev. 1. EPRI NP-6864-L.
7.71 Cattant, F., et al. "C.P.N. De Dampierre SUC 1/2-Tranche
1-Boucle." EdF Report D.5004/CTT/R.87.16. April 17, 1987.
7.72 Boursier, J. M., 0. deBoouvier, J. M. Gras, D. Noel, R. Rios, and E
Vaillant. "SCC of Alloy 600 in High Temperature Water: A Study
of Mechanisms." EdF Report HT-40/COM 1528-A. Paper
presented at Corrosion Deformation Interaction Conference CDI
1992, Fontevreaud, 5-7 October, 1992.
8
TUBE SUPPORT AND

TUBESHEET CORROSION
(ACIDIC CREVICES-TUBE
DENTING)
W. L. Pearl,NWT
S. G. Sawochka, NWT
INTRODUCTION: DESCRIPTION AND CAUSES OF

TUBE DENTING
PWR steam generators at many sites have experienced accelerated
corrosion of carbon steel tube support structures and local deformation of
the alloy 600 tubing. The term denting is employed to describe the
constricting or mechanical deformation of the steam generator tubes as they
pass through the corroded supports. Such corrosion results from the
formation of an acidic environment resulting from impurity concentration
processes associated with boiling within the tube/tube support crevices.
This chapter includes a discussion of the importance and impact of this
operational problem together with a review of plant experiences and the
structural aspects of denting. The literature pertaining to the corrosion of
carbon and chromium steels in high temperature water is briefly reviewed,
and the influence of pH, steel composition, dissolved anions and cations,
and other environmental variables on the corrosion kinetics and oxide
morphology are examined. Mechanisms are proposed which explain field
and laboratory observations.
PWR Steam Generator History
Steam Generator Design. Tube-and-shell PWR steam generator designs

are described in Chapter 2 together with a brief overview of operating
experience. Westinghouse (W), Combustion Engineering (C-E), Framatome
(FRA), Mitsubishi (MHI), Babcock & Wilcox Canada (B&W), and Kraftwerk
Union (KWU) have each produced several design variations of the inverted
U-tube steam generator. There have been and continue to be significant
differences in the problems encountered in different plants as a consequence
of both differences in operation and steam generator and secondary system
design. A feature of particular relevance to denting corrosion has been the
use of different support plate designs. Carbon steel has been used in drilled
hole, lattice bar, and trefoil designs which exhibit different denting rates
depending on design and operating conditions. Improved quatrefoil and
lattice bar designs utilizing chromium and austenitic steels are more open
and exhibit much less of a tendency for denting-related corrosion.
Steam Generator Materials. There are several variations in steam generator

construction materials. Alloy 600 (15% Cr) and alloy 800 (20% Cr) have
been used for most tubing, although 300 series austenitic stainless steel
was used in early steam generators. Alloy 690 (30% Cr) has been specified
for many replacement steam generators. Alloy 400 was used in early
CANDU steam generators (similar in design to PWR steam generators).
Tube Support and Tubesheet Corrosion (Acidic Crevices-Tube Denting) 8-3
Tube support plates (TSPs) and lattice bar structures were initially made
of carbon steel, but as a result of the denting process, carbon steel has been
replaced by 12% Cr (A405, A409) or 13% Cr (A410) ferritic stainless steels
in new designs. Austenitic stainless steel (Type 347) is used for lattice bar
structures in KWU steam generators, but this introduces problems due to
differential expansion relative to the ferritic shell which have to be addressed
by design. A508 Class 3 (<0.25% Cr) ferritic steel or an equivalent steel is
typically used for the tubesheet.
Denting Experience. Denting was the most important single corrosion

process encountered with PWR steam generators in terms of loss of
availability and the numbers of tubes affected through 1980 (8.1). Industry
experience is provided in Table 2-2. Denting has occurred both with and
without prior phosphate exposure.
Westinghouse and C-E designs experienced denting as a result of their
carbon steel support structure design and a corrosive environment; the
feedwater was contaminated by condenser leakage. FRA and MHI designs
have been less affected, probably because of improved condenser history.
In addition, most FRA plants are at fresh water sites. KWU designs, which
use stainless steel in a grid-type support structure, have not been affected.
B&W Canada designs, which operate at a lower temperature at fresh water
sites, have reported no denting.
Failure Analysis of Support Plate Sections

Several detailed examinations have been conducted on samples of
dented tube/tube support plate intersections removed from PWR steam
generators (8.2-8.4). There is general agreement between the studies that
denting is due to the growth of magnetite corrosion products in the crevice
between the support plate and the heat transfer tube. The magnetite volume
is approximately equal to twice the volume of metal consumed. The oxide
morphology shown in Figure 8-1 (8.3) was similar to the scale morphology
observed by Potter and Mann (8.5) on mild steel corroded in acid chloride
solutions. The oxide exhibited fine-grained magnetite growing at the metal/
oxide interface as well as elongation of the ghost images of the carbides
and metal grains in a direction perpendicular to the metal/oxide interface.
Low levels of chloride were detected in the oxides in each study.
Overlaying the magnetite corrosion product was a deposit of iron
phosphate containing magnetite, metallic copper, and silica particles with
traces of sodium, calcium, and manganese. The presence of these trace
elements is attributed to impurities in the feedwater and pickup from the
feed train, except for manganese which is, due to its high solubility, believed
to originate from the tube support plate (8.6). No evidence was found for
participation of phosphate in the denting process, other than as a filler for
the initial tube/TSP gap. However, it was suggested that copper may have
a significant influence on the corrosion process (8.3).
FIGURE 8-1
Details of Scale and Deposit Showing Banded Structure and
Variation in Porosity (8.3)
0.
eL
AREA FOR
X-RAY ANALYSIS w
I-
UJ
4Z
METAL
Small amounts (<2%) of hematite were found in the scale (8.3) in the
form of discrete bands. The potential-pH diagram for the Fe-H 20 system
(Figure 8-2) shows that Fe 2 0 3 will be the stable corrosion product at
intermediate and high pH and high oxygen partial pressures. Thus, it is
possible that short-term changes in pH, oxygen concentration, or transport
of other oxidants (e.g., CuO) may have occurred in the steam generator as
a result of changes in plant operating conditions.
FIGURE 8-2
Potential/pH Diagram for Fe-H2 0 at 5500 (299 0 C) (8.7)
2.50
2.00
1.50
FeO 4=
NIN
1.00 - I,
I IN
i 0.50 IN
W_ 0.00 " , -
n0a.-_"-_ (a) N. , .(a
_ , '" " (b)
0 4 8 12 16
pH
Effects on Tubes, Tube Supports, and Tubesheet

The denting process, in which the growth of magnetite in the tube/
tube support progresses almost linearly with time, produces high levels of
stress on both the tube and the support structures. The ratio of "oxide
volume produced" to "volume of metal corroded" is referred to as the
Pilling-Bedworth (P-B) ratio. The theoretical value of this ratio for Fe 304
grown from carbon steel is 2.2. An analysis of cross-sectional areas of oxide
formed by corrosion of carbon steel cylinders contained in alloy 600 capsules
resulted in a P-B ratio of 2.5 ± 0.3 (8.8). The significance of the ratio is
illustrated in Figure 8-3, which shows the extent of radial tube deformation
in several capsule tests as a function of the metal corroded. In this case,
approximately 0.007 inches of metal corrosion was required to fill the radial
gap. Additional corrosion caused capsule bulging, which progressed at a
linear rate.
FIGURE 8-3
Correlation of Metal Loss Versus Capsule Bulging (4.8)
251
(Corrosion) 1.36 (Bulge) + 6.8
0
20
0
0
o
0
rr
10
50
0 5 10 15
Radial Bulge (mils)
In a steam generator the tube is constricted by the growing oxide as the

support structure is corroded. The result is a reduction in tube cross section
with possible initiation of stress corrosion cracking due to the high local
stresses; stretching and rupture of tube support plate ligaments; or through-
ligament corrosion or rupture of lattice type structural members. Figure 8-4
(8.9) is a photograph of a section of a support plate from a decommissioned
steam generator, showing the extent and distribution of denting with
ligament fracture. Tube collapse and flow slot closure are evident. Many
ligaments are cracked on both the hot and cold leg sides. Tube collapse is
most apparent near the support plate "hard spot" (without flow slots).
In one detailed inspection effort (8.4), secondary side corrosion in the
form of shallow pitting, wastage, and intergranular stress corrosion was
found on the tubing in the tube support plate crevice region. The severity
of attack appeared greater in more heavily dented tubes. The tube wall
loss was generally less than 20%. Axial OD cracks were noted at regions of
high tensile hoop stress when the calculated strains were greater than 60%.
FIGURE 8-4
Top of 7th Support Plate After U-Bend Removal (t.9)
Column 61
Cold Leg
Side
Hot Leg
Side
Center Post
Column 44
One of the early reported side effects of denting was the deformation
of inner-row-tubing U-bends which resulted in ID and OD stress corrosion
cracking in Row 1 and OD cracking in Row 2 tubes. In this process, the
growth of the uppermost tube support plate led to flow slot closure and
offsetting of the two U-bend legs in an inward direction, producing an over-
bend condition in the tubing. This led to cracking near the apex of the bend
as the tube cross section became ovalized (8.10). Figure 8-5 (8.9) illustrates
a U-bend apex stress corrosion crack.
IGSCC on the primary side (ID) also has been found in some heavily
dented intersections, e.g., Figure 8-6. No ID cracking was observed below
10% strain, with all the inspected specimens above 20% strain containing
such cracks. Based on such operating experience as well as laboratory tests,
calculated tube strains of 10-20% have been employed as tube plugging
criteria. If preventive plugging of dented tubes had not been widespread,
IGSCC might have been observed more frequently. Improved technology
of measuring tube shape, together with increased favorable operating
experience, has refined the relationship of tube strain and the need for tube
plugging, and has led to responsive tube plugging criteria.
FIGURE 8-5
Effect of Flow Slot Closure on Appearance and Behavior of Row 1
U-Bend's 5th Flow Slot (8.9)
FIGURE 8-6
Hot Leg Specimen at 1st Support Plate Showing ID Cracking
(26.9% Strain) (8.4)
5 mm
Corrosion at the tube/tubesheet interface generally has been much less

extensive than that found higher in the tube bundle (8.10-8.11). Very few
tubes have required plugging because of tubesheet denting (8.11). The major
effect of such corrosion has been stress corrosion cracking. Denting at this
location also can interfere with sleeving and tube removal activities.
CAUSES OF DENTING
Corrosion Processes
The impact of denting in PWR steam generators led to extensive research
into this mode of corrosion, and a search for remedial measures. The
processes leading to denting are broadly understood, and are shown
schematically in Figure 8-7 (8.12). The concentration and corrosion processes
governing the many forms of steam generator degradation are discussed

in Chapter 5. An overview of the corrosion mechanisms involved in denting
is given below.
FIGURE 8.7
Schematic Diagram Showing Mechanism of Corrosion at Tube/Tube
Support Plate Crevice (8.12)
Hydrolysable Chloride e.g. MgCi 2

Transported to Crevice and
Concentrated by Boiling Heat
Transfer, Giving Acid Conditions
Or
Heat Flux 02 or Cu
zi•lC-athodic Reactions
(O2 + 2H 2 0 + 4e-i. 40H)
Corrosion (Cu 2 + 2e- Cu Metal)
Product / /
,Cathodic Reaction (2H*+ 2e- H)
Anodic Reaction (Fea. Fe + 2e-)

(3Fel'+ 4H12 0- Fe 3 O 4 + 41 2 )
Mild Steel Tube Support Plate
~ and Copper
C2 or Maneit.Heatt
Heat Flux• C I- or Oxygenated Salt Water
with Copper Ions
Alloy 600 Tube
Nature of the Oxide/Corrosion Product Layer. Several types of oxide/

corrosion product layers have been observed on carbon steel and low
chromium ferritic steels in high temperature solutions of various pHs. The
growth mechanisms of these oxide layers differ considerably. Their
morphologies are summarized in Figure 8-8. In neutral and moderately
acid or alkaline solutions (room temperature pH = 4 to about 12), a duplex
scale can be formed (Figure 8-8a) which has protective characteristics, i.e.,
the, corrosion rate decreases with time. The duplex scale consists of a fine
grain inner layer, which grows'at the metal/oxide interface, and occupies
the same volume as the metal consumed, and an outer layer of magnetite
crystals.
In neutral or alkaline solutions, another form of protective oxide
("Bloom"-type oxide) also can form. This consists of a thin single layer of
protective magnetite crystals (Figure 8-8d).
In more concentrated acidic and alkaline solutions, the inner layer can
break down to give fast corrosion, often at a linear rate (Figure 8-8b). This
breakdown is particularly influenced by the chromium content of the steel.
In acid conditions, a dense adherent scale forms which produces a net
volume increase and denting. In some cases a laminated oxide forms
(Figure 8-8c).
The fast linear growth oxide (formed at low crevice pH) consists of
either a single layer (Figure 8-8b) or a multi-layered laminated structure
(Figure 8-8c). Features that are present in the metal become elongated in
the direction perpendicular to the surface as they are consumed to form
oxide, and there tends to be little lateral expansion. Other studies of acid
chloride corrosion of carbon steel have confirmed Potter and Mann's original
observations of a fast linear growth rate and a laminated scale, although
the corrosion rates vary widely.
The factors affecting oxide morphology and corrosion kinetics at 572°F
(300°C) in acid solutions are described below. The intermediate pH range
has been extensively reviewed (8.13) and will not be considered further
since corrosion rates are low and do not give rise to problems in boilers.
,Similarly, although alkaline corrosion rates are relatively high and alkaline
denting has been experienced (Chapter .11), the impact of such attack on
plant availability has been minimal compared to the impact of acid denting.
A proposed denting mechanism is shown schematically in Figure 8-9.
Iron corrodes at the metal surface to form iron-bearing species through the
typical reactions
Fe + H 2 0 -> Fe(OH)+ +H + 2e" (8-1a)
Fe + 2H 2 0 -> Fe(OH) 2 + 2H+ + 2e- (8-1b)
FIGURE 8-8
Schematic Representation of Corrosion Product Morphologies of
Iron and Chromium Alloys Exposed to High Temperature Water (8.12)
S S * * , . , .. , e Inner Layer of
Iron/Chrome Spinel
ft I - I
NXNXNNNXNX•\N \\NX\ \N\\\"
Metal
(a) Duplex Oxide (Potter-Mann)
Water Water
r * 5' . S
S * *' *
, __ Original
• •. .. - '.... . •.......... •.
Interface
* - . . S -
\\\ \ -X-,\N\\
Metal Metal
(b) Low Protectivity (c) Laminated Scale

Porous Scale
(d) Bloom Oxide

FIGURE 8-9
Schematic Representation of the Proposed Mechanism of Fast
Linear Magnetite Growth on Steel in Acid Chloride Solution (8.14)
Alloy 600 Tube
S b 40, .I zZl Z Cathodic Sites

a e , 2H++ 2e-- H2
" " " 4 ' d Split Filled With Solution

or0 " A, Direction of Electron Flow
ft 0 b, 0. Large Crystal Ripening
, . • at the Expense of
' ~- 1b . Porous Magnetite
b) .** *4b _ .
-. Split Initiation
St *, , 3Fe(OH)++H 2 0
_-•- •* e_ , • * =H 2 +3H++ Fe0 4 (Anodic Reaction)
4b .0 0,0. 0.'e• ,pao. .
ad. Iron Solubility Limit
-. . N >.Gradient
SMetal "Expansive "Denting" Force

Fe + H 2 0= Fe(OH)++ H++ 2e- (Anodic Reaction)
Legend:
e Fe 3 O4 Deposited From Solution
Porous Fe 3 0 4 Grown
*at Metal Surface
* Large Grain Solution

Deposited Magnetite
As a result, an iron hydroxide compound forms at the metal surface,

and a high concentration of iron-bearing ions is present in the pores. A
short distance from the metal surface, oxidation proceeds and magnetite is
formed and precipitates:
3Fe(OH) + + H 20 -> H 2 + 3H+ + Fe3 0 4 (8-2)
The cathodic reaction (2H+ + 2e- ->H 2) is thought to occur mainly on
outer surfaces but can also occur at splits in the oxide and on other surfaces
which are better cathodic dischargers than magnetite. The net result is that
a pH gradient is established across the oxide as a result of the separation of
cathodic and anodic sites with the lowest pH next to the corroding metal.
Chloride will become enriched at the metal/oxide interface as a result of
the local build-up of cations and the need to maintain charge neutrality.
The hydrogen evolution, noted in the laboratory and at plants where
denting is occurring, is consistent with Equations 8-1 and 8-2.
The pH, iron, and chloride gradients eventually reach dynamic
equilibrium. Magnetite always deposits within the existing oxide away
from the metal/oxide interface where its solubility limit is exceeded.
Solution transport through the thin porous layer near the metal surface is
thought to be corrosion-rate-controlling, yielding a constant corrosion rate.
Magnetite growth generates compressive stress within the oxide as the
crystals push against one another (8.15. This causes strain in the direction
of least restraint, (perpendicular to the metal surface), leads to the apparent
stretching of discontinuities in the oxide (e.g., retained metal grain
boundaries), and disrupts the oxide, providing easy oxidant ingress.
Crevice Chemistry. As discussed above, acid denting is the result of

corrosion of carbon steel supports which remain in an active corrosion state
as a result of crevice environmental conditions. Specifically, a high chloride/
low pH condition develops within the crevice and prevents formation of a
passive film. In addition, the concentrated crevice solution provides an
electrically conducting solution for ionic transport.
Of particular pertinence is the underlying mechanism which
leads to
the formation of highly acidic solutions in crevices at seawater and brackish
water sites. Several early models of crevice chemistry indicated that
concentration of steam generator bulk water impurities, which resulted from
condenser inleakage at seawater sites, led to the formation of acidic solutions
as a result of precipitation of species (8.16-8.17) such as magnesium
hydroxide, calcium, and/or magnesium silicates. (In effectively all cases,
silica is present in the bulk water primarily as a result of makeup water
impurity transport, rather than as a result of condenser inleakage.)
More recently, a comprehensive high temperature solution chemistry
code (MULTEQ) has been developed by EPRI (8.18-8.19) for estimating
crevice chemistry during liquid evaporation in partially occluded regions

of the steam generator, such as tube-to-tube support crevices and sludge
piles. Results developed from this code confirm the general trends predicted
by the earlier investigators.
An example of the variation in predicted crevice pH at 280'C as seawater
contaminants in the steam generator bulk water are concentrated, is shown
in Figure 8-10 (developed using MULTEQ MULTEQ-REDOX Version PC 2.0
with database 2.74). Results are given for a bulk water chloride
concentration of 0.1 ppm with steam removed during the concentration
process as well as with steam and precipitates removed. The latter situation
seems more reasonable since, as bulk water moves into the crevice and
FIGURE 8-10
Effect of Precipitate Removal on Predicted Crevice pH With 0.1 ppm
Chloride in the Bulk Water From Seawater Contamination (MULTEQ
Version 2.0/386; Database 2.74)
~H
NpH
• pH l
6- 6 . . . . .
* NPH l].• . . . . . . . .eNpH"K• , . . . ... ... ..... . ...
] ................. .............. .
----
-- --......
IE4- . . . ... .
3 - . . . . . . . .•. . . . . . . . . .............. ... ..--- .. ..•
... .. ... .. ! ....
. ...................! ~ ..... .......
......W ....
2 rrr j r--r-rj- r rjrr----j--rr r r r -- r-

-6 -5 -4 -3 --2 -1 0 1 2
Log1 0 Ionic Strength
I-Precipitates Retained -Precipitates Removed

O
impurities begin to concentrate, many precipitates are formed at relatively

low concentration factors above the bulk coolant concentration. These
probably deposit near the mouth of the crevice and thus are removed from
contact (equilibrium) with concentrated solutions found deeper in the
crevice. As shown, the crevice pH is predicted to be very acidic with
precipitate removal.
A major consideration in estimating crevice chemistry conditions in
operating steam generators, is the extent to which the crevice concentration
process proceeds. As illustrated in Figure 8-11, the concentration process is
a constant pressure process and can proceed only until impurity
concentrations in the crevice solution are sufficient to depress the solution
vapor pressure to the secondary side pressure at the local primary
temperature. The concentration process can be considered to begin at
Point A (which corresponds to the low impurity concentrations present in
the bulk water) and to proceed horizontally to the limiting condition at
Point B, in this case the hot leg tubesheet (primary coolant) temperature
TH. Depending on the impurity, it also is possible for dryout to occur with
precipitation from solution of all impurity species and evaporation of the
liquid prior to reaching the limiting condition at Point B. At intermediate
locations in the steam generator, the limiting condition will be between
Point A and B. In the actual situation, solutions may not concentrate to the
boiling point elevation limit. They may only proceed to the point where
the impurity input rate from the bulk water corresponds to the impurity
loss from the crevice as a result of steam solubility, liquid carryover in the
exiting steam, and diffusion from the crevice liquid to the bulk water. This
is, in part, a function of the porosity of the oxides filling the crevice.
Unfortunately, various terminologies have been employed by
investigators to describe the extent of the concentration process within the
crevice, and the limiting condition imposed by the boiling point elevation
limit.
In the following discussion, the "available superheat" is defined as the
local primary coolant temperature minus the secondary side temperature
corresponding to the secondary side saturation pressure during normal
operation. This corresponds to the condition when boiling terminates as a
result of the solution vapor pressure, at the local primary coolant
temperature, reaching the secondary side pressure. The term "crevice
superheat" is defined as the difference between the temperature of the
crevice solution and the saturation temperature corresponding to the
secondary side pressure. In general, crevice solution concentrations will
be limited to approximately 20 molal on the hot leg side of the drilled-hole-
support steam generator and 5-10 molal on the cold leg side. As noted,
local concentrations may not reach those corresponding to the boiling point
elevation limit.
FIGURE 8-11
Relation of Crevice Concentration and Steam Generator Conditions
(A = Bulk Water Concentration; B = Concentration Limit) (8.19)
Increased
Solution
Concentration
~1)
C,,
Co
'I,
0~
C,,
240 260 280 300 TH 320
Temperature, 0C
Effect of Dissolved Oxygen. Pathania and Balakrishnan (8.20) indicate

that the presence of oxygen can increase acidification by segregation of the
location of the anodic and cathodic reactions. In particular, if the oxygen
reduction (cathodic reaction) occurs outside or near the mouth of the crevice,
and the corrosion reaction (anodic reaction) takes place within the crevice,
an increase in hydrogen ion concentration and, as required for
electroneutrality, an increase in anion concentration (e.g., chloride) result.
02 + 2H 2 0 + 4e = 40H (cathodic reaction) (8-3)
3Fe + 4H 2 0 = Fe 3 0 4 + 8H+ + 8e- (anodic reaction) (8-4)
Each mole of oxygen reduced outside the crevice generates four moles
of H+ and causes the CI- or comparable anions to enter into the crevice.
This hypothesis was supported in single-tube model boiler tests with
seawater (8.21). Specifically, the corrosion rate increased by approximately
a factor of 2 in the presence of 40 ppb oxygen at 1.7 ppm chloride. This
represents a change in crevice pH from 4.6 to 3.4 in the presence of oxygen
according to Pathania and Balakrishnan's calculations-(8.20).
Mann and Castle (8.22) found that the presence of dissolved oxygen or
copper provides a satisfactory explanation for denting observed at fresh-
water-cooled units where condenser leaks may increase alkalinity in the
steam generator. Under these conditions, the reduction of oxygen or
oxidized copper species provides a mechanism for generating acidic
solutions within the tube/support plate crevice.
They further found that oxidized copper species (ions or oxides) are
much more effective than dissolved oxygen at initiating corrosion because
they are not volatile and may penetrate into the crevice before reacting.
The ability of copper to initiate corrosion in neutral or alkaline chloride
solutions is greatly enhanced byprior exposure to dissolved oxygen. Dosing
hydrazine into the experimental system eventually suppressed the ability
of copper to initiate corrosion under these conditions. Since dissolved
oxygen is stripped from solution into the steam phase in the steam generator,
the crevice appears to. not be much more susceptible to corrosion by
dissolved oxygen in neutral or alkaline chloride solutions than an unheated
steel surface.
Laboratory Denting Experiments. Environments which have been used

to produce denting in laboratory simulations have included a wide range
of hydrolyzable and neutral chlorides both oxygenated and deoxygenated.
In some cases, metal oxides and phosphates were added. Experiments to
produce and/or inhibit the growth of oxide and resulting mechanical
deformation associated with denting have been the focus of several studies.
Consequently, in some cases, extraction of corrosion rate data or
determination of the synergistic effect of corrodents is difficult. In addition,
a wide variety of test methods have been employed. An overview of key
issues relating to the laboratory studies is given below.
Heat Flux and Concentration Effects. The cause of denting is the

formation and concentration of acid chlorides in crevices between the tube
support plate and the steam generator tube. The source of the chlorides in
most cases has been the ingress of seawater through condenser tube leaks.
The concentration process, resulting from boiling within the crevice, has
been shown to produce concentration factors of greater than 60,000 in the
laboratory. For example, a bulk water concentration of 50 ppb of chloride
can produce a crevice concentration of greater than 3,000 ppm.
Concentration factors can be much greater at lower bulk water
concentrations and much less at higher bulk water concentrations (8.20
8.23) because of the solution concentration limit imposed by the boiling
point elevation.
Wootten, et al. (8.24) calculated that the heat flux through a steam
generator tube wall can lead to formation of solutions ten times more
concentrated than seawater at the boiling point elevation limit, suggesting
that ionic impurities from a seawater condenser leak can theoretically
concentrate by orders of magnitude under the effects of local heat flux and
boiling. Heat flux and dryout effects in crevices were investigated by Baum
and Curlee (8.25-8.26) who showed that the crevice geometry is of primary
importance in creating dryout regions. Concentration factors up to 20,000
were observed during their studies.
Studies of concentration factors also have been carried out by Mann
and Castle (8.27). They showed in autoclave tests that concentration of
sodium chloride in a heated crevice was a linear process which takes place
with a concentrating efficiency of 0.46-0.80 of the theoretical rate. The rate
of accumulation was found to be a function of the feedwater (or the steam
generator bulk water) concentration and heat flux. Concentration factors
of greater than 60,000 were measured.
Bawden 8.28) showed that acid chlorides could be formed in heated
crevices from seawater inleakage, whereas typical river or lake water
normally becomes acidic only when there is inleakage of acid sulfate from
makeup water or cooling tower cooling water.
Computer codes that calculate pH of the seawater solution as the salinity
increases (8.16-.8. 8.28) have been prepared by several investigators.
Two major types of tests were carried out in the investigation of denting
corrosion: isothermal tests in which a corrodent at a relatively high
concentration was used, and heat transfer tests in which a much lower level
of corrodent was used but local boiling provided a concentrating effect much
like an actual steam generator. The isothermal tests (corrosion coupons in
autoclaves, tube capsules, electrochemical tests, etc.) were used to screen
possible corrodents and possible environment-modifying remedial
measures, and to determine the contaminant level required to initiate and
propagate corrosion. Such tests are relatively inexpensive and allow initial
test conditions at room temperature to be known. They suffer, however,
from a lack of solution refreshment during testing and probable depletion
or modification of the corrodent solution. The heat transfer tests, e.g.,
refreshed autoclaves, model boilers, etc., overcome the problems of
refreshment and depletion of the corrodent but are more expensive and
more complex. Although such tests more closely simulate conditions in
the steam generator, corrodent solution properties in the crevices, where
corrosion takes place, cannot be readily determined.
Isothermal Capsule Tests
Capsule Design. To simulate a crevice, a close-fitting cylindrical slug of

support plate material is inserted into a length of alloy 600 steam generator
tubing with alloy 600 end caps forming a capsule. The crevice is established
by the clearance between the slug OD and the tube wall ID. This reverses
the geometry of a real support plate crevice and makes a simple system for
measuring tube bulging (equivalent to denting). Since there is no thermal-
hydraulic mechanism to concentrate impurities in the crevice, the capsules
are filled with concentrated solutions to simulate solutions in steam
generator crevices. An external heat source is used to maintain temperature.
Nonprotective magnetite formed on the OD of the slug first fills the crevice,
then bulges the capsule.
Reference Chemistry. For isothermal tests and heat transfer tests, the
general approach was to develop a reference denting chemistry and compare
denting rates for a wide range of other chemistries to that in the reference
chemistry. The denting chemistries most commonly used were either strong
acid chlorides which usually consisted of metal chlorides (CuCl2 , NiC12,
FeC12, and FeC13 ) and hydrochloric acid, weak acid chlorides such as MgC12,
seawater, or combinations of these.
Pathania and Balakrishnan (8.20) developed a chemistry model to
calculate the chemistry of the environment in the isothermal test capsules
as the tests proceeded. They reported that a change in high temperature
pH occurred in a short time, and pH reached a steady value in less than ten
days. The calculated high temperature pH of a 0.02 N chloride solution of
seawater and several other chlorides after ten days is given in Table 8-1.
TABLE 8-1
Isothermal Capsule Test pH After Ten Days (8.20)
Chloride
Normality Medium pH
0.02 HCI 3.38
0.02 FeCI2 3.63
0.02 CuCI2 3.32
0.02 MgCI 2 4.70
0.56 Seawater 4.82
Corrosion Process. The capsule corrosion process can be considered to

occur in at least two stages. First, corrosion of the plug leads to oxide growth
into the open annular space between the plug and the inside diameter of
the alloy 600 tube. This is designated as the crevice fill period. As noted
previously, the volume of the oxide formed during the unrestrained
corrosion process is approximately twice that of the corroded metal. The
time to fill the crevice corresponds to the time to initiate denting in a PWR
steam generator. The corrosion rate during this period would be similar to
the oxide growth rate for unrestrained specimens such as coupons, which
can be used to establish oxidation rates.
After the oxide fills the annular gap and contacts the alloy 600 tube
surface, further corrosion leads to outward bulging of the tube. In some
cases, oxide growth continues at a linear rate. In other cases, decrease in
corrosion (and bulging) rate occurs at initiation of bulging. Broomfield, et
al. (8.12) found a complex relationship in corrosion rates when using a stack
of carbon steel coupons in a special restraining test assembly. Beineke, et
al. (8.29) reported that the corrosion rate in capsules was reduced by a factor
of ten by introduction of restraint when the crevice became filled as shown
in Figure 8-12.
FIGURE 8-12
Comparison of Calculated Corrosion Rates During Crevice Fill
Period Versus Bulging Period for Isothermal Capsule Tests (8.29)
56
A 120
48 -0.25 M F--• 0.MCum

44.1 44 -- ~Seawater
M NiC• 10
4550 1288 C) 565°F (296-C) r A
936a
eS32 -Ga
11 Diametral
mo
E -- u0 m
E
024
16 - Crevice Fill Period 40 g
12
00 5 10 15 20 25 30 35 40 45 50 55
Exposure Time (days)

Pathania and McVey (8.30) used two types of specimens in studying

carbon steel corrosion in concentrated seawater, i.e., carbon steel coupons
and a tapered carbon steel sleeve fitted over an alloy 600 tube to form a
crevice. The coupon corrosion rates were nearly an order of magnitude
higher than that in the crevice.
In summary, the capsule corrosion rates generally are lower than those
measured for coupon tests. The effect of restraint by the capsule wall maybe
is a major factor in this difference. Comparison of several separate studies
suggests that the effect of restraint in the capsule tests is to reduce apparent
oxide growth rates tenfold at 572°F (300'C). However, since weight loss
measurements and oxide porosity or density measurements generally were
not made, it is not possible to quantify carbon steel corrosion rates with
and without restraint.
Corrosion in Various Chloride Solutions. Kunig and Economy (8.31)

evaluated Westinghouse isothermal capsule test data to determine which
contaminants caused or contributed to accelerated denting and to determine
the threshold levels at which these effects occurred. Their findings are
summarized in Table 8-2. It can be seen that strong acid chlorides (HC1,
CuC12 , FeC12 ) produced nonprotective magnetite (NPM) at 0.02 N chloride,
and tube deformation at 0.05 N chloride. A weak acid chloride (e.g., MgC12 )
took about the same concentration to generate NPM, but required 2 N
chloride to cause tube deformation. The neutral chloride solution (seawater)
required 2.9 N chloride to produce tube deformation.
Sawochka, et al. (8.23) attempted to correlate the corrosion data from
Westinghouse (W) and Combustion Engineering (C-E) denting studies. The
isothermal capsule data included those with fill solutions containing
impurities such as seawater, NaCl, metal chlorides, HCI, H 2S0 4 , MgC12 ,
etc. at temperatures of 228 to 296°C.
A general corrosion rate increase with a change from neutral to acidic
conditions was apparent (Figure 8-13). Although the general finding was
similar to that reported by Mann (8.13), significant data scatter was present.
Similarly, corrosion rates were shown to increase with chloride concentration
(Figure 8-14). When results obtained with seawater were compared to those
with strong acid chlorides (e.g., CuC12 , NiC12, FeC12 , FeC13, and hydrochloric
acid), two separate clusters of corrosion data were noted as a function of
pH (Figure 8-15) and chloride concentration (Figure 8-16). This made it
apparent that the corrosion/chemistry relations differed in neutral and
strong acid chloride solutions. Similar relations appeared to exist between
corrosion rate and pH when seawater and MgC12 capsule data were
compared.
TABLE 8-2
Summary of Westinghouse Isothermal Capsule Tests
at 550°F (288°C) (8.31)
PRESENCE OF NPM 1 ON CARBON STEEL
Chloride (N)
Cation 0.005 0.02 0.05 0.1 0.5 2.0

H+ No Yes Yes 2 Yes
Cu++ No Yes Yes Yes
Fe++ No Yes Yes
Fe+++ Yes
Mg++ No Yes Yes Yes Yes Yes
Na+ No3 No3
PRESENCE OF NPM 1 IN SEAWATER TESTS

X-Fold 0.1 0.5 1.0 5.0ý 10.0
CI (N) 0.06 0.3 0.6 2.9 5.7
No Yes Yes 4 Yes Yes
1
NPM = Nonprotective magnetite.
2
Yes = Sufficient NPM to fill crevice and produce tube deformation.
3
4
W---h or without CuO or NiO.
With or without SiO2
Based on the capsule data, the following series of correlations (8.23)

were developed employing linear regression analyses and utilizing the
chloride/pH relationship based on the Hitch code (8.16):
Seawater log CR = 1.115 + 1.754 log [Cl] (8-5)
Magnesium log CR = 1.59 + 0.826 log [CI] (8-6)
Chloride
Strong Acid log CR = 3.898 + 1.702 log [Cl'] (8-7)
Chlorides
where,
CR = Corrosion Rate, mils/year
[ClI] = Chloride Ion Concentration, molal
FIGURE 8-13
Corrosion Rate and pH (8.23)
100,000
10,000
0
C
1,000
:8 0
100
0
E
B
10 0 8
03
0
1.0
0
0.1
2 4 5 6 10
pH
Using a similar approach, Pathania and Balakrishnan (8.20)

independently correlated the W and C-E data and noted that the chemistry
and corrosion were affected by the form in which copper was added to the
system. Their calculations showed that if copper were added in the form
of CuC12 rather than CuO, the resulting pH was lower.
The time-to-denting correlation they developed for the isothermal
capsule test results based on linear kinetics was of the form:
CRL = 1.34 x 106 [H+][Cl-1 (8-8)
Based on parabolic kinetics, their corrosion rate (CRp) correlation was:
CRp = 7.74 x 104 [H+][Cl-]0 .5 (8-9)
FIGURE 8-14
Corrosion Rate and Chloride Concentration (8.23)
100,000
10,000
1,000
I 0
0
0
a
0 0
100 0
0
10 0 0
2 0 0
8 00 0
0
8
1.0
8
0 0
0g
0o
0
0 8
0
0
0
0 0
0
0.1 0 0
0
0
0.01
0.0001 0.001 0.01 0.1 1.0 10 100 1,000
Chloride, M
They were able to combine all isothermal capsule test data (strong acid
chlorides, seawater, etc.) into the correlation resulting in a corrosion rate
within a factor of 7 at the 95% confidence level for their linear rate, and a
factor of 5 for the parabolic rate. In comparison, the correlations done by
Sawochka, et al. (8.23) considered the seawater and strong acid chloride
data separately with corrosion rates within a factor of 2.2 at the 90%
confidence level for both cases.
Early autoclave coupon tests were performed by Vaia, et al. (8.32) who
investigated the oxidation rates of mild steel and 12% Cr steels in several
solutions. The carbon steel results are given in Table 8-3 along with the
calculated pH at 300'C (8.20).
FIGURE 8-15
Comparison of Seawater (x) and Strong Acid Chlorides (o) on
Corrosion Rate/pH Relation (8.23)
10,000
1,000
100
m 10
0
1t0
0.1
0.01
2 3 4 5
pH
Apparently, a protective inner oxide formed on the carbon steel coupons

exposed to the near neutral NaCl/MgC12 and concentrated seawater
solutions, resulting in the leveling off of the weight loss curve after about
40 days as shown in Figure 8-17. However, in solutions containing nickel
chloride, linear corrosion (Figure 8-18) with oxide deposits showing a
succession of nonprotective layers was observed. These results are
consistent with the findings of Sawochka, et al. (8.23), i.e., corrosion
mechanisms in neutral chloride solutions (such as seawater) differ from
those in strong acid chlorides.
Other evidence of this difference can be inferred from the results of
Beineke, et al. (8.29). Beineke's objective was to determine the effects of
chemistry on corrosion after denting had been established. He later found
that the extent to which the bulging rate was reduced by the addition of
various alkaline solutions was determined by the aggressiveness of the
denting environment prior to isothermal soaking and did not appear to be
a function of the composition of the soaking solution. For example, a dent
Tube Support and Tubesheet Corrosion (Acidic Crevices-Tube Denting) 8.27
FIGURE 8-16
Comparison of Seawater (x) and Strong Acid Chloride (o) on
Corrosion Rate/Chloride Relation (8.23)
10.000 r
1,000 P
0
0
1001- 0
10 F
8
1.0 P x
X
0.1 F
I
0.01
0.001 0.01 0.1 1.0 10
Chloride, M
produced by reactions involving Cu++ could not be arrested by isothermal

soaking, but dents produced from seawater (MgC12 ) or FeC12 could be
arrested by a change in chemistry. (The effectiveness of soaks, flushes, and
neutralization is discussed in Chapter 17).
Corrosion in different chloride solutions also were addressed by Kunig
and Economy (8.31). They showed that the corrosion rate of carbon steel
exposed to seawater in capsules at 190,000 ppm chloride was higher than
in 19,000 ppm seawater chloride, but both were considerably less than in
more dilute solutions of NiC12 or CuC12 . This was further demonstrated in
a series of capsule tests in which seawater additions were made to the (0.1 M
NiC12 + 0.1 M CuC12 ) environment. Although increasing the seawater
concentration can adversely affect corrosion, it can also slow aggressive
corrosion in strong acid chloride solutions. When the seawater is
concentrated to ten times its original strength, the addition of 0.1 M NiC12
+ 0.1 M CuCI 2 has little or no influence on capsule deformation (Figure 8-19).
TABLE 8-3
Corrosion Rates of Mild Steel in Autoclave Tests (8.31)
CI- Weight Loss Weight

Conc. Calc. pH Temp. mg/cm 2 Loss
Molal at 300 0 C °F/°C (day) Curve
NaCl/MgCI2 _- 6.4
0.12 4.8 518/270 30 (84) Plateau
5x Seawater
2.9 4.2 540/282 44(140) Parabolic
0.07 x Seawater + 0.0257 M FeCI2 + 0.025 M NiCI2
0.15 3.4 550/288 600 (28) Linear
NiCI2
0.16 3.3 550/288 950 (28) Linear
FIGURE 8.17
Descaled Weight Loss Versus Exposure for Carbon Steel in a
Seawater Solution and a NaCI/MgCI 2 Solution at Elevated
Temperature (8.32)
50
"40
6 30
.•
_• 20
l1
10
Q 0
0 20 40 60 80 100 120 140

Exposure (days)
Tube Support and Tubesheet Corrosion (Acidic Crevices-Tube Denting) 8.29
FIGURE 8-18
Descaled Weight Loss Versus Exposure for Carbon Steel in a NiCI2
Solution and a Seawater/ FeCI 2 /NiCI 2 Solution at 550 0C (288 0 C) (8.32)
1000
900
800
700
0
0)
E 600
ef~
C,,
0
-j
0)
a,
500
-u
a,
(a
U
(I,
a, 400
300
200
100
0
0 10 20 30 40 50
Exposure (days)
FIGURE 8-19
Tube Deformation in Capsule Tests As a Function of Composition
for Seawater Solutions (8.31)
0.6 Ii i
a)
4)
Seawater 288°C 20
15 A+0.1 M Ni C12
S 0."4•- +016
0 +OAMCu 12
EE
E
a, S eaw ater-- 4
0)
0
o 0
0 4 5 6 7 8 9 10
Seawater Concentration (X fold)
In the field, copper is introduced to the steam generators by corrosion

of copper alloys in the condenser and feedwater heaters, particularly in the
presence of dissolved oxygen. Copper primarily enters the steam generator
as an oxide or hydroxide rather than as cupric chloride even in the presence
of caustic. Therefore, for predictive purposes, Pathania and Balakrishnan
claimed (8.20) that tests performed in a seawater or seawater + CuO system
are more representative of steam generators than those in a CuC12 + seawater
system.
Effect of Temperature. Economy and Smith-Magowan (8.33) determined

the effect of temperature on corrosion rate in strong acid chloride solutions
based on capsule test data (Figure 8-20 and 8-21).
Unfortunately, comparable tests with seawater at various temperatures
were not performed to determine whether a similar effect existed for
seawater contamination.
As a result of analysis work carried out to define the dependence of
corrosion rate on temperature, Partridge (8.34) compiled the data shown in
Figure 8-22. In this figure, he attempts to relate the ratio of corrosion rates
at various temperatures with respect to a typical hot leg temperature of
600'F, based on a differing range of activation energy during the crevice fill
and during the bulging (after-fill) periods. His ratio data are consistent
S
with Economy and Smith-Magowan's data (Figure 8-20).
FIGURE 8-20
Effect of Temperature on Bulging in 5,800 ppm NiCl 2 and CuCI2
Solutions (0.16 M Cl) (8.33)
Temperature (0 C)
200 220 240 260 280 300
0,
0I 10
(08
N 4
I 2
500 600
Temperature (OF)
FIGURE 8-21
Effect of Temperature on Capsule Bulge Size (8.33)
16- 16Legend:
SOCUC12
E 12-
c 8--
0
E 0
.9
5000 F
Exposure Time (weeKs)

FIGURE 8-22
Temperature Dependence of Denting (8.24)
U-
ai
Stage in Denting Range of Activation Typical Range of Corol

Cc Process Energy, 0, kcarmole at 600'F (alter initiation), rn•
Crevice Fill 10 60 to 90
Bulging (after 15 to 20 30 to 50
540 550 560 570 580 590 600 610 620 630 640 650 660
Typical Typical
Cold Leg Hot Leg
Temperature, oF
Alternate Support Plate Materials. Corrosion of alternate tube support

plate materials has been investigated in many solutions. The relative
corrosion rates of carbon steel and type 405 (12% Cr), and the estimated pH
at 300'C from various studies, are given in Table 8-4 (8.20). As shown, the
initial or pre-bulge corrosion rate for Type 405 is generally a factor of 2 or 3
less than that of carbon steel in concentrated seawater, and a factor of ten to
hundred less in more aggressive solutions containing HC1, FeC12, or NiC12.
Such 1f2% Cr steels form a duplex type of oxide with an inner layer of
iron-chromium spinel and an outer layer of magnetite crystals. Both general
corrosion and localized pitting occur in acid chloride solutions. In
concentrated solutions denting may result (8.12). As noted above, the change
in corrosion rate on exposure to acid chlorides compared to more neutral
chlorides is much greater for carbon steels than for Type 405 (note ratio of
corrosion rates in Table 8-4). This implies that the large improvements (10
to 100 X) noted in these test programs may not be realized in steam generator
crevices that are more likely to contain concentrated seawater than strong
acid chlorides.
TABLE 8-4
Corrosion Rate of Type 405 Stainless Steel in Acid Chloride Solutions (8.18) CD
Estimateed Initial Corrosion Rate 0

-t
pH at Cl (mils/y) C.R. Type 405
Reference Type of Test Environment 3000C M Carbon Steel Type 4 05 C.R. Carbon Steel Z)
Kunig & Capsule HCI + NaCI 3.6 1.0 278 4.7 0.017
Economy 2880C HCI + NaCI + CuO 3.4 1.0 278 5.2 0.018 0D
(8.29)
cn
HCI + NaCI 6.2 0.1 0.12 0.04 0.34 zr
HCI + NaCI + CuO 6.1 0.1 0.105 0.111 1.0

HCI + NaCI 6.2 1.0 0.48 0.15 0.3
HCI + NaCI + CuO 5.9 1.0 0.64 0.12 0.18
Vaia et al. Coupons NaCI + MgCI2 4.8 0.12 4.8 1.9 0.39
(8.31) 270-288°C 5X Seawater 4.2 2.9 5.4 1.7 0.31
0
FeCl 2 + NiCI2 + 3.4 0.15 518 11.5 0.022
Seawater
NiCI2 3.3 0.008 614 66.5 0.11
Broomfield Coupons NiCI2 3.3 0.002 29.3 0.025 0.00085
et al. (8.33) 3000C NiCI2 3.3 0.02 397 0.081 0.0002
NiCl2 3.3 0.2 569 0.32 0.00056
NiCI2 3.3 2.0 725 2.3 0.0032
NiCI2 3.3 4.04 1450 4.8 0.0033
NiCI2 + CuO 3.3 4.04 5177 6.8 0.0013 )D
NiCI2 + FeCI2 + Seawater 3.4 1.0 380 0.69 0.0018
NiCI 2 + FeCI 2 + Seawater 3.4 1.0 328 0.65 0.002
Studies by Vaia, et al. (8.32) show that the corrosion rates of other 400
ferritic stainless steels are comparable to that of Type 405. Type 347 austenitic
stainless also showed added resistance to the acid chlorides (8.12).
McKubre and Leach (8.35) compared'the galvanic corrosion of various
support alloys with that of carbon steel coupled to alloy 600 tubing in dilute
hydrochloric acid, seawater and river water environments (Table 8-5).
Their findings were comparable to those of others, i.e., as the total
alloying content (Cu + Ni + Mo + Mn + Si) increases, the corrosion rate
decreases.
TABLE 8-5
Relative Improvement Factor in Corrosion Rates:
Carbon Steel/Alloy (8.35)
Corrosion Rates: Carbon Steel/Alloy

Electrolyte lox loX
10- 3 N 10- 3 N Susquehanna Millstone
Aflty HCI H2 _SO River Seawater
405 1.5 3.5 5.8 4.5
409 - 3.6 2.8 3.9
410 1.4 2.9 1.3 2.1
430 2.9 4.9 --
508 - 2.7 2.3 2.3
347 5.6 26 --
800 48 47 -
1018 1 (mpy) 48 56 29 85
1Corrosion rates for 1018 are representative of the largest sustainable values at
2500C for the given conditions of crevice geometry and electrolyte composition.
Effects of Other Anions. Evaluations of the impact of anions other than

chloride on NPM growth have been limited. However, sulfate has been
considered as a potential dent-producing corrodent in both capsule and
heat transfer tests. In no case, however, was denting produced nor was the
presence of a nonprotective magnetite oxide layer (suggesting incipient
denting) observed in capsule testing with H 2SO 4 at concentrations up to
1 N (48,000 ppm sulfate) and CuSO 4 at 0.1 N (4,800 ppm sulfate).
Potter and Mann (8.),showed that bromide ion (Br-) will produce the
characteristic nonprotective oxide. It also has been suggested that fluoride
might support denting. However, it is unlikely that high concentrations of
fluoride could be present in acidic crevices due to the high volatility of HE
In another study (8.36), 1018 carbon steel and alloy 508 (typical of
tubesheet steel) were exposed in capsules to a variety of corrodents
(Table 8-6). Of the various corrodents, only MgC12 produced nonprotective
magnetite. Pitting of alloy 600 and the ferrous alloys was the predominant
form of attack.
TABLE 8-6
Average Pit Depths in Capsules
Time = 8 weeks, Temp. = 564 0 F (296'C) (8.36)
Avg. Pit Depth, mils

AISI 1081 A508
Test Solution Alloy 600 Carbon Steel Alloy Steel
0.2 M Na 2 SO 4 0.5 1.6 9.4
CaSO 4 (Sat. @ 25 0C) 0.7 1.4 6.8
0.2 M CaCl2 1.0 2.3 2.2
0.2 M NaCI 1.0 6.0 2.8
1
0.2 M MgCI2 1.0 NPM 1.6
0.2 M MgSO 4 0.8 1.5 9.3
Turkey Point Seawater 0.8-2.1 (var) 4.6 9.8
0.1 M NaHCO 3 0.4-3.0 (var) 3.5 3.6
+ 0.1 M NaCI
0.1 M Si0 2 0.7 3.1 0.6
0.1 M NaHCO 3 1.7 1.8 0.8
0.1 M NaCI + 0.1 M Si0 2 0.6 1.8 13.2
0.1 M MgCl2 1.1 2.6 4.1
ANO-2 Water, 1OX 1.0 1.8 1.1
Concentration
ANO-2, 1 OX With H2 SO4 0.6 3.1 3.2
'Nonprotective magnetite.
Heated Crevice Tests

Crevice assemblies with internal electrical cartridge heaters were used
in several investigations of denting. In general, the apparatus consisted of
an autoclave containing an alloy 600 steam generator tube with an electrical
cartridge heater inside the tube and a carbon steel ring surrounding the
tube to simulate a support plate. As such, the heated crevice tests use a
constant heat flux that can produce a higher superheat than the constant
temperature heat source present in an operating PWR steam generator (8.20).
Wootten, et al. (8.24) utilized 0.01 fold seawater solution (200 ppm Cl) and
found that denting could be established with CuO in the bottom of the
autoclave. When the CuO was removed, the denting rate decreased
(Figure 8-23). Denting re-initiated on addition of CuO.
Von Neida, et al. (8.37) showed that deoxygenated 1% seawater required
the presence of copper and/or nickel oxides to cause denting. Denting
rates near 100 mpy (Figure 8-24) were obtained in a variety of tests with
seawater and copper oxide sludge (8.33).
FIGURE 8-23
Progression of Denting in Heated Crevice Device (8.24)
0.01 Fold Seawater 0.01 Fold Seawater

+CuO 0.01 Fold Seawater +CuO
14
X
I i/
E 12
10
2
8
E
6
4
0
2
2
2 4 6 8 10 12 14 16 18
Time, Weeks
FIGURE 8-24 -
Denting Rate From Heated Crevice Tests With Magentite-Prepacked
Crevices (f = 14,300 Btu/ft 2 -h, TSAT = 540'F [282'C]) (8.33)
- 20
Em
D
0,) 16
Z 12
E
.5
0) 8
CD
4)
<4
0
00 10 20 30 40 50 60 70
Exposure Time (days)
Vaia, et al. (8.32) repeated the seawater + CuO experiment of Wootten,

et al. but found a lower corrosion rate. This may have been due to a different
superheat leading to differing concentrations within the crevice. Since
concentration effects are significant, differences in available superheat can
lead to large variations in crevice concentrations.
Castle and Mann (8.38) performed heated crevice tests with alkaline
ammonia-dosed water (pH 8.5 to 9) at 536°F (280'C) and found only low
corrosion rates in open crevices at up to 48 ppm C1- (as NH 4 C1). In packed
crevices, rapid corrosion occurred with deoxygenated ammonium chloride
solution; the corrosion rate was pro ortional to the chloride concentration
and the heat flux. At 150 kW m7 the corrosion rate was 8.4 gm/wk
(17.2 mpy) per ppm of C1-. With an oxygenated NaCl solution at pH 8.5 to
9.0, rapid corrosion was also observed showing that acid conditions were
generated. Only heat flux was reported in these tests.
. Cylindrical shrouds incorporating crevices (Figure 8-25a) were used by

Brown, et al. (8.39) to examine the effects of sludge packing. A corrosion
rate equivalent to 1,540 gm/wk (3,151 mpy) was observed after 39 days in
288°C lake water with 29 ppm chloride, under a magnetite sludge with
small amounts of Ni, Cu, Cr, and Zn as oxides and phosphates. The original
solution pH (25°C) dropped from 8.0 to 5.3 in <20 days.
FIGURE 8-25
Shroud Assemblies
(a) Cylindrical Shroud (b) Conical Shroud

Assembly Assembly
Alloy 600 Steam
In further investigations, Dautovich, et al. (8.40) used 50% lake water

containing 3 ppm C1 in a conical shroud arrangement packed with sludge
(illustrated in Figure 8-25b). Pits up to 100 gtm were observed after 6 weeks
of exposure, but denting was not.
Economy and Smith-Magowan (8.33) summarized their heated crevice
tests by noting that high heat flux crevices competed more effectively than
low heat flux crevices for contaminants in the bulk solution. At higher
steaming rates associated with higher heat flux, crevices concentrated

contaminants more rapidly. This observation is consistent with plant data
which show more extensive denting on the hot leg where the available
superheat and heat flux are greater than on the cold leg. Additional testing
(8.33) revealed a decreasing trend in denting rate as heat flux was further
increased.
They concluded that the "... heat flux serves principally as a dent-
initiating factor but that the progress of denting is ultimately governed by
other factors."
Pathania and Balakrishnan (8.20) developed the relation between crevice
pH and crevice chloride concentration in seawater shown in Figure 8-26.
The corrosion rate for linear kinetics with the estimated 95% confidence
interval (±2 a) is included. A similar plot for a mixture of cupric chloride
and seawater (1:2 ratio) is shown in Figure 8-27.
FIGURE 8-26
The Relationship Between Crevice Chloride Concentration, Crevice
pH, and Carbon Steel Corrosion Rate for Seawater (8.20)
6.0 4.0
10_. 3.0
o5.0 S.7-.-
/ 102 2.0
11.0
a.
Q)
C, +20'
-,
16c;io• .
2•
-" 0
0
0 Wn
1(1(
1A ý
.
_
102 103 104 101
Crevice Chloride Concentration, ppm

FIGURE 8-27
The Relationship Between Crevice Chloride Concentration, Crevice
pH, and Carbon Steel Corrosion Rate for Cupric Chloride-Seawater
Mixture (1:2) (8.20)
5 O.IJ
4.0
4 3.0
o,-
C)
74 102- 2.0 0"
ID
0~
0 lol- 1.0
/ 10o 0
2 I I I I I [
4
I I I I II -1.0
10 10 5
102 10 106
Crevice Chloride Concentration
Model Boiler Tests. The term model boiler is used to describe an apparatus
which includes a circulating boiler system with a refreshed corrodent supply.
A single tube boiler system described by Esposito, et al. (8.41) was qualified
by carrying out repetitive tests with seawater (0.2 ppm C17) and copper
chloride (0.1 ppm C1). A denting rate of 45 mpy was consistently achieved
(Figure 8-28). Dent initiation was generally observed in 10-20 days in these
tests (8.42).
Model boilers using conical shrouds were instrumented with strain
gauges by Beineke, et al. (8.29) to measure corrosion. These corrosion rates,
however, cannot be directly compared with the denting rates measured by
other authors, since they only monitor motion at one point on the
circumference of the tube. However, the data confirm that even at very
low heat fluxes (e.g., 10,000-Btu/ft 2 -h) denting is initiated in typically less
than two months from test initiation at bulk water concentrations of 100 to
200 ppm chloride.
FIGURE 8-28
Statistical Evaluation of Denting Rate From Reference Denting Tests
(8.42)
15
14
13
12
11
C
10
C
9
8
E 7
6
0) 5
4
3
2
1
20 40 60 80 100 120 140

Time After Onset of Denting (Days)
White and Berry (8.43) reported that NaCI (15 ppm) did not cause
denting in a deaerated model boiler (with or without phosphates), nor did
sea salt with a mixed Fe/Ni oxide (Table 8-7). However, copper and nickel
chlorides produced denting which was moderated by the presence of
phosphate in the case of nickel chloride. The results of White and Berry as
well as those of Wootten for pure NaCl agree with laboratory results
obtained by Ashford, et al. (8.44) in their investigations of corrosion in
conventional boilers. They measured the hydrogen evolution from the
corrosion of a mild steel heated crevice in 100 ppm NaC1 (both aerated and
deaerated) at 752-932°F (400-500'C). Oxygen caused rapid corrosion in NaCI
solutions because it causes the formation of an acid chloride environment
at the metal surface. The oxide was mainly magnetite, with small amounts
of hematite.
TABLE 8-7
Corrosion Rates in Pot Boiler Tests (8.43)
Primary Temperature-635 0 F (335 0 C)
Secondary Temperature-550°F (2880 C)
Chemistry Denting
(15 ppm chloride salt)
P04 + NaCI None
NaCI None
CuO + CuCI 2 Slight - Moderate
P04 + CuO + CuC12 None - Very Slight
Fe 3 0 4 + NiCI2 Moderate - Heavy
NiFe 2 0 4 + Sea Salt None
To interpret the pot boiler, single tube model boiler and model steam
generator data developed by W and C-E, Sawochka, et al. (8.23) and Pathania
and Balakrishnan (8.20) employed the observed corrosion rates to predict
the crevice chemistry that must have been present based on the chemistry/
corrosion correlations developed from the capsule data. From this, they
calculated a concentration factor for each type of test. The concentration
factor between the bulk water and crevice varied as a function of bulk water
chloride concentration. At high bulk water chloride concentrations, i.e.,
5 ppm or greater, the concentration factor approached that corresponding
to the available superheat. At lower bulk water concentrations, the
calculated concentration factors were significantly less than those possible,
based on available superheat indicating that thermal-hydraulic/diffusion
effects were establishing concentration levels in the crevice. Sawochka, et
al. (8.23) developed the following equations for correlating the concentration
factor (CF) with bulk water chloride concentration for seawater
contamination:
Pot boilers (C-E):
log CF = 4.129 - 0.6252 log ClB (8-10)
Single tube model boilers (W):
log CF = 4.911 - 0.5121 log ClB (8-11)
Model steam generators (C-E):
log CF = 4.645 - 0.8725 log ClB (8-12)
where,
CIB = chloride concentration in bulk water, ppm
Figure 8-29 compares these concentration factors as a function of bulk

water chloride concentration. As noted above, a significant decrease in
concentration factor was observed as the chloride concentration increased.
The pot boiler concentration factors are considerably lower than those
for the model steam generators or single tube model boilers. Sawochka, et
al. attributed this difference to the lower heat flux in the pot boiler apparatus.
Since the heat flux in the latter is reasonably similar to that in a commercial
steam generator, concentration factors for these apparatus are considered
more applicable to prediction of corrosion in commercial units.
FIGURE 8-29
Comparison of Concentration Factors From Pot Boilers, Single Tube
Model Boilers, and Model Steam Generator (8.23)
101
106
105 -Single Tube Model Boiler
-/
._
4
10
10
0 Model Steam
Generator
Pot Boiler
103_
102 j
0.01 0.1 1.0 10 100
Chloride Concentration, ppm

Pathania and Balakrishnan (8.20) combined the W and C-E pot and
model boiler heat transfer tests in seawater (with and without CuO) in
relating concentration factor to bulk chloride as shown in Figure 8-30. They
equated this relationship in the equation:
log CF = 4.3304 - 0.808 log CIB (8-13)
Since the crevice chloride concentration is defined as the product of the
bulk chloride and the concentration factor, Pathania and Balakrishnan noted
(8.20) from Eq. 8-13 that "...the crevice chloride concentration tended to be
almost constant." In other words the equilibrium crevice chloride
concentration has only a weak dependence on the bulk chloride
concentration.
Based on linear regression of the model steam generator and single
tube model boiler tests using the data obtained at <2 ppm bulk water
chloride and the chloride/corrosion rate correlation of Eq. 8-5, Sawochka,
et al. predicted PWR steam generator corrosion rates during seawater ingress
to be
log CR = 1.714 + 0.9277 log ClB (8-14)
FIGURE 8-30
Relationship Between Bulk (Blowdown) Chloride Concentration and
0
Calculated Concentration Factors in Westinghouse STMB and
Combustion Engineering Model and Pot Boiler Tests (8.20)
0 W STMB
FI ~< GE MB
02 0 CE PB
00
C
0
C() 10
0 0
102
3102 0
0.01 0.1 1.0 10 100
Bulk Chloride (ppm)
Effect of Superheat. A great deal of attention in studying denting was

placed on available superheat and crevice superheat. In reviewing the
literature, it is important to recognize how each investigator defines these
terms. In heated crevice tests, a linear relationship was found between the
heat flux and crevice superheat (within prepacked crevices) 8.(3). In model
boilers, available superheat is defined as the local primary temperature
minus the secondary saturation temperature (Tpri - Tsat). Experiments at
an available superheat of -65 0F (36 0C) gave crevice superheats of 5 to 10'F
(2.5 to 5.5°C) (8.33).
The relation between crevice superheat and heat source obviously is
dependent on the thermohydraulic conditions in the crevice, solution
composition, amount and density of deposits, relative stage of denting, etc.
(8.33). Figure 8-31 depicts the differences in temperature at the tube support
(as measured by an in situ thermocouple extending into the crevice, [Tts])
and saturation temperature when a model boiler was operated at -30,000
Btu/h-1 ft-2 at an available superheat of 60'F (31.5°C). During operation
with AVT water and no corrodent additions, the T (Ttsp -Tsat) was constant
at ~-5F (2.8QC). Almost immediately upon addition of corrodent, the T
increased significantly, indicating formation of a concentrated solution in
the crevice.
FIGURE 8-31
Effect of Chemistry on TSP Crevice Superheat (8.33)
12
II
11 - Boiler Operation Boiler Operation With Reference II 0
4, -
10- - With Deionized H20 Denting Chemistry 0 I
9 - 0 o 0 II
0
8
0 II
I-. 7
I-
Note Separation I
6
0 in Time Axis \4 1
5 0000000 0
0 II
4 -
II
3
2
II
II
I I I II I I I I I1 I I I Ii I
0
-10 -8 -6 -4 -2 0 2 . 4 6 8 10 12 14 16 18 20 38 40 42
Days Before and After Addition of Denting Chemistry
Pathania and Balakrishnan (8.20) related the crevice superheat to crevice

chloride concentration (CIC) for seawater concentrations in the range 0.005x
to 20x by the following equation:
log T = -3.8697 + 0.93 log C1C (8-15)
The calculated crevice superheats for the pot and model boiler seawater
tests included in their correlation studies were 1.9-3.0°C. The effect of crevice
superheat on the corrosion rate in the presence of seawater was as follows:
log CR = -1.32 + 1.656 log T (8-16)
Model boilers provide an excellent means of monitoring the effect of
available superheat on denting corrosion. This is done experimentally by
adjusting either the primary side temperature or secondary side pressure.
In one series of tests in which Tsat was held constant, Wolfe, et al. (8.42)
showed that the average denting rate in their test facility increased
approximately linearly with available superheat over a range of 35-65°F
(19-36°C). Very little or sporadic denting was obtained in tests where the
available crevice superheat was less than 35°F (20'C). Results are
summarized in Figure 8-32 (8.33). Tests at lower feedwater concentrations
showed that at superheats above about 70'F (39°C) the denting rate again
decreased (Figure 8-33) (8.33).
FIGURE 8-32
Denting As a Function of Available Model Boiler Superheat (8.33)
C,
15 20 25 30 35
5.0o -- 1 1 1 1 1
Make-up Tank Reference
Denting Chemistry
-a 4.0 - 0.2 ppm CI From Seawater
E 0.1 ppm CI From CuCI2 0
"h 2 ppm NH3
0 0
• 3.0 _ 50 ppb N2H4
TSAT'--525 F"
tr 2. (292 C°) 0 8
(D/ 0
0
0 0
1.0
0
0 1
25 30 35 40 45 50 55 60 65
Available Superheat, Tprm - Tsat, F°

FIGURE 8-33
Effect of Model Boiler Available Superheat on Denting Rate at a
Five-Fold Dilution of Reference Chemistry Conditions (8.33)
C,
20 25 30 35 40 45 50
0.4 I IIII
3ý 0.3
/
/I
0.2 Q
o¢ I
5, I
o I \
00 0.1 I
0I
40 50 60 70 80 90
Available Superheat, Tprm - sat, F
Mann and Castle (845) found that the rate of NaCi hideout in a corroded
crevice in heated crevice tests was one-fifth of the rate of hideout in a packed
uncorroded crevice. This was attributed to a reduction in the rate of
evaporation in the crevice due to steam blanketing. Mann and Castle's
work also supported the hypothesis that available superheat controls the
crevice concentration. Salt accumulated in the crevice until an equilibrium
weight was reached. Equilibrium was determined primarily by the available
superheat and free volume in the crevice; bulk water concentration had
only a minor effect.
Summary of Corrosion Rate Test Data. In deoxygenated seawater, there

is general agreement that acid chlorides will form as impurities are
concentrated, and the formation of nonprotective magnetite will result in
denting. In addition, a similar result will occur if neutral chlorides and an
oxidant (e.g., 02, CuO, etc.) are provided. Corrodent concentrations to cause
denting may be different for the two cases.
Examination of the oxide growth rates in isothermal tests shows that
oxide growth rates can be an order of magnitude higher when unrestrained
than when restrained, as in a crevice. The accelerating effect of temperature

is shown by the Economy and Smith-Magowan (8.33) results. Denting rates
in model boiler tests cannot be directly related to bulk water corrodent
concentration since concentration processes govern final crevice chemistry.
One of the interesting corroborating observations from operating plants
is the fact that denting is generally more severe on the hot leg than on the
cold leg. A cursory evaluation of this result would lead to the conclusion
that the corrosion rate is a strong function of temperature, as described
above. While this is true, the more important variable probably is available
superheat. The laboratory data clearly predict that at cold leg superheat
conditions (~35°F/20°C) denting rates will be quite low, whereas at hot leg
conditions denting rates of 50 to 100 mpy may be expected at high bulk
water impurity levels.
Evaluation of the laboratory studies shows a reasonable degree of
correlation in the corrosion rate data. Sufficient evidence was developed
to demonstrate a difference between corrosion/chemistry relations with
neutral and seawater solutions versus strong acid solutions such as NiC12
and CuC12 . The corrosion rates produced in the laboratory model boiler
experiments with seawater are consistent with the rates observed in
operating steam generators once crevice concentration factors are
considered. Measured corrosion rates in isothermal (nonconcentrating) tests
can be correlated with corrosion rates in model boilers (concentrating tests)
by considering concentration factors.
PREDICTION OF DENT INITIATION

Though other factors may contribute to the occurrence of denting, a
plant with condenser cooling water that forms acidic solutions on
concentration is a candidate for denting. The concentration of acid chloride
in the crevice is the single most important factor in the initiation and
progression of denting. Brackish water, seawater, and acidified cooling
tower water generally become acidic as they concentrate. Even water
containing neutral chlorides can form acidic crevices in the presence of
oxygen. When a plant experiences a condenser leak, impurities concentrate
in the steam generator bulk water and further concentrate in the crevices.
In a study of four plants experiencing condenser inleakage, the three plants
whose cooling water concentrated acidic, experienced denting. The fourth
plant, with alkaline-forming cooling water, did not experience denting.
Several attempts have been made to predict the time to dent initiation
in PWR steam generators from the isothermal and heat transfer data.
DeSilva and Wolfe (8.46) designed a model for correlating denting with
cooling water based primarily on two considerations:
(a) existence of a threshold level of chloride ingress to initiate,

(b) ratio of acidchloride to.neutral chloride concentrations.
In this model, the denting susceptibility coefficient, Q, is proportional
to the product of the integrated acid chloride concentration and the mole
fraction of total chloride present as acid chloride or
[acid chloride conc.]

Q = K [acid chloride conc.] x
[total chloride conc.]
Fundamentally, the model assumes that the corrosion rate of the support
plate material (hence denting of the tubing) is dependent on the chloride
composition of the solution contacting the corroding surface. If the material
surface is contacted with neutral chloride, little corrosion occurs; if acid
chloride is present, corrosion is expected.
The first test of the model was a comparison of predicted versus actual
denting of isothermal capsules and then of model boiler tubes, where the
ratio of acid chloride to neutral chloride concentrations could be measured.
Agreement between the predicted and actual denting was good.
In a plant, however, there is no direct measure of neutral or acid chloride
concentrations and, in fact, the measured chloride could be defined as
neutral chloride in most cases. For the purpose of the model, the oxygen
concentration in the feedwater was assumed to be an equivalent measure
of hydrolyzable metal chloride (acid chloride) in the steam generator. The
model was found to be useful in its prediction of denting only if plant
chloride ingress was present in sufficient concentrations to initiate denting.
A blowdown level of 15 ppm chloride days per month for two consecutive
months was assumed to be sufficient to meet this criterion. This greatly
exceeds values achieved at the present time in operating units.
Based on their correlation of isothermal and heat transfer corrosion test
data in seawater 8.(23), Sawochka, et al. developed the following equation
to estimate the time to fill a 15 mil crevice and initiate denting as a function
of the bulk liquid chloride concentration in ppm:
log tD = -0.53 - 0.928 log CIB (8-17)
Employing a similar approach, Pathania and Balakrishnan (8.20)
reported the following relation for estimating the average time for filling a
radius crevice gap of (Yrd mils prior to the onset of denting at a plant with
seawater cooling:
log tD = log Yr - 1.187 - 0.296 log CIB (8-18)
and to fill a 15 mil crevice:
log tD = - 0.011 - 0.296 log ClB (8419)
Table 8-8 was prepared from various sources for the time-to-denting at
11 brackish and seawater plants in an attempt to benchmark the two
correlations. The time-to-denting and chloride concentration from brackish/
seawater plants are plotted in Figure 8-34 along with the predicted times to
denting from the two groups. Reasonable agreement is indicated in the
range covered by the laboratory data.
TABLE 8-8
Denting Data From Operating Plants on Brackish Water and
Seawater
Avg. Time to Denting Avg. Blowdown

Plant Code EFPD EFPY Ref.
Millstone 2 M-2 649 1.8 480 8.47
Maine Yankee MY 1319 3.6 120 8.47
St. Lucie 1 SL-1 421 1.2 39 8.48
Calvert Cliffs 1 CC-1 1700 4.7 55 8.47
Calvert Cliffs 2 CC-2 1400 3.8 32 8.47
Ringhals 2 R-2 368 1.0 67 8.20
Indian Point 3 IP-3 578 1.6 51 8.48
Salem 1 S-1 428 1.2 173 8.48
KoRi 1 KR-1 204 0.5 84 8.48
W-2 720 2.0 59 8.46
W-3 600 1.6 94 8.46
Two important factors must be considered in trying to interpret field

data in this manner. First, in some plants denting is initiated before it is
detected since inspections usually occur only during annual refueling
outages, and then only a fraction of the tubes are inspected. The net result
is to reduce the indicated time to plant denting (lowering the points in
Figure 8-34). Second, most chloride analyses prior to 1980 were performed
using a procedure with a detection limit of about 50 ppb (8.49). Since many
plants reported data as <50 ppb, the method of accounting for such data,
when calculating the average blowdown chloride level, can have a major
0
effect on the predictions and plant experience.
As can be seen in Figure 8-34, the two correlations overlap in the 100 to
200 ppb Cl level. However, the correlation by Sawochka, et al. indicates a
much greater increase in the time to denting with decreasing blowdown
chloride concentrations. This is of particular importance relative to the use

of the EPRI PWR Secondary Water Chemistry Guidelines (8.50) and the
availability of chloride analytical equipment for analyses near 1 ppb
chloride. Specifically, the Guidelines allow one week of operation above
20 ppb or four hours above 100 ppb chloride prior to a power reduction to
30%. The predictions have been extrapolated to low concentrations,
although both groups noted that the predictive capability of the correlations
was highly suspect at concentrations below 100 ppb, since this was the limit
of the available database.
FIGURE 8-34
Predicted Versus Observed Time-to-Denting in Plants With Brackish
or Seawater Contamination
10.0
Sawochka, et al (8.23)
Pathania and
0 CC-2 o0 MY Balakrishnan
(8.20)
0
o 0 M-2
C
IP-3 0 0 W-3
C
a) 0 SL-1 S-1
0 R-2
E
O KR-1
0.1 I I I 1I
10 100 1000
Blowdown Chloride From Seawater, ppb
Pathania and Balakrishnan (8.20) developed a similar time to initiation

of denting based on the source of chloride being a neutral salt. The average
time for filling a radius crevice gap of (Yr) mils prior to the onset of denting
for NaC1 was as follows:
log tD = log Yr + 0.870 - 0.192 log ClB (8-19)
For the same chloride concentration, the times to denting in NaCI are
considerably higher than in seawater. For example, at 100 ppb chloride,
the time to denting in NaCl is 37 years compared to 1.5 years in seawater.
CONCLUSIONS
The laboratory data base documents the following conclusions:
* Denting is caused by the fast linear growth of nonprotective magnetite
within the tube/support crevice. A concentrated acid chloride
environment is formed in the crevice by the combination of local boiling
combined with either breakdown of a hydrolyzable chloride (e.g.,
MgC12 ) or the interaction of a neutral chloride (NaC1) and an oxidizing
species, such as Cu+ 2 or dissolved oxygen.
* Most denting in PWR steam generators can be related to development
of acid crevice solutions as a result of seawater or brackish water ingress.
* A different corrosion/chemistry relation exists for neutral and seawater
solutions compared to strong acid solutions as NiC12 and CuC12 .
Apparently, MgC12 solutions are closer to the seawater type.
* Denting is accelerated at increased temperature and at increased
concentrations of acid chlorides. In the laboratory, it is at a maximum
at an available crevice superheat of nominally 60-80'F (33-44°C).
* Studies with acidic sulfates have produced fairly high corrosion rates
but have not produced non-protective magnetite deposits nor denting.
" Chromium and silicon alloying additions to steel are beneficial in
reducing the corrosion rate in acid conditions, the former by promoting
duplex scale formation.
" Moderate corrosion of A405 (12% Cr) steel and denting of tubes and
mechanical restraints have been observed in model boiler and laboratory
tests. However, denting rates are much lower than for carbon steel,
and more concentrated chloride environments are required to give fast
linear magnetite growth.
* Attempts to predict the time to initiation of denting in PWR steam
generators from laboratory isothermal and heat transfer data have been
reasonably successful at plants where average impurity concentrations
have been high. The applicability of the correlations is limited by the
lack of experimental data at bulk water chloride concentrations
(<10 ppb) expected in commercial steam generators.
IMPROVED OPERATING AND MAINTENANCE

PRACTICES
Other chapters of this book and the Design and Operation Guidelines to
Minimize Steam GeneratorCorrosion discuss the various remedial measures
effective against denting corrosion. They are listed in Table 8-9 for reference.
TABLE 8-9
Improved Operating and Maintenance Practices
Chapter/
Guideline
Factor Remedy Reference
Reduce chloride " Condenser replacement Ch. 4 (1) (6)
levels entering the " Condenser maintenance Ch. 4 (1)
steam generator * Install condensate polishing Ch. 4 (2) (6)
" Optimize blowdown Ch. 4, 17,
system effectiveness 18 (3) (4)
* Improve polisher/ Ch. 17 (2)
makeup demin. performance (6)
Reduce acidity of " Eliminate copper- Ch. 4 (6)
corrodents bearing alloys
" Eliminate ingress of Ch. 10,15,17
chlorides (above) (5)
* Prevent air inleakage Ch. 18 (1)
(6) (5)
Reduce oxidizing " Eliminate sources of Ch. 18 (1)
potentials at crevice oxidants: Air ingress, (4) (5) (6)
aerated makeup water, (7) (8)
aerated auxiliary feedwater,
aeration during hot standby, etc.
" Optimize N2 H4 feed system Ch. 4 (5)
" Provide mechanical Ch. 4,17
deaeration (1) (4) (5) (6)
" Eliminate copper- Ch.4, 10 (6)
bearing alloys
Reduce inventory of " Sludge lancing Ch. 4, 18 (4)
corrodents in steam " Cyclic pressure flushing Ch. 4, 17 (4)
generator (sludge pile " High pH soaks/flushes Ch. 17 (4)
and crevice deposits) * Chemical cleaning Ch. 18 (9)
IMPROVED STEAM GENERATOR DESIGNS

In modern PWR steam generators, modifications have been made to
mitigate denting. The first modification addressed tube/support plate
crevice hydraulics. This involves using plates with holes of quatrefoil or
trefoil shape or supporting the tubes by a lattice bar structure. Experiments
have confirmed that this can reduce crevice superheat (8.15, 5,25). The
second modification is to change the support plate steel to a higher
chromium steel since such alloys are more corrosion resistant than carbon
steel. Examples are Types 405, 509, and 410S (11-13% Cr), and Type 347
stainless steels.
REFERENCES
8.1 Tatone, 0. S., and R. L. Tapping. "Steam Generator Tube
Performance: Experience With Water-Cooled Nuclear Power
Reactors During 1985." Atomic Energy of Canada, Ltd., (AECL
9724).
8.2 Morgan, E. P., F. W. Pement, J. N. Esposito, and R. G. Aspden.
"Examination of Denting and Characterization of Associated
Materials in the Plate-Tube Intersections of Westinghouse Nuclear
Steam Generators." Westinghouse Scientific Paper 76-7D2-SGEXM-
P1. 1976.
8.3 Bell, P. S., J. E. Forrest, D. F Libaert, and G. M. W. Mann. "PWR
Denting-The Examination of a Corroded Mild Steel Support
Plate." CEGB Note No. RD/L/N 87/80. June 1980.
8.4 Kurtz, R. J., et al. "Steam Generator Tube Integrity Program/Steam
Generator Group Project." U. S. Nuclear Regulatory Commission
(NUREG/CR-5117). May 1990.
8.5 Potter, E. C., and G. M. W. Mann. "The Fast Linear Growth of
Magnetite on Mild Steel in High Temperature Aqueous Conditions."
British CorrosionJournal 1 (1965): 26.
8.6 Kirsch, H. "Scale Formation on Different Types of Steel Under
Conditions of High Pressure, Hot Water/Steam Circulation in
Power Stations." Werkstoffe und Korrosion22, no. 6 (1971): 527.
8.7 Daniel, P. L., and S. L. Harper. "Use of Pourbaix Diagrams to Infer
Local Pitting Conditions." EPRI NP-4831. October 1986.
8.8 Beineke, T. A., et al. "Neutralization of Crevice Acids." EPRI
NP-3054. May 1983.
8.9 Schwenk, E. B. "Steam Generator Group Project, Task 10-
Secondary Side Examination." U.S. Nuclear Regulatory
Commission (NUREG/CR-4850). June 1987.
8.10 Green, S. J., and J. P. N. Paine. "Materials Performance in Nuclear

Pressurized Water Reactor Steam Generators." Nuclear Technology
55 (1981): 10.
8.11 Dow, B. L. "Steam Generator Progress Report Rev. 6." EPRI,
September 1990.
8.12 Broomfield, J. P., J. E. Forrest, D. R. Holmes, and M. I. Manning.
"Oxide Growth Mechanisms on Chromium Alloy Steels." EPRI
NP-4647. July 1986.
8.13 Mann, G. M. W. "The Oxidation of Iron-Base Alloys Containing
Less than 12% Cr in High Temperature Aqueous Solutions." NACE
Conference, Guildford, January 1973.
8.14 Steam Generator Reference Book. EPRI Steam Generator Owners
Group, May 1, 1985.
8.15 Huijbregts, W. M. M. "Determination of the Corrosion
Susceptibility of Evaporator Tubes." VGB Kraftwerkstechnik 61, no.
9 (September 1981): 681-689.
8.16 Sawochka, S. G., W. L. Pearl, and H. T. Pham. "PWR Secondary
System Chemistry During Air and Cooling Water Inleakage."
American Society of Mechanical Engineers, 76-WA/NE-14 (1976.)
8.17 Balakrishnan, P. V. "Effect of Condenser Water Inleakage on Steam
Generator Water Chemistry." International Water Conference 38
(November 1977): 131-145.
8.18 Alexander, J. H. and L. Luu. "MULTEQ: Equilibrium of an
Electrolytic Solution With Vapor-Liquid Partitioning and
Precipitation, Volume 1: User's Manual (Rev. 1)." EPRI NP-5561-
CCM. May 1989.
8.19 Proceedings of EPRI MULTEQ Users' Workshop, Clearwater Beach,
Florida, February 19-21, 1991.
8.20 Pathania, R. S. and P. V., Balakrishnan. "Correlation of Tube Support
Structure Studies." EPRI NP-4672. July 1986.
8.21 Economy, G. and C. R. Wolfe. "Causes of Denting, Volume 1:
Summary Report." EPRI NP-3275. May 1984.
8.22 Mann, G. M. W. and R. Castle. "The Effects of Oxygen, Copper,
and Acid Chlorides on Denting Corrosion." EPRI NP-4648. July
1986.
8.23 Sawochka, S. G., K. A. Friedman, S. S. Choi, and J. K. Pyo.
"Correlation of Tube Support Corrosion Studies." EPRI NP-4818.
October 1986.
8.24 Wootten, M. J., G. Economy, A. R. Pebler, and W. T. Lindsay Jr.
"Laboratory Investigations of the Denting Phenomenon in Nuclear
Steam Generators." MaterialsPerformance30 (December 1978).
8.25 Curlee, N. J., Jr. and A. J. Baum. "Single-Tube Thermal and

Hydraulic Tube Support Test," Vol. 1. EPRI NP-2046. September
1981.
8.26 Baum, A. J. and N. J. Curlee, Jr. "An Experimental and Analytical
Investigation of Dryout and Chemical Concentration in Confined
Geometries." ASME Century 2, Nuclear Engineering Conference,
San Francisco, 1980 (80-C2/NE-6).
8.27 Mann, G. M. W. and R. Castle. "Salt Concentration in Heated
Crevices and Simulated Scale." EPRI NP-3050. October 1983.
8.28 Bawden, R. J. "Rationale for Chemical Control of Feed and Boiler
Water, Task 4. The pH of PWR Steam Generator Water Under
Various Fault Conditions." CEGB Note No. RD/L/N 100/79, 1979.
8.29 Beineke, T. A., et al. "Tests of Isothermal Soaking Procedures for
Limiting Tube Denting in Nuclear Steam Generators." EPRI
NP-1761. April 1981
8.30 Pathania, R. S. and E. G. McVey. "The Effect of Cooling Water
Leakage on Corrosion of Steam Generator Materials." Corrosion/
79, NACE, Atlanta, Georgia, March 1979 (Paper 97).
8.31 Kunig, R. H., and G. Economy . "Causes of Denting, Vol. 4:
Isothermal Capsule Tests." EPRI NP-3275. December 1983.
8.32 Vaia, A. R., G. Economy, M. J. Wootten, and R. G. Aspden. "Denting
of Steam Generator Tubes in PWR Plants." Materials Performance
(February 1980): 9.
8.33 Economy, G., and D. Smith-Magowan. "Causes of Denting, Volume
2: Laboratory Test Results." EPRI NP-3275. May 1984
8.34 Proceedings: 1984 Workshop on Secondary-Side Stress Corrosion
Cracking and Intergranular Corrosion of PWR Steam Generator
Tubing. EPRI NP-4478. March 1986.
8.35 McKubre, M. C. H., and S. C. Leach. "Crevice Corrosion of Support
Alloys in the Secondary Environments of Nuclear Steam
Generators." EPRI NP-5017. March 1987.
8.36 Klisiewicz, J. W., et al. "Determination and Verification of Required
Water Chemistry Limits," Vol. 1. EPRI NP-3274. March 1984.
8.37 Von Neida, G. E., G. Economy, and M. J. Wootten. "Denting in
Nuclear Steam Generators-Laboratory Evaluation of Carbon Steel
Corrosion Under Heat Transfer Conditions." Corrosion/80, NACE,
Chicago, 1980 (Paper 175).
8.38 Castle, R. and G. M. W. Mann. "Denting Corrosion of PWR Steam
Generators. Heated Crevice Experiments With Feedwater
Containing Dissolved Oxygen." CEGB Note No. RD/L/N88/80,
1981.
8.39 Brown, J., F. Gonzalez, G. Itey, and A. McKay. "Studies of Denting

in Steam Generators With Simulated Fresh Water Inleakage."
International Conference on Materials Performance in Nuclear
Steam Generators, St. Petersburg, Florida, October 1980.
8.40 Dautovich, D. P., J. Brown, D. Iley, and A. McKay. "Corrosion
Studies of Nuclear Steam Generators Under Fault Water
Conditions." Corrosion/80, NACE, Chicago, 1980 (Paper 182).
8.41 Esposito, J. N., et al. "Laboratory Investigation of Corrosion
Occurring Under Heat Transfer Conditions." Water Chemistry II,
BNES, Bournemouth, 1980 (Paper 10).
8.42 Wolfe, C. R., J. N. Esposito, S. M. Wozniak, and D. D. Whyte.
"Neutralization of Steam Generator Denting," Vol. 1. EPRI NP-3023.
September 1983.
8.43 White, E. L., and W. E. Berry. "Effects of Copper and Nickel
Compounds on the Corrosion in PWR Steam Generator Materials."
International Conference on Materials Performance in Nuclear
Steam Generators, St. Petersburg, Florida, October 1980.
8.44 Ashford, J. H., R. Garnsey, and G. M. W. Mann. "Corrosion of Mild
Steel Under Heat Transfer in High Temperature Aerated Sodium
Chloride Solutions." Corrosion Science 14 (1974): 515.
8.45 Mann, G. M. W. and R. Castle. "Hideout and Return of Chloride
Salts in Heated Crevices Prototypic of Support Plates in Steam
Generators." EPRI NP-5015. January 1987.
8.46 DeSilva, S. G., and C. R. Wolfe. "Causes of Denting, Vol. 3: Plant
Chemistry Correlations." EPRI NP-3275. May 1984.
8.47 C-E Plant Data Supplied by K. Craig through EPRI cited in
Reference 8.18.
8.48 Pearl, W. L., S. G. Sawochka, and S. S. Choi. "Evaluation of Field
Applications of Boric Acid in PWR Steam Generators." EPRI
NP-3278. March 1984.
8.49 "PWR Secondary Water Chemistry Guidelines." EPRI NP-2704-
SR. October 1982.
8.50 "PWR Secondary Water Chemistry Guidelines, Rev. 2." EPRI
NP-6239. December 1988.
9
TUBE WASTAGE AND

PHOSPHATE SECONDARY
WATER CHEMISTRY
W. T Lindsay,Consultant
J. P.N. Paine, EPRI
Y. Solomon, EPRI (Deceased)
1993
INTRODUCTION
Wastage is a form of generalized corrosion in which metal is more or
less uniformly removed over an area of macroscopic dimensions. Tube
wastage (frequently called "thinning") was first observed in nuclear steam
generators when these units were being treated with sodium phosphate.
Corrosion was on the outside of the tubes in regions where sodium
phosphate could be concentrated by locally operating evaporative processes.
Typical locations corresponded to the position of semi-permanent steam/
water interfaces, such as just below the top of a sludge pile, or near the
boundaries of partly restricted, high-quality zones defined by mechanical
obstructions, e.g., upper bundle supports.
In contrast to many other forms of corrosion on alloy 600, wastage by
concentrated phosphate solutions is transgranular, i.e., there is no selective
reaction at grain boundaries. Continuing wastage can thin the tube wall
substantially. Eventually, a small area can become so thin that it can no
longer support the pressure difference between primary and secondary
systems, at which time a pinhole leak will occur as a result of ductile rupture.
Relatively few tubes have leaked as a result of wastage due to
phosphates, although many thousands of tubes have experienced this form
of corrosion. Eddy current inspection is sensitive in detecting large areas
of metal removal; consequently, wastage could be identified readily and its
progression followed with confidence. Extensive tube involvement
combined with progression, that was only slowed but not arrested by
refinements in phosphate treatment, led the domestic industry to convert
almost unanimously to all-volatile treatment (AVT) chemistry in 1974-1975.
The phenomenon was observed on the prevalent tubing in use at the time,
alloy 600. However, other austenitic alloys, both nickel and iron based, are
subject to phosphate wastage. Wastage continues to be a concern for steam
generators utilizing phosphate treatment. Both alloys 600 and 800 are
utilized in these generators and both alloys continue to experience a slow
corrosion progression.
Concentrated phosphate solutions are not the only media that can cause
wastage of alloy 600 tubing. A number of acid solutions can produce the
same general result, most notably acid sulfates (9.1). Laboratory tests have
shown that sulfates and sulfuric acid are among the products of hydrolytic
decomposition of cation exchange resins, suggesting that resin ingress could
be another cause of this form of corrosion. While not confirmed, it is
suspected that acid sulfates may be involved in a number of cases of wastage
that have been detected in recent years on peripheral tubes, mainly at the
lower support plates on the cold leg (Chapter 16). Shallow wastage has
also been observed in the laboratory on tubing in contact with concentrated
organic acid (%2) and caustic (9.3).
Tube Wastage and Phosphate Secondary Water Chemistry 9-3
PWR PLANT EXPERIENCE

From 1972 to 1976 wastage affected 17 of 22 Westinghouse or
Combustion Engineering designed phosphate-water-controlled units tubed
with alloy 600. Table 9-1 lists the affected units (9.4). Affected areas were
typically found at the top of sludge piles where wet and dry zones could
concentrate corrosive chemicals. Some units have had significant wastage
at other elevations in the steam generator bundle: underneath support
plates, in U-bends, and underneath heavy tube scales in the free span.
Neither the low corrosion rate at units such as San Onofre and Jos6 Cabrera,
which still use phosphate water chemistry control, nor the high rate of attack
which occurred 20 years ago at Palisades, Mihama 1, and others has been
satisfactorily explained. Point Lepreau, a CANDU power plant, is also
currently operating "successfully" with phosphate chemistry control.
As of 1987, five alloy 800 units in Germany (manufactured by Kraftwerk
Union) continued with phosphate chemistry (9.5). The German units'
operational history dates from 1972; however, wastage did not occur until
1979. The KWU Unit experience indicated that wastage was not seen until
sludge loading in the steam generators built up to high enough levels to
allow concentration of chemicals in wet and dry zones in the top layers of
the sludge. Table 9-2 lists the relevant experience (9., 9.6).
The KWU phosphate chemistry specifications were much different from
the specifications for 'the domestic units that experienced wastage with
alloy 600. (Jos6 Cabrera and Point Lepreau which have alloy 600 tubes have
continued to use the KWU treatment practice with notable results.) A
comparison (.9,5 9.7) of the Kraftwerk Union and Westinghouse water
chemistry specification for comparable periods of time and for comparable
power plants is shown in Table 9-3. In an attempt to slow or arrest wastage,
Westinghouse modified their specifications over time, tightening the
allowable phosphate control band. Significant differences exist in the
phosphate residual in the generator and in the allowable condenser
inleakage.
Nature of Phosphate Wastage

Figure 9-1 shows macrophotographs of the exterior appearance of a
tube removed from a steam generator which experienced phosphate
wastage. The corrosion is in the triangularly shaped, roughened appearing
area, bounded on the bottom by a line corresponding to the elevation of the
top of the tubesheet. This area appeared relatively bright when the tube
was examined in the laboratory, but it was probably covered initially by a
green layer of mixed nickel-iron-chromium phosphates, which was typically
observed in other situations where the corrosion products were relatively

undisturbed. Wall thinning is evident from the single .wall radiographs
shown in Figure 9-2, in which a similar pattern can be perceived. The
maximum depth of wastage on this tube was about 4.3 mils (108 jtm),
approximately 8% of the reference wall thickness.
In its initial stages, phosphate wastage begins as an assemblage of
shallow pits, which gradually broaden and merge to form a semi-continuous
area of attack. Figure 9-3 is an SEM micrograph of the tube surface after
cathodic descaling. This shows the transition from attacked to unattacked
metal, where the pattern of overlapping pits is clearly visible (9.8).
Field Experience
Phosphate wastage was first observed in domestic commercial PWR
plants when phosphate treatment was changed to a low sodium-to-
phosphate molar ratio control, in which the molar ratio of Na/PO4 was
maintained at about 2.0. This change was in response to a series of caustic
stress corrosion cracking events that were attributed to operation with
uncontrolled high sodium-to-phosphate ratios (above 2.8), from which free
caustic could result. The incidence of caustic stress corrosion cracking
dropped markedly, but the general corrosion we now know as phosphate
wastage began to be observed within approximately a year after the change.
Weeks (%.9) has given a general account of some of the observations. Baschek
and Sandona (9.10) provided a detailed report of the events at Beznau I
and Beznau 2.
The location of phosphate wastage in PWR steam generators of the
Westinghouse design was, for the most part, on hot leg tubes at the elevations
where the tubes entered sludge piles accumulated on top of the tubesheet.
Where the sludge pile was deep, the zone of wastage extended into the pile
about an inch or so. Where the pile was less deep, the wastage extended
down to the tubesheet surface, as in the case of the tube shown in Figure 9-1;
but, in most cases, it did not penetrate to any great extent into the tube-to-
tubesheet crevice. Although eddy current inspections did not reveal wastage
corrosion in U.S. plants at the locations of tube-to-support plate intersections
(in drilled-hole, carbon steel support plates), subsequent inspections of tubes
removed from operating generators did show some very slight surface
corrosion at this location as well, along with the presence of green phosphate
deposits.
The German experience with wastage was somewhat similar. Figure 9-4
schematically shows wastage on a tube removed from a Borssele steam
generator in 1979 (9.5).
TABLE 9-1
Steam Generators Affected by Phosphate Wastage
No. Tubes
Months No. Tubes Plugged Thot
Plant With PO0(1) Plugged (Month) (OF)
Palisades 32(2) 3,738(3) 117 599
Mihama 1 35 1,329(3) 38 603
H. B. Robinson 13 yrs. 571 4 604
Surry 1 24 468 20 590
Mihama 2 28 266 10 607
Beznau 2 32 262 8 597
Point Beach 2 28 207 7 611
Surry 2 16 190 12 606
Point Beach 1 40 176 4 611
Turkey Point 3 22 112 5 605
Takahama 1 6 98 16 613
Turkey Point 4 13 71 5 602
Ginna 50 54 1 610
Connecticut Yankee 75 34 0.5 574
San Onofre 1 19 yrs. 29 0.1 597
(PO 4 still in use) lowered
Jose Cabrera (Zorita)19 yrs. 17 0.1 596
(PO 4 still in use) lowered
Beznau 1 29 9 0.3 600
Plants With Some Coordinated Phosphate Exposure Not Plugging

Any Tubes for Phosphate-Induced Wastage
Zion 1 17
Indian Point 2 16
Kewaunee 6
Zion 2 4
Prairie Island 1 3
1
Estimated calendar months when plant operated with sodium-phosphate-
conditioned secondary water.
2
Secondary water treatment: sodium phosphate with sodium sulfite.
3
Mostly batwing area corrosion.
TABLE 9-2
Phosphate Experience at German Units
No. Tubes No. Tubes

Initial Plugged Plugged Thot
Plant Criticality (.1979-1983) (1984-1989) (OF)
KKS 01/72 216 93 597
KCB o03/73 91 9 607
KWB-A 07/74 14 444(1) 606
KWB-B 03/76 4 2 607
GKN 05/76 5 3 610
KKU 09/78 0 0 613
KKG 01/79 1 N/A 623
1Mostly batwing corrosion.
TABLE 9-3
Sodium Phosphate Water Chemistry Specifications
Kraftwerk Union
2
SG Bulk Water Practice 1972-1980(1) Westinghouse Practice( )
Blowdown Sec. Actual 1976-71 1972-73 1974
pH Value 8.8-9.5 8.5-10.6
Conductivity, <gm <50
Phosphate, ppm 2-6 2-3 .10-80
Na/P04 Molar Ratio <2.6 1.8-2.0 <2.6 2.0-2.6 2.3-2.6
Chloride <1 ppm avg. <30 ppb <75 ppm
Silica <4 <5
1 Reference 9.5.
2
Reference: Westinghouse Steam Generator Symposium, April 1973.
FIGURE 9-1
Phosphate Attack on Alloy 600 Tube Surface (2.5x)
Flow
Tubesheet
Face
00 900
Fill
-KTubesheet
Face
2700
1800
FIGURE 9-2
Single Wall Radiographs on Tube Shown in Figure 9-1
(Courtesy of Westinghouse Electric Corporation)
-. 315* 00 1. 45*
Flow
'%,1We lap 4. m
In steam generators of the Combustion Engineering design, wastage

was first noted and was most extensive in the vicinity of antivibration bars,
0
which were relatively wide and oriented in such a way as to define a
persisting region with a high proportion of steam blanketing. Corrosion
was concentrated at the boundaries of this region, where a liquid/vapor
interface was presumably fluctuating over a short length of tubing.
Among the plants with early indications of phosphate wastage were

Beznau 1 in Switzerland, Connecticut Yankee and Palisades in the
United States, and Mihama 1 in Japan. However, it soon became clear that
other PWR nuclear steam generators on phosphate water treatment with
low sodium-to-phosphate ratios were being affected. The rate of corrosion
and the extent of tube involvement varied widely from plant to plant,
depending on many factors, such as the amount and distribution of sludge
accumulation, the details of treatment chemistry and its control, operating
history, etc. Corrosion depth on tubes, removed from Beznau 1 in 1972
after 6,830 operating hours with low ratio phosphate, was 0.001 to 0.003
inches (25-75 Itm). Approximately 0.008 inches (0.2 mm) of attack was found
in 1973 after 14,368 operating hours (%3). This corresponds to about 0.001
to 0.004 inches/yr (33 to 97 Vtm/yr) for the earlier period, and up to
0.005 inches/yr (124 rtm/yr) for the later period.
FIGURE 9-3
Scanning Electron Micrograph Showing Tube Wastage and
Intergranular Attack in Descaled Alloy 600 Tubing
FIGURE 9-4
Wastage on a Tube Removed From a Borsele Steam Generator
in 1979 (9.5)
E
E
2. 2/3 of circumference
oz~
0)
Affected Area
0*j
Sludge Pile Area
C0
Top of Tube Sheet
Top of Tube Sheet
Wall Thickness (pm)

1200-
1000-
800-
42% Wall Thinning
600-
400-
200-
3.0 4,0 5,Omm Above Tube Sheet

I3I 4I 59m AoeTb he
As recognition grew that the new phenomenon was widespread and

of a generic nature, major laboratory investigations were launched by the
two U.S. vendors of recirculating-type steam generators, and KWU. It was
demonstrated in pot boilers how the location of attack was related to
concentrating processes at steam/water interfaces. Emphasis was placed
on investigating the chemistry of low-ratio phosphate solutions in high-
temperature water, in addition to corrosion studies and model boiler tests.
The laboratory work showed that corrosivity of concentrated sodium
phosphate solutions was related to both concentration and sodium-to-
phosphate molar ratio in a complicated way, as described briefly in the
following section.
The laboratory results led to recommendations for control in domestic
units in the relatively narrow band of molar ratios between 2.3 and 2.6.
The lower limit was related to the rapidly increasing rate of wastage at
lower ratios, and the higher limit was selected to avoid free caustic and
concomitant caustic stress corrosion cracking. The KWU response was
somewhat different, emphasizing instead minimum concentration of P0 4,
and minimization of sludge building. A number of domestic plant operators
were successful in controlling their steam generator chemistries within this
restricted range of compositions; eddy current inspection data indicated
that the rate of attack, at best, was slowed. It appeared that attack would
not cease unless the operators constantly operated within a narrow control
band at the high end of the allowed range. Furthermore, a constantly
increasing sludge burden, augmented by precipitated phosphate
compounds, made such control very difficult. The KWU response has
enjoyed more favorable results. Wastage was slowed to manageable
proportions as operators removed sludge piles by lancing and chemical
cleaning and reduced their P0 4 residuals to 0.5-2 ppm at Na/PG4 ratio of
2.0-2.2.
The above difficulties led the vendors of the affected domestic plants
to recommend that the operators adopt an all volatile water treatment (AVT)
based on the use of ammonia and hydrazine. This recommendation was
almost universally followed in the United States with two exceptions.
Extensive dephosphatizing procedures were also recommended to facilitate
the changeover to AVT, but these were only partially implemented by most
operators. Steam generators manufactured by Kraftwerk Union continue
to use phosphate water treatment. These steam generators, tubed with
alloy 800 tubes, have experienced a slower rate of wastage corrosion. It is
not clear whether the slower rate is a function of the alloy, the very low
concentration of phosphate used, or the relatively good inleakage history
at several of these units. It is probable that all these factors are involved.
LABORATORY STUDIES
Sodium Phosphate Solution Chemistry

Although sodium phosphates had been used for internal treatment of
drum boilers on fossil-fueled plants for many decades, detailed knowledge
of the high-temperature chemistry of these solutions at sodium-to-
phosphate molar ratios below about 2.7 was not developed, until prompted
by the phosphate wastage phenomenon, in nuclear steam generators. The
classic study by Ravich and Shcherbakova (9.11) had shown the existence
of a congruent point at Na/PO4 = 2.8 at 572'F (300'C). This was the basis
for "congruent point control" (9.12), which aimed at avoiding solution
compositions above a ratio of 2.8, because precipitation of sodium-deficient
solids from such solutions produces free caustic with associated possibilities
for severe corrosion and stress corrosion cracking.
Since waistage attack on alloy 600 began after a change in control strategy
to much lower Na/PG4 ratios, chemistry investigations focused on the
behavior of the sodium-phosphate-water system at Na/PG4 ratios around
2. An extensive series of solubility measurements reported by Panson et al.
(9.13) and Economy et al. (9.14) is summarized by Figures 9-5 and 9-6.
On Figure 9-5, the equilibrium composition of the dilute aqueous
solution is plotted against the composition of precipitates or concentrated
phases, which were inferred from differences between starting and
equilibrium concentrations in the solution. The horizontal segment indicates
an invariant solution composition at Na/PO4 = 2.15, which was interpreted
to mean that two solid phases (Solid Solution I and Solid Solution II) were
present, in addition to the liquid and vapor phases. Actually, the phase
indicated as Solution II (Figure 9-5) was later discovered by Broadbent et
al. (9.15) to be a very concentrated liquid solution (see below), and 2.15
turned out to be the upper limit of the Na/PG 4 molar ratio at 572°F (300'C)
for a region of liquid-liquid immiscibility. Nevertheless, the horizontal
segment does have the characteristics of an invariant composition, at which
a liquid solution would dry up on removal of water if both concentrated
phases were solids.
Figure 9-5 shows that precipitation of Solid Solution I occurs as water
is removed from saturated solution compositions below the congruent ratio
of 2.8, shifting the composition of the remaining solution toward lower Na/
PG 4 ratios until the, invariant solution composition is reached. The liquid
solution maintains this composition as it disappears on evaporation and is
replaced by some combination of the two more concentrated phases, one
being Solid Solution I and the other being the very concentrated liquid
corresponding to Solution II. The relative proportions of the two
concentrated phases depend on the starting composition of the dilute

solution, which is evaporated toward dryness. Starting compositions close
to Na/PG4 = 2.15 produce a large proportion of Solution II, while starting
compositions far removed from this produce mostly Solid Solution I.
FIGURE 9-5
Sodium Phosphate Water System at 5720 F (3000 C) Composition of
Liquid Versus Composition of Concentrated Phases
3.8
---
nmlcI 0IIPure IPhosphate
Panson and I I I I
3.6 -aEconomy A With 0.3 and 0.6 M NaCI
0 With 0.1 MNa2SO4
3.4 - Ravich - 0 Pure Phosphate
3.2
CO /
/,
z 3.0 / l
-- 0 /
(D /
2.8
__ / t
/0
0~
2.6 /
C:
2.4
0 Congruent
Line
2.2 //A/
C,) 2.0
__Z/
/ S A
1.8 -- / Concentrated Phases -
Solid
1.6 - Solution II I + II Solution I -
/
.I " I I I I I I I
1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
Molar Ratio in Concentrated Phases (MNa/MPo 4 )
Figure 9-5 also shows that starting compositions below Na/PO4 = 2.15
cause a further decrease in the dilute solution molar ratio as water is
removed and only concentrated Solution II forms. Figure 9-6 shows that
as the molar ratio of the concentrated phases decreases to very low values,
the equilibrium concentration of the dilute solution increases rapidly. The
practical significance of these effects was taken to mean that sodium

phosphate compositions with Na/PO4 = 2.15 or lower should be avoided
in steam generators. Concentration of such a solution would lead to a quite
high concentration, possibly corresponding to the thermodynamic limit for
superheat locally present, with a still lower Na/PG4 ratio, which was
expected to be more acid, or at least less alkaline, and highly corrosive.
This interpretation was basically correct, although the details of the phase
diagram from which it was derived were partly mistaken. The selection of
Na/PG4 = 2.3 at the lower end of the allowed composition range was based
on this interpretation, with allowance for margin.
FIGURE 9-6
Sodium Phosphate Water System at 572 0 F (3000 C) Phosphate
Solubility As a Function of Composition of Concentrated Phases
Cd)
CL,
0~
1.6 1.8 2.0 2:2 2.4 2.6 2.8 3.0 3.2

Molar Ratio in Concentrated Phases (MNa/Mpo4)
Subsequent investigation of the sodium-phosphate-water system by

Balakrishnan (9.16), with a recirculating experimental loop containing a
boiling test section, gave results in agreement with Figure 9-4, but it was
the detailed study by Broadbent et al. (9.15) that showed the existence of
liquid-liquid immiscibility and revealed much of the complexity of the
system at temperatures of 482 to 662 0F (250 to 350 0 C). Taylor et al. (9.17)
have contributed a study of the equilibrium solid phases that adds further
detail.
Marshall (9.18) carried out additional solubility measurements that
further elucidate the system, and he summarized his results with those of
others that pertain to liquid-liquid immiscibility. Figure 9-7 (9.18) shows
the region of two liquid phases as the area within the two rising legs of the
U-shaped curves at constant Na/P04 molar ratio. For example, at 572°F
(300'C), the dashed line drawn between the two legs of the Na/PO4 = 2.0
curve indicates that two liquid solutions, one with about 12 wt% dissolved
solid and the other with about 59 wt.% dissolved solid, are present within
the miscibility gap. No liquid-liquid immiscibility exists in the region below
the curve labeled for Na/PO4 = 2.16; that is, higher ratios do not allow the
presence of the more concentrated liquid phase.
The significance of the concentrated liquid phase for phosphate wastage
in steam generators is that it can be formed as soon as the more dilute
solution reaches the concentration at the lower concentration limit of the
immiscibility range. This may be a solution with a boiling point elevation
that is relatively small, which means that the more concentrated phase can
be formed (i.e., precipitated) in locations that have less than the full
superheat that is potentially available. If then, as has been suggested (9.19),
it is the more concentrated liquid of the two-liquid system that is responsible
for attack on alloy 600, a number of puzzling results from both corrosion
tests (see below) and field experience may be explained. For one, the slowing
of progress of wastage in operating units when Na/PG4 ratios were
controlled in the range between 2.3 and 2.6, but not its cessation, is
understandable, since starting compositions in this range will always
produce some of the corrosive phase when they are concentrated by
evaporation of water. (See the above discussion of Figures 9-5 and 9-6 and
the invariant composition.) However, this does not explain the dramatic
decrease in corrosion rates at KWU-built units which have continued to
operate at Na/PO4 ratios of 2.0-2.2 and Jose Cabrera who decreased the
ratio from 2.5 (1973-75) to 2.1 (1975-82), but decreased P0 4 concentration
from 4-9 ppm to <3 ppm.
FIGURE 9-7
Two Liquid Phases and Solution Solid Boundaries for Aqueous
Solution Mixtures of Sodium Phosphate Salts of Molar Ratios,
Na/PO 4 , from 1.0 to 3.0 at 392°F (2000 C) to 752 0 F (400 0C)
400
375
350
325
600
300o
E
550
275
250
225
ni 3.0 " "-.

J 200
30 40 50 80
Dissolved Solid (wt. %)
Potassium Phosphate Solution Chemistry

Marshall and coworkers at Oak Ridge national laboratory have found
that a region of liquid-liquid immiscibility exists in the potassium phosphate
system, as well as in the sodium phosphate system, but only at temperatures
above 680'F (360'C) (92,18 9.24, 9.25). Consequently, the potassium
phosphate system cannot form a concentrated second liquid phase in PWR
steam generators. Since the concentrated second liquid phase of sodium
phosphate has beeh suspected as a factor in accelerated attack on steam
generator tubing (see the previous discussion under Sodium Phosphate

Solution Chemistry), Marshall has suggested that potassium phosphate
treatment should be considered an alternative to AVT, conveying the benefits
of phosphate treatment without the detrimental effects of sodium phosphate.
However, Marshall et al. (9.24) also found that solubility in the
potassium phosphate system is much higher than in the sodium phosphate
system: for example, the solubility of K2 HPO4 at 482-662°F (250-350'C) is
77 to 78 wt%. A very concentrated liquid solution, with a boiling point
equal to the highest possible superheat, would be expected to eventually
fill steam generator crevices. Although Marshall's suggestion can be fully
evaluated only by a series of corrosion tests in concentrated potassium
phosphate solutions at steam generator crevice conditions, it is likely that
steam generator tubing would suffer attack, perhaps severe, if potassium
phosphate treatment were employed. This possibility is consistent with
the experience of the power industry several decades ago, when potassium
phosphate treatment was tried briefly in fossil-fueled drum boilers. Severe
corrosion was experienced, and the trials were quickly discontinued.
Potassium phosphate treatment should not be employed in PWR steam
generators without extensive prior corrosion testing and evaluation in model
boilers. The prospect for its successful use is doubtful.
Sodium Phosphate Corrosion Tests

Numerous corrosion tests were performed in pot boilers, model boilers,
and capsules by NSSS vendors. The information developed by this work
was initially made available only to the operators of PWR plants with
affected steam generators. The results of an extensive program of
immersion-type corrosion tests by Pessall et al. Were subsequently published
(9.20). Pessall's work studied the corrosion resistance of alloy 600 in sodium
phosphate solutions of a variety of concentrations and compositions at
temperatures between 525 and 617°F (275 and 325°C). Specimens were
exposed under fully and partly immersed conditions, as well as in the vapor
above the solutions. Some tests were run with initial concentrations
approximately at the solubility limit that corresponded to the chosen
temperature and composition. Others were run with an initial excess of
phosphate, with the expectation that retrograde solubility would reduce
the concentration to the high temperature solubility limit as the test vessel
was heated up and solids precipitated. However, the complexities of
phosphate chemistry are such that it was not always possible to be certain
about the composition and concentration of the solution that would result.
Although hot samples for solution analysis were taken periodically from
some of the tests, all corrosion results were reported on the basis of initial
conditions. Exposure times up to 4,700 hours were evaluated.
Figure 9-8 shows a summary of results at 572°F (300'C) for fully

immersed specimens. A regression analysis showed that the best fit to the
data for all phosphate concentration and Na/PO4 ratios was a parabolic
expression. The solid lines represent the fit of the parabolic rate law to
each of the tests. These show the penetrations predicted after one year,
which range from a low of 0.02 mils (0.5 gtm) for the composite fit to the
data at the lowest concentrations and/or highest Na/PG4 ratios, to a high
of 1.6 mils (47 pm) for the test with initial 6 molal phosphate at Na/
P0 4 = 1.6.
The parabolic rate law is of the following form:
D = A t1/ 2 (9-1)
In the above equation, time (t) is in years and A is the parabolic rate
constant, which depends on the test conditions.
The dependence of the rate constant on phosphate concentration was
found to be linear, at least for values of Na/PG4 between 2.0 and 3.0. This
was the case also for 527 and 617'F (275 and 325°C). The linear relation for
572°F (300'C) is given by the following:
A = 0.02 + 0.26 m (9-2)
in which m is phosphate molality, and A is the parabolic rate constant
of Eq. 9-1. The constant 0.02 is consistent with published data for the
corrosion rate of alloy 600 in pure water.
Figure 9-9 shows the dependence of corrosion rate on Na/PG4 molar
ratio for data obtained at 617'F (325°C), also with fully immersed specimens.
A maximum appears at Na/PO4 = 1.6 for both parabolically and linearly
predicted penetration at one year. Similar trends were obtained at 572 and
527°F (300 and 275'C), but the 617'F (325°C) data extend to Na/PO4 as low
as 1.0, thus more clearly defining the maximum. This complex dependency
on Na/PG4 below 2.0 was unexpected. Pessall et al. noted that species
distribution in sodium phosphate solutions shows a maximum at about
1.6 Na/PG4 for the monohydrogen phosphate ion, HPG 4 2 -, and they
speculated that this particular ion may have been the aggressive specie
responsible for the maximum in corrosion rate at the same value of Na/
PG 4 . However, there is no direct evidence, nor is there any rationale, to
support such a speculation. It now seems more likely, in view of what we
know about the existence of liquid-liquid immiscibility in the system, that
the test conditions employed at Na/PG4 = 1.6 were more conducive to
formation of the concentrated liquid phase than were the other conditions,
and that it was this phase that was the most aggressive medium.
FIGURE 9-8
Average Corrosion Penetration Versus Time Relationships for Alloy 600 in Deaerated Sodium Phosphate
Solutions at 572 0 F (3000 C)
.24- i
.2 - . 80 r Molality ppm P0 4
.22 -r= 1.6, 6m 1.6 6.0 316,800

.20 A0 2.0 1.2 97,400 -
.40 U 2.0 0.15 14,000
.18 0 2.3 0.4 35,900 -

0 2.3 5.0 272,500
.16 - 600 1000 1400 2
V 2.3 0.15 27,900
.14 r 2.0, 1.2m 0 0.15 13,900 --
E A 2.6 0.4 71,600 0
12 -co @
0 3.0 0.15 13,900
.2 3.0 0.4 35,700
.06 --.
08•
Imis0 o ~
r =2.3, 0.4m
_
-
0~
.02 - ' • 0.02 mils (1 y r) (D
.02~ ~
= 3.0,0.15m, u. m /-
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500
Exposure Time (Hours),.

FIGURE 9-9
Average Corrosion Penetrations of Alloy 600 in Deaerated Sodium
Phosphate Solutions at 615°F (325 0 C) As a Function of the Na/PO 4
Molar Ratio
Initial Time of Exposure

r Molality Hours
5 1.0 6 387
1.35 6 312
1.6 6 480
2.0 5.2 504
4) 2.3 5 451
4 - 2.7 5 432
t 3
Linear Rate
a)
2 --
M
Parabolic
Rate
I I
1.0 1.35 1.6 2.0 2.3 2.7 3.0
Na/P0 4
It has sometimes been speculated that sodium pyrophosphate is

responsible for accelerated corrosion in low ratio Na/PO4 solutions, but
Pessall et al. showed by Raman spectroscopy that no pyrophosphate ions
were present in the solutions, and that the conversion of dissolved
pyrophosphate species to orthophosphate species is very rapid in high-
temperature water.
In Figure 9-10, one-year corrosion penetrations, predicted by the
parabolic rate law, are plotted against temperature for two values of Na/
P0 4 molar ratio. Many of the test conditions involved solutions at the
solubility limit for some of the solids in the Na/PG 4 water system, in which
retrograde solubility-temperature relations are the norm. Consequently,
the effects of temperature on concentration (and, indirectly, on corrosion)
oppose the direct effects of temperature on corrosion rates, and maxima
appear in the combined result.
FIGURE 9-10
Temperature Dependence of the Corrosion Rate of Alloy 600 Tubing
Exposed to Deaerated Phosphate Solutions at or Near Saturation
Concenctration
(0C)
inn 275 300 325
8.0
6.0
4.0+
(r - 1.6)
2.0 6m (r -1.6)
1.0
.8
.6
.4 (r -2.13)
CU (r-2.13) 0.21m
0.6m - - -
0
.2
.1 -(r- 2.13) /
.08 - 1.8m,/
.06
Na/P04 Ratio =r
.04 ! Phosphate Concentration
in Molality =m
.02
I I I I I I
.01
525 550 575 600 625 650
Temperature (°F)
An interesting result of this work was the finding that.partly immersed

specimens showed higher corrosion rates, based on weight loss and
immersed area, than did fully immersed specimens. Attack was especially
pronounced at and near the immersion line. No explanation could be offered
at the time, but it now seems possible that this may also be related to the
formation of the concentrated liquid phase near the interface, where some
degree of wetting and drying could have been taking place.
Pessall et al. (9.20) also studied the corrosion behavior of alloy 690,
alloy 800, AISI Type 304 stainless steel, and Croloy (2.25 Cr-I Mo) under
similar conditions, but less comprehensively. All of these showed high
corrosion rates. Alloy 690 and alloy 800 were generally comparable to alloy
600, while 304 and Croloy were less corrosion resistant.
In other published work, Bignold (9.21) concluded from short term

exposure tests, supplemented by polarization resistance tests and
polarization experiments, that alloy 600 can undergo rapid pitting corrosion
in highly concentrated sodium hydrogen phosphate solutions at Na/P0 4 =
2.3, while corrosion is less localized but also rapid under more acid
conditions. Broadbent et al. (9.22) reported on a study of the corrosion of
mild steel in concentrated sodium phosphate solutions at Na/P0 4 from 1.0
to 3.0 and temperatures from 482 to 662°F (250 to 350'C). Corrosion was
rapid, but less than in sodium hydroxide of comparable concentrations. A
complicated pattern was observed for the dependence of corrosion rates
on concentration and composition of the solution, which was not readily
explicable in terms of solubility and phase relations. They consider that
the solubility of iron in the phosphate solutions is an important rate-
determining factor in the corrosion of mild steel.
Interactions of Phosphate With Magnetite

Although possibly not a causative factor in tube wastage, the interaction
of sodium phosphate solutions with magnetite and steam generator sludge
is intimately bound up with the problem of control of the chemistry of steam
generators being treated with phosphates. It had been known for a long
time that sodium phosphate solutions react with magnetite and other solids
that are found in boilers, but these reactions had not been extensively
characterized at the time phosphate treatment was widely used in nuclear
steam generators. Consequently, chemical investigations were directed
toward that end.
Economy et al. (9.14) reported on a study of the effects of phosphate
concentration, phosphate composition, and temperature on these reactions.
They used a flowing apparatus, with the phosphate solutions passed
through a column of magnetite held at steam generator temperatures. The
effluent solution was monitored to determine the uptake or release of
phosphate as the reactions within the column proceeded. A substantial
degree of reaction was observed. The maximum uptake, in a special test
designed to determine the saturation limit, was 0.2 oz (0.7 grams) of
phosphate per gram of magnetite. The reaction product contained a sodium
iron phosphate compound, with an unindexed X-ray diffraction pattern, in
which the molar ratio Na/P0 4 was 2.13 and the molar ratio of phosphate
to iron was 0.6.
The ratio of sodium to phosphate in the solid reaction product was
always less than the ratio in the influent solution, since the effluent solution
was depleted more in phosphate than in sodium. Hence, the reaction of
sodium phosphate solutions with iron corrosion products can raise the Na/
P0 4 ratio in steam generator water, thereby confounding composition
control efforts and posing a risk of entering the free caustic region.
This work verified that phosphate hideout is partly chemical, as a result

of chemical reactions, as well as partly physical, as a result of limited
solubilities in the sodium-phosphate-water system. Balakrishnan (9.16)
found depletion of phosphate, in his recirculating loop with boiling test
section, to be partly due to reactions with corrosion product deposits on
the surfaces of the loop, and he observed a rise in the Na/P0 4 ratio as
deposition proceeded, even for solutions that were initially below the
congruent molar ratio of 2.8.
An interesting discovery by Economy et al. was that the reaction could
be nearly completely reversed by passing a sufficient amount of either pure
water or lower concentration solutions through their reaction column.
However, the recovered solubles were generally richer in sodium than
phosphate, relative to the decomposing sodium iron phosphate compound.
This means that repeated cycles of uptake and release (hideout and return
by chemical reaction processes) can lead to a gradual accumulation of more-
or-less permanently hidden-out phosphate in the form of highly insoluble
iron phosphate compounds.
One objective of these studies was to determine the optimum conditions
for decomposition of sodium iron phosphate compounds, in order to
facilitate the conversion to AVT operation. A series of experiments showed
that decomposition was speeded by reducing the temperature, as well as
by passing highly diluted feed through the column containing phosphate-
loaded magnetite. The most rapid decomposition and release of sodium
and phosphate was found to take place between 300 and 400'F (150 and
200'C). Recommendations for conversion to AVT included soaks and feed-
and-bleed operations in this temperature range. Additional work'by Connor
and Panson (9.23), for the purpose of evaluating phosphate soaks for denting
control, has verified much of the above.
DISCUSSION
Subsequent experience, after the conversion to AVT and implementation
of the KWU treatment modifications, has shown that tube wastage by
phosphates has been essentially eliminated as a problem in nuclear steam
generators. However, the phenomenon did not immediately disappear in
all plants, since many units were heavily loaded with phosphate-containing
sludge. In some cases, only token efforts were made for removal of
accumulated phosphated material, whether by soaks and feed-and-bleed
in the recommended temperature range, or by sludge lancing. In other
cases, the undesirable rock-like properties of phosphated sludge defeated
determined attempts at removal. Consequently, eddy current inspections
continued to show some progress of wastage on some tubes in some plants,

even for several inspection intervals following conversion to AVT. Even
now, phosphate return is detected on shutdown of some units, indicating
that elimination of the hidden-out material can be a very prolonged process
with normal operating practices.
The two domestic plants that continued with phosphate operation have
not escaped a variety of other corrosion-related difficulties, some of which
may have been compounded by phosphate treatment. Sleeving repairs at
San Onofre 1, for example, were seriously impeded by the presence of
phosphate-loaded sludge at elevations where sleeve brazing was attempted.
The use of phosphate continued until 1987-1988 for operating PWR
plants of KWU-design, despite heavy sludge accumulations and some
reports of tube thinning. KWU recommended lower levels of phosphate
than were used in U.S. PWRs and carefully controlled Na/PO4 ratio to
minimize the occurrence of conditions that could lead to either wastage or
caustic stress corrosion cracking. Nevertheless, KWU has recommended
AVT for new stations, and older units have successfully changed water
treatment to high pH AVT.
The Point Lepreau plant in New Brunswick, Canada, and the Jose
Cabrera plant in Spain have been operating since startup with phosphate
chemistry.
CONCLUSION
Despite the amount of work that has been devoted to the subject of
phosphate wastage and phosphate chemistry, and the extensive experience
that the industry accumulated during the period of phosphate use in nuclear
steam generators, a number of unknowns still remain. One question
frequently asked, because it pertains to current ideas about concentrating
processes in crevices and related configurations, is "Why was tube wastage
by phosphate not seen within tube-to-tubesheet crevices, and why was it
confined to a relatively narrow band at or close to the top of sludge piles?"
The most plausible explanation is that concentration of sodium phosphates
creates a thick, pasty mass (Solid Solution I and Solution II) that can
penetrate sludge pores and crevices with difficulty, if at all. Probably, also
concentrated phosphate solutions have high surface tensions that do not
promote penetration, but no definitive experiments or measurements have
been made to check these suggestions.
Another question relates to the discrepancy between corrosion rates
measured in the laboratory and those experienced in the field. In general,
rates measured in the laboratory were lower. Pessall et al. (9.20) suggest
that disruption of protective films may have prevented the development of
parabolic kinetics in operation units. They further state that, in the majority
of cases when high rates of attack on alloy 600 have been observed in
phosphate-containing steam generators, the region of attack has been
characterized by the presence of a greenish scale. In the laboratory tests,
bright green scale was associated with the highest rates of corrosion and
was predominantly present on all immersed samples in solutions with Na/
PG 4 = 1.6 or less. In contrast, tests at Na/PG4 = 2 were characterized in
general by yellow and yellow-green scales, while tests at Na/PO4 above
2.3 produced scales that tended toward brown, gray, and black. It must be
pointed out, also, that the amount and precision of the laboratory data for
tests at Na/PG4 = 1.6 or less do not support the applicability of parabolic
kinetics to these conditions with any great confidence.
The following statements can be made to summarize what is known
about phosphate wastage to date.
* Phosphate-induced tube wastage is now a relatively well understood
phenomenon in the control bands practiced by domestic utilities.
" Control of phosphate chemistry to avoid wastage, while at the same
time avoiding conditions that can cause caustic stress corrosion cracking,
is very difficult.
" Phosphate treatment adds to sludge burden, which further enhances
hideout, concentration, generation of caustic, and prospects for
corrosion of one kind or another.
Although five German, one Spanish, one Canadian, and two domestic
units continued to use phosphate chemistry control with apparently
successful results, there are still a large number of unanswered questions.
A survey by Pearl provides an update on phosphate treatment practices
and operational performance since 1974 (9.26). It should be consulted before
implementing a phosphate treatment program. The Doel Unit 4 Steam
Generators were converted to phosphate water chemistry after a full bundle
chemical cleaning in May 1992. Concurrently, model boiler tests were started
under EPRI and Laborelec sponsorship.
Below is a list of questions that need answers prior to a change from
AVT to a phosphate chemistry control program.
* Are the phosphate units free of continued wastage?
* Are there any other corrosion problems aggravated or alleviated by
the phosphate treatment? Is the difficulty of phosphate control worth
the possible benefit of lower maintenance costs if corrosion problems
can be controlled?
* What is the acceptable band for control of phosphate?
* Is phosphate control more forgiving of chemical upsets than AVT?
* What should the water chemistry limits be for phosphate control?
REFERENCES
9.1 Cullen, W. H., H. J. Partridge, Jr., and J. P. N. Paine. "IGA/IGSCC
of Alloy 600 in Acid Sulfate Solutions." Paper presented at the
Fifth International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems-Water Reactors, Monterey,
California, August 1991.
9.2 Baum, A., et al. "Proceedings: Workshop on the Role of Sulfur
Species." EPRI NP-6710-SD. March 1990.
9.3 "Measurements of Surface-Induced Microplasticity in Alloy 600 C-
Rings." EPRI NP-6705-SD. March 1990.
9.4 Steam Generator Progress Report, Rev. 4. EPRI, November 1988.
9.5 Odar, S. "Control of Phosphate Wastage in Steam Generators:
German Experience." In Proceedings: IGSCC/IGA Mechanisms
Workshop. Washington, D.C., April 1987.
9.6 Bouecke, R., R. Riess, G. Schucktanz, and L. Stieding. "Experience
With Steam Generators of the KWU Concept." Paper presented at
the International Symposium on Reliability of Reactor Pressure
Components, Stuttgart, March 1983. IAEA-SM-W69/25.
9.7 Paine, J. P. N. "Overview of Steam Generator Operating Chemistry
Performance." In Proceedings of Steam Generatorand Heat Exchanger
Conference, Canadian Nuclear Society, Toronto, Canada, 1990.
9.8 Green, S. J., and J. P. N. Paine. "Materials Performance in Nuclear
55 (October 1981): 10-29.
9.9 Weeks, J. R. CorrosionProblems in Energy Conversionand Generation.
Princeton, N. J.: Electrochemical Society, Inc., 1974, p. 322.
9.10 Baschek, H., and E. Sandona. "The Steam Generator Failure History
of the Nuclear Power Plants Beznau 1 and Beznau 2." Paper
presented at Educational Seminar, Colloquium on Steam Generator
Tube Failures, Southwest Research Institute, 1974.
9.11 Ravich M. I., and L. G. Shcherbakova. Izvest. Sektora Fiz.-Khim.
Analiza Inst., Obshch. Neorg. Khim., Akad. Nauk SSSR, 26 (1955):
248.
9.12 Marcy, V. M., and S. L. Halstead. Combustion (January 1964): 45.
9.13 Panson, A. J., G. Economy, Chia-Tsun Liu, T. S. Bulischeck, and W. T.
Lindsay, Jr. Journalof the Electrochemical Society 122 (1975): 915.
9.14 Economy, G. A., J. Panson, Chia-Tsun Liu, J. N. Esposito, and W. T.
Lindsay, Jr. Proceedings of the 36th International Water Conference,
Pittsburgh, Pa., 1975, p. 161.
9.15 Broadbent, D., G. G. Lewis, and E. A. M. Wetton. Journal of the
Chemical Society., Dalton Ser. (1966): 464.
9.16 Balakrishnan, P.V. CanadianJournalof Chemical Engineering55 (1977):

592.
9.17 Taylor, P., P. R. Tremaine, and M. G. Bailey. Inorganic Chemistry 18
(1979): 2947.
9.18 U. S. Department of Energy. "High Temperature Aqueous
Potassium and Sodium Phosphate Solutions-Two-Liquid-Phase
Boundaries and Critical Phenomena, 275-400 C; Potential
Applications for Steam Generators." ORNL-5842. Washington,
D.C.: Government Printing Office, December 1981.
9.19 Garnsey, R. Journalof British Nuclear Energy Society 18 (1979): 17.
9.20 Pessall, N., A. B. Dunlap, and D. W. Feldman. Corrosion33, no. 4
(1977): 130.
9.21 Bignold, G. J. "The Behaviour of Inconel 600 in High Temperature
Phosphate Solutions." In Proceedings at CEFA Seminar on Corrosion
in Water-Cooled Nuclear Reactors. Behaviour of Inconel 600, Incoloy
800 and Stainless Steel, Brussels, Belgium, February 24-26, 1975.
9.22 Broadbent, D., G. G. Lewis, and E. A. M. Wetton. "The Chemistry
of High Temperature Phosphate Solutions in Relation to Steam
Generators." In Water Chemistry of Nuclear Reactor Systems BNES
Proceedingsof the 1st InternationalConference. October 1977, pp. 53-
62.
9.23 Connor, W. M., and A. J. Panson. "Investigation of Phosphate-
Sludge Interactions." EPRI NP-2963. March 1983.
9.24 Marshall, W. L., C. E. Hall, and R. E. Mesmer. Journal of Inorganic
and Nuclear Chemistry 43 (1981): 449.
9.25 Marshall, W. L. Journalof Chemical and EngineeringData,27 (1982):
175.
9.26 Pearl, W., and S. G. Sawochka. "Review of Field Use and Corrosion
Experience With Phosphate Chemistry." EPRI NP-2347. June 1991.
10
TUBE PITTING
J. F Sykes, TU Electric
J. P. N. Paine,EPRI
M. J. Angwin, EPRI
1991
INTRODUCTION
Scope of Problem
Pitting corrosion affected the integrity of about 20% of the tubes in the
original steam generators at Indian Point 3 and Millstone 2. Major tube
pitting was first discovered in 1981 at Indian Point 3. In addition to these
units, pitting has also been reported or confirmed in steam generators at
Connecticut Yankee, Trojan, Indian Point 2, Ko-Ri, and the retired Surry
Unit 2 steam generator examined at Battelle Pacific Northwest Laboratory.
St. Lucie and Calvert Cliffs have removed and examined tubes with pit-
like eddy current indications. However, pitting was not found. Other plants
have occasionally reported minor shallow pitting, which was discovered
when tubes were pulled for other reasons.
The term "pitting" seems to accurately reflect the physical appearance
of the degradation observed in steam generator tubes. However, the term
"underscale" or "crevice corrosion" may be more precise, since it appears
that the sludge and/or tube scale are necessary to develop the attack.
Agrawal et al. (10.1) commented that the corrosion attack probably started
because of crevice corrosion; at some point, the attack became intensely
localized which resulted in pitting.
Impact of Tube Pitting on Plant Performance

Pitting reduces the local tube wall thickness and thus the margin to
tube leakage. The impact of pitting on steam generator performance.
depends largely on pit depth and area or volume, which in turn depends
on pit growth rate. From Millstone 2 eddy current data, pit growth rate
during an operating cycle varied up to 34% of tube wall thickness. Based
on laboratory tests, it is believed that once a pitting reaction is arrested, it
generally does not reinitiate at the same pit location again. Because of this
tendency for multiple initiation sites, eddy current signals correlate with
an average value of about 10% of tube wall thickness per cycle (10.2). These
values were derived before corrective actions were implemented to mitigate
pitting. Pit growth is known to occur in service. In April 1982, Indian
Point 3 experienced a through-wall leak, which developed while steaming.
It has also been suggested that pits can grow in layup. However, the
necessary inspection data to prove this suggestion are not available.
Inspection of the Indian Point 3 tube which leaked during operation
showed that the leak occurred at one large (0.3 inch [8 mm] diameter) pit.
There was no cracking or tube deformation associated with the leak location.
In addition, a hydrotest was conducted on a pitted steam generator tube
which had a maximum pit depth of 66% of the tube wall thickness. The
Tube Pitting 10-3
tube deformed, but did not leak when pressurized to 10,000 psi (10.3). Virgin
tube material tested at the same time burst at 11,000 psi. Another tube from
Indian Point 3 had an oval-shaped wastage area of 0.8 x 0.5 inch (20 x
12.5 mm) (10.4). The maximum depth of attack was 92% of the tube wall
thickness. There was some deformation of the tube at the thinnest cross
section but this was considered due to stresses induced by the tube removal
process. There was no indication of tube deformation related to plant
operations.
In addition to the consequences of a leak in service, pitting may impact
power plant performance in various other ways such as: (1) require plugging
of tubes and thus removal of tubes from service, (2) require more frequent
shutdowns for eddy current inspections, (3) require lengthy shutdown for
repairs, i.e., sleeving, and (4) require additional evaluation of steam
generator integrity of pitted tubes.
NDE DETECTION
Pitting on the cold leg side is detected using a bobbin coil probe operated
in the differential mode. However, the examination is not routine. Pits
present several challenges to conventional eddy current methods, including
the necessity of eliminating signals from copper-rich scales.
Location of Pit Indications

Most of the pit eddy current indications at Indian Point 3, Millstone 2,
and Connecticut Yankee were located toward the center of the tube bundle
on the cold leg side of the steam generator between the tubesheet and the
first support plate. This is graphically illustrated in Figure 10-1. The eddy
current tests indicated inhomogeneities (pits) at varying elevations within
the affected region. The height of the indications varied within the tube
bundle. For Indian Point 3, the indications averaged about 10 inches (25 cm)
above the tubesheet; for Millstone 2, the indications averaged about 5 inches
(13 cm) above the tubesheet; and for Connecticut Yankee, the indications
averaged about 3 inches (7.5 cm) above the tubesheet. The length of tubing
affected varied from tube to tube but was generally on the order of several
inches. Although most of the eddy current indications were close to the
tubesheet, the leak which occurred at Indian Point 3 during the 1981
hydrotest was just below the first support plate, approximately 40 inches
(102 cm) above the tubesheet. The cause of the leak was never determined.
FIGURE 10-1 .a
Cumulative Tube Plugging and Sleeving Map for MP2 Steam Generator No. 1 Cold Side Through
the 1985 Outage
IUn
I I I I I -7 I *
-
*~ ~ I I I r ill- I I
I
1_ ýT L J
]r ! ] I F ............,7 74,: 7n•ji %7'i 7 7 o .%

o, ...
2:
0
........ } • ,•
~t~wnnnn~~ • ~-C ~409.
500174
------1-
4 8 0
c44iU-~nr3X'u'o 7
fJ P ,0 Ion&.nnn
811 I I II I ., ,- '. . . I I
4-- - 4 .:~
%I ----- -- 4-- J4~4Vt~~ ~ ~~9:.6
~8En
*7,7,~0O N-n
711
37,783
nnson-
7oO~o%0 on 7 "'0
7
-o.
IL 4003
ag"7,0
17-i
-i1--
.. .... .. ........ . ,, , . oo ,.,. 0
I 777,88-77
7 77,
~,47777777-~
7371F"3Y'774777~'3"'&',733'33'7½P77784t'
3 8 773 3 O 7'o..~."n'
7999 ~.5.!..
~ . - 7,9 ~"6 007-0~03375777777~ gO 7'
777 7-33 77~77
3 7777 ~ -
7777707777778 n-.3737737343777383707 77--*-7-7--9 9 -377--777773-37--,
377773n-77--77-'773-77
0
3(340 77~747 77,7,033
7~71'ibfl~W
-99
****
770,747777773
.....S2~-7~' 0 -
o;*~*
07j 0-
' ......
7j-77,7777
373,77
737713,7 7,~ '375 . .
9
7 7 7 77
77~
8: 373077,7n 7,7 3 '17t77.8.~t88!778~-0 n708,t0:- 708,
0 !onSnu ,~. 35177-
7~
Al
60
t -' - *%gRt of
%~004!'!0f 7
S -2
58
57 -
•'ii¸, 777-737 Li
45
7,7<
7,0 -
37
,' 'I 337
773737qQ011
7;.; 7,090 ~9fl8
•:•
7
00n *~*
~37-77nn
577 74%
-I- -
-F---
17 877 7 7 7 7
7p3 r7, n &nnor
3~7375
26-- -c n%~8O~o% 50777

n00~3
I
8~77%%5%C
7--
2Ž -
772 ,9%
38<
00 :0%
o ~nnn7o7
0008 7700,
S~ on 7~7 18
o j','o•rn+o 7 ,-
330n.on 7 ... 7.... .... .. I'.,np,,. tr-,t, 70 . •00-07
3333333
I • [1 IIN2umberŽ7Ž237733333 753738 555555666nn850 7737778 8888I8I2 3 843,3 37L58587888556858888n56
88587890,78.35677g[01
023Ž7-307B97123 Ž838
637B013Ž85677i7
tlI7838
$787,7,72848857B9135790n8I7902q5 78,S860Ž8578,7238787Ž75887Ž73_n78Ž77
eol35 69D •r 777877,75
707,5Ž7315877871gn Ž
BQ2••5 /8477178737758788
Line Number
Tube Pitting 10-5
Millstone 2 and Indian Point 3 have also found pit-like indications on

the hot leg side of the steam generator. These indications are in the central
portion of the tube bundle but they are closer to the tubesheet. Eddy current
inspections at Millstone 2 and Indian Point 3 also indicated large voltage
signals. These indications were in the same location as pits but were
determined to be wastage corrosion.
Prior to the discovery of pitting in September 1981, Indian Point 3 had
inspected a few tubes in the central portion of the cold leg during a
September 1979 inspection. Millstone 2 conducted an inspection of some
cold leg tubes in a December 1980 inspection. There were no indications of
pitting at either power plant during these previous inspections. This may
mean that (1) pits were not present, (2) pits were not detected because of
the cold leg sampling plan, (3) pits were not detected because of probe and
technique insensitivity to pitting defects, or (4) defects were masked by
interfering signals from copper.
Detection and Early Indications of Tube Pitting

When conventional eddy current methods are used to detect pits, several
problems occur. One problem is the sometimes close proximity of pits'to
their nearest neighbors. The eddy current signal from a closely spaced pit
merges with that of its nearest neighbor and this causes difficulty in
interpreting the signal. A second problem is related to copper deposits
within the pits. The signal coming from metallic copper located on the
exterior surface of a tube can sometimes be suppressed by mixing. At Calvert
Cliffs, a tube with an indication was removed and the indication was
subsequently determined to be due to copper interference (10.6). A number
of indications disappeared on tubes removed from the retired Surry steam
generator after chemical cleaning (10.7). However, when copper plates out
on the bottom of a pit, mixing is less effective; the copper signal causes
difficulties in detecting and measuring these kinds of pits. Figure 10-2a
shows the complex signal wave shape that results from closely spaced
defects plus copper. It may be compared to the clean "figure 8" shaped
signal that is generated by an isolated.pit in the absence of copper as shown
in Figure 10-2b. Figure 10-3 shows the results of laboratory eddy current
examinations of tube R12C66 from Indian Point 3, comparing the tube
signals with calibration standards. During destructive examination, pits of
up to 35% of the tube wall were found on tubes which had no reported
eddy current indications in the plant. In another case, a tube which was
classified as having a possible indication was found to have a pit which
penetrated 60% of the wall thickness (10.8).
FIGURE 10-2
Eddy Current Trace of Defects With and Without Copper
(a) Closely Spaced Defects With Copper
(b) Single Pit Without Copper

Tube Pitting 10-7
FIGURE 10-3
Results of Laboratory Eddy Current Inspection of Indian Point 3
Steam Generator Tube R1 2C66
(a) Calibration Standard Machined

Defects of 20, 40, 80, and 100% of
the Wall (Top Left to Right)
(b) Calibration Standard With

Copper Wire on the Tube
(c) Indication on Tube About

18 Inches (45 cm) Above the
Secondary Face of the
Tubesheet
The early detection and sizing problems have been largely eliminated
through the development of a method by Edwards and Lareau that is
optimized for pitting (10.9). In this method, the excitation frequencies are
increased to 200 kHz and 600 kHz. The benefit of this frequency mix can
be seen in Figure 10-4. This figure shows that the same defects are more
readily detected with the 600/200 mix in the presence of copper than with
the 400/100 mix.
FIGURE 10-4
Results of Eddy Current Testing Showing the Effect of Tube
Frequency and Copper Deposition on Detection of Pitting (10.9)
No
Copper
Copper
Over
Pit
% Thru-
Wall 57 38 30 18
Frequency: 400/100 KHZ Support Mix

Probe A720
57 38 30 18
600/200 KHZ Copper Mix

A740HF
Tube Pitting 10-9
The modifications to the standard practice have the following benefits:

* The higher frequency signals are less affected by copper plated on the
outside of the tubes.
* The higher frequency produces a narrower focus, thus improving the
separability of pit signals.
With the use of this method, the accuracy of the inspection is reported
to be greatly improved. Depths estimated in one eddy current examination
agree with depths measured from two pulled tubes to within 5%.
PWR FIELD DATA
Introduction
There appear to be two types of pits. Major pits are those which are
associated with copper-rich deposits and result in penetrations with a depth
of 30% of the tube wall thickness, while minor pits are those which result in
penetrations with a depth usually 10% of the tube wall thickness. Major
pitting has been found or reported in steam generator tubes from Indian
Point 3, Millstone 2, Connecticut Yankee, Trojan, Indian Point 2, and Surry
Unit 2 (retired steam generator). Minor pitting has been found on a number
of units and is usually of an incidental nature. The extent of pitting in the
units with major pitting varies. Indian Point 3 and Millstone 2 have
extensive pitting which has impacted multiple tubes. Although occasionally
deep, the pitting at Connecticut Yankee and Trojan appears to be less
extensive on the whole, than the pitting at Indian Point 3 or. Millstone 2.
Indian Point 2 has reported pitting but has not removed a tube to confirm
eddy current indications. Pitting is one of several corrosion degradation
mechanisms found on the retired Surry Unit 2 steam generator.
Classification of the extent of corrosion attack by degradation mechanism
is not available at this time. In addition, several units were suspected of
having pitting, but subsequent examination indicated that pitting was not
present. These units included Calvert Cliffs 2 (10.6) and St. Lucie 1. This
section will summarize the results of selected available examinations on
tubes removed from units with pitting.
Major Pitting
Indian Point 3. Indian Point 3 has a Westinghouse nuclear steam supply

system. The unit started up commercially with AVT secondary water
chemistry in August 1976. Environmental Technical Specifications
prevented the use of Westinghouse-recommended hydrazine- and
ammonia-treated water during plant layup until September 1981. Layup
prior to September 1981 consisted of filling with demineralized water and
sparging with nitrogen.
In 1978, denting was discovered at Indian Point 3. About three years
later, in January 1981, a turbine blade was thrown puncturing the condenser
shell and about 30 tubes. This failure resulted in about 8,000 gallons
(30,000 liters) of Hudson River Water (partially brackish) intruding into the
steam generators. Indian Point 3 returned to power in April 1981 and
operated without incident until September 1981.
Indian Point 3 was shut down in September 1981 for scheduled plant
maintenance. During this outage (September- November), a primary-to-
secondary leak developed in the alloy 600 steam generator tubing during
hydrotesting of the No. 31 steam generator. Subsequent eddy current
inspection identified multiple indications in all four steam generators at
heights ranging from 6 to 19 inches (15 to 48 cm) above the top of the
tubesheet on the cold leg side. Four tubes were removed in order to
determine the cause of the eddy current indications. A section of tube
R12C66 with a 74% eddy current indication was examined (10.8).
Pits were found on the secondary side tube surface over a 3-inch (8-cm)
section of tubing at an elevation between 16 to 19 inches (41 to 48 cm) above
the top of the tubesheet, an area expected to be near or just above the top of
the sludge pile region. A radiograph of the affected region is shown in
Figure 10-5. As can be seen from this figure, the pits are segregated into
two clusters with larger pits in the topmost cluster. The pits were filled
with corrosion deposit and confined to a shallow wastage region on the
affected portion of the tube as shown in Figure 10-6. The wastage region
was a lightly etched region of the tube. The largest diameter pits were
about 0.070 inch (1.8 mm) in diameter and 0.030 inch (0.8 mm) deep (60%)
based on metallographically measured pit depths. Figure 10-7 shows that
the pit corrosion deposit was laminated. Electron microprobe mapping of
the deposit showed that the laminar appearance was caused by the presence
of metallic copper layers. In addition, the corrosion deposit was enriched
in chromium and depleted in nickel and iron compared to the base metal.
Sulfur and chlorine were also found in the deposit along with trace amounts
of other elements. X-ray diffraction analysis of the pit deposit identified
the chromium-enriched deposit as CrOOH. Considerable difficulty was
Tube Pitting 10-11
encountered in efforts to descale the pit with alkaline permanganate

ammonium citrate (APAC). Examination of the base of a descaled pit
showed very shallow intergranular ditching.
FIGURE 10-5
Radiograph of Indian Point 3 Steam Generator Tube R12C66 (10.6)
Cluster of large pits
Cluster of small pits

FIGURE 10-6
Photograph of Pits on Indian Point 3 Steam Generator Tube R13C66
Note shallow
"wastage" region
(light shaded)
Note corrosion
deposit in pits
52frn
FIGURE 10-7
Photomicrograph of Pit on Indian Point 3 Steam Generator Tube
R12C66 (10.8)
Tube Pitting 10-13
Following the Fall 1981 outage, Indian Point 3 returned to power and
operated normally until two days before a scheduled shutdown in March
1982, when a primary-to-secondary leak occurred while steaming. The
leaking tube, R19C47 on the cold leg side, was removed and found to have
multiple pits. The hot leg section of tube R32C42 was also removed because
of a 99% eddy current indication. Destructive examination of this tube
identified intergranular cracking. The principal crack was axially oriented
and about 0.5 inch (1.3 cm) long. The crack passed through some pits, but
crack initiation did not appear to be associated with pitting. During the
outage, tube R2C72 was removed through the secondary side hand hole.
This method of tube pulling allowed the tube scale to be retained over the
pitted region. Destructive examination of the leaking tube identified a
through-wall pit as the cause of leakage. In the tube radiographs, pits were
faintly visible beneath the scale. The scale was up to 0.060 inch (1.5 mm)
thick and covered the first 3 inches (8 cm) of tubing above the tubesheet.
Electron microprobe analysis of the scale showed that the scale was rich in
silicon, nickel, iron, copper, and zinc. The nickel and iron probably resulted
from the precipitation of dissolved ions produced in the pit. Chromium,
on the other hand, hydrolyzes (Cr+++ + 2H 20 CrOOH + 3H+) in the pit
resulting in a chromium-rich pit deposit and acidic conditions. Figure 10-8
shows the metallographic cross section and microprobe chemical analysis
of the tube scale.
During the March 1982 outage, a related examination was performed
on a plugged section of blowdown transfer pipe (10.10). The pipe section,
removed at the September 1981 outage, was examined and ion exchange
resin beads were found. Though no direct link between the resin beads
and the pitting process has been made, it cannot be eliminated as a possible
precursor.
The steam generators were routinely inspected during refueling outages
between 1982 and 1985. During the June 1985 refueling outage, large voltage
signals were found on the hot leg side of tubes R30C46 and R15C47 from
steam generator No. 32. These tubes were part of a small group of six tubes
which had eddy current indications of 10 volts. The deepest eddy current
indication was 81% and 25.1 volts. Destructive examination (10.4) of the
tubes identified wastage corrosion on both tubes with a maximum depth
of 96%. The wastage area was partially filled with corrosion deposit which
was similar in composition to that found on other tubes from Indian Point 3.
In addition to the wastage and pitting, some intergranular attack was also
found. There were also some areas of nickel-sulfur and copper-sulfur-rich
material and some nickel-rich metallic filaments, possibly alloy 600.
Figure 10-9 shows the appearance of the tube after descaling to remove
corrosion deposits.
FIGURE 10-8
Photomicrograph and Electron Microprobe Chemical Analysis of Pit
and Scale Found on Indian Point 3 Steam Generator Tube RI C72
(a) Photomicrogrph of Pit and Scale
Si Fe Cr NM BSE Cu Zn S Mn CI
(b) Electron Microprobe Chemical Analysis of Pit and Scale

(Note enrichment of Ni, Fe, and Si in scale above pit.
Cr enrichment is only in the pit.)
Tube Pitting 10-15
FIGURE 10-9
Optical Photograph of Tube Showing Wastage (Kink in wastage zone
probably resulted from tube removal stresses.) (10.4)
Millstone 2. Millstone 2 has a Combustion Engineering nuclear steam

supply system. The unit is cooled with ocean water. Millstone 2 started up
with AVT secondary water chemistry in December 1975. Denting was
discovered in May 1977.
Following the discovery of pitting in Indian Point 3 in September 1981,
Millstone 2 shut down in December 1981 for a scheduled outage. During
this outage the alloy-600-tubed steam generators were eddy current
inspected. Multiple indications were found and Northeast Utilities decided
to remove cold leg sections of three tubes for destructive examination (10.11).
The destructive examination identified pitting as a circumferential band of
pits located about 5 inches above the top of the tubesheet. The maximum
measured pit depth was 85% of the tube wall thickness which corresponded
to an eddy-current-measured depth of 83%. The physical and chemical
features of the pits were similar to those found in Indian Point 3.
Figure 10-10 shows the metallographic cross section of a large pit and the
EDS analysis of the pit corrosion deposit. Metallic copper laminations,
chromium-rich oxide, and sulfur were present in the pits. Slight
intergranular attack was observed at the bottom of the pit. This feature
was more prominent in Millstone 2 pits than in Indian Point 3 pits.
FIGURE 10-10
Photomicrograph and EDS Chemical Analysis of Pits From
Millstone 2 Steam Generator Tube 68/58 and 64/54 (10.11)
(a) Photomicrograph of Pit

From Tube 68/58 (Note
metallic copper and light
laminations in the pit
corrosion deposit.)
(b) EDS Analysis of Pit Deposit

From Pit on Tube 64/54
Tube Pitting 10-17
Unlike Indian Point 3, Millstone 2 did not have any major transient
chemistry conditions in the steam generators before pitting was discovered.
Millstone 2 had condenser leakage for about the first six months of operation.
The condenser was overhauled in May 1976, and full-flow condensate
polishing was added in November 1976.
In March 1983, Millstone 2 experienced a primary to secondary leak
resulting in a forced outage. Subsequent eddy current inspection indicated
some tubes had large volume defects (10.5). During a June 1983 outage,
eddy current inspections of the hot leg tubes indicated distorted signals
associated with the tubesheet (DTS) and some shallow (20%) large volume
defects. Three hot leg tube sections were removed and examined by
Combustion Engineering and Babcock & Wilcox.
The destructive examinations (10.12, 10.13) identified pitting, wastage,
denting, and intergranular attack in the damage zone. The attack was
concentrated as a circumferential ring of pitting/wastage between 0 to
1.0 inch (2.5 cm) above the top of the tubesheet. Most of the attack was
within 0.5 inch (1.27 cm) above the top of the tubesheet, Figure 10-11.
Distinct but generally shallow (maximum measured depth = 50%) pits were
scattered on the tube surface above the wastage zone. Denting was observed
on profilometry traces of the tube surface which corresponded to the
distorted tubesheet signal at the top of the tubesheet. Areas of intergranular
attack were found within the wastage zone and within some pits.
FIGURE 10-11
Photograph Showing Circumferential Band of Attack Above the
Tubesheet on Millstone 2 Hot Leg Tube Section (10.13)
Destructive examination of the tubes showed that the deepest area of

wastage was about 25% of the tube wall thickness. Metallographic cross
section of the pits identified intergranular attack and, in one case, a large
metallic copper particle within the pit. Figure 10-12 shows the cross section
of selected areas.
FIGURE 10-12
Optical Photomicrograph Showing Intergranular Attack on
Millstone 2 Hot Leg Tube Section 92/54 (10.13)
Residual Dposit
Cu Particle
,. Specimen M1
Grain Boundary Deterioration

0
I I IaJ 50 pm
Orthophosphoric Acid
Etch
Orthophosphoric Acid
Etch
ID Corrosion Due to
Decontamination Process
I I I k 1J 0.5 mm
Tube ID
Tube Pitting 10-19
Figure 10-13 schematically summarizes the results of Energy Dispersive

Spectroscopy (EDS) chemical analyses of deposits in the wastage and pitted
area on Millstone 2 hot leg sections. This figure shows that the deposits in
the wastage zone were primarily chromium and iron oxides and nickel
sulfide. The pit deposits consisted primarily of chromium oxides, metallic
copper, and copper sulfide. In addition, it was observed that the fraction of
iron in the deposit decreased with distance from the tubesheet interface.
The highest fraction of iron in the deposits was associated with the wastage
and intergranular attack zones. The copper and silicon content in the deposit
increased with distance from the tubesheet interface. Sulfur was associated
with nickel in the intergranular attack zone and with copper in the pitted
areas. The wastage area had both copper and nickel sulfur compounds.
FIGURE 10-13
Schematic of Corrosion Products and Deposits Observed in Pits
and on Tube Surfaces (10.13)
Compounds Observed
On Tube Surface -Top of Sludge Pile
Compounds Observed
CrFe' •,"In Pits
Ni Sulfide Cr-Oxide, Cr(OH), 3H1O20

Metallic Cu • "CuSulfide
Pits Metallic Cu
Cr & Fe Oxides, Ni-Sulfide
Tubesheet
Connecticut Yankee. Connecticut Yankee has a Westinghouse nuclear

steam supply system. The unit went commercial in January 1968. Cooling
water is supplied from the Connecticut River (10-20 ppm Cl, 10 ppm SO 4 ).
In May 1975, the unit switched from coordinated phosphate to AVT. A
year later denting was found. Pitting was discovered in 1983. Based on
eddy current indications, the pitting was most severe on the cold leg (40%
of tubes had indications) beneath the sludge-covered portion of the tube
above the tubesheet. Pitting indications were also found on hot leg tubes
above the first support plate (10.14).
Two sections of hot leg tubes were removed for destructive examination.
The pits were filled with chromium-rich deposit and had intergranular
attack and some intergranular spikes on the bottom of the pit with grain
dropping into the pit deposit (10.14).
Trojan. Trojan has a Westinghouse nuclear steam supply system. Trojan

went commercial in 1975 and started up with AVT water treatment. Eddy
current testing during the 1986 refueling outage indicated several new
discrete indications in two of the four steam generators. Several of these
indications were located within the sludge pile on both hot and cold leg
tube sections. Three sections of hot leg tubing were removed for destructive
examination (10.15). Two of the tube sections had eddy current indications
of 22 and 35% of the tube wall thickness, while the third tube had a copper
signal and no indication of wall penetration.
Some of the pits found on tube R12C63 had relatively deep intergranular
penetrations associated with the defect. Figure 10-14 shows an extreme
example of this morphological feature. Chemical analysis of the pit deposit
indicated that the deposit was rich in chromium, and depleted in nickel
and iron compared to the base metal. One unusual feature of the Trojan
pits was the presence of bands of nickel sulfide as shown in Figure 10-15.
One pit had bands of copper sulfide instead of nickel sulfide.
Retired Surry Steam Generator. Surry replaced steam generators in 1980

because of severe denting. One of the four Surry Unit 2 steam generators
was examined at Battelle Northwest as part of a group project managed by
the NRC. Surry Unit 2 has a Westinghouse nuclear steam supply system
and the unit went commercial in May 1973. The unit switched from
coordinated phosphate to AVT treatment in 1975. Denting was discovered
later the same year. Prior to the discovery of denting, there had been a
number of condenser leaks which affected the steam generator water
chemistry. Surry is cooled with brackish water. Examination of one of the
retired steam generators began in 1982. Layup conditions of the retired
steam generators during the period from removal to the beginning of the
Tube Pitting 10-21
examination is unknown, but layup was probably poorly controlled. The

primary focus of the examination was comparative eddy current testing;
however, some destructive examination work was also performed.
Destructive examination (10.16) of removed tubes identified the
following types of attack: circumferential cracking, circumferential trenches,
intergranular attack, pits, and wastage at the top of the tubesheet; primary
side cracking beneath explosive plugs; primary and secondary side cracking
underneath dents; primary and secondary side cracking in the tube U-bends;
and denting at the tube support plates. From this list it can be seen that
many other forms of corrosion occurred in addition to pitting.
Pitting and wastage were the predominant defects at the top of the
tubesheet. Unlike Millstone 2 and Indian Point 3, pitting was more severe
on the hot leg tubes than on the cold leg tubes. The maximum pit depth
measured on the hot leg tube was 87% of the wall thickness, while the
maximum depth on the cold leg tube was 20% of the wall thickness. The
pitting was concentrated within one inch above the top of the tubesheet.
Chemical analyses of the pit deposits identified similar findings to those
found in pits at other plants. The sludge was found to be different for the
hot leg and cold leg. The hot leg sludge identified Fe30 4 while the cold leg
sludge identified Fe 2 0 3 .
Some of the tubes with pits were subjected to burst testing. The results
of this testing are available in Table 10-1 (10.17).
These results indicate that, in spite of comparatively deep pits, the burst
pressure of the tubing is considerably above the operating pressure.
FIGURE 10-14
Optical Photomicrograph of Intergranular Attack Associated With
Pitting on a Trojan Steam Generator Tube (10.15)
'.R '9'
p
FIGURE 10-15a
Optical Photomicrograph of Pit Deposit From Trojan
FIGURE 10-15b
Nickel and Sulfur X-Ray Maps of Metallic Filaments in Pit Deposit
Note that photographs and X-ray maps represent different pits. (10.15)
Tube Pitting 10-23
TABLE 10-1
Summary of Burst Test Results of Retired Surry Steam Generator
Tubes With Pits (10.17)
Specimen Maximum Pit Burst Pressure( 1 )

Number Depth, % Comments
628C 65 10,078 Pinhole leak
__(2)
826C 57 Stripped end fitting
712C 38 10,652 Axial cracks?
799C 51 6,930
615C 37 10,152
790C 57 10,300
812C 44 9,988
642C 36 10,498
661C 75 9,195 Crack in Pit
795C 29 10,800
792C 38 10,128
1
Burst pressure of virgin tube not cited. Typical burst pressure is about
10,000
2
psi.
Not a valid test. Specimen retested.
Calvert Cliffs. Calvert Cliffs is equipped with a Combustion Engineering

nuclear steam supply system and is cooled with brackish water. The unit
which went commercial in May 1975 started up on AVT secondary water.
Denting was found in 1981. Calvert Cliffs had an early history of condenser
inleakage. During a 1983 outage, eddy current identified a few new
indications which were possibly due to pitting. The new indications were
low voltage with a maximum indicated depth of 48% of the wall thickness.
Two sections of tubing with these indications were removed for laboratory
examination.
The eddy current indications obtained in the plant were confirmed on
each of the tubes. However, subsequent examinations identified corrosion-
related degradation at only one of the three eddy current indication sites.
The attack on this tube consisted of shallow (1 to 2 grain deep) wastage
over a 1/2 inch 2 (3.2 cm 2) area. The other indications were associated with
copper-rich deposits on the tube surface.
PWR Plant Experience-Minor Pitting

Shallow pitting has been reported on alloy 600 tubes removed from
steam generators designed and built by all three American PWR NSSS
vendors as well as foreign vendors. For example, pits have been found on
steam generator tubes at Oconee 1, 2, and 3 (B&W) (10.18), Palisades (C-E)
(10.19), and Beznau 1 (W) (10.20). The extent of minor pitting is unknown
since the presence of pits has been only incidental on most tubes removed
from service. Therefore, it has not been highlighted during tube
examinations and information on pit-corrosion-deposit morphology and
chemical composition is not available. The small pits have not been
restricted to a specific location within a steam generator; their occurrence
may be of a more random nature than the pitting found at Indian Point 3
and Millstone 2.
The Canadians have conducted a laboratory pitting test program as
reported in References 10.21 and 10.22. TWo older German plants (Stade
and Biblis-B) have experienced some pitting in alloy 800 tubes. The Germans
have conducted laboratory test programs on pitting (IQ% 10.24 . However,
since they have changed their design for better circulation of inlet water,
they do not consider pitting a major problem.
Summary of PWR Field Data

The pits found on steam generator tubes removed from Indian Point 3,
Millstone 2, Connecticut Yankee, and the retired Surry Unit 2 steam
generator have similar characteristics. The pits are usually found in clusters
above the tubesheet and within the sludge pile. The pit corrosion product
is rich in chromium. There are also indications of intergranular attack at
the base of the pit, in some cases deep.
LABORATORY TESTS
Introduction
Laboratory test results specifically related to steam generator tube
pitting corrosion have been published in References 10.25 through 10.32.
Reference 10.25 provides a summary of pitting information from various
laboratories and utilities. The work reported in References 10.26 through
10.32 was funded by either EPRI or the Steam Generator Owners Group I
or II. Laboratory projects on pitting are listed in Table 10-2.
The terminology used in electrochemical testing can be confusing. For
the purpose of discussion, Figure 10-16 illustrates the location of selected
electrochemical potentials for a typical active-passive metal.
Tube Pitting 10-25
TABLE 10-2
Summary of Contracted Work on Steam Generator Tube Pitting
Project Contractor Type of Test Reference

S124-1 Westinghouse Beaker 10.26
S138-6/ Battelle Columbus Division Beaker, Autoclave 10.27
S308-1(1)
S308-3 Ohio State University Electrochemical 10.28
S308-4 Combustion Engineering Pot Boiler 10.29
S308-5 Brookhaven National Laboratory Electrochemical 10.30
S308-6 Battelle Columbus Division Autoclave 10.31
RP2163-3 University of Leuven Electrochemical 10.32
1
Work initiated under SGOG I, S138-6, completed under S308-1.
FIGURE 10-16
Schematic Diagram of a Cyclic Anodic Polarization Curve of an
Active/Passive Alloy (Modified From Reference 10.32)
and/
Electrochemical or
Potential Critical Trans assivg
-Pitting
E, .Potential
Pit Initiation
Potential
E.. Protection
Potential Passive
Primary
- Passive
Potential
Active
E~mj.+"- Corrosion
Potential
Current Density, Log Scale
Key
EpI-Pit initiation potential
Ep -Pitting potential, pit propagation
Erp-Pit repassivation (protection) potential
Summary of SGOG I Work (10.26, 10.27)

Based on the results of laboratory testing, Reference 10.27 concluded
that steam generator type pits were produced at medium to high
temperatures, 302-572°F (150-300'C), in the presence of chloride and copper
ions or other oxidants.
Low temperature testing was able to produce pitting on alloy 600;
however, the pits were generally free of corrosion products. Some tests
with dilute concentrations of copper chloride and sodium thiosulfate did
result in pits having corrosion products; however, the characteristics of the
corrosion product morphology were not considered typical of steam
generator pit morphologies. Increasing the pH above 8.5 or reducing the
concentration of dissolved oxygen were found to retard or prevent pit
growth (10.26).
Laboratory testing at intermediate temperatures, 212-302°F (100-150 0C),
resulted in pitting in short times and with many of the features found in
steam generator tube pits.
High temperature, 482-572'F (250-300'C), laboratory tests by a number
of investigators (10.25) were found to produce pits with some of the steam-
generator-pit morphological features. These tests were usually accelerated
by using concentrated solutions of seawater, cupric chloride, and sodium
sulfate. Sludge was also added to some tests. In these tests, severe pitting
was found beneath sludge-covered portions of the tube. In tests in cupric
chloride, it was found that coupling alloy 600 steam generator tubing to
carbon steel retarded pit growth on alloy 600 (i.e., galvanic protection).
Galvanic protection probably explains why numerous denting tests
(concentrated aggressive solutions) have not produced pitting; that is, the
sacrificial corrosion of the steel protected alloy 600 even in very aggressive
bulk solutions.
A number of high temperature autoclave tests conducted by Battelle
(10.27) were performed with the specimen in contact with titanium autoclave
liners. Pitting on these samples was found on portions of the tube covered
by sludge. Corrosion, which was considered to be galvanic corrosion, was
found on the tube surface contacting the titanium liners. Both the pits and
the galvanic scale had features similar to steam generator tube pits.
Intergranular attack on the pit base was observed to be associated with
sulfur species. In addition, it was observed that the corrosion product in
laboratory pits was enriched in sulfur, even though sulfur was not present
in the test solutions. Finally, high temperature autoclave testing showed
that steam generator pit features could be developed without thermal
cycling (startup and shutdown). The tests indicated that severe pitting did
not occur with 1,000 ppm CuC12 but did occur with 10,000 ppm CuC12.
Tube Pitting 10-27
Ohio State University Test Results (10.28)

OSU testing was primarily concerned with the determination of pit
initiation and propagation conditions. Electrochemical techniques were
used to obtain pitting data at various temperatures in various environments.
Figure 10-17 summarizes the results of pit potential measurements (Ep)
on alloy 600 in NaCl and the corrosion potential (Ec) of alloy 600 in CuC12
solutions as a function of temperature. This figure shows that the pitting
potential decreases as temperature increases and, that at any given
temperature, Ep is 150 mV lower for the 0.1 m (5,840 ppm) NaCl solution
than for the 0.01 m (584 ppm) solution.
FIGURE 10-17
Corrosion Potential of Alloy 600 in CuCI2 Solutions As a Function of
Temperature (10.28)
500
(3.0) L(60)
A100y 2000 ppm CuCI2
400 (30)1(85)
500 1000
(35?
100
(168 )
1'00(1pp68) Hours
S\ ,l 1 OPP
68) m
(168))
\0P 'b
>-0- 20p0pppm
00 0
O8 (8
VSOPppm
0 (o-00 - 18
-100- OP
02 N"m
C 0- (80)
20ppm CuCI2
-300 FAn-4An~PC
1O0 200 30O

Temperature (0 C)
The corrosion potential of alloy 600 in CuC12 solutions over the entire
range of values tested is almost always above the NaCI pit initiation curves.
This means that for dilute concentrations of CuC12 , pitting can be expected
at any temperature.
The influence of CuC12 on pitting behavior as a function of temperature
is shown in Figure 10-18. This figure shows that as temperature increases,
the concentration of CuC12 needed to initiate and propagate pits decreases.
The most pronounced change occurs over the 122-392°F (50-200 0C) range.
At 536°F (280°C) pitting was not observed in a 10 ppm CuC12 solution but
it was observed in a 20 ppm CuC12 solution.
FIGURE 10-18
Pitting Susceptibility of Alloy 600 in Cupric Chloride As a Function
of Temperature (10.28)
101
Aerated{ 3 Moderate Pitting

Solution[ No Pitting
0 Severe Pitting
60 Deoer. 13 Moderate Pitting
Solution 0 Stight Pitting
I0-- d85 0 No Pitting
3001 ()168 08
Numbers=Exposure Time (hours)
i0
lo 035 0168 (3168 08
o
0
U- . .
Temperature (OC)
Tube Pitting 10-29
Finally, Figure 10-19 shows the results of testing aimed at determining

the kinetics of pit growth in a 100 ppm CuC12 solution at between 302-
5360 F (150-2800 C). The test data were obtained from samples which were
not electrochemically polarized. This figure shows that pit growth occurs
rapidly and is followed by a period of almost constant growth. The initial
pit growth rate was greater at 302OF (150'C) than at 530°F (2800C).
Reference 10.28 contains pitting potential information in NaCl,
measured corrosion potentials in CuC12, and calculated copper ion
equilibrium potentials, as well as detailed analytic information on the pitting
phenomena in cupric chloride.
Metallographic/visual and SEM/EDS examination of the pits showed
that the pits generally had more lateral than perpendicular growth. Most
of the pits contained corrosion products which were composed primarily
of chromium oxide with lesser amounts of Ni, Fe, Cu, Cl, S, and Ti. Some
pits were free of corrosion products. These pits tended to be those formed
at lower temperatures and those formed in more concentrated CuCI 2
solutions. However, the density of pits formed in CuC12 solution increased
with temperature and solution concentration.
FIGURE 10-19
Pit Depth As a Function of Time at 302 and 536 0 F (150 and 280 0 C) in
Deaerated 100 ppm CuCi2 (Modified From Reference 10.28)
103
13
. . . . . ... I . . . I'
Deaerated 100 ppm CuCI2
Slope=O.39
S " -Slope=O.40'
0)
101
10210
Time(minutes)
Combustion Engineering Test Results (10.29)

Combustion Engineering pot boiler tests were aimed primarily at
developing prototypic pits under fault-water conditions, and then
evaluating the effects of remedial actions on continued pit growth. Two
pot boilers were operated with actual heat fluxes through several test phases
exposing the tubes to a variety of fault-water conditions.
The actual test conditions for both pot boilers are summarized in
Table 10-3. The basic difference between the two pot boilers was that 1-0
operated with seawater and CuCI2 as a fault-water additive while 2-M
operated with a seawater fault only until the last two phases of testing. In
addition, two tubes from pot boiler 1-0 were chemically cleaned.
TABLE 10-3
Summary of Combustion Engineering Pot Boiler Test Conditions
Pot Boiler 1-0

Conductivity Oxygen Boron Duration
Test Phase(1 ) Specific Cation (ppb) (ppb) Days
IA AVT 7.86 30.0 - 10 -- 4
lB Aerated AVT 8.25 28.9 - 125 - 13
2 Aerated SW+CU 5.48 90.0 280 100-2,000 - 86
3 SW+CU 5.84 85.0 269 10 - 84
4 Aerated SW+CU 4.99 85.2 292 30-1,000 - 40
5 4+Boron 5.7 82.9 326 30-1,000 17.2 23
Pot Boiler 2-M

Conductivity Oxygen Duration
Test Phase( )
2
Specific Cation (ppb_) Day-s
1 AVT 8.75 19.1 - 10 10
2 Aerated SW 5.9 104 287 20-2,000 85
3 SW 7.0 85.2 311 10 41
4 Aerated SW 8.01 64.5 228 20-2,000 39
5 Aerated SW+CU 6.49 70.2 273 35-1,500 75
6 Aerated SW+CU 8.85 124 285 20-1,500 21
All chemistries deaerated unless otherwise indicated.

SW = seawater
CU = CuCI2
AVT = deaerated All Volatile Treatment
1
Chloride averaged about 22 ppm during phases 2-5.
2
Chloride averaged about 22 ppm during phases 2-6.
Tube Pitting 10-31
The eddy current examination results of pot boiler 1-0 following each
phase of testing are summarized in Table 10-4. Table 10-5 summarizes the
results of eddy current testing on pot boiler 2-M following each phase of
testing.
The eddy current data indicates that electrochemically generated pre-
pits did not grow under any of the test conditions found in either pot boiler.
In addition, the tests conducted in aerated solutions resulted in deeper and
more severe pits than those pits developed in the pot boiler with deaerated
solutions. Pits were more severe on specimen tested in aerated CuC12 , but
pits were also found on specimen tested in aerated seawater. When the
seawater environment was changed to a deaerated environment, pit growth
ceased and new pits did not initiate. However, if the environment was
subsequently aerated, pit growth resumed and new pits would initiate.
Boric acid and chemical cleaning on pot boiler test 1-0 did not indicate any
beneficial effect of either action in subsequent exposure periods to pit
initiating environments. The C-E pot boiler test data also showed that the
higher pH environment with morpholine may be effective in preventing
pit initiation.
Tables 10-4 and 10-5 show that pits did initiate on non-pitted tubes in
pot boiler tests 1-0 and 2-M. While new pits did initiate, the pits did not
propagate completely during subsequent exposure periods. Thus it appears
that pits grew rapidly and then ceased growing after pit initiation. Pit
growth rate information is difficult to derive from the C-E pot boiler tests
because the pits generally did not continue to grow during subsequent
exposures, and because the chemistries were often changed during
subsequent exposure periods. Nevertheless, Figure 10-19 includes test data
from both C-E pot boilers. The data are in harmony with projected data
from the OSU tests.
Figure 10-20 illustrates the appearance of attack on one tube beneath a
surface previously covered by an umbrella. Destructive examination of
pot boiler test tubes confirmed that the pits were prototypic in nature. The
pits were filled with chromium-rich deposit. Some pits were found to have
layers of metallic copper. Figure 10-21 shows the metallic copper layers
found in one pit. In addition, the pit deposits consistently contained low
levels of sulfur and titanium. Intergranular attack and grain dropping into
the pit deposit were also observed as shown in Figure 10-22. Within the
same pit, some areas contained severe intergranular attack while other areas
contained very little attack. Electron microprobe traces of the concentration
of sulfur were performed in representative areas with and without
intergranular attack. The microprobe traces are shown in Figure 10-22.
These traces show that sulfur enrichment is present in the areas with the
most severe intergranular attack.
TABLE 10-4
Pot Boiler Test 1.0 Non-Destructive Examination/ECT Results (10.29)
Defect Height TEST PHASE

Location A.T.S.(1 ) 1 2 (Interim) 2 (Final) 3 (Interim) 3 (Final) 4 5
(inches) (%)(2) (%N) (%N) (%N) (%N) (%N) (%N)
1HL
Top Cup 12.0 57/14 65/3.2 73/3.7 73/2.6 73/3.2 68/4.1
9.0 69(3) 59/6 69/4.1 67/4.4 64/2.6 67/4.1 61/2.8
8.8 67 59/6 69/2.1 69/2.9 67/0.4 64/2.3 68/1.5
Bottom Cup 4.5 68 68/8 77/5.5 7.0/6.5 67/3.4 75/5.0 69/4.2
4.3 75 72/12 71/6.2 73/7.1 71/3.6 75/5.5 70/3.8
Collar 4.0 34/4.3 32/2.4 42/3.6 39/3.0
3.0 44/4.2 58/1.7 58/0.3 62/2.0 58/1.6
1.5 65 54/3 58/2.2 67/1.4 68/1.1 60/3.0 64/2.4
1.3 - - 20/1.8 (4)
1.0 66 59/4 65/2.2 54/1.0 54/0.5 62/2.5 (4)
-- S Dent
-- S 67/1.4 70/0.6 77/0.8 (4)
1CL
Top Cup 10.5 46/11 56/3.6 72/5.0 73/2.8 73/6.7 71/2.8
9.5 62 62/5 60/3.5 58/4.1 55/2.2 60/3.4 58/1.8
9.0 58 51/3 58/2.2 58/3.0 57/1.6 64/2.8 58/1.6
8.0 20/3 51/2.0 52/4.7 54/2.8 53/7.0 55/8.0
7.8 64/4.0 100/1.6
TABLE 10-4 (cont'd)

Location A. T.S 2 (Interim) 2 (Final) 3 (Interim) 3 (Final) 4 5
(inches) (%)(2) (%vN) (%N) (%N) (%N) (%N) (%N)
1CL
Top Cup 7.3 21/3 51/2.0 57/2.0 56/1.1 55/3.6 49/2.1
5.5 64 60/4 61/2.4 61/3.2 60/1.8 55/2.6 76/4.1
4.5 62 67.4 66/3.4 66/4.2 64/2.3 74/3.0 70/2.2
Collar 1.5 74 74/5 76/3.9 73/4.8 69/2.4 71/3.5 70/2.7
0.5 68 64/4 72/4.1 68/3.9 71/2.4 73/4.5 67/2.5
"TS Dent Dent Dent
2HL
Top Cup 11.0 54/12 62/6.2 68/8.4 58/5.3 56/14.5 63/6.0
Bottom Cup 7.8 20/5 47/3.0 (4)
7.5 69/0.8 68/5.4 63/2.8 60/9.5 60/4.2
7.3 51/12 53/4.6 61/6.8 58/3.4 58/1.4 68/2.3
Collar 3.8 30/7 43/2.3 49/2.3 42/1.4 45/4.3 46/1.8
3.0 36/7 38/2.2 49/3.1 47/1.3 46/4.5 44/2.2
2.5 61/2.8 63/1.4 64/3.8 63/4.0
2.0 41/7 45/2.5 63/4.7 68/2.2 65/6.0 57/3.6
1.0 mult./7 (4)
0.5 39/7 mult.(3) (4)
Ca
TABLE 10-4 (cont'd)

Location A. T.S.0) 1 2 (Interim) 2 (Final 3 (Interim) 3 (Final) 4 5
(inches) (%)(2) (%N) (%N) (%N) (%N) (%N) (%N)
2HL
Collar TTS 20/1 42/2.2 74/8.6 70/5.7 80/16.8 71/5.6
TTS 58/3.2 58/2.9 71/11.4 68/7.9
-- S Dent Dent
2CL
Top Cup 12.0 55/10 58/1.1 56/4.7 58/2.6 73/7.0 67/3.3
11.8 70/2 55/3.4 76/2.1 76/1.0 51/2.3 77/0.9
Bottom Cup 8.0 20/2 37/2.9 47/4.8 47/2.2 47/7.7 46/3.7
7.3 50/6 53/2.1 53/2.1 53/1.3 62/5.0 58/1.6
7.3 20/1.7 (4)
7.3 20/0.9 38/1.8
5
Collar 3.0 N.Q.( )
TTS
Dent Dent Dent Dent
3HL
Top Cup 12.0 20/9 30/1.4 20/2.0 17/0.9 55/3.4 37/2.0
11.5 47/2.0 54/2.9 54/1.5 48/6.5 51/2.1
TABLE 10-4 (cont'd)

Location A. T. S.(1 ) 1 2 (Interim) 2 (Final) 3 (Interim) 3 (Final) 4 5
(inches) (%)(2) (%N) (%N) (%VN) (%VN) (%N) (%N)
3HL
Bottom Cup 7.5 - 36/14 - - - - -
7.3 - 34/2 55/2.3 62/3.8 62/2.1 58/10.5 70/6.7

7.3 - - - - - - 37/1.4
Collar TTS - Dent Dent Dent Dent -
3CL
Top Cup 11.0 - 21/12 - - - - -
9.5 - 68/3 - - - - -
Bottom Cup 8.0 - 20/3 - - - - -

7.0 - 55/3 - - - - -
Collar TTS - Dent Dent Dent Dent - -

Bottom Cup 7.5 - - - 38/1.6 37/0.6 45/5.3 48/2.5
7.5 - - - - - 26/5.5 42/2.3
4HL
Top Cup 11.3 - 37/14 32/5.6 - - - -
11.0 - 51/4 57/3.5 - - - 42/3.0

Bottom Cup 8.0 - 20/2 25/3.2 - - - 24/2.4
7.3 - 29/8 61/5.6 - - - -
U'
TABLE 10-4 (cont'd)
Location A.T. S.(1 ) 1 2 (Interim) 2 (Final) 3 (Interim) 3 (Final) 4 5

(inches) (%)(2) (%N) (%N) (%N) (%N) (%N)
(%N)
4HL
5
Collar 3.0 N.Q.( )
TTS Dent Dent Dent
4CL
Top Cup 10.5 20(1) 28/2.5
Bottom Cup 7.0 21/8 44/2.4
6.5 49/4 51/1.7
Collar 4.0 22/0.6
1.0 43/3 (4)
TTS 56/5 48/3.1
1Voltage was not reported

2
Heights above tubesheet are not exact because a hand-pulled probe was used. The values shown were
determined by ECT in preliminary exams. The locations are given for the purpose of identifying defects relative to
each other.
3
Prepits.
4
Defect indicated on previous exam(s) - not called on current exam.
5
N. 0. - not quantifiable (too distorted or too small).
0 0 0
TABLE 10-5
Pot Boiler Test 2-M Non-Destructive Examination/ECT Results (10.29)

Location A. T. S. 1 ) 2 (Interim) 2(Final) 3 4 5 (Interim) 5 (Interim) 5 (Final) 6
(inches) (%)(2) (%N) (%N) (%N) (%N) (%N) (%N) (%/N)
1HL
Top Cup 11.4 71/4.8 70/3.1 65/7.2 77/5.4 77/12.7 77/8.6 77/12.7
10.7 60/1.5 60/1.7 66/1.0 62/2.2 55/0.9 53/1.3 58/0.7 65/1.2
10.2 61/2.6 61/3.1 60/1.7 62/3.2 63/1.5 61/2.6 54/0.8 64/1.7
Bottom Cup 7.6, - 62/4.3(2) 56/2.4 61/4.7 60/4.7 64/8.6 58/5.8 "57/6.6
5.3 63/3.4 58/4.1 59/2.5 58/5.1 60/2.0 60/3.5 62/2.6 61/3.0
4.9 51/2.6 54/2.7 54/1.6 53/3.4 55/1.3 53/2.3 53/1.9 58/1.7
Collar 1.9 61/2.3 63/2.6 62/1.5 63/3.1 61/1.2 58/2.1 58/1.3 67/1.9
1.3 77/1.2 74/1.8 74/1.0 71/2.1 76/1.1 77/1.3 93/1.0 76/1.2
1CL
Top Cup 12.1 - 52/2.2 78/3.3
10.4 75/1.6 63/1.4 66/0.8 70/2.1 70/0.9 70/1.2 70/0.8 73/1.1
10.0 59/2.0 59/2.6 59/1.6 53/2.9 59/1.2 59/2.4 57/1.3 54/1.4
Bottom Cup 5.9 39/3.7 35/2.2 44/2.9 CD
5.2 65/2.5 63/3.0 59/1.9 55/4.1 52/1.8 58/2.9 55/1.8 56/2.1
4.7 68/2.9 68/3.6 60/2.0 61/3.8 62/1.7 64/2.9 61/2.2 69/2.1
Collar 1.7 63/3.8 63/4.5 62/2.5 60/5.1 60/2.2 60/3.2 61/2.5 63/2.5
1.4 71/2.8 70/3.2 70/1.7 69/3.6 69/1.4 68/2.4 65/1.8 73/1.8
(3) (3) -16
TTS 20/2.1
TABLE 10-5 (cont'd)
41-
Location A. T.S.0) 2 (Interim) 2 (Final) 4 5 (Interim) 5 (Interim) 5 (Final) 6
(inches) (%)(2) (%N) (%N) (%N) (%N) (%N) (%N) (%N) CD
2HL
Bottom Cup 7.8 - 61/3.2 (2) 62/2.0 62/3.8 60/2.8 64/5.9 60/4.3 57/3.4
2CL
Top Cup 13.0 - 59/9.0 60/11.6
Bottom Cup 8.6 - 38/6.1 51/10.5
Collar 0.9 - 81/3.0 91/3.1
0.7 - 57/7.1 60/6.0
0.7 - - 20/8.1
3HL
Bottom Cup 7.8 20/1.6(2) 56/0.8 58/6.5 59/6.9 62/9.3 59/7.9 63/11.4
3CL
Top Cup 11.9 - -- - 55/2.8 74/2.0 77/3.6 62/3.3 68/3.1
Bottom Cup 7.3 34/3.1 55/3.7 70/4.4
00
TABLE 10-5 (cont'd)


Location A. T. S.(1 ) 2 (Interim) 2 (Final) 3 4 5 (Interim) 5 (Interim) 5 (Final) 6
(inches) (%)(2) (%N) (%N) (%N) (%N) (%N) (%N) (%N)
4HLL
Top Cup 11.6 - 60/5.9
Bottom Cup 7.5 - 20/1.1(2)
Top Cup 11.3 28/2.5 (4)

11.3 41/4.5 (4)
Bottom Cup 6.5 41/4.3 (4)
Collar . TTS 20/1.1
Top Cup 11.8 61/3.3 78/4.0
Bottom Cup 8.6 35/2.9 64/2.0
491
Bottom Cup 7.7 66/1.6 63/7.6
1
H7
Heights above tubesheet are not exact because a hand-pulled probe was used. The values shown are averaged over all Pot 2-M exams.
The locations are given for the purpose of identifying defects relative to each other.
2
Defect indicated by ECT as "possible".
3
Defect indicated on previous exam(s) - too small to be called on current exam. 0~
4
Flaws found on 10/4/85 - Recheck of Tube 4. Flaws not called in ECT inspection 9/10/85.
FIGURE 10-20
Tube Showing Pits Beneath Lower Umbrella
Brookhaven National Laboratory (10.30)

Brookhaven testing was aimed at determining the pit initiation,
propagation, and repassivation characteristics of alloy 600 as a function of
several chloride and oxygen concentration levels. Electrochemical corrosion
techniques were used to monitor the potential and to polarize the sample
to induce pitting. Tests were conducted at 77 and 203'F (25 and 95°C) in
NaC1 solutions over the range 0.01 to 1.0 M chloride. The solutions were
deaerated, aerated, and oxygenated.
Figure 10-23 shows the pit initiation and propagation potentials as a
function of chloride concentration at 77 and 203'F (25 and 95°C). These
potentials decrease with increasing concentration of chloride at both 77 and
203'F (25 and 95°C). In addition, the pit initiation potential is about 40-
130 mV lower than the pit propagation potential. Thus, pits will not continue
to propagate unless the potential is above the pit propagation potential.
The test data from OSU and BNL are similar where they can be compared
(0.1 m NaC1 95°C, compare Figure 10-17 and 10-23). Figure 10- 24 illustrates
Tube Pitting 10-41
the domains of pitting based on chloride concentration and fraction of

oxygen in the sparge. This figure shows that pitting will not initiate at low
chloride concentrations over the entire oxygen range which was evaluated.
For low concentrations of oxygen, pitting will not initiate with concentrated
chloride solutions. Pit propagation will occur at higher chloride
concentrations.
FIGURE 10-21
SEM Photographs and Distribution of Copper in a Pit-Pot Boiler
Test 1-0 (10.29)
FIGURE 10-22
Optical Photograph of Pit and Electron Microbe Traces of Sulfur
Concentration in Regions With and Without IGA (10.29)
SArea 1
Area 3
Wt % S vs. Location
In Base Metal and InOxide, Area 1
13
12-
11-
( 7
• 45
4
3
2
0
-100 100 300 500
No. Microns (Zero Metal/ox Transition)
Wt %Ci vs. Location
InBase Metal and InOxide, Area 1
13
6
*) 10
5
3
2
1
100 300
No. Microns (Zero Meta~lox Transition)
Tube Pitting 10-43
FIGURE 10-23
Pitting Potential Versus Log (CI) at 77 0 F (25 0 C) and 203°F (95 0 C) for
Alloy 600 With Lab Air Sparge (10.30)
700•+. I I I I
600
500
50 400
EJ
300
EL 200
100
-100
-200 1_
0.01M 0.1M 1.0M
Log [C1]
FIGURE 10-24
Chloride and Oxygen Levels Required for Pitting Initiation and
Propagation of Alloy 600 at 203°F (95 0C) (10.30)
0.25 1 1 1 1 I
o Pitting Propagation
* Pitting Initiation
0.20
0Ep, prop
3C
=\
o
n
~~~~Pittin
ntito
-But o Propagation
s 0.o .• B
7p, . Lt
E _
% Oxygen in Sparge
Battelle Columbus Division's Test Results (10.31)

The objectives of this test program were to determine the effect of
operating and environmental factors on pitting corrosion and to evaluate
the effectiveness of several remedial actions.
Battelle conducted static autoclave tests in concentrated cupric chloride
solutions using a variety of experimental conditions. Refreshed autoclave
tests were conducted in dilute cupric chloride solutions with various
specimen types and chemical additives to the solutions.
Figure 10-25 is a plot of the maximum measured pit depth versus
concentration of CuCI2 for aerated and deaerated solutions at 320 and 536'F
(160 and 280'C). This figure shows that the maximum pit depth increases
as the log of the concentration of CuC12 over the range I to 100 ppm CuC12 .
FIGURE 10-25
Plot of Pit Depth Versus Log Concentration of Cupric Chloride
(10.31)
700
--No Sludge--
280 C 160 C About 650 pm /
o O Aerated /
600 - a Deaerated /
Once Through Refreshed
Autoclave Test Loop Extrapolation /
t ,u0,,00 ppm
500 Exposure From 95 to 160 /
Hours, Most About 100 Hours •uC
E
400 10,000 ppm

• CuCt 2
d_ I Two Data Points
E
E 300
•, - Static
Autoclave
Intercepts Line at About * 4;7 Test
200 - 20 ppm CuCI 2 No
Sludge
100
0 1 l I.,. hlp1
100 l0 u00
pI"00
01 10 100 1000 10000
Concentration of Cupric Chloride (ppm)
Tube Pitting 10-45
Extrapolation of this linear relationship to 10,000 ppm CuC12 predicts pit

depths 650 prm (26 mils) deep. Pits of this depth were measured on
specimens exposed to sludge (25% Cu + 25% Fe 3 04 + 50% Cu) in static
autoclave tests. The pits on non-sludge covered portions were about 250 rtm
(10 mils) deep, which corresponds to a CuC12 concentrate of 20 ppm based
on the linear relationship. These results suggested that the concentrated
chemistry could be maintained in the sludge, possibly because of excess
reactants. Accordingly, a less concentrated solution may have formed above
the sludge. Figure 10-25 also shows that the role of temperature and the
degree of aeration were not clearly separable from the concentration
dependence.
The results of selected Battelle pit growth rate information have been
included on Figure 10-19 along with OSU and C-E test results. According
to Figure 10-19 the data from each laboratory are consistent.
Figure 10-26 shows the appearance of a specimen partially exposed to
sludge in a 536°F (280'C) test in CuC12 . Large and deep pits were found on
the sludge-covered surface of the tube while numerous shallow pits were
found above the area covered by sludge. Destructive examination identified
metallic copper laminations in the pits under the tube deposit, which was
in agreement with observations by Combustion Engineering. The pits above
the sludge also had metallic copper but this was globular in shape.
Figure 10-27 illustrates the copper laminations found in one such pit. In
one case, the attack resembled the deposition of copper found along the
grain boundaries. Actual deposition of copper was observed in some
sections as shown in Figure 10-28. This grain boundary scale was similar
in appearance to that found in one steam generator tube pit from Millstone 2.
The pit deposits were similar in composition to the deposits tested at
Ohio State and Combustion Engineering. A few pits had caps of corrosion
product above the pit. These deposits consisted primarily of nickel and/or
iron, thus indicating that the more soluble nickel and iron constituents of
alloy 600 precipitate outside the pit, while the less soluble chromium
precipitates (hydrolyzes) in the pit. Figure 10-29 shows EDS spectrum traces
and computer analysis of deposits in and above the pit.
The Battelle testing concluded that sludge was an important factor in
the formation of prototypic pits. Since the tests which resulted in these pits
were performed at elevated temperatures in concentrated chemistries, it
was concluded that prototypic pits could form at temperatures near 5360F
(280'C) in concentrated chloride chemistries. Tests conducted without
sludge at lower temperatures, or in less concentrated solutions, did not
result in pits with morphological features similar to those found in steam
generator pits. In addition, low-temperature-initiated pits did not propagate
and did not end up with prototypic features in the pit deposit when they
were exposed to high-temperature test conditions. This suggests that the

formation of certain morphological features requires the presence of a set
of specific environmental conditions. The kinetics of pit growth in the
Battelle testing were similar to the kinetics previously shown for Ohio State
testing.
FIGURE 10-26
Photograph of Tube Specimen Tested in Cupric Chloride Partially
Exposed to Sludge (10.31)
Small Shallow Pits
No Sludge
Sludge
Large Deep Pits
End Scale
Tube Pitting 10-47
FIGURE 10-27
Photomicropraph of Pits in and Above the Sludge (10.31)
'Metallic
Copper-
Colored
Particles
(a) Optical Photomicrograph of Pit Above Sludge
(b) Optical Photomicrograph of Pit on Sludge-Covered Portion of Tube

FIGURE 10-28
SEM Photograph and EDS Analysis Showing Copper Deposition
Along Grain Boundaries (10.31)
(a) SEM Photograph of Base Material and Deposit
c
'U
Element wLQ
Al 2.6
Si 1.5
S 0.6
15k- Cl 0.5
Cr 1.4
Fe 1.0
Ni 31.0
49 10k- Cu 61.6
5000- c
- U C
0 joL" U -S'
0.000 5.000 10.000 15.000 20.000

Energy (keV)
(b) EDS Spectrum and Analysis of Area Marked in the SEM Photograph
Tube Pitting 10-49
FIGURE 10-29
EDS Spectra and Chemical Analyses of Deposits Associated
With a Pit (10.31)
Elemnent wV&
10K 2.7
Al
Si 1.0
S 1.2
CI 0.6
1T 0.6
o
0 Cr 81.5
0 5000 12.5
Ni
R N
0
0.000 5.000 10.000 15.000 20.000
Energy (keV)
(a) Pit Deposit
15k
Element
Al 2.2
Si 1.2
S 1.6
10
Cr 1.4
Fe 59.2
0 Ni 34.5
5000
0 L."
0.000 5.000 10.000 15.000 20.00C
Energy (keV)
(b) Deposit Above Pit

The effect of sulfur oxyanions was evaluated by adding various sulfur

compounds to a 1% CuC12 test solution inside an alloy 600 capsule. The
capsules were sparged with inert gas, sealed, heated to 320'F (160'C) for
three weeks, and then examined. The results of this test indicated that the
addition of sulfur oxyanion resulted in less severe pitting, in general, than
in tests with CuC12 only. There was no observed pH dependence associated
with pit depth.
The effectiveness of potential chemical inhibitors was evaluated by
conducting once-through autoclave tests at 320 and 536°F (160 and 280'C)
in pitting solutions containing the potential inhibitor. It appeared as though
the inhibitors (boric acid, silicate, and monobasic phosphate) were ineffective
at 320'F (160'C) with the possible exception of NaH 2 PO 4 . However, at
536°F (280 0C) boric acid, Na 2 SiO 3, and NH 4OH all produced less severe
pitting in cupric chloride than in tubes without the inhibitors present. These
tests did not assess the effectiveness of the inhibitor on retarding the growth
of existing pits but only on the reduction of pit initiation and pit propagation.
This result is somewhat different from the Combustion Engineering results,
in which new pits apparently initiated in the presence of boric acid.
The Battelle testing discovered two unusual morphological features
during metallographic examination of specimens from several tests. A
number of pits that were sectioned contained metallic filaments in the pit
deposit which were rich in nickel; however, the chemical analysis did not
correspond to that of alloy 600. Figure 10-30 illustrates the cross section
and analysis of one such pit. Because the filament size was small,
fluorescence of the surrounding oxide deposit might have occurred. Some
pits from Indian Point 3 had similar metallic filaments in the pit deposits.
The second metallographic feature which was unusual was the presence of
titanium carbonitride inclusions in the center of undercutting within the
pit. Figure 10-31 shows the appearance of a particle within an undercut
area and the EDS chemical analysis of that particle. The particle was
considered to be a titanium carbonitride inclusion. These inclusions are
common in alloy 600.
University of Leuven Test Results (10.32)

This project summarized the results of electrochemical and analytical
investigations concerning the corrosion of stainless and nickel-based alloys
in high temperature water. These investigations included a significant
amount of work related to steam generator corrosion damage and pitting
corrosion in particular. A variety of electrochemical techniques (e.g.,
potential dynamic, open circuit potential, etc.) were used to obtain data.
The specimens were examined at the end of the test using various surface
analytical tools (e.g., scanning Auger, Energy Dispersive Spectroscopy
[EDS], etc.).
Tube Pitting 10-51
FIGURE 10-30
SEM Photograph and EDS Analysis of a Pit Deposit Showing
Metallic Filaments in the Deposit (10.31)
Element w/o
Al 218
Si 1.6
CI 1.2
Cr 7.7
Fe 0.6
Ni 80.4
Cu 5.8
FIGURE 10-31
SEM Photograph and EDS Chemical Analysis of an Undercut Area
Within a Pit (10.31)
w/o
Element Area I Area 2
Al 3.4 3.0
S 1.8 -
CI 2.5 1.5
Ti 1.9 65.9
Cr 75.7 18.9
Fe 2.6 -
Ni 80.4 3.0
Si - 1.6
Nb 6.1
The University of Leuven discovered that the pitting and repassivation

potentials decreased as the temperature increased. Associated with this
relationship was the observation that the number of pits increased with
increasing temperature, but that the depths of the pits did not increase with
temperature. Thus, more lateral growth was observed with the pits as the
temperature increased. At temperatures above 392°F (200'C), alloy 600 tends
to experience general depassivation in aggressive environments rather than
pitting. Pit growth was found to be highest at intermediate temperatures
(302'F [150°C]) in accord with the results from Ohio State, Combustion
Engineering, and Battelle testing. Repassivation of the pits was easier at
temperatures above 392°F (200'C) than at the intermediate temperatures.
In addition to the extensive work on the temperature dependence of
pitting, the University of Leuven research made the following key
observations:
* The pitting potential decreased as the concentration of chloride
increased.
" The primary effect of oxidizing species was to raise the potential into
the zone where pitting occurs.
* The presence of sludge piles aggravate the pitting situation in an almost
uncontrollable manner.
" Intergranular attack on some pits may be related to the formation of
CrO 4 = or HCrO4- ions in the pit cavity.
" There is a passive pit inhibiting tendency in the solutions when pH is
around 10 (similar to Combustion Engineering and Battelle
observations).
" Sulfate may prevent pit initiation but accelerates growth of existing
pits.
* Phosphate is effective in preventing local breakdown of the passive
film at high temperatures, but it is not effective in arresting pre-existing
pits.
CAUSES OF PITTING
The results of the laboratory testing, field examinations, and published
data suggest a few phenomenological causative factors associated with
steam generator tube pitting: sludge, tube scale, oxidizing species, acid
chloride, and steaming conditions.
J
/
I
Tube Pitting 10-53
Sludge and Tube Scale

With a few exceptions, both nondestructive and destructive examination
field data have shown a correspondence between the location of pitting/
wastage and the location of sludge/scale above the tubesheet. Some pits
have been located above the first support plate or even on the free span
away from possible influence of sludge/tube scale. In general, the pitting
has been most severe and most extensive in the central parts of the steam
generator in the region corresponding to low flow and sludge accumulation.
Sludge can be formed after layup or during plant operation. Sludge
forms after layup as the result of exposure of open piping systems to air
during maintenance and testing and due to residuals from maintenance
activities. These corrosion products, primarily iron and copper oxides, are
transmitted into the steam generator during pipe line flushing and startup
and accumulate as sludge.
Thermal hydraulic conditions prevent most of this sludge from being
removed by steam generator blowdown; thus it accumulates until it is
physically removed. Sludge which forms during plant operation is
primarily the result of erosion corrosion and the ingress of contaminants in
the feedwater.
Scale can form because of the presence of scale forming elements/
compounds in the water which will plate onto the tube surface. Silicates
are the most common scale-forming compounds. The Indian Point 3 tubes
removed with the scale intact showed the presence of a silicon-rich layer
on the tube surface (Figure 10-8). Other scale-forming compounds are
commonly found such as calcium sulfate and magnesium-containing
precipitates. Sludge and scale can insulate the tube surface and can provide
a physical barrier which will allow aggressive species in the water to
concentrate next to the tube at the exclusion of protective species. Some
data (10.32) have shown that dryout may occur within 1-3 cm (1/2 to 1 inch)
below the top of the sludge, enhancing chemical concentrations. Bridging
of scales between tubes above the nominal sludge pile is commonly found
when sludge lancing has been ineffective.
Combustion Engineering and Battelle laboratory test data consistently
have shown that at high temperatures, (554°F [290°C]), prototypic pits form
in areas covered by sludge or in areas where sludge and corrosion deposit
accumulation would be expected. Non-sludge covered areas of the
specimen in the Combustion Engineering and Battelle tests had small non-
prototypic pits. Both Ohio State and Battelle produced pits on alloy 600 in
tests conducted without sludge or scale present. In these tests, the pits
were morphologically different than steam generator pits.
Oxidizing Conditions
There is limited direct evidence of oxidizing conditions in operating
steam generators. A high fraction of the sludge at Indian Point 3 and
Millstone 2 consisted of reduced copper and copper oxides. Copper metal
probably resulted because of electrochemical reduction of either copper
oxide or a complex form of copper. Based on thermodynamic data, Cu20
is stable at operating temperatures and CuO/Cu are more stable at low
temperatures (10.36). In addition to the sludge evidence, many units
operated with aerated auxiliary feedwater during hot standby
Air could also be introduced into the feedwater as a result of condenser
leakage.
Babcock & Wilcox (10.35) suggested that oxidizing conditions may have
developed locally in the steam generator after the crevice acidified below
the pH corresponding to the solubility of copper oxides. Once the pH was
acidic, the copper oxide dissolved and participated in the electrochemical
actions associated with scale deposition and pitting. Also, if the acid
condition developed as a result of bulk water chemistry, it would be difficult
to explain the Indian Point 3 in-service leak two days before a scheduled
shutdown when bulk water conditions should be reducing. However, local
chemistry conditions could result in sustained corrosion attack.
The C-E model boiler laboratory data clearly indicate that pitting was
related to the presence of oxidizing conditions in the water. Their model
boiler exhibited pit growth during exposure periods in which air was added
to the bulk water. Similarly, pits did not grow during periods in which the
bulk water was completely deaerated. In addition, the Battelle and OSU
test data showed that oxidizing conditions even with low levels of cupric
chloride resulted in pitting.
Acid Chloride
Babcock and Wilcox calculated high temperature Pourbaix diagrams
for nickel, chromium, iron, and copper (10.34). These diagrams identify
the domain of thermodynamic stability of the individual species and oxides.
The diagram for nickel, chromium, iron, and copper are shown in
Figure 10-32. By superimposing one diagram upon another, it is possible
to approximate the potential pH region in which certain species would
coexist. Then, by comparing the results of the destructive examination of
steam generator tubes, the conditions (potential and pH) consistent with
pitting can be estimated.
Table 10-6 summarizes the relationship between the location of corrosion
attack, observed chemical compounds,. and the inferred chemistry
conditions (derived from the Pourbaix diagram based on destructive
examination of two tubes removed from the hot leg at Millstone 2).
Tube Pitting 10-55
FIGURE 10-32
Pourbaix Diagram for Nickel, Chromium, and Copper in 550°F
(2800C) Water (10.24)
2.50
ti+ I I 1JN0(OH)3")
2,00 -
1.5o I-OH)
1.00.0
I0C50 (+ 041
-lH-4 " 0 40 1. C
-150 0 0-
- xx0.50-do)
2.50 -• . CO)
',C' ",C-2+ '

c.C.0
C.20
CuO.
,,, . . .
------------
------
o12 4
TABLE 10-6
Relationship Between Observed Chemical Compounds and Inferred
Chemistry
0
Corrosion Products Inferred Conditions
Location and Deposits pH, Potential (E. Volts)
Wastage band and pit Metallic nickel dissolves; pH<3.3, E>-0.3
interiors nickel oxide is absent
Tube surface and pit Chromium-rich oxides pH>1.0, E<line "b"( 1)
interiors including Cr(OH) 3 3H20
Tube surface, wastage Iron oxides pH>3, assuming
band, and interior of E - line "d"
pits that are near
the tubesheet
Interior of pits that Absence of iron oxides pH<3
are away from the
tubesheet
Tube surface and Metallic copper pH<4, E < -0.1
pit interiors
Tube surface and Nickel sulfide pH>2, E - line "a"
interior of pits that
are near the tubesheet
Interior of pits that Copper sulfide pH<2, E - line "d"
are awayfrom the
tubesheet
1Line "d" in the potential-pH diagrams is the hydrogen evolution potential as a

function of pH; line "b" is the oxygen evolution potential.
In general, Table 10-6 indicates that the pH needs to be between 1 and

4 and the potential between -0.1 and -0.3 volts (SHE).
Both Indian Point 3 and Millstone 2 had condenser leakage. The
chloride levels during the major Indian Point-3 feedwater transition in 1981
approached 400 ppm Cl-. In addition to a history of condenser leakage, the
affected units had experienced tube denting. Denting at seawater sites is
usually ascribed to the presence of acid chloride in the tube-to-tube support
plate crevice (10.34). Although the denting chemistry became sufficiently
aggressive to corrode the steel support plate, pitting has not been found on
steam generator tube surfaces beneath dents. This is probably due to
cathodic protection of the tube by the corroding steel.
Tube Pitting 10-57
Laboratory data developed by Combustion Engineering, Ohio State,

and Battelle showed that, while pits could initiate and propagate in a CuC12
solution having as little as 1 ppm CuC12, more concentrated solutions or a
concentrating mechanism were required to develop pits with reduced
copper present in the pit deposit. Thus, the solutions responsible for steam
generator pitting were probably a moderately concentrated acid chloride
solution.
Operating Temperature
The temperature at which pitting occurs in a steam generator has been
thoroughly discussed. Some have attributed steam generator tube pitting
to be a layup phenomenon occurring at a low temperature 176'F (80'C).
Others attribute it to be strictly an operating temperature phenomenon;
others, again, suggest that there are both low and high temperature aspects
to pitting.
The field data suggest high temperature propagation of pits. This is
derived from (1) through-wall propagation of a pit at Indian Point 3 while
the unit was steaming, and (2) eddy current data which show initiation
and apparent propagation of some pitted areas from cycle to cycle following
improvements in layup practice and water chemistry control. Steaming
conditions allow impurities in the water to concentrate, resulting in the
formation of more aggressive solutions. The laboratory data suggest that
pitting is an operating temperature phenomenon. This conclusion is based
on the test results at Ohio State, Combustion Engineering, University of
Leuven, and Battelle which have consistently found that the chromium-
rich oxide forms only at higher temperatures. Combustion Engineering
and Battelle test data reproduced metallic copper only in tests conducted
at high temperatures (5360F [280°C]). The deposit found at low temperatures
is usually friable and easily removed by descaling, even if it has been
exposed to high temperatures. On the other hand, deposits formed at high
temperatures (536°F [280°C]) have been found to be adherent, non-friable,
and difficult to descale, i.e., similar to steam generator pit deposits.
Theory of Steam Generator Tube Pitting

Several theories of steam generator tube pitting have been proposed
(10.8 j012 IM 17.IM2 j and 10.35). The various theories have
been derived from a combination of field data from destructive
examinations, analytical data used to generate high temperature Pourbaix
diagrams, and laboratory test data. However, the essential points of the
various theories are in basic agreement.
Location of Pitting. The location of pitting on steam generator tubing

seems to be determined by several competing reactions as follows: cathodic
protection from the steel tubesheet; steam blanketing due to boiling within
the sludge pile; anodic dissolution/pitting because of the presence of a
concentrated chemical beneath the scale/sludge; and reduction reactions
on the tube surface above the sludge/tube scale. Since the reactions compete
with each other, the elevation of pitting can vary from tube to tube because
of differences in the influence of each reaction on each tube. Figure 10-33
illustrates the various factors related to the location of pitting corrosion on
the steam generator tubes.
FIGURE 10-33
Relationship of Factors That Determine the Location of Pitting in a
Steam Generator (10.27)
Secondary Water
Impurities AVT or
'Wet' (e.g. Cl', Sox, Auxiliary
Wet rCa" ".i Crevice
Pit
Wet Wet/Dry
Wt Chemical Conc. L2ioPropaga-
C--e-v--- Anodic
.L.... Sludge--. ", Region

Scale -. ' ,
Tube .
Tube
Dry Sheet Cathodic
Non Steaming Primary Water

Steaming
Tc
Tube Pitting 10-59
Pit Initiation. Several explanations have been proposed for the conditions
leading to pit initiation. The electrochemical work performed by Ohio State
concerned pitting potentials close to the passive-to-transpassive transition
zone (Figure 10-16). The Brookhaven electrochemical tests determined both
a pit initiation potential and a pitting potential (continued pit propagation).
Pits which initiated at the pit initiation potential did not continue to grow
after initiation and repassivation. Most of the Ohio State and all of the
Brookhaven tests were conducted on scale-free surfaces and without the
presence of sludge. This is potentially an important difference since steam
generator pits are always found beneath scale/sludge. The pitting potentials
were found to be lower with increasing temperature and with increasing
concentrations of chloride (OSU testing). This shows that pits are most
likely to initiate at higher temperatures and in concentrated solutions.
Babcock & Wilcox suggested (10.35) that the pits initiated at potentials
close to the active-to-passive transition zone. This suggestion is based on
the competing forces vying for the location of pitting as previously
discussed. The protection effect of the steel tubesheet is considered to pull
the potential on the tube down towards active corrosion. This explanation
seems reasonable and helps to explain the presence of wastage on the hot
leg tubes. Wastage is close to the active corrosion peak (lower potentials),
whereas the pits are associated with local breakdown in the passive film
(higher potentials) instead of a general metal loss with wastage. The lower
potentials would also harmonize with the crevice corrosion concept of attack
associated with the sludge/scale occluded surface of the tube.
Pit Propagation. The plant eddy current data show that there is
considerable scatter in pit growth rates from cycle to cycle. In general, and
prior to the implementation of corrective actions, the pit depth continued
to increase during each cycle. There was some scatter, however. The
through-wall propagation of a pit at Indian Point 3 confirms that pits
propagate at operating temperatures. Earlier in the understanding of pitting,
it was thought that pit propagation may have been associated with startup
and shutdown cycles because of the layered nature of pit deposits. The
laboratory data showed that pits with multi-layered deposits could be
produced in one cycle.
The pitting phenomenon has been associated with cold leg steam
generator tubes while the wastage phenomenon has been associated with
hot leg tubes. It is possible that the phenomena developed separately and
independently from each other. For example, the corrective actions
implemented by the utilities may have changed the distribution of corrosive
chemical species in the steam generator so that they concentrated in the hot
leg sludge pile instead of in the sludge pile on the cold leg side. Or perhaps
corrective actions removed some of the oxidant present in the system

resulting in a decrease in corrosion potential. This effect could actually
increase the rate of corrosion attack because of the shift in potential from
passive to active corrosion. It is even possible that the hot leg phenomenon
always existed and was simply missed because of eddy current probe
insensitivity or sampling variability. Severe attack on the hot leg tubes is
limited to a few tubes and, therefore, less likely to be detected during a
routine inspection. In retrospect, it seems reasonable to expect that corrosion
attack should occur on hot leg tubes if it occurred on the cold leg.
A number of morphological differences have been observed on steam
generator tubes removed from service. Some of these differences have
apparently developed during the course of plant operation and may be
reflecting changes in the electrochemical potential, and thus movement into
a more or less passive state. For example, the presence of intergranular
spikes within pits suggests a modification in the corrosion mechanism
during plant operation. This modification may have resulted from
intentional plant modifications in the water chemistry or unintentional
modifications due to time-induced natural dilution of corrosive chemical
species. The change in concentration of chemical species in the deposit
might also result in a change in corrosion degradation morphology. The
presence of copper and nickel sulfides in the pit deposits, as well as high
sulfur concentration in intergranular spikes, suggest an increasing sulfur
concentration in the pits. The laboratory test data have clearly associated
sulfur with intergranular attack.
Figure 10-34 illustrates the various physical and chemical phenomena
associated with pit initiation and propagation. Figure 10-35 illustrates the
details of reactions occurring beneath the tube scale/sludge. This figure
shows the various morphological features associated with pit propagation
and the apparent relationship of morphology to causative species.
The rates of pit propagation based on the laboratory data are
summarized in Table 10-7. From this table it appears that pits go through a
period of rapid growth followed by a period of very slow or negligible
growth. Laboratory testing has shown that pit propagation can be reduced
by inspecting the test (dryout), raising the pH, raising the temperature,
and by deaerating the solution. The inhibitors which have been evaluated
were found to be ineffective in retarding pit propagation. Initially, the rate
of pit growth seems to be higher with concentrated solutions than with
dilute solutions.
The C-E testing has shown that it is difficult to re-initiate pits following
an inspection. In addition, the C-E and Battelle testing have shown that
pits generated by a prepitting procedure grow very little during subsequent
exposure to environmental conditions that are conducive to the initiation
and propagation of new pits. This information suggests that existing pits
Tube Pitting 10-61
FIGURE 10-34
Summary of Physical and Chemical Phenomena Associated With
Initiation and Propagation (Modified From Reference 10.13)
Localized boiling concentrates solution

under the tube/sludge scale interface.
I|
Hydrolysis, precipitation, volatilization,
and electrochemical processes cause
the concentrate to become acidic.
| |
Chlorides (sulfates) break down CuO from the sludge dissolves

passive oxide film at defect sites. in the acidic concentrate.
Cupric ions reduce to metallic cuprous

Alloy dissolves in pit.
copper on and near
the tube surface.
I
I
Charge imbalance drives cations
(Ni2+ and Fe2+) out of the pit and
attracts anions (Cl- and S04 2-) into the pit.
and H+ within
SHydr'olysis the pit
of Cr3+ forms lowers the pH in
and chromium-rich the pit.I
oxides
Pitting stifles when

• H rises
• onductivity rises
may lose the acidic chemistry necessary to continue pit propagation and
that re-exposure of the tube to pit-producing environments usually does
not re-establish pitting conditions in the original pits. It is not known why
this should occur. One possible explanation is that pit propagation is
integrally linked to pit initiation. Thus, if pit propagation were to be
impeded for any reason (as discussed above), then initiation would be no
more likely an event than the original initiation phase of the pitting. In
fact, re-initiation may be less likely, because the pitted sample has already
consumed the initiation sites. Pit initiation and propagation has been shown
to occur at titanium carbonitride inclusions. If the final shape of the pit
does not leave any exposed titanium carbonitride inclusions, then re-
initiation would be less likely. This explanation would also account for the
apparently rapid growth of pits, since, once a pit has initiated, the driving
force of initiation carries the pit propagation phase until the pit is stifled.
FIGURE 10-35
Summary of Reactions Associated With Initiation and Pit
Propagation Beneath a Scale/Sludge Deposit (10.27)
Non-Crevice Secondary Primary Water

Water
-02 + 2H 20 + 4.- - 40H- (Aerated)
Boiling
(Percolation)
Impurities C.• 2C-.C

(cull, CuCI211
Dffusion
- Cu O IH 2 0 C u + + + 2 0H-
H2 Thic n ss a
2H+ + 2e- Be t
Copper
eCoCI 2 + H+ + 2c. Cull + 2e- - Cu Layers Time
Crevice Anode"
-- M-- Ni++ + Fe++ + Cr+++ +- e-
Potential
Chloride Transport
0D lnto Pit CuCI2-
Out of Pit M l+
Scale - Tuben----
Tube Pitting 10-63
TABLE 10-7
Summary of Selected Laboratory Pit Depth Data As a
Function of Time
150 0C 280 0 C
Average( 1 )
1
Max Pit Average( ) Max Pit Growth
Time Depth Growth Rate Time Depth Rate
kh rs) mlIs (mils/day) t rs (milsI (mils/day)
4 5.6 34 4 4.4 26
8 6.3 19 7.5 4.7 15
16 8.5 13 15 5.4 9
33 11.6 8 31 7.7 6
40(3) 12.6 7.5 118(3) 12.6 2.6
160(3) 16.6 2.5 1056(2) 30.2 0.7
1056(2) 16.4 0.4
2066(2) 26.3 0.3
1
Growth rate normalized to time zero.
2
Pot boiler data. 1-0 contained CaCI2. 2-M contained no CuCI2.
3
Battelle data
Characteristics of Units Most Likely to Develop Pitting

Based on the field and laboratory data accumulated to date, units with
the following characteristics are considered most likely to develop pitting
(ranked in approximate order of importance):
" Units with a permanent copper bearing sludge pile.
* Units with copper-base alloys in the secondary water system.
* Units with a large sludge pile.
" Units with a history of air inleakage.
* Units which use aerated water during hot standby.
" Units which have had dry layups.
* Units with a history of condenser inleakage or units with a major
condenser leak.
" Units cooled with brackish water or seawater.
* Units with active denting.
UTILITY-IMPLEMENTED CORRECTIVE ACTIONS

Major pitting of alloy 600 steam generator tubing has caused the New
York Power Authority and Northeast Utilities to implement a number of
corrective actions. Other affected utilities have taken similar steps. The
initial corrective actions consisted of relatively simple fixes such as reducing
condenser inleakage. Subsequent actions included secondary side chemical
cleaning (Millstone 2) and the installation of full flow condensate polishing
and steam generator replacement (Indian Point 3). Sleeving has also been
used extensively to allow tubes with pits to remain in service. The utility-
implemented/planned corrective actions and the rationale for the actions
are summarized in Table 10-8. This table shows those actions which are
primarily of an equipment, procedural, or maintenance nature.
It is very difficult to quantify the benefit of any given corrective action
on continued pit propagation. However, it is possible to quantify the effect
of chemical cleaning on eddy current detection of pitting. Figure 10-36
illustrates eddy current traces on pitted tubes before and after chemical
cleaning at Millstone 2. It is clear from these traces that the signal is much
less distorted after chemical cleaning than before chemical cleaning. Much
of the noise in the pre-chemical cleaning signal was due to interference by
copper present in the sludge and tube scale. Because of the increased
sensitivity of the eddy current probe, new flaws which had not been detected
previously were detectable after chemical cleaning. The number of tubes
with 40% through-wall defects and the associated voltage signal are shown
in Figure 10-37.
The laboratory and field data suggest that pit propagation may occur
in a few days. Thus, once the necessary conditions are present, pit
propagation may occur quickly. This suggests that it may be more prudent
to attempt to prevent pit initiation rather than pit propagation. A number
of the corrective actions implemented by the utilities are directed toward
preventing or eliminating conditions leading to pit initiation.
RECOMMENDATIONS TO PREVENT/MITIGATE
STEAM GENERATOR TUBE PITTING
The utilities have already implemented those actions (see "Utility-
Implemented Corrective Actions" above) which are likely to have the most
beneficial effect for preventing the formation of new pits or for mitigating
the consequences of existing pits. The corrective actions can be classified
according to the relationship to the causative factor associated with pitting,
i.e., sludge/tube scale, acid chloride, oxidizing conditions, and temperature.
Tube Pitting 10-65
TABLE 10-8
Summary of Utility-Implemented Corrective Actions for Pitting
Corrosion
Corrective Action Rationale Behind Action

Equipment
Steam Generator Replacement Performance of original SGs
considered unreliable; therefore,
replace SG to remove the problem.
Retube Condenser With Titanium Eliminate major source of copper for
transport into the steam generator;
provides more reliable condenser
tubing.
Retube Feedwater Heaters With 439SS Eliminate major source of copper for
transport into the steam generator.
Retube Moisture Separator Reheaters Eliminate source of copper for transport
into the steam generator.
Install Makeup System Deaerators Remove dissolved oxygen from
demineralized water to reduce the
oxidizing tendency of the water.
Install Blowdown Recovery System Minimize requirements for makeup
water and therefore possible
contamination of the water.
Install Condensate Startup Loop Clean condensate water of particulates
to minimize transfer of sludge, formed
during an outage, into the steam
generator.
Install Condensate Polishing System Purify secondary water and remove
contaminants before they enter the
steam generators.
Install Condensate Sludge Collection Remove sludge and particulate
System material before itenters the steam
generator to minimize the formation of
tubesheet sludge.
Install Filters in the Water Makeup Plant Same as above.
TABLE 10-8 (cont'd)

Summary of Utility-Implemented Corrective Actions for Pitting
Corrosion
Corrective Action Rationale Behind Action

Procedural
Implement SGOG Water Minimize damage to steam generators
Chemistry Guidelines by requiring specific corrective actions
depending on the severity of the
out-of-specification condition and the
reactor power.
Improve Secondary Water Chemistry Minimize ingress of chemical
impurities by refinement in chemistry
limits for aggressive chemicals.
Improve Steam Generator Minimize exposure of the secondary
Layup Procedure side of the steam generator to air,
maintain reducing condition on
secondary side.
Maintenance
Chemical Cleaning Reduce the inventory of sludge/scale
by chemical dissolution. (Millstone 2
used a modified version of the EPRI
chemical cleaning solution.)
Sleeving Increase the useful life of the tube by
"increasing" the wall thickness required
to perforate the tube (Indian Point 3
used thermally treated Alloy 600 while
Millstone 2 used a bimetallic Alloy 625
OD and Alloy 690 ID).
Increase Eddy Current Inspection Increase the likelihood of identifying
Scope and Frequency defects which should be sleeved or
plugged before an additional cycle.
Sludge Lancing Reduce the inventory of sludge.
Maintain Condenser Leak Tight Minimize the introduction of species
which might become acidic in the
steam generator.
Maintain Condenser Air Tight Minimize the introduction of species
which are or might become oxidizing.
Tube Pitting 10-67
FIGURE 10-36
Eddy Current Traces of Millstone 2 Tube Before and After Chemical
Cleaning (10.9)
- -r - - - - -i
04 ~ WET ~ 04 M 3 WE?. wH:N. "a - I ; P UN kLma ppIt
.mw " 6 . ..... 1
Pai-r CHU-M CS.EAm 0Owo- -
MW- 4w~
ww- ISO
On.w - N v
MM- " S
CSfl4 - as3
CT . US I b h3
RMNI -$1CM
IM- O.ECT (2)

* 4 of H4. 6
,,I25S3V9p
9FL 4 19m
req 1 2 3 4 5 6 7 8
600 5
100 I I t6i
I
FIGURE 10-37
Eddy Current Indications 40% Detected After Chemical Cleaning at
Millstone 2 (Indications were not identified prior to the cleaning.)
(10.9) 0
400
346
(I)
a)
300
~
0 242
U" 200 186
E
z
100
I 67
0-1 1-2 2-3 3-4

Volts
Sludge and Scale

Sludge should be removed regularly. Sludge and/or tube scale allow
chemical impurities to concentrate in the water, occlude the surface from
bulk water chemistry additives, and enable concentration of chemical
impurities in the dryout region. Sludge and tube scale can be removed by
chemical cleaning or by sludge lancing for existing steam generators.
Consideration should be given in the design of new or replacement steam
generators to eliminate low flow regions in the steam generator, or
alternately improve accessibility to central tubes for cleaning using
mechanical devices.
Acid Chloride
Plant operations should be suspended until leaking condensers are
repaired. This applies particularly to brackish water or seawater-cooled
units. The condenser cooling water contains impurities (e.g., chloride,
sulfate) which can acidify in the steam generator, and which have been
shown to be related to pitting. After condenser leakage incidents, hot soaks
with deaerated water should be performed until hideout return is negligible.
Tube Pitting 10-69
Feedwater purification equipment should be carefully maintained to

minimize resin bead carryover. Cation resin beads have been shown to
cause shallow pitting of alloy 600 in laboratory testing. These resin beads
are unstable at steam generator operating temperatures and will decompose
if exposed to hot water. Ion exchange resin beads containing sulfate ions
from regeneration chemicals and sulfonate radicals would be released as
acid into the generator.
Oxidizing Conditions
Oxygen ingress to the steam generator must be kept at a minimum to
prevent pitting. One of the most common sources of oxygen is the entry of
aerated auxiliary feedwater during hot standby when hydrazine residuals
are low. This allows oxygen to enter while the generator is at intermediate
temperatures when pit initiation is easiest. Use of deaerated feedwater is,
therefore, recommended for use in hot standby and other layup conditions.
Copper oxide and/or copper cations are oxidizing agents and have a similar
effect to oxygen. Replacement of copper-based alloys with more corrosion
resistant alloys such as stainless steel or titanium will eliminate a major
contributor to pitting corrosion.
Temperature
Pitting is considered to be a high temperature phenomenon. However,
there is laboratory data which suggest that pit initiation and propagation
may be easier at intermediate temperatures. Thus, steam generators should
be heated up and cooled down as rapidly as possible (consistent with plant
hideout return requirements) with good quality water. The likelihood of
pit initiation and propagation is further enhanced during these intermediate
temperature periods because of poor water chemistry control. However,
fast heatup encourages hideout and fast cooldown limits hideout return. It
is extremely important that blowdown purity be within the water chemistry
guidelines before heating or cooling the generator; that is, the combination
of poor water chemistry and intermediate temperatures makes the
conditions for pit initiation greater than either factor considered alone.
Miscellaneous Recommended Actions
Layup Chemistry. Secondary water layup chemistries and procedures

should be used to minimize the entrance of chloride and air into the steam
generator. The pH should be kept high by recirculation and N 2 H 4 /NH 3
addition. Recirculation during layup can aid in particulate impurity
removal. Laboratory data show that pitting can. initiate at layup
temperatures in the presence of faulted water conditions. In particular,

copper chloride and air have been shown to be consistently detrimental to
the pitting resistance of alloy 600 tubing. Repeated flushing and filling
(deaerated water and inert gas blanketing) will remove impurities from
the crevice regions. Introduction of air can be effectively mitigated by
maintaining the steam generator in wet layup with deaerated water
recirculation. Regular and frequent sampling of the water recirculation
line will ensure detection of high impurity levels. Planning for maintenance
in the steam generator should specifically consider minimizing the time
the steam generator is in a non-wet layup condition. Draindown should be
eliminated except when considered necessary (i.e., sludge lancing,
secondary-side maintenance, etc.). Nitrogen blanketing should be used to
the maximum extent possible during layup and when drained down.
Sludge lancing and eddy current testing should take place with a nitrogen
overpressure.
Eddy Current Inspection. Eddy current testing should be performed

regularly on hot and cold leg tubes in the center of the tube bundle between
the tubesheet and first support plate using techniques proven to reliably
detect pitting. Plants which have seawater or brackish cooling water are
most susceptible and should, therefore, implement this recommendation
to identify the presence or absence of tube pitting and wastage so that
corrective actions can be taken to mitigate the consequences of tube
degradation.
Sleeving. Sleeving should be considered as a delaying or preventative

measure to stop through-wall penetration by active pits. Sleeves are installed
inside the pitted tubes and provide a backup barrier between the primary
and secondary fluids., The sleeve may arrest further pit growth once a pit
perforates the steam generator tube wall, or it may merely retard continued
pit growth through the tube (plus sleeve) wall. Sleeves may be made of
one alloy or a bimetallic alloy. In bimetallic sleeves, the OD alloy is selected
for pitting corrosion resistance, and the ID alloy is selected for primary side
corrosion resistance. Selection of the sleeve materials and the installation
procedure should be carefully reviewed to ensure that induced stresses in
the steam generator tube from the sleeve expanding procedure do not result
in primary side stress corrosion. Additionally, life cycle testing is needed
for estimation of time-to-through-wall penetration for sleeved tubes.
Tube Pitting 10-71
Tube Examination. Utilities considering removal of a tube for destructive

examination should consider the following:
* Remove tube through secondary side hand hole in order to retain tube
scale intact.
* Remove the tube adjacent to the tube with a defect indication to confirm
that an indication is truly unique to a given tube.
* Conduct inspection following tube removal of the sludge, scale, and
surrounding tubes to determine sludge profiles and the presence or
absence of scale, etc.
* Collect and analyze liquid samples from the sludge immediately above
the tubesheet.
• Collect and analyze a sludge sample immediately above the tubesheet.
* Conduct eddy current test after the tube has been removed for
comparison with in situ eddy current inspection test results.
• Carefully document the orientation of the tube in the steam generator
and mark individual segments as they are removed. Maintain careful
documentation throughout the examination.
REFERENCES
10.1 , Agrawal, A. K., J. F Sykes, W. N. Stiegelmeyer, and W. E. Berry.
"Pitting of Alloy 600 Steam Generator Tubes at Indian Point 3." In
Proceedings of the International Symposium on Environmental
Degradationof Materials in Nuclear Power Systems-Water Reactors,
Myrtle Beach, South Carolina, NACE, August 22-25, 1983, p. 241.
10.2 Fackelmann, J. "Pitting Experience With Millstone 2 Steam
Generators." Paper presented at EPRI/SGOG Contractors Review
Meeting, Scottsdale, Ariz., November 7-8, 1983.
10.3 Aspden, R. G. "Examination of Steam Generator Tube R12 C46
From Indian Point 3." Westinghouse Report SG-82-07-024. July
13, 1982.
10.4 Nilsen, N. and J. Gillen. "Pitting Experience at Indian Point 3."
EPRI/SGOG Pitting Workshop, White Plains, New York, November
20-22, 1985.
10.5 Anderson, F "Millstone 2 Operating Experience." Op. cit. Reference
10.4.
10.6 Mc Inteer, W. A. "Calvert Cliffs Unit I Tube Examination." EPRI
NP-4904. November 1986.
10.7 Ferris, R. H. "NDE Validation Studies." Battelle Northwest, Steam
Generator Group Project, Task 13.
10.8 Agrawal, A. K., W. N. Stiegelmeyer, J. F. Sykes, and W. E. Berry

"Evaluation of Steam Generator Tube R12C66 From Indian Point 3."
EPRI NP-3029. May 1983.
10.9 Edwards, L. J., and J. P. Lareau. "Detection of Steam Generator
Tube Pitting in the Presence of Secondary Side Copper." Op. cit.
Reference 10.4.
10.10 Nealy, C. L. "Analysis of Sludge From Indian Point 3." EPRI
NP-3027.
10.11 Habicht, P. "Destructive Examination of Steam Generator Tubes
With Pits From Millstone 2." Paper presented at EPRI SGOG
Workshop on Pitting in Steam Generators, Pittsburgh, PA, October
28-29, 1983.
10.12 Hall, J. F "Destructive Examination of Steam Generator Tubes
Removed From Millstone Point Unit 2." EPRI NP-4737-LD. July
1986.
10.13 Inman, S. C. "Examination of Steam Generator Tube Sections From
the Millstone Point Unit 2 Nuclear Power Plant." EPRI NP-4829-
LD. October 1986.
10.14 Gosselin, A. "Connecticut Yankee OperatingExperience." Op. cit.
Reference 10.4
10.15 Powell, D. E. "Destructive Examination of Steam Generator Tubes
From Trojan Nuclear Plant 1986 Refueling Outage."
10.16 Paine, J. P. N. "Destructive Examination Results From the Retired
Steam Generator at Battelle." Paper presented to the Steam
Generator Reliability Project Technical Advisory Group, Tampa,
Florida, January 20-22, 1987.
10.17 Battelle Northwest, Steam Generator Group Project, Task 4, Status
Report. May 1987.
10.18 Green, S. J. and J. P. N. Paine. "Materials Performance in Nuclear
55, no. 1 (October 1981): 12-13.
10.19 Weeks, J. R. "Corrosion of Steam Generator Tubing in Operating
Pressurized Water Reactors." Corrosion Problems in Energy
Conversion and Generation, Electrochemical Society, Princeton, N.J.
(1974): 322-345.
10.20 Fletcher, W. D., and D. D. Malinowski. "Operating Experience With
Westinghouse Steam Generators." Nuclear Technology 28 (March
1976): 356-373.
10.21 Pathania, R. S., and E. G. McVey. "The Effect of Cooling Water
Leakage on Corrosion of Steam Generator Materials." Materials
Performance 19, no. 7 (July 1980): 34-41.
Tube Pitting 10-73
10.22 King, P. J., and D. P. Dautovich. "Pitting Corrosion of Nuclear Steam

Generator Materials." Nuclear Technology, pp. 196-206.
10.23 Hickling, J., and N. Wieling. "Electrochemical Investigations of
the Resistance of Alloy 600, Incoloy 800, and 347 Stainless Steel to
Pitting Corrosion in Faulted PWR Secondary Water at 150-250'C."
Corrosion 37, no. 3 (March 1981): 147-152.
10.24 Stellwag, B., W. Beyer, and N. Wieling. "Influence of Steam
Generator Water Treatment, Temperature, Cl-, and 02 Content on
the Pitting Performance of Alloy 800." Paper presented at the
International Conference on Corrosion in PWRs, Myrtle Beach, S.
C., August 1983.
10.25 "Workshop: Pitting Corrosion in Steam Generators," edited by M.
J. Angwin. EPRI NP-3574-SR. October 1984.
10.26 Whyte, D. D., and R. C. Burchell. "Laboratory Program to Examine
Effects of Layup Conditions on Pitting of Alloy 600." EPRI NP-3012.
April 1983.
10.27 Sykes, J. F., A. K. Agrawal, and W. E. Berry. "Pitting Corrosion of
Alloy 600 Steam Generator Tubing: Results of Laboratory Scoping
Study." EPRI NP-3905. February 1985.
10.28 Smialowska, S., and J. R. Park. "Steam Generator Tubing: Pitting
Corrosion." EPRI NP-5037. February 1987.
10.29 Hall, J. F., R. S. Frisk, K. E. Marugg, A. S. O'Neill. "Investigation of
Causes of and Corrective Actions for Pitting in Steam Generator
Tubes Prototypical Tests." Research Project S-308-4.
10.30 Isaacs, H. S., et al. "Localized Electrochemical Corrosion of Nickel-
Base Alloys." EPRI NP-4754. September 1986.
10.31 Sykes, J. E, and A. K. Agrawal. "Pitting in Steam Generator Tubing:
Causes and Corrective Actions." EPRI NP-5207. June 1987.
10.32 Bogaerts, W. E, and C. Bettendorf. "Electrochemistry and Corrosion
10.33 Curlee, N. J., and A. J. Baum. "Single-Tube Thermal and Hydraulic
Tube Support Test." EPRI NP-2046. September 1981.
10.34 Mann, G. A. "The Effects of Oxygen, Copper, and Acid Chloride
on Denting Corrosion." EPRI NP-4648. July 1986.
10.35 Daniel, P. L., and S. L. Harper. "Use of Pourbaix Diagrams to Infer
Local Pitting Conditions." EPRI NP-4831. October 1986.
10.36 Cobble, J. EPRI Contractors' Review Meeting, Columbus, Ohio,
October 1984.
11
TUBE SUPPORTS AND

TUBESHEET CORROSION-
ALKALINE DENTING
J. P. N. Paine, EPRI
K. D. Stavropoulos, Dominion Engineering,Inc.
M. J. Partridge,Dominion Engineering,Inc.
1993
1 1-2 Steam Generator Reference Book, Revision 1
INTRODUCTION
A number of PWR plants have reported cases of steam generator tube
denting that are believed to result from alkaline crevice concentrations.
Denting occurs because corrosion products produced in crevices have larger
volumes than the steel they replace. As a result, the corrosion products
exert pressure on the tube and it becomes deformed.
Even small amounts of denting can lead to small plastic strains and to
yield level stresses in tubes, which, in turn, can lead to primary and
secondary side stress corrosion cracking (SCC). Therefore, it is important
to understand the mechanisms involved in alkaline denting and to take
appropriate steps to minimize its occurrence.
Denting caused by acidic conditions occurred at numerous plants and
became a serious problem in the mid to late 1970s. Hot leg tube supports
were mainly affected, with infrequent and minor denting at the hot leg
tubesheet. Denting attributed to alkaline environments, on the other hand,
has occurred mainly at the top of the hot leg tubesheet, with lesser amounts
of denting at lower hot leg tube supports.
PLANT EXPERIENCE
Denting in PWR Plants

Alkaline conditions are suspected to have caused carbon steel corrosion
and denting at Fort Calhoun, D.C. Cook Units 1 and 2, Point Beach Units I
and 2, and at San Onofre Unit 1, and at other freshwater-cooled sites.
Fort Calhoun. This unit which is located on the Missouri River uses a
once-through cooling system. Some of its tube supports have been corroded
and caused tube denting. In 1984, a tube rupture occurred at a dent in a
horizontal run of a tube at the top of the steam generator where the tube
contacted a vertical support strap (11., 11.2). The rupture was attributed
to caustic-induced intergranular stress corrosion cracking (IGSCC).
Analyses of impurities in Missouri River water indicate that alkaline
conditions occur in crevices; thus the denting was attributed to the alkalinity
of the cooling water (11.1. 11.2).
D.C. Cook Units 1 and 2. These units are located on Lake Michigan and
use a once-through cooling system. Initially, Unit-I was operating on
phosphate water chemistry, while Unit 2 has used only all volatile treatment
(AVT). During the early years of operation, condenser leaks were fairly
Tube Supports and Tubesheet Corrosion - Alkaline Denting 11-3
frequent in both units. Analyses of the effects of the impurities introduced

by inleakage of Lake Michigan water indicate moderately alkaline crevices.
Unit 1 started commercial operation in 1975. Denting at the top of the
tubesheet was reported at the first in-service inspection in 1976 (11.3). Unit 2
has experienced denting at the top of the tubesheet and at the first tube
support. Both units have seen intergranular attack/stress corrosion cracking
(IGA/SCC) in tubesheet crevices. Examinations of removed tubes from
these units indicate that the IGA/SCC and denting occurred under alkaline
conditions.
Point Beach Units 1 and 2. These units are located on Lake Michigan and
use once-through cooling. Initially, both units operated on phosphate water
chemistry. During the early years of operation, Unit 1 experienced
significant condenser inleakage. Unit 2, on the other hand, is reported to
have had relatively minor condenser inleakage (11.4). Both units have
experienced minor amounts of denting, which restricted the movement of
eddy current (EC) probes inside the tube and, in a few cases, led to tube
plugging (11.4). Eddy Current Testing (ECT) indicates that the dents have
been at support plates, rather than at the tubesheet. As noted for Cook
Unit 1, analyses of the effects of Lake Michigan impurities indicate that
they result in moderately alkaline conditions. Similarly, evaluation of
leachate from tubesheet crevices at Point Beach indicate that the impurities
in the crevices were moderately alkaline forming (11.5). Examinations of
removed tubes have indicated that caustic conditions caused, or at least
initiated, IGA/SCC in tubesheet crevices (11.5, 11.6). However, it is possible
that caustic conditions were present during an early phosphate water
chemistry period, during which IGA/SCC initiated, and that the IGA/SCC
then propagated under less alkaline conditions during AVT water chemistry.
San Onofre Unit 1. This unit is located at a seawater site and uses once-
through cooling. Although most plants switched to AVT in the 1970s, San
Onofre has remained on phosphate water chemistry. The unit has
experienced some condenser inleakage, and has occasionally added sodium
hydroxide to neutralize acidic conditions caused by sea water ingress. Unit 1
has experienced denting at the lower tube support plates 11(.7), and denting
and IGA/SCC at the top of the tubesheet. It is suspected that denting was
the result of alkaline conditions associated with the use of sodium hydroxide.
Summary of Plant Experience

IGA/SCC has often occurred at plants believed to have alkaline
tubesheet crevices. Occasionally, denting has occurred in the same plants,
most often at the top of the tubesheet, but also at tube support plates. Thus,
plant experience indicates that alkaline conditions can cause both IGA/
SCC and denting.
The concentration of alkalinity required to induce alkaline denting and
IGA/SCC is not clear. Evaluations of deposits and oxides at tube surfaces
and in cracks of removed tubes sometimes indicate that strong alkaline
conditions (e.g., pHT >10) were present in the areas experiencing IGA/
SCC. However, more often these evaluations indicate that mild alkaline
conditions were present (e.g., pHT -7); however, such low concentrations
have not been observed to cause denting or IGA/SCC in the laboratory.
Similarly, MULTEQ predictions of crevice conditions, which are based on
cooling water composition (e.g., Lake Michigan water) and hideout return
data, indicate that crevice conditions are sometimes strongly alkaline, but
more often relatively mild. Thus, plant experience tends to suggest that
denting and/or IGA/SCC may occur in plants under less strong alkaline
conditions than observed in the laboratory.
Possible Interaction Between Alkaline Denting and IGA/SCC

As discussed in Chapter 13, the occurrence of IGA in caustic
environments is moderately accelerated by increased stress, which raises
the potential for IGSCC. The slow continuous application of plastic strains,
with stresses at the work-hardened yield level, may be especially severe.
For example, the results of constant extension rate tests (CERT tests) of one
heat of alloy 600 in pure water environments showed critical strain rates
above which SCC was not observed; below this level SCC increased as strain
rate decreased (11.8). At 600'F (315 0C), the strain rate below which SCC
occurred was about 5 x 10- 8 /sec. Thus, caustic crevice environments could
potentially lead to severe IGA and IGSCC, since caustic can cause denting,
which applies high stresses and slow strain rates to the tubing in an
environment conducive to IGA/SCC.
Despite the potential for simultaneous occurrences of caustic denting
and IGA/SCC, it has not been a widespread problem. Combined alkaline
denting and IGA/SCC has been observed in one well known case, Fort
Calhoun, but could possibly have been involved in several other cases.
Based on this experience, laboratory test results, and theoretical
considerations, it is judged that, if alkaline denting occurs, IGA/SCC
aggravated by the denting is a real possibility. Thus, the possibility that
denting may accelerate the potential for IGA/SCC provides an additional
incentive to prevent strongly alkaline conditions from developing in
crevices.
CAUSES AND MECHANISM OF ALKALINE

DENTING
Alkaline denting, similar to acidic denting (Chapter 8), occurs when
local electrochemical conditions encourage iron or steel corrosion and the
formation of non-protective magnetite. These conditions were reviewed
by the Central Electricity Generating Board (CEGB) (L1.9). As discussed in
that report, non-protective magnetite can form in alkaline conditions.
Figure 11-1, adapted from (11.10), is a pH-potential plot for iron corrosion
at 572°F (300'C). The figure indicates that magnetite is not stable at a pHT
over about 10 for fully deaerated conditions. This implies the possibility
FIGURE 11-1
Potential pH Diagram for Fe-H20 at 572 0 F (3000 C)
(Activity of ions = 10-6, of gases = 1, of solids = 1)
2.O.
1.0*
, "' - ,ci.n
0..
as0
Oxide
VX..I - ,-Formation
Denting
&/
i In 12 i .4
pH
that the oxide film breaks down at about the same pH, which leads to non-
protective magnetite, high corrosion rates, and denting. This type of
corrosion followed by denting can occur in the absence of oxidizing
conditions, i.e., it can occur under reducing conditions of the type normally
achieved in PWR steam generators. Occurrence of denting under alkaline,
non-oxidizing conditions has been demonstrated in several laboratory tests,
as described later in this chapter.
Corrosion kinetics of carbon steel and chromium steels are illustrated
in Figures 11-2 and 11-3 (11.9). These figures indicate that certain trends
are connected with alkaline corrosion of carbon and low alloy steels. The
rate of corrosion increases as the concentration of caustics becomes elevated.
This is expected to encourage the rate of denting in tubesheets, as compared
to tube supports, since the available superheat, and, hence the equilibrium
concentration of caustic, is higher in tubesheet crevices. Furthermore, as
the temperature increases, so does the rate of corrosion, which will affect
the amount of denting in tubesheet crevices.
The rate of corrosion is higher in alkaline environments for the low
chromium alloy steels in tubesheets (such as SA508 CI.2) than for the carbon
steel used for tube support plates. This may be a factor in the increased
incidence of dents at tubesheet level. The behavior is different in acidic
environments where low alloy steels, which have a higher chromium
content, are more resistant to the formation of non-protective magnetite
and denting than carbon steels (11.11).
Another fact to note from Figures 11-2 and 11-3 is that the rate of
corrosion of steels in caustics is quite rapid. For example, at 572°F (300'C)'
and 13% NaOH, Figure 11-2 indicates that carbon steel corrodes at a rate of
30 jtm (1.2 mils) (about 24 mg/cm 2 ) per 1,000 hours. Assuming that the
oxide volume is about twice that of the metal corroded, this rate of corrosion
indicates that it would only require about a year to cause enough corrosion
to fill a typical tube-to-tubesheet or tube-to-tube-support crevice gap of
250 4xm (10 mils). Thus, fairly short times under caustic conditions have
the potential of leading to denting and IGA or SCC.
Other pertinent considerations regarding corrosion conditions in
crevices include:
Hideout of sodium and potassium is more efficient than that of anions
since acids such as HC1, H 2 SO 4, etc., are more volatile than alkaline
compounds such as NaOH and KOH. Thus crevices tend to become
0
alkaline as superheat increases, which is more often the case for
tubesheet crevices than for tube support crevices, since the superheat
is higher in tubesheet crevices.
FIGURE 11-2
Corrosion Kinetics of Carbon Steel in NaOH Solution
60
EO5
CD)
E 40
Oxidation of Carbon Steel
S30 In 13% NaOH
- 250'C
6 20 o 280*C
LL * 300 *C
o 340 C
0 x 3550C
Plotted Against Square
Time, h Root Time
(a) Temperature Dependence
60
50 Oxidation of Carbon Steel at

E 340°C In Aqueous NaOH
0) 40
E * 5%
30 o 13% By Weight
x 20%
'0
20
0
Plotted Against Square
ai) Root Time
LL I0
Time, h
(b) NaOH Dependence
, 50
E
Distribution of Oxidized
'CQM40
E Iron After Oxidation of
Carbon Steel In 13% NaOH
at 340°C
:2 20 Plotted Against Square
Root Time
0
(2) 10
LL
Time, h
(c) Fe Distribution
FIGURE 11-3
Corrosion Kinetics of Cr Steels in 15% NaOH at 316 0 C
70 -
60 .
so
E
S30 -
.-
20
I0
Time, h
Because tube support crevices are relatively shallow, it is expected that

impurities will flush out more easily in this location in a flushing
application than in tubesheet crevices. Thus, the accumulation of
impurities in tubesheet crevices is expected to be more certain and
continuous than in tube support crevices.
The concentration processes of impurities taking place in crevices of
PWR steam generators are described in Chapter 5. If the impurities
introduced into the steam generators are such that this concentration results
in strongly alkaline conditions developing in crevices, then there is a
potential for alkaline denting to occur, as well as other forms of corrosion
such as IGA/SCC. Experience indicates that the potential for alkaline
denting is highest in units with tubesheet crevices. This may be because
tubesheet crevices collect more impurities than support plate crevices since
they are harder to flush out, operate at higher temperatures, have crevices
with higher available superheats, and have more susceptible (to alkaline
Tube Supports and Tubesheet Corrosion -Alkaline Denting 11-9
corrosion) low chromium alloy steels. Alkaline denting can occur in drilled
hole support plate crevices and in crevices where strip-type supports contact
tubes. It is believed that this possibility is increased if the support material
is carbon steel, as in earlier steam generators, rather than the stainless steels
used in later steam generators since, as shown in Figure 11-3 (11.9), corrosion
rates of stainless steels are much lower than those of carbon steels. However,
tests have not been performed to conclusively determine whether alkaline
denting could occur with stainless steel support materials.
LABORATORY TEST RESULTS

The occurrence of alkaline-induced denting has been demonstrated in
several test programs. Table 11-1 summarizes tests using capsules, model
boilers, and similar devices (11.12-11.21). The table includes tests where
denting occurred, as well as tests where caustic conditions did not initiate
denting. Table 11-2 covers the same tests, and adds some additional
information regarding heat transfer conditions and the use of hydrazine.
The most notable observations are as follows: (a) Significant denting
occurred in several tests with known, strong alkaline conditions. (b) A
variety of alkaline environments, such as sodium carbonate, sodium
hydroxide, mixtures of sodium carbonate and sodium hydroxide, sodium
phosphate, mixtures of sodium and potassium hydroxide, and mixtures of
sodium, potassium, and sodium carbonate, tested favorably. (c) Denting
was detected in as little as 32 days, with typical materials and temperatures,
and possible chemical concentrations. (d) In some tests, IGA and SCC were
observed along with denting (11.12), although a cause and effect relationship
between denting and IGA/SCC was not documented. (e) Some sludge
ingredients, especially chromium oxide, appear to accelerate alkaline
denting. (f) For some of the test cases, it is possible to determine why denting
did not occur; in test 3 of Table 11-1, for example, calcium hydroxide was
present and apparently prevented the development of strongly alkaline
conditions. In other cases, it is more difficult to explain the lack of denting
(e.g., tests 7 through 10 in the same table). There are a number of speculations
as to the causes.
In some cases, the test duration was kept short because aggressive
conditions caused tube leaks. This could be the reason why denting did
not occur, Tubesheet Crevice Simulation Apparatus (TCSA) tests (test 8)
using strong caustics, while denting was observed during earlier tests in
the same device but with milder alkaline conditions (test 5).
TABLE 11-1
Alkaline Denting Tests
Type of Temp. Tubesheet Radial Crevice Feed T-S lime Dent/Bulge

TeFl Device ýE Mateidal Gain. Prelad
C•
Chemicals Sua Dayýs mils Reference
la Single Tube 620 Pr. 4130 0.008 none Na2CO3.06 - 1.7 ppm SPS or Fe304 32-91 0.2 to 3.5 WCAP-10273
Model Boiler 520 Sec. Feb. 1983
0
(O
lb Single Tube 620 Pr. 4130 0.008 none NaOH .3 - 1.0 ppm SPS 78 7 WCAP-10273
Model Boiler 520 Sec. Feb. 1983
CO
1c Single Tube 620 Pr. 4130 0.008 none Na2CO3.06 - 1.7 ppm + With & w/o SPS 60-74 0.6 to 6 WCAP-10273
Model Boiler 520 Sec. NaOH .3- .5 ppm Feb. 1983
("
2a Inconel 600 SA508 0.0125 40%NaOH+10%KOH* none none 270 5 NP-3040
Capsule with "+ items listed under May-83 CO
CuO+ZnO 270 12
Steel Slug "TTS Sludge' Cr203 270 34
CuO+ZnO+NaNO3 270 9 to 21 CO
CuO+ZnO+SiO2 270 9 CO
2b Inconel 650 SA508 0.0125 40%NaOH+10%KOH* none none 337 7 NP-3040 CO

Capsule with *+ items listed under CuO+ZnO+Na2SO4 337 7 May-83
Steel Slug "TFS Sludge" CuO+ZnO+NaCl 337 7
Na2HPO4 180 1.8
Na2CO3 180 1.4
mix ** 180 0
**mix = NaNO3+CaSO4+SiO2+NaCI+Na2HPO4
3 Single Tube 620 Pr. 4130 0.008 none .75ppm Ca(OH)2+ SPS 60 none NP-3060
Model Boiler 520 Sec. 50ppb N2H4 May-83
none 25ppm CaC3.+MgCO3+ SPS 53 none
50ppb N2H4
none .75ppm Ca(OH)2+ SPS 58 none
.3ppm Na2CO3+50ppb N2H4
4 Two Tube 642 Pr. A508 Cl 3 0.008 none 1 to 1Oppm NaOH+.lppm N2H4+ SPS at TiTS 16 to none NP-4053
Model Boiler 560 Sec. I ppm Na2CO3 or 126 Jun-85
1.5 ppm Na2SO4
5 Tubesheet 600 Pr. A508 C12 0.0075 none AVT,then screen+Fe304 14 NP-4272
Crevice 525 Sec. none P04+FrWtr+SPS, then screen+Fe304 126 Oct-85
0
Simulation none AVTFrWtr+Fe304 screen+Fe304 188
Apparatus
(TCSA) none AVT, then screen+Fe304 21 NP-4272
none AVT+NaIKOH+NaCO3, then screen+Fe304 24 Oct-85
none AVT+Na2C03+Fe3O4 screen+Fe304 106
6 Tube- 610 AISI 1025 0.008 none 8ppm NaOH+2ppm KOH none 37 10 NP-4802
Tubesheet 610 A508 0.008 none,Cr203,Fe3O4,SPS 8ppm NaOH+2ppm KOH none 186 2 to 7 Sep-86
Crevice 605 A508 0.008 none 10 ppm Na2CO3 none 66 3
Simulator 605 A508 0.008 none,Cr203,Fe304,SPS 10 ppm Na2CO3 none 180 1.5 to 4
(CRESI) 650 A508 0.008 Cr203,Cr2O3+FeS 8ppm NaOH+2ppm KOH none 67 1 to4
7 Inverted 620 Pr. 4130 0.008 40% NaOH+10% KOH .tppm OH* Cr203 60 none NP-4978 0
Single Tube 535 Sec. SA 508 0.008 none 1.7ppm CO3 Cr203 10 none Dec-86 .1
Model Boiler SA 508 0.008 none 1.7ppm CO3 Cr203 90 none 1P
SA 508 0.008 10% NaOH+2.5% Na2SO4 .5ppm OH+.2ppm S04" Cr203 87 none
SA 508 0.008 10% Na2CO3+4% NaOH 3ppm C03+.5ppm OH* SPS 91 none
*+ 50 ppb N2H4
CJ
8 Tubesheet 630 Pr. A 508 C12 0.0075 Na2CO3+Fe3O4, Cr203 Na2CO3+Fe3O4+ZrO2+ none 89 none NP-5263
Crevice 540 Sec. to N2H4 Jul-87
Simulation 0.0125 40% NaOH+1.6% Na2SO4 NaOH + Na2SO4 + none 42 none
Apparatus N2H4
(TCSA)
9 Two Tube 640 Pr. A508 Cl 3 0.008 40% NaOH+Na2SiO2 .lppm NaOH+ Na2SiO3+ 2 tubes-none, 137 none NP-5377
Model Boiler 563 Sec.- .25ppm NH4+.4ppm N2H4 2 tubes-Fe304 Aug-87
40% NaOH+Na2SO4 .1 ppm NaOH+ Na2SiO3+ 2 tubes-none, 137 none
.25ppm NH4+.4ppm N2H4 2 tubes-Fe304
none Loire River water SPS 325 none
*Sec. temp. reduced to 550 for 2nd part of Loire River water test.
10 Inverted 615 Pr. SA508 0.008 10% NaOH+2.5% Na2SO4 . .Sppm NaOH+.2ppm Na2SO4+ none 87 none NP-4978
Single Tube 535 Sec. and Cr203 on tube 100ppb N2H4 Dec-87
Model Boiler 10% NaOH+2.5% Na2SO4 .lppm NaOH+.04ppm Na2SO4+ none 91 none
and Cr203 on tube 50ppb N2H4
Notes: SPS = standard plant sludge, TTS = top of tube sheet

TABLE 11-2
Alkaline Denting Test Heat Transfer & Hydrazine Conditions DO
Test Heat Transfer Condition Dentino'? Hvdrazine Comments

Used?
1 Single Tube Model Boiler Tubesheet immersed in boiler. Yes No
Westinghouse - WCAP-10273 Tubesheet cooler than tube.
2 Inconel Capsule with Steel Slug Isothermal. Tubesheet and slug at Yes No
Westinghouse - NP-3040 same temperature.
3 Single Tube Model Boiler Tubesheet immersed in boiler. No No Mild alkaline conditions. Thus
Westinghouse - NP-3060 Tubesheet cooler than tube. results are not conclusive re
heat transfer conditions.
4 Two Tube Model Boiler External band heaters. However, No Yes Primary - secondary leaks-
CEA - AJAX - NP-4053 Figure 2-12 in NP-4053 indicates boric acid entered system after
tubesheet cooler than tube. several hundred hours
5 Tubesheet Crevice Simulation External band heaters. However, Yes Yes Similar denting in mild
* Apparatus - TCSA 1&2 - CE - CE states that isothermal conditions (mild) alkaline (FrWtr) as in caustic.
NP- 4272 are obtained (page 1-2 of NP-4272)
6 Tubesheet Crevice Simulator External band heaters. Tubesheet Yes No

CRESI - Westinghouse - hotter than tube.
NP-4802
7 Inverted Single Tube Model ID of tubesheet heated by coolant. No Yes

Boiler - Westinghouse - OD exposed to air. Tube sheet cooler
NP-4978, Vol. 1 than tube.
8 Tubesheet Crevice Simulation External band heaters. However, No Yes One of two tests had frequent
Apparatus - TCSA 4&5 - CE - CE states that isothermal conditions primary to secondary leaks.
NP-5263 are obtained (page 1-2 of NP-4272) Other test did not.
9 Two Tube Model Boiler External band heaters. However, No Yes

CEA - AJAX - NP-5377 Figure 2-12 in NP-4053 indicates
tubesheet cooler than tube.
10 Inverted Single Tube Model ID of tubesheet heated by coolant. No Yes

Boiler - Westinghouse - OD exposed to air. Tubesheet cooler
NP-4978, Vol. 2 than tube.
It has been postulated that caustic solutions tend to form on the hotter
metal surface in a crevice and remains on this surface except when
transferred by surface tension, wicking, or capillary flow to the colder
surface or to deposits. If the tube is at a higher temperature than the
tubesheet simulant, denting tends not to occur. If this model is correct,
denting should be accelerated by conditions where the tubesheet exceeds
the temperature of the tube. The test results shown in Table 11-2, however,
do not indicate a clear correlation between the temperature of the tube/
tubesheet and the occurrence of denting. Denting occurred in test 1, for
example, with the tube hotter than the tubesheet, but not in test 7, where
the temperature conditions were the same.
The use of hydrazine may tend to suppress denting by ensuring that
oxidizing conditions do not occur. With one exception, this hypothesis is
supported by the test results summarized in Table 11-2.
It is also possible that denting is accelerated by transients that cause
defects to develop in oxides as a result of thermal expansion effects. The
defects may allow corrodents new access to the metal surface, which
accelerates the development of oxide films. According to this theory, the
absence of denting in the laboratory was probably due to a reduced number
of transients in the test.
Relatively minor changes in the electrochemical potential or pH can
lead to large changes in oxide stability, which could increase the likelihood
of denting. However, this relationship is not yet fully understood.
In summary, the reasons why denting occurred in some tests, but not
in others, although the test conditions conducive to denting appeared to be
similar, require further investigation. However, based on the tests in which
denting did occur, it is concluded that there is a significant risk of denting
at tubesheet crevices and tube support crevices if strongly alkaline conditions
develop and persist for long periods of time.
As noted earlier in the discussion of plant experience, the possible
simultaneous occurrence of denting and IGA/SCC as the result of caustic
is a concern; this relationship has been observed both in plants and in
laboratory tests.
1 1-14 Steam Generator Reference Book, Revision I
IMPROVED OPERATING AND MAINTENANCE

PRACTICES
The main approach for avoiding alkaline denting is similar to what is
required to avoid IGA/SCC, acid denting, and pitting, namely, keeping the
ingress of impurities and oxidants to very low levels. This includes both
minimizing the ingress of impurities via the condensers and makeup
systems, and operating polishers and other ion exchangers using optimized
methods that minimize leakage of resin fines and ionic impurities. Methods
to accomplish these objectives are discussed especially in Chapter 21. There
are two parts to this approach:
(1) Minimizing the input of impurities, such as sodium and potassium.
This is directed at preventing high pH conditions in crevices where
impurities collect. If high pH environments can be avoided, then the
potential for caustic-induced denting and IGA/SCC is eliminated.
(2) Minimizing the input of oxygen and reducible corrosion products, such
as copper oxides and hematite. This is directed at preventing the
development of elevated (oxidizing) potentials in crevices, which can
accelerate IGA/SCC. It is suspected, but has not yet been proven in
the laboratory, that elevated potentials also can increase the generation
of corrosion products which tends to accelerate denting.
On-line additions of boric acid (11.22) is another recommended water
chemistry approach that has the potential for preventing alkaline denting.
Detailed investigations of the chemistry that develops in crevices of model
boilers during the simultaneous addition of caustics and boric acid show
that boric acid combines with the caustics and prevents the development
of strongly alkaline conditions. For this reason, it is expected that boric
acid will prevent alkaline denting, as well as the initiation of caustic-induced
IGA/SCC.
Model boiler tests (11.23) have shown that caustics are introduced into
tubesheet crevices primarily by the sludge located above the crevice. From
this location, accumulated caustics enter the crevice during shut down
operations when thermal hydraulic conditions cause the steam bubble in
the crevice to collapse and allows water to enter. (During power operation,
a steam bubble is normally present in the crevice, and entrance of liquids is
minimal.) Therefore, it is important to minimize the accumulation of sludge
on the tubesheet. Chapter 23 discussed various approaches for minimizing
the accumulation of sludge such as:
* periodic sludge lancing to remove accumulated corrosion products;
* reduction of air inleakage into secondary systems in order to reduce
corrosion of secondary system materials and the resulting transport of
corrosion products to the steam generators;
Tube Supports and Tubesheet Corrosion -Alkaline Denting 11-15
* use of full flow condensate polishers to remove corrosion products from

the condensate, especially during startups, thereby reducing the input
of corrosion products into the steam generator;
* elimination of copper from secondary systems to reduce (and eventually
eliminate) the input of copper into the steam generators. This also allows
the use of higher pHs in the secondary system, thereby reducing iron
corrosion rates and input of iron corrosion products; and
* use of morpholine or other amines to reduce corrosion of iron surfaces
in the secondary system, thus reducing the input of iron corrosion
products into the steam generator.
IMPROVED STEAM GENERATOR DESIGNS

Steam generator design features that will minimize the potential for
alkaline denting are similar to those that minimize the potential for other
forms of corrosion, such as IGA/SCC and acid-induced denting.
Recommended design features are listed in Chapter 4 and discussed in detail
in Chapter 24. Certain features are considered especially important with
regard to alkaline denting.
Full-depth expansion of the tubes into the tubesheet will eliminate the
deep crevices that exist in some early steam generator designs. These
crevices have been found to be collectors of impurities and have encouraged
corrosion, including alkaline denting.
The tube-to-tubesheet joint must be verified with regard to joint
tightness, remaining crevice depth, and residual stresses.
The tightness of the joint produced by the expansion should be sufficient
to prevent the ingress of alkaline or other species. The concern is that the
expansion process may not have closed the crevice entirely. (Depending
on materials and their coefficients of expansion, the tubesheet/tube joint
could become loose as a result of thermal cycling.) Aggressive chemicals
can creep into the remaining small crevices and cause corrosion of the tube
or tubesheet. Tests indicate that hydraulic expansion processes may not
always provide full protection in this regard, and that a superimposed
mechanical expansion is also required (11.24). However, this aspect of tube-
to-tubesheet joints has not been thoroughly reviewed in the open literature.
The tightness of hydraulic and explosive expansion joints of the types
previously and currently in use is not well known.
The size of any crevices left at the top of the tube-to-tubesheet joint, if
any, should be verified to confirm that they do not collect aggressive
impurities. This determination needs to take into account the thermal-
hydraulic design of the steam generator, especially the cross flow at various
locations across the top of the tubesheet, the degree of subcooling, the heat
flux, and the potential for sludge accumulation to aggravate the potential
for impurity concentration.
The residual stresses in the joint should be shown to be compatible
with expected ID and OD environments, considering applied stresses (due
to pressure, heat transfer, thermal expansion, bending by flow distribution
baffle, etc.) and possible impurity concentration effects in any remaining
crevices.
It is recommended that drilled hole support crevices be eliminated since
they are known to collect impurities and are subject to both acid and alkaline
denting (11.23). It is further recommended that stainless steels be used for
tube support structures. The data in Figure 11-3 indicate that corrosion
rates in alkaline environments are significantly reduced for stainless steels
as compared to carbon and low alloy steels. It is therefore hypothesized
that the use of stainless steels will greatly reduce the potential for alkaline
denting; however, this hypothesis has not yet been fully tested.
The use of a PWSCC-resistant alloy, such as thermally treated alloy 690,
will essentially eliminate the risk of primary side attack, even if some
unanticipated denting did occur. This alloy, with proper heat treatment, is
also expected to provide the maximum, practical resistance to secondary
side attack, though immunity is not possible if concentrated impurity
environments develop, especially if these are combined with high stresses
and applied strains due to denting.
SUMMARY
To date, alkaline denting has not been a major problem in PWR steam
generators, and it will most likely not become a wide-spread problem in
the future. However, the possibility of alkaline denting provides an
additional incentive to improve plant operations to minimize ingress of
caustic-forming impurities, oxidants, and sludge-forming corrosion
products, and to improve steam generator design. New steam generator
designs should eliminate crevices, use SCC-resistant materials, and materials
that will prevent the occurrence of denting.
REFERENCES
11.1 Craig, K. R. "Failure Analysis Tube R20L84 Steam Generator B
Fort Calhoun." In Proceedings: 1985 EPRI Workshop on Remedial
Actions for Secondary -Side IntergranularCorrosion. EPRI NP-4929.
December 1986.
11.2 Krupowicz, J. J. et al. "Caustic-Induced Magnetite Formation in
A508 Tubesheet Crevices." In Proceedings of the Second
International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems-Water Reactors, Monterey
California, September 9-12, 1985. American Nuclear Society, La
Grange Park, Illinois, 1986, p. 240.
11.3 NRC letter dated June 3, 1976 to AEP, Docket No. 50-315, and
enclosed IE Inspection Report No. 050-315/76-08.
11.4 Frieling, G. D. "Operating Experience With Inconel 600 Tube
Cracking, Point Beach Nuclear Plant, Units 1 and 2." Paper
presented at SGOG Contractors Meeting for Inconel-600 Cracking
in Tubesheet Crevices, September 9-10, 1980, Palo Alto, CA.
11.5 Agrawal, A. K., W. N. Stiegelmeyer, E. Vondrias, and J. Means.
"Point Beach-1 Steam Generator: Tubesheet Crevice Chemistry."
EPRI NP-5660-LD. February 1988.
11.6 "Summary of Meeting Held on November 5,1979, With Wisconsin
Electric Power Company and Westinghouse Electric Corporation
to Discuss Recent Inspection of Steam Generator Tubes at Point
Beach Unit No. 1." NRC meeting report. Docket No. 50-266. Nov.
16, 1979.
11.7 "Steam Generator Inspection, San Onofre Nuclear Generating
Station, Unit 1." SCE report. In NRC public document room. No.
8201270236 810925. January 30, 1977.
11.8 Kim, U. C., and D. van Rooyen. "Stain Rate and Temperature
Effects on the Stress Corrosion Cracking of Inconel 600 Steam
Generator Tubing in the Primary Water Conditions." In
Proceedings of the Second International Symposium on
Environmental Degradation of Materials in Nuclear Power
Systems-Water Reactors. J. T. A. Roberts, J. R. Weeks and G. J.
Theus, Symposium Committee, American Nuclear Society, La
Grange Park, Illinois, 1986
NP-4647. July 1986.
11.10 "Computer-Calculated Potential pH Diagrams to 3000 C, Vol. 2:

Handbook of Diagrams." EPRI NP-3137. June 1983.
11.11 Economy, G., and D. Smith-Magowan. "Causes of Denting, Vol. 1:
Summary Report." EPRI NP-3275. May 1984.
11.12 Wolfe, C. R., and J. B. Prestegiacomo. "Effects of Calcium Hydroxide
and Carbonates on IGA and SCC of Alloy 600." Westinghouse
Electric Corporation. WCAP-10273. February 1983.
11.13 "Neutralization of Tubesheet Crevice Corrosion." EPRI NP-3040.
May 1983.
11.14 Balavage, J. R. "Effect of Calcium Hydroxide and Carbonates on
IGA and SCC of Alloy 600." EPRI NP-3060. May 1983.
11.15 Daret, J. "Intergranular Attack of Alloy 600: Simulation Tests."
EPRI NP-4053. June 1985.
11.16 Krupowicz, J. J., and D. B. Scott. "Simulation of Intergranular Attack
on Alloy 600 tubing in Tubesheet Crevices." EPRI NP-4272. October
1985.
11.17 Connor, W. M., D. Smith-Magowan, G. Economy, R. H. Kunig, R.
G. Aspden. "Evaluation of Environmental Effects on Intergranular
Attack of Alloy 600." EPRI NP-4802. September 1986.
11.18 Hermer, R. E., and C. R. Wolfe. "Evaluation of Intergranular Attack
on Alloy 600, Vol. 1: Evaluation of Causes." EPRI NP-4978.
December 1986.
"Production of Intergranular Attack of Alloy 600, Alloy 690, and
Alloy 800 Tubing in Tubesheet Crevices." EPRI NP-5263. July 1987.
11.20 Daret, J. "Intergranular Attack of Alloy 600 Tubing: Simulation
Tests." EPRI NP-5377. August 1987.
on Alloy 600, Vol. 2: Effectiveness of Boric Acid." EPRI NP-4978.
December 1987.
11.22 Piskor, S. R. "Boric Acid Application Guidelines for Intergranular
Corrosion Inhibition," Rev. 1. EPRI NP-5558-SL. December 1987.
11.23 Campan, J.-L., and C. E. Shoemaker. "Sodium Hideout Studies in
Steam Generator Crevices." In Proceedings of the Third
Materials in Nuclear Power Systems-Water Reactors, edited by
G. J. Theus and J. R. Weeks. Warrendale, PA: The Metallurgical
Society, 1988, p. 199-207.
11.24 Engstr6m, J., and R. Bouecke. "Testing to Improve Tube-to-
Tubesheet Joint Design for Steam Generators." In Proceedings: 1990
EPRI Workshop on Circumferential Cracking of Steam Generator
Tubes. EPRI NP-7198-S. March 1991.
12
SECONDARY SIDE
INTERGRANULAR ATTACK AND
STRESS CORROSION CRACKING
-PLANT EXPERIENCE ON AVT
WATER CHEMISTRY
J. P.N. Paine,EPRI
M. J. Partridge,Dominion Engineering,Inc.
D. B. Lowenstein, Dominion Engineering,Inc.
1993
INTRODUCTION
A variety of operating PWRs throughout the world have experienced
intergranular corrosion (IGC) on both secondary and primary side surfaces
of steam generator heat transfer tubes. This chapter focuses on secondary
side or steam-side field experiences in plants operating with all volatile
water treatment (AVT). Laboratory experience is covered in other chapters
relating to specific corrodents. Experience at plants operating on phosphate
water chemistry is covered in Chapter 9. Laboratory experience and
experience of plants operating on phosphate water chemistry are mentioned
in this chapter only when necessary to provide further background.
Intergranular corrosion on the secondary side is a pervasive form of
corrosion in PWR steam generators. A great number of steam generators
operating at both freshwater and seawater-cooled locations have
experienced one or another form of IGC. This applies particularly to
recirculating, steam generators (RSGs). Intergranular corrosion has also
occurred at once-through steam generators (OTSGs), although these are
affected less severely than RSGs.
The occurrence of secondary side intergranular corrosion, especially in
RSGs, has been strongly correlated with the presence of built-in crevices
and crevices formed by the deposition of sludge, such as on top of the
tubesheet.
Tubesheet Crevices
Early RSGs of one manufacturer used part-depth rolling, for about
2- 4 inches (5-10 cm) to join the tubes to the tubesheet. The other domestic
RSG manufacturer has always used full-depth explosive expansion. The
part-depth expansion left a deep crevice, approximately 18 to 20 inches (46
to 51 cm) in depth with an 0.008 inch (0.2 mm) radial gap, at the top of the
tubesheet. This crevice is open to the secondary side environment.
Full-depth tubesheet crevices are not present in later designs since, in
the early 1970s, the manufacturer decided to expand tubes for the full height
of the tubesheet. However, because of design tolerances and manufacturing
variations, some small crevices can remain at the top of the tubesheet. Full-
depth expansion has been accomplished by using three different techniques.
Initially, explosive expansion was performed for a group of plants. Next,
full-depth expansion by mechanical rolling was achieved at a number of
plants. Beginning in the second half of the 1970s, full-depth expansion was
performed using hydraulic methods.
In Japan, three RSG plants with part-depth rolled tubes have been
expanded additionally to close the tubesheet crevice. This work was done
over a period of years after several years of operation. Approximately two
Secondary Side Intergranular Attack and Stress Corrosion Cracking 12-3
thirds of the tubes in each plant were not expanded for the full height of the
tubesheet, which left a 4 inch (10 cm) long crevice (12.1). Approximately
one third of the tubes were expanded nearly to the top of the tubesheet,
leaving only a 0.2 inch (5 mm) crevice.
OTSGs use part-depth rolls for both the top and bottom tubesheets,
and thus have deep crevices at both locations. Intergranular corrosion in
these crevices has been relatively minor. This is probably mainly the result
of the thermal hydraulic conditions associated with the once through design.
However, the full vessel stress relief applied to these units may also be a
factor in the relative absence of intergranular corrosion.
Tube Support Crevices

Steam generator designs of one domestic RSG manufacturer used
drilled-hole tube support plates (TSPs) until the second half of the 1970s.
In the drilled-hole configuration, crevices that are about 0.75 inches (19 mm)
high with nominal radial gaps of about 0.012 inches (0.3 mm) form around
the tubes. In later designs, starting in the mid-1970s, the TSP design was
changed to broached holes to minimize crevices. In addition, the support
plates were changed from carbon steel to stainless steel.
The second domestic RSG manufacturer has used mainly eggcrate tube
supports with flat support strips. However, one early plant used drilled-
hole TSPs, and other early designs, up to the middle of the 1970s, used
drilled-hole TSPs for partial tube supports high in the tube bundle. In the
mid 1970s, the use of drilled-hole TSPs was completely abandoned. In
addition, the eggcrate support strips were changed from carbon steel to
stainless steel.
OTSGs have used broached holes for almost all of their TSPs, which
are made of carbon steel.
Sludge Pile Crevices

Sludge accumulates on top of tubesheets and sometimes on tube
supports as a result of agglomeration and deposition of dissolved and
particulate solids carried into the steam generator by the feedwater. The
amount of sludge and its specific composition determines the type of
crevices that form between the sludge pile and the tubes. This situation is
governed by a complex set of conditions: (1) steam generator design
features, such as location and capacity of blowdown pipes, recirculation
ratio, and the presence of a flow distribution baffle; (2) secondary plant
design features such as sources of cooling and makeup water (which
determine the impurities entering the feedwater), the presence of copper in
the system, integrity of the condenser, use of full-flow condensate polishers,
makeup water system design, etc.; and (3) secondary system and steam
generator operating history and practices, such as type of water chemistry

used (AVT or phosphate), pH, amount of cooling water and air inleakage
allowed, frequency of sludge lancing, etc. Many early steam generator and
plant designs, as well as operating practices, were such that large sludge
piles developed and formed crevices that are conducive to intergranular
corrosion. Later designs and operating practices have been modified to
prevent this from happening.
Progression of Intergranular Corrosion

The progression rate of intergranular corrosion has been found to vary
widely, depending on the type of attack. In some cases, progression rates
have been sufficiently rapid to require unscheduled outages to plug or repair
leaking tubes, mid-cycle inspections, or plant temperature reductions to
reduce degradation rates. In a number of cases, steam generators have
been replaced as a result of significant secondary side intergranular
corrosion. Most of the corrosion has been confined to tubesheet and tube
support plate crevice locations so that leaks have been small. However,
several large leaks or ruptures have also occurred. For example, a large
leak required a plant shutdown in July 1988 at Almaraz 1; this leak resulted
from secondary side cracks initiating in a sludge pile located above the
flow distribution baffle. Other examples are a rupture in 1989 and a large
leak in 1992 at McGuire 1; the rupture and leak were due to secondary side
cracks in cold leg/preheater free span areas.
The number of units experiencing intergranular corrosion continues to
increase each year. The extent of the problem at individual plants also
continues to grow. Table 12-1 provides a listing of utility experience as of
early 1993, and shows the locations where intergranular corrosion has
occurred in various units throughout the world. No remedial measure,
except plant temperature reduction, has been shown to be effective for units
in general. Use of on-line additions of boric acid has also been considered
a possible general remedial measure. Although on-line boric acid additions
have correlated, on the average, with somewhat reduced intergranular
corrosion rates, rapid intergranular corrosion has nevertheless been
experienced at some plants using boric acid. Thus, boric acid cannot be
considered to be a proven remedial measure for intergranular corrosion.
Initially, plants affected by intergranular corrosion in tubesheet crevices
were primarily situated at freshwater-cooled locations. In those early days,
plants with tubesheet crevices at seawater sites did not exhibit the same
level of intergranular corrosion as those at freshwater sites. However, more
recently, the intergranular corrosion behavior of freshwater and seawater
plants has been similar. In fact, in some cases, intergranular corrosion has
been more severe at seawater plants than at freshwater plants.
TABLE 12.1
List of PWR Steam Generators With MA Alloy 600 Tubing That Have Experienced Secondary Side
Intergranular Corrosion Using AVT Water Chemistry CD
Tubesheet TTS I Tube Postulated

Date Hot Leg SG Cooling \Water Crevice Sludge SupportsI Other Causes and
No, Plant nit LComml. Temp. -- IIfg. Water I Treatment] IGA/SCC] IGA/SCC" IGA/SCCI IG••SCC Remarks
rD
1 Obrigheim (Orig) Mar-69 594 K Fresh AVT NA X
2 Beznau Dec-69 599 W Fresh P04/AVT X X Caustic
3 Ginna Jul-70 601 W Fresh P04/AVT X Caustic
4 Point Beach (Orig) Dec-70 597 W Fresh P04/AVT X X Caustic
5 Palisades Dec-71 599 CE Fresh P04/AVT NA X X Reduced sulfur species
6 Beznau 2 Mar-72 597 W Fresh P04/AVT X Caustic
7 Mihama 2 Jul-72 599 M Salt P04/AVT X Caustic
8 Point Beach 2 Oct-72 597 W Fresh P04/AVT X X Caustic
9 Maine Yankee Dec-72 598 CE Salt AVT NA x x (D
10 Surry (Orig) Dec-72 605 W Salt P04/AVT Severe denting CD
11 Turkey Point 3 (Orig) 3 Dec-72 602 W Salt P04/AVT Severe denting

12 Surry (Orig) 2 May-73 605 W Salt P04/AVT Severe denting
13 Oconee Jul-73 604 B Fresh AVT X Reduced sulfur species
NA
14 Ft. Calhoun Sep-73 594 CE Fresh AVT x Caustic 4.
15 Turkey Point 4 (Orig) 4 Sep-73 602 W Salt P04/AVT Severe denting
X
16 Prairie Island Dec-73 590 W Fresh P04/AVT
X
17 Zion Dec-73 591 W Fresh P04/AVT NDE only 0-
X
18 Zion 2 Dec-73 591 W Fresh P04/AVVT NDE Only
19 Kewaunee Jun-74 590 W Fresh P04/AVT X x Caustic
20 Three Mile Island Sep-74 604 B Fresh AVT Reduced sulfur species
21 Takahama Nov-74 608 W Salt AVT* Caustic
22 Arkansas Nuclear One Dec-74 603 B Fresh AVT Reduced sulfur species
23 Prairie Island 2 Dec-74 590 W Fresh AVT X__
24 Doel Feb-75 598 C Salt P04/AVT X
25 Rancho Seco Apr-75 608 B Fresh AVT Reduced sulfur species
NA
26 Calvert Cliffs May-75 599 CE Salt AVT
X X X Acid sulfates, lead
27 Ringhals (Orig) 2 May-75 609 W Salt AVT* Mild alkaline
28 Cook Aug-75 599 W Fresh AVT X X Caustic
29 Genkai Oct-75 599 M Salt AVT X Caustic - oxidizing
._L
TABLE 12.1 Cont'd
Intergranular Corrosion Using AVT Water Chemistry
Tubesheet TTS Tube Postulated
Date Hot Leg SG Cooling Water Crevice Sludge Supports Other Causes and
No. Plant Unit lComml. Ternm. F Mtfg Water Treatment I IGA/SCC IS
IGA/SCC I IGAISCC Remarks
30 Tihange 1 Oct-75 611 C Fresh AVT X X Mixed mode - lead
31 Takahama 2 Nov-75 608 M Salt AVT X X Caustic - oxidizing
32 Doel 2 Dec-75 598 C Salt AVT X X
2
33 Millstone Dec-75 604 CE Salt AVT NA x Acidic, then caustic
34 Trojan 1 May-76 615 w Fresh AVT NA Caustic
35 Indian Point (Orig) 3 Aug-76 595 w Salt AVT
36 Mihama 3 Dec-76 608 M Salt AVT NA
37 St. Lucie Dec-76 601 CE Salt AVT NA x Acidic and lead
38 Beaver Valley Apr-77 607 w Fresh AVT NA Caustic
39 Calvert Cliffs 2 Apr-77 599 CE Salt AVT NA Acid sulfates
40 Bruce A Sep-77 579 H-P Fresh AVT X Lead
41 Farley Dec-77 607 w Fresh AVT NA Caustic
42 Fessenheim Dec-77 611 F Fresh AVT NA Caustic
43 Fessenheim 2 Mar-78 611 F Fresh AVT NA Caustic
44 Cook (Orig) 2 Jun-78 606 w Fresh AVT X Caustic
45 North Anna 1 Jun-78 614/8 W Fresh AVT NA Acidic conditions
46 Bugey 2 Mar-79 613 F Fresh AVT NA Caustic, possibly lead
47 Bugey 3 Mar-79 613 F Fresh AVT NA Caustic, possibly lead
48 Ohi 1 Mar-79 613 w Salt AVT NA Caustic - oxidizing
49 Bugey 4 Jul-79 613 F Fresh AVT NA Caustic
50 Bugey 5 Jan-80 613 F Fresh AVT NA Caustic
51 Arkansas Nuclear One 2 Mar-80 607 CE Fresh AVT NA Reduced sulfur species
52 Gravelines 61 Dec-80 613 F Salt AVT NA
53 Gravelines Dec-80 613 F Salt AVT NA Caustic
2
54 North Anna Dec-80 614/8 w Fresh AVT NA Acidic conditions
55 Tricastin Dec-80 613 F Fresh AVT NA
2
56 Tricastin Dec-80 613 F Fresh AVT NA
2
57 Dampierre Feb-81 613 F Fresh AVT NA
58 Dampierre 3 May-81 613 F Fresh AVT NA Caustic
0
TABLE 12-1 Cont'd
Tubesheet I TTS Tube Postulated
Date Hot Leg SG Cooling IWater Crevice Sludqe
rjD
Supports Other ICauses and
No. Plant 0-
Unit Comm'l. Temp. F I Mfg. I Water] Treatment, IGA/SCC IGA/SCC IG /SCC IGA/SCCI Remarks
59 Tricastin 3 May-81 613 F Fresh AVT NA X X
60 Gravelines B3 Jun-81 613 F Salt AVT NA X X
61 Farley 2 Jul-81 607 w Fresh AVT NA X X Caustic
62 Sequoyah 1 Jul-81 609 w Fresh AVT NA X
63 Ringhals 3 Sep-81 610 W Salt AVT NA
1
64 Almaraz Oct-81 616 w Fresh AVT NA Caustic
65 Gravelines 14 Oct-81 613 F Salt AVT NA
66 Salem 2 Oct-81 602 w Salt AVT NA (D
67 Dampierre 4 Nov-81 613 F Fresh AVT NA
68 Tricastin 4 Nov-81 613 F Fresh AVT NA Caustic
69 Blayais Dec-81 613 F Salt AVr NA
70 McGuire Dec-81 618 W Fresh AVT NA X CL free span rupture (Ju
71 Sequoyah 2 Jun-82 609 w Fresh AVT NA
72 Doel 3 Oct-82 613 C Salt AVr NA
73 Blayais 2 Feb-83 613 F Salt AVT NA
74 Krsko 1 Feb-83 616 w Fresh AVT NA Caustic
75 Tihange 2 Mar-83 613 C Fresh AVr NA
76 St. Lucie 2 Jun-83 601 CE Salt AVT NA
77 San Onotre 2 Aug-83 611 CE Salt AVT NA x
78 St. Laurent B1 Aug-83 NA Caustic
613 F Fresh AVT
79 St. Laurent B2 NA Caustic
Aug-83 613 F Fresh AVT
80 Asco 1 Sep-83 620 w Fresh AVT NA Caustic & acidic
81 Ringhals 4 Nov-83 610 U Salt AVT NA
82 San Onofre 3 Jan-84 611 CE Salt AVT NA
83 Summer Jan-84 619 w Fresh AVT NA Minor IGA only
84 Almaraz 2 Feb-84 616 w Fresh AVT NA
85 McGuire 2 Mar-84 618 w Fresh AV'r NA X
86 Chinon B2 Aug-84 613 F Fresh AVT NA
87 Angra 1 Dec-84 616 w Salt AVT NA NDE only
I.k
TABLE 12-1 Cont'd
Tubesheet I TTS I Tube Postulated

Date Hot Leg SG Cooling Water Crevice Sludqe Supports Other Causes and
No. Plant Unit CommI. Temp. OF Mfg. Water Treatment IGA/SCC IGA/SCC IGA/SCC IGA/SCC Remarks
88 Catawba 1 Jun-85 618 W Fresh AVT NA X
89 Doel 4 Jul-85 626/19 C Salt AVT NA X X X Acidic, lead
90 Byron 1 Sep-85 618/08 W Fresh AVT NA X
91 Tihange 3 Sep-85 626 C Fresh AVT NA X
92 Asco 2 Feb-86 620 W Fresh AVT NA X X Caustic
93 Palo Verde 1 Feb-86 621 CE Fresh AVT NA X
94 Diablo Canyon 2 Mar-86 603 W Salt AVT NA X
95 Palo Verde 2 Sep-86 621 CE Fresh AVT NA X X
96 Shearon Harris 1 May-87 620 W Fresh AVT NA X X
97 Braidwood 1 Jul-88 618/08 W Fresh AVT -NA X
Legend:
B = Babcock & Wilcox AVT = All volatile treatment
C = Cockerill P04 = Phosphate treatment
CE = Combustion Engineering X = Present; based on NDE and/or metallurgical evaluation
F = Framatome *= Very limited period with P04 water treatment
H-P = Howden-Parsons NA = Not applicable
K = Siemens/KWU CL = Cold leg
M = Mitsubishi TTS = Top of tubesheet
U = Uddcomb
W = Westinghouse
The three plants with low-temperature mill annealed tubing that have
experienced the most severe occurrence of intergranular corrosion at TSPs
are all seawater-cooled plants in Japan (Genkai Unit 1, Ohi Unit 1, and
Takahama Unit 2).
Among older domestic RSG plants with high temperature mill annealed
tubing, intergranular corrosion is relatively severe at both seawater-cooled
plants (e.g., Millstone Unit 2, St. Lucie Unit 1, and Calvert Cliffs Units 1
and 2) and at freshwater-cooled plants (e.g., Palisades Unit I and ANO
Unit 2).
For modern plants with high integrity condensers, the differences in
intergranular corrosion experience between freshwater- and seawater-sited
units is small. These differences may be due more to the quality of makeup
water and the performance of various demineralizers than the type of
cooling water.
Definition of Terms
Intergranular corrosion has been found to take several forms. Therefore,
considerable confusion exists in the literature as to precisely what is meant
when an author identifies a particular form of intergranular corrosion. At
least three forms have been identified; these are illustrated by Figures 12-1,
12-2, and 12-3.
FIGURE 12-1
IGSCC in Alloy 600 C-Rings (Etched Sample)
Figure 12-1 (12.2) illustrates typical intergranular stress corrosion

cracking (IGSCC). In the case of IGSCC, the corrosion morphology consists
of single or multiple major cracks with minor to moderate amounts of
branching. In most cases, cracks propagate intergranularly in alloy 600
steam generator tubing, although occasional cases of transgranular cracking
are observed. The rate of IGSCC may be very high, as much as 0.04 inches
(1.0 mm) per day, depending on the corrodent, electrochemical potential,
stress level, and the material properties. Experience suggests that IGSCC
requires a high stress level in order to propagate rapidly. At lower stress
levels, propagation rates may approach zero or the corrosion may take
another intergranular form.
This form of intergranular corrosion has been described as general
intergranular attack (IGA) or volumetric IGA and is illustrated in Figure 12-2
(12.3). Its morphology is characterized by relatively uniform attack of all
grain boundaries over the surface of the tubing. A distinguishing feature
between IGA and IGSCC is the lack of evidence for a significant stress
contribution to the morphology of IGA. The close relationship between
IGA and IGSCC is further illustrated in Figure 12-2 by an apparent stress-
assisted finger of corrosion penetrating from the IGA layer into the tube
FIGURE 12-2
IGA of Alloy 600 Steam Generator Tubing (Etched Sample)
material. IGA propagation rates of 0.00006 inches (0.0015 mm) per day have
been measured in laboratory tests and are consistent with most steam
generator experience.
The third form of intergranular corrosion, called intergranular
penetration (IGP), has been variously described as a mixture or a hybrid
between the other two forms, or as the initiation stage of corrosion when it
is not yet clear which form it will eventually take. IGP has been experienced
over a significant range of morphologies. One typical morphology is shown
in Figure 12-3, which illustrates a finger of intergranular corrosion extending
below a surface layer of IGA (12.4). In the case of IGP,a stress factor appears
to be involved. However, it has been postulated that the propagation rate
is considerably slower than that of IGSCC, and that the propagation of IGP
is only weakly dependent upon stress. It is possible that the stress required
for activation of IGP is nominally at or below the material yield strength.
FIGURE 12-3
IGP of Alloy 600 Steam Generator Tube
The confusion in definitions and the interrelationship between the forms

of intergranular corrosion are illustrated in Figure 12-4, which shows IGSCC,
IGA, and IGP on the same steam generator tube segment in close proximity
to each other. Figure 12-4a (12_.5 shows an apparent effect of stress and
might be termed IGSCC. Figures 12-4b and c show fingers of corrosion,
IGP, extending below more general surface degradation IGA. Also
illustrated is the difficulty in observing intergranular corrosion of alloy 600
in unetched micrographs. On the left hand of Figures 12-4a, b, and c are
unetched metallographic samples which show IGSCC but fail to adequately
define the extent of IGA or IGP. These are brought to view only by proper
etching techniques.
In several cases, secondary side corrosion has included mixed mode
attack in which some transgranular cracks have occurred as well as
intergranular attack. (Pure transgranular stress corrosion cracking has been
observed in a few laboratory tests in lead-contaminated environments
[Chapter 151, but has not been observed in removed tubes.) A typical case
of mixed mode attack is shown in Figure 12-5 (12.6). Tests indicate that a
mixed mode attack can be caused by the presence of lead contamination
(lead has sometimes been found in deposits associated with mixed mode
attack).
For convenience, the whole set of intergranular corrosion modes,
including intergranular/transgranular mixed mode attack, is referred to as
IGA/SCC in this chapter.
Summary of Plant Experience
Locations. Within the steam generator, intergranular corrosion has been

found in a variety of locations. The most important of them are the tube-
to-tubesheet crevice (an annular gap remaining in some early steam
generators after manufacture) as shown in Figure 12-6; within sludge piles
on top of the tubesheet and, in a few cases, on top of tube supports; and
tube-to-tube support plate crevices in drilled-hole tube supports, as shown
in Figure 12-7. Significant intergranular corrosion at this latter location was
first found in one domestic plant with sensitized tubing, then at several
Japanese sites with mill annealed tubing, and, recently, at several foreign
and domestic sites in increasing amounts.
In addition to the locations mentioned above where most of the attack
has occurred, intergranular corrosion has also occurred on the OD of
expansion transitions located at the top of the tubesheet in non-sludge-pile
areas (as well as in sludge pile areas), at hot leg and cold leg tube-to-tube
support strip intersections in units with eggcrate supports, at vertical
support locations in the top portion of the tube bundle, and in' hot leg, cold
leg, and U-bend free span areas.
FIGURE 12-4
Intergranular Corrosion of Alloy 600 Steam Generator Tube
Alloys. Most of the steam generators experiencing secondary side

intergranular corrosion are tubed with mill annealed alloy 600. However,
in laboratory tests, intergranular corrosion has been produced in mill
annealed alloys 600, 800, and 690, and in thermally treated alloys 600 and
690. Based on these laboratory tests, there appears to be little difference
between the performance of these alloys in the mill annealed metallurgical
condition. Thermally treated materials, especially alloy 690, were
significantly less susceptible.
FIGURE 12.5
Mixed Mode Corrosion of Alloy 600 Steam Generator Tube
(Exam of St. Lucie Tube)
FIGURE 12-6
Typical Steam Generator Tube-Tubesheet Annular Crevice
Primary face
tube sheet
Secondary Side Intergranular Attack and Stress.Corrosion Cracking 12-15
FIGURE 12-7
Typical Steam Generator Tube-Tube Support Plate Annular Crevice
Tube
SnI
I I 0 .012" x .75" nominal
Crc in I crevice dimensions
lcatking Corrosion
oanpresent
111 Lplate I -Tube support

I I
Secondary side intergranular corrosion in PWR steam generators has

affected both low temperature mill annealed tubing and high temperature
mill annealed tubing. Tests in both deaerated acidic (12.7) and deaerated
caustic (12.8) environments indicate that there is little difference in material
susceptibility to these environments, where the materials are strained to
about the same degree (e.g., C-rings witl.two legs touching). On the other
hand, even short periods of time (e.g., six minutes or more) in the carbide
precipitation temperature range (i.e., minor amounts of sensitization)
significantly reduce the susceptibility of low-temperature mill annealed
material in caustic environments (12.9). In addition, IGA experience in
Japanese plants also indicates that carbide precipitation results in reduced
susceptibility, since low-temperature mill annealed tubes with high ferrite
readings (indicative of carbide precipitation) experience less IGA than tubes
with low ferrite readings (12.10).
Contaminants. Several types of environmental contaminants have been

postulated to explain the occurrence of intergranular corrosion (12.5), such
as:
" high concentrations of sodium and/or potassium hydroxide;
" reaction products from the reduction of sulfate ion with hydrazine or
hydrogen at metal surfaces (This reaction is postulated to produce
reactive, partially reduced, sulfur species; however, these species do
not affect mill annealed tubing but can corrode sensitized tubing,
especially at layup temperatures under oxidizing conditions.);
* acidic products of thermal decomposition of ion exchange cation resins
and neutral organic impurities that pass through makeup and
condensate demineralizers (acidic sulfur species and organic acids);
* highly concentrated salt solutions at neutral or near neutral pH (These
salt solutions are the natural consequences of condenser leakage
concentrated to high levels of salt by the boiling processes of the steam
generator.);
" alkaline carbonates and/or their reaction or hydrolysis products
(believed to affect the nature of the passive film on the alloy surface);
" steam environments doped with impurities such as caustics, acids, or
lead;
" lead and its compounds (12.8); and
* thermal decomposition products of neutral clays and colloids (Such
clays and colloids can pass through polishers and may decompose and
release hydroxyl ions.).
Temperature and Superheat. IGA/SCC has occurred preferentially in

higher temperature locations, such as hot leg tubesheet crevices, sludge
piles, and at tube support crevices. These locations also have higher
available superheat. Which of these factors is more important is not clear,
though the increased rates of IGA/SCC exhibited by higher temperature
plants, with similar superheats, implies that temperature has a great
influence.
The effect of temperature on rates of secondary side IGA/SCC is
believed to be controlled by an Arrhenius or apparent activation-energy-
type effect, with typical apparent activation energies around 50 kcal/mole.
The differences in available superheat at different locations can affect
corrosion rates by influencing the stability (presence) of a corrodent, and
by controlling its concentration. For example, caustic species such as sodium
hydroxide and sodium acetate are stable in liquid form at superheats higher
than those available in steam generators, i.e., higher than 75'F (42QC), and
thus can be present in any crevices in steam generators. Typical acidic
species, however, can sustain maximum superheats of only about 30'F (17°C)
(acid sulfates) to 45 0F (25°C) (acid chlorides) maximum; even at these levels

they volatilize and must be replenished. Thus, an acidic crevice may not be
stable in some hot leg areas at full power. Hence, the ability of corrodents
to exist as a liquid phase at available superheats can affect the occurrence
of IGA/SCC at different locations. For example, acidic species are unlikely
to affect tubes in hot leg tubesheet crevices since the tubesheet crevices.
have superheats exceeding the stability limits of the acid species; however,
they can affect tubes in hot or cold leg tube support areas since these areas
have lower superheats.
The available superheat also controls the concentration of impurities
that are stable, with higher concentrations being required for stability at
higher superheats. The changes in concentration at differing locations with
different superheats can affect the rate and character of the corrosion at
these locations.
Stresses. Laboratory tests indicate that tensile stresses strongly accelerate

the rate of SCC, and moderately affect the rate of IGA. Most plant experience
of IGA/SCC has occurred on straight tube lengths where applied stresses
are low. However, residual stresses from straightening and polishing may
be a significant factor. When interpreting the results of plant experience,
including results of examinations of removed tubes, it should be considered
that stresses at the OD of plant tubes are generally significantly lower than
those in laboratory test specimens such as U-bends. These lower stresses
tend to reduce the amount of IGSCC relative to IGA as well as increase
plant initiation times relative to laboratory results.
Cold Work. Plant experience and laboratory tests indicate that the thin
layer of cold-worked material produced on the OD surface of mill annealed
tubes by polishing is more susceptible to IGA/SCC than unpolished or
thermally treated material. The shallow IGA that readily occurs in this
hardened layer (12.7) may increase susceptibility to further attack, i.e., to
deeper SCC.
Electrochemical Potential and Presence of Oxidants. Laboratory tests

indicate that electrochemical potentials that are 100 to 200 mV above the
corrosion potential in deaerated environments strongly accelerate the rates
of SCC in caustic and acidic environments. The actual values of
electrochemical potential present at typical sites of corrosion are not well
known; they can be affected by deposits, dissolved species, galvanic couples,
etc. Plants attempt to prevent elevation of the corrosion potential by
preventing the ingress of oxidants such as dissolved oxygen, copper oxides,
and dissolved copper ions. However, in some cases, the occurrence of IGA/
SCC has been attributed to periods of operation with elevated corrosion

potential. For example, in Japan, the IGA/SCC occurring in several severely
affected plants is believed to have initiated early in plant life when oxidizing
conditions were present due to air inleakage, and polishers were not
available to remove copper and oxidizing corrosion products (12.11).
Focus. This chapter will focus on (1) operating plant experience, (2) results
of destructive examinations of intergranular corrosion and laboratory
examination of deposits and corrosion films, and (3) recommendations to
prevent/arrest intergranular corrosion.
PLANT EXPERIENCE: DEFINITION OF PROBLEM

A considerable amount of information has been gained by examining
operating steam generators which have experienced intergranular corrosion.
This experience is separated into the following categories:
" correlation of IGA/SCC experience with operating history;
" failure analysis of samples, including identification of crevice chemistry
to the extent possible based on examination of removed tubes and
deposits; and
* effects of IGA/SCC on tube integrity.
Operating History
Only limited attempts have been made to correlate field experience
with intergranular corrosion. These efforts have not been overly successful,
although several fairly distinct factors that generate higher than average
susceptibility have been suggested. Below is a list of plant factors that have
been partially investigated:
* temperature (hot leL versus cold ley location, and hot ley temnerature
variations among plants);
0 seawater site versus freshwater site;
0 secondary plant materials;
0 impurity ingress;
0 AVT or coordinated phosphate water treatment;
0 period of operation before first occurrence, and aging of material;
0 presence/use of full flow condensate polishers;
0 use of boric acid;
6 use of morpholine; and
0 position in tube bundle.
Temperature (Hot Leg Versus Cold Leg Location, and Hot Leg
Temperature). Based on eddy current examinations and limited numbers
of laboratory examinations, intergranular corrosion has occurred mainly
and more extensively on hot leg tubes. Primary water flows through hot
leg tubes at a temperature of about 590-626°F (310-330'C), which is the
equilibrium temperature within tubesheet crevices (if present) and within
tubesheet sludge piles. In the cold leg tubesheet area, where corrosion has
been less frequently observed, crevice temperatures range from 530-560'F
(277-2930 C).
In an attempt to reduce the corrosion rate in tubesheet crevices at Point
Beach Unit 1, power levels were reduced to 78%, which reduced the hot leg
temperature from 5970 F to 557°F (314'C to 292QC). Based on the number of
detected new defects, it was clear that the corrosion rate had decreased
dramatically. While these data are limited, they are consistent with
experience at other plants (e.g., Ringhals Unit 2) which have reduced
temperature, and also with laboratory test data that show that both IGA
and IGSCC rates are dependent upon temperature.
Reviews of the intergranular corrosion performance of the total set of
LTMA-tubed steam generators have shown that the rate of IGA/SCC in
tubesheet crevices and in tube support plate crevices increases, on the
average, quite strongly with an increase in hot leg temperature (Figures 12-8,
12-9) (1Z2j 12.13). The apparent activation energies of about 50 kcal/mole
exhibited by plant performance appear reasonable in light of laboratory
tests of caustic and acidic environments (12.I14 12.15).
The main exceptions to the "hot leg only" occurrence of IGA/SCC are
discussed below.
* Attack has occurred at vertical supports in HTMA-tubed units, located
in the top region of the steam generator. This attack has been associated
with dry out areas at the supports, and has been severe enough to lead
to ruptures and to preventive plugging of substantial numbers of tubes
in affected regions.
* IGA/SCC has been experienced at the top of tubesheets, in sludge piles,
and at tube supports in cold legs of several units. Possible causes for
this relatively severe cold leg attack are discussed later.
* In a few cases, corrosion of sensitized tubing was attributed to the
presence of reduced sulfur species. IGA and IGSCC of sensitized tubing
caused by reduced sulfur species are known to initiate and propagate
at ambient temperatures.
* At Ringhals Unit 2, five cold leg tubes were removed in May 1984, after
about ten years of operation (12.16). This plant had relatively high
temperatures (hot leg temperature of 609'F [320.5'C] and cold leg
temperature of 5520F [289QC)1. Examination of the tubesheet crevice
FIGURE 12-8
Time to 1% IGA/SCC Versus Inverse Temperature for Part-Depth-Rolled Plants
615 605 Hot Leg Temperature, OF 595 585
C)
l [•lower
: bound*
o 56 Kcafmole
EFPY 10 (D
Has not reached 1% at this time. Actual time may be greater

1
1.71 E-03 1.72E-03 1.73E-03 1.74E-03 1.75E-03 1.76E-03 1.77E-03
Inverse Hot Leg Temerpature, 1/°K
NOTES: Does not include plants that remained on phosphate water chemistry. Includes all tube mills except Sandvik.
FIGURE 12-9
Time to 1% IGA/SCC Versus Inverse Temperture
-Plants With Alloy 600 LTMA Tubing, Drilled-Hole TSPs, and AVT Water Chemistry
100
Apparent Activation Energy: 0 = 54 + 16 kcal/mole

(0
"1% failures not yet achieved 95%
LL 11%failures achieved ••
50%
00
0 a 0)
10
8_ 0)
(L
0
oE ~-
Temperature (Scale is 1/T in 1/°R)

region showed that all five of the cold leg tubes exhibited IGA, to a
depth of one to five grains, and all five exhibited IGSCC, to a maximum
depth of 0.009 inches (0.23 mm). The investigator concluded that the
relative depths of hot leg and cold leg IGSCC at this plant were
consistent with an apparent activation energy of 40 kcal/mole.
IGSCC has been observed in cold leg free span regions at McGuire
Units I and 2 and in hot leg free spans at Palo Verde Unit 2 (1993). It
was first seen in 1989, when a rupture occurred in a cold leg tube at
McGuire 1 (12.17), after eight years of operation. Additional similar
defects have been subsequently detected in both McGuire Units I and
2. Laboratory examination indicates that the long axial cracks have
been caused by linkup of a series of IGSCC cracks located along a
shallow groove or line of abrasion. Relatively rapid crack growth rates
in the through-wall direction have been observed, up to a maximum of
about 3 mils (75 gtm) per month, with an average growth rate of about
1 mil (25 ýtm), or slightly less, per month.
Seawater Site Versus Freshwater Site

The differences in experience with regard to occurrence of IGA/SCC at
seawater versus freshwater-cooled units can be summarized as follows:
" IGA/SCC occurred first in tubesheet crevices or in sludge piles on the
tubesheet at freshwater-cooled units with tubesheet crevices. This
experience is attributed to the occurrence of cooling water inleakage
combined with the alkaline-forming tendency of the impurities
introduced by the inleakage.
" Subsequent to the early experience with IGA/SCC at freshwater units
with tubesheet crevices, IGA/SCC has become a major concern at
several seawater-cooled units, both with and without tubesheet crevices.
Moreover, the IGA/SCC behavior of later freshwater and seawater units
has been quite similar. This change in behavior, i.e., similar occurrence
of IGA/SCC at seawater and freshwater units, is attributed to the
significant improvements made by the industry with regard to
condenser leak tightness. This has minimized the introduction of
alkaline-forming, IGA/SCC-inducing impurities at freshwater sites. It
has also minimized the introduction of acid chlorides at seawater sites,
which may tend to inhibit IGA/SCC. As a result, it is postulated that
IGA/SCC experience of modern units, located at both freshwater and
seawater sites, tends to be dominated by impurities entering via the
makeup system and by demineralizer throw. These factors tend to be
similar at all sites, regardless of the type of cooling water.
IGAISCC at Tubesheet Crevices of Part-Depth Rolled Units With LTMA

Tubing. IGA/SCC in tubesheet crevices is a phenomenon that occurred
earlier at freshwater-sited power stations than at seawater sites. It has
affected all freshwater RSG units with tubesheet crevices, and has resulted
in replacement of steam generators at two freshwater units (Point Beach
Unit I and Beznau Unit 1), extensive sleeving at four freshwater units (Point
Beach Unit 2, Cook Unit 2, Ginna, and Kewaunee), and planned
replacements at three units (Beznau Unit 1, Ginna, and Point Beach Unit 2).
While the problem started earlier at freshwater sites, IGA/SCC has also
severely affected several seawater units. It has resulted in the replacement
of steam generators at one unit (Ringhals Unit 2) and is a factor in the
extensive sleeving and planned replacement at two units (Takahama Unit 2
and Genkai Unit 1). IGA/SCC has also resulted in extensive sleeving of a
seawater unit that remained on phosphate water chemistry until plant
shutdown, San Onofre Unit 1.
As time passed, most utilities took steps to increase condenser integrity
and otherwise improve water chemistry. As a result, the IGA/SCC behavior
of freshwater and seawater units has become quite similar. Worldwide
experience is illustrated in Figure 12-10. This Weibull plot was prepared
using standard statistical methods (two parameter fits, using suspended
items and median ranks 12.18). It shows times to cumulative fractions of
plants affected (at the 1% of tubes level) by IGA/SCC in the tubesheet
crevices, for both freshwater plants and seawater plants with tubesheet
crevices. To minimize effects of the temperature variable, times have been
adjusted to a hot leg temperature of 600'F (316'C) using an activation energy
of 50 kcal/mole. Figure 12-10 shows that freshwater plants have tended to
experience IGA/SCC in tubesheet crevices earlier than seawater plants, but
that plants of both types tend to be affected as plants age. The reasons for
this pattern of behavior are described at the beginning of this section. These
are related to the changes that have occurred in condenser integrity, which
have changed the types of impurities dominating the crevice environments
as condenser integrity has increased.
The examples below illustrate how water chemistry history has affected
the occurrence of IGA/SCC at seawater-cooled plants with tubesheet
crevices.
Several Japanese units have experienced IGA/SCC in tubesheet crevices
and/or tube-to-tube support plate intersections. An extensive study
was performed to identify the causes of the IGA/SCC (12.11).
Researchers concluded that the IGA/SCC was the result of free caustics
and elevated corrosion potentials. The free caustics were identified as
mainly coming from welding slag residuals and fumes left from original
construction, as well as elevated corrosion potentials originating from
FIGURE 12-10
Seawater Versus Freshwater IGC Experience at Plants With
Tubesheet Crevices (Time to 1% Tubes Plugged/Sleeved)
Weibull Plot
.90
0.63
0.5
15 0.2
C
0
.10
0.05
0.02
.01*
1 10 100
Adjusted Service Time (EFPY)
(Adjusted to 600.F using Q=50 kcal/mol)

high condensate oxygen levels early in plant life. It was observed that
seawater inleakage had not occurred at the affected plants (This is a
result of condenser integrity, including thorough inspections and
conservative preventive plugging practices). The study also showed
that IGA/SCC had not occurred at similar plants which experienced
seawater inleakage. Thus, it appears that, if a plant has 'tight condensers,
the presence of seawater cooling is irrelevant with regard to IGA/SCC.
On the other hand, if a plant experiences seawater inleakage, it is less
likely to experience IGA/SCC, though other problems can be seriously
aggravated (e.g., denting and pitting).
Japanese plant experience has shown that operation for a short period
using phosphate water chemistry early in life tends to reduce the risk
of IGA/SCC occurring later in life, as compared to plants which have
always operated on AVT (12.11) (Chapter 9). Examination of removed
tubes showed that residual phosphates were still present in crevices.
The reduced amount of IGA/SCC is attributed to the ability of the
remaining phosphates to neutralize caustics accumulating in the
crevices.
The experience of Ringhals Unit 2 is more difficult to explain. This
unit is on a seawater location. The original steam generators
experienced significant denting early in life as a result of condenser
inleakage, suggesting that tube support plate crevices were acidic and
dominated by chlorides. In 1980-81, the condensers were replaced with
leak-tight, titanium-tubed units, and since that time seawater ingress
has been non-existent. One would expect that with time the crevice
acidity would have been reduced due to loss of chloride by diffusion,
and that a neutral environment would have been achieved. In fact, the
denting rate was reduced to essentially zero, documenting the success
of the improved water purity control. Subsequent to this improvement,
however, IGSCC was found in a large number of tubes in tubesheet
crevices. Several different crevice contaminants have been suggested.
In fact, the corrodent may be a result of all or several of the contaminants
listed below:
-sulfuric acid accidentally introduced into the secondary water supply;
-organic acids or other species from the makeup water system (formic,
acetic and propionic acid salts have been identified);
-selective elution of buffering species from the crevices with time,
perhaps leaving a caustic environment;
-- caustics resulting from operation with high Na/Cl (sodium/chloride)
ratios in, the blowdown; and
-steam doped with one of the aggressive chemicals mentioned
previously.
Based on crevice flushing data from Ringhals Unit 2, one can

conclude with a high degree of reliability that the fluid composition in
the Ringhals Unit 2 tubesheet crevices does not resemble a concentrate
of seawater. This is reasonable, since makeup water at Ringhals is not
derived from seawater, and the condensers are now leak tight. Based
on MULTEQ analyses of the composition of soluble deposits on tube
surfaces of pulled tubes, it appears that the crevice environment at
temperature was moderately high in pH, just below a pHT of 10, i.e.,
moderately alkaline.
Doel Units I and 2 are located on the estuary of the Schelde River, a
brackish river source. Only a few occurrences of IGSCC on the
secondary side of the tubing have been detected. The units operate
with full-flow condensate polishers (FFCP) and careful attention to
water chemistry. Analysis of hideout return shows an abundance of
silica paralleling the return of sodium and potassium. It is possible
that crevice chemistry is dominated by the effluent of the FFCP units
and not by seawater inleakage, or that crevice chemistry is buffered
and dominated by silica compounds.
IGA/SCC in Hot Legs of Combustion Engineering Units. These units

have full-depth explosively expanded tubes (i.e., no tubesheet crevice), and
mostly eggcrate type supports. IGA/SCC has been experienced in sludge
piles and, to a lesser extent, at eggcrate supports, at both freshwater- and
seawater-cooled plants, although degradation was seen first in seawater-
cooled plants. (It should be noted that Palisades has been excluded from
this comparison, since it has sensitized tubing and drilled-hole tube support
plates, and is not typical of the other Combustion Engineering units.) The
IGA/SCC at seawater units has been attributed to acid species at St. Lucie
Unit 1 and Calvert Cliffs Units 1 and 2, and to caustic at Millstone Unit 2.
The IGA/SCC at ANO Unit 2, a freshwater unit, has been attributed to
acidic species.
IGA/SCC at Hot Leg, Drilled Hole Tube Support Plate Crevices of Units
With LTMA Tubing. The experience of freshwater and seawater units is
illustrated in Figure 12-11. As shown in the figure, the behavior of the two
groups of plants is not very different, with the seawater plant data spanning
the freshwater plant data. It should be noted that the most severely affected
units, Takahama 2, Genkai 1, and Ohi 1, are seawater-cooled units.
The similar behavior of the seawater- and freshwater-cooled units is
attributed to the fact that IGA/SCC at TSPs occurs primarily in relatively
high temperature plants, and thus in more modern plants, most of which
have high integrity condensers. The presence of high-integrity condensers
is believed to make the type of cooling water essentially irrelevant with
respect to occurrence of IGA/SCC, as discussed earlier.
FIGURE 12-11
Seawater Versus Freshwater IGC Experience at TSPs
-Plants With LTMA Tubing and Drilled-Hole TSPs (Time to 1% Tubes Plugged/Sleeved)
Weibull Plot
.90 /
S -0.63
C Seawater PlantsL 13
1301
o03
0
t.10
.00.0
A13
Seric (EFPY)ts Seaate
0.02 -
C33
1 10
Service Time (EFPY)
(Adjusted to 615OF using Q=50 kcal/mofl)4
Secondary Plant Materials

Laboratory testing has shown that electrochemical potentials above the
free corrosion potential accelerate the rate of IGA and IGSCC in both caustic
and acidic environments. In caustic environments, the potential is about
100 mV above the free corrosion potential for maximum IGA, as shown in
Figure 12-12, and about 150 mV above the free corrosion potential for
maximum IGSCC (12.19). Similar effects are reported for acidic
environments (12.15). Since the introduction of secondary plant corrosion
products into the steam generator can raise the potential at tube surfaces in
sludge piles and crevices, it is possible that secondary system materials can
have a significant influence on the tendency for IGA or IGSCC to initiate.
For example, the presence of copper oxides in the sludge is believed to
promote IGSCC in sodium hydroxide solutions. Additions of other oxides,
such as magnetite and chromic oxide, have been shown to increase the rate
of IGA in some tests.
FIGURE 12-12
Potential Area Where IGA and IGSCC Develop for Alloy 600 in
Carbonate-Contaminated NaOH Solutions
300
200 S
Soc
z
E 100 SO
0c_+
CU IGA
C __
0 IGA
102 103 104

Current (pA)
IGA and IGSCC have been found in units with a mixture of secondary
plant designs and materials ranging from all ferrous to predominantly
copper alloys. Time to a given level of IGA/SCC in plants with feedring
steam generators with and without copper alloys in the secondary system
have been evaluated. There are indications that plants with copper alloys
tend to experience the problem earlier (Figure 12-13) (12.20). However,
similar evaluations for units with preheater steam generators show little
difference between plants with and without copper. Other plant experience
suggests that copper can have an effect. For example, among low
temperature plants at freshwater sites with tubesheet crevices, IGA/SCC
has been more severe at Point Beach and Kewaunee, which have copper
alloys, than at Prairie Island and Zion, with all ferrous systems. Similarly,
for full-depth rolled plants at freshwater sites with high temperatures, IGA/
SCC occurred earlier at Catawba Unit I than at McGuire Unit 1, which
correlates with an increased amount of copper in Catawba Unit 1. Originally,
both units had copper-nickel-tubed MSRs, but the tubes were replaced with
stainless steel tubes early in life at McGuire Unit 1. On the other hand,
some units with no copper have experienced relatively early IGA/SCC at
TSPs, e.g., Ringhals Unit 3 and Doel Unit 4.
Impurity Ingress
Developing useful correlations between impurity ingress and the
corrosion performance of steam generator tubes is not an easy task. It is
difficult to determine accurately and completely the levels of all relevant
impurities, to determine what crevice environments are produced as a result
of the impurity ingress, and to consider the many reactions possible among
the concentrated impurities, differing volatilities of the various impurities,
etc. Two main approaches for monitoring and assessing impurity ingress
have been pursued, i.e., to measure levels of contaminants in the blowdown
water and to gather and analyze hideout return data.
Unfortunately, data on concentrations of applicable contaminants in
the blowdown are often limited or lacking. The results of an early study
using blowdown data (12.21) did not find a strong correlation between
impurity concentrations in the blowdown and corrosion. However, there
was a possible correlation between IGA/SCC occurrence and the cumulative
number of days with abnormal levels of sodium or cation conductivity. It
appears somewhat unlikely that more useful correlations will be possible
using blowdown data, since it is difficult to quantify all appropriate variables
(e.g., some important chemical parameters may not have been monitored),
and since it is difficult to identify and treat upsets, which may be more
influential than average performance.
FIGURE 12-13
0L
Effect of Copper on IGC at Drilled TSPs in Non-Preheater Plants (Time to 0.2% Tubes Repaired) 0
Weibull Plot
!
.90 |I G
E 'V
'V
- 43 Plants With Copper (25
EC affected*) 'd
Median time to 0.2%
- 6 Plants Without Copper (3
Copper plants - 4.96 EFPY []
Ak affected*) _C•_
EY" 0.63
*Affected =with 0.2% or more tubes with Eli

CO 0.5
repairable defects due to IGA/SCC at TSPs
0~ \0] -0
--------
C: I I I
0
Median time to 0.2%
Eý C,opper free plants - 6.45 EFPY
U-
1
0C
0.2
DO
C El.
'V
]1
t-
.10 I I I1
0 I I I I I I
1 10
Time forO.2% TSP IGA/SCC
EFPY Adjusted to 620°F
0
In recent years, efforts to correlate IGA/SCC with impurities have been

based on hideout return studies, where the pH of crevices is calculated
using hideout return data and the MULTEQ computer program. One recent
study concluded that the correlation between crevice chemistry parameters
and IGA/SCC was not consistent (12.22). However, a later study concluded
that a few types of IGA/SCC show some correlation with crevice pH
calculated using hideout return data and MULTEQ. Increased corrosion
correlated with an increase in the product of effective full power time and
net pH (EFPY x net pH), where net pH is the difference between the crevice
pH at temperature and neutral pH at temperature (12.23). This result
indicates that, at least in some cases, increased exposure to highly alkaline
conditions is a factor in the observed IGA/SCC.
Condenser leakage can be a major source of impurity ingress. To date,
there has been no industrywide effort to develop a detailed understanding
of the effects of condenser inleakage on IGA/SCC performance. However,
at least a few cases exist where condenser inleakage clearly has been a
contributing factor.
Beznau Unit 1, which has tubesheet crevices, initially started up on
AVT-type water chemistry. Significant IGA/SCC occurred within the first
two years of operation. This corrosion was attributed to alkaline impurities
introduced by condenser inleakage. Many plants that are located at
freshwater sites, e.g., on the Great Lakes, have experienced IGA/SCC in
their tubesheet crevices. This IGA/SCC is generally attributed to alkalinity
developed by concentration of impurities introduced by condenser
inleakage. The Japanese attribute the absence of IGA/SCC at several of
their plants in part to the occurrence of seawater inleakage (12.11).
In 1992, molar ratio control was identified as a factor in the occurrence
or absence of IGA/SCC (12.24). Molar ratio control involves controlling
the ratio of strong cations to strong anions among the impurities
concentrating in crevice areas. Typically, impurities measured in the
blowdown and in hideout return are used in an effort to achieve
approximately neutral conditions in crevice areas. The concept is based on
the absence of IGA/SCC observed in three Japanese units that have operated
for up to 15 years; these units have practiced molar ratio control since initial
startup (12.25). Molar ratio control has been the standard practice in Japan
at least since the early 1980s. Yet, this practice has not prevented the
continued progression of IGA/SCC at three severely affected units, nor the
onset of IGA/SCC at another Japanese unit. Thus, the ability of molar ratio
control to prevent the progression of IGA/SCC once it has initiated (i.e., at
plants where molar ratio control was not implemented from initial startup)
is not clear. Molar ratio control is currently being implemented at several
plants in the USA; its effect on the progression of IGA/SCC at these plants
is expected to become apparent within a few years.
Water Treatment History

As discussed in Chapter 9, plants that started up in the late 1960s and
early 1970s using phosphate water chemistry experienced problems with
IGA/SCC and wastage. As a result, most plants switched to AVT about
1974. Around 1979, some of these plants began to detect IGA/SCC in
susceptible locations, such as in tubesheet crevices. Later, plants that had
never operated with phosphate water chemistry also began to experience
IGA/SCC. It has been difficult to determine how much the prior phosphate
treatment has influenced the potential for IGA/SCC. In fact, the experience
at different plants appears in some cases to be contradictory.
Units tubed with alloy 600 that continued to use phosphate water
chemistry treatment after 1974, (San Onofre Unit 1, H. B. Robinson Unit 2,
and Jose Cabrera [Zorita]) all experienced IGA/SCC. All three units also
experienced chemistry excursions in which excess NaOH developed in the
steam generators. In two cases, the intergranular corrosion has been
extensive, which has lead to large scale sleeving. It has also been a factor in
steam generator replacement. This experience indicates that remaining on
phosphate water chemistry would probably not have prevented the
occurrence of IGA/SCC at plants that switched to AVT, though it would
very likely have changed the specific mechanism and details of the attack.
With the exception of one tube, units tubed with alloy 800 tubing
(manufactured by Kraftwerk Union/Siemens) have not reported
intergranular corrosion. This includes units which originally operated with
phosphates as well as units which have always used AVT. Freedom from
IGA/SCC is probably mostly due to the steam generator design, use of a
different type alloy (nuclear grade alloy 800), absence of tubesheet crevices,
and the use of lattice (rather than drilled-hole) tube support structures which
minimize crevices at the supports. It is also probably related to the specific
parameters of the water chemistry control program used during the
phosphate water chemistry period, such as low sodium to phosphate molar
ratio, relatively low concentrations of phosphate, and low input levels of
impurities into the secondary system. The experience at Kraftwerk Union/
Siemens units is similar to that at Zorita, which has remained on low
concentration phosphates, and which has experienced very minor IGA/
SCC (Zorita's good experience may be due in part to its low hot leg
temperature of 583 0F). The IGA/SCC at Zorita mainly occurred during a
period of time when sodium hydroxide transients were reported; there has
been only minor occurrence of IGA/SCC since then.
Beznau Unit 1, which started up on AVT, experienced extensive IGA/
SCC in the tubesheet crevices during its first two years of operation. During
the three years that followed, when the plant was using sodium-phosphate-
treated water, the corrosion rate was either arrested or greatly reduced.
The operator, however, also changed operating and maintenance practices

so that water purity was greatly improved. Since converting back to AVT,
slow progression of intergranular corrosion has again been reported. For
this reason the steam generators are scheduled for replacement in 1993.
Many plants that operated on phosphate water chemistry and then
switched to AVT have subsequently experienced significant amounts of
IGA/SCC. For example, many tubes at Point Beach Units 1 and 2 and
Ringhals Unit 2 have all experienced IGA/SCC in tubesheet crevices.
Detailed evaluation of crevice deposits and fracture surfaces for Point Beach
Unit 1 have indicated the probability that sodium phosphate reacted with
magnetite in the tubesheet crevice to release sodium hydroxide, and that
the sodium hydroxide initiated the corrosion (which appears to have
continued to propagate under mild alkaline conditions) 12(L.26).
In some cases, sister units at the same station have been exposed to
different amounts of phosphate water chemistry and have had different
IGA/SCC experience.
In the case of Prairie Island Unit 1, prior phosphate correlates with
increased degradation. Unit 1 operated for a period of time on phosphate
water chemistry, while Prairie Island Unit 2 did not. Although the Unit 1
steam generator has the same type of tubing as Unit 2 it has experienced
significantly more severe IGA/SCC than Unit 2 (12.27).
For some units, however, prior phosphate correlates with decreased
degradation. Cook Unit 1, Doel Unit 1, Mihama Unit 2, and Takahama
Unit 1 operated on phosphate water chemistry only for a short time, while
their sister units, Cook Unit 2, Doel Unit 2, Mihama Unit 3, and Takahama
Unit 2 did not. The units on prior phosphate chemistry have experienced
significantly less severe IGA/SCC than their sister units (12.27). However,
it should be noted that, in some cases, there are other differences in addition
to water chemistry that may also be a factor, such as differences in
temperature and tube supplier. Evaluation of the phosphate effect indicates
that units that had prior phosphate experience took about 1.4 times longer
to reach 1% IGA/SCC at TSPs than those that have only used AVT (12.20).
More than one half of the plants with mill annealed alloy 600 tubing
that are now operating onAVT are experiencing IGA/SCC. It appears that,
for high temperature plants, this form of attack at drilled-hole tube support
plates, and possibly also at lattice-type supports, will become quite
extensive. This corrosion is attributed to the accumulation of impurities in
crevices despite operation with high purity feedwater. Moreover, most units
lack a buffer that would prevent the concentrated impurities from
developing an either highly caustic or highly acidic environment in the
crevices. In addition, evidence is accumulating that indicates that crack
propagation, and possibly initiation, can probably occur even with near
neutral crevice conditions.
In summary, IGA/SCC was a serious problem in early domestic alloy

600 units on phosphate water chemistry, and it continues to be a serious
problem at plants that switched from phosphates to AVT. In addition, plants
with mill annealed alloy 600 tubing that have only operated on AVT are
experiencing significant IGA/SCC. The only group of older plants with
RSGs not experiencing significant IGA/SCC are the Kraftwerk Union/
Siemens plants. These plants have used a modified phosphate water
chemistry and/or a high pH AVT water chemistry, and are equipped with
alloy 800 tubes and few crevices. It is not known whether a modified
phosphate water chemistry could be developed that would prevent the
occurrence of IGA/SCC in alloy 600 units with significant crevices.
Period of Operation Before First Occurrence-Aging of Material

As cited above, Beznau Unit 1 experienced IGA/SCC within two years
of startup. Zion Units I and 2, on the other hand, did not experience IGA/
SCC until after almost ten years of operation. These units operate at
relatively low hot leg temperatures, 598°F (314'C) and 591'F (311'C),
respectively. The Beznau and Zion experience indicates that the time to
occurrence is a function of many variables, especially those related to
secondary water purity.
For higher temperature plants, the occurrence of IGA/SCC early in life
appears to be more frequent than for lower temperature plants. For example,
several high temperature plants, i.e., hot leg temperatures of 610'F (321'C)
or more, are reporting IGA/SCC at support plates after only seven or eight
years of operation, while many lower temperature plants, i.e., hot leg
temperatures below 600'F (316'C), have not reported any IGA/SCC at
supports even after 15 or more years.
To date, aging of tube material has not been positively identified as a
factor in determining susceptibility to IGA/SCC. Examination of removed
tubes has not conclusively shown that changes are a function of extended
service exposure. Similarly, investigation of laboratory samples has given
mixed results. Examination of model boiler tubes that operated for many
years indicated that some relatively minor amounts of sensitization occurred
in some tubes after long times at typical hot leg temperatures (12.28). In
another investigation, no microstructural changes were noted in samples
that had been exposed in an autoclave at 630'F (332°C) for 73,250 hours
(12.14). However, Huey tests of these samples indicated that some
sensitization had occurred.
If sensitization is occurring due to service, it is probably a
low-temperature sensitization process with grain boundary precipitation
of chromium carbides. The following consequences may result:
An increasing susceptibility to intergranular corrosion in

low-temperature oxidizing environments involving partially reduced
sulfur species may develop. Sensitized material is known to be
susceptible in such environments, while mill annealed material is not.
" An increased resistance of alloy 600 to intergranular corrosion in caustic
environments may develop since grain boundary carbides appear to
provide increased resistance to caustics (see discussion on
microstructure later in this chapter). Whether the increase would be
significant is not known.
Presence/Use of Full Flow Condensate Polishers, Problems With

Other Ion Exchangers
Both IGA and IGSCC have occurred in units with and without full flow
condensate polishers (FFCP). Although no industrywide correlation
between polishers and intergranular corrosion has been established, except
for the case of sulfur attack on the sensitized tubing material found in OTSGs,
some experience indicates that the occurrence of IGA/SCC has been
influenced by polishers.
A detailed review of Japanese experience indicates that the use of
condensate polishers from initial plant startup is beneficial in preventing
the occurrence of IGA/SCC (12.11). According to the Japanese, the polishers
minimize the introduction of caustics and oxidants such as copper oxides
and hematite into the steam generators.
In at least one case, rapid, caustic-induced IGA/SCC has been related
to the effect of condensate polishers on water chemistry. The unit was
chemically cleaned (copper removal only) during a refueling outage.
Following the outage, a condensate polishing system went into operation
for the first time at this unit. During the initial part of the fuel cycle, sodium
leakage from the polishers was unusually high. After a few months of
operation, a forced outage occurred due to IGSCC just above the hot leg
flow distribution baffle. Inspection revealed that several hundred tubes
had developed detectable IGA/SCC. Examination of pulled tubes indicated
that mild alkaline conditions may have been involved.
As indicated above, the use of condensate polishers may or may not
contribute to the prevention of IGA/SCC. It should be noted that the design
of the polishers and the manner in which they are operated probably are
more important than whether polishers have been installed or not. Detailed
attention should be given to the design and operation of condensate
polishers to ensure that leakage of impurities and resin fines is minimized,
and that anions and cations are kept in balance.
Problems with other ion exchangers can also be a factor in the occurrence
of IGA/SCC. For example, sodium hydroxide ingress is suspected of having
caused some IGA/SCC at tubesheet crevices and TSPs at Jos6 Cabrera
(12.29); the sodium hydroxide ingress was possibly the result of a
malfunction of the makeup demineralizer resin regeneration system.
Use of Boric Acid

A number of plants have operated for long periods of time using on-line
additions of boric acid to inhibit denting. In some of these plants, e.g.,
Maine Yankee and Fort Calhoun, little IGA/SCC has been detected.
However, it is not clear whether the absence of significant IGA/SCC is due
to the use of boric acid or the conditions that caused denting in the first
place (e.g., presence of chlorides). Other improvements to the water
chemistry in response to denting could also have been an influential factor.
On-line additions of boric acid were initiated around 1985 in three
Japanese plants that are experiencing significant IGA/SCC at TSPs. The
rate of new IGA/SCC indications appears to have decreased with the use
of boric acid (Figure 12-14) (12.18). Farley Unit 2 and Cook Unit 2 were
also reported as experiencing a reduction in IGA/SCC at TSPs after initiating
on-line boric acid, which tends to support the hypothesis that boric acid is
responsible for the reduced rate of corrosion in these plants (12.30). A more
recent study of available experience concluded that use of boric acid was
effective at mitigating IGA/SCC at tube supports of some (but not all) units.
It was effective for tubesheet crevices but not for areas on top of the tubesheet
(12.31). The study further noted that, in many cases, other water chemistry
changes had been made at the time when boric acid was in use, such as
increasing the concentration of hydrazine in the feedwater. Thus it was not
possible to separate the effects of these changes from the effect of boric
acid.
While boric acid has appeared to be beneficial in some cases, significant
IGA/SCC has still occurred at several units on boric acid, especially high-
temperature units, such as North Anna Unit 1. At this unit, examination of
removed tubes indicates that dented TSP intersections have developed
secondary side as well as primary side cracks, despite the introduction of
boric acid after the first fuel cycle. Secondary side IGA/SCC has also
occurred at Asco Units 1 and 2 which began using on-line boric acid early
in life to prevent denting. Still they have experienced a relatively rapid
0
increase in the number of tubes affected by IGA/SCC at TSP9. In addition,
severe IGA/SCC at TSPs has continued to progress at plants such as Trojan
and Catawba Unit 1, although a boric acid injection program was initiated
as soon as secondary side IGA/SCC was detected.
FIGURE 12-14
IGA/SCC at TSPs at Japanese Plants
Weibull Plot
.90 - _
b=2.7
--------------- - -- - - - - - - - 1ý - - - - - - - - - - - - 0 .63
b=2.2 Wt orcAi
orcAi
PlantO0Wt
b-2.3-0.
.10
b=2.2
Fraction
on -------------------------------------------- ----------------------
----- 0.05
Tubes
M No Boric Acid
.001
10 100
Service Time (EFPY)
Laboratory tests show that on-line boric acid is beneficial in preventing

the initiation of caustic-induced IGA/SCC. It can also be beneficial in
reducing the rate of growth of already initiated cracks in caustic
environments. This laboratory experience coupled with the mixed plant
experience cited above leads to the conclusion that use of boric acid should
0
be helpful at plants experiencing caustic-induced IGA/SCC, but may not
be of significant benefit at plants' experiencing acidic-induced IGA/SCC.
Use of Morpholine
Morpholine is applied widely to control secondary system pH. Because
it partitions more equally between the liquid and vapor phases than
ammonia (ammonia remains mostly in the steam), it is more effective at
keeping pH elevated in steam drains and in other areas where steam
condenses. As a result, use of morpholine results in less corrosion of carbon
steel surfaces in steam drain systems, and lower input of iron corrosion
products into the steam generator.
To date, morpholine has not exhibited any negative effects, although it
decomposes to form various organic acids (12.32). It is expected that the
reduction in iron corrosion product input reduces the formation of sludge
piles, and thus reduces the risk of IGA/SCC in sludge piles. The decreased
need for sludge lancing should also be of benefit, reducing the exposure to
oxidizing conditions. It is also expected that the reduction in iron corrosion
product input will reduce the rate of clogging of crevice areas, and will
thus reduce the amount of IGA/SCC at supports.
Despite the expected benefits of morpholine, it should be noted that at
least one domestic plant which started up on morpholine, Beaver Valley
Unit 1, is experiencing significant amounts of IGA/SCC at TSP crevices
and in sludge piles. The MSR in this plant has copper alloy tubing and the
plant has no polishers. Which (if any) of these variables is involved in the
IGA/SCC is not known. It should also be recognized that many French
units have used morpholine-for many years but are nevertheless
experiencing an increasing amount of IGA/SCC at TSPs and in sludge piles.
In summary, it is possible that morpholine may help to minimize the
likelihood of IGA/SCC by reducing the accumulation of sludge and
clogging of crevices. This, in turn, will reduce the need for sludge lancing.
Elevation in Tube Bundle

A strong decrease in the number of ECT indications at TSP crevices is
commonly noted on the hot leg side as elevation increases, i.e., there are
more IGA/SCC indications at lower elevations than higher up in the tube
bundle. The rate of decrease from one tube support plate elevation to the
next higher is typically a factor of two or three. The reasons for the decrease
are not yet fully understood but may include:
* decrease in temperature of about 6°F (30 C) from one tube support
elevation to the next;
* decrease in superheat in crevices as elevation increases, with
corresponding decrease in concentration of impurities in equilibrium
with the superheat;
* decrease in thermal stress associated with decreasing heat flux as
elevation increases;
* the stability of liquid phases or liquid films in the crevice may be limited
to particular ranges in superheat;
* increase in flow velocity through, and differential pressure across, the
support as elevation increases, tending to improve flushing of crevices
at higher elevations;
* different patterns in accumulation of tube deposits as support elevation
increases; in some cases, however, the data indicate that tube scale tends
to increase with elevation; and
* change in oxidant availability or electrochemical potential as elevation
increases. (This could be the result of changes in concentrations of
copper metal and oxides at different elevations as the result of the
retrograde solubility of some copper ions.)
Somewhat surprisingly, a similar decrease in the number of IGA/SCC
indications as elevation increases has been observed by ECT in the cold leg
of one Combustion Engineering unit. Since temperature and heat flux
increase as elevation increases on the cold leg side, the first three factors
listed above clearly do not apply to the cold leg situation (They would result
in the opposite variation to that observed). The other variables are all
speculative. A decrease in electrochemical potential or oxidant availability
appears to be the most likely explanation for the cold leg variation. This is
supported by observations in some plants that more copper metal and
copper oxide are present on the cold leg side than on the hot leg side.
Summary for Units With Mill Annealed Alloy 600 Tubing

As discussed above, quantitative correlations of plant factors for
susceptibility to IGA/SCC, except for correlation with temperature, have
not been particularly successful. However, review of plant operating and
design factors has provided some useful understanding of the factors
involved in the occurrence of secondary side IGA/SCC. A summary of the
information is provided below.
* Two domestic plants with mill annealed alloy 600 tubing that remained
on phosphate water chemistry have experienced severe IGA/SCC,
despite low operating temperatures (Chapter 9). This indicates that
phosphate water chemistry of the type practiced in the USA may not
be suitable for steam generators with mill annealed alloy 600 tubing,
especially if significant sludge develops in the unit and if drilled-hole
tube supports are present. However, the successful experience with
phosphate water chemistry in Kraftwerk Union/Siemens units and at
Zorita indicates that operation with phosphates is possible with
appropriate water chemistry parameters and appropriate steam
generator and secondary system design details.
" For early plants with tubesheet crevices, freshwater sites with alkaline-
forming cooling water generally experienced more IGA/SCC than
seawater plants, and at an earlier time in plant life. However, there
have been some significant exceptions (Ringhals Unit 2 and several
Japanese plants). The reason for these exceptions may be that chemistry
in the affected plants are dominated by makeup and polisher impurities,
rather than by seawater inleakage.
* For modern plants with leak-tight condensers, the type of cooling water
does not appear to be an important factor with regard to occurrence of
IGA/SCC. For these plants, variables such as hot leg temperature,
makeup water impurities, polisher/demineralizer usage and leakage,
presence of copper in the secondary system, molar ratio of impurities
in the feedwater and, possibly, use of boric acid are believed to be the
most important factors affecting the susceptibility to IGA/SCC.
* Specific causative factors involved in IGA/SCC at modern
high-temperature plants with leak tight condensers have not been
clearly identified. Based on detailed studies by the utilities involved,
caustic is suspected in Japan and France. Acidic environments are
suspected in several other cases (e.g., Doel Unit 4, Tihange Unit 1, North
Anna Unit 1, and ANO Unit 2). Elevated electrochemical potential has
been identified as a major factor by the Japanese, and may be an
important factor in other cases. However, current Japanese experience
indicates that IGA/SCC can continue, once initiated, even with a fully
reducing environment.
" In many cases, a source of sodium or potassium ions in excess of that
neutralized by appropriate anions appears to be a factor that IGA/SCC-
affected plants have in common. Studies of the anion-cation balance in
crevice deposits, however, have often failed to produce satisfactory
quantitative results.
" An increasing number of studies are identifying acidic conditions, lead
contamination, and near neutral conditions as possible environments
involved in IGA/SCC propagation.
FAILURE ANALYSIS OF SAMPLES

Examination of tubing samples removed from operating steam
generators has confirmed the presence of IGA/SCC in numerous steam
generators. The corrosion has been detected mainly on the hot side of the
tube bundle. Initially, it was found mostly within the tube-to-tubesheet
crevice, and with less frequency within the sludge pile or under deposits
residing on tube surfaces in the region above the sludge pile. Recently, it
has also been discovered in many tube-to-tube support crevices, particularly
in higher temperature tubesheet crevices, it may extend relatively uniformly
over essentially the full depth of the crevice, about 20 inches (500 mm), or it
may occur on relatively localized patches of tubing within the crevice. In
several instances, IGSCC is associated with IGA; both can be initiated on
the tube's secondary side. In sludge piles, IGSCC is generally axial in
orientation and relatively short in length, though it sometimes extends
irregularly for several inches. In tube supports, the IGSCC generally is
axial in orientation and limited in length to less than the height of the
supported region (3/4 inches in Westinghouse units and 2 inches in
Combustion Engineering units). However, if denting occurs, IGSCC can
extend beyond the support boundary.
Nondestructive evaluations of the integrity of the tubing in tubesheet
crevices and in sludge piles have been relatively successful in quantifying
the depth of IGA/SCC once the IGA/SCC reaches significant depths. At
shallow depths, however, detecting IGA/SCC is difficult, since IGA/SCC
is characterized by extremely tight penetrations of grain boundaries.
Quantifying the depth of IGA/SCC in tube support plate crevices has been
more difficult than in tubesheet crevices and sludge piles. This is due to
the complexity of the geometry, materials interactions, and the nature of
the corrosion itself.
A large number of tubes have been pulled at various plants and
examined. Based on the results of these examinations, there are numerous
reports available describing the morphology of the observed IGA/SCC,
the nature of the deposits and corrosion films, as well as other related
information. Results of tube pull examinations representing different areas
of the steam generator and plant type are summarized below.
Tubesheet Crevices-Recirculating Steam Generators

Freshwater-cooled units have often behaved quite differently than those
cooled by seawater. Therefore, these two categories of plants are treated
separately.
Freshwater Plants. Secondary side IGA/SCC first occurred in tubesheet

crevices of units operating on AVT. In 1971, it was detected at Beznau Unit 1
which initially started up on AVT-type water chemistry (zero solids). Within
the first two years of operation, a significant number of tubes were affected
by IGA/SCC. Sections were removed from the hot leg side of a Beznau
Unit 1 steam generator in 1971 and 1972 to determine the cause of tube
leaks. Examinations showed axially oriented IGA and IGSCC with
extensive, branching networks. Microchemical studies of deposits on the
tube surfaces showed the presence of hardness salts and the alkali metal
ions Na+ and K+. It was concluded that the corrosion was caused by the
alkaline deposits. The surface pH on the tube samples ranged from 5.5-12.0
with the bulk of the measurements indicating alkalinity. The Swiss utility
subsequently converted their secondary water treatment practice to
coordinated phosphate chemistry control and greatly improved their water
chemistry and air inleakage control programs. Subsequent operation, first
on phosphate chemistry control and since 1974 on AVT, resulted in much
purer feedwater, and the rate of corrosion was significantly reduced.
However, corrosion has continued to occur at relatively high rates and has
resulted in repair of more than 20% of the tube population.
Several other freshwater-sited plants operated first with phosphate
water chemistry and then with AVT. After several years on AVT these units
began to experience IGA/SCC in tubesheet crevices. The time period after
the switch to AVT and before significant numbers of tubes were affected
has varied from unit to unit, e.g., from about five years for Point Beach
Unit I to ten or more years for other units; however, all of the units in this
category have now experienced IGA/SCC in tubesheet crevices. The
variation in IGA/SCC behavior may be due to temperature differences,
differences in secondary system water purity, which is affected by plant
design, and varying operation practices, or differences in the amounts of
sodium phosphate in crevice areas which, in turn, are residuals from earlier
years of operation on phosphate water chemistry.
Tubes removed from this group of plants have exhibited IGA, IGP, and
IGSCC, with IGA being the most prevalent mode of corrosion. Typical
examples of the observed corrosion are shown in Figures 12-1 to 12-4.
Analyses of crevice deposits and corrosion films have generally supported
the conclusion that the IGA/SCC has been due to caustic. Some of the
specific observations in this regard are as follows:
" Significant amounts of sodium and potassium have been detected on
fracture surfaces and in adjacent deposits, although not in all cases.
* Chromium depletion of oxide layers has been observed in some cases.
* The pH of wetted tube surfaces has generally been alkaline, sometimes
highly alkaline.
Examination of crevice deposits many years after switching from

phosphate water chemistry has shown significant amounts of phosphate
still present in the crevices. This illustrates the difficulty of eluting material
from deep tubesheet crevices.
Much of the IGA and IGP observed in the tubesheet crevices has been
relatively uniform (non directional), indicating little influence of stress.
However, in tubes removed from several units, long axially oriented IGSCC
has been found in the straight sections of the tubes, e.g., in Kewaunee and
Prairie Island Unit 1 (Figure 12-15) (12.33). The IGSCC indications have
measured as much as 16 inches (41 cm) or more in length, and tend to occur
in three locations around the circumference. This finding is similar to that
discussed later in this chapter for Ringhals Unit 2. During the failure
analysis, three axial groupings of striations were observed around the
circumference of the tube. The major cracks appeared to have propagated
along the~e striations. The stress and cold-work situation leads to relatively
rapid growth in the length direction of partially through-wall cracks, which
then grow more slowly through the wall. Abrasive polishing of the OD
surface of tubes develops a disturbed cold-work layer of about 0.001 inch
(25 mm) in depth, in which cracks can apparently initiate and grow easily.
In addition, polishing and straightening of tubes sometimes lead to relatively
high residual hoop stresses on the OD. Pressure stresses act throughout
the wall and tend to keep cracks growing, even after they grow out of the
higher stresses and cold work on the surface.
FIGURE 12-15a
Long Axial IGSCC in Alloy 600 Steam Generator Tube (OD Surface)
FIGURE 12-15b
Long Axial IGSCC in Alloy 600 Steam Generator Tube
(Cross Section View)
Intergranular corrosion has often affected much of the depth of the

crevice; it sometimes reaches down to the roll transition at the bottom of
the crevice. The corrosion is often less extensive close to the top of the
crevice. When the corrosion reaches down to the roll transition, it can be
circumferentially oriented, as found by detailed evaluation of a group of
tubes removed from a retired Point Beach Unit 1 steam generator
(Figures 12-16 and 12-17) (12.34). The main results of the Point Beach Unit 1
analyses are listed below (12.26).
" The depth of IGA in the tube walls in the crevice ranged between 20 to
85% of the wall thickness.
" The corrosion attack in the tube walls above the crevice region was
minimal.
" The tube deposits in the crevices contained primarily Fe304, Na2SO4,
NaFePO4, and NaHFeO2. The sodium content of the deposits was about
10%.
* It was concluded that caustic was the most likely corrodent, and that
its presence could be due to a reaction of sodium phosphates with
magnetite.
* The depth of attack of the tubesheet material was minimal.
* Examination of the tube microstructure has shown the tubing to be
typical mill annealed material and not sensitized.
FIGURE 12-16
SCC and IGA at Roll Transition of Part-Depth-Rolled Alloy 600 Tube
F-87-1199 To 1201
Examination of removed hot leg tubes from Kewaunee, another

freshwater-cooled plant, showed long cracks in the crevice region (12.35).
The depth of the cracks in the crevice ranged up to 74% of the wall thickness.
The corrosion attack in the tube walls immediately above the'tubesheet
was deeper than in the crevice and consisted of multiple IGSCC with strong
IGA characteristics. It was concluded that the attack was most likely due
to alkaline conditions. This conclusion was based on the evidence of
moderate chromium depletion and the presence of compounds, especially
silicates, which are not stable in acidic or neutral conditions. Modified Huey
tests showed that the tubes were not sensitized.
A cold leg tube removed from Ginna in 1980 (12.36) exhibited no
intergranular corrosion. It was not determined whether this lack of
corrosion, as compared to the corrosion observed at hot leg crevices, was
due to differences in the environment, or whether it was merely a response
to a lower operating temperature. The concentrations of sodium and
potassium in crevice deposits on the hot and cold leg side were relatively
similar.
FIGURE 12-17
IGP and IGA at Roll Transition of Part-Depth-Rolled Alloy 600 Tube
F-'87-1193 I-*j 50um

Fully Expanded
F-87-1195 - 50urn
Approx. 10 mm Above Taper

Seawater Plants on AVT. Table 12-2 provides the IGA/SCC experience of

units that initially had deep tubesheet crevices (part-depth rolls) and are
located at seawater or brackish water sites.
TABLE 12-2
IGA/SCC Experience: Seawater, AVT, Part-Depth Rolled Plants,
Tubesheet Crevices
Comm. Prior Hot Leg Status of IGA/SCC in

Unit Oper. Phos. (OF) Tubesheet Crevice
Mihama 2 7/72 Yes 599 Most of crevice re-expanded,
moderate IGA/SCC
Surry 1 12/72 Yes 605 SG replaced because of denting;
minor IGA observed
Turkey Pt. 3 12/72 Yes 602 SG replaced because of denting;
no reported IGA/SCC
Surry 2 5/73 Yes 606 SG replaced because of denting;
no reported IGA/SCC
Turkey Pt. 4 9/73 Yes 602 SG replaced because of denting;
no reported IGA/SCC
Indian Pt. 2 7/74 Yes 576 No reported IGA/SCC
Takahama 1 11/74 Yes 608 Most of crevice re-expanded;
moderate IGA/SCC
Doel 1 2/75 Yes 598 No reported IGA/SCC
Ringhals 2 5/75 Yes 609 Severe IGA/SCC;
led to SG replacement
Genkai 1 7/75 No 599 Severe IGA/SCC
Takahama 2 11/75 No 608 Most of crevice re-expanded;
severe IGA/SCC
Doel 2 12/75 No 598 Minor IGA/SCC
Indian Pt. 3 8/76 No 595 SG replaced because of pitting;
minor IGA observed
None of the six domestic units with tubesheet crevices and seawater or
brackish cooling water have reported significant IGA/SCC in the
crevices. However, steam generators at five of these units have been
replaced as a result of denting or pitting, and the sixth unit has
experienced significant degradation caused by these phenomena.
Denting and pitting are both indicative of acidic, oxidative conditions

caused by seawater inleakage and either copper in the secondary
systems or air inleakage. This experience suggests that, if seawater
inleakage occurs, IGA/SCC will not be a significant problem, at least
in the short term. This conclusion was confirmed by comparison of
tube and tubesheet samples removed from the Surry Unit I and Point
Beach Unit I steam generators; these samples showed only minor IGA
of the Surry Unit I tubes, apparently as the result of chloride ingress
(12.26), while substantial IGA had occurred in the Point Beach tubes,
where chloride ingress was not as significant.
" Ringhals Unit 2 steam generators have been replaced because of the
IGA/SCC experienced in the tubesheet crevices. Possible factors in
this experience are discussed later.
" All four of the Japanese plants with tubesheet crevices (part-depth rolls)
have experienced IGA/SCC, despite their seawater location. As a result
of the IGA/SCC, the tubes in three of these units were expanded after
several years of operation to close the crevice for most of the tubesheet
height; however, enough of a crevice remains to make IGA/SCC at the
top of the crevice a severe problem at some of the plants. Seawater
inleakage in theseplants is minimal as a result of extensive condenser
inspections and maintenance. This experience indicates that IGA/SCC
still can occur in seawater plants even if seawater inleakage is minimal.
" Two units, Doel Units 1 and 2, have experienced either none or very
minor IGA/SCC despite many years of operation. This indicates that
it is possible to avoid serious problems for at least 15 to 16 years in low-
temperature plants 598°F (314'C). The feedwater in these two plants
contains substantial amounts of silica and these units are equipped with
full-flow condensate polishers.
* All units with hot leg temperatures above 600'F (315.6°C) that have
operated for more than ten years have experienced moderate to severe
IGA/SCC in tubesheet crevices.
Of these units, Ringhals Unit 2 has been one of the most severely
affected. Large numbers of tubes from both the hot leg and cold leg sides
have been removed and examined. The results of these examinations are
as follows (12.37):
The corrosion in the crevices was typically IGSCC, with only minor
IGA. Minor IGA extended for the full depth of the crevice. Generally,
IGSCC was localized along axial lines and was deeply penetrating; in
some cases it extended for much of the crevice depth. Figure 12-18
illustrates a typical crack pattern. The axial cracks were found to be
associated with surface striations, similar to those observed at Point
Beach Unit 2 and Prairie Island Unit 1.
FIGURE 12-18
Crack Patterns Observed in Alloy 600 Steam Generator Tube
Metallographic
OD-defects Examination
Examined cross sections:
400 94mm 180mm 239mm 313mm
Deepest crack observed:
1.1mm
Axial cross sections:

4 at 90deg 33-46mm
- at 270deg 33-46mm
300
Deepest crack obsedved:
.6mm
0 3 Intergranular corrosion
U) on the 0D-surface on all
0 cross sections.
CL
Fa 200
,• 2 ID-cracks confirmed.
100 I 1 ID-defects
II'' i,
n II
,I I
0 I I I I I
0 90 180 270 380 0 90 180 270 360

Angular Position (deg)
* For several tubes which had not experienced primary to secondary

leaks, and thus had not been affected by primary water leakage, the
sodium to phosphate molar ratio was about 7.
* For several tubes which had not experienced primary to secondary
leaks, and thus had not been flushed, tubes with extensive cracking
had low soluble to insoluble phosphate ion ratios in their crevice
deposits compared to less affected tubes, and had high concentrations
of iron phosphates. It was concluded that iron phosphates may have
existed in the crevice. It should be noted that the reaction of sodium
phosphate with iron can result in the formation of sodium hydroxide.
• The pH of the soluble deposits was moderately alkaline (sometimes

about pH 10). Using MULTEQ, the pHT of a typical, concentrated
solution in the crevice was below 10.
It is difficult to postulate an alkaline crevice environment in a seawater-
situated steam generator, unless full-flow condensate polishers are used,
or all inleakage of seawater is prevented, and makeup water dominates the
water chemistry. Seawater is known to result in acidic conditions after
concentration. Analyses have suggested that the crevice environment in
the Ringhals Unit 2 tubesheet crevices was perhaps more characteristic of
the makeup water supply than of seawater.
The superheat in tubesheet crevices is so high, about 75°F (42 0C), that
acidic species are not stable; they volatilize or precipitate at about 30 to
45°F (17 to 25°C). Thus, it is believed that the acidic conditions associated
with seawater inleakage do not result in IGA/SCC in tubesheet crevices.
This conclusion is supported by the absence of detected IGA/SCC at plants
with known seawater leaks, by the presence of only very minor IGA (3% of
wall thickness), and the presence of chlorides in the tubesheet crevice area
of tubes removed from a retired Surry 1 steam generator (12.26).
Examination of a removed tube from Takahama Unit 2 (12.38) with a
primary-to-secondary leak shows the presence of copper metal and
magnetite in the tubesheet crevice, an indication that conditions were fully
reduced. Analyses of hideout return data indicate that the crevice pH during
operation should have been essentially neutral. Thus, it appears that crack
propagation may have occurred under neutral, fully reduced conditions.
This is the same conclusion reached by the Japanese regarding IGA/SCC
propagation at tube support plate crevices, as discussed later.
Top of Tubesheet of Full-Depth Expanded Plants (OD of Expansion

Transition at Top of Tubesheet)-Recirculating Steam Generators
Combustion Engineering Units. Circumferential cracks have been

detected by ECT at the top of the tubesheet in several hundred hot leg and
several hundred cold leg tubes at Millstone Unit 2. 'This unit is seawater-
cooled. Three cold leg tubes and one from the hot leg side were removed
from Millstone Unit 2 and destructively examined. The cracks were found
to be GD-initiated in the explosive transition region. The causative species
involved in crack initiation were not positively identified. Caustic and acidic
conditions are both suspected, based on chemistry studies from different
periods of operation. Evaluation of chromium-to-nickel ratios at fracture
faces indicate that crack propagation may have taken place under acidic
conditions. In addition, some lead and transgranular cracking were noted;
thus, lead may have played a role. The stresses involved in the cracking
are believed to be due to denting at the top of the tubesheet and tube bending
caused by tube bundle shifting following denting at eggcrate supports.
Several years before the circumferential cracks were detected at
Millstone Unit 2, examination of a pulled tube from that unit showed a
band of IGA immediately above the top of the tubesheet. The tube had
been located in the sludge pile region. A band of wastage and a band of
pits were found above the band of IGA (Figure 12-19) (12.39), with isolated
pits above that elevation. This behavior suggests a gradient in
electrochemical potential, which led to the different forms of attack; this.
gradient may be affected by the tubesheet-to-tube material difference.
FIGURE 12-19
Schematic of Hot Leg Tube Corrosion at Millstone Point Unit 2
Schematic of Typical
Hot Leg Tube Corrosion
0
0
Pits
} Band of Pits
Wastage {
/ IGA
Tubesheet
Recently, circumferential ODSCC at the top of the tubesheet has also

been detected in numerous tubes at ANO Unit 2 and St. Lucie Unit 1. ANO
Unit 2 is a freshwater plant and St. Lucie is a seawater-cooled plant.
Evaluation of tubes removed from ANO Unit 2 indicate that the SCC was
intergranular, that it was due to acidic conditions, with lead and sulfur
present, and that no IGA was present (12.40).
Westinghouse-Type Units. Circumferentially oriented IGSCC initiating

from the OD was first detected in a French unit at explosive expansion
transitions. Subsequently, it was detected in the sludge pile zones of several
French kiss-rolled units. The specific factors causing the OD cracks have
not been identified. It should be noted that the OD cracks were always
associated with ID cracks, which are more numerous and have received
most of the attention. In a few cases, the ID and OD cracks have been
associated with denting at the top of the tubesheet in the sludge pile region.
This denting may have been caused by corrosion of iron shot introduced
into the steam generator prior to plant start up.
In recent years, circumferential IGA/SCC has been observed at the OD
of standard and kiss roll transitions in a number of recently examined tubes
in Spanish, domestic, and French units with freshwater cooling, and one
Belgian unit with seawater cooling. Much of this IGA/SCC was not detected
by ECT performed in the plant. For Asco tubes, evaluation of cracks in
some cases indicated chromium depletion, carbonates, and metal
hydroxides; these were believed to indicate that caustic was the probable
cracking agent, although some lead was observed and could have been a
possible causative agent as well (12.41). Another analysis did not find
evidence of chromium depletion or presence of caustics, and thus did not
confirm the hypothesis that cracking was due to caustics (12.42). In the
case of Doel Unit 4, which is a seawater-cooled plant that has experienced
severe circumferential IGA/SCC at the top of the tubesheet, examination
of pulled tubes indicates that the cracking was due to acidic conditions,
probably aggravated by lead (12.43). The circumferential IGA/SCC has,
for the most part, not been associated with sludge nor with denting. Thus,
it apparently is the result of residual stresses caused by tube expansion,
although the tubes at Doel Unit 4 had been kiss rolled. (The French and
Belgians adopted kiss rolling specifically to reduce such stresses.) The
environment leading to the IGA/SCC is apparently the result of
concentration of impurities in the small crevices and restricted flow zones
that may exist at the tube-to-tubesheet joint of some tubes, even in the
absence of significant sludge accumulation.
Hot Leg Sludge Piles-Recirculating Steam Generators
Freshwater Plants on AVT. Axially oriented IGA/SCC has been detected

by ECT in many freshwater plants in sludge pile areas on the tubesheet
and, in several cases, on flow distribution baffles and TSPs. In older units
with tubesheet crevices, the IGA/SCC in the sludge pile has been less severe
than that in the crevices. In newer units, without tubesheet crevices, the
IGA/SCC has sometimes been more severe in sludge piles than in tube
support plates and, in other cases, less severe. This difference in behavior
is probably related to differences in the size of the sludge pile and its
composition. In plants with large sludge piles and high operating
temperatures, IGA/SCC in the sludge piles has become a serious issue.
Examination of pulled tubes has led to the following observations:
* The IGA/SCC tends to be axially oriented. However, it can sometimes
span a significant circumferential extent.
" A variety of causative species have been identified at different plants.
In a few cases, mixed mode transgranular/intergranular cracking has
been seen, and lead has been identified (e.g., Bugey Units 2 and 3). In
some cases, caustic has been identified as the probable cause (e.g., Krsko
and Cook Unit 2). In still other cases, acidic environments have been
identified (e.g., Tihange Unit I and ANO Unit 2).
Seawater Plants on AVT. IGA/SCC in sludge piles of seawater units is

becoming a major concern. This group of plants mainly consists of older
Combustion Engineering units. Some older Westinghouse steam generators
at seawater sites that have been replaced with new steam generators did
not experience any detectable IGA/SCC attack prior to replacement.
Apparently, other Westinghouse-type units (e.g., in Japan and Sweden) have
very small sludge piles, and have not experienced attack, though units in
France, Belgium, and Brazil have seen some attack in recent years. The
older Combustion Engineering units have experienced the following IGA/
SCC problems in the sludge pile on the hot leg side:
" Increasing numbers of IGA/SCC indications have been detected in the
sludge pile area. Average growth rates of these defects have been in
the range of 4% of the wall thickness per year (12.44).
* Examinations of pulled tubes have shown patches of IGA, and
occasional axially oriented IGSCC. In one plant, St. Lucie Unit 1, mixed
mode cracking was discovered. Copper has generally been found on
-the tube surfaces. Lead has also been noted on crack surfaces on a
number of tubes (12.6).
* In one case, IGSCC was found to be associated with "kinking" of a

tube as a result of denting-induced deformation (Millstone Unit 2)
(12.39).
* At Millstone Unit 2, bands of IGA, wastage, and pits were located
immediately above the top of the tubesheet (12.39) (Figure 12-19).
* Reviews of plant water chemistry and hideout data have suggested
that IGA/SCC initiated mostly under acidic conditions, and has
continued to grow under near neutral conditions. In one case, Millstone
Unit 2, an increase in the amount of cracking following one cycle was
attributed to probable caustic conditions.
Hot Leg Tube Supports-Recirculating Steam Generators
Freshwater Plants on AVT. In this category of plants, IGA/SCC at tube

supports is becoming a major industry concern. With few exceptions, older,
low-temperature plants have not experienced significant IGA/SCC in this
area. Newer plants, however, are experiencing a relatively rapid increase
in the numbers of tubes affected.
Among freshwater units with initial operating hot leg temperatures
below 600'F (316'F), only Palisades has experienced severe IGA/SCC at
tube supports. The Palisades case is considered unique because of its
sensitized tubing and its water chemistry history (phosphate sodium sulfite
in early years). In addition, it is the only Combustion Engineering plant
with all drilled hole tube supports. Evaluations of the IGA/SCC at Palisades
indicate that most of the damage may have occurred early in plant life,
possibly during layup periods, due to reduced sulfur species. Progression
of the IGA/SCC has been relatively slow. Nevertheless, the steam generators
were replaced in 1990.
After about 20 years of operation, Kewaunee, a freshwater unit with a
low hot leg temperature of 590'F (310'C), has recently begun to detect
significant IGA/SCC at TSPs. As a result, about 1.7% of the tubes will be
repaired in the spring 1993. Evaluations of IGA/SCC in tubesheet crevices
and at sludge piles indicate that the IGA/SCC is due to alkaline conditions;
the same cause probably applies to the IGA/SCC at TSPs.
As plants age, an increasing number of units with intermediate
temperatures are experiencing IGA/SCC at supports. For instance, Cook
Unit 2, a freshwater unit with an intermediate hot leg temperature of 606'F
(319'C), experienced significant IGA/SCC at tube supports. The steam
generators of this unit were replaced in 1988 as a result of IGA/SCC in
tubesheet crevices and at tube support plates. Examination of pulled fubes
did not conclusively identify the causative agent (12.45), but it was presumed
to be caustic. A range of countermeasures was implemented, including
reductions in power level and temperature, replacement of condenser tubes,

and use of on-line boric acid; defect growth slowed considerably before the
steam generators were replaced.
Other freshwater units with similar temperatures, Farley Units 1 and
2, and Beaver Valley Unit 1 (all with hot leg temperatures of 607°F [319°C]),
have also experienced significant amounts of IGA/SCC at TSPs. Based on
water chemistry history, it was concluded that the corrosion at Farley Unit 2
was due to caustic (12.46). Examination of pulled tubes has not been able
to confirm the corrodent causing the attack; but various types of evidence
point to caustic, mild alkalis, lead, and acids (12.47, 12.48). However, the
utility believes that the use of boric acid, together with other remedial actions
such as crevice flushing and molar ratio control, have reduced the
appearance of new defects (12.49). However, a significant number of tubes
continues to require repair each outage. At Beaver Valley Unit 1, IGA/SCC
has also become a serious problem, with large numbers of tubes requiring
repair, despite the use of boric acid.
ANO Unit 2, a freshwater unit with eggcrate supports, HTMA tubing,
and a hot leg temperature of 607'F (319'C), has recently started to experience
IGA/SCC at the supports. Examination of pulled tubes indicates that the
attack is due to acidic environments (12.40).
An increasing number of high-temperature freshwater units with mill
annealed (non-sensitized) tubing and hot leg temperatures over 610'F
(321'C) have started to experience significant IGA/SCC at tube supports.
To date, this problem has mainly affected units of Westinghouse design,
with drilled-hole tube support plates (attack has recently been detected at
Palo Verde Unit 2, a Combustion Engineering unit). The first high-
temperature freshwater units reporting this problem were North Anna
Unit 1 and Fessenheim Unit 1 in 1985 and 1986. Since then, it has affected
many other domestic units (e.g., Trojan, Catawba Unit 1, North Anna Unit 2,
Byron Unit 1, V. C. Summer, Braidwood Unit 1, McGuire Unit 1), most
French units with drilled-hole TSPs, all four Spanish units with Model D
steam generators, two units in Belgium, and a unit in Yugoslavia. Results
of pulled tube examinations have provided the following information:
* The cracks are axially oriented except in association with denting. The
cracks do not extend beyond the edge of the TSP.
* In many cases, caustics have been identified as probable causative agents
based on water chemistry history, and because nickel enrichment and
chromium depletion, carbonate, and metal hydroxides have been
detected at crack faces. Often, lead has been identified as a possible
contributing factor; lead has been observed on OD deposits and crack
surfaces, and occasional, mixed-mode crack morphologies have been
reported. Finally, acid environments have also been implicated, based
on the detected presence of sulfur and chlorine, the close proximity of

pits to the cracking, and evidence of resins in the deposits.
* In the case of North Anna Unit 1, the cracks are associated with denting
and are believed to be the result of acidic conditions (12.50).
Seawater Units on AVT. In this group of plants, one Combustion

Engineering unit (St. Lucie Unit 1) and several units of Westinghouse design
have been significantly affected. St. Lucie Unit I is operating with a hot leg
temperature of 601'F (316'C), and has experienced IGA/SCC in both hot
and cold leg sludge piles and at hot and cold leg tube supports. Examination
of pulled tubes and the evaluation of water chemistry data indicate that
some mixed mode attack has occurred, and that IGA/SCC was initiated by
acidic conditions, but has continued to propagate under near neutral
conditions (12.6). Several other seawater units of Combustion Engineering
design have reported a few tubes with attack at supports, including
Millstone Unit 2, Maine Yankee, and Calvert Cliffs Unit 1.
Several seawater units of Westinghouse design have experienced IGA/
SCC at TSPs. This includes three Japanese units (Takahama Unit 2, Genkai
Unit 1, and Ohi Unit 1) which have experienced attack at numerous tubes
(20% or more of tubes affected in each unit), and three others that have
experienced less severe IGA/SCC attack (Takahama Unit 1, Mihama Unit 2,
and Mihama Unit 3). It also includes units in Sweden (Ringhals Units 3
and 4), France (Graveline Units 1, 3 and 4, Blayais 1), Belgium (Doel Units 3
and 4) and Brazil (Angra Unit 1). Results of investigations are summarized
below.
Following an extensive evaluation, the Japanese made the observations
noted below (12.1, 12.51):
* The IGA/SCC initiated as the result of a combination of caustic and
oxidizing conditions during the first cycles of operation. The caustic is
believed to have come mainly from welding residuals, and the oxidizing
conditions from air inleakage and transport of copper and iron corrosion
products into the steam generators. Caustic initiation was indicated
by chromium depletion seen at crack faces near the tube surface.
* Continued crack propagation appears to be occurring under fully
reduced, near neutral conditions, i.e., despite the elimination of the
caustic oxidizing conditions believed to have initiated the corrosion.
Propagation under near neutral conditions is indicated by the lack of
chromium depletion at crack faces near the crack tip, by analyses of
0
hideout return data, and by conditions measured in a crevice in a side
stream model boiler at Ohi Unit 1.
" Use of full flow condensate polishers beginning with initial plant startup
correlates with no attack. The polishers are considered to have assisted
in preventing caustics and oxidizing corrosion products from entering

the steam generators, and also assisted in maintaining molar ratio.
" The use of on-line boric acid additions, together with high hydrazine,
appears to have helped reduce the rate of IGA/SCC at TSPs.
Nevertheless, significant numbers of new defects are being detected at
each inspection at affected units.
• There has been little IGA/SCC in Japanese plants that have experienced
seawater inleakage. The IGA/SCC has been reported to be dominated
by construction, makeup water, and auxiliary boiler impurities. It is
also interesting to note that IGA/SCC at TSPs has been less at plants
with prior phosphate operation. The Japanese have attributed this to
the buffering ability of resolubilized residual phosphate compounds
(12.11).
In at least some cases (e.g., Doel 4), the environment involved in
initiation, as well as propagation, appears to be acidic to neutral, with
possible lead involvement, based on the simultaneous presence of IGSCC,
wastage and pitting, chromium enrichment in oxide layers, the composition
of deposits, and hideout return analyses (12.43).
Hot Leg Free Spans in Recirculating Steam Generators

Serious IGA/SCC was recently detected on hot leg free spans at Doel
Unit 4 (Westinghouse design) and Palo Verde Unit 2 (C-E design). Both are
high-temperature units.
Examination of removed tubes has shown the IGA/SCC of hot leg free
spans at Doel Unit 4 to consist mainly of multiple axial cracks, with cellular
cracks in scratches (12.43). The crack depths are greatest at lower elevations,
with maximum depths of about 30% through wall. The cracks occur in
irregular patches, and do not appear to be associated with deposits. X-ray
measurements of archive tubes indicate that residual hoop stresses range
up to 44 ksi (300 MPa). The affected steam generator is known to have
been contaminated with significant amounts of lead. Analysis of crevice
areas indicates that degradation in those areas was due to acidic
environments; however, the implication for the free spans is uncertain.
Recently, a leaker outage occurred at Palo Verde Unit 2. Preliminary
ECT examinations have detected several free span cracks, which apparently
initiated at the OD surface.$Removal and examination of tubes is planned.
U-Bend Support Region in Recirculating Steam Generators

IGA/SCC in this region has occurred in Combustion Engineering units,
in Babcock & Wilcox design RSGs in Canada (pressurized heavy water
reactor plants), and, recently, in a heat-treated Westinghouse-design unit.
A significant number of tubes in Combustion Engineering units have

been affected; in some cases, preventive plugging has been performed to
minimize risks of forced outages. In at least one instance, IGSCC in this
region has resulted in a tube rupture and a forced outage (Ft. Calhoun,
May 1984). In this case, the rupture occurred in a horizontal run of a tube
where it contacts a vertical support. The tube had previously been weakened
by denting. The problem appears to be the result of concentration of
impurities in a flow-starved area where the vertical supports are relatively
wide and restrict flow. Analysis of deposits indicated the presence of
sulfates, nitrates, calcium, copper, and silicon. No sodium or potassium
was found. Nevertheless, based on evaluation of the impurities entering
via condenser leaks, it was concluded that the problem was due to caustic
stress corrosion caused by condenser leakage.
The Bruce A units in Canada, which have RSGs of Babcock & Wilcox
design with mill annealed alloy 600 tubing, have experienced significant
amounts of SCC at scallop bar supports in the U-bend area. In the assembled
condition, the scallop bar supports have a configuration similar to a drilled
hole, and thus provide impurity concentration sites (the tube supports in
these RSGs use broached holes). Examination of removed tubes, and hideout
return analyses, indicate that the worst cracking was associated with high
levels of lead and occurred in a near neutral environment (12.52).
The row 1 and 2 U-bends at Tihange were stress relief heat treated
several years ago as a preventive measure against PWSCC at the inner
surface. Recently, the U-bends have started to experience SCC on the OD.
The cracks are located at the ends of the heat treated area, just above the
top tube support and below the U-bend to straight tube tangent point. The
specific mechanisms involved in the attack have not been determined.
Cold Legs in Recirculating Steam Generators

Limited amounts of IGA/SCC have been detected at cold leg free spans,
tube supports, sludge piles, and crevices.
Free Spans. In March 1989, a tube in McGuire Unit 1 developed an

approximately 4-inch- (10 cm) long rupture in a cold leg area (12.17). The
top end of the crack extended above the lowest preheater baffle plate, and
the lower end extended about 2-3/4 inches (7 cm) down into the free span
area below the baffle plate. Detailed examination of the tube indicated that
the rupture was the result of a series of IGSCC cracks linked to one another.
The series of cracks initiated on the OD surface along a long axial score or
line of abrasion of about 0.5 mil (13 jtm) depth. The tube material in the
score region was abnormally hard, with a peak residual tensile stress of
92 ksi (630 MPa) a few mils below the scored surface. No specific corrodents
were identified. Subsequently, another leaker outage occurred due to a

similar crack in McGuire Unit 1, and additional tubes with similar cracks
have been observed by ECT in both McGuire Unit I and Unit 2. Evaluation
of additional pulled tubes confirms that the cracks are associated with score
marks or lines of abrasion, which resulted in high local cold work and
residual stresses. No corrodents have been identified and it is speculated
that the cracking is a neutral solution or PWSCC type mechanism.
Supports. At one Combustion Engineering unit (St. Lucie Unit 1),

significant IGA/SCC has been detected at tube supports on the cold leg
side. The number of indications is greatest at the bottom support and
decreases as elevation increases. This behavior is unexpected, since
temperature increases as elevation increases; therefore, it would be
reasonable to expect the number of indications to increase, rather than
decrease, as elevation increases on the cold leg side. The reasons for this
unexpected pattern of corrosion have not been systematically explored. This
could possibly be explained by gradients in the electrochemical potential
or in the concentration of some species, such as an oxidant, along the height
of the tubes. Differences in pressure drop or sludge accumulation at different
elevations are other possible reasons.
At the Asco units in Spain, which are of Westinghouse design, IGA/
SCC has been detected at significant numbers of TSP intersections in the
cold leg, but to a lesser extent than in the hot leg. The indications are located
mainly at higher elevations, i.e., in the higher temperature areas of the cold
leg.
Top of Tubesheet. As discussed earlier, circumferential cracks have been

noted at the top of the tubesheet in cold legs of one Combustion Engineering
unit, Millstone Unit 2. The cracking in this unit has been about as severe
on the cold leg as on the hot leg. The reasons for the approximately equal
severity on the hot leg and cold leg, despite the large temperature difference,
have not been determined. It is speculated that this is the result of the
increased stability of aggressive species, such as acids and/or some copper
species, in the lower superheat on the cold leg side. It could also be due to
the lower pHT for similar concentrations at lower temperatures. Another
Combustion Engineering unit, St. Lucie Unit 1, has also experienced a
significant amount of attack in the cold leg top of tubesheet area; in this
case, however, the IGA/SCC is axially oriented in the sludge pile.
Tubesheet Crevices. Cold leg tubes have been removed from several
plants. In the case of Ginna and Doel Unit 2, no IGA/SCC was observed.
In the case of Ringhals Unit 2, the tubes were found to have shallow IGSCC
(12.16). The investigator of the Ringhals Unit 2 tubes concluded that the
IGSCC was occurring in the same manner as on the hot leg, but at a slower
rate as a result of the lower temperature. Based on the relative rates of the
attack on both the hot and the cold leg, the investigator concluded that the
apparent activation energy was about 40 kcal/mole (12.16). Evaluation of
deposits in the cold leg crevice showed them to be similar to those in the
hot leg, with solutes from the deposits ranging from neutral to mildly
alkaline. Specific corrodents were not identified.
Upper Tubesheet Crevice and Upper Tube Supports in Once-Through

Steam Generators
Several OTSG units have experienced IGA/SCC from the secondary
side in the upper tubesheet crevices. Examination of pulled tubes from
Arkansas Nuclear One, Unit 1, indicated that sulfide was present in the
crevices. It was concluded that the corrosion was due to reduced sulfur
species attack of the sensitized tubing.
It should be noted that the tubes in OTSGs were stress relieved as part
of the manufacturing process and consequently are in the sensitized or
chromium-depleted condition. In this state, alloy 600 tubing is especially
susceptible to intergranular corrosion by oxidizable sulfur compounds
(partially reduced sulfur species). This corrosion occurs more typically at
layup temperatures than at operating temperatures.
Lower Tubesheet Crevice and Sludge Pile in Once-Through Steam

Generators
A tube was removed from the lower tubesheet (cold leg end) of an
Oconee Unit I steam generator in 1981 (12.53). Minor IGP was detected on
this tube penetrating to less than 20% of the tube wall. The presence of
sulfur species was believed to be the cause of the IGA of the sensitized
tubing.
Summary of Failure Analysis Results

Examination of pulled tubes has shown that IGSCC, IGA, IGP, and
mixed-mode SCC have all occurred. It is not clear which of these will appear
at any given plant. However, recent experience with higher-temperature
plants, over 600'F (316'C), indicates that IGSCC in these units is the most
common form of significant attack, i.e., IGA or IGP rarely penetrate to
significant depth before IGSCC becomes the controlling mode of attack.
Some general observations are noted below.
0 Older, lower-temperature RSGs of the Westinghouse type that have
operated for long periods of time at alkaline-forming freshwater

locations have tended to experience IGA, and to develop IGP or IGSCC
only after the IGA has penetrated to significant depths. This corrosion
has mainly occurred in tubesheet crevices, with only a relatively small
number of eddy current indications at tube support locations, the nature
of which has generally not been determined by tube pull examinations.
Generally, the causative environment has been identified as caustic.
* Older, lower-temperature RSGs of the Combustion Engineering type
have experienced IGA, IGSCC, and mixed-mode SCC in sludge pile
regions and at tube supports. Most of this attack appears to be due to
acidic environments, though lead has been identified in several cases
and caustic is believed to have been important in one case.
* Higher-temperature RSGs with high integrity condensers at both
freshwater sites and seawater sites have generally experienced IGSCC,
with limited amounts of IGA or IGR The performance of these units
has not been significantly different. The IGA/SCC has mainly occurred
in sludge piles and at TSPs. (Most of these units do not have tubesheet
crevices. The original steam generators at Ringhals 2 were an exception,
in that they operated at high temperature and had tubesheet crevices.)
The causative environments have been identified as caustic, acid, and
lead. It is believed that the crevice environments in these plants with
high integrity condensers are dominated by impurities leaking past
demineralizers (makeup demineralizers, condensate polishers, and
blowdown demineralizers) or resulting from the demineralizers
themselves, rather than by cdoling water inleakage. The types of
impurities leaking past demineralizers can include neutral organics and
neutral clays, which can breakdown in steam generator crevices to form
acidic species and caustics, respectively.
* At seawater and freshwater sites where intergranular corrosion has been
observed in the first year or two of operation, IGSCC seems to dominate.
This tendency may be a function of the relative corrosion rates of IGSCC
versus IGA.
* The IGSCC and mixed mode SCC observed in straight tube areas has
generally been axial in orientation. The axial orientation is probably
due to the applied plus residual stresses being greater in the hoop
direction than the axial direction. The OD IGSCC observed at expansion
transitions has generally been circumferential in orientation. This
orientation is probably the result of axial residual stresses. In cases
where denting has occurred, OD IGSCC can be circumferential as well
as axial in orientation.
* Microchemical examinations have provided indirect evidence for the
corrosive environments. However, additional work is required to clarify
the situation. Based on work to date, a variety of environments seem

to have been involved:
-Caustic, acid, and lead are frequently implicated.
-Elevated ECP appears involved in initiation in some severe cases.
-Crack propagation may continue even in reduced, near neutral,
environments, despite the use of boric acid.
Based on metallographic and eddy current examinations, reference
morphologies have been established for laboratory studies. The amenability
of intergranular corrosion to detection by eddy current or other NDE
methods is strongly dependent upon the corrodent and the resulting width
or conductivity of the corroded grain boundary. IGA of sensitized tubing
produced by sulfur compounds (e.g., Palisades, Oconee) appears to be the
easiest to reproduce in the laboratory. It is readily detectable by eddy current.
IGA and IGSCC produced in alkaline environments are much more difficult
to prepare and to detect, apparently due to slow growth rates of IGA, erratic
initiation processes, and the tight nature of the attack. Extreme variability
in results has been the norm.
The Ringhals Unit 2 experience suggests that small stress or strain
gradients may cause the IGSCC to align in an axial direction. Similar
examinations of tubing with primary-side-initiated IGSCC (e.g., Doel Unit 2
roll transitions, Trojan U-bends) suggest that the surface stress state and
the hoop stress imposed by primary side pressurization can often cause
IGSCC to assume a distinct axial directionality.
Microchemical examinations have provided indirect evidence for the
corrosive environments causing IGA/SCC but, as discussed above, much
additional work is required to clarify the situation.
During the course of the Steam Generator Owners Groups and the
subsequent Steam Generator Reliability Project, considerable effort was
spent examining field data generated by tube failure analyses, deposit
analyses, analysis of crevice flush water, and evaluation of operating site
chemistry history. Based on field experience, a great deal of information is
now available on intergranular corrosion; yet, there is limited knowledge
about how to specifically identify the conditions under which the corrosion
will most likely proceed in steam generators. It appears that a wide variety
of environmental conditions can lead to intergranular corrosion of alloy 600,
and, based on laboratory experience, of alloy 800 and alloy 690. It appears
that the various manifestations of intergranular corrosion, i.e., IGSCC, IGA,
IGP, and mixed-mode attack, are varying responses of the tube alloy to a
wide range of environmental conditions: the principal variables being
temperature, pH, chemical composition of the environment, concentration
of chemicals, electrochemical potential, and stress levels or strain state of
the tubing.
EFFECTS OF INTERGRANULAR CORROSION ON

TUBING MATERIALS
Examination of tube segments removed from operating plants has raised
questions relative to the structural integrity of the tubing material below
the bands of relatively uniformly degraded material with intergranular
corrosion. In order to answer these concerns and to determine the integrity
of the remaining tube wall, tensile tests and hydraulic biaxial burst tests
have been conducted on samples removed from several operating steam
generators. Data from Point Beach Unit I are shown in Figure 12-20 and
Table 12-3 (12.3). The data show that the remaining tube wall is sound and
suffers from no incipient degradation in the grain boundaries. These data,
combined with the location of intergranular corrosion in the tubesheet
crevice, serve to provide considerable margin for alleviating concerns
relative to tube burst or rapid leaking from intergranular corrosion in
tubesheet crevices. On the other hand, the type of degradation experienced
at Ringhals Unit 2, i.e., long, axially oriented cracks, has been shown to
produce high volume leaks upon through-wall penetration. However, these
axially oriented cracks have also occurred on confined tubes within the
crevice, and have not led to safety concerns with regard to tube burst.
Intergranular corrosion of the IGA, IGP,IGSCC, or mixed-mode cracking
in tube support plate crevices is, like these forms of corrosion in tubesheet
crevices, not of significant safety concern from a tube burst standpoint. This
is because the tube support plate provides structural reinforcement which,
according to laboratory testing, prevents the defects from reducing the burst
pressure of the tube. However, the corrosion can lead to increased leakage,
which must be considered from both operational and safety viewpoints.
IGA/SCC in sludge pile areas is of more safety concern than when it
occurs in tubesheet or tube support plate crevices. Experience at one plant
(Almaraz Unit 1, July 1988) showed that this form of attack can lead to
rapid, large leaks as a result of the linkup of axially aligned multiple smaller
cracks. However, appropriate periodic inspections can reduce the risk of
rupture caused by IGA/SCC in sludge pile areas considerably.
IGA/SCC at tube supports in Combustion Engineering units is of
greater safety concern than similar attack in drilled-hole tube support plate
crevices. This is because the eggcrate-type supports in these units do not
entirely capture the tube, and large ruptures are possible if the corrosion
proceeds deep enough. However, appropriate periodic inspections can keep
the risk of rupture caused by IGA/SCC at eggcrate supports to low levels.
FIGURE 12-20
Tensile Stress Strain Curves for Point Beach Unit 1 Steam Generator
Tubes
100
.f Tube AR22-C37
80/
._60
ILl Tube AR15-C45
20--
0 10 20 30 40 50 60
Engineering Strain (%)
Note: Tube AR22-C37 had been stretched by about 10% during removal.
TABLE 12.3
Hydraulic Burst Pressure for Steam Generator Tubing From
Point Beach Unit 1
Max. Burst Expect'd Burst Press.

Depth Pressure (Effect. Wall Thickn.
Sample Location (%) With No IGA)
Top of tubesheet negligible >11,799(1) 11,000 ± 1,000
10-15 inch. below 40-50 6,800 6,000 - 7,200(2)
tubesheet top
Typical virgin tube 0 11,000+ 1,000 11,000 ± 1,000
1
2
Thespecimen did not burst at the maximum pressure of the burst apparatus.
Because first sample did not burst at 11,700 psi, a burst pressure of 12,000 psi
was used to calculate expected burst pressures for other samples.
REMEDIAL MEASURES
Laboratory studies, as well as field experience, suggest several areas
for remedial measures. Figure 12-21 shows the interrelationship of various
factors that may affect IGA/SCC. A remedial measure would be expected
to interrupt or change one or more of the factors discussed below.
FIGURE 12-21
Factors Possibly Affecting IGSCC and IGA
Constant load
Constant deflection
Constant plastic deformation
Cyclic load
Residual stress
Composition Composition
Temperature
Grain size
Time
Processing history
Additives
Thermal treatment
Potential
Stress
It has been shown with reasonable certainty that increasing the service
stress, imposing a dynamic strain, or the presence of a high residual stress
level can all be major factors in accelerating initiation or propagation of
IGSCC. With regard to IGA, laboratory results indicate that increasing levels
of stress accelerate IGA a moderate amount, but not nearly as much as
stress influences IGSCC.
Because of the high stress dependency of IGSCC, stress relief of the
tubing is a possible remedial measure when residual stress is the main
causative factor. Stress relief is accomplished for new steam generators by
using thermally treated alloy 600 tubing which has been thermally aged at
1300'F (705'C) for 16 hours, and by stress relief of tight radius U-bends.
Also, Babcock & Wilcox uses a full bundle stress relief following once-
through steam generator fabrication to provide stress relief. For operational
steam generators with mill annealed tubing, performing stress relief is a
much more difficult operation. However, two approaches have been
proposed to treat other problems that might be of some benefit against
secondary side IGSCC. One approach would be to perform a global heat
treatment of the whole steam generator (12.54); this approach has been
suggested to treat roll transitions to reduce PWSCC, but it might also be
effective for the entire steam generator. Another approach would be to
perform local stress relief of tube support plate intersections using heaters
inserted into the tubes; this approach has been used on a trial basis to treat
dented TSP intersections. The cost effectiveness and practicality of these
measures as ways to prevent IGA/SCC, especially at tube supports, have
not been determined.
Microstructure
It is well known that microstructure has a major impact on susceptibility
to IGSCC in primary water environments (Chapter 7). However, the
relationship between microstructure and secondary side IGA/SCC
resistance is less certain. Chapters 13, 14, and 15 discuss the effects of
microstructure and heat treatment on IGA/SCC of alloy 600 in secondary
side environments, as determined from laboratory investigations. As
discussed in these chapters, thermal treatment of the tubing is effective in
reducing susceptibility to caustic IGSCC, PWSCC, and acidic IGSCC.
However, much of this benefit appears to be due to stress relief and
annealing of cold-worked structures, and not due to the carbide
microstructure. In addition, preliminary work indicates that at least some
microstructures (high mill anneal), that have improved PWSCC resistance,
appear to have no reduction in susceptibility to acidic environments.
Based on field experience, there are some indications that microstructure

influences IGA/SCC to a limited extent.
" A detailed evaluation of 15 tubes pulled from Ringhals Unit 2 showed
that IGSCC on the tube OD in the tubesheet crevice area was deeper in
tubes with intragranular carbides than in tubes with grain boundary
carbides (0.40 mm versus 0.12 mm).
" The Japanese report (12.11) on the examination of pulled tubes from
plants that are severely affected by IGA/SCC indicates that tubes
without attack show a great amount of grain boundary carbides and
high ferrite scope readings (Ferrite scopes indicate the amount of
magnetic material present). For tubes in plants experiencing significant
IGA/SCC, high ferrite scope readings of tubes correlate with low attack,
and low ferrite scope readings correlate with high attack, where the
severity of attack was determined by ECT test. However, there are
tubes with low ferrite scope readings in plants with no IGA/SCC. The
Japanese conclude that, while there is a correlation between material
factors and IGA/SCC, environmental factors dominate the situation.
* Significant numbers of tubes at St. Laurent B1, steam generator 2, have
experienced IGA/SCC at TSPs. The affected tubes have relatively low
mechanical strength and carbon levels, though they meet specification
requirements (12.55). There has also been significant PWSCC at roll
transitions. Tubes with PWSCC are less likely to have experienced IGA/
SCC than the general population. The opposite is also true, i.e., the
tubes that have experienced IGA/SCC are less likely to have
experienced PWSCC than the general population. This inverse
correlation is assumed to be due to a microstructural effect; however,
the details are not yet known.
Changing the microstructure, together with stress relief, to enhance
corrosion resistance has been applied to new steam generators since the
late 1970s, i.e., since the introduction of thermal treatment. Babcock & Wilcox
used to stress relieve the entire steam generator; however, this resulted in
sensitization of the tubes, and thus it has some drawbacks. Changing the
microstructure in operating steam generators would be a difficult operation.
Conceptually, it could be achieved using a global heat treatment approach
of the type suggested in Reference 12.54. However, it would involve
significant uncertainties and risks, and has therefore not been attempted.
Environment
Modification of the crevice environment probably offers the greatest
promise for a reduced rate of IGA/SCC. Modification of the crevice
environment could include several factors such as elimination of the crevice
or minimizing its size, lowering temperature, adding a pH neutralizer,

modifying the alloy protective film, removing the corrodent by flushing or
soaking, removing crevice-forming deposits or sludge, adjusting the
electrochemical potential, changing the concentration and/or ratio of bulk
water contaminants, etc. Laboratory studies have confirmed the benefit of
several of these approaches, and some of them have been applied to
operating steam generators.
Minimization of Tubesheet Crevices. The Japanese minimized the depth

of the tubesheet crevices in several units with part-depth rolled tubes by
expanding the tubes into the tubesheet several years after operation (12.1).
This expansion was accomplished in phases over several refueling outages.
For the inner two thirds of the tube bundle, the expansion was performed
using a combined hydraulic expansion plus roll expansion, which left about
a 4-inch- (100 mm) long crevice at the top. For the outer one third of the
tube bundle, the expansion was performed using hard rolling for the bottom
one fourth of the tubesheet, and then using hydraulic expansion, mechanical
rolling, and shot peening, which left about a 0.2 inch (5 mm) crevice at the
top of the tubesheet. The crevice closure has successfully prevented 'the
growth of IGA/SCC on the OD in the expanded areas. However, some
significant IGA/SCC has been detected in crevices left at the top of the
newly expanded tubes.
Lower Temperature. The hot leg operating temperature of Point Beach

Unit I was kept low for several years which effectively reduced the rate of
IGA progression. The hot leg temperature was lowered from 597°F (314'C)
to about 557°F (292 0C), which is close to the normal cold leg temperature.
Similar temperature reductions have been used at Ringhals Unit 2 and Cook
Unit 2. Significant temperature reductions often result in reduced power
production. The effectiveness of temperature reductions can be estimated
using the Arrhenius equation with an apparent activation energy of about
50 kcal/mole (about a factor of two decrease in corrosion rate for each 18'F
[10 °C] temperature reduction).
Adding pH Neutralizers. Based on laboratory data, use of on-line boric

acid is expected to significantly reduce the occurrence and rate of growth
of caustic-induced IGSCC. Some plant experience appears to support the
idea that use of boric acid reduces IGA/SCC. However, other plant data
suggests that, in some cases, there is little benefit. In addition, tests and
analyses indicate that boric acid will not be effective at inhibiting
acid-induced or lead-assisted IGA/SCC.
Flushing or Soaking. These off-line maintenance procedures have been

used at several plants. Optimization of crevice flushing procedures for
each unit and periodic repetitive application of the optimized procedures
are recommended for plants with deep tubesheet crevices. It should be
recognized that flushing tubesheet crevices or adding a pH neutralizer to
the crevices may be difficult if denting has occurred at the top of the
tubesheet, thus blocking access to the crevice. Hot flushes or soaks may
also be useful at plants with sludge piles or tube support plate crevices;
however, the cost effectiveness of the procedures for these areas is not as
well known as for deep tubesheet crevices, and the procedures have not
been widely used.
Inleakage and Particulate Control. The concentration of contaminants

accessible to crevices is primarily controlled by eliminating or reducing
ingress of contaminants to the steam generator. With regard to tubesheet
crevices, it can also be reduced by controlling the concentrating capability
of the sludge pile above the crevice. Reducing ingress of impurities is best
accomplished by preventing condenser leaks, routing drains to the
condenser, properly treating makeup water sources, careful attention to
polisher and demineralizer regeneration, etc. Use of full-flow condensate
polishers for control of ionic species has not been shown to be as effective
in controlling those species that probably are responsible for IGA and
IGSCC, as was shown for control of the chlorides responsible for denting.
Polishers themselves are potential sources of SO4= and Na+ and do not
effectively remove silica, neutral organics, and neutral clays. However, the
successful performance of Japanese units that have used full-flow polishers
from initial startup indicates that use of polishers from the beginning of life
can be beneficial, if they are used under rigorous control such that they do
not contribute impurities .themselves.
Control of the Concentrating Capability of the Sludge Pile. This requires

effective, periodic sludge lancing, minimization of particulate transport by
preventing air ingress and/or providing for condensate filtration such as
by powdered resin condensate polishers, and elimination of chemical species
which tend to promote agglomeration. In this latter category, phosphates,
silicates, and copper oxides are believed to be active. Chemical cleaning
could also be employed to remove the sludge on the tubesheet. Use of
morpholine to reduce secondary side corrosion and transport of corrosion
products to the steam generator is also effective. The above items are
discussed in greater detail in later chapters.
Ratio Control. Japanese utilities have utilized ratio control of two major
inleakage contaminants to minimize corrosion. They attempt to maintain
a Na+/C1- ratio <0.7 so that C1-, an acid forming specie, is always present
in excess of Na+, an alkali forming specie. This practice may be a factor in
the lack of detected IGA/SCC at several Japanese plants after 15 or more
years of operation. The success of these units indicates that molar ratio
control is worth pursuing as a remedial approach. However, the continued
development of IGA/SCC at other severely affected Japanese units, despite
use of molar ratio control, indicates that molar ratio control cannot be relied
upon to stop the growth of IGA/SCC once it has initiated.
Avoidance of Oxidizing Conditions. Oxidizing conditions can lead to

rapid attack under either caustic or acidic conditions. Since it is difficult to
avoid caustic and acidic conditions, it is important to minimize the
occurrence of oxidizing conditions. This can best be achieved by minimizing
the ingress of dissolved oxygen and other oxidizers such as copper oxide
and hematite. Minimizing the ingress of these species involves practices
such as the use of condensate polishers to filter out oxides, clean up of
secondary systems following layup periods, maintaining tightness of the
secondary system against air inleakage, replacement of copper alloy heat
exchanger tubes with ferrous materials, and use of high levels of hydrazine
in the feedwater.
Protection of Sensitized Material. Avoidance of oxidizing conditions in

steam generators with sensitized tubing is of increased importance because
of the high vulnerability of sensitized tubing to IGA/SCC under oxidizing
acidic conditions, especially at low temperatures. Normal practices to
protect such tubing include minimizing the input of sulfur species, and the
use of nitrogen blankets for many maintenance operations.
Inhibitors. Inhibitors, such as titanium dioxide, that could be added to the

feedwater for the purpose of reducing the initiation or progression of IGA/
SCC are being investigated (12.56). Testing of inhibitors is at an early stage,
and whether useful ones will be found is not yet known.
Recommendations. It is recommended that utility operators make

strenuous efforts to reduce transport of ionic and particulate impurities
into steam generators. Addition of boric acid to the feedwater is also
recommended for plants with known caustic forming feedwater.
Minimizing oxidizing conditions by use of high levels of hydrazine and
minimizing the transport of oxidizing corrosion products is also
recommended. For plants with deep tubesheet crevices, use of optimized

flushing and soaking procedures is recommended. Other remedial measures
require further evaluation before an unqualified recommendation of the
effectiveness can be given.
REFERENCES
12.1 Takamatsu, H. "Japanese Steam Generator Operating Experiences
and Results of Related R & D." Paper presented at EPRI SGRP
Technical Advisory Group Meeting, Philadelphia, June 26-28,1990.
Corrosion Resistance of Inconel Alloy 600." EPRI NP-3051. July
1983.
12.3 Pement, F. W., et al. "Examination of Three Steam Generator Tubes
From the Point Beach Unit I Nuclear Power Plant." EPRI NP-2958-
LD. March 1983, pp. 3-8, 4-21, 4-95.
12.4 Pement, F. W., and P. J. Kuchirka. "Examination of Steam Generator
Tube A (18-37) from the Point Beach Unit 2 Nuclear Power Plant."
EPRI NP-2539-LD. August 1982.
12.5 Paine, J. P. N. "Operating Experience and Intergranular Corrosion
of Inconel Alloy 600 Steam Generator Tubing." NACE Corrosion/
82 no. 204 (March 1982).
12.6 Frye, C. R. "Laboratory Evaluation of Steam Generator Tubes 120/
12, 79/9, and 59/95 From St. Lucie Unit 1." EPRI NP-5397-LD.
August 1987,
12.7 Cullen, W. H., et al. "IGA/IGSCC of Alloy 600 in Acid Sulfate
Solutions." In Proceedings of International Symposium,
Fontevraud II, Contribution of Materials Investigation to the
Resolution of Problems Encountered in PWR Plants SFEN
(September 1990): 332.
12.8 Miglin, B. P, and J. M. Sarver. "Investigation of Lead As a Cause
of IGA at Support Plate Intersections." EPRI NP-7367-S. June 1991.
12.9 Nagano, H., et al.. "Effects of Environmental and Material Factors
on the Intergranular Attack of Alloy 600." In Proceedings: 1985
EPRI Workshop on Remedial Actions for Secondary-Side
Intergranular Corrosion. EPRI NP-4929. December 1986.
12.10 Takamatsu, H. "Status of IGA in Japanese Plants and Results of S/
G Pulled Tube Examinations From Japanese Plants." In
Proceedings: 1985 EPRI Workshop on Remedial Actions for
Secondary-Side Intergranular Corrosion. EPRI NP-4929. December
1986.
12.11 Kishida, A., et al. "The Causes and Remedial Measures of Steam
Generator Tube Intergranular Attack in Japanese PWR." In
Environmental Degradation of Materials in Nuclear Power Systems
- Water Reactors, edited by G. Theus and J. R. Weeks. The
Metallurgical Society (1988): 465-471,
12.12 Ogren, R. A., and J. A. Gorman. "Correlation of Temperature With
Steam Generator Tube Corrosion Experience." Dominion
Engineering, Inc., DEI-290. April 1990.
12.13 Gorman, J. A. "Contribution to Alloy 600 Experts Meeting." Paper
presented at EPRI Alloy 600 Experts Meeting, April 6-9,1993, Airlie,
Virginia. To be published by EPRI.
12.14 Jacko, R. J. "Corrosion Evaluation of Thermally Treated Alloy 600
Tubing in Primary and Faulted Secondary Water Environments."
EPRI NP-6721-SD. June 1990.
12.15 Newman, J. F "Stress Corrosion of Alloys 600 and 690 in Acidic
Sulfate Solutions at Elevated Temperatures." EPRI NP-3043.
October 1983.
12.16 Norring, K. "Examination of Five Tubes Pulled From the Cold Leg
of Steam Generator I at Ringhals 2." Studsvik/EI-84/157. October
24, 1984.
12.17 Frye, C. R., et al. "Laboratory Examination of Pulled RSG Tube
Sections From McGuire Nuclear Station." Babcock & Wilcox,
RDD:90:5459-01:01. August 1989.
12.18 Gorman, J. A., et al. "Statistical Analysis of Steam Generator
Degradation." EPRI NP-7493. September 1991.
12.19 Serra, E. "Stress Corrosion Cracking of Alloy 600,." EPRI NP-2114-
SR. November 1981.
12.20 Gorman, J.A. "Correlation of Plant IGA/SCC at TSPs With Copper,
Boric Acid, and Prior Use of Phosphates." Paper presented at 1992
EPRI IGA/SCC Workshop, San Antonio, TX, Dec. 8-10, 1992. To be
published by EPRI.
12.21 Michaels, W. J., et al. "Intergranular Corrosion of Steam Generator
Tubes-Field Study" EPRI NP-4457. February 1986.
12.22 Sawochka, S. G., and S. S. Choi. "MULTEQ-IGA/SCC
Correlations." In Proceedings: 1991 EPRI Workshop on Secondary-
Side Intergranular Corrosion Mechanisms. EPRI TR-101103.
August 1992.
12.23 Hobart, S. A., and M. J. Partridge. "Status Report: Correlation of
Hideout Return Data With IGA/SCC." Paper presented at 1992
EPRI IGA/SCC Workshop, San Antonio, TX, Dec. 8-10, 1992. To be
published.
12.24 "Interim PWR Secondary Water Chemistry Recommendation for
IGA/SCC Control." EPRI TR-101230. September 1992.
12.25 Rootham, M. W. "Cation/Anion Ratio Control." Paper presented

at 1992 EPRI Plant Chemists' Meeting, San Diego, CA, Nov. 18-20,
1992.
12.26 Agrawal; A. K., et al. "Point Beach-1 Steam Generator: Tubesheet
Crevice Chemistry." EPRI NP-5660-LD. February 1988.
12.27 Dow, B. L., Jr. Steam Generator Progress Report, Rev. 6. EPRI.
September 1990.
12.28 Vaccaro, F. P. "Lower Temperature/Longer Time Aging of Alloy
600: Possible Effects of Initiation of Caustic IGA/SCC." In 1987
EPRI Workshop on Secondary-Side Intergranular Corrosion
Mechanisms: Proceedings. EPRI NP-5971. September 1988.
12.29 Pearl, W. L. "Utilization of Phosphate Chemistry." EPRI NP-7347.
June 1991.
12.30 Hermer, R. E. "Boric Acid Application Guidelines for Intergranular
Corrosion Inhibition," Rev. 1. EPRI NP-5558-SL. August 1990.
12.31 Partridge, M. J., et al. "Correlation of Secondary-Side IGA/SCC
Degradation of Recirculating Steam Generator Tubing With the On-
Line Addition of Boric Acid." EPRI TR-101010. August 1992.
12.32 Riddle, J. M., et al. "Survey of Domestic and Foreign PWR
Experience With Morpholine in Chemistry Control by All-Volatile
Treatment." EPRI NP-4671. July 1986.
12.33 Kuchirka, P. J., and A. Madeyski. "Examination of Tubes R4C19HL,
R6C18HL, and R16C33HL From Steam Generator 12 of the Prairie
Island Nuclear Station Unit 1." EPRI NP-4745-LD. August 1986.
12.34 Frye, C. R. "Examination of Roller Expansion Transitions in Point
Beach Unit 1 Steam Generator Tubing." EPRI NP-6192-SD. July
1989.
12.35. "Kewaunee Steam Generator Mid-Cycle Report." Westinghouse
Electric Corporation, WCAP-12791. December 1990.
12.36 Pement, F. W., and P. J. Kuchirka. "Examination of Three Steam
Generator Tubes From the Ginna Nuclear Power Plant." EPRI
NP-2534-LD. August 1982.
12.37 Norring, K.. "Examination of 15 Tubes From Steam Generator 1 at
Ringhals 2." Studsvik/EI-84/56. March 26, 1984.
12.38 Attachment to Utility Experience Report, Takahama 2, June 30,1989,
by Kansai Electric Power. Paper presented at EPRI SGRP Technical
Advisory Group Meeting, Montreal, CN, July 1989.
12.39 Anderson, F C. "Millstone 2 Steam Generator Experience," Vol. 2.
EPRI NP-5971. September 1988.
12.40 Eubanks, C. "Pulled Tube Analysis." Paper presented at 1992 EPRI
IGA/SCC Workshop, San Antonio, TX, Dec. 8-10, 1992. To be
published.
12.41 Toribio, E., et al. "Examination of SG Tubes Removed From ASCO

Units 1 and 2." EPRI NP-7198-S. March 1991.
12.42 Lancha Hernandez, A. M., and M. D. G6mez Bricefio. "Examen
Del Tubo R8C57 Del Generador Vapor A De La Unidad II De La C.
N. De Asco." Ciemat ITN/ME-30/DP-89. December 1989.
12.43 Laire, C., G. Platbrood, and J. Stubbe. "Secondary Side IGA/SCC
in Belgian Plants: Tube Examination and Possible Mechanisms."
Paper presented at 1992 EPRI IGA/SCC Workshop, San Antonio,
TX, Dec. 8-10, 1992. To be published.
12.44 Craig, K. R, "St. Lucie Unit I Steam Generator Condition." Paper
presented at EPRI Contractor Meeting, Alexandria, VA, December
15, 1988.
12.45 "NRC Presentation Report on Steam Generator Tube Integrity for
DC Cook Unit 2." Westinghouse Electric Corp., Power Systems,
Pittsburgh, PA, WCAP-11330. September 1986.
12.46 Carr, W. "Farley Nuclear Plant-Unit 2 Tube Support Plate
Cracking." In 1987 EPRI Workshop on Secondary-Side
Intergranular Corrosion Mechanisms: Proceedings., Vol. 2. EPRI
NP-5971. September 1988.
12.47 "Updated Tube Examination Results for Farley Unit 2 Steam
Generator Tubes R16-C50HL and R16-C53HL." Westinghouse
Electric Corp. Energy Systems, Pittsburgh, PA, SG-89-12-024.
12.48 "Summary Letter Report Regarding the Examination of Farley
Unit 1 Steam Generator Tubes R21-C48 Cold Leg and R20-C26 Hot
Leg." Westinghouse Electric Corp., Energy Systems, Pittsburgh,
PA, SG-90-02-002.
12.49 Wood, R. "Hideout Return Data Evaluation/Corrosion
Correlation." Paper presented at 1992 EPRI PWR Plant Chemists'
Meeting, San Diego, Nov. 18-20, 1992.
12.50 Albertin, L., and P. Kuchirka. "Examination of Tubes R3C41HL
and R9C58HL of Steam Generator C, North Anna Unit 1." EPRI
NP-5420-LD. October 1987.
12.51 Takamatsu, H., et al. "IGA/IGSCC Propagation Behaviors of Alloy
600." Paper presented at the Fourth International Symposium on
Environmental Cracking of Materials in Nuclear Power Systems-
Water Reactors, Jekyll Island, Georgia, August 6-10, 1989. NACE,
1990.
12.52 King, P. J., and F. Gonzales. "Cracking at Ontario Hydro's Bruce
NGS 'A'." Paper presented at 1992 EPRI IGA/SCC Workshop, San
Antonio, TX, Dec. 8-10, 1992. To be published
12.53 Pardue, E. B. S., et al. "Evaluation of the Lower Tubesheet Region
of an Oconee 1 Steam Generator Tube." EPRI NP-3026-LD. July
1983.
12.54 Frederick, G., et al. "Qualification of Remedial Methods to Prevent

Primary-Side Stress Corrosion Cracking of Steam Generator Tubing,
Volume 3: Global Heat Treatment." EPRI NP-5249. June 1987.
12.55 Scott, P. M. "IGA/SCC Plant Data, Analysis, and Predictive Tools."
Paper presented at 1992 EPRI IGA/SCC Workshop, San Antonio,
TX, Dec. 8-10, 1992. To be published.
12.56 "Section III: Research Reports-Inhibitors and Remedies." Papers
presented at 1992 EPRI IGA/SCC Workshop, San Antonio, TX, Dec.
8-10, 1992. To be published.
13
INTERGRANULAR CORROSION
OF ALLOY 600 FROM
CAUSTIC COMPOUNDS
A. P. L. Turner,Dominion Engineering,Inc.
J. 'P N. Paine,EPRI
INTRODUCTION
Intergranular corrosion in the form of intergranular attack (IGA),
intergranular penetration (IGP), and intergranular stress corrosion cracking
(IGSCC) has been experienced on the secondary side of PWR steam
generator tubes in tubesheet crevices, sludge piles, and tube support
crevices. A typical example of IGA is shown in Figure 13-1. The
distinguishing characteristic of IGA is general attack of nearly all grain
boundaries to a uniform depth over a substantial area of the tube surface.
A variation of IGA in which there is localized deep attack of the grain
FIGURE 13-1
IGA of Alloy 600 Steam Generator Tubing (Etched Sample)
boundaries is called IGP. In some cases, IGP occurs as localized fingers of

grain boundary attack extending below a layer of general IGA. In other
cases, the IGP takes the form of localized regions of deep attack on an
otherwise undamaged surface (Figure 13-2). It is postulated that IGP
represents an early stage of IGSCC or a transition stage between IGA and
IGSCC. However, degradation identified as IGP is not limited to a single
grain boundary path except at its deepest point. A typical example of IGSCC
is shown in Figure 13-3. IGSCC consists of discrete deep cracks with
relatively minor amounts of branching. It has been observed that the various
Intergranular Corrosion of Alloy 600 From Caustic Compounds 13-3
forms of intergranular corrosion can occur individually or in combination.

Frequently, IGSCC extends into material below surfaces exhibiting
generalized IGA. In other cases IGSCC occurs with no general IGA.
Sometimes, it is difficult to identify observed intergranular corrosion, as
one of the three types defined here, and classification can become somewhat
arbitrary.
Analysis of sample tubes removed from operating steam generators
(Chapter 12) and calculations based on blowdown chemistry have indicated
that the crevice environment, where intergranular corrosion has occurred,
is frequently alkaline. Boiling point elevation calculations show that NaOH
and KOH can become highly concentrated liquids in heated crevices.
Therefore, caustic attack was quickly identified as being one possible cause
of alloy 600 steam generator tube degradation. As a result, a substantial
number of laboratory studies have been undertaken in caustic environments
(e.g., 0.1% to 50% NaOH) to attempt to reproduce the IGA, IGSCC, and IGP
processes that are observed in operating steam generators.
FIGURE 13-2
Nature of OD Attack on Alloy 600 Capsules Exposed for
One Year at 5991F (315°C)
FIGURE 13-3
Intergranular stress corrosion cracking consists of long, narrow
cracks showing limited branching.
Laboratory studies have included electrochemical tests of polarization

and passivation, exposure of stressed samples in a variety of environments
to determine the effects of environmental and material variables on rates of
attack, and model boiler tests to simulate the process of concentrating caustic
species in hideout locations. Various studies have investigated the effects
of caustic concentration, temperature, stress, material variables,
electrochemical potential, and effects of additions of other chemical species
on the intergranular corrosion process. In most of these tests, accelerators
(i.e., high temperatures, stress, and concentration of the corrodent) have
been applied to promote fast initiation and propagation of defects. A
number of these studies have reproduced IGA and IGSCC degradation that
exhibits characteristics that are similar to the damage found on pulled steam
generator tubes. Model boiler tests have produced IGA/IGSCC degradation
when small amounts of caustic have been added to the feedwater. It is
clear that a large number of variables combine to form IGA/IGSCC on
alloy 600 steam generator tubes. Therefore, it is difficult to determine from
the laboratory results the importance of the various factors involved and
how well the laboratory-accelerated conditions correlate with conditions
in actual operating steam generators.
Samples from laboratory tests under known conditions provide
information regarding relationships between corrosion conditions and
degradation morphology. Such studies have helped to establish correlations
between the characteristics of surface deposits and crack surface chemistry,
and the corrosive environment that was responsible for the attack. Some
distinguishing characteristics of caustic attack have been established by such
studies (i.e., depletion of chromium from surfaces exposed to caustic
conditions). This information can be used to assist in the analysis of pulled
tubes to identify the probable causative agents of degradation'experienced
in service..
Other studies have been directed to evaluate remedial measures that
would prevent or slow degradation in steam generators where concentration
of caustic compounds is believed to occur. These studies have included
investigation of additives such as boric acid and studies of how the additive
can best be delivered to the target area.
INVESTIGATION OF CAUSTIC CONDITIONS THAT

CAUSE INTERGRANULAR CORROSION
A large number of studies have been performed that show that caustic
solutions can cause IGA and IGSCC at temperatures similar to those in
PWR steam generators. IGA and IGSCC have been produced in alloy 600
samples exposed to pure hydroxides (NaOH and KOH) and hydroxides
containing various additives. Several studies have investigated the variables

of chemistry, material condition and microstructure, temperature, and
electrochemical potential. Considerable information on the behavior of
alloy 600 in caustic environments is available, and, in many cases, reliable
trends have been established. Because of the number of variables involved,
significant uncertainties persist.
Behavior in Pure Caustic

In a study published in 1990 (13.1) the intergranular corrosion behavior
of alloy 600 was investigated by exposing stressed C-ring samples to NaOH
solutions for periods up to 7,000 hours at open circuit potentials. C-ring
samples from five different heats of alloy 600 were exposed to 1, 10, or 50%
NaOH solutions at temperatures ranging from 550 to 650'F (288 to 343'C).
The tests included samples from all five heats in both the mill annealed
(MA) and thermally treated (TT) conditions. The effects of applied stress
and prior plastic deformation were investigated by the use of three different
deflections for the C-rings. Previous studies (13.2 13.3, 13.4, 13.5) done in
pure NaOH, pure KOH, or mixed NaOH-KOH environments at comparable
temperatures on material in the same heat treatment conditions, provide
additional data on behavior in pure caustic environments.
1% Caustic. Jacko (13.1) studied intergranular corrosion in low

concentration NaOH (1%) by exposing MA and TT material for up to 7,000
hours at 600'F (316'C). Most samples experienced shallow IGA
approximately one grain deep. This was attributed to cold work and high
residual stresses in a ground surface layer. Uniform IGA in MA samples
was typically 10 to 30 gim deep. Significant IGSCC (defined by the author
as being greater than 110 gtm deep) was observed in some of the MA samples
that were prestrained more than 0.2%. The maximum depth of IGSCC
observed was 540 gim in a sample prestrained 3.5% and exposed for 7,000
hours. Cracking greater than 110 jtm deep was found in several MA samples
in as little as 2,000 hours. The maximum apparent IGSCC crack growth
rates (crack depth divided by exposure time) were 0.077 jim/h and
0.055 jim/h for 7,000 hr and 2,000 hr exposures, respectively. For
comparison, a previous investigation (13.5) reported an average crack
growth rate of 0.20 jim/h in 1% NaOH at 315'C (599°F) with the
electrochemical potential controlled to +150 mV above the free corrosion
potential.
Degradation of MA alloy 600 in 1% NaOH is. more rapid when the
temperature is raised to 650'F (343°C) (13.1. At this temperature, the
dominant mode of attack is by discrete IGSCC cracks. General IGA was
not observed to extend to more than 20 jim in depth for any exposure up to
4,500 hours. IGSCC was found in 1,000 hours of exposure in samples of

MA material prestrained 3.5%. MA samples strained to lower values
exhibited IGSCC after 3,000 hours of exposure. 86% of MA samples exposed
for 4,500 hours had initiated IGSCC greater than 110 ýtm deep, but times to
crack initiation were widely scattered. The maximum crack depth observed
in samples tested at 650°F (343 0C) in 1% NaOH was through wall after
4,500 hours exposure. Average crack growth rates in MA material were
higher than at 600°F (316'C), and the percentage of MA C-rings that
experienced cracking was also greater at the higher temperature. The
fraction of samples that had experienced IGSCC after 4,500 hours of
exposure appeared to be independent of applied strain in the C-ring, but
crack growth rates were greater in highly strained. No cracking (IGA or
IGSCC) of significant depth was observed in TT C-rings in 1% NaOH at
650'F (343QC) with any level of prestrain. IGSCC crack depths were less
than 2% through wall (20 jim) in all TT samples (Figure 13-4).
FIGURE 13-4
Histogram of Maximum IGSCC Depths for MA and TT Samples in
10% NaOH at 630°F (3320 C) (13.1)
70
60
50
U)
-1
40
0~
CO)
30
z
20
10
0
<2% 2-5% 5-10% 10-25% 25-50% 50-75% >75%
IGSCC DEPTH (% wall)

In summary, IGSCC in 1% caustic solutions is accelerated by increasing

the temperature from 600°F (316 0 C) to 650'F (343°C). Shallow IGA is
observed but IGSCC is the more dominant mode of attack. Both average
crack depths and the number of cracked samples increase with time. TT
material shows more resistance to IGSCC in 1% caustic than does MA
material.
10% Caustic. A number of studies of intergranular corrosion have been

performed using 10% NaOH, 10% KOH, or a mixture of the two with a
total concentration of 10%. In Jacko's 1990 study (13.1), C-ring samples of
MA, TT, and mill annealed and sensitized (MAS) material were exposed to
10% NaOH at three temperatures, 288, 302, and 630'F (332°C). Some of
these tests repeated tests done earlier on the same heats of material used by
Airey (13.2). Some other data for 10% NaOH are available from work by
Van Rooyen and Bandy (13.5).
Jacko's experiments (13).1 included exposures of 1,000 to 8,000 hours
in 10% NaOH at 550'F (288 0C). Twelve of 15 C-rings of MA material with
long exposure times (Ž4,000 hours) experienced cracking in excess of 110 jtm.
The average deepest crack depth in these samples was about 50% of the
wall thickness (550 jtm). Crack depths for the various exposure times were
analyzed assuming that crack growth occurs at a constant rate following
an initiation time (Figure 13-5). Based on this analysis, the initiation time
was estimated to be 1,000 to 2,000 hours and the average crack growth rate
was approximately 0.06 jtm/h.
Additional samples were examined after 1,000 and 2,000 hour
exposures. No IGSCC was found after 1,000 hours of exposure, and only
one sample of MA material had significant cracking after 2,000 hours
indicating that the average initiation time for IGSCC is probably somewhat
greater than 2,000 hours at 550'F (288°C) in 10% NaOH. No cracking in
excess of 110 jim depth was found in 15 TT C-ring samples from the same
heats of material exposed in the, MA condition. The maximum depth of
IGSCC found in the TT samples was 50-70 gm. Samples in the MAS
condition were also exposed to 10% NaOH at 550°F (288°C) for up to 8,000
hours. The maximum crack depth in MAS samples was 540 jtm after 8,000
hours. This was less than the maximum depth of 680 jtm in MA material,
but significantly greater than the maximum depth in TT material. Detailed
comparison of the TT, MAS, and MA sample results indicate that the MAS
material, on the average, experienced less severe cracking than the MA
material, but more than TT material under the same conditions.
Exposures to 10% NaOH at 575°F (302'C) were carried out for up to
8,000 hours. A number of samples were examined after relatively short
exposures (1,000 to 3,000 hours) to determine initiation time. The initiation
time was found to be variable with the fraction of the population that
experienced cracking increasing with time. For MA samples, there was
virtually no initiation time for samples prestrained 3.5%. The average
initiation time for samples strained 0.2% was approximately 1,000 hours.
These results are essentially consistent with an earlier study (13.3), which
showed that at 572°F (300'C) the initiation time for intergranular corrosion
in 10% caustic (9% NaOH, 1% KOH) was approximately 240 hours for
samples that had been prestrained into the plastic regime (estimated stress
level 450 ± 30 MPa [65 ± 5 ksi]). The amount of IGSCC observed in Jacko's
experiments at 575°F (302°C) in 10% NaOH was anomalously small in
comparison with results at 550 and 600'F (288 and 316'C). The kinetics of
crack growth were analyzed and the mean growth rate for MA material
was determined to be 0.049 iim/h. No cracking as deep as 110 gim was
observed in C-ring samples of TT material at 575°F (302'C) in 10% NaOH.
The deepest cracks observed were 30 itm. Average crack growth rate for
TT samples was approximately 0.04 gm/h. No significant IGA was found
on any sample of either MA or TT material for these exposure conditions.
FIGURE 13-5
Maximum IGSCC Depth of Mill Annealed Alloy 600 As a Function of
Exposure Time at 550°F (288 0 C) in 10% NaOH (13.1)
1000
800
600
U)
400
200
0 2,000 4,000 5,000 6,000
Exposure Time (h)

Tests performed in 10% NaOH at 630'F (332°C) show that this

environment is very aggressive for alloy 600 material. Both MA and TT
samples experience significant IGSCC under these conditions. Intergranular
corrosion initiates in MA material in approximately 1,000 hours. Uniform
IGA 20 to 70 jtm deep is observed in MA samples after 3,000 hours of
exposure. Discrete IGSCC cracks extending deeper than the IGA layer are
observed after 5,000 and 7,000 hour exposures. All MA C-rings prestrained
3.5% experienced cracking in 3,000 hours. Seventy percent of all MA samples
experienced significant IGSCC (>110 jtm deep) under these exposure
conditions. The average crack growth rate for MA samples was 0.091 gim/
h. C-rings of TT material experienced significant IGA and IGSCC in 10%
NaOH at 630'F (332°C'). Eight of 54 TT samples experienced IGSCC greater
than 110 jtm in depth. The average crack growth rate for TT material was
0.015 jim/hr.
The important observations from the results of exposures to 10% caustic
environments included:
" The initiation time for IGSCC decreases with increasing prestrain of C-
ring samples.
* Rate of attack generally increases with increasing temperature, but
anomalously few samples experienced IGSCC at 575°F (302°C) in
comparison with results at 550 and 600°F (288 and 316°C).
* Mill annealed and sensitized (MAS) material exhibits resistance to
intergranular corrosion that is intermediate to that of MA and TT
material.
50% Caustic. Tests in 50% NaOH at 550'F (288 0C) produced very little
attack on either MA or TT samples. The deepest penetration of IGA/IGSCC
on any sample was less than one grain diameter. These results, in contrast
to those for 10% NaOH at 550°F (288°C), show that the rate of attack
decreases dramatically with increasing NaOH concentration at this
temperature. Similar results showing very low rates of attack in
concentrated caustic had been obtained in previous studies (13.2,13.3). Airey
(13.2) found that the rate of IGSCC decreased strongly with increasing
concentration from 10 to 50% NaOH at 600°F (316°C). In a study by Pinard-
Legry and Plante (13.3) no IGA or IGSCC were observed in samples exposed
to 45% caustic (40% NaOH, 5% KOH) with 5% Na 2 SO 4 at 608'F (320°C) for
120 hours, whereas well defined IGA was produced by a 120 hour exposure
to 10% caustic (9% NaOH, 1% KOH) with 1% Na 2 SO 4 at 608°F (320°C).
These results indicate that the inverse dependence of rate of attack on caustic
concentration extends to a temperature of at least 608'F (320°C). The reason
for the decreasing rate of attack when the caustic concentration is increased
from 10% to 50% is not known, but. may result from a shift of the
electrochemical potential in these open circuit experiments.
Intergranular Corrosion of Alloy 600 From Caustic Compounds 13-1 1
At higher temperature, the rate of attack increases with increasing

caustic concentration. Tests in 50% NaOH at 630'F (332°C) produced
extensive IGA and IGSCC in both MA and TT material in 2,500 hours
(Figure 13-6). Many MA samples were cracked through wall in less than
1,000 hours. Significant IGSCC was also produced in 12 of 14 TT C-ring
samples exposed to 50% NaOH at 332°C (630'F). The average crack depth
for TT samples was 750 gim while that for the MA samples was 1,100 gim
(through wall). The histograms in Figure 13-6 show two peaks for both
MA and TT material. There are small peaks at less than 5% through wall
and larger peaks at greater than 75% through wall. This appears to indicate
that some samples are resistant to the initiation of IGSCC, but, once cracks
reach a significant depth (i.e., 5 to 10% through wall), they grow rapidly
such that the probability of finding IGSCC of intermediate depth is low.
FIGURE 13-6
Histogram of Maximum IGSCC Depths for MA and TT Samples in
50% NaOH at 630°F (3320 C) (13.1)
50 F]MA
~TT
40
(0
-J
C,) 30
U-
0
wo
20 -
z
10 I
0
<2% 2-5% 5-10% 10-25% 25-50% 50-75% >75%
IGSCC DEPTH (%wall)

Extensive IGA was a characteristic result of exposure to 50% NaOH at

630'F (332°C). The average depth of IGA was 420 gm for MA samples and
80 gm for TT samples. The computed average rates of IGSCC penetration
were 0.59. gm/h for TT samples and 0.74 jim/h for MA samples. Airey
(13.2) has previously reported that the rate of IGSCC increased when the
concentration of NaOH was increased from 10 to 50% at 650'F (343°C).
The most significant results for alloy 600 exposed to 50% caustic
solutions are:
* 50% caustic is less aggressive than 10% caustic at 550'F (288°C), but
more aggressive at 630'F (332°C).
* 50% caustic is very aggressive to MA and TT alloy 600 at 630'F (332°C).
TT material experiences rapid IGA and IGSCC in 50% caustic at 630'F
(332 0C).
Effects of Other Dissolved Species

The effects of additions of various ionic species to caustic environments
have been studied extensively Additives have been found to both accelerate
and retard the rate of IGA attack of alloy 600. In some cases, addition of an
ionic species is found to accelerate attack at one caustic concentration or
temperature and retard the rate of attack under other conditions. Additives
that accelerate attack have been widely used to accelerate testing, while
those that retard attack have been studied for possible use in remedial
measures to slow the rate of degradation of tubes in operating steam
generators.
Many investigators have found that additions of carbonate and sulfate
to caustic solutions increase intergranular corrosion of alloy 600. Model
boiler studies (13.6) have shown that carbonates either alone or in
combination with sodium hydroxide produce rapid cracking of mill
annealed alloy 600. Several autoclave or capsule exposure studies (13.3
13.4 ý13. 13.7) report that addition of carbonate as Na 2 CO 3 or sulfate as
Na 2SO 4 accelerate the rate of intergranular corrosion in alloy 600 at
temperatures near 572°F (300'C).
A study by Conner, et al. (13.7) investigated the behavior of alloy 600 in
a large number of caustic environments including solutions with additions
of carbonate, nitrate, sulfate, silicate, phosphate, chloride, and fluoride. For
these tests, a reference environment of 40% NaOH + 10% KOH at 650'F
(343 0C) was used to establish a baseline to which the behavior in solutions
containing additives were compared. Carbonate was shown to increase
the rate of IGA relative to the reference solution, but sulfate addition was
found to have little effect. This result is in disagreement with a number of
other studies that have shown enhancement of IGA and IGSCC from sulfate
additions. It is possible that the effect of sulfate depends on the caustic

concentration and temperature. It should be noted that 50% caustic with
no additives has been shown to be a very aggressive environment at 650'F
(3430 C).
A series of tests by Van Rooyen and Bandy (13.5) at Brookhaven National
Laboratory (BNL) used sealed capsules containing solutions of NaOH alone
or with Na 2 SO 4, Na2 CO 3, or Na2SiO3 as additives. In these tests, the sealed
capsules were heated in an autoclave containing pure water. Because the
caustic solutions have elevated boiling points, the pressure inside the
capsules at any temperature was less than the pressure in the autoclave,
which contained pure water. Therefore, the capsule walls were loaded in
compression during the exposure. The loading tends to suppress IGSCC
so that IGA was the dominant mode of attack observed in these tests. The
BNL capsule tests showed that both sulfates and carbonates accelerate IGA
in sodium hydroxide. Solutions with 1% sulfate or carbonate were shown
to be slightly more aggressive than solutions with only 0.1% sulfate or
carbonate.
The BNL studies (13_4 13.5) included tests on stressed C-rings under
controlled electrochemical potential in NaOH containing carbonate, sulfate,
or silicate additions. Most tests used a 10% NaOH concentration with 0.1%
or 1% of the contaminant, but tests were also performed with 1% NaOH +
1% Na 2 CO3 , 1% NaOH + 10% Na2SO 4 , 10% NaOH + 10% Na2 SiO3, and
1% NaOH + 10% Na 2SiO 3 . The results showed that the solutions with
higher concentrations of carbonate and sulfate additions (1% versus 0.1%)
were more aggressive. A solution of 10% Na2SO 4 and 1% NaOH caused
rapid IGA but no IGSCC within the potential range -102 to +384 mV at
315 0C (599°F). No IGA or IGSCC was observed in the tests in 1% or 10%
NaOH with addition of 10% Na 2 SiO 3 which were conducted at the free
corrosion potential.
Studies conducted by Pinard-Legry and Plante at Commissariat 6
l'Energie Atomique (CEA) (13.3) have investigated the effects of additions
of carbonate, sulfate, chloride, and silicate to caustic solutions on the rates
of intergranular corrosion. Studies included controlled electrochemical
potential exposure tests on C-ring samples of alloy 600 samples in MA, TT,
and sensitized (MAS) conditions. Samples were loaded until they deformed
plastically, and the resulting stress level was estimated to be 450 ± 30 MPa
(65 ± 5 ksi). Potentials ranged from -20 to +20 mV relative to the free
corrosion potential. Exposures were for 120 or 240 hours. Temperatures
were 572, 608, or 662°F (300, 320, or 350'C). Environments included 10%
caustic (9% NaOH + 1% KOH) with additions of sulfate, carbonate, and
chloride. Additional exposure experiments using C-rings stressed to
200 MPa (29 ksi) without control of electrochemical potential were carried
out on MA and TT samples in caustic composed of 36% (NaOH + KOH)

with sulfate or silicate additions at 6627F (350'C). These low stress level
experiments were performed for the purpose of investigating IGA without
IGSCC. Tests in caustic containing sulfate produced more uniform IGA
than'those in a carbonate environment. In carbonate at 572°F (3000C), many
small closed cracks were observed mixed with the IGA. At 608°F (320'C)
and 662°F (350'C), more and deeper IGSCC cracks were observed. Solutions
containing carbonate and sulfate in combination were no more aggressive
than those containing these species individually.
Additions of silica (SiO 2 ) or sodium metasilicate (Na2 SiO 3) to caustic
solutions have mixed effects depending on the caustic concentration and
the temperature. In the experiments by Conner, et al., (13.7) capsule and C-
ring tests at 650'F (343°C) indicated that additions of silica to the reference
50% caustic solution decreased the rates of IGA but increased the rate of
IGSCC. In tests by BNL (13.5) at 599°F (315'C) with addition of 10% Na 2 SiO 3
to NaOH solutions (1% and 10% NaOH), neither IGA nor IGSCC was
observed. Studies by CEA (13.3) included exposures of C-rings of MA and
TT material stressed to 200 MPa (29 ksi) in solutions of 36% NaOH + KOH
(NaOH/KOH = 4) with addition of 9% Na2 SiO 3 . The low stress was used
to produce IGA without IGSCC. Samples were exposed to the solution at
662°F (350'C) embedded in magnetite powder or above a bed of magnetite
powder in solution. The samples were exposed for either 10 or 28 days.
The tests indicated that the silicate addition promotes IGA. Other CEA
tests in 10% caustic have shown that silicate had an inhibiting effect on
intergranular corrosion. When taken together, these results indicate that
the addition of silica or silicate to high concentration solutions of caustic
(40 to 50%)'makes the solution more aggressive to alloy 600; at lower
concentrations of caustic, however, the silicate additions reduce the
aggressiveness of the caustic solution.
The addition of boron in the form of boric acid (H 3BO 3 ) to caustic
solutions has been found to reduce the rates of attack of alloy 600 by
intergranular corrosion. Boric acid was first used in steam generators as a
denting inhibitor. However, evidence indicates that it may also be an
effective inhibitor of intergranular corrosion by caustic compounds in crevice
locations (13.8). Boric acid addition by either continuous feed or by crevice
flushing is used as a remedial measure to retard the rate of secondary side
degradation of tubes in operating steam generators. Several laboratory
studies have investigated the retarding effect of boron additions to caustic
solutions regarding the initiation and growth of IGA and IGSCC.
Model boiler studies with feed water faulted with carbonates and
caustics with and without addition of boric acid have been used to
demonstrate the intergranular corrosion-inhibiting effects of boric acid.
Work by Hermer, Prestigiacomo, and Wolfe (13.9) showed that addition of

boric acid to makeup water for a model boiler faulted with carbonates was
effective in preventing IGSCC. The reference carbonate solution (0.3 ppm
Na2 CO 3 in the makeup tank) reproducibly causes through wall cracking in
15 to 17 days. The model boiler with boric acid addition (2 ppm boron as
boric acid in the makeup) operated for 91 days without developing a through
wall crack. Another model boiler test by Daret (13.10) in the Ajax facility
showed no cracking at crevices after over 100 days of operation with makeup
water containing 4 ppm NaOH with boric acid added at a boron to sodium
molar ratio in the range 5 to 18. Operation of the model boiler with 4 ppm
NaOH added to the makeup water promotes rapid IGSCC.
The effect of boron additions to caustic on the rate of intergranular
corrosion was investigated by CEA (13.3) using C-ring samples exposed
for "120 or 240 hours to 50% caustic with additions of boric acid, and 10%
caustic with sulfate, boric acid, or sodium tetraborate. Exposures in caustic
with boron addition in the form of boric acid and sodium tetraborate were
performed to test the inhibiting effects of the boron compounds. In one set
of experiments performed in 50% caustic, no IGA or IGSCC occurred in the
baseline caustic solution or in the solutions containing boron so that the
inhibiting effects of boron could not be evaluated. Another set of
experiments was carried out using 10% caustic. In these experiments, IGA
was observed in the reference solution of 9% NaOH + 1% KOH + 1%
Na 2SO 4 . No localized corrosion was observed in the solutions containing
boron.
Experiments by Hirano and Takaku (13.11) have also shown an
inhibiting effect of boric acid on intergranular corrosion. In a solution of
20% NaOH + 4% Na 2 CO 3 at 350 0C (662'F), addition of 20% H 3 BO 3 caused
the rates of IGA and IGSCC on C-ring samples to be reduced by
approximately a factor of 2.
The IGA/SCC inhibiting effect of boric acid has been attributed to either
the formation of a protective coating on the alloy 600, or reduction of the
solution pH by the boric acid.
Other soluble compounds have been added to caustic solutions in C-ring
or capsule exposure tests. In the capsule exposure tests of Conner, et al.
(13.7), chloride, phosphate, and fluoride had little effect on the rate of IGA.
The addition of nitrate was found to decrease both IGA and IGSCC, but it
increased the rate of general corrosion. Tests by CEA (L3.3) with 1% chloride
additions to 10% caustic solution also showed that chloride had relatively
little effect on the rate of intergranular corrosion.
In summary, experiments to evaluate how the rates of IGA and IGSCC

are affected by the additions of other species to caustic solutions are as
follows:
" Carbonate and sulfate additions to caustic generally increase the rate
of intergranular corrosion.
* Boron addition as either boric acid or sodium tetraborate generally
retards the rate of intergranular corrosion.
* Silica and silicate additions have variable effects depending on the
concentration of the caustic and possibly on the temperature of the test.
* Most other soluble neutral salts that have been tested have relatively
little effect on intergranular corrosion rates when tested with large
amounts of caustic.
Effects of Presence of Solid Oxides

Locations that experience IGA or IGSCC in a steam generator are
frequently associated with the presence of sludge. This could be composed
of corrosion products carried into the steam generator with the feedwater
or by corrosion products from adjacent materials (e.g., tube support
corrosion products). Thus, there has been considerable interest in the effects
of solid oxides on the intergranular corrosion behavior of alloy 600 in caustic
solutions. Several laboratory studies of intergranular corrosion of alloy
600 in caustic solutions have investigated the effects of oxide powders on
the rates of IGA and IGSCC.
Studies by Conner, et al. (13.7) investigated the effects of a number of
solid oxides on the attack of alloy 600 by caustic solutions. These studies
included exposure of samples in the presence of Cr 203, CuO, and ZnO. In
this study the addition of Cr20 3 to the reference 50% caustic solution in
capsule samples had a pronounced accelerating effect on the rate of IGA.
The acceleration of attack by Cr 2 0 3 was even greater than the acceleration
observed by Na 2 CO 3 , which was the soluble additive that caused the
greatest increase in IGA in that set of tests. Capsule tests were performed
by Van Rooyen and Bandy (13.5) with exposures up to one year with 10%
NaOH + 1% Na 2SO 4 on the OD and 10% NaOH with oxide additives inside
the sealed capsules. The tests showed that 10% NaOH with 10% Fe30 4
added caused less intergranular corrosion than 10% NaOH + 1% Na 2 S0 4.
Capsules containing 10% NaOH with 10% Cr 2 0 3 showed much more attack
than the capsules containing NaOH and Fe 3 0 4 . Both IGSCC and IGA were
observed in samples exposed for 6 and 12 months to 10% Cr 2 0 3 in 10%
NaOH. However, these experiments did not show that Cr 2 0 3 increased
rates of IGA and IGSCC relative to the 10% NaOH + 1% Na 2 SO 4 when
comparisons were made between the attack on the ID and attack on the OD
of the capsules.
In the tests by Conner, et al. (13.7) IGA was suppressed when 7% CuO
+ 3% ZnO were added to capsule samples; however, the severity of IGSCC
in stressed C-ring samples was increased. When the CuO and ZnO addition
was combined with SiO 2 addition (7% CuO + 3% ZnO + 12% SiO 2 in 40%
NaOH + 10% KOH) the effect was more pronounced. C-ring samples in
this environment showed essentially no IGA, but suffered rapid IGSCC.
Airey (13.2) found that addition of CuO to lower concentration caustic
solution (10% NaOH) enhanced IGSCC. He also observed that in the
presence of CuO, the rates of IGSCC were comparable for MA and TT
material. However, Jacko (13.1) found that 1% CuO in 10% NaOH
accelerated IGSCC of alloy 600, but found a benefit of thermal treatment
for tests with exposures of 3,000 to 4,600 hours in this environment at 630'F
(332 0C).
It is generally concluded that the effects observed from the addition of
oxides such as CuO are due to changes in the electrochemical potential in
the presence of the oxide. CuO can act as an oxidizing agent for alloy 600.
This causes a shift to higher electrochemical potentials. Jacko (13.1) made
direct measurements of electrochemical potential in caustic solutions
containing the oxides commonly found in sludge. The environments
investigated included Fe 3 0 4 , NiO, Cr 2 0 3 , and CuO separately and in
combination. Only CuO was found to cause a significant shift in the
electrochemical potential. The shift varied with the concentration of CuO
from insignificant when 0.01% CuO was added to +400 mV with addition
of 10% CuO. With addition of 1% CuO to 10% NaOH the shift in
electrochemical potential was +210 to +250 mV. Controlled electrochemical
potential tests show that the maximum rate of IGSCC occurs at a potential
approximately +150 mV above the free corrosion potential (13.5). IGSCC
tests on C-rings at 630'F (3320 C) showed approximately the same rate of
IGSCC in the environment with 1% CuO as was observed in pure 10% NaOH
with the electrochemical potential controlled in the +210 to +250 mV range.
Thus, the hypothesis that the effect of CuO is to increase the electrochemical
potential is consistent with the observed corrosion behavior and direct
measurements of potential. The suppression of IGA is also consistent with
an increase in electrochemical potential because IGA is found in controlled
potential tests to occur in a band of potentials from -50 to +50 mV relative
to the free corrosion potential.
Work by CEA (13.3) included exposures of C-rings of MA and TT
material in solutions of 36% NaOH + KOH (NaOH/KOH = 4) with additions
of either 3.6% Na 2 SO 4 or 9% Na 2SiO 3 with the samples embedded in
magnetite powder or in the solution above a bed of magnetite powder.
Exposure was for either 10 or 28 days. The magnetite enhanced the IGA. It
was concluded from electrochemical potential measurements in the oxide,
that the effect of magnetite is not due to a shift in the electrochemical

potential. Therefore, it was concluded that the effect was probably due to
changes in the local chemistry in the presence of the oxide. Samples
embedded in magnetite exposed to 36% caustic containing silicate suffered
depassivation and generalized corrosion.
In summary, the rate of intergranular corrosion of alloy 600 in caustic
environments can be strongly affected by the presence of some metal oxides.
The largest effects have been observed from CuO and Cr 20 3 . The effect of
CuO is believed to be that it acts as an oxidant so that it increases the
electrochemical potential. Direct measurements of electrochemical potential
in caustic containing CuO appear to confirm this hypothesis. Some
investigators have found that magnetite, Fe30 4 , accelerates intergranular
corrosion, but others believe that it has no effect. Potential measurements
in magnetite show no significant shift in electrochemical potential relative
to the same conditions in the absence of magnetite. Therefore, the effect of
magnetite is assumed to change in the local chemistry; it may depend on
the impurities in the magnetite powder used.
Effects of Concentration
A variety of studies, where different concentrations of caustic were
systematically investigated, show that the dependence of the rate of attack
of alloy 600 tubing on concentration is complex and dependent on
temperature. At moderate temperatures (near 572OF [300 0 C]), the rate of
attack appears to peak at a concentration between 10 and 50% caustic. At
higher temperatures above about 625°F (330'C), the rate of attack appears
to increase monotonically with increasing temperature.
Van Rooyen and Bandy (13.5) studied the dependence of the IGA rate
on the NaOH concentration using capsule tests at 689°F (365°C). In these
tests, there was very little influence of the NaOH concentrations. However,
it should be noted that because of the nature of the tests, the samples exposed
to higher NaOH concentration were also exposed to higher compressive
stresses. In C-ring tests, Van Rooyen and Bandy found that the IGSCC
propagation rate had a strong concentration dependence up to a
concentration of 10% NaOH at 315'C at a potential of +150 mV relative to a
nickel electrode (Figure 13-7).
Jacko (13.1) found that 50% NaOH solutions were less aggressive than
10% NaOH solutions at 550'F (288°C), but that 50% NaOH was very much
more aggressive than 10% NaOH at 630°F (332°C). Pinard-Legry and Plante
observed no attack in 50% caustic solution after 240 hours at 608'F (320'C),
but well defined IGA after the same exposure time in 10% caustic at the
same temperature.
These results indicate that the rate of intergranular attack of alloy 600
by caustic solutions peaks at a concentration between 10 and 50% for
exposure temperatures up to approximately 608'F (320'C). At higher
temperatures, caustic appears to become significantly more aggressive as
the concentration increases to at least 50% concentration.
FIGURE 13-7
Crack Growth Rate in Alloy 600 at 600°F (315 0 C) at 150 mV Potential
Relative to a Nickel Electrode in NaOH Solutions As a Function of
NaOH Concentration
10
6 1.0
=-
C:
0r)
Ca
0ý
-•0.1
0.01
0.1 1.0 10
% NaOH
Effects of Temperature
Intergranular corrosion in the form of IGA and IGSCC are thermally
activated processes that occur more rapidly under a given set of conditions
as the temperature increases. The value of the apparent activation energy
can be determined from the variation of rate of attack as a function of
temperature. Several investigators have reported activation energy values
for IGA and IGSCC.
From his experiments in pure NaOH Jacko (13.1) determined the
activation energies as listed in Table 13-1.
TABLE 13-1
Activation Energy Values for IGA and IGSCC in Pure NaOH
IGSCC
1% NaOH 27 kcal/mole
IGA
These results are illustrated in Figures 13-8 and 13-9 which show the
rates of IGSCC and IGA penetration, respectively, versus reciprocal
temperature for the three caustic concentrations studied. The magnitudes
of the slopes of the lines are proportional to the apparent activation energies
of the degradation processes. It should be noted that the line for 50%
concentration is steeper than the lines for lower concentrations and crosses
them. This is an additional demonstration of the observation that 50%
caustic is more aggressive than lower concentrations at high temperatures,
but less aggressive at low temperatures. The nature of the data in the results
of Jacko's experiments is such that the high apparent activation energies
for IGA and IGSCC calculated for 50% NaOH have a high degree of
uncertainty. The data show that there is a near discontinuous increase in
rates of attack between 600'F (316'C) and 630'F (332°C). Therefore, the
calculated apparent activation energies for 50% NaOH may not be applicable
over the entire temperature range of the experiments.
An apparent activation energy for IGSCC was computed from the
results of studies by Pinard-Legry and Plante (13.3) in a 10% caustic plus
sulfate environment (9% NaOH + 1% KOH + 1% Na 2SO 4 ). An activation
energy was computed from crack depths at 572, 608, and 662°F (300, 320,
and 350'C) after exposure for 240 hours. These results give an apparent
activation energy value of 30 kcal/mole for IGSCC. This value agrees very
well with the 32 kcal/mole value obtained by Jacko in 10% NaOH.
Van Rooyen and Bandy (L3.5) measured the temperature dependence
of IGSCC rates using a long capsule in a temperature gradient furnace.
The capsule contained 10% NaOH on the inside with air outside. It was
exposed for 69 days in a tube furnace with a temperature gradient. Cracking
measurements were made at different locations on the tube with
temperatures ranging from 444 to 599°F (229 to 315'C). The data give an
apparent activation energy of 6 kcal/mole. This activation energy is
considerably lower than values obtained in the other experiments discussed

above which used separate samples exposed in the same chemical
environment at different temperatures. This may indicate that the metal
temperature on the ID of the capsule was more uniform than the furnace
temperature because of heat transfer within the capsule.
FIGURE 13-8
Temperature Dependence of the IGSCC Propagation Rate in Caustic
Environments for Mill Annealed Alloy 600 (13.1)
Temperature (°C)
340 320 300
1.000
%
50% NaOH
0.100
10% NaOH
CO
C:
0)
1% NaOH
2
CL
0.010 '-
0)
0.001 L I I I
1.60 1.65 1.70 1.75 1.80

0
1,000f!( K)
The apparent activation energy for IGA was obtained by Van Rooyen
and Bandy from isothermal capsules exposed at temperatures ranging from
558 to 689°F (292 to 365 0C). The capsules were filled with 10% NaOH and
were exposed for 40 days in autoclaves filled with pure water. The data for
IGA rate show straight line behavior on a plot versus 1/T (Figure 13-10)
with a slope indicating an activation energy of 18 kcal/mole. Because the
10% NaOH inside the capsules has an elevated boiling temperature (lower
pressure at given temperature), the capsules are in compression during the

autoclave exposure. The apparent activation energy measured in these
experiments is significantly lower than the 25 kcal/mole determined for
IGAin 10% NaOH in jacko's experiments. It is possible that the compressive
stresses on the capsule ID in the experiments of Van Rooyen and Bandy
influenced the results.
FIGURE 13-9
Temperature Dependence of the IGA Propagation Rate in Caustic
Environments for Mill Annealed Alloy 600 (13.1)
Temperature (0C)
340 320 300
1.000
0.100
U,
Cr'
0.010
0.001
1.60 1.65 1.70 1.75 1.80
1,0001T (°K)
On the basis of available experimental data, the best estimate for the
activation energy for IGSCC of alloy 600 in caustic environments with
concentration of 10% or less is 30 ± 2 kcal/mole. IGSCC in concentrated
solutions appears to have a much higher activation energy, possibly as high
as 72 kcal/mole.
The activation energy for IGA appears to be somewhat lower than that
for IGSCC, but it shows the same apparent increase for concentrated
solutions that is observed for IGSCC. The apparent increases in activation
energy observed for both IGSCC and IGA in 50% NaOH may indicate that
there is a change in the rate limiting process as the caustic concentration is
increased from 10% to 50%.
FIGURE13-10
Temperature Dependence of IGA Rate in 10% NaOH From Capsule
Tests Showing Activation Energy of 18 kcal/mole (13.5)
Temperature (°C)
292 315 340 385
0.1
(D 0.01
CC
0.001
1.80 1.70 1.80
1,000/T (-K)
Effects of Stress and Prestrain

IGSCC is generally accepted as being a tensile stress assisted process,
but there is very little quantitative information regarding the stress
dependence of IGSCC rate. Most laboratory tests of IGSCC in caustic
environments have been performed with applied stress levels near the initial
yield strength of the material (e.g., 90% yield) or by loading the sample
such that it undergoes some plastic prestrain. Since the plastic strain changes
the properties of the material by strain hardening, comparison of the

behavior of different samples strained different amounts does not yield
information about the independent effect of stress. In relatively short
duration laboratory tests, when samples are loaded to stresses significantly
below the yield strength of the material, it is generally found that the
behavior is dominated by IGA (13.7), if any degradation is observed at all.
This may indicate the existence of a threshold stress below which IGSCC
does not occur, or it may only indicate that the rate of IGSCC crack growth
is a strong function of stress, such that longer exposure times are required
to observe cracking at lower stresses.
Additional evidence that IGSCC is a strong function of stress comes
from the distribution of cracking on a C-ring test sample. C-ring samples
have a gradient in stress with a maximum at the sample apex. It is usually
observed that IGSCC in a sample is limited to a narrow band at the apex.
For plastically strained samples, the region of cracking may not include
even the entire region of plastically strained material.
Most measurements on stressed C-ring samples that compare behavior
of samples loaded by different amounts are actually comparing samples
loaded to near yield with samples that have undergone various amounts of
plastic deformation (13.1). The plastically strained samples are usually held
at their maximum deflection, so that the stress in the plastically strained
portions remains at the current flow stress of the material, except for any
relaxation that occurs. The stress in the plastically strained samples is higher
than the initial yield strength of the material because of work hardening
that occurs during the plastic deformation. However the material strength
is also higher. In interpreting the results of such experiments, it is impossible
to unambiguously determine the individual effects of absolute stress, stress
as a percentage of the current material flow stress, and plastic prestrain.
Results of experiments on stressed C-ring samples frequently report
the loads on the C-rings in terms of a pseudoelastic stress calculated from
the imposed displacement of the sample by a formula given in ASTM
Standard G38 (13.12). When the stress value calculated by this method
exceeds the yield strength of the material (e.g., 150% yield), the pseudoelastic
stress value represents the stress that would have been reached if the sample
had remained elastic. The actual applied stress in a plastically strained
sample is much lower than the pseudoelastic stress. The values of the actual
stress and plastic strain in the sample depend on the stress-strain
characteristics of the material.
Jacko made careful measuremrents of the stress strain behavior for the
C-ring samples used in his experiments. Those experiments used samples
strained to four levels: 0.13,0.20, 0.42, and approximately 3.5%. The lower
three strain levels correspond to pseudoelastic stresses, per the ASTM
formula, of 90,150, and 275% yield. The highest strain, 3.5%, is the condition
with the C-ring compressed "'til the legs touch" (TLT). The actual stresses
estimated for the three lowest deflection conditions are 40, 42-55, and 42-
55 ksi. No stress was estimated for the TLT condition, but those samples
were significantly work hardened. The actual range of stresses covered by
the lowest three deflection conditions is only about 30% of the mean value
of 50 ksi. This is considerably less than the range of strains which cover
more than a factor of 3 for the lowest three values and a factor of 27 including
the TLT samples. The various C-ring deflections used in the Jacko
experiments, therefore, represent more of a range of plastic prestrain than a
range of stress.
Plastic prestrain increases the susceptibility of alloy 600 to IGSCC. In
Jacko's experiments, a greater percentage of the C-rings with the largest
prestrain exhibited cracking than of the lower prestrain samples. However,
in most cases where a significant number of the TLT specimens experienced
IGSCC, at least some of the low prestrain samples also experienced IGSCC.
As shown in Figure 13-11, the behavior of the C-ring samples in 1% NaOH
at 600'F (316'C) can be described by a log normal distribution of crack
initiation times. The effect of the prestrain is that larger prestrains have
shorter mean times to initiation. Similar results were found for other
conditions.
FIGURE 13-11
Log Normal Probability Plot of IGSCC Initiation for Mill Annealed
Alloy 600 C-Rings in 1% NaOH at 6001F (316 0C) (13.1)
99
o 90
3.5% STRAIN /0
U)
uJ 70
'-50
• 30
0
10
- U0.2% STRAIN
1
100 1000 10,000 100,000
EXPOSURE TIME (h)

An effect of prestrain on the rate of IGSCC crack growth after initiation,

was not evident in Jacko's test results. The higher prestrain samples may
have had slightly higher growth rates, but the differences between samples
with different prestrains were smaller than the scatter in the results.
After initiation of a sufficiently deep IGSCC crack, the driving force on
the crack can be described by the crack tip stress intensity KI = o-ga where G
is the stress and "a" is the crack depth. Experiments reported by Takamatsu
(13.13) show that the rate of IGSCC crack growth in relatively dilute NaOH
solutions (.004 to .4% NaOH) increases rapidly with increasing stress
intensity in a threshold region where KI is between 3 and 6 MPaIm (2.8
and 5.6 ksi Iin). The stress intensity at the crack growth threshold depends
on the concentration of NaOH. When the stress intensity is greater than
about 6 MPa lm (5.6 ksi 4in), the crack growth rate is relatively insensitive
to K1.
The effect of stress on initiation and growth of IGA is not well defined.
IGA is usually considered not to be a strongly stress dependent process.
However, in laboratory experiments (13.2, 13.3), IGA is often limited to
stressed regions of the sample (i.e., the apex of C-rings) but is observed on
both tensile and compressive surfaces. These results probably indicate that
the mechanism of IGA is accelerated by localized plastic flow as, for example,
in a passivating film rupture mechanism. The required small amount of
plastic flow can be driven by either tensile or compressive stresses.
IGA may also occur in the absence of stress as the result of selective
dissolution of grain boundaries which, in turn, is the result of binding energy,
composition, or electrochemical potential differences between the grain
boundaries and the grain interiors. IGA has been observed on samples
exposed to caustic environments with no applied stress. However, these
samples may have had significant residual stresses in the surface layers
that experienced IGA.
The occurrence of IGA under compressive stress clearly indicates that
a high stress intensity at the tip of the IGA penetration is not required for
the IGA process to proceed because compressive stresses force cracks closed
and do not produce high stress intensities.
In summary, IGSCC is generally assumed to require a tensile stress that
is a substantial fraction of the yield strength of the material. Samples used
in short duration laboratory experiments are generally stressed to near yield
or into the plastic regime. The time required to initiate IGSCC is found to
decrease as the plastic prestrain in C-ring samples is increased, but the effect
is small in comparison to the scatter in time to initiation for different samples
prestrained the same amount. IGSCC may initiate at lower stress levels
after long exposures typical of the exposure times in operating steam
generators. IGA can occur at low stresses or in the absence of applied
stresses, but initiation of IGA is found to be accelerated in the highly stressed

portions of C-ring samples on both the tension and compression surfaces.
Effect of Electrochemical Potential

Electrochemical tests have been used to investigate the electrochemical
behavior of alloy 600 in caustic solutions and to define the ranges of
electrochemical potential that promote IGA and IGSCC in caustic
environments. Two types of electrochemical corrosion tests have been
perfoimed to help define the conditions that promote intergranular
corrosion damage of alloy 600: potentiodynamic polarization measurements
and controlled electrochemical potential exposure tests.
Through the combination of these two types of tests, the occurrence of
intergranular corrosion (IGA/IGSCC) can be correlated with the
electrochemical behavior of alloy 600 in caustic environments.
A typical polarization curve for mill annealed (MA) alloy 600 in caustic
solution is shown in Figure 13-12. Voltages are relative to a nickel electrode.
The curve exhibits an activation-controlled region, an anodic current peak
at approximately 100 mV, and an active-passive transition region extending
from the peak to approximately 150 mV. At potentials above 200 mV,
passivation of the alloy 600 is well established.
FIGURE 13-12
Polarization Curve for Alloy 600 in 10% NaOH+1% Na2CO3 Solution
at 572 0 F (3000 C) Showing Potential Regions Where IGA and IGSCC
Are Observed (13.5)
200 S
scc
60 scc
100 +
" IGA
-100
0
Current (pA)
13-28 .Steam Generator Reference Book, Revision 1
C-ring samples were strained to 150% of the yield strain and tested at
temperatures near 572°F (300'C) in 10% NaOH + 1% Na2 CO 3 . In these
controlled potential exposure tests, IGA was observed to occur for a band
of potentials from -50 to +50 mV relative to Ni as indicated in Figure 13-12
(13.5). In another study, the rate of IGA was found to be essentially
0
independent of potential in the range -20 to +20 mV (13.3). In a 10% NaOH
+ 0.1% Na 2CO 3 solution at 572°F (300'C), the IGA rate was observed to
increase as the potential increased from approximately +30 to +80 mV anodic
relative to a nickel electrode (Figure 13-13) (135 13.14). The IGApenetration
rate nears 0.6 jim/h at +80 mV.
FIGURE 13-13
Effect of Electrochemical Potential on the Rate of IGA for Mill
Annealed Alloy 600 in 10% NaOH + 0.1% Na2CO3 Solution at 572°F
(3000 C) (Potential is relative to a nickel electrode.) (13.5)
0.6
0
0.5 L
0.41-
S
0.3L
(.9
ID
Ca
CE S
0.2 _@
0
0
0.1 L
0 a a I I
-80 -40 -20 0 20 40 60 80
Potential, mV vs Ni
The polarization curve for alloy 600 in caustic at 572°F (300'C) exhibits
an anodic current peak near +100 mV. IGA and IGSCC are observed together
on both sides of this peak. IGSCC begins at potentials below the anodic
current peak. The rate increases with increasing anodic potential up to a
maximum rate peak between +150 and +200 mV depending on the
environment (Figure 13-14). The potential for the peak rate of IGSCC is in
the active passive transition region of the polarization curve. The peak
average IGSCC crack growth rate of MA alloy 600 in 10% NaOH containing
0.1% Na 2CO 3 at 572'F (300'C) is approximately 6 jtm/hr. The IGSCC rate
falls sharply at higher potentials as surface passivation becomes more firmly
established.
FIGURE 13-14
Effect of Electrochemical Potential on the Rate of IGSCC
Propagation for Alloy 600 in Three Heat Treatment Conditions
(10% NaOH + 0.1% Na2CO3 Solution at 572'F [300 0 C]) (13.5)
21
4
Cz
2
0
2
CD
0
0
Potential (mV vs Ni)
The correspondence between the laboratory results for the effect of

electrochemical potential on intergranular corrosion and actual conditions,
which exist in operating steam generators, is not well defined. In the absence
of oxidants, such as dissolved oxygen and copper oxides, it is considered
that the electrochemical potential in an operating steam generator should
be close to the free corrosion potential. This is a suitable potential for
occurrence of IGA, which is observed in laboratory studies in a range from
-50 to +50 mV with respect to the corrosion potential. IGSCC is observed
in laboratory experiments to be most rapid at elevated potentials (i.e., +150
to +200 mV). As discussed in Chapter 12, the electrochemical potential at
corrosion sites in operating steam generators can be elevated by the presence
of oxidants. In some cases, the occurrence of IGSCC in operating steam

generators has been attributed to periods of operation at elevated
electrochemical potential. However, it should be noted that the exposure
times used in the controlled electrochemical potential laboratory tests have
all been less than 300 hours. It is possible that the range of potentials at
which IGSCC can occur in highly stressed samples, during much longer
exposures is considerably broader than observed in the short laboratory
tests, and may extend to negative values of the electrochemical potential.
Thus, occurrence of IGSCC in the field may not require elevated
electrochemical potentials.
Effects of Heat Treatment and Material Microstructure

As with PWSCC, susceptibility of alloy 600 to IGA and IGSCC in caustic
environments is sensitive to the heat treatment that the material has received.
MA material shows greater susceptibility to intergranular corrosion by
caustics in the temperature range near 572°F (300'C) than most other heat
treatment conditions. The mill annealed and sensitized (MAS)
microstructure that is developed by relatively short-time heat treatments
in the carbide precipitation regime (approximately 1,050-1,200'F
[565-650°C]) has been found to be more resistant to intergranular corrosion
in caustic environments than the MA condition (13.1). This sensitized
microstructure is more susceptible to intergranular corrosion at low
temperature in sulfur-containing environments (e.g., polythionic acid) than
other standard heat treatments, because of chromium depletion at the grain
boundaries associated with the carbide precipitation. The thermally treated
(TT) condition of alloy 600 is achieved by a heating in the carbide
precipitation regime for a sufficient time to allow the chromium
concentration at the grain boundaries to be replenished by diffusion from
the grain interiors. A typical thermal treatment is 15 hours at approximately
1,300°F (700 0C). Thermally treated material generally exhibits superior
resistance to intergranular corrosion in caustic environments relative to both
MA and MAS, but the factor of improvement depends on the temperature,
the environment, and the microstructure established by the mill anneal prior
to thermal treatment.
The effectiveness of thermal treatment on corrosion behavior of alloy
600 in caustic environments has been extensively investigated. In all of the
testing performed by Jacko (13.1), equal numbers of MA and TT samples
from the same heats of material were exposed to each set of conditions
used in the tests. Thermal treatment was found to increase the average
time for initiation of IGSCC and reduce the average IGSCC growth rate as
shown in Figure 13-15. For 1% NaOH, and 10% NaOH at temperatures of
600'F (316'C) and below, the IGSCC propagation rate for MA material is 9
to 20 times greater than for TT material. Under the same conditions, the
fraction of TT samples, in which IGSCC initiated in a given exposure time,
was at least a factor of five less than for MA samples. Under more aggressive
conditions, 10% or 50% NaOH at 630'F (332°C) and higher, the IGSCC crack
propagation rates were only 1.25 to 6 times greater for MA material than
for TT material. Under these more aggressive conditions, the fraction of TT
samples experiencing IGSCC was generally less than the fraction of MA
samples under the same conditions, but the differences were not very large.
Thermally treated samples also showed increased resistance to IGA relative
to MA samples, but the factors of improvement were not as great as for
IGSCC. In many conditions, degradation was limited to a thin surface layer
FIGURE 13-15
Comparison of the Rates of IGSCC Propagation in Mill Annealed and
Thermally Treated Alloy 600 in 10% NaOH As a Function of
Temperature (13.1)
Temperature (°C)
340 320 300

1.000
10% NaOH
0
U
0.100
B MA
S0.01
CU
0 0 0
o MA 0.2% Strain
D MA Average All Specimens 9I
STT 0.2% Strain
* TT Average All Specimens
0.001 I
1.60 1.65 1.70 1.75 1.80
1ooo/T (OK)
(approximately 20 .tm)in samples of all material conditions. This is

presumed to be a layer with high residual stresses beneath a ground surface.
Cracking occurred rapidly in this layer in many test environments, but did
not penetrate deeper after longer exposures. Under these conditions, the
IGA depths were similar for MA and TT samples and IGA was not an
important mode of degradation relative to IGSCC. When IGA was an
important mode of degradation, TT material showed significantly greater
resistance than MA material, except at temperatures above 630'F (332°C)
in 10% and 50% NaOH where TT and MA material behavior was
comparable.
In a study by CEA (13.3), C-rings were stressed to 200 MPa to assess
the effects of thermal treatment on resistance to IGA and to examine heat-
to-heat variations in behavior. The tests were performed in solutions of
36% NaOH + KOH (NaOH/KOH = 4) with additions of either 3.6% Na 2 SO 4
or 9% Na 2 SiO 3 . Samples were exposed to the solution embedded in
magnetite powder or above a bed of magnetite powder in the solution for
either 10 or 28 days. The results showed that thermal treatment was
beneficial for all heats of material tested, but that it was more effective on
some heats than others. In addition, thermal treatment was less effective in
suppressing IGA in high concentration solutions of caustic than in less
aggressive media. This is consistent with the results of Jacko (13.1) who
also found that TT material suffered significant attack in concentrated caustic
solutions.
Van Rooyen and Bandy (13.5) studied the behavior of mill annealed
(MA), mill annealed and sensitized (MAS), solution annealed (SA), solution
annealed and sensitized (SAS), and cold worked (CW) samples of alloy
600. Controlled potential tests of C-rings in various caustic environments
showed that the MA heat treatment condition was the most susceptible to
IGA and IGSCC. Figure 13-14 shows that the rate of IGSCC crack
propagation is much higher in MA material than in SA or SAS material.
However, these data may be showing a combined effect of the susceptibility
of the material microstructure and stress. All samples were strained the
same amount, and because the yield stress in the SA and SAS material is
much lower than that of the MA material, the applied stresses in the C-rings
are lower for the SA and SAS material than for the MA material. However,
the results correlate with other tests that show that sensitization heat
treatment improves the resistance of alloy 600 to IGA and IGSCC in caustic
environments.
Samples in the MAS condition were exposed to 10% NaOH at 550'F
(288°C) for up to 8,000 hours by Jacko (13.1). The maximum crack depth in
MAS samples was 540 gim after 8,000 hours. This was less than the
maximum depth of 680 gim in MA material, but significantly greater than
the maximum depth of 70 jtm in TT material. Detailed comparison of the

TT, MAS, and MA sample results by Jacko indicated that the MAS material,
on the average, experienced less severe cracking than the MA material, but
more than TT material under the same conditions.
The observed effect of thermal treatment of alloy 600 on the corrosion
resistance in caustic environments is similar to its effect on resistance to
PWSCC (Chapter 7). Thermal treatment generally increases the time
required to produce cracking of alloy 600 in pure water, but the magnitude
of the effect is different from one heat of material to another. Some heats of
alloy 600 show little if any improvement in PWSCC resistance from thermal
treatment.
In addition to the effects of heat treatment, several studies have
investigated heat to heat variations of intergranular corrosion resistance.
Jacko found approximately a factor of two difference in rates of IGSCC
attack between the best and worst performing heats of material in both the
MA and TT conditions. However, there was no particular correlation
between performance of heats in the MA condition and their performance
in the TT condition. In addition, no significant correlation could be
established between material composition (e.g., carbon content) or
microstructural parameters (e.g., grain size, qualitative assessment of grain
boundary carbide density) and IGSCC rate. No quantitative metallographic
characterization of the samples was performed, so it is possible that subtle
.differences in microstructure that correlate with IGSCC resistance were
overlooked.
Studies of different heats of material under conditions that favored IGA
were done by Pinard-Legry and Plante (13.3). Although significant
differences in performance between different heats were noted, no
correlation was found between IGA resistance and alloy composition. No
correlation was noted between the IGA resistance of heats in caustic and
the IGSCC resistance in pure water as measured by constant extension rate
tests (CERT) and reverse U-bend tests.
MODEL BOILER TESTS

A variety of studies have been carried out in which laboratory model
boilers have been used to simulate steam generators so that the mechanisms
of concentration of impurity chemicals in hideout locations can be studied.
Unlike autoclave and other exposure testing techniques, the model boiler
tests make it possible to determine whether aggressive chemical
environments can actually be created from the small amounts of
contaminants deliberately added to the model boiler feedwater.
In an early study (L3.6), it was determined that 0.17 ppm of carbonate

added to the makeup water as sodium carbonate (Na2 CO 3 ) in the presence
of a simulated plant sludge would reproducibly cause through wall cracking
in 15 to 20 days. The simulated plant sludge contained copper and nickel
oxides as oxidizing species in addition to magnetite. In these experiments,
NaOH without carbonates in the presence of sludge, and carbonates in the
absence of sludge failed to produced significant cracking in 50 to 60 days.
The tubes in these experiments were subjected only to through wall pressure
stresses of approximately 11 ksi.
Tests in single tube model boilers operating with a 620'F (327°C) primary
temperature and a 535°F (2790C) secondary temperature using environments
contaminated with combinations of NaOH, KOH, Na2 SO 4, and Na2 CO 3
produced both IGA and IGSCC in MA alloy 600 tubes (13.16). In these
tests, the model boiler arrangement simulated both the tubesheet crevice
and tube support crevices. Only IGA was observed in the simulated
tubesheet crevices, but both IGA and IGSCC were produced in the simulated
tube support crevices. Portions of the tube surface within the simulated
tubesheet crevices were covered by either Cr 2 03 or a simulated plant sludge.
Simulated tube support crevices were tested packed with Cr 20 3 , Fe 30 4 ,
and empty. Porous alloy 600 frits were used above and below the simulated
tube supports to retain the prepacked Cr 20 3 or Fe30 4 . Frits were also used
above and below the empty simulated tube support crevices. The maximum
rate of IGA observed in the tubesheet was 1.06 jim/day in a crevice prefilled.
with 40% NaOH/10% KOH, with Cr 20 3 on the tube surface. Tubesheet
crevices prefilled with 10% NaOH/2.5% Na 2SO 4 and 10% Na 2 CO 3 /4%
NaOH experienced lower rates of IGA (0.73 jim/day and 0.61 jim/day
respectively). IGA at a rate of 0.51 jtm/day occurred in a simulated
tubesheet crevice that was not prefilled with an aggressive solution but
was operated with Na 2 CO 3 added to the feed water.
The environments in the simulated tube support crevices were created
by concentration of impurities in the feed water due to the boiler operation.
Higher rates of IGA were observed in the simulated tube support crevices
than in the simulated tubesheet crevices. The maximum rate of IGA
observed for all tests was 4.0 jim/day in an empty (however, closed by
porous frits) simulated tube support crevice in a Na 2CO 3 environment.
High rates of IGA were observed in tube support crevices that were empty
or filled with Cr 20 3 . Crevices packed with Fe30 4 showed generally lower
rates of IGA than empty or Cr 20 3 filled crevices.
IGSCC was observed in these model boiler tests in simulated tube
support crevices that were either empty or Cr 20 3 filled. The maximum
observed rate of IGSCC penetration was approximately 127 gim/day in an
Na 2 CO 3 environment. IGSCC was also observed in tests conducted with
NaOH and NaOH/Na 2SO 4 environments. The only case of IGSCC reported
in the simulated tubesheet crevice was at the top of the roll transition (i.e.,
in the presence of high stresses) in the test where the crevice was prefilled
with 40% NaOH/10% KOH.
Long-term model boiler tests lasting for up to 6,000 hours have been
using simulated caustic and river water as pollutants (13.16, 13.17). The
model boilers in these experiments have a simulated full-depth tubesheet
crevice and a simulated sludge pile above the tubesheet. In the tests, the
sludge pile was composed of either a synthetic mixture which models actual
plant sludge, including a high copper content, or magnetite powder
(typically technical grade containing sulfate, hematite, and other impurities).
In early tests which were started with a clean crevice, tubes in boilers
contaminated with caustic and carbonates, or caustic and sulfates,
experienced rapid IGSCC in the sludge pile region before attack in the
tubesheet crevice could occur. In subsequent tests, the crevices were prefilled
with a 40% caustic solution. Even using this method, caustic escaping from
the crevice caused early IGSCC failure in the sludge pile. However, by
sleeving the tubes after the IGSCC cracking occurred, it was possible to
continue the tests for more than 3,000 hours. Examination of the tubes at
this time showed that IGA typical of that found in operating plants had
been produced in the simulated tubesheet crevice.
In the caustic tests both IGA and IGSCC were observed. There was a
clear distinction between the surface condition of sample tubes that
experienced IGA and those that experienced IGSCC. The tubes affected by
IGSCC had thick layers formed of a mixture of oxide and deposits. These
layers had iron enrichment in the outer portions and chromium enrichment
near the oxide/metal interface. Tubes affected by IGA had only a very thin
oxide layer and no significant enrichment of any of the alloy constituents.
However, sulfur was found on the surfaces of tubes that experienced IGA
whether or not sulfur was purposely added to the environment. The thick
oxide coatings on the samples suffering IGSCC and the thin oxides on
samples suffering IGA are consistent with the controlled electrochemical
potential tests which showed that IGSCC predominates in the active-passive
transition region, while IGA is favored in the active region when the
potential is insufficiently positive to cause passivation.
Mill annealed and thermally treated alloy 600 tubes were included in
the model boiler experiments using caustic environments. Mill annealed
and thermally treated alloy 600 and mill annealed alloy 690 tubes were
used in the river water experiments. Of these, the MA alloy 600 tubes were
the most susceptible to IGA and IGSCC. MA alloy 600 samples exhibited
IGA to depths greater than 20% through wall after 3,300 hours of operation
in NaOH/Na 2 SiO 3 and NaOH/Na 2 SO 4 environments. Attack was
somewhat greater in the tests containing sulfate additions than in tests

containing silicate additions. The IGA on the TT alloy 600 tubes was
approximately 5% of the wall thickness and was the same in the
environments containing silicate and those containing sulfate.
In the river-water-faulted tests only minor IGA was observed on the
MA tubes after 7,000 hours of operation. The TT alloy 600 material and the
alloy 690 material suffered no attack in the boilers faulted with river water.
The results of the model boiler tests which simulated concentration of
caustic environments in steam generators can be summarized as follows:
* IGSCC of alloy 600 can be caused reproducibly by operation of a model
boiler with feed water containing Na 2 CO3 in the presence of simulated
plant sludge.
* Tubes experience IGA in model boilers with simulated tubesheet
crevices prefilled with combinations of NaOH, KOH, Na 2 SO 4, and
Na2 CO 3 when the tube surface is coated with Cr 2 0 3 or simulated plant
sludge.
* Both IGA and IGSCC can be produced in model tube support crevices
in model boilers operating with additions of the combinations NaOH,
KOH, Na 2SO 4 , and Na 2CO 3 . Degradation is experienced both in
simulated tube support crevices initially packed with oxides or sludge
or initially empty when the crevices are closed by porous frits to assist
the concentration of caustic impurities.
MECHANISMS OF IGA/IGSCC IN CAUSTIC

Laboratory evidence indicates that IGA/IGSCC in caustic occurs by
anodic dissolution of the grain boundary (13_•_ 13.5). It is also generally
accepted that oxide film rupture by localized microplastic deformation (slip
dissolution) is an important part of the IGSCC process (.13_113.5). However,
a detailed model of the mechanism that is consistent with all of the
experimental evidence has not yet been proposed. The processes of IGA
and IGSCC in caustic solutions are known to be affected by a number of
material and environmental variables, but the details of these effects, and
the ways in which they relate to the anodic dissolution and oxide film
rupture processes, are not well established. The following effects have been
proposed to play a role in IGA/IGSCC degradation of alloy 600:
0 dealloying (chromium depletion),
0 segregation of impurities within the alloy to grain boundaries,
0 modification of passivating surface films near grain boundaries by
chemical species from the grain boundary or environment,
* enhanced plasticity along grain boundaries, and

* electrochemical potential differences between the grain boundary and
grain interior induced by grain boundary structure and chemistry.
Oxide Film Rupture

The results of electrochemical tests which correlate occurrence of IGA
and IGSCC with anodic polarization curves are interpreted as providing
support for the oxide film rupture mechanism of IGSCC (13.5). According
to this model, IGSCC occurs because the passivating film on the material is
repeatedly ruptured by localized plastic strain events allowing anodic
dissolution at the site of the rupture until repassivation can take place.
IGSCC begins to occur when the potential becomes sufficiently anodic to
cause anodic dissolution and dies off rapidly when the potential is far
enough into the passive region to quickly form a strong passivating film.
This is considered to be consistent with the film rupture anodic dissolution
model for IGSCC (13.5).
The film rupture anodic dissolution mechanism is based on the fact
that a clean metal surface is rapidly dissolved in aqueous Caustic solutions.
The stability of the metal in such solutions is dependent on an inert oxide
layer covering the surface and passivating it. The passivating oxide films
on metals are usually brittle so that they will not deform plastically along
with the underlying metal, but will fracture instead. Thus, plastic
deformation of the metal substrate ruptures the oxide and allows the metal
to be dissolved. The magnitude of the plastic deformation that is presumed
to be responsible for the film rupture is very small and can consist of the
emission of only a few dislocations from the metal near a crack tip. Once a
crack has been formed, the process is self perpetuating. The presence of
the crack creates a high stress intensity at the crack tip which has associated
high local stresses. In a metal, these high stresses relax by local plastic
deformation. The deformation ruptures any passivating oxide film formed
on the crack surface. This allows dissolution of the crack tip until a new
film can form. The dissolution changes the local stress conditions causing
additional plastic deformation and then the process repeats itself.
The film rupture, anodic dissolution process depends strongly on the
kinetics of the normal pAssivation process and the properties of the
passivating layers that are formed. Under conditions where no passivation
occurs, dissolution will not usually be localized at regions of high stress
intensity such as crack tips. General wastage, rather than cracking, is usually
observed when passivation does not occur. However, it is noted that IGA
is observed in unstressed samples at electrochemical potentials that are
below the passivation potential and extend to cathodic potentials (±50 mV
relative to the free corrosion potential). When passivation is very rapid

(e.g., at high anodic potentials) rupture of the passivating layer is healed
before significant local dissolution can occur. The relative rates of dissolution
of depassivated and passivated regions of the surface, and the propensity
for cracking, are determined by the composition and structure of the oxide
layer that forms on the surface. This in turn is controlled by the chemistry
of the environment. Some species may prevent effective passivation by
causing the oxide layer to be porous and nonprotective, which promotes
general wastage over a cracking mechanism. Chemical species in solution
can also affect the mechanical strength of the passivating film. Some species
strengthen the film so that it prevents local surface deformation and resists
rupture; other species produce an oxide film structure that is particularly
susceptible to rupture.
One means for studying the nature and kinetics of the passivation
process is through the electrochemical polarization experiments.
Experiments have been performed to determine the effects of temperature
and additions of various contaminants to caustic solution on the polarization
characteristics (13.5). Increasing the temperature above 572'F (300'C)
increases the anodic current density in the passive region, but has little
effect on either the potential for the active-passive transition or the current
density at the anodic current peak. Similarly, the addition of carbonates
and sulfates to NaOH solutions primarily affects the current density in the
passive region of the polarization curve. Addition of 1% Na 2 CO 3 or 1%
Na 2SO 4 to 10% NaOH increases the passive region current density by
approximately a factor of 5. Since the effects of impurities on the polarization
curve are primarily to increase the passive currents without affecting the
critical potentials, it is deduced that these impurities change the nature of
the passivating film rather than affect the electrochemical potential. Such
changes in the passivating film affect the rates of IGA and IGSCC by either
making the film less protective when intact, or changing its strength to make
it more susceptible to rupture by local plastic deformation.
Grain Boundary Properties

Corrosion-induced cracking of alloy 600 in caustic solutions is always
intergranular in nature. (Environments containing lead that can induce
transgranular cracking of alloy 600 are discussed in Chapter 11). A number
of factors have been proposed to explain why grain boundaries are preferred
sites for film rupture and anodic dissolution, leading to intergranular
corrosion. The following explanations have been proposed:
" Microplastic strain can be concentrated along grain boundaries.
* Grain boundaries have higher energy than surrounding material and
thus may dissolve more rapidly.
" Grain boundary composition can differ from bulk metal composition.
" The composition and properties of passivating films at grain boundaries
may differ from those over the remainder of the surface.
Plasticity at Grain Boundaries. Grain boundaries are barriers to plastic

deformation so that residual stresses are created near grain boundaries when
a metal is plastically deformed. These residual stresses represent stored
energy that can be released if local conditions change as, for example, by
cracking of a nearby segment of grain boundary. In this way, when an
increment of intergranular cracking occurs, it causes additional microplastic
strain in its vicinity to release the built-up residual stresses along the grain
boundary. This localized plastic deformation can cause rupture of the oxide
film on the crack surface at the crack tip, allowing rapid anodic dissolution
to occur. This advances the crack and starts the process again. In this
scenario, the driving force for the plastic deformation is stored energy
associated with residual stresses along the grain boundary.
If applied stresses are added to the residual stresses, the process can be
accentuated because the applied stresses assist the residual stresses in
producing the localized plasticity. Since plastic deformation is driven by
shear stresses, both tension and compression loading can accelerate localized
deformation along grain boundaries near an advancing crack. This is
probably the reason why IGA is often observed to be accelerated on both
the tension and compression surfaces of C-ring samples (13.3)..
The film rupture mechanism for stress corrosion cracking (SCC) relies
on film rupture caused by the stress intensity at a crack tip. If a crack has
already formed, a tensile stress has the effect of opening up the crack and
creating a high stress intensity at the crack tip. Since the stresses surrounding
an atomically sharp crack tip in a tensile stress field would be very high if
the deformation remained elastic, local plastic strain occurs in the region of
high stress intensity to relax the stresses and blunt the crack. Thus, under
applied tension, the localized plastic strain required for film rupture is an
inherent effect of a crack. This is true whether the crack tip is near a grain
boundary or in a grain interior. The reason why grain boundaries are
preferred paths for crack propagation when IGSCC is observed, may be
associated with either the structure or chemical composition of the grain
boundaries.
Grain Boundary Energy. Intergranular attack occurs in a potential band

from approximately -50 mV to approximately +50 mV relative to the
corrosion potential (Figure 13-12) (L3.5). This region incorporates the
activation-controlled region of the polarization curve and is significantly
below the anodic current peak that occurs at approximately +100 mV. This
indicates that IGA occurs by an active dissolution mechanism. An active
dissolution mechanism for IGA is consistent with the observation

(Figure 13-13) that the IGA rate increases as the electrochemical potential
approaches the anodic current peak potential.
It is postulated that IGA occurs because the composition and structure
of the grain boundaries are different from the bulk of the material. The
crystal structure mismatch at a grain boundary implies that atoms adjacent
to the boundary are less tightly bound to the metal than atoms in the grain
interior (13.1. 13.5). This binding energy difference alone could be
responsible for rapid dissolution of the boundary when its passivation is
locally breached. In addition to changes in the binding energy, the grain
boundary structure and composition may change the electrochemical
potential of the boundary. Grain boundaries may be anodic relative to the
remainder of the material giving rise to local corrosion cells where the grain
boundary is preferentially dissolved (13.5).
Jacko (13.1) has used model alloys to simulate grain boundaries in alloy
600. These alloys were enriched in phosphorous and boron and heat treated
to obtain an amorphous structure. Electrochemical polarization
measurements in NaOH solutions showed that anodic current densities
for the model grain boundary alloys were greater than those for alloy 600
in the potential regime near the corrosion potential. Under some conditions,
the current density differences between the model grain boundary alloys
and alloy 600 indicated that the ratio of the grain boundary corrosion rate
to the bulk alloy 600 corrosion rate could be as high as 100 to 600. The
ratios of anodic current densities between the model alloys and alloy 600
exhibited dependencies on NaOH concentration and temperature that are
consistent with the concentration and temperature dependencies of IGA
and IGSCC rates in Jacko's experiments in caustic (13.1).
Chemical Segregation at Grain Boundaries. Another proposed reason

for localization of cracking along grain boundaries is that chemical
composition differences between the grain boundary and the bulk of the
metal interfere with the formation of effective passivating films. It is
observed that sensitization heat treatments improve resistance to
intergranular corrosion in caustic (.13_113.3). It is known that sensitization
depletes chromium from the regions near the grain boundaries as
intergranular carbides grow. Therefore, it appears that chromium depletion
is not a cause of intergranular corrosion of alloy 600 in caustics. However,
other chemical composition effects could be responsible for changing the
nature of oxide films at grain boundaries.
Segregation of various impurity species to grain boundaries is a well
established phenomenon in steels. It is difficult to determine whether
segregation of impurities to grain boundaries plays a role in intergranular
corrosion because the aggressive environment alters the grain boundary

surface during and after the cracking process. However, segregation of
impurities to grain boundaries is expected on theoretical grounds. It has
been proposed that impurity segregation at grain boundaries in alloy 600
changes the way that oxide films form at the boundaries to make the films
less protective. Impurity segregation is also invoked to explain grain
boundaries being anodic relative to the bulk of the material.
Composition differences at grain boundaries due to impurity
segregation can be confined to a few atomic planes near the boundary.
Therefore, it is extremely difficult to experimentally measure the grain
boundary composition. Measurements of the grain boundary surfaces after
they have experienced corrosion-induced cracking cannot unambiguously
determine the grain boundary chemistry, for it is probable that the thin
layer of altered composition is dissolved or changed either during the crack
propagation process or by exposure to solution behind the crack tip. In
most respects, the role of chemical segregation to grain boundaries in
intergranular corrosion of alloy 600 remains speculative.
Grain Boundary Carbides. Jacko (13.1) studied the electrochemical

properties of bulk Cr 7C 3 carbides to help determine the role of grain
boundary carbides in IGA and IGSCC processes in alloy 600. Large Cr 7C 3
carbides were formed by hot isostatic pressing of powder compacts.
Polarization curves were obtained from the carbides in 10% and 50% NaOH
solutions at 600°F (316'C) and 630'F (332'C). The polarization curves
indicated that the anodic currents are greater for the carbides than the alloy
600 in the potential regime near the corrosion potential. This indicates that
preferential dissolution of the carbide could play a role in the penetration
of cracking along the grain boundary, but appears to conflict with the result
that TT material which has a higher density grain boundary carbides than
MA material has substantially better IGA and IGSCC resistance.
These results support the hypothesis that the beneficial effect of the
grain boundary carbides on IGSCC resistance is mechanical rather than
electrochemical. According to this hypothesis, the carbides lock the grain
boundaries which prevents grain boundary sliding as a deformation
mechanism and reduces local plastic deformation adjacent to the grain
boundary. This mechanical strengthening of the grain boundary region, as
a result of the intergranular carbides, decreases the rate of oxide film rupture
and slip dissolution.
Alternatively, the presence of the carbides on the boundaries may have
some electrochemical effect that cannot be determined by measurements
on bulk carbides prepared by processes different from their precipitation
from an alloy 600 matrix.
Dealloying
Experimental evidence of chromium depletion from surfaces of both
laboratory and removed tube samples has been interpreted to mean that
IGA in caustic results from a dealloying mechanism. However, a cause
and effect relationship between dealloying and IGA has not been clearly
established.
Dealloying is the depletion of chromium and to a lesser extent iron
from the alloy 600. Surface analysis of a tube removed from a steam
generator at Point Beach Unit I showed depletion of chromium and iron
from the surface layer (13.18). Table 13-2 shows the in-depth composition
profile by Auger Electron Spectroscopy (AES) (for metals only) of the Point
Beach tube.
TABLE 13-2
AES In-Depth Composition Profile of the Corrosion Product on a
Boundary Near the Outside Surface of Alloy 600 Tubing From Point
Beach Unit 1. The Metals Are Normalized to 100%. (13.18)
Depth Normalized Atom Percent

,-- Ni Fe Cr
0 87.4 6.8 4.8
1 82.4 8.0 9.6
3 84.0 6.5 9.5
5 84.0 7.0 9.0
Surface analysis of laboratory samples has shown nearly complete

depletion of chromium from samples that have been exposed to caustic
environments that induce IGA (13.18). Correlations between the surface
analyses and rates of IGA indicate that the amount of dealloying observed
is greatest in those samples that experience the most rapid IGA• Chemical
equilibrium diagrams (Pourbaix diagrams) indicate that, of the major
constituents of alloy 600, chromium is the least noble and nickel the most
noble in high pH environments as shown in Figure 13-16. This is consistent
with the observed loss of chromium by dealloying. Experiments have also
shown that surface films formed on alloy 600 in caustic environments that
cause IGA are very thin (13.18). This indicatesthat the material does not
effectively passivate under caustic conditions that cause IGA. Chromium
depletion has also been observed in sample tubes from steam generators
that are suspected of having experienced caustic attack (13.18).
FIGURE 13-16
Pourbaix Diagrams for Iron, Nickel, and Chromium at 2880 C (5500 F)
1.0 I I I
0.5 FeO4
0 Fe 2O3 -
I '°
1.5
1-.0
.15 Fe H• P Ni02
-1.5 ~0.5 \ -
4 8 12
H :p "(b).
-..
~3
1.0 -1.0 -
0.5 t 4 C' -1.5 N1
HCrOO
20 4 12
pH
-.0.5
c o"(a
- )
3.
-2.0 I j
4 8 12
pH
Other laboratory experiments have been performed to test the cause

and effect relationship between IGA and dealloying (13.20). In these
experiments, samples were exposed to conditions that caused through
thickness IGSCC. These samples were then fractured in the vacuum
chamber of AES equipment and the exposed fracture surfaces were analyzed
for alloy content. In these experiments, no evidence of dealloying was found
on the newly created fracture surfaces, and the depth of altered chemical
composition on the fracture surfaces was extremely thin, less than 0.2 gm.
These results raise the question whether the dealloying observed in alloy
600 samples exposed to caustic is an active component of the IGA process,
or whether it is an incidental effect that occurs in the same environments
that cause the cracking.
Whether the dealloying is part of the mechanism of intergranular
corrosion in caustic, or simply a related phenomenon, it appears to be
characteristic of exposure of alloy 600 to caustic environments. Therefore,
a chromium-depleted surface layer is a signature of caustic attack. This
fact can be used to identify pulled tubes that have been exposed to a caustic
environment.
REFERENCES
13.2 "Optimization of Metallurgical Variables to Improve the Stress
Corrosion Resistance of Inconel 600." EPRI NP-3051. July 1983.
13.3 Pinard-Legry, G., and G. Plante. "Intergranular Attack of Alloy
600: Laboratory Investigations." EPRI NP-4223. August 1985.
13.4 Roberge, R., R. Bandy, and D. van Rooyen. "IGA of Alloy 600 in
High-Temperature Solutions of Sodium Hydroxide Contaminated
With Carbonate." EPRI NP-3059. May 1983.
13.5 Bandy, R. "Mechanisms of Intergranular Attack and Stress
Corrosion Cracking of Alloy 600 by High-Temperature Caustic
Solutions Containing Impurities." EPRI NP-5129. July 1987.
13.6 "Effects of Calcium Hydroxide and Carbonates on IGA and SCC
of Alloy 600." Westinghouse Electric Corporation, Pittsburgh, PA.
WCAP-10273. February 1983.
13.7 Connor, W. M., R. G. Aspden, R. Hermer, and N. Pessall.
"Neutralization of Tubesheet Crevice Corrosion." EPRI NP-3040.
May 1983.
13.8 Hermer, R. E. "Boric Acid Application Guidelines for Intergranular
Corrosion Inhibition." EPRI NP-5558. December 1987.
13.9 "Causes and Remedies of Alloy 600 IGA/SCC." Westinghouse

Electric Corporation, Pittsburgh, PA. WCAP-10523. July 1984.
13.10 Daret, J. "Boric Acid Inhibition of IGA." In 1987 EPRI Workshop on
Secondary-Side IntergranularCorrosionMechanisms: Proceedings,Vol.
2. EPRI NP-5971. September 1988, p. B2-1 to B2-24.
13.11 Hirano, H. and H. Takaku. "Intergranular Attack of Inconel Alloy
600 in Sodium Hydroxide Solutions Contaminated With Na 2CO 3
at 350'C." Energy and Environment Laboratory Report No. 285056.
13.12 "Standard Practice for Making and Using C-Ring Stress-Corrosion
Test Specimens." Annual Book of ASTM Standards, Vol. 3.02.
Philadelphia, PA: American Society for Testing and Materials, 1990.
13.13 Takamatsu, H. "Japanese Steam Generator Operating Experiences
and Results of Related R & D." Paper presented at EPRI Technical
Advisory Group Meeting, Philadelphia, PA, June 26-28, 1990.
13.14 Bandy, R., R. Roberge, and D. van Rooyen. "Intergranular Attack
and Stress Corrosion Cracking of Alloy 600 in High-Temperature
Caustic Solutions Containing Contaminants." EPRI NP-4051. June
1985.
on Alloy 600." EPRI NP-4978. December 1987.
13.16 Daret, J. "Intergranular Attack of Alloy 600: Simulation Tests."
13.17 Daret, J. "Intergranular Attack of Alloy 600 Tubing: Simulation
Tests." EPRI NP-5377. August 1987.
13.18 Lumsden, J. "The Possible Role of Dealloying on Initiation of IGC."
In 1987 EPRI Workshop on Secondary-Side IntergranularCorrosion
Mechanisms: Proceedings," Vol. 2. EPRI NP-5971. September 1988,
p. A6-1 to A6-29.
13.19 "Use of Pourbaix Diagrams to Infer Local Pitting Conditions." EPRI
NP-4831. October 1986.
13.20 McIntyre, P. "Auger Analysis of IGA-Affected Grain Boundaries."
In 1987 EPRI Workshop on Secondary-Side IntergranularCorrosion
Mechanisms: Proceedings, Vol. 2. EPRI NP-5971. September 1988,
p. A7-1 to A7-19.
14
INTERGRANULAR CORROSION
FROM ACIDIC COMPOUNDS
15
LEAD CRACKING OF ALLOY 600
A. K. Agrawal, Battelle Columbus Division
1989
INTRODUCTION
Alloy 600 steam generator tubes are susceptible to stress corrosion
cracking in high temperature, aqueous environments. The susceptibility
varies with the presence of certain contaminants, stress, and type of heat
treatment. Alloy 600 tubes have experienced cracking on the secondary
side (i.e., steam side) in most older PWR plants. The mode (i.e., morphology)
of cracking in the tubes usually is intergranular. The cracking has occurred
mainly at those locations in the steam generators where the contaminants
have concentrated from the secondary water by local boiling, e.g., in tube/
tubesheet or tube/tube support structure crevices, sludge pile regions, and
under heavy scales. The contaminants that have been considered
responsible for producing the secondary-side-initiated intergranular
cracking (IGSCC) in the tubes are caustic, acidic sulfate, reduced sulfate
species, and perhaps carbonaceous matter (carbonates).
In recent years, mixed mode cracking, i.e., IGSCC and transgranular
cracking (TGSCC), also has been observed in tubes in a few PWR plants. In
these cases, lead has been found on the tube surface and at crack faces.
Therefore, it is likely that lead or its compounds are corrodents or accelerants
in the mixed mode cracking cases. Lead is present as a contaminant in
practically every steam generator, but the concentration of soluble lead
usually is extremely small in comparison to the other contaminants.
The cracking of alloy 600 has been less thoroughly investigated in the
presence of lead than with the other corrodents mentioned above. Thus,
the purpose of this review is to collect and integrate the data that are
available. (After this review was written, several organizations have
initiated major programs investigating tube alloy cracking [alloys 600, 600T1,
800 NG, 690TT] in the presence of lead salts. All alloys are susceptible to
cracking in some lead-contaminated waters.)
PLANT EXPERIENCE
Sources of Lead
The specific sources of lead in the steam generators have remained
largely uninvestigated. Vaccaro and Miglin (15.1) have hypothesized the
general sources of lead to be plant makeup water, condenser cooling water
leaks, copper alloys in condensers, copper alloys in feedwater heaters, seals
and gaskets in pumps and pipe fittings, Babbitt alloys, i.e., Pb-containing
alloys in pumps and turbines, paints and preservatives, marking pencils,
and lead chromate tinting in polyethylene wrappings.
Lead Cracking of Alloy 600 15-3
However, there are other potential sources of lead, e.g., brazes and
solders used on pipes, metallic streaks left on equipment by the use of lead
mallets, and any radiation shielding material (lead brick or sheet) left
inadvertently inside equipment.
The concentration of lead is practically negligible in unpolluted, natural
waters which usually are used around power plants and in the cooling
water circuit. However, industrially polluted water, if used, could be a
significant source of lead. The concentration of lead in the heat exchanger
tubing made of copper alloys, in general, is <0.05 weight percent. The lead
concentration is considerably higher, 0.2 to 4 weight percent, in the copper
alloys (Muntz metal and bronzes) used in valves and for the tubesheet.
The corrosion of copper alloys would certainly introduce some lead into
the water streams.
The pumps are likely to be one major source of lead in the streams,
since they contain leaded materials such as bearings, lubricants, seals,
gaskets, etc. Release of lead from these materials probably occurs by leaching
and/or wear process(es).
Lead-containing paints have been used in the past to prevent corrosion
of steel components during transportation and storage. Kori 7 and St.
Lucie I are two plants, in which the source of lead contamination has been
traced to the paints on the interior of their respective turbines (15.1).
Typical Lead Levels

Lead has been found in steam generators at a number of domestic and
foreign PWR plants. The types of steam generators affected include
recirculating as well as once-through units. Plants known to have lead in
their steam generators (as of 1987) are listed in Table 15-1 (15.3-15.21):
The lead was found in these plants either in the tubesheet sludge, tube/
tubesheet crevice deposits, surface scale on alloy 600 tubes, and/or in steam
generator blowdown.
The concentration of lead was in a narrow range of 100 to 500 ppm in
the sludge deposits of most steam generators. In a few plants higher
concentrations of lead have been reported, e.g., Turkey Point 4 (900 ppm),
Prairie Island 1 (1,400 ppm), and Surry 2 (2,400 ppm). The concentrations
measured in the surface scale of tubes is typically <1,000 ppm, but lead
levels >1,000 ppm have been reported at Millstone 2, Point Beach 1, and
Zorita. An even higher concentration, Pb >10,000 ppm, has been detected
in very localized areas of the tube surface at Indian Point 3 and Farley 2.
At several plants, lead has been analyzed in filterables collected from
the steam generator blowdown (15.22). Lead was detected in every steam
generator or plant investigated. Its concentration in the filterables at these
plants ranged from 500 to 2,400 ppm.
TABLE 15-1
Lead Has Been Found in the Steam Generators of a
Number of Plants
Arkansas One-1 Palisades

Beznau-1 Point Beach-1
Calvert Cliffs-1 Prairie Island-1
Farley-2 Prairie Island-2
Genkai-1 Ringhals-2
Genkai-2 Robinson
Ginna San Onofre
Indian Point-2 St. Lucie-1
Indian Point-3 Surry-2
Kori-7 Tihange-1
Millstone-2 Trojan
North Anna-1 Turkey Point-4
Obrigheim Zion
Ohi-1 Zorita
Attempts to monitor lead content in streams at PWR plants have been

unsuccessful in the past because of the extremely low concentrations,
<1 ppb. Recently, lead has been monitored successfully in various streams
at Farley 1, Millstone 2, Prairie Island 2, and St. Lucie 1, using special
techniques (15.2). At each plant several streams were monitored, and lead
was found in trace amounts in all of them. The following streams were
monitored:
Farley 1: condensate pump discharge, heater drain pump
discharge, steam generator feedpump inlet, and final feed water.
Millstone 2: condensate pump discharge, condensate polisher
effluent, feedwater heaters, heater drain pump discharge, and
moisture separator reheaters.
Prairie Island 2: condensate pump discharge, heater drain pump
discharge, feedwater heater inlet, feedwater, and steam generator
blowdowns.
St. Lucie 1: condensate pump discharge, heater drain pump discharge,
heater, and feedwater.
Typically, the concentration of lead in most streams was <10 ppt at
Millstone 2 and Prairie Island 2, and <20 ppt at Farley 1 and St. Lucie 1.
The two streams that showed relatively high lead concentrations were the
heater drain pump discharge at Farley 1, and the steam generator blowdown
at Prairie Island 2. The average values of lead concentrations in these
streams were 170 ppt and 140 ppt, respectively. Of these four units, only
Prairie Island 2 is free of copper components in the balance of plant.
The reason for the high concentration of lead in the heater drain pump
discharge at Farley 1 has not yet been fully investigated. The presence of
high concentrations of lead in the blowdown at Prairie Island 2 indicates
that any lead brought in by the feedwater accumulates in the steam
generators. It also suggests that the blowdowns may be effective in
minimizing the lead inventory of the steam generators.
Cracking in Steam Generator Tubes

Steam generator tubes have experienced cracking from the secondary
side in most of the plants listed above. The cracking has been confined
primarily to the tube/tubesheet crevice region, sludge pile, and lower tube/
tube support intersections. The mode of cracking was exclusively IGSCC
in all these plants except at St. Lucie 1. A few tubes at St. Lucie 1 have
exhibited mixed mode cracking in the outer surface grains (15.18), but at
greater depths the cracking was IGSCC, as in the other plants. An example
of mixed mode cracking is shown in Figure 15-1. The cracking agent in
most of the plants has been considered to be either caustic or acid sulfate/
chloride, but not lead.
FIGURE 15-1
Example of Mixed Mode Cracking in Alloy 600
I .• :'" .L•. ,-
S50u
L1 .
Lead has been suspected to be a contributory or a primary cracking

agent at St. Lucie 1 (15.18), Calvert Cliffs 1 (15.5), and Farley 2 (15.6). The
suspicion was based on findings of significant amounts of lead on the
fracture faces of the tubes, and also in the surface scale. The lead was
analyzed on the fracture faces using energy-dispersive X-ray spectroscopy
(EDX) techniques, and the results reported were qualitative.
In the case of St. Lucie 1, the role of lead in promoting cracking was
considered to be only secondary. The environment at the tube/tube support
intersection, which was the cracking location, was determined to be very
acidic from seawater intrusions. The acidic sulfate environment itself was
sufficient to have promoted IGSCC in the tubes. The acid could have
dissolved lead from the scale and redeposited it on the fracture faces.
However, because of the TGSCC component in the cracking, lead was
considered a contributor in the cracking process. The case at Calvert Cliffs 1
was similar to that at St. Lucie 1, in that an acidic sulfate/chloride
environment existed in the region of cracking. In contrast, at Farley 2, a
freshwater plant, the surface scale was extremely rich in lead, up to 7.6
weight percent; therefore, lead was the prime suspect in cracking of the
tubes. In the case of Farley 2, cracking was by IGSCC only.
Eddy Current Detection

Eddy Current (EC) techniques are used in the field to inspect steam
generator tubes for the presence of defects. The technique, however, has
some limitations. The accuracy of EC results is typically ±20 percent of the
tube wall thickness in the case of defects produced by IGSCC. It is not
sensitive enough to detect small defects with penetrations of less than 20
percent through the wall. Sometimes, even larger defects go undetected,
particularly if the IGSCC produces tight cracks between grains. Such an
attack is typically produced on alloy 600 tubes in caustic environments.
The EC field inspections of St. Lucie 1, Farley 2, and Calvert Cliffs 1
steam generators have yielded much better results than the general industry
experience. Table 15-2 gives the results of field EC inspection, laboratory
re-inspection of the removed tubes, and metallographic confirmation of
the defects at St. Lucie 1. The types of EC probes used in the field and the
laboratory are also listed in Table 15-2. It has been surmised that the presence
of lead at the attacked grain boundaries was responsible in some way for
the better than average accuracy with the EC probes. However, also present
at the same grain boundaries were some Mg-Si-O compounds, which could
have formed an insulating or highly resistive film on the attacked surfaces.
Such an insulating film could have contributed to the better detection of
the defects by the EC probes. Therefore, the role of lead remains uncertain
in this better than average detection of the tube defects.
TABLE 15-2
Results of Eddy Current and Metallographic Examinations of Steam
Generator Tubes From St. Lucie 1 (15.19)
Tube Wall Penetration, %

Tube Segment Field Laboratory Laboratory Metallography
(Elevation) MFBC(1 ) MFBC 8xl Probe Results
120/12-2 41 40 _(2) 30
(Sludge Pile)
120/12-7 82 80 90 72
(#2 Eggcrate)
79/91-2 UDS(3 ) 20 50 17
(Sludge Pile- 57 50 60 42
Two Indications)
59/95-4 DSS( 4 ) _(5) 50 52
(#1 Eggcrate)
59/95-5 29 25 30 13
(#2 Eggcrate)
1
Multifrequency bobbin coil probe.
2
Not examined.
3
UDS = undefined signal.
4
DSS = distorted support signal.
5
Not detected.
LABORATORY EXPERIENCE
In the early 1960s, Copson and Dean (15.23) realized that lead, present
in some high temperature lubricants, could leach out into the secondary
water and enter the steam generators as a contaminant. They conducted
stress corrosion cracking (SCC) tests in lead-contaminated water on a typical
commercial heat of alloy 600 (C = 0.05 weight percent) and a relatively pure
laboratory heat of alloy 600 (C = 0.01 weight percent). The heat treatments
on the alloys were: commercial alloy = hot rolled/mill annealed [-1,800°F
(982.2°C)]; and laboratory alloy = hot rolled/2,000°F (1,093.3 0 C) annealed
and water quenched. The SCC tests were conducted in autoclaves at 600'F
(315.6°C), using single and double U-bend specimens. The test medium
was water saturated with air and ammoniated to pH 10. Four different
tests were run, each with a different type of lead contamination: (1) lead
oxide, (2) lead powder, (3) a 70-30 mixture by weight of lead powder and a
petroleum jelly, and (4) petroleum jelly containing approximately 1 percent
lead. The contaminant was placed at the bottom of the autoclave as well as
inside the crevice of the double U-bend specimen.
The alloy 600 specimens experienced cracking in 1,000 to 1,350 hours
in each test. The cracking was most severe with the PbO, which was
followed by the Pb powder, then the 70-30 Pb mixture; the cracking was the
least severe with the petroleum jelly. The mode of cracking was
predominantly TGSCC in the high carbon commercial alloy tested with the
PbO. In all the other tests the cracking mode was mixed, i.e., IGSCC and
TGSCC. It was concluded that lead-contaminated water can produce both
types of cracking in alloy 600, depending upon its heat treatment and the
test condition.
Copson and Dean (15.23) reported results from some other tests in which
the alloy 600 U-bend specimens were kept in air at 600'F (315.6°C) or 650'F
(343.3°C) (melting point of lead) in contact with pure lead. The specimens
experienced no cracking in 1,350 hours of exposure. This suggested that
the presence of both water and lead was required to produce cracking in
the alloy 600.
In the late 1960s, INCO screened a large number of experimental Fe-
Cr-Ni alloys in lead-doped water in order to develop SCC-resistant alloys
for use in pressurized water systems (15.24). The tests were conducted
using single U-bend specimens in deaerated; high purity water at 600'F
(315.6°C). Lead powder was placed at the bottom of the test autoclave.
Specimens of high nickel alloys were found to be very susceptible to SCC,
relative to the lower nickel alloys. The alloys found to be resistant to SCC
in lead-doped water followed an empirical relationship: Ni + 0.75Fe
<69 percent.
The apparent crack velocity was found to be very high in some of the
experimental alloys (15.25). An experimental alloy 600 sample which was
annealed and heat treated at 1,300'F (704.4°C) for 5 hrs, developed a 90 mils
deep crack in 18 weeks. The apparent crack velocity in the lead-doped
water in this case was 260 mpy.
Experimental heats of alloys 600, 690, and 800 also were tested in high
temperature solutions which contained lead salts, either Pb 3 (PO 4 )2 or PbC12
(15.26). In each case, these salts were added to provide one gram lead in
the test autoclave. The solution with PbC12 had a pH of 6, and that with
Pb 3 (PO4 )2 was adjusted to pH 10 by adding Na 2 HPO 4 and NaOH. In both
cases, the solutions were deaerated, the test temperature held at 600'F
(315.6°C), and the exposure period was eight weeks. U-bend specimens of
the alloys were tested in the annealed as Well as sensitized conditions. None
of the specimens of alloy 690 and alloy 800 experienced cracking in any of
the solutions. In contrast, the annealed specimens of alloy 600 experienced

cracking in both solutions. However, the sensitized specimens of alloy 600
were resistant to cracking in the Pb 3 (PO 4) 2 solution, and experienced only
minor cracking in the PbC12 solution. The results with these alloys generally
follow the empirical relationship given above, regarding cracking resistance
and the nickel alloy content. The results also indicate that alloy 600 in the
sensitized condition is more resistant to cracking than in the annealed
condition.
Further tests were conducted with five commercial heats of alloy 600,
in annealed and sensitized conditions (15.27). Single U-bend specimens
were exposed for eight weeks in pH 10, deaerated, Pb 3 (PO 4 )2 solutions at
600OF (315.6 0C). Lead phosphate, equivalent to one gram lead, was added
in each autoclave; these solutions were likely supersaturated with the salt.
All the annealed specimens from the different heats developed numerous
IGSCC cracks. The average depth of these cracks was 4 mils, which gave
an apparent crack velocity of 26 mpy. As expected, the sensitized specimens
were more resistant to cracking. The depth of cracking in the sensitized
specimens ranged from 0 to 1.0 mil, giving an apparent crack velocity of up
to 6 mpy.
The superior cracking resistance of alloy 690 was further evaluated with
longer exposures, using commercial heats of alloy 690 and alloy 600 (15.28).
A variety of heat treatments were used on the U-bend specimens. The 600'F
(315.6°C) test environment was deaerated water of pH 7 doped with lead
powder, and the exposure period was 18 weeks. Cracking did not occur in
any of the specimens of alloy 690. In contrast, all of the specimens of
alloy 600, whether annealed or sensitized, cracked. The cracking was
predominantly IGSCC in the annealed specimens, but predominantly
TGSCC in the sensitized specimens.
Babcock & Wilcox has conducted long-term cracking tests with
commercial heats of alloy 600 steam generator tubing in the presence of
lead in a typical OTSG environment (15.29 15.30). Types of specimens tested
included tube/tubesheet mockup, C-ring, double U-bend, and slotted tube.
Specimens were made from mill annealed and stress relieved tubing which
were probably sensitized. A lead foil was placed in intimate contact with
each specimen to simulate a severe case of lead contamination. The test
environment was deaerated water adjusted to pH 9.3-9.5 with ammonia
and maintained at 550'F (287.8'C), 1,250 psig. The specimens were
examined at different intervals. None of the specimens exhibited cracking
in the first 2,200 hours of exposure. At longer exposures, the highly stressed
specimens cracked but not the others. Tube tensile specimens, U-bend,
and C-ring specimens that were stressed to 125% of the room temperature
yield strength, cracked in 4,200 to 5,200 hours, but those stressed to under
the yield strength did not crack in 9,300 hours. None of the tube/tubesheet
mockup specimens cracked in 6,400 hours of testing.
The mode of cracking was exclusively TGSCC in the tube tensile
specimens, particularly in the areas that were in direct contact with the
lead foil. Some intergranular penetration (IGP) was also observed, but only
in areas of the specimens that were not in direct contact with the lead foil.
The depth of TGSCC was approximately 6.3 mils in 5,200 hours,
corresponding to an apparent crack velocity of 11 mpy. The TGSCC cracking
also occurred at the apex of the inner U-bend specimens. The outer U-bend
of the double U-bend specimens was steel in one case and alloy 600 in the
other, but these did not experience any cracking. The cracking was
predominantly TGSCC in three of the five C-rings, and IGSCC in the other
two C-rings. The maximum crack depth in the C-rings was 1.7 mils, thus
giving an apparent crack velocity of approximately 3 mpy.
Airey (1_ , 15.32) has investigated the effects of alloy purification and
various thermal treatments on the cracking properties of a commercial heat
of alloy 600 tubing in PbO-doped water. The test environment was
deaerated water treated with morpholine and ammonia with PbO powder
at the bottom of the autoclave. Tests were run at 630'F (332.2°C) for 1,350
hours, using C-ring specimens that were stressed to 90 or 150 percent of the
yield strength.
The purification of tubing was done in dry hydrogen by annealing a
batch of the stock at an intermediate stage of tube extrusion. A second
batch was extruded in a normal commercial production. The main effect of
purification of the tubing was to reduce the concentration of carbon and
the other trace elements. The thermal treatments given to the regular mill
annealed tubing specimens were 1,1000 F (593.3'C), 1,200'F (648.9°C), 1,300'F
(704.4°C), or 1,600°F (871.1 0C) for 1, 10, 24, or 100 hours.
All specimens tested at 150 percent of the yield strength cracked
irrespective of the material condition or thermal treatment. The mode of
cracking was predominantly IGSCC in the mill annealed and purified
tubings, whereas the cracking was predominantly TGSCC in the thermally
treated specimens. Most of the specimens that tested at 90 percent of the
yield strength cracked, but the cracking in general was less severe. The
specimens that did not crack were mill annealed and thermally treated at
1,300°F (704.4 0C) for 10 or 24 hours. This thermal treatment appears to
impart some resistance to cracking in lead-doped water.
The superior resistance of thermally treated alloy 600 was also
demonstrated by De and Ghosal (15.33) in tests using U-bends stressed to
yield. Tests were made in recirculating autoclaves at 600'F (315.6°C), in
deaerated or aerated water (8-ppm 02) having 0.6-ppm dissolved lead from
PbO. Specimens that were thermally treated at 1,110°F (598.9°C) for 24 or
Lead Cracking of Alloy 600 15-1 1
100 hours experienced no cracking in 1,600 hours, but the mill annealed
specimens without the thermal treatment did.
Alloy 600 was resistant to cracking in tests conducted by Westinghouse
at lower temperatures of 212-230'F (100-110°C) (i5L 15.34). Wire and
U-bend specimens stressed between 40 to 48 ksi in a boiling solution of
dilute PbNO 3 did not show any evidence of damage in 650 hours (for wire)
or 1,800 hours (for U-bend). Similarly, some rod specimens stressed between
40 to 48 ksi failed to crack when tested in steam at 212'F (100'C), and in
intimate contact with different alkaline/lead environments, namely:
(a) 20 wt% Na 2 SO 4 + 20% Na 2CO 3 + 0.05% PbCrO 4 for 194 hours,
(b) 20% Na 2 SO 4 + 20% Na 2CO 3 + 2% Na 2 HPO 4 + 0.05% PbCrO 4 for
650 hours, and
(c) 20% Na 2SO 4 + 20% Na 2CO 3 + 2% Na 2 HPO4 + 0.1% PbCrO4 for
170 hours.
It is likely that the exposure times were not long enough in these tests
to initiate cracking, since 2,000 hours or more is required for the cracking to
occur at 600'F (315.6°C), as described earlier. Very shallow IGPs were
produced in U-bend specimens of mill annealed alloy 600 when exposed
for 3,300 hours at 630'F (332.2°C) in a solution of PbSO 4 (15.31). The
concentration of lead in the solution was estimated to be approximately
29 ppm. In the same test solution, extensive IGSCC cracking occurred in a
laboratory heat of relatively pure (C = 30 ppm) alloy 600, and TGSCC cracks
leading to IGSCC cracking developed in alloy X750 and alloy 601. Again,
the test results demonstrate that trace amounts of dissolved lead can produce
cracking in high nickel .alloys, but the mode of cracking and the rate are
dependent upon the alloy composition and the heat treatment.
Klisiewicz et al. (15.35) have tested tubes of alloys 600, 690, and 800
under heat transfer conditions in a model boiler that simulated recirculating
steam generators. The boiler operated for 530 days with all-volatile-treated
water and was faulted with seawater. Artificial copper oxide sludge
containing 0.018% lead impurity was placed on the tubesheet in the boiler.
Cracking occurred on all alloy 600 tubes (8/8) at carbon steel tube/tube
support plate intersections. In most cases, the mode of cracking was IGSCC
but one alloy 600 tube also exhibited mixed mode cracking. Extensive
cracking networks with crack depths of up to 36 percent of wall thickness
were reported. The tubing was in a mill annealed condition, but not
sensitized, as determined from the results of a modified Huey test.
On the crack faces of alloy 600 tubes, lead, copper, and sulfur were
detected using EDX. Therefore, lead and sulfur (probably acidic sulfate)
were considered as agents contributing to the cracking of tubes in this model
boiler. In several of the previous studies cited above, investigators detected
lead on the specimens near the crack sites and/or on the fracture faces.
Therefore, it is reasonable to expect some lead on the fracture faces or near
the crack site, if lead is an active agent in producing the cracking in the
)laboratory or in the field.
No cracking occurred in the alloy 800 tubes (0/2). One alloy 690 mill
annealed tube (1/6) experienced a shallow IGP, about 8 grains or 6 percent
of wall thickness. This single crack initiated in an area that was damaged
by a grinding wheel. Therefore, the stress state in that location was not
normal.
The presence of lead in the test environment does not always lead to
cracking of alloy 600, as demonstrated by the Westinghouse tests in the
phosphate-treated boiler water (15.31). Capsules made from alloy 600
tubing, having machined flats on their outer surfaces, were tested at 600'F
(315.6°C) in boiler water treated with 10 ppm P0 4 with a Na/PO4 ratio of
2.5, and with lead and other contaminants, as follows:
(a) 0.1 g/l PbO,
(b) 0.1 g/l PbO + 0.17 g/l NaCl,
(c) 0.5 g/l PbO + 0.5.g/l NaCl + 0.5 g/l CuO + 0.5 g/l NaAsO 2 ,
(d) 1.0 g/l PbO,
(e) 1.0 g/l PbO and coupled to carbon steel, and
(f) 10 g/l PbO.
The test duration was approximately 3,500 to 4,000 hours. Solutions 1,
2, and 3 did not produce any cracking, whereas the higher levels of lead
promoted cracking in Solutions 4,5, and 6. Some C-ring specimens of mill
annealed and sensitized alloy 600 tubing experienced no cracking in 8,900
hours when tested at 90 percent of yield stress in a static autoclave containing
PbO and 10 mg/ 1 PG 4 treated boiler water at 630°F (332.2°C) (15.36). The
above results indicate that the phosphate ions may have some inhibiting
effects on the lead cracking of alloy 600.
Lead-containing caustic solutions have been shown to be no more
damaging to alloy 600 tubing than pure caustic solutions in one series of
laboratory tests (15.37). Alloy 600 capsules experienced no cracking in
approximately 3,600 hours at 600'F (315.6°C) in a 4% NaOH solution that
contained a mixture of PbO, NaCl, NaAsG 2 and CuO, with each chemical
present at a concentration of 0.5 g/l. However, the capsule cracked in the
4% NaOH solution that did not contain the mixture. Stressed C-ring
specimens of alloy 600 experienced no cracking in 2,200 hours at 620'F
(326.7 0C) when exposed to very concentrated caustic solutions, with or
without the presence of 0.5 g/l PbO in the solutions (15.36). The C-rings
were stressed slightly above yield stress, and were from several heats of
alloy 600 tubing. The caustic solution concentration was 50 percent and
consisted of an equimolar mixture of NaOH and KOH. The C-ring

specimens, however, cracked with or without PbO present at a prolonged
exposure of 4,400 hours.
Some C-rings of alloy 690 also exhibited cracking in the above tests at
2,200 and 4,400 hours when PbO was present, but no cracking occurred
without the PbO contaminant.
LEAD CHEMISTRY
Metallic lead is stable at 77°F (25°C) in neutral and alkaline
environments, in the absence of oxidizing agents, according to its potential-
pH equilibrium diagram (i.e., Pourbaix diagram) as shown in Figure 15-2.
In acidic environments, lead is thermodynamically unstable and
decomposes water with the evolution of hydrogen. However, the
dissolution of lead is hindered in the acidic environment by the poor kinetics
of the hydrogen evolution reaction on the lead surface, and by the nearly
insoluble lead salt film (e.g., PbSO 4 ) that forms over the surface.
FIGURE 15-2
Potential pH Diagram for Lead at 770 F (25 0C), in the Absence of
Substances Forming Insoluble Salts (After Pourbaix, 15.38)
-2 0 2 4 6 8 101 2 14 16
E(V) Pa,6.o I 1
1,6 Passivation
1,2~-~WPbO2 -1,2
0,8 -0,8
0,4 Pb 0,4
Corrosion
- 0,4- •0,4 H-0 -.
-0,8Immunity• o,
-1,2Pb .4,2
-1,6 LPb:: : - -0
~Corrosion
_-4
-7
-2 0 2 4 6 8 10 12 14 pH
The stable ionic specie of lead is Pb++ in acidic solutions under

moderately oxidizing or reducing conditions. The stable species of lead
are PbO and HPbO 2 -under moderately oxidizing conditions in neutral and
alkaline solutions, respectively (15.38). The environment inside a steam
generator is usually non-oxidizing or reducing because of the use of oxygen
scavengers (e.g., N 2H 4) in the feed water. Therefore, the lead species that
are expected inside a steam generator are lead salts, and Pb++, Pb°, PbO or
HPbO 2 -, depending upon the pH in the localized area of interest. Lead
species PbO 2 , Pb+4 , which have oxidation states greater than 2, exist only
under extremely high oxidizing conditions. Consequently, these species
are not considered relevant to the steam generator environment. The
following lead salts are likely to be found in a steam generator: lead acetate,
carbonate, basic carbonate, chloride, fluoride, monoxide, phosphate, sulfate,
and sulfide. These salts have vastly different solubilities in water as shown
in Table 15-3.
TABLE 15-3
Solubilities of Some Lead Salts in Water at Different Temperatures
(15.39, 15.40)
Solubility, Weight %
Salt 25C 40C 50C 1000 3510
PbNO 3 36.9 41.9 45.0 57.1 0(1)
Pb(CH 3 COO) 2 35.6 53.9 67.6 - -
PbCI2 1.08 1.46 1.75 3.10 .81.2
PbF 2 6.6x10- 2 - - - -
PbO 6.8x1 0-3 ...

PbSO 4 4.5x1 0-3 5.6x1 0-3 - - -
PbCO 3 1.1x10- 3 -...

2PbCO 3 Pb(OH) 2 1.6x10- 4 -...
PbS 8.6xl 0-5 -..
Pb 3 (PO 4 )2 1.4x10- 5 -...
1Data not available.
The acetate and chloride are very soluble; the fluoride, monoxide,
sulfate, and carbonate are sparingly soluble; whereas the basic carbonate,
sulfide, and phosphate are practically insoluble.
Lead oxide (PbO) is an amphoteric oxide; it will dissolve in acidic,
neutral, and alkaline solutions according to Pourbaix. The solubility of
PbO, however, is at a minimum (0.222 mg/1) in 77 0F (25°C) water at pH 9.3.

Mann and Deutscher (15.41) have calculated the total concentration of lead
ions at 77°F (25QC) in single and mixed solutions of C1- and SO 4 - -, over a
pH range 2 to 12. The solutions, in addition, were assumed to be in contact
with 0.3% CO 2 from the air. The data indicated that the total concentration
of Pb++ (i.e., solubility of lead compounds) in the pH range 8 to 10 was
very low (<0.6 mg/1), and it was not appreciably affected by the pH or the
presence of Cl- and S04- - ions. The concentration of Pb++ increased
appreciably outside this pH range, and was affected significantly by the
presence of anions. Lead salts form complexes in solutions that contain
high concentrations of certain anions, e.g., Cl-. Stability constants for many
complexes are given in References 15.42 and 15.43.
Only limited solubility data exist for lead salts at higher temperatures,
particularly above 212'F (100'C). As shown in Table 15-2, the data available
for some very soluble salts indicate that their solubilities increase with a
rise in temperature. The sparingly soluble salt PbSO 4 also shows an increase
in solubility with temperature. Possibly, a similar trend is followed by the
other sparingly soluble salts and the nearly insoluble salts. However, the
solubilities are expected to remain very small, even at 572°F (300°C).
The solubility data suggest that lead can be immobilized as solids in a
steam generator by maintaining the feed water pH within the range of 8.5
and 9.5. The plants which have used sodium phosphate in the past for
water treatment are likely to have lead present in their steam generators
primarily as Pb 3 (PO 4 )2 , which has negligible solubility. Ingress of Cl-,
followed by its buildup in restricted crevices, can solubilize lead there, and
can produce relatively concentrated solutions of PbC12 , which could be
damaging to the steam generator tubes. The lead inventory of a steam
generator can be kept to a minimum by blowdown and periodic removal
of the sludge.
DISCRIMINATION BETWEEN CRACKING BY LEAD

AND OTHER CORRODENTS
The criteria that possibly can be used to discriminate between lead
cracking and other types of cracking of alloy 600 steam generator tubes are
(a) presence of lead, (b) cracking mode, (c) sensitization conditions, and
(d) presence or absence of stress.
The investigators always detected lead at the crack sites and on the
fracture faces of the specimens that were analyzed after failure with
lead-doped water. Therefore, the presence of lead on the surface near the
cracks and/or on the fracture faces strongly indicates its involvement in
the cracking process(es).
However, the role of lead would be only secondary in the cracking

process if some other corrosive species, e.g., caustic or acid-sulfate mixtures,
also were detected in the tubes. Wilson et al. (15.37) have shown that lead
essentially remains passive in the case of caustic cracking, i.e., the process
is neither accelerated nor inhibited. An acid-sulfate or PbC12 environment
would produce cracking individually; however, their combined effect on
the cracking process is not known.
Lumsden (15.44) has shown by Auger spectroscopy analysis that, in
relation to the bulk composition of alloy 600, fracture faces are enriched
with nickel in the case of caustic cracking, and with chromium in the case
of acid chloride/sulfate cracking. Whether or not any kind of enrichment
occurs with lead cracking is not known at this time.
The mode of cracking in the laboratory specimens of alloy 600 in the
presence of lead was largely intergranular in the annealed material, and
predominantly transgranular in the sensitized or the thermally treated
material. Therefore, steam generator tubes in the presence of lead could
exhibit either intergranular, transgranular, or mixed mode cracking; the
distribution of intergranular and transgranular components would be
dependent upon the degree of sensitization in the tube material. There is
no known practical environment, besides that containing lead, which
produces transgranular cracking in alloy 600. Therefore, if any transgranular
component is detected in a steam generator tube, it can be assumed that
lead was a contributing factor in the cracking process, provided some lead
is detected on, the tubes.
Cracking of alloy 600 specimens in the presence of lead contamination
has required stresses close to, or higher than, the yield stress. In the absence
of such high stresses, the cracking by lead appears improbable in steam
generator tubes. Those tubes suspected of lead cracking should also be
examined for the presence of denting, kinks, ovality, and loading from any
external sources. However, alloy 600 low-temperature-annealed tubing
typically has high manufacturing residual stresses which may provide
stresses for SCC initiation.
POSSIBLE REMEDIES FOR LEAD CRACKING

Factors which influence lead cracking of alloy 600 are stress level, degree
of sensitization, temperature, and the nature of lead contamination.
The steam generator tubes that are likely to be affected by lead cracking
are those in which high stresses are present, e.g., due to improper roller
expansion in the tubesheet, denting at tube support plate intersections, or
from buckling of the support plates. To mitigate lead cracking in the
expansion joints in the tubesheet region, stresses possibly can be lowered

by using a localized, stress relieving heat treatment. There appears to be
no practical way of lowering stresses in the other parts of a tube bundle.
Mill annealed tubes exhibiting only nominal sensitization are the most
vulnerable to lead cracking, whereas sensitized or thermally treated tubes
are considered less vulnerable. Therefore, the thermally treated tubes are
preferable for new steam generators. The latter also exhibits superior
resistance to cracking in several other environments.
Alloy 600 exhibited no cracking in tests at 212'F (100°C), even in
concentrated solutions of lead salts. At elevated temperatures (550-630'F
[287.8-332.2 0C]), the cracking occurred in the lead-containing environments,
and the time to failure generally decreased with the increase in temperature.
Therefore, lead cracking might be a thermally activated process, as judged
from the limited laboratory data. Possibly, the life of tubes in an afflicted
steam generator can be prolonged at the cost of reduced power, by operating
at temperatures lower than normal.
' Cracking has been produced in alloy 600 in aqueous environments in
the presence of metallic lead, PbO, Pb in hydrocarbons, PbC12 , PbSO 4, and
Pb 3 (PO 4 ) 2 . Experimental results indicate that a certain minimum
concentration of lead is required in dissolved form to produce cracking. In
one instance, approximately 39 ppm Pb in solution, dissolved as PbSO 4,
wis sufficient to produce cracking. However, the critical, minimum
concentration of dissolved lead to produce cracking is not known.
Cracking could possibly be avoided if lead is kept in its least soluble
form. The lowest solubility of lead salts at 77°F (25'C) was shown to occur
in the pH range 8 to 10. Therefore, if the feedwater is maintained at pH 8.5
to 9.5, it would assure, by extension of the 77°F (25°C) data, the lowest
concentration of dissolved lead and the minimum probability for cracking
to occur in tubes.
Cracking in alloy 600 was found to be the least severe when phosphates
were used for the water treatment, or when Pb3 (PO 4) 2 was used as a
contaminant in the tests. This beneficial effect likely resulted from the tying
up of the lead ions by the phosphate ions to form nearly insoluble lead
phosphate (0.14 mg/1 at 77°F [25°C]). The steam generators with lead
cracking might benefit by occasionally treating the feed water with sodium
phosphate. However, considerably more work is required to qualify the
use of sodium phosphate for this application.
Ingress of chloride in the steam generators should be minimized to
prevent denting as well as cracking; the latter could be avoided by,
preventing dissolution of lead as PbC12.
The lead inventory of steam generators can be reduced by practicing
periodic or continuous blowdown and sludge removal. The transport of
new lead into the steam generators via ancillary plant equipment should
be minimized by removing lead-containing components that can be leached,
abraded, or otherwise cause contamination.
REFERENCES
15.1 Vaccaro, F. P., and B. P. Miglin. "Lead-Enhanced IGA and SCC of
Steam Generator Tubing." Paper presented at the EPRI Meeting
on Intergranular Corrosion Mechanisms, Alexandria, VA, April
1987.
15.2 Sawochka, S. G., and M. E. Clouse. "Investigation of Lead Transport
in PWRs. " EPRI Progress Report. July 1987.
15.3 Agrawal, A. K., W. N. Stiegelmeyer, and W. E. Berry. "Evaluation
of Alloy 600 Tube A77-34 from Steam Generator A of Arkansas
Nuclear One, Unit 1." EPRI NP-4504-LD. March 1986.
15.4 Begley, R. T., et al. "Interim Report: Beznau-l." Westinghouse,
Pittsburgh, PA. RP 72-704-TAPSC-R2. On file at EPRI.
15.5 Hall, J. "Destructive Examination Results Tubes R53L57 and R52L94
Steam Generator B-Calvert Cliffs Unit 1." Interim Report by
Combustion Engineering to Baltimore Gas & Electric Co., May 1987.
On file at EPRI.
15.6 Carr, W. "Farley Nuclear Plant Unit 2 Tube Support Plate
Cracking." Paper presented at the EPRI Meeting on Intergranular
Corrosion Mechanisms, Alexandria, VA, April 1987.
15.7 "Ohi-1/Genkai 1 & 2." Paper presented at the EPRI Meeting on
Intergranular Corrosion Mechanisms, Alexandria, VA, April 1987.
15.8 Agrawal, A. K., W. N. Stiegelmeyer, and W. E. Berry. "Evaluation
of Steam Generator Tubes R7C45 and R12C46 From the Ginna
Nuclear Power Plant." EPRI NP-3070-LD. May 1983.
15.9 "Steam Generator Sludge Pile Model Boiler Testing: Sludge
Characterization." EPRI NP-2041. September 1981.
15.10 Inman, S. C. "Examination of Steam Generator Tube Sections From
the Millstone Unit 2 Nuclear Power Plant." Progress Report. On
file at EPRI, 1986.
15.11 Ogren, J. "Examination of Pulled Tubes R3C41 and R9C58 From
SGs of North Anna Unit 1." Paper presented at the EPRI Meeting
on Intergranular Corrosion Mechanisms, Alexandria, VA, April
1987.
15.12 Schenk, H. J. "Investigation of Tube Failures in Inconel 600 Steam
Generator Tubing at KWU Obrigheim." Materials Performance 15,
no. 3 (1976): 25.
15.13 Takamatsu, H. "Status of IGA in Japanese Plants and Results of S/

G Pulled Tube Examinations From Japanese Plants." Paper
presented at the EPRI Workshop on Remedial Actions for
Secondary-Side Intergranular Corrosion, Baltimore, Maryland,
October 1985.
15.14 Agrawal, A. K., W. N. Stiegelmeyer, and W. E. Berry "Evaluation
of Alloy 600 Tubes From Steam Generators A and B of the Palisades
Nuclear Power Plant." EPRI NP-3964-LD. March 1985.
15.15 Pement, E W., P. J. Kuchirka, and C. R. Wolfe. "Examination of
Three Steam Generator Tubes From the Point Beach Unit I Nuclear
Power Plant." EPRI NP-2958-LD. March 1983.
15.16 "Steam Generator Sludge Analyses." EPRI NP-516. April 1979.
15.17 Finnigan, D. J., D. F. Libaert, E. Metcalfe, and I. S. Woolsey.
."Chemical and Metallographic Examination of Tube Specimens
From Ringhals 2 Steam Generator." EPRI NP-2972-LD. March 1983.
15.18 "Failure Analysis of Steam Generator Tubes From St. Lucie Unit
1." Babcock & Wilcox Report to Florida Power and Light Company.
Lynchburg, VA, November 1986. RDD:86:5261-01:01. On file at
EPRI.
15.19 "An Evaluation of Steam Generator Tube Failure in St. Lucie-1
Nuclear Power Plant." Battelle Columbus Division Report to Florida
Power and Light Company, Columbus, OH, July 1987.
15.20 "Point Beach-1 Steam Generator: Tubesheet Crevice Chemistry."
EPRI NP-4782. March 1987.
15.21 Agrawal, A. K., and W. N. Stiegelmeyer. "Evaluation of Steam
Generator Hot Leg Tubes R18C43 and R18C53 From the Cabrera
Nuclear Power Plant." EPRI NP-4759-LD. August 1986.
15.22 Patel, B. M., and J. K. Hayes. "Investigation of Steam Generator
Corrosion Products Under Typical PWR Operating Conditions."
EPRI NP-3068. May 1984.
15.23 Copson, H. R., and S. W. Dean. "Effect of Contaminants on
Resistance to Stress Corrosion Cracking of Ni-Cr Alloy 600 in
Pressurized Water." Corrosion21 (1965): 1.
15.24 INCO IRDC Project Report 192.20R. December 12, 1968. On file at
EPRI.
15.25 INCO IRDC Project Report 358.3. October 1, 1969. On file at EPRI.
15.26 INCO IRDC Project Report 358.11. January 31, 1972. On file at
EPRI.
15.27 INCO IRDC Project Report 192.27R, January 28, 1972. On file at
EPRI.
15.28 INCO IRDC Project Report 358.22, March 13, 1975. On file at EPRI.
15.29 Babcock & Wilcox Report LR:72:2341-09:1, February 29, 1972. On
file at EPRI.
15.30 Babcock & Wilcox Report LR:72:2341-77:2, December 7, 1972. On

file at EPRI.
15.31 "Synopsis Relative to Stress Corrosion Cracking of Inconel 600 in
Aqueous Environments Containing Lead." Report by
Westinghouse. April 6, 1979. On file at EPRI, Palo Alto, CA.
15.32 Airey, G. P. "The Effect of Carbon Content and Thermal Treatment
on the SCC Behavior of Inconel Alloy 600 Steam Generator Tubing."
Corrosion35, (1979): 129.
15.33 De, P. K., and S. K. Ghosal. "A Comparative Study of Stress
Corrosion Cracking of Steam Generator Tube Materials in Water at
315 C." Corrosion37 (1981): 341.
15.34 Kramer, L. D., S. T. Michael, and F W. Pement. "Service Experience
and Stress Corrosion of Inconel 600 Bellow Expansion Joints in
Turbine Steam Environments." Materials Performance 14, no. 9
(1975): 15.
15.35 Klisiewicz, J. W., B. M. Patel, and E. E. Grondahl. "Intergranular
Attack of Nuclear Steam Generator Tubing in the Presence of
Seawater Inleakage." In Proceedings, IX International Congress on
Metallic Corrosion, Toronto, Canada, 1984. See also EPRI NP-3274.
15.36 Pement, F W., I. L. W. Wilson, and R. G. Aspden. "Stress Corrosion
Cracking Studies of High Nickel Austenitic Alloys in Several High
Temperature Aqueous Solutions." Materials Performance 19, no. 4
(1980): 43.
15.37 Wilson, I. L. W., et al. "Caustic Stress-Corrosion Behavior of Fe-Ni-
Cr Nuclear Steam Generator Tubing Alloys." Nuclear Technology
31, no. 10 (1976): 70.
15.38 Pourbaix, M. "Atlas of Electrochemical Equilibria in Aqueous
Solutions." NACE, Houston, TX, 1974.
15.39 Seidell, A. Solubilities of Inorganic and Organic Compounds, Second
Edition, New York: Van Nostrand Co.
15.40 Stephen, H., and T. Stephen. Solubilities of Inorganic and Organic
Compounds, Vol. 1. New York: Pergamon Press, 1979.
15.41 Mann, A. W. and R. L. Deutscher. "Solution Geochemistry of Lead
and Zinc in Water Containing Carbonate, Sulfate, and Chloride
Ions." Chemical Geology 29 (1980): 293.
15.42 Baes, C. E, and R. E. Mesmer. The Hydrolysis of Cations. New York:
John Wiley, 1976.
15.43 Smith, R. M., and A. E. Martell. CriticalStability Constants, Vol. 4.
New York: Plenum Press, 1976.
15.44 Lumsden, J. "Possible Role of Dealloying on Initiation of IGC."
Paper presented at the EPRI Meeting on Intergranular Corrosion
Mechanisms, Alexandria, VA, April 1987.
16
TUBE CORROSION AND WEAR:

U-BEND SUPPORTS (AVBs) AND
TSP INTERSECTIONS
A. P L. Turner,Dominion Engineering,Inc.
1993
INTRODUCTION
Many steam generators have experienced tube wall loss at the
intersections between the heat transfer tubes and tube support structures
(16.1). Tube wall loss has occurred at the horizontal perforated tube support
plates (TSPs) on the vertical legs of the tubes, and at support structures of
various designs that are used to support the tubes and limit vibration in
the return bend regions of recirculating steam generators. Westinghouse-
design steam generators use a single 180 degree U-bend between the hot
and cold legs of the tubes. Combustion Engineering (C-E) generators use
two 90 degree bends connected by a section of horizontal tube between the
hot and cold legs. Vibrations in the Westinghouse-design U-bends are
restrained by antivibration bars (AVBs) installed between the tubes
(Figure 16-1). The bends and horizontal tubes in C-E steam generators are
supported by a number of vertical and diagonal strips between the tubes.
These are commonly referred to as the batwing supports (Figure 16-2). The
bend region supports in C-E steam generators also include horizontal strips
underneath the tube bend. These horizontal strips connect to the vertical
strips to form a two-dimensional grid. Damage at support structures is
caused by corrosion, wastage, mechanical wear, or a synergistic interaction
of the two processes. Most instances of corrosion/wear wall loss at support
structures have occurred in recirculating steam generators, but a few cases
of wear at TSPs have occurred in once-through steam generators.
Wear at U-bend AVBs and batwing supports has occurred in steam
generators of both Westinghouse and Combustion Engineering design. AVB
wear was first observed in San Onofre 1 in 1973 and has since been
experienced in many other steam generators of various designs. Currently,
more than 90 units of Westinghouse design worldwide are affected to some
extent byAVB wear (Table 16-1). Wear scars form slowly on the OD surfaces
of the tubes where they contact the AVB component. The scars grow slowly
in depth but eventually penetrate sufficiently far through the tube wall to
require that the tube be plugged, unless remedial actions are taken. Most
plants affected by AVB wear have plugged relatively few tubes because of
this degradation mechanism. Replacement of the AVBs is an effective
remedial measure that prevents continued growth of the original wear scars,
because the replacement AVBs are placed adjacent to the original wear
locations. Batwing wear was first observed at San Onofre 2 in 1984 and has
since been experienced by at least nine other C-E units (Table 16-2). Batwing
wear appears to affect only tubes near the central open cylinder (stay
cylinder) of C-E steam generators. The remedial actions taken in response
to batwing wear have primarily been to plug affected tubes or to
preventively plug tubes in the susceptible region of the tube bundle.
Tube Corrosion and Wear: U-Bend Supports and TSP Intersections 16-3
FIGURE 16-1
Typical AVB Arrangements in Westinghouse-Design
Steam Generators
Tubes
11 16 46
Model 51 Antivibration Bar Arrangement
AVBs
Tubes
Model F Antivibration Bar Arrangement
Tube wall loss from corrosion and or wear has occurred at tube support
plates (TSPs) in both recirculating and once-through steam generators. This
type of damage appears to have been generated by a number of different
mechanisms. The exact nature of the corrosion/wear damage that can occur
in a given steam generator appears to be dependent on the specific design
features of the generator. Corrosion/wear at TSPs ranges from
predominantly mechanical wear in high cross flow velocity regions near
the feedwater inlet of steam generators with preheater sections (16.2) to
what appears to be vibration-enhanced corrosion wastage at cold leg
supports (cold leg thinning) in some steam generator models (16.3). Fretting
wear has been experienced by a few tubes in Babcock & Wilcox (B&W)
once-through steam generators.
Corrosion/wear mechanisms have been studied by examining sample
tubes removed from affected steam generators. In the laboratory, researchers
have attempted to simulate corrosive environments that can exist at TSP
intersection. In the field, the motion caused by tube vibration has been
measured. Tube damage due to corrosion/wear at support structures can
usually be detected and monitored by eddy current inspection (ECT).
Remedial measures that have been applied to prevent corrosion/wear
damage include modifications of flow patterns, replacement of tube support
structures, and modifications of water chemistry to prevent formation of
corrosive environments.
FIGURE 16-2
Antivibration Support Structure in Combustion Engineering
Steam Generators
Vertical
Strip
Supports
Horizontala
Strip
Supports
Diagonal
Strip
Supports
TABLE 16-1
Plants With Tube Wear at AVB Positions-Westinghouse Design
Almaraz 1 Doel 4 Salem 2

Almaraz 2 Farley 1 San Onofre 1
Angra Farley 2 Sena (Chooz A)
Asco 1 Fessenheim 1 Sendai 1
Beaver Valley 1 Fessenheim 2 Sequoyah 1
Beznau 1 Gravelines 1 Sequoyah 2
Beznau 2 Gravelines 2 Shearon Harris
Blayais 1 Gravelines 3 South Texas 1
Blayais 2 Gravelines 4 South Texas 2
Blayais 3 Ikata 1 St. Laurent B2
Braidwood 1 Jos6 Cabrera (Zorita) Surrey 1(2)
Braidwood 2 Kewaunee Surrey 2(2)
Bugey 2 Kori 2 Takahama 3
Bugey 3 Kori 3 Takahama 4
Bugey 4 Kori 4 Tihange 1
Bugey 5 McGuire 1 Tihange 2
Byron 2 McGuire 2 Tihange 3
Callaway Mihama 2 Tsuruga 2
Catawba 1 Mihama 3 Tricastin 1
Catawba 2 Millstone 3 Tricastin 4
Connecticut Yankee North Anna 1 Turkey Point 3(2)
D. C. Cook 1 North Anna 2 Turkey Point 4(2)
D. C. Cook 2(1) Ohi 2 V.C. Summer
Dampierre 1(1) Paluel 4 Vandellos II
Dampierre 2 Point Beach 1(2) Vogtle 1
Dampierre 3 Prairie Island 1 Wolf Creek
Dampierre 4 Prairie Island 2 Yankee Rowe
Diablo Canyon 1 Ringhals 2(1) Youngkwang 1
Diablo Canyon 2 Ringhals 3 Youngkwang 2
Doel 1 Ringhals 4 Zion 1
Doel 2 Salem 1 Zion 2
Doel 3
1
Original
2
Replacement
TABLE 16-2
Plants With Tube Wear at Batwing Positions
-Combustion Engineering
Arkansas Nuclear One 2 Palo Verde 3

Calvert Cliffs 1 San Onofre 2
Calvert Cliffs 2 San Onofre 3
Palo Verde 1 St. Lucie 2
Palo Verde 2 Waterford 3
AVB WEAR
Plant Experience
Wear at AVBs was first observed at San Onofre Unit 1 (16.4). This
430 MW 3-loop Westinghouse PWR entered commercial service in 1968.
Tube leaks from AVB wear occurred as early as 1972. By 1976, AVB wear
damage had affected several hundred tubes in the San Onofre Unit I steam
generators. Damage was observed primarily in tubes near the outside of
the bundle (large radius U-bends). Similar damage has also occurred at
Connecticut Yankee and Zorita, which have AVBs of the same design as
San Onofre Unit 1 (16.5).
The typical arrangement of AVBs in steam generators of Westinghouse
design is shown in Figure 16-1. The AVBs in the San Onofre Unit 1,
Connecticut Yankee, and Zorita steam generators were made from round
cross section, carbon steel bars. This design is particularly susceptible to
wear of both tubes and AVBs because there is, initially, point contact between
the tubes and the AVBs such that contact forces are concentrated. The low
corrosion resistance of the carbon steel material also makes it susceptible to
fretting corrosion, because rubbing between the tube and the AVB removes
the corrosion product scales from the AVB so that corrosion can continue
on a clean metal surface. However, these processes may be somewhat
ameliorated by the inherent lubricity of magnetite in high temperature water,
as compared to the more abrasive character of chromium-containing spinel
films. Wear of the tube and the AVB increases the clearance between them,
leaving the tubes essentially unsupported by the AVBs. This allows vibration
amplitudes to increase, which, in turn, accelerates the rate of wear.
The extent of wear at Connecticut Yankee and Zorita has required

plugging of only a small number of tubes. There has been no need to replace
AVBs nor to install additionalAVBs. Beznau Units 1 and 2 steam generators
are also equipped with AVBs of round cross section, but they are made
from alloy 600 with a chromium-rich, wear-resistant coating. The Beznau
units have experienced AVB wear of a significant number of tubes, and the
AVBs in both plants have been replaced with new AVBs with a rectangular
cross section.
Newer Westinghouse-design steam generators have been equipped
with AVBs fabricated from rectangular cross section alloy 600 bars that are
coated with a wear-resistant, chromium-rich material (chromized coating).
This provides a better contact geometry that spreads the contact forces over
a larger area and reduces the contact stresses. This design shows improved
resistance to wear at the tube AVB intersections but does not completely
prevent the wear (16.5). As shown in Table 16-1, more than 90 plants with
Westinghouse-design steam generators have experienced AVB wear to some
extent. In fact, wear of this type has been observed in all Westinghouse
steam generator models. Usually, relatively few tubes are affected and the
wear scars grow only slowly in depth so that most affected tubes have not
required plugging. Excluding the three plants that originally had carbon
steel AVBs with a round cross section, the maximum number of tubes
affected in any plant is less than 2% of the entire tube population. The
maximum number of tubes plugged in a unit is 1% On the average, less
than 0.5% of the tubes at a unit are affected by AVB wear and less than 0.2%
of the tubes have required plugging for this cause (16.1). Some of the tubes
in a steam generator will have clearances between the U-bend and the AVBs
that are near the maximum allowed by the fabrication tolerances. These
tubes may not be sufficiently well supported to prevent flow-induced forces
from causing them to vibrate. These are the tubes that are susceptible to
AVB wear. AVB designs that use flat, chromium-plated alloy 600 bars reduce
wear rates of the tubes and the AVBs sufficiently to extend the time needed
for the wear scar to grow to a depth that requires the tube to be plugged.
However, this type of AVB bar cannot completely prevent wear from
occurring.
Steam generators made by Combustion Engineering have also
experienced significant wear in the upper parts of the tube bundle at support
structures. ECT inspections of San Onofre Unit 2 in December 1984
identified wear indications at the intersections of tubes with the vertical
and diagonal supports in the return bend region (16.2). In steam generators
of C-E design, there is an untubed cylinder in the center of the generator
(the central stay cylinder region) above a stay cylinder in the channel head;
this is used to stiffen the tubesheet. The worn tubes in the San Onofre
steam generators were all near the inner periphery of the central stay
cylinder region of the generator. The maximum wear depth observed was
95% through wall. Flow calculations indicated that the cause of the wear
was flow-induced vibration of both the tubes and the carbon steel diagonal
supports; the supports are 0.090 inch thick by 2 inch wide. A number of
tubes with deep indications were staked and plugged in 1985 and additional
tubes have been plugged since then due to further progression of this wear.
Because the batwing supports are only 0.090 inch thick, there is concern
that wear of the supports may cause failure of the support strips, which
would create loose parts that could cause additional damage to the tubes.
Calculations performed by C-E have indicated that the batwing wear
should be confined to a relatively small region of the generator near the
central stay cylinder. In this region, the diagonal support strips have longer
than usual unsupported spans near the center line of the steam generator
because of a reduced number of tubes in this region. This makes the diagonal
strips in this region susceptible to flow-induced vibration. In addition, flow
velocities in the portion of the steam generator near the stay cylinder are
higher because of the lower flow resistance in the untubed cylinder. The
number of tubes in C-E steam generators considered to be susceptible to
wear is less than the tube plugging margin provided in the steam generator
design. Therefore, the primary remedial measure taken in response to
batwing wear has been to preventatively plug the tubes in the susceptible
region. In addition to San Onofre, other plants with C-E steam generators,
including those with newer Series 80 generators, have also experienced
wear at batwing supports. Ten plants have plugged tubes because of this
degradation mechanism. Several units have preventatively plugged several
hundred tubes.
Examination of Sample Tubes

Inspections at San Onofre Unit I in 1975 and 1976 showed rapid
increases in the rate of wear and the number of tubes affected. Two AVBs
and three tube sections were removed from the steam generators for
examination to confirm the nature of the damage detected by ECT.
Examinations of the AVBs showed wear scars at the points of contact with
tubes. The depths of the scars were greatest near the tube bundle periphery.
The metal surfaces were clean and, in some cases, bright with only minor
evidence of corrosion. Hardness measurements on cross sections through
the wear scars showed evidence of strain hardening near the damaged
surfaces. Examination of the most severely damaged tube section showed
no signs of corrosion or other damage except for the wear scars at the AVB
intersection points. Hardness measurements also showed evidence of strain
hardening adjacent to the damaged surfaces (16.4). These results confirm
that the mechanism of material removal is predominantly mechanical wear

from rubbing between the tubes and the AVBs. There is no evidence that
corrosion made a significant contribution to the wear rates of tubes at AVB
intersections in the San Onofre Unit I steam generators.
One U-bend from Zion Unit 1 with AVB wear damage has been
destructively examined (Figure 16-3). Zion Unit I is a Westinghouse plant
with Model 51 steam generators (16.6). The AVBs are chromized, rectangular
cross section bars of the type used in later Westinghouse-design steam
generators. Both the tube and sections of the neighboring AVBs were
removed for examination. The primary purpose of the examination was to
determine whether the AVB wear, occurring in the Zion steam generators,
was the result of mechanical wear or whether it was due to a corrosion-
assisted mechanism. Examinations included profilometry, metallography,
and analysis of deposits on the tube both within AVB intersection regions
and in free span regions. When comparing the deposits in the AVB
intersection regions with those in the free span, the analyses did not detect
any differences in the compositions, which is an indication that chemical
concentration had not occurred at the intersections. Surfaces in wear scar
regions showed a small amount of work hardening that is consistent with
what is expected from a mechanical wear process. Striation marks in the
wear scars indicate that the direction of the relative motion between the
tube and the AVB is perpendicular to the tube axis. There is no evidence of
significant deformation of either AVBs or the tubes in the worn regions,
indicating that the contact forces between the two components are relatively
small. It was confirmed that the AVBs had chromium-rich, anti-wear
coatings as expected. The chromium-rich coatings showed some evidence
of porosity at high magnification which may be the result of IGA of the
coating. The porous coating could also be a site where concentration of
impurities could occur because of restricted coolant flow in the porosity.
On the basis of the results of the sample tube and AVBs, it was concluded
that the formation of the wear scars was predominantly a mechanical wear
process.
A sample tube removed from the return bend region of steam generator
11 of Calvert Cliffs Unit 1 in 1983 had a wear scar up to 32% through wall at
the intersection with the central vertical support (Figure 16-4) (T16,7 16.8).
Calvert Cliffs is a C-E plant with Series 67 steam generators. The nature of
the damage on the sample tube indicated that the mechanism of material
removal was mechanical wear.
FIGURE 16-3
Various Views of Central Area of Section 3 Showing AVB
Intersections and Deposits
.70*
1806
900
00 0
FIGURE 16-4
Batwing Wear on Tube Row 99 Line 143 From Calvert Cliffs Unit 1
Modeling of Flow-Induced Vibrations

Finite element models were used to evaluate the effects of cross flow
over tubes with different radii U-bends and support conditions to determine
whether vibrations caused by fluid-elastic instabilities are consistent with
the observed AVB wear at San Onofre Unit 1 (16.4). The results showed
that tubes in rows 32 to 48 that are not supported at the lower AVB location
will undergo unstable fluid-elastic vibrations, while those tubes that are
supported at the lower AVB locations will be stable. These results indicate
that tubes which have large clearances to the lower AVBs because of
tolerance stackups will be susceptible to fluid-elastic vibration and wear at
the AVB intersections. Tubes with small clearances are not expected to
experience wear. However, no clearance measurements have been made
of in-service steam generators to validate these models.
Repairs and Remedial Measures

To prevent continued wear of steam generator tubes at AVBs, San Onofre
installed additional AVBs adjacent to the original ones in the susceptible
portion of the tube bundle (Figure 16-5). The new AVBs were made from
alloy 600 bars with a rectangular cross section and were chromized. The
original round cross section AVBs were 0.250 inch diameter. In order to
minimize the clearances between the tubes and the AVBs, approximately
15% of the new AVBs installed are made from bars that are 0.267 inch thick.
The remaining 85% of the new AVBs were 0.250 inch thick matching the
original AVB diameter. The additional AVBs provide additional points of
support for the tube U-bends and reduce contact stresses, because there is
line contact between the tube outside surfaces and the flat AVB surfaces.
Corrosion-resistant alloy 600 material and chromium plating was selected
to improve the wear resistance of the AVBs themselves.
FIGURE 16-5
AVB Repair Performed at San Onofre Unit 1
New
11 16 46
Beznau Units I and 2 replaced the original AVBs with new supports
that have a rectangular cross section. At least seven units equipped with
Model 51 steam generators have performed AVB repairs (partial
replacement) (16.5). The structure of the replacements is similar to the
original AVBs. The new supports are installed at locations that are spaced
away from the locations of the original AVBs, which prevents wear at the
original intersections. In some cases, such as San Onofre Unit 1, larger
thickness bars are used at selected locations where the spacing between the
tubes is unusually large, in order to reduce the gaps between the AVB and
the adjacent tubes.
An offset pattern of anti-vibration stakes are used in marine boilers to

stabilize tubes against vibration (Figure 16-6). In this technique, the stakes
are slightly thicker (0.001 to 0.005 inch) than the maximum distance between
adjacent tubes determined by the nominal spacing, the nominal outside
tube diameter, the position tolerance, and the tube diameter tolerance. This
guarantees an interference fit between the stake and the tube when the
tube is at its nominal position. If these oversized stakes were all placed in
line, they would not fit. Therefore, the stake positions for alternate rows
are offset so that the interference can be accommodated by a slight bending
of the tubes, such that the tubes are forced into a slight "S" shape. In this
arrangement, the anti-vibration stakes are all initially in contact with the
adjacent tubes to prevent even low amplitude vibrations. An offset AVB
pattern based on this principle has not been used in PWR steam generators;
however, it could be an effective means for preventing AVB wear.
FIGURE 16-6
Offset AVB Arrangement Used to Avoid Loose Fitup
Tolerance Variations
OFFSET
AVBs
-- TUBES
Conclusions
The widespread occurrence of wear in the bend regions of a number of
different model steam generators from two different manufacturers
demonstrates that current anti-vibration structures are difficult to fabricate
to insure that steam generator tubes are resistant to wear. Because of fit
variations allowed by tolerances on tube and support element dimensions,
some gaps between the tubes and the supports will always exist with current
designs. Complete elimination of tube motion in the U-bend regions will
require that all tubes are tightly restrained. This could be accomplished by
using the elasticity of the tubes in bending to achieve a tightfit on all tubes
as, for example, with the offset bar pattern described above. An offset AVB
arrangement has not been tested in any operating steam generators.
Fortunately, standard ECT inspections of the tubes can readily detect wear
damage at bend region supports, and the depths of the scars can be
accurately sized. Wear scars tend to increase only slowly in depth such
that the rate of growth can be monitored over several successive inspections.
Badly worn tubes must eventually be plugged unless the wear process is
stopped by replacing the adjacent supports in new locations away from the
existing damaged regions. For plants that have an unacceptable progression
rate of AVB wear, replacement of the AVBs has been an effective remedial
measure. When tubes susceptible to wear damage are restricted to a small
portion of the tube bundle by, e.g., batwing wear in C-E steam generators,
preventive plugging may be the most economical solution to eliminate the
possibility of leaks and to reduce the required NDE inspection effort.
COLD LEG THINNING

Plant Experience
Cold leg thinning is a wall loss! mechanism that has been observed at
the lower cold leg TSP intersections in Westinghouse Models 51, 44, and 24
steam generators. Typically, it appears as a thinned region on the tube wall
on one side of the tube; the affected area corresponds to almost the full
height of the TSP, and extends over an approximately 90 degree angular
segment. Affected.tubes with damage at the two lowest support plates are
usually located within a few rows of the periphery of the tube bundle. Some
tubes positioned deeper inside the tube bundle have experienced cold leg
thinning at the third or higher tube supports. Cold leg thinning was first
detected in 1980 (16.3), and has since been found in steam generators at
approximately 20 plants (Table 16-3). At the average plant that has
experienced cold leg thinning, approximately 50 tubes are affected by this
mechanism, and approximately 25 of these have been plugged. The plant
with the most tubes plugged due to cold leg thinning has plugged 172 tubes.
At the present time, no cold leg thinning has been observed in steam
generators with preheater sections, e.g., Westinghouse Model D steam
generators.
TABLE 16-3
Plants With Cold Leg Thinning
Plant Steam Generator Model

Beaver Valley 1 51
Cook 1 51
Cook 2 (orig. SGs) 51
Diablo Canyon 1 51
Farley 1 51
Farley 2 51
Jose Cabrera (Zorita) 24
Kewaunee 51
North Anna 1 51
North Anna 2 51
Point Beach 2 44
Prairie Island 1 51
Prairie Island 2 51
Ringhals 2 51
Salem 1 51
Salem 2 51
Sequoyah 1 51
Sequoyah 2 51
Zion 1 51
Zion 2 51
This list is based on plants without denting that have defects at cold leg TSPs
concentrated at the lower TSPs and near the periphery of the tube bundle. In
only two cases has the degradation mechanism been confirmed by removal of
sample tubes.
The depth growth rate of cold leg thinning defects has been determined
by analyzing ECT data from 489 tubes in 18 steam generators. Only data
that reported 20% through-wall depth or greater for two successive
inspections were used. The majority of the measured average growth rates
were less than 1% of wall thickness per full power month of operation. The
overall average growth rate for the 489 tube sample was 0.45% of wall
thickness per full power month of operation (5.4% per EFPY).
Examination of Sample Tubes

Two tubes exhibiting cold leg thinning have been removed from steam
generators (one tube from each of two different generators) for destructive
examination (16.3). Results from both examinations were similar. An
example of a thinned region from one of the sample tubes is shown in
Figure 6-7. Each tube had a zone of shallow wall loss at the first tube support
and a zone of deeper wall loss at the second tube support. The axial extents
of the thinned regions were approximately 0.6 inches, which is slightly less
than the thickness of the tube supports. The circumferential extent of the
degraded zones were between 0.2 and 0.6 inches. The widest and deepest
thinned zones correspond to slightly more than 90 degrees of the tube
circumference. There was only a single degraded zone at each TSP
intersection. The portions of the degraded zone that showed only shallow
thinning had a series of circumferential ridges that appear to correspond to
the tool marks left on the ID surface of the drilled hole in the mating TSP.
The central portion of the damage zone is an irregularly shaped deeper
crater. The tool mark ridges appear to indicate that rubbing between the
tube and the TSP is a significant part of the damage mechanism. The surface
of the deeper central crater could not have been in contact with the TSP
hole surface, so direct rubbing of the two surfaces cannot have accounted
for the rapid material loss that formed the crater.
Surface chemical analysis techniques were applied to the sample tubes
with cold leg thinning damage to determine whether evidence of a corrosive
chemical environment could be detected in the damaged regions of the
tube surface. The surfaces of the tube in the regions of the TSP intersections
were generally free from deposits. This is interpreted as evidence that the
crevice between the tube and the TSP were not fouled or packed with
corrosion product material, which means that the tube was probably free
to move within the drilled hole of the TSP. No chemical compounds or
surface films suggesting a particular chemical species were found; however,
sulfur was discovered on the surfaces of one tube in greater amounts than
other contaminants. Additionally, sulfur had deposited at the bottom of a
pit near one of the wall-thinned areas. Only minimal sulfur was found on
the surfaces of the other tube examined. Although minimal amounts of
corrosion products were present on, or adjacent to, surfaces damaged by
cold leg thinning, a corrosive environment is still considered to be one of
the causes of this phenomenon. Because the damage is confined to the cold
leg side, the chemical environment is considered important for the cold leg
thinning process to occur. Hot legs and cold legs in Model 51 steam
generators are essentially alike in their mechanical characteristics. Therefore,
a purely mechanical wear phenomenon would be as likely to occur on the
hot leg as the cold leg. However, due to the differences in superheat in TSP
crevices on the hot leg and cold leg sides, the chemical environments are
probably different. It has been suggested that the environment responsible
for cold leg thinning is stable for the 10 to 15'F superheats in cold leg TSP
crevices, but not for the 50 to 60'F superheats in hot leg crevices. One
possible environment that could account for the observed tube wastage is a
neutral or slightly acid sulfate environment.
FIGURE 16-7
Cold Leg Thinning on Removed Sample Tube
Metallography of a tube section including the damaged zone of the

tube showed a normal alloy 600 microstructure. The surface produced by
the cold leg thinning mechanism cut across grains and grain boundaries
and provided no evidence that grain boundaries are preferentially attacked
in the process. There is no significant deformation of alloy 600 adjacent to
the surface of the damaged region, and no significant evidence of work
hardening from hardness measurements. These two results show that the
material removal is not the result of high contact stress wear.
Measurements of Tube Motions

Tubes experiencing cold leg thinning are located on the periphery of
the tube bundle where fluid cross flow velocities are large. Cross flow is at
a maximum at the top of the tubesheet where the tubes are constrained by
the tubesheet. Therefore, at higher elevations where tubes are restrained
by TSPs of various designs, the vibrations excited by the cross flow will
have higher displacement amplitudes. Analysis of possible vibration modes
of the tubes indicates that the elevations of the TSPs where cold leg thinning
is observed may correspond to the levels with relatively high tube
displacements at the TSP from flow-induced vibrations. Therefore, it is
expected that tube motion induced by flow forces contributes to the process
of material removal. In order to evaluate the importance of tube motion to
the cold leg thinning damage, several tubes in an operating steam generator
were instrumented with accelerometers during power operation (16.3). The
tests generated the following conclusions:
* The TSPs provide more support for tubes on the outer row of the tube
bundle than for tubes one row from the periphery. In the most
extensively affected steam generators, there is a weak tendency for cold
leg thinning to affect more tubes one row or more in from the bundle
periphery than in the outer row.
* For the instrumented tubes, the support plate intersections having
deeper indications corresponded to locations exposed to greater tube
motion. The tubes which had the deeper indications exhibited vibration
modes indicative of either free or pinned, rather than clamped,
boundary conditions at the TSPs. This indicates that the crevice between
the tube and the support is probably open so that the tube has some
freedom of movement.
It was, therefore, concluded that tube motion is a contributing cause of
cold leg thinning.
Mechanisms
The fact that surface areas, where maximum wall loss has occurred on
tubes degraded by cold leg thinning, do not conform to the mating TSP
surface, has been interpreted to mean that wear or fretting is not a significant
contributor to the damage process. The apparent enhancement of the wall
thinning rate by tube motion has been explained to be a result of (a) mixing
of the corrosive environment which keeps fresh corrodent in contact with
the tube surface, or (b) a pumping action that increases the rate at which
aggressive fluid is concentrated in the tube/TSP intersection. Both of these
mechanisms could accelerate the corrosion rate. It is also possible that
corrosion could increase through direct rubbing contact between tube and
support plate; in addition, solids could be trapped between tube and TSP
which would enhance the thinning effect.
Fretting corrosion can produce deep pits in the fretting surfaces if wear
particles and corrosion products remain trapped between the rubbing
surfaces. Pits on contacting smooth surfaces is a common characteristic of
mechanical fretting and wear in an air environment. This is accomplished
by the mechanism illustrated in Figure 16-8. Fretting is a wear process in
which the amplitude of the relative sliding between two parts is very small
in comparison to the size of the contact area between them. Because the
sliding distance is so small, loose particles formed by wear cannot escape
from the area between the rubbing surfaces and remain trapped. Since the
metal particles have a large surface-to-volume ratio they are rapidly
oxidized. This increases their volume and makes them more abrasive.
Additional wear occurs on both of the mating surfaces as they continue to
rub against the trapped oxide particles. This produces more wear particles
which allows the process to continue. Since pits are being formed in the
mating surfaces, wear debris is more easily trapped between them so that
the pits become deeper. Around the edges of the pitted area, the original
metal surfaces remain in contact because particles can escape from this
region. Because the particles responsible for formation of pits in fretting
wear are not attached to either surface, they usually are lost when the
surfaces are separated.
FIGURE 16-8
Schematic Illustration of the Formation of Nonconforming Surfaces
by Fretting Corrosion
Amplitude of
Trapped Debris Forms Relative Motion
Pits in Rubbing Surfaces -- j
Wear and
Corrosion
Debris
Contact
The surfaces formed on tubes that have experienced cold leg thinning
are very similar to those formed by fretting wear. The edges of the damaged
region have a pattern that is a mirror image of the machined surface on the
TSP counterface (the hole ID). This strongly suggests that rubbing against
the tube hole ID produced this portion of the surface. The central region of
the damage zone is a pit that was possibly formed by fretting-corrosion
wear by corrosion products trapped between the tube and TSP surfaces. In
the case of cold leg thinning, the material removal is probably by a
synergistic interaction of very mild wear and corrosion such that the role of
the wear is to keep the metal surfaces from forming protective oxide films.
This type of damage is termed fretting-corrosion. The material trapped
between the fretting tube and TSP surfaces for the case of cold leg thinning
is probably a corrosion product with composition determined by the
chemical species present in the concentrated solution that forms in the tube-
TSP crevice. The lack of obviously deformed or work hardened material at
the damaged surface shows that the mechanism of material removal in cold
leg thinning is a synergistic interaction.
Most corrosion processes occur more rapidly at higher temperatures.
Therefore, most of the corrosion damage to steam generator tubes is more
severe on the hot leg portions of the tubes. However, there are some
exceptions, such as cold leg thinning, some cases of phosphate induced
wastage, and pitting, which Occur more rapidly on cold legs. Two
explanations have been offered to explain why cold leg thinning is found
only on the cold leg side of the steam generator: (1) the environment
responsible for the corrosive part of the process cannot exist in crevices
which have a high available superheat. Thus, the cold leg crevices which
have only about 15'F of superheat can concentrate a corrosive environment
that causes tube wastage when assisted by tube motion. In the hot leg TSP
crevices, the additional superheat causes the impurities that are responsible
for cold leg thinning to evaporate or to precipitate as solids. (2) Tubes are
not free to move in the hot leg TSP intersections. Apparently, tube motion
is required within the TSP hole to cause cold leg thinning; therefore, this
type of degradation is not expected to occur if the annular crevices between
the tube and the TSP become filled with deposits or corrosion products. It
is possible that hot leg TSP crevices become fouled very early in steam
generator life. Thus, the tubes cannot move in the TSP holes and the cold
leg thinning mechanism is prevented.
Because the surfaces of sample tubes that have experienced cold leg
thinning are clean and free from deposits, it has not been possible to identify
the chemical species that are responsible for the corrosion portion of the
material removal process. However, on the basis of indirect evidence, it is
suspected that cold leg thinning is caused by neutral or slightly acid sulfate
solutions. In several of the plants that have experienced cold leg thinning,
it is known that sulfates have been introduced into the steam generators by
intrusions of ion exchange resin or contamination from condensate polisher
regeneration chemicals. In some plants, sulfates are found in deposits from
locations other than the TSPs where cold leg thinning is occurring, and
some plants have detected high sulfate levels in hideout return.
Laboratory Corrosion Studies

A number of laboratory studies have been undertaken to evaluate
possible chemical environments in TSP crevices where cold leg thinning
occurs. Isothermal tests of C-ring and capsule specimens in acid sulfate
environments have been conducted by Central Electricity Research
Laboratories (CERL) 1(6.9), Westinghouse (16.10), and Materials Engineering
Associates (MEA) (16.11-16.13). Combustion Engineering has performed
model boiler tests (16.14).
The primary objective of the tests by CERL and Westinghouse was to
determine the conditions that will result in stress corrosion cracking of alloy
600 in sulfate environments. The isothermal tests showed that acid sulfate
solutions will produce cracking when the pH of the solution is reduced
below neutral. Cracking was frequently accompanied by wastage and,
under certain conditions, wastage predominated over SCC. According to
CERL experiments, wastage became more rapid when the temperature was
increased from 554 to 635°F (290 to 350'C); due to the wastage it was difficult
to maintain low pH in the static autoclaves at the higher temperatures. The
MEA study included measurements of the amount of wastage produced
on stressed C-ring samples in a static autoclave at 600'F (315'C) for 100
hours. Tests were conducted in sulfate (SO4 -) and sulfate plus chloride
(Cl-) environments. Wastage rates were correlated with pH and equivalent
sulfate ratio, where equivalent sulfate ratio is defined as:
Equivalent Sulfate Ratio = 2 M SO4- M1C1-
-
2 M S04 -+ M Cl-
and
M = Molality = F(moles solute, kg H 20) (16-2)
The results of the MEA tests are shown in Figure 16-9. The wastage
rate increases with decreasing pH. The logarithm of the wastage rate is
approximately a linear function of pH. For a given pH, wastage is most
rapid in pure sulfate and least rapid in pure chloride.
MEA also evaluated the effectiveness of zinc oxide (ZnO) as an inhibitor
for wastage corrosion in sulfate/chloride solutions. As shown in
Figure 16-10, the addition of ZnO to acidified sulfate/chloride solutions
reduced the rate of wastage at all temperatures tested and has an increasingly
larger effect as the temperature is reduced. The wastage rate reduction is

approximately a factor of 2 at 690'F (370'C) and nearly a factor of 5 at 530'F
(275°C). The tests also showed that ZnO suppressed stress corrosion
cracking (SCC) at the higher test temperatures where SCC was observed in
the sulfate/chloride environments without the ZnO additive. These results
indicate that it may be possible to prevent attack of steam generator tubes
by sulfate solutions by using inhibitors such as ZnO.
FIGURE 16-9
C-Ring Wastage As a Function of pH for Various Equivalent
Sulfate Ratios
10' 10'
4 I - . - - Equivalent Sulfate (S0) Ratios 4
N 1.00 0 0.01
080.7 *0.00
CF 1o' A 0.19 1to
E
- 6
102 U~. - -S.- .4. -~ a~ ~ 4 4 4~U~

1,102
- - .,. - .,. -~ r -~~-----~. ~- - * -I.--.---~..: B
a)
R)
CO
3:
:Liz7if3fEz7~j~
10
: - I- -I i
,v(10
1 - - ... 1 U ~J U
1 ---4- 1
- ........4.........4....~..... ~ - 4.. .4..........~ U ±..
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
pH at 315*C
Combustion Engineering has performed a number of model boiler and

pot boiler tests with environments containing acid sulfate species in
combination with other dissolved and solid contaminants. The model
boilers and pot boilers are different from one another regarding the number
of tubes and the heat flux. (The model boilers have 16 tubes compared to 2
to 4 for the pot boilers, and the heat flux is higher in the model boiler.) Test
materials included alloy 600 in a variety of heat treatment conditions, alloy
690, and alloy 800. Tests and results are summarized in Table 16-5. In
general, all of the environments that contained sulfate and chloride ions
produced wastage along with other types of attack, including IGA, IGSCC,
denting, and pitting. Sulfuric acid alone with very low heat flux caused
only shallow IGA on alloy 600 samples and no corrosion on alloy 690. In
tests which involved a variety of materials, the most extensive corrosion
usually occurred on the mill annealed alloy 600 samples. However, the
environment faulted with exhausted resin fines in the Na+/C1- form
(0.8 ppm C1-, 1.1 ppm Na+, 5.3 ppm S04 -) produced deeper IGA on alloy
800 samples than on alloy 600 samples. An environment faulted with
exhausted resin fines in the H+/OH- form (50 ppb C1-, 200 ppb S04 -) and
acidified Missouri River water (3 ppm Cl-, 35 ppmSO4- -produced wastage
in steam blanketed regions of the model boiler on alloy 600, alloy 690, and
alloy 800 samples. The maximum wastage in these environments occurred
on the alloy 600 samples.
FIGURE 16-10
Weight Loss in Acidified Sulfate/Chloride Solutions With and
Without 2% ZnO Additions
Degrees Fahrenheit
650 600 550
-5.00.,
Alloy 600 &Alloy 690
Z I Weight Loss vs. 1/T
- -5.50 I- TLT Specimens
o IAcidified Sulfate Environment
C 0
,U) -6.00 " 3 - ... o~i oI
o - v 2o
.- J-- - C Oyide
• -6.50 13[ ,
- oo.. I 0
v-
o~C.,
. -8.00
z
-8.500.0016 0.001
0.0017 .
0.0018
I/T, Deg. Kelvin

TABLE 16-4
Summary of Combustion Engineering Model Boiler Tests of
Sulfate Environments (16.15)
Bulk Time Destructive

Impurity Concentration (Days) ECT Examinations
Pot Boiler Tests
H2SO4 300 ppb S04- 2 130 ND Shallow IGA
L. I. Sound Water + 1- 5 ppm CI- 114 80% Wastage, IGA,
NaHSO 4 + 02 30 ppm S042 shallow pitting,
15- 50 ppb 02 denting
Tennessee River + 3- 80 ppm CI 219 95% Wastage, shallow
Na2S0 4 + 80-100 ppm S04- pitting, IGA,
(NH4 )2 SO4 + IGSCC
NH4CI
Model Boiler Tests
Exhausted Resin 0.8 ppm CI- 276 25% Wastage, IGA,
Fines (Na+/Cl 1.1 ppm Na+ denting,
Form) 5.3 ppm S04"2 SP corrosion
Exhausted Resin 50 ppb C-2 462 ND Minor wastage
Fines (H+/OH- 200 ppb S4-2 and IGA
Form)
Acidified Missouri 3 ppm Cl" 358 100% Wastage
River Water 35 ppm SO4- (25-30%),
IGA, IGSCC for
all heat treatments
of alloy 600
Westinghouse performed a three-month-long, single-tube model boiler

test on a sample of mill annealed alloy 600 with acid sulfate chemistry. The
sulfate concentration in the bulk water in this test was maintained in the
range of 7 to 20 ppmSO4- -. The primary side temperature was 610°F (321'C)
at the inlet, and the secondary side temperature was 518°F (279QC). The
test produced 280 ptm (11 mils) of wastage on the tube, and IGA to a depth
of 100 lam (4 mils).
Because pH is not a controlled variable in the hideout locations in the
model boiler tests, it is not clear whether these tests show the same pH
dependence of wastage rate that is observed in the isothermal autoclave
tests. However, it is clear that sulfate and chloride environments have a
tendency to produce wastage as well as IGA and IGSCC. Table 16-5 shows
results of MULTEQ calculations for a model boiler environment in one of

the C-E tests. MULTEQ predicts that the environment produced in hot leg
and cold leg hideout locations for this test was neutral to acidic. Field
experience with cold leg thinning is different from the results of laboratory
tests in that metallography of removed tube samples shows that the cold
leg thinning attack is transgranular wastage with no preferential attack of
the grain boundaries. Laboratory tests in acid sulfate environments,
however, usually produce a combination of wastage and grain boundary
attack. This difference may be the result of the effect of mechanical fretting
in the cold leg thinning mechanism. If fretting prevents the tube surface
from passivating, wastage may be accelerated and preferential grain
boundary attack will be suppressed.
TABLEI 6-5
Combustion Engineering Model Boiler Test Model 10 (16.16)
Mean Median
Concentration Concentration
Species kpprn.
CI- 3.09 2.90
SO4 - - 35.86 34.0
Na+ 16.43 13.05
02 0.00064 0.0000
SiO 2 3.47 2.00
NH4 + 1.34 0.59
N2 H4 0.0056 0.0050
Fe (sol) 0.00544 0.0050
Cu (sol) 0.03597 0.0370
Ni (sol) 0.00093 0.00050
Ca+ + 0.69 0.44
Mg+ + 0.15 0.025
F- 0.19 0.17
K+ 1.96 1.70
N03- 0.09 0.06
Calculated Mean Median

Hideout Chemistry Chemistry
Location pH
At 171C BPE (Cold Leg) 4.9 2.7
At 28 0C BPE (Hot Leg) 5.0 2.0
Impact on Steam Generator Operations and Potential Remedies

Cold leg thinning has been observed in a relatively small percentage of
tubes in Westinghouse-design steam generators that are not equipped with
preheater (economizer) sections. The overwhelming majority of generators
that have experienced cold leg thinning are Model 51; one Model 44 and
one Model 24 steam generator have also been affected. To date there has
been no report of cold leg thinning in the newer, non-preheater Model F
steam generators. Owners of Model D and E series of Westinghouse-design
steam generators, which do have preheater sections, have not reported cold
leg thinning. Typically, less than 100 tubes per plant at the exposed plants
have experienced degradation; even fewer tubes have required plugging.
However, as operating time increases, affected plants continue to find new
tubes targeted by cold leg thinning. There is also a tendency for the tubes
that become damaged later in life to be at higher TSP elevations and further
from the tube bundle periphery than those tubes that experienced damage
early in life. This may mean that a substantial fraction of the tubes in the
generator are potentially susceptible to cold leg thinning before the end of
the steam generator design life. However, the damage progresses slowly,
and, therefore, the number of tubes that will require plugging for cold leg
thinning in a given steam generator will probably be less than the plugging
limit. Cold leg thinning is not considered likely to be the primary life-
limiting degradation mechanism for currently operating steam generators.
Because tube motion is believed to play an important part in the cold
leg thinning mechanism, it is possible that modifications of tubes in the
susceptible region near the periphery of the cold leg at the first two tube
supports will prevent tube motion and thereby the degradation mechanism.
In order to test this possibility, 50 tubes in the Beaver Valley Unit I steam
generators were hydraulically expanded at the first two tube supports to
minimize the tube motion. The annular gap between the tube and the TSP
was reduced to approximately 0.001 inch. This leaves the tube free to expand
relative to the TSP position so that axial forces are not applied to the tube.
The results of this test are still inconclusive.
Cold leg thinning could possibly be controlled by preventing the
formation of corrosive environments in the cold leg TSP crevices. This is
difficult because boiling in restricted flow locations, such as the TSP crevices,
can produce aggressive chemical conditions by concentration of impurities
even when bulk water purity is very good. Laboratory studies, such as the
MEA work described above, have shown that use of inhibitor chemicals
can reduce rates of wastage in acid sulfate environments. Thus, use of
inhibitors is a possible remedy for cold leg thinning degradation, but no
inhibitor has yet been qualified for use in an operating steam generator. At
present, maintaining good water chemistry with a proper cation-anion
balance is the only means to prevent cold leg crevice chemistry from creating
an environment that will promote cold leg thinning. Poor operation of
condensate polishers has been shown to introduce both sulfates and resin
particles that break down to form sulfates. Therefore, operation of ion
exchange systems must be improved to prevent the introduction of
impurities that are known to form environments that can promote cold leg
thinning.
Non-Destructive Detection and Measurement of Cold Leg Thinning

Tube wastage at TSP intersections is easily detected by ECT techniques
before the defect approaches a depth that requires the tube to be plugged.
The defect occurs in a location where the ferromagnetic tube support
produces a strong signal in ECT inspections. Therefore, the signal from the
support and the signal from the defect must be separated so that the depth
of the defect can be quantified. Use of absolute-mode ECT with a frequency
mix to minimize the TSP signal is the most effective way of measuring cold
leg thinning defect depths. It has been found that ECT frequently
overestimates the depth of wear-type defects at supports. While this is
conservative, it can lead to unnecessary tube plugging. In general, detection
and monitoring of cold leg thinning defects has not been a serious problem.
Because relatively few tubes are affected, plugging decisions can be based
on conservative criteria that assure that the tube is plugged before its
strength is significantly degraded.
PREHEATER WEAR
In preheater type steam generators, most of the feedwater is injected
into a baffled enclosure at the bottom of the cold leg half of the generator.
This is a departure from earlier designs in which the feed water was
introduced into the downcomer annulus. The preheater feedwater injection
arrangements resulted in higher cross flow velocities near the preheater
entrance, and a number of problems were encountered with excessive wear
rates (preheater wear). Details of early experiences with wear in preheater
units is conatined in the previous edition of the Steam Generator Reference
Book (16.17)
Preheater wear was first .detected in Ringhals Unit 3 in 1981. The
Ringhals Unit 3 steam generators are Westinghouse Model D3 which have
a split flow preheater arrangement. The first leak, later attributed to
preheater wear, occurred after approximately 2,200 hours of operation at
75% power. It was determined that tubes in the preheater near the feedwater
entrance were experiencing rapid wear at the preheater baffles. Wear was
also found by inspections of similar steam generators at the Almaraz Unit 1

plant. In a Model D3 steam generator, the feed water is directed at an
impingement plate as it enters the preheater. The purpose of the
impingement plate is to break up and diffuse the flow. Measurements and
calculations indicated that the impingement plate arrangement was not
sufficiently dispersing the incoming flow to prevent the excitation of high
amplitude vibrations of the tubes. Investigation of the flow fields and tube
vibrations in the Model D4 steam generators at the Krsko plant showed
that they were also subject to excessive vibration at 100% power level. The
Krsko plant has Model D4 steam generators that have a counterflow design
preheater with a feedwater inlet arrangement that is different from the
impingement plate in the Model D3 generators. However flow-induced
vibrations were also occurring in the counterflow generators at power levels
well below 100% power.
Analyses and model tests performed for the Model D steam generators
identified fluid-elastic instability as the most likely cause of vibrations in
the Model D3 generators, and turbulent buffeting as the most probable
driving force for the Model D4 generators. In all cases, large amplitude
flow-induced vibrations are associated with excessive fluid-flow velocities.
The vibrations are of sufficient amplitude to cause the tubes to impact and
rub against the surfaces of the preheater baffle and TSP holes causing rapid
mechanical wear. The process is considered to be completely mechanical.
The excessive vibrations and wear damage were limited to a few rows of
tubes near the feedwater entrance.
A number of remedial actions have been taken to mitigate the preheater
wear problems in Model D steam generators. The feedwater inlet for
Model D2 and D3 split-flow steam generators has been modified by
incorporating a manifold device in the feedwater inlet, which reduces
turbulence and diffuses the inlet flow. The tubes in Model D4 and D5
counter-flow steam generators in the region near the feedwater inlet have
been stiffened by expanding them within the baffle plate holes at several
elevations in the preheater. After expansion, a 0.003 inch diametrical
clearance remains between the tube and the baffle plate to allow relative
thermal expansion motion. In addition, flow velocities in the counter flow
preheaters of the Model D4 and D5 steam generators have been reduced
by splitting the feedwater flow between the preheater inlet and an auxiliary
feedwater nozzle above the downcomer annulus for the recirculating flow.
These modifications have been successful, and preheater wear has not been
a significant problem in plants with the modified steam generators.
FRETTING IN OTSG TUBES

Fretting has occurred in a limited number of tubes in once-through
steam generators (OTSGs) manufactured by Babcock & Wilcox (16.2). Very
few tubes have been plugged due to this cause and the problem is not
considered to be significant. The fretting damage has been limited to the
uppermost tube support plate in the OTSGs in tubes located within three
rows of the open inspection lane in the tube bundle (a missing row of tubes
from the center to the periphery of the bundle). This region is subject to
higher cross flow velocities than the remainder of the bundle. The
uppermost tube support plate is located in the superheat region of the
generator and the tubes in this region are subjected to a concentration of
suspended solids in the last remaining water droplets. The tube motion
induced by the cross flow and the presence of the solid particles from the
water droplets cause fretting wear damage to the tubes where they contact
the lands of the broached-hole tube support. Both the number of tubes
affected in all B&W OTSGs and the rate of progression of damage are small.
Damage can be monitored by ECT inspection using pancake coil array
probes. It is believed that damage rates can be reduced by improvements
in water chemistry such that the volume of solids deposited on the tubes in
the superheat region is minimized.
SUMMARY OF CONCERNS WITH CORROSION

AND WEAR AT SUPPORT STRUCTURES IN
NUCLEAR STEAM GENERATORS
Tube corrosion and wear at support structures is a significant
degradation mechanism, but has much less impact than the various
degradation mechanisms that lead to cracking or intergranular attack of
the tubes. Tube motions induced by flow forces are an essential part of all
of the degradation mechanisms discussed in this chapter. As discussed
earlier, the phenomena of flow-induced vibration are not completely
understood, and it is evident that empirical rules, used in steam generator
design to avoid excessive vibrations induced by flow forces, are not
sufficiently conservative to assure that the design objective will be met. By
far the most widespread mechanism of wear/corrosion damage in PWR
steam generators is AVB and batwing wear (16.18). This mechanism affects
essentially all models of recirculating steam generators to some degree.
Operating experience shows that newer steam generator designs, completed
after the phenomenon was well known, are no more resistant to this type
of damage than earlier designs. Although AVB wear does not usually require
plugging of a sufficient number of tubes to threaten the operability of the
steam generator, the costs of inspections and remedial measures are
significant. To date, the cold leg thinning mechanism has affected only a
few steam generator designs; currently, approximately 20 plants have been
affected. Because the mechanism is not completely understood, no clear
guidelines exist prescribing how to prevent the mechanism from occurring
in future steam generators. For the plants that are experiencing degradation,
the rate of damage progression is small and does not greatly impact the
plugging rate. As with AVB wear, cold leg thinning requires additional
inspection effort. Preheater wear was the result of unanticipated high flow-
induced vibrations in the initial preheater steam generators. Modifications
to feedwater flow paths appear to have arrested this type of damage; it is
no longer considered to be a significant problem. TSP fretting and wear in
OTSGs also appears to be less significant.
REFERENCES
16.1 Steam Generator Progress Report, Rev. 7. EPRI. September 1991.
16.2 Workshop Proceedings: Fretting and Wear Susceptibility in Nuclear
Steam Generators. Washington D. C., March 19-20, 1986.
16.3 Baum, A. J., C. Dangelo, P. J. Kuchirka, N. Singleton, and M. J.
Sredzienski. "Steam Generator Cold Leg Thinning in Operating
Plants." EPRI NP-5140. April 1987.
16.4 "Steam Generator Inspections-San Onofre Nuclear Generating
Station, Unit 1." January 30, 1977. "Interim Steam Generator
Inspections of All Tubes in Area of Antivibration Bars-San Onofre
Nuclear Generating Station, Unit 1." October 1977. Attachments
to letter from Southern California Edison Company to U. S. Nuclear
Regulatory Commission (Docket Number 50-206) dated October
3, 1977.
16.5 Malinowski, D. D., and W. D. Fletcher. "Update of Operations With
Westinghouse Steam Generators." American Nuclear Society, 23rd
Annual Meeting, June 12-16, 1977, New York, N. Y.
16.6 Aspden, R. G., and E W. Pement. "Destructive Examination of Zion
Unit I Tube and Antivibration Bar Samples." EPRI NP-4375-LD.
December 1985.
16.7 McInteer, W. A. "Calvert Cliffs Unit 1 Tube Examination." EPRI
16.8 Rudell, B. C., and D. J. Van Petten. "Destructive Examination of
Steam Generator 11 Tubing at Calvert Cliffs Power Plant Unit 1."
EPRI NP-7136-SD. April 1991.
16.9 Newman, J. F. "Stress Corrosion of Alloys 600 and 690 in Acidic

October 1983.
16.10 Baum, A. J., W. M. Connor, C. D. Piskor, and W. J. Stenger. "Acid
Sulfate Corrosion in PWR Steam Generators." EPRI NP-7346-SD.
June 1991.
16.11 "Proceedings: EPRI Workshop on the Role of Sulfur Species on the
Secondary Side Degradation of Alloy 600 and Related Alloys." EPRI
NP-6710-SD. March 1990.
16.12 Cullen, W. H., M. J. Partridge, J. A. Gorman, and J. P. N. Paine.
"IGA/IGSCC of Alloy 600 in Acid Sulfate Solutions." Colloque
International, Fontevraud II, September 10-14, 1990.
16.13 W. H. Cullen, and M. J. Partridge. "Acid Sulfate Studies." To be
published.
16.1.4 "Evaluation of Condensate Polishers." EPRI NP-3046. June 1983.
16.15 Lurie, S. W., T. A. Beineke, et al. "C-E Pot Boiler Tests with Sulfate-
Faulted Chemistry." In Workshop on the Role of Sulfur Species on the
Secondary Side Degradationof alloy 600 andRelated Alloys, Washington,
DC, December 1988. EPRI NP-6710-SD. March 1990.
16.16 Krupowicz, J. J., D. B. Scott, R. M. Rentler, and G. C. Fink. "Corrosion
Performance of Alternative Steam Generator Materials and
Designs," Vol. 3. EPRI NP-3044. July 1983.
16.17 Steam GeneratorReference Book, Chapter 12. EPRI. May 1985.
16.18 Hofman, P. J., T. Schettler, and D. A. Steininger. "Pressurized Water
Reactor Steam Generator Tube Fretting and Fatigue Wear
Characteristics." ASME Pressure Vessels and Piping Conference,
July 21-24, 1986, Chicago, Illinois. 86-PVP-2.
17
TUBE FRETTING AND WEAR

18
TUBE FATIGUE
19
OTHER PROBLEMS
20
WATER CHEMISTRY CONTROL

21
IONIC IMPURITY CONTROL

22
WATER CHEMISTRY OPTIONS

23
CORROSION PRODUCT
CONTROL
24
MATERIAL SELECTION AND

ALTERNATIVE DESIGNS FOR
STEAM GENERATORS
A. R. Mcllree, EPRI
C. E. Shoemaker,EPRI
1994
MATERIAL ALTERNATIVES FOR TUBES AND

SUPPORT STRUCTURES
Improved tube and support plate alloys with better corrosion resistance
than those now in' widespread use are available for new steam generators
but each alloy has environmental limitations. Improved tube alloys
evaluated in comparison with mill annealed (MA) alloy 600 include
thermally treated (TT) alloy 600, MA and TT alloy 690, and commercial
and nuclear grade (NG) alloy 800. For support structures, alloys 405, 409,
430, 439, 347, SCR-3 stainless steel and nickel-based alloys (such as alloy
800) are compared to carbon steel. These alloys and their environmental
limitations are discussed below.
Tube Materials
Steam generators contain from 20,000 to 130,000 square feet of heat
transfer tubing surface. Tubing wall thickness (the primary to secondary
pressure boundary) is between 30-50 mils (1-2 mm) thick. If the tubing is
to last 40 years, the corrosion rate must be kept below a few tenths of a mil
(a few thousandths of a millimeter) per year over its entire surface, including
areas where flow is restricted around the tubes on the secondary side. In
these locations, concentrated, corrosive environments may accumulate from
residual salts in the boiling secondary water.
As discussed more completely in Chapter 8, these concentrated
environments form under scale, in support structure crevices, in tubesheet
crevices, in sludge piles, under straps, or in other areas where secondary
water with restricted flow contacts heated tubes. The environments vary
widely in composition depending on the residual salts in the water, and
the intrusion of salts through leaks, upset conditions, etc. These localized
environments have caused stress corrosion cracking (SCC), intergranular
corrosion (IGC) or intergranular attack (IGA), pitting, and wastage of tubes
(24.1). Testing has focused mainly on alloy 600, but, when possible,
alternative tube alloys have been included for comparison. Specified
compositions and mechanical properties of tube alloys are given in
Table 24-1.
Originally, domestic suppliers and their licensees chose alloy 600 for
tubes because of its superior resistance to chloride stress corrosion cracking
(SCC), a troublesome problem with stainless steels. It has fulfilled this need
well, and, at present, no SCC tube failures in steam generators have been
attributed to chlorides. Shortly after alloy 600 was accepted for steam
generator tubing, alloy 800 with higher chromium, and lower nickel contents
was developed. When Coriou (2.2 reported cracking of alloy 600 in water,
Kraftwerk Union (KWU) switched from alloy 600 to alloy 800 for their
Material Selection and Alternative Designs for Steam Generators 24-3
Table 24-1
Composition and Mechanical Properties of Tube Alloys for Nuclear
Steam Generators
Chemical Composition (wt %)

Alloy Alloy Alloy Alloy
2
600(1) 690(1) 800 800 NG( )
Al .50 max .15-.60 .15-.45
B (max) .007
C .025-.05 .015-0.025 0.10 max 0.03 max
Co (max) .02 .02 .10
Cr 15-17 28-31 19-23 20-23
Cu (max) .50 .50 .75 .75
Fe 6-10 7-11 39.5 min bal.
Mn 1.0 max .50 max 1.5 max .4-1.0
Mo .20 max
N (max) .05 .03
Ni 72.0 min 58.0 min 30-35, 32-35
Nb (max) 0.1
P (max) 0.15 0.15 0.20
S (max) .010 .010 .015 .015
Si .50 max .50 max 1.0 max .3-67
Ti .50 max .15-.60 .60 max
Mechanical Properties
MPa (ksi) Percent
Alloy Yield Ultimate Elongation
MA or TT 600 241-389 (35-55) 552 (80) 30 min
MA or TT 690 276-365 (40-53) 552 (80) 30 min
800(3) 375 (54.4) 663 (96.2) 34
1
EPRI Guidelines NP-6743-L, Vol. 1&2.
2
KWU stabilized grade, Ti/C>12, Ti/(C+N)>8, N>.03 (24.53).
3
Typical values.
tubing and developed a stabilized nuclear grade (NG) (24.3). Stabilization

is accomplished by minimizing the carbon content and adding a small
amount of titanium which combines with carbon to prevent the precipitation
of chromium carbides during a stress relief anneal of the steam generator
after fabrication. When chromium carbides precipitate on grain boundaries,
the chromium content of the alloy near the grain boundary may be depleted,
making the alloy more susceptible to intergranular corrosion. Not all the
carbon is combined with titanium in commercial alloy 800 and some carbide
precipitation is possible. Later, alloy 690 with still higher chromium and
nickel contents was developed as an improvement over alloy 800 24_.4).
In the following subsections, the relative corrosion of alternative tube
alloys are compared in representative corrosive environments. Because the
nuclear grade may be more corrosion resistant than the commercial grade
of alloy 800 in the following discussion, the grade will be identified when
known. Fretting, wear, and corrosion fatigue are discussed in Chapters 17
and 18.
Stress Corrosion Cracking. Mill annealed alloy 600 tubing is susceptible

to SCC in a variety of caustic environments and, to a much lesser extent, to
SCC in pure water, primary water, and in other aqueous environments.
Most investigators have not separated the variable crack initiation period
from the crack growth period. The resultant error in crack growth rate is
small in more corrosive environments at high temperature where the crack
initiation time is short. But the error is significant at lower temperature
and in less corrosive environments where the initiation time may be long
compared to the cracking time.
Corrosive test results of alternative tube alloys are compared below for
several environments. They are summarized later in this chapter in
Table 24-31.
Chlorides. Alloy 600 has excellent resistance to SCC in chloride
solutions. Available data indicate that alloy 690 may be even more
corrosion resistant but the data on alloy 800 are conflicting. Laboratory
data at 260'C show that commercial alloy 800 is very susceptible to
chloride SCC but the KWU nuclear grade has resisted chloride SCC
well in German steam generators. A tentative explanation of the
differences is presented below. In the following discussion, available
but limited data will be used to show the effect of oxygen, the
improvement in corrosion resistance of the nuclear grade over the
commercial grade of alloy 800, and the reduction of chloride SCC that
can be obtained by modifying the chloride environmeitts with
phosphates and AVT water.
The susceptibility of alloy 800 to SCC in chlorides has been

summarized by Sedriks, et al. (24.4). Tests in boiling magnesium
chloride at 310'F (154°C) in Table 24-2 found no cracking of alloys 600
and 690 while all specimens of alloy 800 and 304 SS cracked. Tests in
500 ppm chloride (from sodium chloride) at 500F (260°C) with residual
air not removed found similar results (Table 24-3). No 690 specimens
cracked after 18 weeks while all alloy 800 and 304 SS specimens cracked
after one week. Interestingly, specimens in the vapor phase above the
solution behaved similarly (Table 24-4). No alloy 690 specimens cracked
after eight weeks while all alloy 800 and 304 SS specimens cracked after
one week. SCC data in Table 24-5 confirm the test results (24.5). The
test was in steam containing 10 ppm chloride and 20 ppm oxygen at
temperature from 536-806'F (280-430'C). The severe conditions
produced no SCC (0/6) in alloy 690 specimens after 18 weeks, cracking
in 2/9 specimens of alloy 600 after times up to 18 weeks, cracking in 7/
12 specimens of alloy 800 after 16 weeks and cracking in 9/12 specimens
of 304 SS after three weeks.
TABLE 24-2
Stress Corrosion of Tube Alloys (U-Bend Test Specimens) at
310°F (154 0C) in Boiling Magnesium Chloride (24.4)
No. of
No. of Specimens Max.
Specimens Test Cracked/No. Depth of
(No. of Duration Destructively Attack
Alloy Heats) "wk Examined mils (mm)
600 3(1) 12.9 0/3 0
690(1) 9(3) 12.9 0/9 0
800 3(1) 0.4 3/3 114(2.9)
304SS 3(1) 0.1 3/3 98(2.5)
Heat Treatment: Alloy 690, 1 h @ 2,000°F (1,093°C)/WQ + 5 h @ 1,300'F

(704 0C)/AC.
Others, 1 h@ 1,9501F (1,065 0C)/WQ + 5 h @ 1,300'F
(704°C)/AC.
1Heats T-53176, T-53838, T-99016.
Table 24-3
Stress Corrosion Tests of U-Bend Test Specimens at 500OF (2600 C)
in Undeaerated Water Containing 500 ppm Chloride (NaCI) (24.4)
No. of
Aloy Heats) "wk) Examined mils (mm)
690(1) 4(1) 8 0/2 0
18 0/2 0
800 2(1) 1 2/2 87(2.2)
304SS 2(1) 1 2/2 118(3.0)
Heat Treatment: Alloy 690: 1 h @ 1,999.4°F/WA (1,093°C) + 2 h @ 1,202°F/AC

(6500C).
Others: 1 h @ 1,9490 F/AC (1,0650C) + 5 h @ 1,299.20 F/AC
(7040C).
1 Heats Y24A7L
Table 24-4
Stress Corrosion Tests of U-Bend Test Specimens in Vapor Phase
Above Undeaerated 500'F (260'C) Water Containing 500 ppm NaCI
(24.4)
No. of
Alloy Heats) "wk Examined mils (mm)
690(1) 6(3) 8 0/6 0
800 2(1) 1 2/2 98(2.5)
304SS 2(1) 1 2/2 118(3.0)
HeatTreatment: Alloy 690: 1 h @ 1,999.40 F/WQ (1,0930C) + 5 h @

1,2990F/AC (7040C).
Others: 1 h@ 1,949°F/WQ (1,0650C) + 5 h @ 1,2990 F/AC
(704°C).
1
Heats T-53176, T-53838, T-99016.
Table 24-5
Stress Corrosion of Tube Alloys (U-Bend Test Specimens) in Steam Containing 10 pm Chloride and
20 ppm Oxygen (24.5)
Specimens
Cracked/ Max. Depth
Heat Treatment Code No. of Specimens Test Duration Destructively of Attack
Alloy (Surface Treatment) (No. of Heats) "wk Examined mils (mm)
690 A6 (ground) 3(1) 18 0/3 0
A6 (as received) 3(1) 18 0/3 0
600 A7 (ground) 3(1) 16 1/3 94 (2.4)
A7 (ground & annealed) 3(1) 18 0/3 0
A7 (as received) 1 (1) 5 1/1 126 (3.2)
2(1) 18 0/2 0
800 A8 (ground) 6(1) 16 6/6 190 (4.8)
A8 (ground & annealed) 6(1) 16 1/6 63(1.6)
304SS A8 (ground) 6(1) 3 3/6 252 (6.4)
A7 (ground & solution 6(1) 3 6/6 126 (3.2)
annealed)
Test temperature - cycling between 536°F and 805°F (2801C and 4301C).
Code: A6 = 1,726 0F (9410 C)/AC (air cooled)
A7 = 1,850°F (1,010 0C)/AC
A8 = 2,1500 F (1,177 0 C)/AC
Ground = surfaces ground on a 100 mesh grit belt
Ground and annealed = ground + 0.16 h @ 1,800°F (9820 C)/AC
Ground and solution annealed = ground + 0.5 h @ 1,9500 F (1,066 0C)/WQ (water-quenched)
When oxygen was eliminated from the chloride environment as in

Table 24-6, cracking of specimens was greatly reduced in 500 ppm
sodium chloride at 600'F (316'C). For alloy 800, only 1/3 specimens
cracked after 48 weeks and in a second group of three specimens, none
cracked after 96 weeks. Cracking times were also extended for 304 SS.
While not strictly comparable, the data in Table 24-3 indicate that the
presence of even small amounts of oxygen accelerates the chloride SCC.
Blanchet (24.6) found that minimizing the carbon content in alloy
800 and increasing the Ti/C ratio reduced the SCC of alloy 800
(Table 24-7) in chloride solutions containing oxygen at 662°F (350'C).
The critical Ti/C ratio is not defined exactly but is between 17 and 30.
A lower Ti/C ratio of about 12 was defined by Debray and Stieding as
the critical ratio, above which copper and acid sulfates did not attack
alloy 800 after a sensitization heat treatment at I,200'F (540'C) for one
hour. Their data are shown in Figure 24-1 (24.3). The original Ti/C
ratio for the nuclear grade was based on these data. Recent heats of
NG alloy 800 by Yonezawa, et al., have a ratio over 20. In the latter
tests, no chloride cracking was found in NG 800 in the mill annealed
condition. Cracking developed in some specimens, however, when
alloy 800 was sensitized as shown in Table 24-8.. Thus, there is some
doubt that all of the carbon is combined with titanium even with a Ti/
C ratio of 20.
Table 24-6
Stress Corrosion of Tube Alloys (U-Bend Test Specimens)
at 600°F (316 0 C) in Deaerated Water Containing 500 ppm
Sodium Chloride (24.4)
No. of
AII"y Heats) "wk Examined mils (mm)
690(1) 6(3) 48 0/3 0
96 0/3 0
800 6(3) 48 1/3 4.3(0.11)
96 0/3
304SS 2 (1) 48 1/1 0.8 (0.02)
96 0/1
HeatTreatment: 1 h @2,0501F (1,120°C)/WQ (water quenched) + 2 h
@1,250°F (6750 C)/AC (air cooled).
1
Heats NX 1OC1H, NX 4460H, NX 4458H.
Table 24-7
Influence of Carbon Content and Stabilization Ratio of Alloy 800 on
SCC Resistance in Chloride Solution (Deionized Water With 100 mg/
kg C1" and 20 mg/kg 02. T = 662°F (3500 C), Stress RT Yield Strength)
(24.6)
Carbon
Content Ti/C Results Material Condition
0.015% 30 No SCC after 1,000 h Nuclear Grade
0.015% 47 No SCC after 1,000 h Nuclear Grade
0.043% 11 2 out of 3 cracked after 9,000 h Standard grade
FIGURE 24-1
Influence of Carbon Content and Stabilization Ratio on Depth of
Intergranular Corrosion. Alloy 800 Heat Treated 1 hour at 1,202°F
(650°C) tested in (250 ml/1 H2SO4. (24.3)
{400
Depth of IGA
pm
300
200
100
5 10 15 Ti/C 20
.1- Stabilization Ratio
Table 24-8
Stress Corrosion Cracking Resistance in High-Temperature Water Containing Chloride and Oxygen (24.10)
Alloy 690 Alloy 600 Alloy 800L

4 5 6 7
MA(1) SEN(2) TT(3) MA( ) TT( ) MA( ) SEN( )
Double U-Bend 0/2 0/2 0/2 0/2 0/2 0/2 1/2(7)
C1- 100 ppm Max: 470 gm
550.4-F (288-C)
4,000 h
Double C-Ring 0/2 -- 0/2 0/2 0/2 0/2
Cl- 100 ppm
34.
550.4°F (288°C)
4,000 h
Double U-Bend 0/4 0/4 0/4 0/4 0/4 0/4 3/4(2) *0
Cl- 500 ppm Max: 700jtm

5720 F (3000C)
1,000 h
1 Mill annealed 1,742 0F (9500C).

2
Mill annealed + sensitized 1,022 0 F/1 5h (5500C).
3
Mill annealed 1,967 0F (1,0750C) + thermal treatment 1,292 0 F/15h (7000C).
4
Mill annealed 1,787 0F (9750C).
5
Mill annealed 1,688 0F (9200C) + thermal treatment 1,292 0 F/15h (7000C).
6
Mill annealed 1,697 0F (9250C).
7
Mill annealed 1,697 0F (9250C) + sensitized (a) 1,1121F/100h (6000C) or (b) 1,0221F/15h (5500C).
0
Data from other sources show that SCC of alloy 800 due to chlorides
can be reduced by additives such as phosphates and ammonia (AVT
water). Data in Table 24-9 show that additions of phosphates and
chlorides to aerated water prevented SCC of alloy 800 U-bends for at
least eight weeks at 500'F (260°C) (24.4). Without phosphates and under
nearly identical conditions, alloy 800 U-bends cracked in one week
(Table 24-3).
TABLE 24-9
Results of 8-Week Autoclave Tests on Sensitized Single
U-Bend Specimens Exposed at 500°F (2600 C) to a Simulated
Aerated Boiler Water Containing 660 ppm NaCI and 150 ppm
Na2HPO4 (24.4)
No. Total No. Time Depth

Alloy Heats Specimens (Weeks) mils (mm)
304 1 4 1 100 (2.5) M1
800 1 2 8 0
600 1 4 8 0
28CR 2 4 9 8 0
1M = Mixed path of cracking.

2
Alloys containing 2-24% Fe and 0.05%C, normal residuals.
Laboratory data show even more clearly that AVT water is effective
in eliminating SCC of commercial alloy 800 tubes in oxygenated
chlorides (24.7). Data in Table 24-10 show that no specimens of any
material cracked after times up to 74 weeks (12,400 hours). Without
AVT, chloride cracking of commercial alloy 800 specimens would have
been expected in at least a couple of weeks.
In summary, it appears that commercial alloy 800 is susceptible to
SCC in neutral to acidic chlorides containing oxygen. Apparently, even
residual oxygen not flushed out at the start of a test is sufficient to cause
SCC (Table 24-6). Based on available data, it is unclear how much the
corrosion resistance of the nuclear grade is improved over the
commercial grade of alloy 800 in chlorides. It is clear, however, that
phosphates or AVT water reduces the SCC of alloy 800. It is also
significant that the NG alloy 800 tubes did not crack in model steam
generator test #9 that was faulted with seawater and a copper-oxide-
containing sludge during a year of testing in AVT water (24.8). Thus,

the lack of chloride SCC in German steam generators could be due to
several reasons including a low oxygen content, the use of the NG alloy
800, and the presence of phosphates or AVT (24.9). In addition, shot
peening, now used, may provide additional corrosion protection (24.10).
Alloy 690 exhibits the most corrosion resistance to chlorides,
particularly with oxygen present, but alloy 600 is almost as corrosion
resistant.
Caustic. Caustic is one of the most corrosive and widely used
environments to evaluate and compare tube alloys (24.11). Alloy 600 is
susceptible to SCC in caustic but its corrosion resistance is improved
by a thermal treatment that precipitates chromium carbides at grain
boundaries. The thermal treatment was optimized for alloy 600 by
Airey (24.12). The same treatment has been applied to alloy 690 but
may not be the optimum conditions for 690.
A recent summary of the SCC propagation rates of MA and TT
alloy 600 in varying concentrations of caustic by Jacko is given in the
Arrhenius plot in Figure 24-2 (24.13). The caustic concentration and
activation energies for each line are noted. The data show considerable
scatter, which is partly attributable to the fact that the effect of stress
and initiation times were not isolated. For MA alloy 600, 10% caustic is
significantly more corrosive than 1% caustic. The line for 50% caustic
shows it to be more corrosive than 10 % caustic only above about 617'F
(325°C). At lower temperatures, its corrosivity drops below 1% caustic.
The effect of caustic concentration (deaerated) and temperature were
confirmed by Airey in Figures 24-3 and 24-4 (24.14, 24.15). Thermal
treatment reduced the SCC of alloy 690 more than alloy 600 in
Figure 24-4.
Lower SCC propagation rates occur with TT alloy 600. The SCC
rates increase with temperature and diverge for various caustic
concentration from a common point at about 280'C (536°F). SCC rates
are reduced by at least an order of magnitude at 1% and 10% caustic by
thermal treatment and by lesser amounts in 50% caustic at low
temperatures. For an estimated service life of 40 years, the caustic
concentration in tubesheet crevices at 600'F (315'C) should not exceed
1% for a TT alloy 600 tube. For a MA alloy 600 tube, the crevice caustic
concentration should be kept much lower.
0
In alloy comparisons, Berge and Donati reported that MA alloy
690 and commercial alloy 800, later identified to be alloy 800 NG, had
better corrosion resistance than MA and TT alloy 600 below a
concentration of about 6% caustic at 662°F (350'C) in Figure 24-5 (24.16).
Table 24-10
Failure Times and Crack Depths for C-Ring Specimens Exposed to an AVT + 500 ppm Chloride + 02
Solution at 600°F (316 0 C) (24.7)
No.of Specimens CD
No. of Specimens Not Cracked

Tubing. Thermal/Mechanical Cracked/Time Total Exposure U')
(D
Alloy Code Heat No. Manufacturer Treatment to Failure (wk) Time (wk) (D
600 MAC NX-0471 C-E Mill-Annealed (MA) 0 1/38,1/74

600 MAW NX-1054 W MA 0 1/38,1/74
600 L-A NX-1054 W MA + 7050 C (1,300 0 F)/1 5 h/AC 0 1/38,1/79
MA + 7050 C (1,300 0 F)/1 5 h/AC
600 L-C NX-1054 W + deformed plastically 20% 0 1/20, 1/56 (D
unloaded + 7050C (1,300 0F)/2 h/AC 1

600 L-B NX-1054 W MA + 500-C/24 h/AC 1 0 1/38,1/74
(D
600 PS NX-0471 C-E Process Stabilized 0 1/38,1/74
690 MAW-A NX-9780 W MA 0 1/38,1/74
690 MAW-B NX-9780 W MA + 705'C (1,300'F)/4 h/AC 1 0 1/38, 1/74
690 MAW-C NX-9780 W MA + 705'C (1,300°F)/1 5 h/AC 1 0 1/38, 1/74
690 MAW1 NX-9780 W MA 0 2/28 Ci)
690 L42 NX-9780 W MA + 705'C (1,300'F)/4 h/AC 0 2/28 CD

690 LW NX-9780 W MA + 705'C (1,300°F)/1 5 h/AC 0 2/28
690 MAW2 NX-0928 W MA 0 2/28
690 MAS NX-0928 Sawhill MA 0 2/28 (D
690 MAV NX-0928 Vallinox MA 0 2/24

690 L4V NX-0928 Vallinox MA + 705'C (1300'F)/4 h/AC 0 2/24
690 LV NX-0928 Vallinox MA + 705'C (1300°F)/1 5 h/AC 0 2/24
800 MA H-8705A Superior MA 0 1/38, 1/74
800 L H-8705A Superior MA + 705'C (13000 F)/1 5 h/AC 1 0 1/38,1/74
1Heat treatment performed in laboratory.

FIGURE 24-2
Arrhenius Plot of Average SCC Propagation Rate for MA and
Tr Alloy 600 in 1%, 10%, and 50% NaOH (24.13)
I I
340C 320C 300C 280C
I I I I
650F 625F 600F 575F 550F
1.000 SI i
0.100-
E
Cr
0 o
C-)
C) 0.010 -
0.001 1
1.60 1.65 1.70 1.75 1.80
1 NO0UT
(deg K)
FIGURE 24-3
Effect of Caustic Concentration and Thermal Treatment on the
Stress Corrosion Cracking of Alloy 600 (24.14)
600
I
600'F (316°C) C-rings
150% Y.S.
0.02 2000 Hours 500
400
300 a
0C.) 0
1)
C-)
0.01 I- C-N
-- 200
-- 100
Thermally
Treated
10 50
Sodium Hydroxide Concentration (wt/o)
Above 6% caustic all alloys were susceptible to caustic SCC. Thermally

treated alloy 690 was not included in these early tests, but the re-created
curve given in Figure 24-5 also presents newer results (24.17 24.
24.19) which shows that thermally treated alloy 690 does not crack. At
a lower temperature of 572 0F (300°C), Pathania and Chitty found NG
alloy 800 is more resistant than alloy 600 to deaerated caustic up to
about 16.7% in capsules as shown in Table 24-11 (24.20). Incidence of
cracking increased sharply for MA alloy 600 between 3.8 and 9.1%
caustic.
In general agreement, Cordovi reported in Table 24-12 a high
incidence of cracking for MA, solution annealed, and sensitized alloy
600 in deaerated 1% caustic at 600'F (316'C) (24.21). The table shows
that commercial alloy 800 in the two solution annealed conditions (M
and A) had less cracking but did not crack in the MA and sensitized (L)
conditions. No alloy 690 specimens cracked.
FIGURE 24.4
Caustic SCC Dependency for MA and TT Inconel 600 and 690 (24.15)
°C
0,040 1000
800 E
0.030
'4
C•
400
0.020
400 L)
0.010
200
Test Temperature ('F)
FIGURE 24-5
Recreation of Berge/Donati Summary Curve for SCC in Deaerated
NaOH-350 0 C C-rings Stressed to About Yield
169LTMA 690LTMA
000 - . 690TTT 690TT.
69 TT
AllIoy
...:...........
.....! " -+ . "...:i.:690
ýO T.. •' ..... -............... ... • "i " i " f i ? ................... ..i "i TMA
3500 -
(0
.......
}.........
........ Al loy+-------
i..+i " --!--i--•.-................. 800NG- i
.......... ... . .. .....• -i
E 3000 -
: F 6OLTMA &
(0 2500 - S................. .......... ...........

LO
CO
2000 -
. ...........
...... .......
..
.. . . . .... + 7. i• .................
+.... .+.
.... i+
0)
1500 -
............ OCIEMAT i..+.i
i..................
.......
+....... . . . . . . . .
1000 -
..+.......
............. • .+
...~. . . . . ....
+.......
T+ . S+..+ ...+....,:
0OCIEMAT .+•..v
!,,. : /ot.,MA
ý•
500
-,.+•A
E. ...•+ ...........
-+"•i
0
1 10 100 500
NaOH Concentration (g/l)
Table 24-11
Pressurized Capsule Tests of 600 MA and 800 NG at 572 0 F (3000 C) in
Deaerated Caustic for 250 Days (24.20)
Nominal No. Failed/No. Tested

Hoop Stress NaOH Sanicro 30
MPa kp.Lj g NaOH/kg H20 MA 600 (800 NG)
170 25 40 (3.8%) 0/5 1/5
170 25 100(9.1%) 3/5 1/4
170 25 200 (16.7%) 5/5 0/5
TABLE 24-12
Stress Corrosion Cracking of U-Bend Specimens in Deaerated 1%
NaOH at 600°F (316 0 C) in 56 Weeks (24.21)
No. of
Heat Specimens Cracked/No. Depth of
Treatment (No. of Destructively Attack
Alloy Code Heats) Examined mils (mm)
600 MA 4 (2) 4/4 15(0.381)
M 4 (3) 4/4 6(0.152)
A 5 (4) 5/5 21.5 (0.546)
A+L 7 (5) 5/7 7(0.178)
690 M 6 (3) 0/6 0
A 8 (4) 0/8 0
A+L 7 (4) 0/7 0
800 MA 2(1) 0/2 0
M 8 (4) 1/8 2.5 (0.064)
A 5 (3) 3/5 2 (0.051)
A+L 10(5) 0/10 0
Code: MA = Mill-Annealed
M = Cold-Rolled 40% + 0.5 h/1,850'F (1,010°C)/AC.
A = Cold-Rolled 40% + 1.0 h/1,950-2,050°F
(1,066-1,121 0C)/WQ.
L = 2.0 h/1,251 -F (677°C)/AC.
Yonezawa, et al., studied the effects of chromium and nickel contents

on alloys in 10% caustic at 617'F (325°C) with the results shown in
Figure 24-6 (24.10). The approximate compositions of alloys 600, 690,
and 800 have been marked on the appropriate chromium composition
lines. In contrast to the results of others, the extrapolated alloy 800
composition appeared to be the most susceptible to caustic attack. Alloy
690 is the least susceptible to attack. As part of their study, they
evaluated U-bends of various alloys in 10% caustic at 617'F (325°C) for
500 hours and at 649°F (323°C) for 1,500 hours with the results in
Figure 24-7. Results in all alloys are as expected except for shot peened
NG alloy 800 at 617'F (325°C). All four specimens cracked after 500
hours. None of the shot peened alloy 800 specimens cracked at 343°C
after 1,500 hours. For some reason, alloy 800 was more susceptible to
SCC in 10% caustic at 617°F (325 0C) than at 649°F (343°C).
FIGURE 24-6
Effect of Cr and Ni Content on the Caustic SCC Resistanceof
0.02%C-x%Ni-15 -30% Cr-Fe Alloy in the Deaerated 10% NaOH
Solution at 617°F (325 0 C) for 500 hr (Single U-Bend) (24.10)
o MA(1100l Cx 0.5h W.Q.)

o 0.02%C-x%Ni-1 5 - 30%Cr
1200 0 o:xNi-15%Cr-Fe
e x%Ni-20%Cr-Fe
' x%Ni-30%Cr-Fe
1000
00
800V
600F
0)
00c
400-
• 2O~r
200 V
A I I
I I I
40 50 60 70 80
Ni Content (wt%)
FIGURE 24-7
SCC Resistance in Deaerated 10% Sodium Hydroxide
Nelson and Floreen in Table 24-13 found cracking in 5/5 commercial

alloy 800 C-rings in 10% caustic at 600'F (316 0C) after 49-80 weeks (24.7).
Only 1 of 4 alloy 800 TT C-rings cracked. In the same test, nearly all
MA and TT alloy 600 C-rings cracked in 10-80 weeks. About half of the
MA alloy 690 C-rings cracked in 10-29 weeks but no TT alloy 690 C-rings
cracked, showing that it had the best corrosion resistance to caustic.
TABLE 24.13
SCC Failure Times for C-Rings Exposed to 10% Caustic and 10% Caustic Plus 750 ppm Sulfate at 600°F
(316 0C) (24.7)
Weeks in Tests( 1 )
Caustic + Total Cracked C-Rings
Alloy Heat No. Condition 10% Caustic 750 ppm SO4 <20 Wks >20 Wks
600 NX 0471 MA 22, 22 25, 49 0 4
NX 1054 MA 10,10 10,10 4 0
MA +1300OF (705oC)/l 5 AC (TT) 50, 80 49, 49 0 4
TT + Plastically deformed 20% unloaded
then 13000F (7050C)/2 hrs AC 15,53 49, 49 0 0
MA + 932°F (5000 C)/24 hrs AC 50, 80 49, 49 0 4 CD
690 NX 9780 MA 10, 10 10,10 4 0
NX 9780 same 10, 10 15,15 3 1 MJ
NX 9780 MA + 13000F (705oC)/4 AC 50, 80 49, 49 0 0

NX 9780 same 30, 30 29, 29 0 0
(D
NX 9780 MA + 1300OF (705oC)/15 AC 50, 80 49, 49 0 0
NX 9780 same 30, 30 29, 29 0 0
NX 0928 MA 15,15 29 2 1
NX 0928 MA 30, 30 29, 29 0 0
NX 0928 MA 15, 30 29, 29 1 0
NX 0928 MA + 13000F (7050C)/4 AC 30, 30 29, 29 0 0
NX 0928 MA + 13000F (705°C)/15 AC 30, 30 29, 29 0 0
800 H 8705A MA 50, 80 49, 49 0 4
MA + 13000F (705oC)/15 AC 60, 60 49, 49 0 1(2)
1
Two C-rings, 150% of yield stress for each test.
2
Single specimen exposed for 49 weeks.
0
Unusual cracking susceptibility for NG alloy 800 was also found in

tests by Hickling, Forchhammer, and Effertz in their tests of capsules
presented in Table 24-14 (24.22). Tests in 3.4% caustic (0.84 N) at 662°F
(350 0C) found through-wall cracking both with and without a low
temperature sensitization. There was no cracking of MA 800, as
received, at lower or higher concentrations of caustic. Cracking was
minor in other environments. In the same study, cracking of MA alloy
600 was minor at 590°F (310 0 C) but substantial at 6620 F (350'C). Low
temperature sensitization appeared to reduce the cracking.
Pathania and Cleland studied the effects of various thermal
treatments on the caustic SCC of the nuclear grade of alloy 800 with the
results in Table 24-15 (24.23). Tests of C-rings in 13% caustic at 572'F
(300'C) were at the corrosion potential and at 60 mV' anodic to the
corrosion potential. The anodic corrosion potential increased the SCC.
Several of the thermal treatments reduced the SCC at the applied
potential but at the same time increased the SCC a small amount at the
anodic potential. Limited metallography appeared to show carbide
precipitation at grain boundaries of alloy 800 during the thermal
treatments of the nuclear grade.
In 50% caustic (not deaerated) at 572°F (300'C), alloy 690 shows
the most corrosion resistance in Table 24-16 (24.4). None of the 14
U-bends that were annealed and thermally processed cracked. Only
half of the U-bends that were cold reduced 40% cracked in 3.9 weeks.
Alternate annealing practices included solution annealing and
sensitization. U-bends of solution annealed (Al) alloy 600 cracked
within 2.6 and 3.9 weeks, although the maximum depth of attack varied
considerably. Surprisingly, duplicate samples of cold-worked material
(MA+CR) did not crack in the 3.9 weeks of testing. U-bends of sensitized
material (AI+L and A 3 +L) suffered only minor attack within the test
period. Thermally treated alloy 600 was not included in the test.
In related work, threshold stress intensities above which SCC can
occur in 50% caustic are given in Table 24-17 (24.24). All alloys were in
the susceptible solution annealed condition. Alloy 600 had the highest
value followed by decreasing values for alloy 690, alloy 800 and 304 SS.
Although the data are not extensive, they show consistently that
TT alloy 690 has the most corrosion resistance to caustic solutions. This
is evident even though the thermal treatment for alloy 690 has not been
optimized as it has for alloy 600. There are too little data to decide
whether TT alloy 600 or MA alloy 800 is next best in corrosion resistance.
The data of Berge and Donati (Figure 24-5) indicate that alloy 800 NG
is more corrosion resistant than TT alloy 600 below about 6% caustic,
but other investigations cited in this review have consistently found at
TABLE 24-14
Mean Crack Depths (gm) for Alloy 800 After 3,000 Hours in Various Solutions at 590°F (310 0 C) and 662 0 F
(350°C) (24.22)
590°F (310°C) 662°F (35000)

Material Condition Material Condition
Na+Conc. Soln. Conc. As-Received+LTS at 752'F (4000C) for As-Received+ LTS at 752 0F (4000C) for
CD
NO0. LrnY p_._ Oh 1i000 h 10.000 h 0 hr 1,000 hr. 10,000 hr.
__(1) __ _ -
NaOH 0.42 0.42 13.3 230 62
NaOH 0.84 0.84 13.4 >1,000 495 >1,000
0
NaOH 1.26 1.26 13.5 - .J_(2) -
Na 3 PO 4 0.42 0.14 12.5 44 42 37

Na 3 PO 4 0.84 0.28 12.7 57 42 - 35 41
Na 3 PO 4 1.26 0.42 12.8 28 - 113 42 46
Na 2 HPO 4 0.42 0.21 10.9 22
Na2HPO 4 0.84 0.42 10.9
Na 2 HPO 4 1.26 0.63 10.7 23 - 27
1 = No penetrations >20 4m.

2
No valid result.
TABLE 24-15
Thermal Treatment Versus Maximum Crack Depth for Alloy 800 in 15
% NaOH(1) at 572-F (300 0 C) (24.23)
Thermal Maximum Crack Depth (Itm)

Treatment + 60 mV 0 mV
X( 2 ) X( 3 ) S(4 ) X( 2 ) X( 3 ) X( 4 )
5
Mill-Annealed( ) 95,80,78, 67 20 113,110, 62 57
55,50,45 12.5
850C, 1 sec 26,0,0 8.6 15 0
850C, 10 sec 58, 33,10 30.3 29 0
850C, 60 sec 20, 20, 20 20 0 0 -
850C, 300 sec 25, 25,12.5 21 7.5 0 -
800C, 1 sec 0,0,0 0 0 25 -
800C, 10 sec 0,0,0 0 0 25 -
800C, 60 sec 25, 22.5, 20 22.5 2.5 0
800C, 300 sec 25, 20, 17.5 20.8 3.8 0
700C 10 sec 38, 20,17.5 25.2 11.2 0
700C 60 sec 25, 20, 20 21.7 2.9 0
700C, 300 sec 0,0,0 0 0 0
700C, 3,000 sec 10,0,0 3.3 5.8 0
600C, 300 sec 15,12.5, 0 9.2 8 78 -
600C, 3,000 sec 0,0,0 0 0 46 -
600C 8 h 0,0,0 0 0 30 -
500C, 8 h 19,19,13 17 3.5 43 -
1Cover gas, 10% H2 + 90% N2; exposure time 14 days; applied stress, 470 MPa;
and 4.4% strain.
2
Maximum crack depth per each specimen.
3
Mean of maximum crack depths.
4
Standard deviation.
59850C, 900 sec.
TABLE 24-16
The Effects of Heat Treatment, Cold Work, and Welding on the Stress
Corrosion of Tube Alloys (U-Bend Test Specimens) in Undeaerated
50% NaOH at 572 0 F (3000 C) (24.4)
No. of
Heat Specimens Test Cracked/No. Depth of
Treatment (No. of Duration Destructively Attack
Alloy Code Heats) (wk) Examined mils (mm)
690 MA 2(1) 3.9 0/3 0
MA+L 2(1) 3.9 0/1 0
MA+A 1 2(1) 3.9 0/2 0
MA+CR 1 (1) 1.7 1/1 110 (2.79)
1 (1) 3.9 0/2 0
A1 +L 2(1) 3.9 0/2 0
A3 +L 2(1) 3.9 0/2 0
A2 +W(1) (12.7 mm) 2 (1) 3.9 0/2 0
A 3 +W (25.4 mm) 2 (1) 3.9 0/2 0
800 AI+L 2(1) 1.7 2/2 42(1.06)
MA 2(1) 1.3 2/2 85(2.15)
600 CR+A 1 1 (1) 2.6 1/1 42(1.06)
1 (1) 3.9 1/1 2(0.06)
MA+CR 2(1) 3.9 0/2 0
A1 +L 2 (1) 3.9 2/2 .8 (0.02
A3 +L 2 (1) 3.9 2/2 2 (0.06)
304SS A1 2 (1) 0.4 2/2 120 (3.05)
AI+L 1 (1) 0.4 1/1 120 (3.05)
201 MA 2(1) 3.9 0/2 0
Code: MA = mill-annealed
CR = cold-rolled 40%
A1 = 1 h @2,000°F (1,093°C)/WQ
A2 = 1 h @2,1001F (1,150°C)/WQ
A3 = 1 h @2,200'F (1,205°C)/WQ
L = 2 h @1,200-F (650°C)/AC
1
W (thickness, mm) = manual gas tungsten arc welded at indicated thickness
matching filler.
TABLE 24-17
Threshold Stress Intensities for the Stress Corrosion Cracking
of Tube Alloys in Deaerated 50% NaOH at 6001F (316'C) (24.24)
Ni Content Threshold Stress lntensity

Material1 Wt% MPa-m ksi'inch
Type 304 9 <1.1 <1
Alloy 800 34 14.3 13
Alloy 690 60 22.0 20
Alloy 600 76 30.8 28
1
Heat Treatments: Alloy 600: 1 h/2,050°F (1,121 0 C)/WQ
Other alloys: 1 h/1,950°F (1,065 0C)/WQ
least some cracking of alloy 800 in dilute caustic. There are not enough
data to determine whether NG 800 is more corrosion resistant in caustic
than the commercial grade. Certainly, it is clear that MA alloy 600 has
the least resistance to caustic and that any of the above alloys would be
an improvement.
The low incidence of cracking of alloy 800 tubing in German nuclear
steam generators provokes the question about the degree of alcalinity
in heated crevices in steam generators. If crevices were strongly alkaline
as some predict, more SCC of alloy 800 should occur. The use of
phosphates (Na2HPO4) may prevent the formation of strongly alkaline
solutions. Or the phosphate ion itself may also reduce SCC. If so, then
the switch to AVT water chemistry, which has now occurred in several
German steam generators, could remove this safeguard and may permit
stronger alkaline solutions to form in heated crevices with the possible
increase of SCC of alloy 800. In a similar comparison with alloy 600,
Na2HPO4 which caused wastage may have prevented caustic SCC.
Only time and more crevice solution analyses will answer these
uncertainties.
The SCC of mill annealed and thermally treated alloy 600 C-rings
was studied in 10% caustic with copper oxide. Results of these tests
are given in Table 24-18 (24.13). The results are complex. Without
copper oxide, the incidence of cracking was less for the TT material for
C-rings stressed at 150% of yield. No difference in behavior was
observed for highly stressed TLT C-rings. With 1% copper oxide, all
TT specimens had a low incidence of SCC. All MA C-rings cracked as
before. With 10% copper oxide only MA C-rings at 150% of yield had a
low incidence of SCC. In only one instance (10% CuO, TLT specimens)
is there any evidence that copper oxide increased the SCC of alloy 600.
Copper oxide and other additives influence the corrosion potential of
alloy 600 as shown in Table 24-19 (24.13). It was previously established
(24.25) that alloy 600 is most susceptible to SCC in the active/positive
potential region, which, for the current tests, was in the sharply defined
range of 200 mV (versus Ni). Smaller amounts of copper oxide in the
range of 0.1%, fix the corrosion potential very near this 200 mV value.
In an actual steam generator the copper oxide is essentially outside the
crevice. Its influence on the potential in the crevice is unknown, as is
the effect of copper oxide on SCC.
Compared to 10% NaOH alone, additives of 10% SiO2 doubled the
elongation of MA alloy 600 tubing while 10% copper oxide cut the
elongation in half in slow strain rate tests in Table 24-20 (24.14). The
silica increased the elongation almost to the value obtained in air,
indicating that it largely offset the effect of caustic.
TABLE 24-18
Cracking of Mill Annealed and Thermally Treated Alloy 600
C-Rings in 10% NaOH Plus Copper Oxide at 630°F (3320 C)
(Number Cracked/Number of Specimens Tested 2,000-4,000
Hours) (24.13)
MA __
1501 TLT 2 1501 TLT 2

0 15/16 5/8 1/17 5/8
1% CuO 9/10 5/5 1/10 1/5
10% CuO 1/5 10/10 2/4 8/10
1150 % of yield at C-ring apex.
2
TLT = Two legs of C-ring touch.
TABLE 24-19 t
Summary of Corrosion Potential Measurements 1 in Caustic and
Sludge Component Solutions Versus Ni in Pure 10% NaOH
(24.13)
Addition Temp E Alloy 600 E Ni Exposure

(Wt %/) "C (MY) (MY) 01[r)
Base Solution 288 0 -1 77
10% Fe 3 0 4 288 0 -1 77
10% CuO 288 +415 +480 150
1%CuO 288 +277 +278 126
0.1% CuO 288 +240 +237 168
0.01% CuO 288 0 -1 126
Cu Metal 288 +14[a] +1 3[a]' 480
10% Fe 3 04 + 0.1%CuO 288 +10 +9 311
10% Cr 2 03 288 +12 +9 6
10% Simulated Sludge 2 288 +240 +237 168
1External couple was formed between the test materials and a strip of Cu
with approximately equal surface area. The potential reported is that for
the
2
Cu/alloy 600 or Cu/Ni couple versus the Ni reference electrode.
60%Fe3O4, 33%Cu, 5%CuO, 2%NiO, 1%Cr 2 03
Water. It is still surprising that a corrosion resistant alloy such as alloy

600 has even a small susceptibility to SCC in pure water, AVT water,
and primary water. Hydrogen and cold work accelerate the SCC, as
shown in Table 24-21. A thermal treatment reduces the SCC both in
the presence and the absence of hydrogen. Up to about 25% area
reduction, cold work had little effect on the SCC. Comparison of data
in Table 24-21 and 24-22 indicate that primary water is not as corrosive
as pure water (24.14).
A comparison of alternative materials (creviced U-bend specimens)
in aerated water is given in Table 24-23 (24.4). Alloy 690 shows at least
at ten-fold reduction in SCC compared to alloys 600, 800 and type 304
stainless steel. Alloy 690 specimens had no failures in the 48-week test
period, whereas all other specimens had cracked in less than four weeks.
Crack initiation times for mill annealed and sensitized alloy 600 in AVT
water are shown as an effect of temperature in Figure 24-8 (24.26). Crack
initiation times decrease with increasing temperature and increase with
sensitization.
U
CD
TABLE 24-20
Slow Strain Rate Test Results on Mill Annealed Inconel Alloy 600 Tubing (Heat 9837) in Caustic
Environments (24.14)
Ultimate True
Strain Total Tensile Fracture Reduction
Test Temp. Rate Extension1 Stress Stress Elongation of Area
Environment OF(Q) (sec- 1 1 in. (mm) psi (MPa) psi (MPa) % %
(D
Air 600 (316) 2.5 x 10- 6 0.622 (15.8) 111,800 (771) 168,200 (1,160) 36.6 35
10% NaOH 630 (332) 2.5 x 1 0 -7 0.20 (5.2) 94,300 (512) 64,500 (445) 15.0 17
10% NaOH 630 (332) 2.5 x 10-7 - 81,800 (564) 94,400 (651) 31.2 20
+10% SiO 2
10% NaOH 630 (332) 2.5 x 1 0 -7 0.19 (4.9) 87,400 (603) 63,400 (437) 8.5 32 14
+10% CuO
1Extension rate x time.
0
TABLE 24-21
Summary of SCC Performance of Split-Tube U-Bends of
Alloy 600 in Pure Water at 680°F (3600 C) (24.14)
Alloy 600 Cold Work

Material % Area No Hydrogen With Hydrogen 1
Condition Reduction % Failed Hours % Failed Hours
Mill Annealed
(a) 0 33 6500 90 1000
(b) 25 33 6500 100 1000
(b) 35 83 3500 100 1000
(b) 45 83 3000 100 1000
Thermally Treated
(a) 0 0 9000 0 1500
(b) 25 33 6500 0 1500
(b) 35 83 6500 67 1500
120 psi hydrogen.

(a) = heats 9837, 9930, 9763, 9788, 9854
(b) = heats 0418,1024,1118
Recently, there has been an interest in raising the lithium content

of primary water to reduce radioactive deposits on primary surfaces
such as in the channel head (24.27). Any reduction in these deposits
would reduce the radiation exposure of personnel who must enter the
channel head to inspect steam generator tubes. SCC tests of alloy 600
tubes were carried out in primary water with varying lithium contents
as part of this study (.24.1 24.28). The pH of primary water varies
with the lithium content as expected. But the pH also varies rather
substantially with temperature, as shownin Figure 24-9 (24.29), which
confounds the accelerated testing at higher temperatures. A
temperature increase from 5720 to 680'F (300 to 360°C) increases the
pH 1.4 to 1.5 units. As a result, it is questionable whether the initial
data obtained at 680'F (360°C) in Figure 24-10 (24.13) can be related to
steam generator operation at lower temperature. Nevertheless, the data
show that 7.6 ppm lithium at the beginning of life (BOL) composition
of 1,200 ppm B caused more SCC than 2 and 5 ppm lithium.
Interestingly, a test at 22 ppm lithium in BOL primary water (pH 9.2 at
680'F [360°C]) caused no cracking of any specimen. The 22 ppm lithium
could not be used in a PWR, however, because of other considerations.
TABLE 24-22
SCC Performance of Split-Tube U-Bends of Alloy 600 in Primary
Water1 at 680°F (3600 C) (24.14)
Alloy 600 Cold Work

Material % Area With Hydrogen 2
Condition Reduction % Failed Hours
Mill Annealed
(a) 0 50 4000
(b) 25 67 4500
(b) 35 100 500
(b) 45 100 500
Thermally Treated
(a) 0 0 4500
(b) 25 17 4500
(b) 35 50 3500
(b) 45 67 4500
1
Middle-of-cycle primary water with 500 ppm boron and 11 ppm lithium.
220 psi hydrogen.
(a) = duplicate specimens from heats 9837,9930,9763,9788,9854
(b) = duplicate specimens from heats 0418,1024,1118
An attempt to establish the initiation times for SCC of alloy 600

specimens as a function of ppm lithium and pH gave the results in
Figures 24-11 and 24-12 (24.13). While the data show considerable
scatter, lines representing times to initiate cracking of 50% of the U-bend
specimens show a logarithmic decline with increasing lithium
concentration and pH. The initiation times would be expected to
increase substantially at lower temperatures.
General corrosion rates in high-velocity-flowing primary water with
hydrogen are shown in Figure 24-13 (24.4). The corrosion rate of alloys
690 and 800 are an order of magnitude lower than for alloy 600. These
lower corrosion rates for 690 and 800 should reduce the radioactive
deposits in the channel head.
0 0
TABLE 24-23
Stress Corrosion Tests in Underaerated Water at 600'F (316 0C) to Evaluate the Role of Heat Treatment,
Cold Work, and Welding on the "Oxygen-Plus-Crevice" Effect (24.4)
Specimens
Heat No. of Test Cracked/ Max. Depth
Treatment Specimens Duration Destructively of Attack
Alloy Code (No. of Heats) "wk) Examined mils (mm)
690(1) MA 2(1) 48 0/2 0
MA+L 2(1) 48 0/2 0 -i
MA+A 1 2(1) 48 0/2 .0

MA+CR 2(1) 48 0/2 0
A1 +L 2(1) 48 0/2 0
A3 +L 2(1) 48 0/2 0
0-
A2 +W( 2 ) (12.7) 4(1) 48 0/4 0
A2 +W (25.4) 2(1) 48 o/2. 0
MA+W (3.8) + L1 2(1) 48 0/2 0
600 A1 +L2 2(1) 2 2/2 102 (2.6)
(D
800 Al +L2 2(1) 4 2/2 79 (2.0)
304SS AlI+L2 2(1) 4 2/2 118 (3.0)
Code: MA = mill=annealed A3 = 1h @2,200'F (1,205°C)/WQ

CR = cold-rolled 40% L = 2h @1,200-F (650°C)/AC
A1 = 1 h @2,000'F (1,093°C)/WQ L1 = 4h @1,1 00'F (593°C)/AC
A2 = 1 h @2,100°F(1,150'C)/WQ L2 = 5h @1,300'F(704'C)/AC
1 Double U-bend (crevice) test specimens.
2W (thickness, mm) = manual gas tungsten arc welded at indicated thickness matching filler.
FIGURE 24-8
Crack Initiation Time Versus Temperature (24.26)
(Steam)
371
360
Test'
Terminated
343 ATest
332 Terminated
(Steam) 927°C Anneal & Severe Pickle

316 -0 A As Above & 600*C/4 Hours
E Test
Terminated
U)
28 -
Test
e " 2Terminate(1
0 6 10 is 20 25 30 35 40 45 50 55 60 65
Crack Initiation Time (Hr X 1000)
FIGURE 24-9
Calculation of High Temperture pH in Primary Water Containing
2,300 ppm B As Boric Acid (24.28)
WWIA~ i i
0
9.0o
3600C
8.0+ 330'C
I-
300°C
------------------
7.0+
6.0+
b.u I i i i i I
1 2 3 4 5 6 7 8 9 10
U Concentration, ppm
- 360C
FIGURE 24-10
Cumulative % With SCC Versus Li Concentration in BOL
Primary Water (24.13)
Cumulative Probability of SCC Initiation

99 1 1 /
98
95
U
U
03
40
E
U 3D
1OD 1000 1000D
Exposure Time (Hr)

0 2.0 ppm Li, 1200 ppm B, 360C
X 5.0 ppm Li, 1200 ppm B, 360C
07.6 ppm Li, 1200 ppm B, 360C
0 22 ppm Li, 1200 ppm B, 360C
Sulfates. From about neutral to strongly alkaline environments,

available data indicate that sulfates do not contribute to the SCC of
MA alloy 600 tubing. Nelson and Floreen found no effect of 750 ppm
sulfate in 10% caustic compared to 10% caustic alone (24.7). Airey found,
however, that both MA and TT alloy 600 were susceptible to attack in
acidic sulfate solutions as shown in Table 24-24 (24.14). The thermally
treated tubing was more resistant to cracking in weakly acid solutions.
For some specimens cracking increased with the strain; cracks were
deeper on TLT specimens than those stressed only 150% of yield.
All metallurgical conditions of alloy 600 experienced SCC in a model
steam generator faulted with 36 ppm sulfate and 3 ppm chlorides as
indicated in Table 24-25 (24.30). Sensitized alloy 600 had the highest
incidence of cracking. Mill annealed and process-stabilized alloy 600
with the least grain boundary carbides had the lowest SCC. There was
no SCC of MA 690 and NG 800. All cracking in this test was
FIGURE 24-11
50% SCC Initiation Time for Heat 1019 MA Versus Li Concentration
in 680'F (3601C) Primary Water. The error bars denote a •1 standard
deviation range corresponding to the times for 16% SCC initiation
and 84% initiation. (24.13)
* Beginning of Fuel Cycle Chemistry

* End of Fuel Cycle Water Chemistry
i 2 3 4 5 6 8 9 10
Li (ppm)
circumferential. Presumably due to a longitudinal stress which was

inadvertently and probably unevenly applied to tubes frozen in the
first support plate held by shrouds. The shrouds were raised off the
tubesheet by magnetite formation in the crevice between the shroud
base and the tubesheet, applying an axial strain on the tubes. The high
incidence of cracking of thermally treated alloy 600 tubing in a mildly
corrosive environment may have been due to the fact that the tubing
did not adequately respond to a thermal treatment.
FIGURE 24-12
50% SCC Initiation Time for Heat 1019 MA Versus High Temperature
pH in 680°F (3600) (24.13)
Beginning of Fuel Cycle Chemistry

* End of Fuel Cycle Water Chemistry
1=-
8.0 8.1 8.2 8ý3 8.4 8.8 8.6 8.7 8.8 8.9 9.0
pH at 680 deg F
Newmann found that a sensitized alloy 600 rod was more

susceptible to SCC over a wide potential range compared to a non-
sensitized (quenched) rod in acid solution as shown in Figure 24-14
(24.31). The sensitized alloy had a low susceptibility to SCC only at or
below the corrosion potential. There was evidence of sulfate reduction
to sulfide in one of the acid tests.
Sulfur. Thiosulfates, polythionates, polysulfides, and other anions are
sources of soluble sulfur. Because reduced sulfur species have an
appreciable volatility at elevated temperatures and low pH, cracking
by sulfur may primarily be a low temperature problem in the secondary
system. When volatiles cannot escape, as in the primary system, high
temperature SCC due to soluble sulfur is also a possibility.
FIGURE 24-13
Effect of Cromium Content on Material Lost to Stream in 600.8 0 F
(316 0 C) Deaerated Water Flowing at a Velocity of 5.5 m/s (24.4)
Alloy
600
14
12
Ell- 10
E28
-R
.M 4
CU
2 2
0
20 25
Chromium Content (%)
Studies of alloy 600 at low temperature have shown that increasing

sensitization is associated with increasing vulnerability to SCC in
polythionic acids as summarized in Figure 24-15 (24.32). A thermal
treatment for one hour at 1,292TF (700'C) for the one specimen should
provide near maximum sensitization (24.33). It is also apparent in the
figure that increasing sensitization decreases the minimum stress level
and the time for cracking. In a similar test alloy 690 showed no SCC
(24.4).
Lead. Some time ago (24.34), it was found that traces of lead or lead
compounds promoted the SCC of alloy 600. As part of the study,
examination of an alloy 600 specimen cracked by Coriou in pure water
found small amounts of lead in and around the crack. Sulfur, another
suspected cause of the cracking, was also found outside the crack. Both
lead and sulfur could cause cracking due to the formation of low melting
compositions that can penetrate metal grain boundaries. SCC due to
lead has remained a concern. A summary of recent test results for alloy
600 and alloy 690 in various lead containing environments and pH
values is shown in Table 24-26 (24.35). The results show that alloy 690
is generally more corrosion resistant than alloy 600. However, alloy
690 becomes subject to a wastage attack and a mixed mode intergranular
and transgranular SCC at high values of pH. Yet, alloy 690 TT generally
has equal or greater resistance to degradation than alloy 600 TT.
Additional test data obtained in high temperature water (24.36) indicate
that alloy 800 is also resistant to cracking by lead and lead compounds.
TABLE 24-24
SCC Performance of 630OF (3320 C) Alloy 600 C-Rings in Sulfate
Environments (24.14)
Na 2 SO 4
Conc. Initial Exposure Stress SCC Depth (1rm) 1
2
Test (pprnJ pH Time (hr) Levels MA TT
1 80,000 10 6,000 150 <10 <10
80,000 10 6,000 TLT <10 <10
2 80,000 7 4,000 150 <10 <10
80,000 7 4,000 TLT <10-50 <10
3 80,000 3 5,000 150 <10-350 <10
80,000 3 5,000 TLT 250-500 <10
4 80,000 2 4,000 150 200-350 30-300
80,000 2 4,000 TLT 250-1,000 <10-250
5 10,000 3 4,000 150 300-350 <10
10,000 3 4,000 TLT 200-800 50-330
1MA= Mill annealed.

TT = Thermally treated at 1,300°F (704oC) for 15 hours.
2150 = C-ring stressed to 150% of yield stress (-0.3% plastic deformation).
TLT = C-Ring stressed until the legs touched (-14% plastic deformation).
TABLE 24-25
Summary of Stress Corrosion Cracking of Tubes in a 19-Tube Model
Steam Generator FaultedI With a Simulated Cooling Tower Water
and Sludge (24.30)
Material Number Cracked/Number Tested

Mill annealed alloy 600 1/4
Thermally treated alloy 600 2/4
Sensitized alloy 600 3/3
Processed stabilized alloy 600 1/3
1
Water was faulted to 36 ppm sulfate, 3 ppm chloride, and a copper-containing
sludge.
FIGURE 24-14
Stress Corrosion Susceptibility As a Function of Potential for
Sensitized and Non-Sensitized Alloy 600 (24.31)
1.0
W
0
-0
0)
V 0.5
ia / Sensitized
Alloy 600
0)i
/
CO /
Non-Sensitzed Alloy 600_.

. . .. . . . . . .
I I I I
-400 t -200 0 200 400 600 . 800

Free Corrosion Potential, .V (SHE)
Potential
FIGURE 24-15
Cracking Time in Polythionic Acids for Alloy 600 C-Rings With Two
Different Heat Treatments (24.32)
400
104-F (40-C)
50
350
40
l 280
0 210 -30,
140 20
70 1 1 110
0 10 30 100 400
lime (Hr)
I MA (little sensitized during cooling) II MA + 1hr 1,292 0 F (700'C) (highly sensitized)
Intergranular Corrosion. Currently, several environments are suspected

of causing intergranular corrosion (IGC) or intergranular attack (IGA): (a)
caustic and modified caustic environments, (b) acidic sulfate attack (with
possible sulfate reduction) of sensitized tubing, and (c) near neutral
environments with lead compounds.
Until recently, most emphasis was placed on caustic because of the
widespread occurrence of IGAin laboratory caustic tests. Recent summaries
of IGA rates in caustic for stressed and unstressed specimens are given in
Figures 24-16 and 24-17 (24.13). The rates are shown as a function of caustic
concentration and temperature.
Examinations of tubes with IGA that were removed from steam
generators have provided some evidence of a strongly caustic environment
as explained more completely in Chapters 11 and 13. As a result, the search
for causes has been broadened. The possible association of sulfate wastage
and IGA in some steam generators suggests that the acid attack of sensitized
tubing cannot be eliminated as a possible cause, at least for the OTSGs.
Current on-going work by Cullen (24.37) is expected to shed considerable
light on IGA and SCC of both alloy 600 and alloy 690 in acidic environments.
In addition, because lead has been detected on some tubes removed from
steam generators, it also is suspected to be a cause or contributing cause of
IGA or SCC. However, the morphology of corrosion damage caused by
lead, particularly in laboratory tests (24.35), appears to influence SCC more
than IGA. Nevertheless, it is possible that lead compounds act synergistically
with dilute caustic to cause IGA.
TABLE 24-26
Maximum Depth of IGA or SCC Attack Observed in Tests of C-Rings With 3% Strain Exposed to 615.20 F
(3240 C) Liquid for up to 4,073 Hours (24.35)
pH 77 0F (250Q Alloy 600 MA( 1)

3 aD
Environment Alloy 600 TTr( 2 ) Alloy 690 TT( ) (n
0.1m PbSO4 + 0.1n NaHSO 4 1.7 0 0 0
0.1m PbSO4 + 0.1n NaHSO 4 - 1.7 0(4) 0(4) 0
0.3m H3 BO3
Morpholine + 0.1n PbSO 4 + 1.7 0(4) 0(4) 0
0.1m NaHSO 4 + 0.3M H3 B0 3
0.2m PbC12 3.9 11% IGA 0 0
rD
AVT + 0.1m PbC12 + 0.1m PbO 4.1 16% IGA 0 0(5)
w
O(4) 0
AVT + 0.1 m PbSO 4 6.6 0 0
AVT + 0.1 m Pb(H2 BO3 )2 7.4 84% SCC slight IGA 0
Morpholine + 1.0m NaOH 8.6 46% IGA/SCC 48% SCC 0
+ 0.1n PbO + 3.0m H3 B0 3
AVT + 0.1m PbO 9.9 83% SCC 0 0
6 0(5)
AVT + 0.1in PbSiO 3 10.2 50% IGA/SCC 69% M( )
5
1.0m NaOH 13.0 0 slight IGA slight( )
1.0n NaOH + 0.1n PbS .13.0 5% IGA/SCC 0 0
6
1.0n NaOH + 0.1n PbO 13.0 60% IGA 16% IGA/SCC 1,2% M( ) •
1
Mill annealed 1,700.6 0 F (9270C). 4
Pitting observed.
2 5
Thermally treated 1,700.6 0 F (9271C) + 1,299.2'F/1 6 hr (7040C). Wastage.
3
Thermally treated 1,981.4 0F (1,0830C) + 1,301-1,328°F/hr (705-7200C). 6
Mixed TG/IGSCC.
FIGURE 24-16
Arrhenius Plot of Average IGA Rate for Stressed, Mill Annealed
Alloy 600 in 1%, 10%, and 50% NaOH (24.13)
340C 320C 300C 280C
65tF 62FSF SOOF' 5715F 551OF
CO
5 1.70 1. 80
1 000ET (deg K)
Until the plant cause or causes of IGA are clearly established, it will be
difficult to compare the corrosion resistance of the candidate tube alloys to
IGA attack more than in a very general way. In all of the above
environments, however, TT alloy 690 appears to have superior corrosion
resistance, except in the case of caustic plus lead.
Pitting. Pitting has been associated with copper oxide intrusion into steam
generators and may occur in sludge piles, in crevices, or under scale. Causes
of pitting are discussed more extensively in Chapter 10. Small amounts of
pitting were found on tubes in a model steam generator faulted with about
30 ppm seawater and a copper-oxide-containing sludge (24.15). There was
less pitting on tubes in a companion model steam generator faulted to the
same equivalence with a simulated cooling tower water (about 36 ppm

sulfate and 3 ppm chloride) (24.32). A copper-oxide-containing sludge was
added to both models. Comparative pit depths on tube materials in the
seawater-faulted model are given in Table 24-27. Pits formed nearly
uniformly on all alloys. There was no clear difference in the pitting resistance
of any of the alloys. The pits were shallow and it is possible that differences
might have occurred as the pits increased in depth.
Yonezawa, et al., (24.10) reported differences in pitting resistance for
deeper pits as shown in Figures 24-18 and 24-19. MA alloy 600 had the
deepest pits, while TT alloy 690 and shot peened nuclear grade alloy 800
had only shallow pits.
Based on this limited data, it appears that alloys containing
molybdenum should also be considered for sleeves and other applications
where pitting is important.
FIGURE 24-17
Arrhenius Plot of Average IGA Rate for Unstressed, Mill Annealed
Alloy 600 in 1%, 10%, and 50% NaOH (24.13)
^^^
1.n()
.vv 1 I
jx 10%Na
0.100+
E '4
'2)
('2
0.010+
0.001-
1.60 1.65 1.70 1.75 1.80
1I00/T (deg K)
TABLE 24.27
Depth of Pits on Tube Alloys Exposed to a Seawater-Faulted
Environment in a Model Steam Generator (24,8)
Material Bundle, in. (mm) Sludge Pile, in. (mm)

Condition Avg. Depth Max. Depth Avg. Depth Max. Depth
A600 MA 0.001 (25) 0.009 (229) 0.003 (76) 0.011 (279)
A600-TT 0.001 (25) 0.006 (152) 0.003 (76) 0.010 (254)
A600 PS 0.003 (76) 0.007 (178) 0.003 (76) 0.013 (330)
A600 Sens. 0.002 (51) 0.007 (178) 0.003 (76) 0.008 (203)
NG A800 0.001 (25) 0.002 (51) 0.004 (102) 0.013 (330)
A690 MA 0.001 (25) 0.003 (76) 0.003 (76) 0.013 (330)
MA = Mill Annealed
TT = Thermally Treated
PS = Process Stabilized
Sens. = Sensitized
NG = Nuclear Grade
Wastage. Wastage or general corrosion of alloy 600 due to phosphates

was an early corrosion problem in steam generators. Typically, wastage
was found near the top of the sludge pile. When it appeared that phosphate
wastage could not be controlled without increasing the pH excessively and
the possibility of caustic SCC, most utilities switched their feedwater
chemistry to an all volatile treatment (AVT) (24.32). Elimination of
phosphates in the feedwater and the steam generator gradually stopped
the wastage. Available data on wastage of alternative tube materials in
phosphates are presented in Table 24-28 (24.38). For a Na/P ratio above 2,
phosphate corrosion of all alloys in Table 24-28 even at 617'F (325°C) is
relatively low in the absence of oxygen, copper oxide, and heat flux;
corrosion increased with phosphate concentration and temperature, as
expected. From these data, wastage due to phosphates in the absence of
oxidizing agents would not be expected to be a problem. The corrosion
resistance of alloy 600 in these tests appeared to be slightly greater than
that for alloys 690 and 800.
FIGURE 24-18
Pit Depth After Exposure in High Temperature Water Containing
Chloride Medium for 3,000 Hours (24.10)
900 -
800 -
E 700 "
•- 600-
0
500-
E 400-
E 300-
: 200-
100-
nfl
MA TT A 6T SP
Alloy 690 Alloy 660 Alloy 800L
TEST CONDITIONS
* Temperature -2880
" Cl- -100 ppm (CuCI2 [25 ppm as ClI]
+NiCl 2 [50 ppm as Cl-+ seawater
[50 ppm as Cl-])
" D02 -7-8 ppm
" pH -4.2
" Exposure Time -3,000 hrs
0
FIGURE 24-19
Pit Depth on the Surface of Heated Crevice Area for Test Materials
After Model Boiler Test (24.10)
Tube Support Test

Plate Model-Tb
(405S.S) \ Tb
29
Hot Leg Cold Leg

Primary
Water
TEST CONDITIONS
" Temperature of Primary Side -3400
* Temperature of Secondary Side -289°C -
" Heat Flux -23xl 04 Kcal/m 2 h
" Water Chemistry -AVT4>AVT+seawater (2 ppm as CI)
(750 hrs) (2,250 hrs)
TABLE 24-28
Corrosion of Various Alloys in High Temperature Sodium Phosphate Solutions (24.38)
C~)
2 1
Test Environment Descaled Weight Loss Per Year, mils (mgq/din )
Concen- Max.
tration Temp Exposure
Na/P (Molality) (63) Time, Alloy 600 Alloy 690 Alloy 800 304 SS
1.35 6.0 325 315 0.32 (680) 0.75 (1600) 0.29 (620) 1.0 (2139) CD
1.6 6.0 325 480 1.3 (2800) 2.6 (5600) 3.0 (6400) 2.8 (5989)
2.0 5.0 325 504 0.24(510) 0.28 (600) 0.28 (600) 0.28 (599)
2.2 0.18 325 504 0.040 (86) 0.020 (43) 0.020 (43) 0.12 (257)
2.3 5.0 325 451 0.22 (470) 0.22 (471) 0.56 (1200) 5.2 (11123)
1.6 6.0 275 1344 1.1 (2400) 1.4 (3000)
2.3 5.0 275 1226 0.040 (86) 0.07 (150) 0.12 (260)
2.3 0.4 275 1560 0.25 (540) 0.16 (340)
2.6 0.15 275 1176 0.020 (43) 0.47 (1000) 0.046 (100)
12,140 mg/dm 2 is equivalent to about 1 mil thickness.

Sulfate wastage was found in a model steam generator that was faulted
with 36 ppm sulfates, 3 ppm chloride, and a copper-oxide-containing sludge
(24.39). The sulfate wastage was located below the first support plate in
the model steam generator. The wastage may have occurred largely towards
the end of the test when sludge buildup and fouling of the support plate
were significant. During this period, the seawater-faulted model had to be
shut down and lanced to increase secondary water circulation. The wastage
data, summarized in Table 24-29, show that all metallurgical conditions of
alloy 600 are corroded more than alloys 690 and 800 (24.30). Several plants
have now experienced wastage on the cold leg side of steam generators
which is believed to be due to sulfates (24.40).
Baum recently presented results (24.41) of single-tube model boiler tests
of alloy 600 MA performed using various combinations of ammonium and
sodium sulfates and chloride compounds, with and without copper or
copper bearing sludges. Accelerated wastage was produced under
conditions analyzed by MULTEQ to be a result of low crevice pH. Wastage
occurred with or without copper present. Intergranular attack or stress
corrosion cracking was not observed in these test conditions. In autoclave
tests of alloy 600 and 690 in sulfate solutions, Cullen noted that wastage
appears to be a stronger function of sulfate concentration than of pH (24.41).
A summary of these results, given in Table 24-30, show that there is
surprisingly little difference between alloys 600 (MA and TT) and 690 TT.
These tests did not include alloy 800, but it is judged that its behavior would
be similar. Based on model boiler test results reported in Reference 24.39,
alloy 690 TT and alloy 800 NG are considered equivalent (Table 24-30).
Discussion of Tube Alloys

A relative ranking of the corrosion resistance of alloys 600, 690, 800,
and stainless steel alloys in the several environments discussed is given in
Table 24-31. Relative rankings are attempted for corrosion issues of stress
corrosion cracking, intergranular corrosion, pitting, and wastage.
Environments chosen include acid, neutral, and alkaline conditions. There
are several cases where data are not available and estimates are indicated.
Thermally treated alloy 690 appears to be the most corrosion resistant
tube alloy. It is already far better qualified than was either alloy 600 or
alloy 800 when they were first selected for use in steam generators.
Nevertheless, the environmental condition where thermally treated alloy
690 may not have optimum stress corrosion cracking resistance is a highly
alkaline environment containing lead. French studies have shown that
thermal treatment at 1,292°F (700'C) improves SCC resistance in caustics if
the anneal is greater than 1,940'F (1,060°C) (24.18). It might still be desirable
to determine if this is true for Pb-contaminated caustics.
TABLE 24-29
Corrosion Thinning of Heat Transfer Tubes Due to 36 ppm Sulfates in
a Model Steam Generator Test at 541 'F (283°C) (24.30)
Max. %
0
Tube Wall
Tube Material Support Location/Material Thinning( 1 )
A600 Mill Annealed
Tube 10(2) HL DSp(3 )A690 MA Bushing 37
Tube 2(2) CL DSPType 405SS Support 26
Tube 2(2) HL Type 347SS Eggcrate 7
Tube 12(4) HL Type 347SS Eggcrate 9
A600 Sensitized
Tube 17 CL DSPType 405SS Support 5
Tube 6 CL DSPCS Bushing 43
Tube 17 HL Type 347SS Eggcrate 15
A600 TT
Tube 18 HL DSPSCR-3 Bushing 13
A600 PS
Tube 19 CL DSPType 347SS Bushing 2
A690
Tube 11 HL DSPCarbon Steel Bushing 5
A800
Tube 9 HL DSPType 347SS Bushing 1
1
Tubes 0.75 inch OD x 0.042 wall thickness.
2
Heat No. NX1056
4
Heat No. 74836
3
HL = hot leg, CL = cold leg, DSP = drilled support plate.
TABLE 24-30
AmountsI of Wastage in 100 hrs for Various Forms of Alloys 600 and
690 (24.41)
Equiv.
Sulfate pH at
Test # Ratio 315 0C 600 LTMA 600 HTMA 600 TT 690 TT
59A 0.67 2.70 3.47 3.01 2.38 3.26
60 0.16 3.15 0.48 0.69 0.62 0.71
61A 1.00 1.20 9.17 8.92 8.84 9.62
62A 0.00 3.67 -0.01 0.02 0.00 0.01
63A 0.00 2.67 0.04 0.02 0.01 0.01
64 0.17 2.22 1.50 1.63 1.60 1.72
65 0.01 2.65 0.10 0.09 0.13 0.11
66 0.01 2.84 0.09 0.11 0.09 0.11
1
Amounts are given in percent of original test specimen weights.
It is questionable whether thermally treated alloy 600 or nuclear grade

alloy 800 is the next best choice for corrosion-resistant tubes. If lead is a
potential corrodent, then the choice could be alloy 800. If acid chlorides or
caustic are potential environments, then alloy 600 TT may be a better choice.
Data on the corrosion resistance of alloy 800 NG in acid chlorides are limited.
In alkaline environments, there appears to be no problem of chloride
corrosion of commercial or nuclear grade of alloy 800. If lead, acid chloride,
and caustic environments may be present, then alloy 690 TT should be
considered.
All of the above alloys are superior to MA alloy 600 in pure and primary
water. NG alloy 800 has not failed from primary water cracking after up to
about 12 years of service (24.42). In fact, the failure rate of NG alloy 800
from all causes has been exceptionally low as indicated in Table 24-32.
During the two-year period, the percent of alloy 800 tubes that developed
defects was 0.005% due only to wastage, compared to 0.217% for alloy 600
from many causes.
The steam generator service experience for TT alloy 600 has been very
favorable and goes back for more than ten years to the replacement
generators at Surry and Turkey Point. These units replaced steam generators
in the 1980-81 and 1982-83 time periods, respectively. The replacement steam
generators were of a model F-type design which have used hydraulic
expansion to remove tube-to-tubesheet crevices and type 405 stainless steel
TABLE 24-31
Relative Ranking of the Corrosion Resistance of Alloys 600, 690,
800, and Stainless Steel
Corrosion Issue Alloy 600 Alloy 600 Alloy 690 800

(freely corroding) MA -- TT NG SS
Stress Corrosion Cracking
Chloride
" Acid 1 1 1 2-3 (4)
" Acid+S04 (trace) 2-3 2 1 (1-2) (3)
* Neutral (or AVT) 1 1 1 2 5
Caustic
* Below 6% 3 2 1 2 5
* 10-50% 4 2 2 3 5
Water
" Pure, primary, 4 2-3 1 1 1
and AVT w/H2
* Pure w/02 4 3 1 (2) 5
Sulphur compounds
" Acid sulphate 3 1 2 5
" Alkaline sulphate 1 (1) (2)
" Acid reduced 3-4 2 (1) (2) (4)
compounds
" Alkaline reduced 2 (3) 3 3 (5)
compounds
Lead
" Acid 4 4 1 3 (4)
" Neutral (or AVT) 3-4 3 1 1 1
" Alkaline 3-4 3 4-5 5 5
Intergranular Corrosion
Acid 3-4 2-3 1-2 (4-5) (4-5)
Alkaline 4 2 1 4 (5)
Pitting in Chlorides 3-4 3-4 2 3 4
Wastage
Phosphates 3 (3) (2) (2) 3
Sulphates 3 3 2 2 (3)
Rankings: 1 - Best; 5 - Worst; 0 Estimates; U - Unknown

Code: MA = Mill Annealed; TT = Thermally Treated; NG = Nuclear Grade;
SS = Type 304 Stainless Steel
broached hole tube supports. Two TT alloy 600 tubes with top of tubesheet
eddy current indications were pulled (from Surry) and examined after 9.5
years of operation (24.43). There was no evidence of corrosion degradation
on either tube. Most of the tube plugging in TT alloy 600 steam generators
has been associated with fretting and wear at antivibration bars (24.44).
The steam generator experience with alloy 690 tubing, while in total is
less than TT alloy 600 or NG alloy 800, is still appreciable. Four MA alloy
690 implanted tubes have been in service at Farley 1 and Diablo Canyon 1
since August 1977 and April 1984, respectively (24.45). Both sets of implants
have received periodic eddy current inspection and have no indication of
tube degradation. Full bundles of TT alloy 690 have been in service at
Cook 2, Indian Point 3, and Ringhals 2 since 1989 with no indication of
corrosion degradation in any unit. Since these three original TT alloy 690
replacement bundles were installed, bundles have gone into service at
Dampierre 1, Millstone 2, North Anna 1, and Ohi 3 and 4.
Tube Support and Tubesheet Materials

Corrosion of tube support plates of steel, although a potentially serious
problem, has actually been more of a nuisance. The purpose of support
plates is to prevent tube vibration as explained more completely in Chapter
16. Corrosion of support plates has been a problem in the heated crevices
between the support plate hole and the supported tube as discussed in
Chapters 8, 11, and 12.
Briefly, support plate corrosion in crevices can cause two types of
damage: denting of the supported tubes and fragmentation of the support
plate. Denting (described in Chapter 8) occurs when. corrosion products,
which have a larger volume than the metal from which they are formed,
accumulate in the crevice. After the crevice gap is filled, the tube is dented
inwardly, eventually preventing internal inspection of the tube. As a result,
the tube is usually plugged although there may be no loss of its integrity.
Fragmentation of the support plate may occur when corrosion extends
through the narrow ligaments between the tube holes and the flow holes or
between tube holes. Although it apparently has not happened, star-shaped
pieces of support plate with sharp edges could be separated as shown in
Figure 24-20, and damage the tubes due to motion from the rapidly moving
steam and water (24.46). The piece in Figure 24-20 came from a section of a
support plate that was removed from a steam generator for examination.
Eddy current measurements indicated only a few mils of denting. The
corrosion products apparently wedge the fragments together so that the
structure has remained intact. With more open lattice supports fragments
may be more of a problem.
TABLE 24-32
Experience With Steam Generator Tube Materials in 1983 and 1984 (24.42)
1983 1984 CD
No. of % of No. of % of
2
Tube No. of No. 1 of Tubes Tubes No. of No. of Tubes Tubes Failure '-1
0D
Material Reactors Tubes Plugged Plugged Reactors Tubes Plugged Plugged Mecha-
nism
Alloy 600 106 1,394,262 3,110 0.217 117 1,447615 3,249 0.187 SCC/IGA, M
W, D, Fr,
P, F, E
Alloy 800 13 164,678 175 0.106 14 178,878 81 0.045 W
Stainless 5 81,090 6 0.007 6 87,738 5 0.006 Fr
Steel
Monel 400 9 210,606 0 0 11 257,082 0 0
1
KWR Obrigheim is included in both Alloy 600 and Alloy 800 categories since both sets of steam generators were serviced
in 1983.
2SCC = Stress Corrosion Cracking, IGA = Intergranular Attack, W = Wastage, D = Denting, Fr = Fretting, F = Fatigue
P = Pitting, E = Erosion
FIGURE 24-20
"Star-Shaped" Ligament Section: (a) Cut Surface, (b) Fracture
Surface. Support Plate 1, Steam Generator 23, Indian Point 2 (24.46)
b Tube Hole a
Flow Hole
bb
In new and replacement steam generators, more corrosion resistant

support plates are needed. Nominal compositions of candidate support
alloys are given in Table 24-33 24.47). They include carbon steels, stainless
steels and nickel base alloys. The descending order has an increasing alloy
content. In general, the higher the alloy content, the greater the corrosion
resistance. In the following discussion, the corrosion resistance of the various
alloys will be compared to carbon steel. Then, based on preliminary
estimates of a few types of corrosive environments in steam generators,
minimum alloy content for supports will be matched to the environment
for a 40-year service life.
Corrosion tests on support materials were carried out in model steam
generators, model boilers, capsules, autoclaves, and a once-through loop
with multiple electrochemical test cells (24.48, 24.49). All tests operated in
the temperature range of 482-572°F (250-300'C).
Corrosion of Support Materials in Chlorides. Chlorides alone catalyze,

but do not cause support plate corrosion; oxidants are also needed (21.,
24.51). Oxidants consisted of oxygen, copper (II), nickel (II) and acid.
Corrosion tests in chlorides were carried out in a model steam generator
and in miniature autoclaves.
U'
TABLE 24-33 c-fl
Hardness and Nominal Composition of Tubesheet Steel and Support Alloys for Steam Generators (24.47)
Chemical Composition (%)1

Code Type Hardness C Cr Mn Mo Ni SSi Other
C
1018 Carbon Steel 66 RB 0.2 0.7 - - - -t
A508 Tube Sheet 0.2 0.1 1.4 0.5 0.7 0.24

-t
Steel
4130 Cr-Mo Steel 20-25 RC2 0.3 0.9 0.5 0.3 0.1 0.27
409 Ferritic SS 75 RB 0.08 11 1.0 1.0 Ti - 6xC C
C
410 Martensitic 0.13 12 0.53 0.13 .17 .33
405 Ferritic SS 75 RB 0.08 13 1.0 1.0 0.1-0.3 A1
430 Ferritic 0.05 17 0.49 -- .17 .38
347 Austenitic SS 85 RB 0.08 18 2.0 -- 11 1.0 Cb+Ta= 1OxC
439 Ferritic SS 78-80 RB 0.03 18 0.4 - 0.2 0.04 Ti = 12xC
SCR-3 Austenitic SS 76.5 RB 0.03 25 1.5 - 25 1.8 1.5V, 0.3Ti
800 Hi Ni-Aust. SS 74 RB 0.06 20 0.9 - 32 0.2 0.5 Al, 0.5Ti
600 Nickel Alloy 82 RB 0.03 15.5 0.2 - Bal 0.2 10 Fe
1Balance iron.
2
Annealed condition.
Model 9 Steam Generator. Krupowicz, et al., used two identical 19-

tube model steam generators (MSG) constructed with various support
alloys and alternative support designs for corrosion tests (24.48). Model
9 was faulted with seawater to about 30 ppm chloride for 285 days with
a chloride/sulfate weight ratio of about 48, while Model 10, discussed
below, was faulted at nearly the same sulfate concentration equivalent
to the chloride above and a sulfate/chloride ratio of about 13 for 358
days. Both model steam generators operated at 541'F (282°C) and
1,070 psi (7.38 mPa). Ten grams of sludge containing 30% copper oxide
was added every 8-hour shift. The crevice heat flux varied from 36,000
to 70,000 Btu/hr/ft 2 (110-220 kW/m 2 ) on the cold leg to 38,000 to 98,000
Btu/hr/ft 2 (120-310 kW/m 2) on the hot leg. The secondary loop was
pressurized preventing air inleakage. Maximum corrosion of support
materials, largely on the hot leg, is presented in Table 24-34 (24.39).
TABLE 24-34
Corrosion of Support Structures in Model Steam Generator
10 Faulted With Simulated Cooling Tower Water ABND
Sludge (About 38-ppm Sulfate Present for 358 Days of,
Operation)l (24.39)
Max. Corrosion
mm (mils) Comments
Carbon Steel Bushings 2,000 (7) in one location
405 SS Support Plate 254 (10) 30-mil cracks
409 SS Lattice 1,100 (43) pits expanded to
crevice corrosion
347 Bushings, Lattices 250 (10) corrosion and
shallow cracks
Alloy 600 Bushings, Lattices shallow Pits
SCR-3 Bushings, Lattices shallow Pits
Alloy 690 Bushings nil
113,580 ppm-sulfate days, 1,060 ppm-chloride days, 12.9 ppm copper
days, 10.7 kg sludge containing 30% copper oxide added.
A destructive examination of Model 9 (24.8) after the test found

that carbon steel bushings, mounted in a 405 SS support plate, had
corroded through the 120 mil thick (3,050 gtm) wall in many places, as
shown in Figure 24-21 with only a few mils of denting. Apparently
much of the dissolved iron did not form oxide locally in the crevice
causing extensive denting.
In other holes without bushings, corrosion of the 405 SS support
plate extended through the ligament between the support hole and the
flow hole or about 74 mils (1,880 gim), (Figure 24-22). The 405 SS support
plate cracked and fragmented as shown in Figure 24-23. Where cracking
did not occur, the tubes were frozen in the support plate.
The 409 SS lattice support corroded to about the same depth as the
405 SS as shown in Figure 24-24. The corrosion products fouled the
lattice supports as shown in Figure 24-25. Since the 90 mil thick lattice
strips were on a slanted grid, two tubes were never on both sides of the
strip at the same location. Through-wall corrosion was thus from only
one side. An overall view of the lattice assembly, still together in
Figure 24-25, shows the extent of the corrosion and the fouling of the
lattice support structure.
The 409 SS lattice support contained a loose strip that was replaced
once during the test. Thus, the morphology of the corrosion in the
crevice with the heated tube could be observed as it developed (24.51).
The initial pitting attack after 576 ppm chloride days developed into a
general crevice corrosion after 2,544 ppm chloride days.
FIGURE 24-21
Section Through Corroded Carbon Steel Bushing (24.8)
FIGURE 24-22
Transverse Section of Corroded Ligament Between Tube Hole and
Flow Hole in 405 SS Support Plate (24.8)
FIGURE 24-23
Type 405 SS Drilled Support Plate With 0.75 in. (19 mm) Tubes
Fragmented by Corrosion (24.8)
0
FIGURE 24-24
Type 409 SS Lattice Supports 0.09 in. (2.3 mm) Thick Fragmented by
Corrosion (24.8)
FIGURE 24-25
Corrosion and Fouling of Lattice Supports From Chloride-Faulted
Test (24.8)
Type 409 SS Eggcrate Support

Corrosion was negligible on the 347 SS lattice structure which had

a small amount of general corrosion only in one location and several
shallow oxide-filled cracks but no fouling. Lattice supports of SCR-3
and alloy 600 had negligible corrosion.
In this severe test, there was little difference in the corrosion rate of
405 SS and 409 SS. The carbon steel bushings were not thick enough to
obtain a corrosion rate. Higher alloys of 347 SS, SCR-3, alloys 600, and
690 had negligible corrosion beyond a few minor pits.
The A508 tubesheet steel showed an interesting corrosion pattern
in Figure 24-26. For a minimum rolled tube in the tubesheet with a
depth of 9.5 inches (24 cm), crevice corrosion was most severe with
about 10 mils denting and an estimated 10 mils corrosion near the
secondary face of the tubesheet. There was also a small amount of
corrosion at the bottom at the crevice. The corrosion resistance of the
tubesheet steel was substantially better than the 405 SSand 409 SS.
Denting shown in Figure 24-26 may have sealed the tubesheet crevices.
Miniature Autoclave Tests. Broomfield, et al., measured the corrosion
of small metal coupons in several concentrations of nickel chloride and
nickel chloride plus additives at 572°F (300'C) (24.52). Individual
coupons were tested separately in miniature autoclaves. Nickel chloride
is a mild oxidizing agent with the deposition of pure nickel. Nickel
chloride oxidizes steel readily but tends to passivate chromium-
containing steels at low concentrations. Increasing the chloride
concentration has the opposite effect and destroys the passive film with
a sharp increase in corrosion. The data in Table 24-35 are based on the
long-term extrapolation formulas developed by Broomfield, et al.,
(24.52) but do not include the initial corrosion term.
Steel corrosion increases with the concentration of nickel chloride
but a concentration is reached where additional copper oxide, copper
chloride, iron chloride, or seawater do not increase the rate further.
This upper limit may be at least partly due to the precipitates of nickel,
copper, and magnetite on the nickel oxide, limiting ionic diffusion to
the metal surface and the related corrosion.
The corrosion of 405 SS is lower than steel up to a concentration of
0.1 M NiC12 (7,100 ppm chloride). There is a transition concentration
range between 0.1 M NiC12 and 1.0 M NiC12 (71,000 ppm chloride),
above which the corrosion of 405 SS accelerates rapidly to substantial
amounts. Above this concentration range, copper oxide increases the
corrosion more than copper chloride or iron chloride.
FIGURE 24-26
Denting Oxide at the Top of a Tubesheet Hole From a Chloride-Faulted Test (24.8) m
TUBE 6
0
A-i
~0
Top of
Tubesheet
TABLE 24-35
Corrosion of 1018 Carbon Steel and 405 Stainless Steel in Nickel Chloride Solutions
at 572 0 F (3000 C) (24.52)
Cl Conc. 1018 Carbon Steel 405SS Effectiveness Ratio

Environment mm (mils) mm (mils) Fe/405
Water 0 14(0.6) 2(0.1) (7)
ýJ
0.001 M NiCI2 71 856 (34) 8 (0.3) 107
0.01 M NiCI2 710 11,290 (444) 35(1.4) 322
0.1 M NiCI 2 7,100 15,573 (613) 46(1.8) 336
1.0 M NiCI2 71,000 20,547 (808) 1,314 (52) 16
2.0 M NiCI2 142,000 45,745 (1801) 2,102 (83) 21
In Seawater Plus
0.1 M NiCI2 47,400 9,976 (393) 324 (13) 31
0.1 M FeCI2
0.1 M NiCI2 47,400 9,490 (374) 420 (16) 23
0.1 M CuC12 rD
2.0 M NiCI 2 143,000 5,800(230) 3,100 (130) 2

+CuO
Additional tests studied the corrosion of other alloys containing

more and less chromium than 405 SS in 0.1 M NiC12 (7,100 ppm
chloride) where the corrosion of 405 SS is low. Corrosion decreased
with increasing chromium content. Tubesheet steel corrosion with its
high silicon content and essentially no chromium was exceptionally
low and comparable to 405 SS. Restraint on the specimen did not stop
the corrosion although access of solution may have been restricted.
Model Steam Generator Test in Sulfates. Model 10 steam generator, the

companion to Model 9, was faulted to about 40 ppm sulfate with a
simulated, acidified cooling tower water, as noted previously. Chlorides
averaged about 3 ppm. The model was operated concurrently with Model
9 but for the somewhat longer period of 358 days. Sludge additions were
the same.
The maximum corrosion of carbon steel in sulfates was substantially
less than that in chlorides (Table 24-36). The general corrosion of 405 SS
was still lower, about 10 mils or 10% of that in chlorides, although some
30-mil-deep cracks were found. The pattern of attack on 409 SS lattice
supports was similar to that in chlorides but not as extensive. Nevertheless,
TABLE 24-36
Corrosion of Support Structures in Model Steam Generators 9
Faulted With Seawater and Sludge (About 30 ppm chloride present
for 285 days of operation)l (24.8)
Corrosion
Corrosion Rate/Year
Support mm (Mis) mm (mi_).
Carbon Steel Bushings 3,048 (120) >3,900 (>1 54)
405SS Support Plate 1,880 (74) 2,406 (95)
409SS Lattice Supports 2,286 (90) 2,926 (115)
347 Bushings and Lattices 254 (10) 325 (13)
SCR-3 Bushings and Lattices nil
Alloy 600 Bushings and Lattices nil
Alloy 690 Bushings nil
18,700 ppm-chloride days, 180 ppm-sulfate days, 9.3 ppm-copper days, 8.5 kg
sludge containing 30% copper oxide added.
the corrosion of 409 SS was greater than 405 SS and more than 50% of that
for carbon steel. The corrosion of 347 SS was low but included shallow
cracking. There was only shallow pitting on alloy 600, SCR-3, and alloy
690 bushings and lattice supports.
Electrochemical Tests. When two dissimilar metals are in contact with

each other in a corrosive environment, galvanic corrosion can occur. The
cathodic metal promotes the corrosion of the anodic metal. Tubes and
supports are different alloys and these alloys are in electrical contact with
each other. Galvanic corrosion between tubes and support alloys was
studied by McKubre, et al., in a loop of alloy 600 tubing at 482°F (250'C)
(24.53). Multiple cells on the loop consisted of alloy 600 tubes with a support
alloy plug covered in all exposed areas except those selected for study. The
tubing was electrically insulated from the plug but the two materials were
connected through an external zero impedance ammeter used to measure
the galvanic current. Dilute acid and other solutions were pumped into
the crevices and exited into the flowing AVT water in the loop. Several
cells were monitored simultaneously
Type A cells with a hole in the center of the support alloy plug allowed
measurement of the galvanic current from the support alloys in the crevice
to the alloy 600 tubes outside the crevice. Normally, these currents are
responsible for the important ionic migration of anions into the crevice
though in these tests the crevice solution was controlled by pumping
solution into the crevice. Type B cells contained both tube and support
alloys in the crevice where galvanic currents measured were substantially
higher.
In most circumstances, the support alloys were anodic or corroded in
preference to the tubing. For the tests, the crevice solution was 0.001 M
HC1 (35 ppm Cl), which is believed to be just below the pH of a denting
crevice. With solution flowing through the Type A crevice, the temperature
was increased in steps to 536°F (280'C) with the results given in Figure 24-27.
The galvanic currents increased sharply at first to about 212'F (100°C) then
more gradually at higher temperatures. Under the acidic conditions, 405 SS,
410 SS, and 430 SS had a galvanic corrosion current about half that of carbon
steel for about a two-fold reduction of corrosion. The 347 SS and alloy 800
had much lower corrosion currents representing 1-2 orders of magnitude
of corrosion reduction. The data were relatively similar but the corrosion
currents were much higher for Type B cells. Carbon steel provides cathodic
protection to the tubes in and near the crevice. Substituting more corrosion-
resistant support alloys for carbon steel will reduce this cathodic protection.
However, a systematic study of cathodic protection in steam generator
environments has not been performed.
FIGURE 24-27
The Influence of Temperature on the Initial Corrosion Rate of
Unpumped Type A Crevices Initially Containing 0.001 M HCI (24.53)
12
2.5 r- 11
10
2.0 9
C 7
E 1.5 -a,
)66
0
0 1.0
0 -•4
0
3
0.5
2
0 0-
-1
0 50 100 150 200 250 300
Temperature (°C)
The corrosion of support alloys as a function of passivating alloy content

is shown in Figure 24-28. Data from both Type A and Type B cells are in
good agreement and indicate that a passivating metal content of at least
25% is needed in chlorides to reduce the corrosion of support alloys to a
low level. The passivating alloy content included the nickel content. These
data indicate that 347 SS may have inadequate corrosion resistance for
supports in acidic chloride environments. Higher alloys such as SCR-3 or
800 would be better for these environments.
In a summary of their work, McKubre, et al., (24.53) have shown that
the corrosion reduction by support alloys compared to carbon steel is nearly
a constant factor dependent on the alloy for a few representative
environments. The data are shown in Table 24-37. The corrosion rates of
carbon steel 1018 in the various environments are listed in Table 24-37 for
reference. The crevice concentration of the natural waters was
approximately ten times their normal concentration which is somewhat
arbitrary but estimated to be a representative condition from a small
condenser leak.
FIGURE 24-28
Corrosion Rate of Pumped Type A and Type B Crevices at 482 0 F
(250 0 C) As a Function of Total Passivating-Element Concentrations
in the Alloy (24.53)
11 +) @94.4 1 1
10 X Crevice Type A
,'+ Crevice Type B
9
Pumped With 10-3 M HCI
8
- (410) (Alloy)
CU 7
CD
CL 6
M2
5
4 (405) (430)
3
02
0) I I(
0
0 2
k: (3
I7)(8 0
1
0
-1
-2
0 10 20 30 40 50 60
Total Passivating-Element Concentration (%)
Corrosivity of Representative Environments. Pathania and Balakrishnan

have recently correlated available data to predict the onset of denting 24.54).
In Figure 24-29, the correlation between laboratory and sea water plant data
is in good agreement at higher concentrations. At the Water Chemistry
Guidelines (24.29) level of 20 ppb chloride from seawater, plants can expect
denting to initiate after one to two-and-a-half years with copper-containing
secondary systems. If denting is to be prevented for a 40-year service life in
a steam generator, a support alloy for a seawater plant must be 8-40 times
more corrosion resistant than carbon steel.
A similar correlation for freshwater, with little plant data, permits
estimates that a typical freshwater-cooled plant might initiate denting in 20
to 60 years (Figure 24-30). Denting could be prevented by a support alloy
that gave only zero to two-fold improvement over carbon steel.
TABLE 24-37
Ratio of the Alloy Corrosion/Steel Corrosion-Support Structure
Alloy Improvement Factors-The Corrosion Rate of 1018 Carbon
Steel Divided by Corrosion Rate of Support Alloy (24.53)
Cell Type: A B B B B
Crevice: Outside Inside Inside Inside Inside
Electrolyte: 10-3 N 10-3 N 10-3 N Susquehanna Millstone
HCI HCI H2S04 River Seawater
Support
Alloy Improvement Factor
347 20 5.6 26 - -
405 3.7 1.5 3.5 5.8 4.5
409 - - 3.6 2.8 3.9
410 1.8 1.4 2.9 1.3 2.1
430 4.6 2.9 4.9 - -
508 - - 2.7 2.1 2.3

800 20 48 47
1018(1) 10 48 56 29 85 (mpy)
250 1200 1400 760 2150 (ýtm/yr)
11018 corrosion rates quoted here are representative of the largest sustainable
values, at 482 0 F (2500C), for the given conditions of crevice geometry and
electrolyte composition. There is a wide variation of the absolute corrosion rates
depending on electrolyte flow rate, exposure time, etc. However, the ratios of
corrosion rates were relatively constant.
Corrosivity in plants using an acidified cooling tower water can be

estimated in a very preliminary way by comparing corrosion of alloys in
Model 9 and 10 steam generators. Corrosion of steel in the two steam
generators cannot be used because the true steel corrosion could not be
estimated in Model 9. The alloy with the next most corrosion in both steam
generators was 409 SS. A comparison of its corrosion in the two
environments provides a factor of 2.7. Using a factor of 3 and the data from
Figure 24-30, it can be estimated that plants with fresh acidified cooling
tower water might experience denting after 3-8 years. For a 40-year service
life, a corrosion improvement factor of 5-13 would be needed compared to
carbon steel.
FIGURE 24-29
Predicted Versus Observed Time-to-Denting in Plants on Seawater
or Brackish Water Sites. (Plants use all volatile treatment.) (24.54)
10
Sea/Brackish Water Plants 0 Denting
0 No Denting
Average Time 0
0 CC-1 0
CC-2 MY W-6
_o 2W-2
W-3 Denting
©S
CLSI-i *R-2M-
0 Denting
.Possible
E
*i- 0.5
1 2 5 10 20 50 100 200 500 1000

Blowdown Chloride From Seawater, ppb
The corrosivity of the above environments may be modified by the

intrusion of ion exchange resins, the regenerating chemicals that slip through
the resin beds, residual salts in makeup water, the water chemistry control
system, the presence of copper alloys in the feedwater system and the
accidental intrusions of other chemicals. Nevertheless, utilities should use
past records to estimate corrosion rates in their steam generators. Then,
based on the improvements in their systems, their needs for corrosion
resistance of supports in a new steam generator can be estimated. Because
of their variability, the contribution of these environmental factors cannot
be estimated.
Effects of Support Hole Design on Corrosion. Corrosion of the various

designs can best be compared with the 405 SS support plates in the Model 9
steam generator operated with seawater faulting (24.8). Figures 24-23 and
24-24 show that a substantial amount of corrosion occurred with all designs.
FIGURE 24-30
Predicted Versus Observed Times to Denting in Plants on
Freshwater Sites. (Plants use all volatile treatment.) (24.54)
100
Freshwater Plants
Average lime
50 Denting
Denting
Minimum Possible
>- 20 Time
CF
0
.2
C_ 10
0 No Denting
E
F_ 5
0
W-5
0
W-4
I II I I I I
2 5 10 20 50 100 200 500 1000

Blowdown Chloride From NaCI, ppb
The corroded ligament between the drilled hole and the flow hole was about
74 mils thick. For the broached holes, the comparable ligament between
the land and the adjacent hole was 134 mils. There was at least 90 mils
corrosion for the concave (conforming) land and near through-wall
corrosion for the flat land. There was no corrosion for two of three convex
lands. Corrosion on the third land though narrow, approached 90 mils also.
Corrosion occurred on the third land probably because it contacted the tube.
While not strictly comparable, it is also worth noting that lattice supports
of 409 SS, an alloy with a composition similar to 405 SS, also corroded
severely in the Model 9 steam generator. The 90 mil (2.3 mm) thick lattice
supports fragmented in 0.8 of a year as noted previously and shown in
Figure 24-24. The extent of the corrosion and fouling is shown in
Figure 24-25.
Thus, the data indicate that alternate designs evaluated thus far have
not reduced corrosion in a highly fouled model steam generator. Benefits
were realized, however, by thicker ligaments.
Matching Service Environments to Support Alloy Composition. There

is now at least a preliminary basis for matching alloys to environments to
obtain a 40-year service life in an average steam generator. Corrosion
reduction factors compared to carbon steel have been developed for several
alloys. In addition, corrosivity of three representative environments have
been estimated from available laboratory and plant data. Based on these
estimates, corrosion reduction factors needed for these environments have
been calculated.
A preliminary matching of alloy composition for the three types of
cooling waters is shown in Table 24-38. The proposed alloys represent the
minimum alloy content for the environment. For example, for cooling with
an acidified cooling tower water, 347 SS should be adequate. Alloy 800
would be better; 405 or 409 SS would probably not last 40 years without
through-wall corrosion.
TABLE 24-38
Suggested Support Alloys for a 40-Year Service Life in Steam
Generators
Alloy
Corrosion- Corrosion-
Reduction Proposed Reduction
Cooling Water Factor Needed Alloy Factor
Seawater 8-40 800 -50
Acidified Cooling 5-13 347 20-25
Tower Water
Freshwater 0-2 405 - 409SS 3-5
While these conclusions appear reasonable, they can only be preliminary

because actual steam generator environments and corrosion should be
characterized in more detail. There are several possible ways to establish
the corrosivity of a steam generator environment but they are beyond the
scope of this discussion. The problem is complicated by short-term
variation, as well as permanent changes in the steam generator
environments suggesting that a more routine or possibly an integrated type
of environment be characterized. The more accurately the environment is
characterized, however, the better it can be dealt with either by improved
chemistry or improved materials.
REFERENCES
24.1 Paine, J. P. N., and S. J. Green. "Materials Performance in Nuclear
Steam Generators." Presented at ANS, St. Petersburgh, FL, October
1980; IAEA, Vienna, Austria, October 1980; IAEA, Vienna, Austria,
October 1980; and ANS, Chicago, Illinois, February 1981. Nuclear
Technology 55 (October 1981): 10-29.
24.2 Coriou, H., L. Grall, Y. Le Gall, and S. Vettier. "Corrosion Fissurante
Sous Contrainte De L'Inconel Dans UEau A Haute Temperature."
Third Collogue de Metallurgie Corrosion. Centre d'Etudes Nucleaires
de Saclay, France. Amsterdam: North Holland Publishing Co.,
1959, p. 161.
24.3 Debray, W., and L. Sieding. INCO Power Conference, Lausanne,
Switzerland, 1972, paper 3.
24.4 Sedriks, A. J., J. W. Schultz, and M. A. Cordovi. "Alloy 690-A
New Corrosion Resistant Material for High Temperature
Application." Presented at the 15th INCO Power Conference,
Lausanne, Switzerland, October 5-7, 1977.
24.5 Hammond, J. P., P. Patriarca, G. M. Slaughter, and W. A. Maxwell.
MaterialsPerformance, (November 1975): 41.
24.6 Blanchet, J., et al. Mem. Sci. Rev. Met. (April 1978): 237.
24.7 Nelson, J. L., and S. Floreen. "Evaluation of Alternative Alloys for
PWR Steam Generator Tubing." EPRI NP-3703. October 1984.
24.8 Krupowicz, J. J., D. B. Scott, and G. C. Fink. "Corrosion Performance
of Alternative Steam Generator Materials and Designs, Volume 2:
Posttest Examination of a Seawater-Faulted Alternative Materials
Model Steam Generator." EPRI NP-3044. July 1983.
24.9 Stellwag, B., N. Wieling, and L. Stieding. "Corrosion Resistance of
SG Materials Alloy 800 and Alloy 690-A Comparative Study"
Proceedings of the Second International Symposium on
Systems-Water Reactors, Monterey, California, September 1985.
24.10 Yonezawa, T., K. Onimura, N. Sasaguri, T. Kusakabe, H. Nagano,
K. Yamanka, T. Minami, and M. Inoue. "Effect of Heat Treatment
on Corrosion Resistance of Alloy 690." Proceedings of the Second
Materials in Nuclear Power Systems-Water Reactors ANS, AIME,
NACE, Monterey, California, September 1985.
24.11 Sedriks, A. John. 1979. Corrosionof Stainless Steels. John Wiley &
Sons. Page 172.
24.12 Airey, G. P. "Carbide Dissolution and Precipitation Kinetics of
Inconel 600." EPRI NP-2093. October 1981.

24.14 Airey. G. P. "Optimization of Metallurgical Variables to Improve
1983.
24.15 Airey, G. P., A. R. Vaia, and R. G. Aspden. "A Stress Corrosion
Cracking Evaluation of Inconel Alloy 800 for Steam Generator
Tubing Applications." Nuclear Technology 55 (November 1981): 436-
447.
24.16 Berge, P. H., and J. R. Donati. "Materials Requirements for
Pressurized Water Reactor Steam Generator Tubing." Nuclear
Technology 55 (October 1981): 88-104.
24.17 Bricefho, D. G., and M. L. Castafio. "Inconel 690 TT and Incoloy 800
in Sulfur, Copper, and Lead Environments." Proceedings: 1991
EPRI Workshop on Secondary-Side Intergranular Corrosion
Mechanisms. EPRI TR-101103. August 1992.
24.18 Santarini, G., et al. "Alloy 690: Recent Corrosion Results."
Proceedings: 1989 EPRI Alloy 690 Workshop. EPRI NP-6750-SD.
April 1990.
24.19 Crum, J. R., et al. "Effect of Different Thermal Treatments on The
Corrosion Resistance of Alloy 690 Tubing." EPRI NP-6703-SD.
March 1990.
24.20 Pathania, R. S., and J. A. Chitty. "Stress Corrosion Cracking of Steam
Generator Tube Materials in Sodium Hydroxide Solutions."
Corrosion34 (November 1978): 369-378.
24.21 Cordovi, M. A. "The Corrosion Resistance of Alloy 690 in PWR
Steam Generator Environments." Proceedings: Workshop on
Thermally Treated Alloy 690 Tubes for Nuclear Steam Generators.
EPRI NP-4665-SR. July 1986.
24.22 Hickling, J., P.Forchhammer, and P. M. Effertz. "Experimental Work
in the SCC of S. G. Tubing Materials in Modified PWR
Environments." Proceedings: First International Symposium on
Systems-Water Reactors, AIME, ANS, NACE, Myrtle Beach, South
Carolina, August 1983, p. 658.
24.23 Pathania, R. S., and R. D. Cleland. "The Effect of Thermal
Treatments on Stress Corrosion Cracking of Alloy 800 in a Caustic
Environment." Corrosion41 (October 1985): 575-581.
24.24 Sedriks, A. J., S. Floreen, and A. R. McIlree. Corrosion 32 (1976):
147.
24.25 Pessal, N. "Prediction of Stress Corrosion Cracking in 10% Caustic

Soda Solutions at 315'C (600'F)." CorrosionScience 20 (1980).
24.26 Theus, G., R. M. Emanuelson, and S. F. Chou. "Stress Corrosion
Cracking of Alloy 600 and Alloy 690 in All Volatile Treated Water
at Elevated Temperatures." EPRI NP-3061. May 1983.
24.27 Bergmann, C. A., D. E. Durkosh, W. T. Lindsay, and J. Roesmer.
"Roll of Coolant Chemistry in PWR Radiation Field Buildup." EPRI
NP-4247. October 1985.
24.28 Jacko, R. "Effect of Lithium Hydroxide on Primary Water Stress
Corrosion Cracking of Alloy 600 Tubing." EPRI NP-7396-S.
September 1991.
24.29 PWR Primary Water Chemistry Guidelines: Revision 2, November
1990.
24.30 Krupowicz, J. J., E. J. Silva, G. C. Fink, and J. E Hall. "Supplementary
Examination of Alternative Materials in a Model Steam Generator,
Volume 1: Tubesheet and Tube Bundle Examinations." EPRI
NP-5928-SP. October 1988.
24.31 Newmann, J. F. "Stress Corrosion of Alloys 600 and 690 in Acidic
October 1983.
24.32 Fletcher, W. D, and D. D. Malinowski. "Operating Experience with
Westinghouse Steam Generators." Nuclear Technology 28 (1976): 356-
373.
the Stress Corrosion Cracking Resistances of Inconel 600." EPRI
NP-1354. March 1980.
24.34 Copson, H. R., and S. W. Dean. "Effects of Contaminants on
Resistance to Stress Corrosion Cracking of NiCr Alloy 600 in
Pressurized Water." Corrosion,no. 1, 21 (January 1965).
24.35 Miglin, B. P., and J. M. Sarver. "Investigation of Lead as a Cause of
Stress Corrosion Cracking at Support Plate Intersections." EPRI
NP-7367-S. June 1991.
24.36 Pement, F. W., J. L. Wilson, and R. G. Aspden. "Stress Corrosion
Cracking Studies of High Nickel Austenitic Alloys in Several High
Temperature Aqueous Solutions." Materials Performance (April
1980).
24.37 Cullen, W. H., "Susceptibility of Alloys 600 and 690 to Acid Sulfate
Solutions." Proceedings: 1991 EPRI Workshop on Secondary-Side
Intergranular Corrosion Mechanisms. EPRI TR-101103. August
1992.
24.38 Pessall, N., A. B. Dunlap, and D. W. Feldman. "The Corrosion

Resistance of Inconel Alloy 600 in High Temperature Phosphate
Solutions." Corrosion 33 (1977): 130-144.
24.39 Krupowicz, J. J., D. B. Scott, R. M. Rentler, and G. C. Fink. "Corrosion
Performance of Alternative Steam Generator Materials and Designs,
Volume 3: Posttest Examination of a Freshwater Faulted Alternative
Materials Model Steam Generator." EPRI NP-3044. July 1983.
24.40 Baum, A. J., C. Dongelo, P. J. Kuchirka, N. Siegleton, and M. J.
Siedzienski. "Steam Generator Cold Leg Thinning." To be
published.
24.41 Proceedings: 1991 EPRI Workshop on Secondary Side Intergranular
Corrosion Mechanisms. EPRI TR-101103. August 1992.
24.42 Tatone, 0. S., P. Meindl, and G. F Taylor. "Steam Generator Tube
Performance; Experience With Water-Cooled Nuclear Power
Reactors During 1983 and 1984." AECL-9170. January 1986.
24.43 Gold, R., et al. "Characterization of the Resistance to PWSCC of
Hydraulic Tube-Tubesheet Expansions." EPRI TR-100865. July
1992.
24.44 Dow, B. L., Jr. "Steam Generator Progress Report Revision 8." EPRI.
October 1992.
24.45 Gold, R., et al. "Alloy 690 for Steam Generator Tubing
Applications." EPRI NP-6997-SD. October 1990.
24.46 Rothstein, S. "Sampling and Analysis of Lower Steam Generator
Support Plate-Indian Point 2." Proceedings: Support-Structure
Corrosion in Steam Generators. EPRI NP-2791. January 1983.
24.47 Metals Handbook. Desk Edition, American Society of Metals. May
1985.
24.48 Krupowicz, J. J., J. E. Casteel, and B. W. Kinyon. "The Design and
Construction of Model Steam Generators for Corrosion Testing of
Alternative Materials." EPRI NP-1965. August 1981.
24.49 Economy, G., D. Smith-Magawon, S. G. Desilva, R. M. Kunig, M.
Ambrosino Jr., and C. R. Wolfe. "Causes of Denting." EPRI NP-3275
(6 volumes). December 1983, May 1984.
24.50 Mann, G. M. W., and R. Castle. "The Effects of Oxygen, Copper
and Acid Chlorides on Denting Corrosion." EPRI NP-4648. July
1986.
24.51 Shoemaker, C. E. "Selecting Support Structure Alloys For Nuclear
Steam Generators." Proceedings of the Second International
Symposium on Environmental Degradation of Materials in Nuclear
Power Systems-Water Reactors ANS, AIME, NACE, Monterey,
California, September 1985.

NP-4647. July 1986
24.53 McKubre, M. C. H., and S. C. Leach. "Crevice Corrosion of Lattice
Support Alloys in the Secondary Environments of Nuclear Steam
Generators Supplemental Support." EPRI NP-5017. March 1987.
24.54 Pathania, R. S., and P. V. Balakrishnan. "Correlation of Tube Support
Structure Studies." EPRI NP-4672. July 1986.
25
PWSCC REMEDIES
A. R. Mcllree, EPRI
E. S. Hunt, Dominion Engineering,Inc.
INTRODUCTION
Primary water stress corrosion cracking (PWSCC) has become a
significant problem at many plants, as discussed in Chapter 7. Locations at
which PWSCC has been reported are shown in Figure 25-1. These locations
include: (1) expanded regions within the tubesheet, (2) tubesheet expansion
transitions, (3) dented tube support plate intersections, and (4) inner row
U-bends. With few exceptions, all of these locations involve high stresses
induced in the tubing during fabrication or by denting or live deflections
which occur during operation.
Remedial measures are used to prevent PWSCC from occurring in the
first place, and to permit plant operation after PWSCC has been detected.
This chapter addresses both preventive and corrective remedial measures.
Table 25-1 summarizes the main PWSCC remedial measures. These
remedial measures are grouped in terms of the three factors required for
PWSCC to occur. The table also indicates the current status of development,
and the extent to which the remedial measure has been applied in the field
through 1989.
The subject of PWSCC remedial measures has been covered in a number
of comprehensive reports and conference proceedings (25.1-25.4).
REGULATORY REQUIREMENTS
The need for PWSCC remedial measures in domestic plants is generally
based on individual plant technical specifications and regulatory
requirements. For plants in the United States, NRC Regulatory Guide 1.121
(25.5) contains the following main requirements pertaining to operating
with PWSCC damage: (1) the margin of safety against tube rupture under
normal operating conditions should not be less than three at any location
where defects have been detected, (2) the margin of safety against tube
rupture under postulated accident conditions such as a LOCA, steam line
break, or feedwater line break concurrent with the Safe Shutdown
Earthquake (SSE) should be consistent with the margin of safety determined
by the stress limits specified in paragraph NB-3225 of Section III of the ASME
Boiler and Pressure Vessel Code, and (3) any increase in primary-to-
secondary leakage rate should be gradual. These criteria apply regardless
of the PWSCC location.
Below is a brief overview of the need for PWSCC remedial measures
based on regulatory requirements for each location in the steam generator
where significant PWSCC has been detected.
PWSCC Remedies 25-3
FIGURE 25-1
PWSCC Locations
At Scratches inTube
Above Top of Tubesheet
Expansion Transitions
TABLE 25-1 U'
ja
Summary of Remedial Measures for PWSCC for Already Fabricated Steam Generators
Plants Treated
Remedial Measures Current Status Through 1989
Reduce Tensile Stress
- Stress relieve U-Bends Offered by several vendors numerous
- Stress relieve expansion transitions Development not completed none
- Shot peen expanded area and transitions Offered by several vendors numerous
- Rotopeen expansion transitions Offered by several vendors numerous
- Re-expand tubing to improve transitions (1)
- Secondary water chemistry to reduce denting numerous
- Secondary chemical cleaning to arrest denting none
0
Reduce Material Susceptibility
- Heat treat tubes to improve microstructure Tests encouraging none
Reduce Aggressive Environment
- Reduce primary coolant temperature No development required one
- Reduce hydrogen concentration Tests completed (2)
3
- Install sleeves( ) Offered by several vendors numerous
- Install plugs Offered by several vendors numerous
- Electroplate tube wall with nickel Application on trial basis two
1Older Japanese plants have had tubing re-expanded.

2
The general practice is now to keep the hydrogen at the low end of the specified range.
3
Many plants have sleeves but no sleeves have been installed as PWSCC remedial measure.
PWSCC Remedies 25-5
PWSCC in Expanded Region

With the exception of the tube bulging which occurred at Obrigheim
(Chapter 7, Figure 7-5), there is little technical risk associated with PWSCC
of steam generator tubing over the expanded length of tube within the
tubesheet, except within a short length of tube below the expansion
transition. The depth below the expansion transition, where PWSCC ceases
to be a concern, is the depth at which the tube will meet NRC Regulatory
Guide 1.121 requirements even if it is severed circumferentially. This
minimum distance can be established based on geometric constraints
preventing tube pull out (P* criterion), or on the force resisting pull out
developed by the remaining length of undegraded tube (F* criterion).
P* Criterion. Remedial measures are not required for PWSCC if (a) it occurs
below the depth at which geometric constraints will prevent the tube from
coming out of the tubesheet even if severed circumferentially, and (b) if it
does not result in coolant leakage exceeding technical specification
requirements. This is defined as the P* (pull out) depth. In theory, the
distance which a tube can pull out of the tubesheet is limited by the
proximity of adjacent tubes in higher number rows including provision for
PWSCC inspection uncertainties, tube dimensions, tube spacing and
tolerances. This length is typically 1.25-1.50 inches.
The P* criterion is not applicable to the outermost tube in each column,
or to any other tubes which are not backed up by a tube in the next higher
row. Furthermore, the criterion cannot be used for cases where the tube in
the same column in the next higher row is also degraded. This means that
a sound, undegraded tube must be present immediately outside the
degraded tube in order to prevent pullout of the degraded tube, should it
become separated. Given these limitations, the P*criterion is not the current
criterion of choice by most utilities.
F* Criterion. Remedial measures are not required for PWSCC if it occurs

below the depth at which the applied force trying to eject the tube from the
tubesheet under normal and accident conditions is less than the force
resisting pull out of the circumferentially severed tube. This is defined as
the F* (pull out force) depth. The force resisting tube pull out is developed
by the residual contact stress between the tube and tubesheet after
expansion. The F* depth is determined based on the following factors:
* Operating primary-to-secondary differential pressure increased by a
factor of 3 to comply with NRC Regulatory Guide 1.121 requirements.
* Maximum primary to secondary differential pressure during a steam
line break transient, increased by a factor of 1.43 to comply with NRC
Regulatory Guide 1.121 requirements. (Note: Paragraph NB-3225 of

Section III of the ASME Boiler and Pressure Vessel Code refers to
Appendix F of Section III which limits the accident condition primary
shear stress to 70% of the ultimate shear strength. For purposes of
establishing a minimum required engagement length of PWSCC free
tube in the tubesheet, this has been interpreted to mean that the tube
should not pull out of the tubesheet at pressures up to 1.0/0.7 = 1.43
times the accident condition differential pressure.)
* Lower bound radial contact stress between the tube and tubesheet after
expansion.
* Changes in radial contact stress between the tube and tubesheet
resulting from operating differential pressure, operating temperature,
and tubesheet strains.
" Lower bound friction coefficient between the tube and tubesheet.
* PWSCC inspection uncertainties.
F* depths licensed for several domestic plants fall in the range of 1.6 to
2.0 inches.
In summary, a technical case can be made for continued operation with
most PWSCC defects within the expanded tubesheet depth, provided the
defects are located 1.5-2.0 inches or more below the expansion transition.
Many plants have found that a significant percentage of expanded region
PWSCC defects can be accepted using this approach.
PWSCC at Expansion Transitions

PWSCC defects with depths exceeding 40% of the wall thickness and
located within the expansion transition, P* region, or F* region, cannot be
accepted using the criteria described above. However, laboratory data
and pulled tube experience at many plants have shown that expansion
transition PWSCC tends to be axially oriented and short, with the crack
length limited to the portion of the transition which is in high residual hoop
tension. Laboratory data and field experience, such as illustrated in Chapter
7 (Figures 7-7 to 7-10), suggest that expansion'transition cracks increase in
length relatively slowly over time.
Based on the fact that most axial expansion transition cracks do not
increase significantly in length over time, or at worst grow very slowly,
several European utilities have developed criteria which permit operation
with short axial cracks which do not have excessive leakage (25.6-25.7).
Using this approach, tubes which are found to have axial cracks of about
0.55 inches (14 mm) or longer in 0.875 inch (22.2 mm) diameter tubing are
plugged, but shorter axial cracks are left in service without remedial
PWSCC Remedies 25-7
measures. In several other countries (France and Spain), a leak before risk
of breaking (LBRB) criterion has been established and implemented. None
of these criteria permit operation with circumferential cracks.
There are major risks associated with continued operation with short
axial cracks in the expansion transition region: (1) The presence of axial
cracks makes inspection for circumferential cracks more difficult. (2) There
is a potential that leakage from the axial cracks, and possibly hidden short
circumferential cracks, will increase the radiation release following a steam
line break accident. (3) there is some increase in primary-to-secondary-
side leakage, which increases secondary side radiation and contamination
levels. (4) The leakage rate from a crack may increase during operation,
thereby necessitating a shutdown.
EPRI and utilities are in the process of evaluating alternate tube plugging
criteria for domestic plants which would permit operation with some
PWSCC in and above the expansion transition region. However, these
criteria have not yet been generally accepted by the NRC.
PWSCC at Dented Tube Support Plate Intersections

PWSCC at tube support plate intersections has not been a significant
problem subsequent to the severe denting encountered in the 1970s at plants
such as Surry and Turkey Point. However, PWSCC has recently been
reported in dented tube support plate intersections at North Anna Units 1
and 2. The approach being taken at North Anna is to plug tubes with
PWSCC exceeding 40% of wall thickness. Some consideration has also been
given to developing a criteria which would permit operating with axial
cracks at dented tube support plate intersections provided that the cracks
remain within the span of the tube support plate thickness.
PWSCC at U-Bends
Tubes which develop PWSCC in the U-bend region and exceed the 40%
through wall criteria must be plugged. However, stress relief is an effective
and practical preventive remedial measure at this location.
REMEDIAL MEASURE STRATEGIES BY

PLANT/MATERIAL TYPE
The need to apply PWSCC remedial measures depends upon a number
of plant specific design and operating factors (25.8-25.14).
Early Steam Generators With Lower Strength Tubing

Most early Westinghouse PWR plants (1968 through 1973 startup), and
all Combustion Engineering plants, were supplied with steam generators
which were manufactured using alloy 600 tubing that was mill annealed at
relatively high temperatures (1,800-1,950'F [980-1,065°C]). It is generally
considered that this tubing, in combination with lower operating
temperatures, has relatively good resistance to PWSCC (25.15-25.16). While
more resistant to PWSCC than later plants, these steam generators are not
immune to PWSCC. This is evidenced by the fact that tubing in these
generators has developed PWSCC. In some plants with severe denting
such as Surry and Turkey Point, PWSCC occurred rapidly. In other
Westinghouse plants, such as Connecticut Yankee, San Onofre 1, and Ginna,
PWSCC began to occur about 18-19 years after plant startup. At Millstone 2,
a Combustion Engineering plant, PWSCC occurred in dented and deflected
tubes about 15 years after plant startup.
PWSCC preventive measures should probably be deferred for these
plants until it becomes clear that PWSCC will be a life limiting problem.
(1) Many steam generators have already been replaced and others are likely
to be replaced due to denting and other secondary side corrosion problems.
(2) With the exception of severe denting, or large tube deflections, there has
been little PWSCC after 15-20 years of operation, and PWSCC is unlikely to
develop suddenly. (3) Generally, these steam generators have large plugging
margins.
Steam Generators With Higher Strength, Non-Thermally Treated,

Alloy 600 Tubing
Starting in the mid-1970s, the yield strength of Westinghouse steam
generator tubing was increased, primarily by lowering the mill anneal
temperatures to less than 1,800'F (980'C). Experience has shown that this
material is more susceptible to PWSCC than material mill annealed at higher
temperature. Remedial measures should be applied to prevent PWSCC in
plants with this material unless the plant has operated for 5-10 years without
problems, or unless plans are being made to replace the steam generators
for other reasons.
PWSCC Remedies 25-9
Expanded Zone Within the Tubesheet. Remedial measures need not be

applied to the expanded length of the tube within the tubesheet depth which
lies below the P* or F* length. The only exception would be for the case of
large bulges such as those observed at Obrigheim.
Part-Depth Roll Expansion. Operating experience at earlier plants with

part-depth roll expansions has shown that short, axially oriented cracks
develop in the transition region. However, these cracks tend not to grow
circumferentially, or to a length where they pose a significant leakage
problem. Also, since the tubes are captured in the tubesheet, there is little
risk of major leakage from a tube failure at this location. The current practice
with this type of expansion in Europe is to plug or electroplate the tubes
when leakage exceeds operational limits. In Japan, re-explansion of the
tubing above the original roll transition has shown promise, and has been
applied at plants such as Takahama Unit 1. Sleeving is also an acceptable
remedial measure and has been used at many plants with part-depth roll
expansion in the United States.
Full-Depth Expansion. PWSCC in tubes with full-depth expansion is of

concern since the cracking occurs at the top of the tubesheet; an axial rupture
or double-ended tube failure at this location could result in large leakage.
There are several types of full-depth expansions:
Full-Depth Roll Expansion. Full depth roll expansion results in the
same type of short axial cracks at roll transitions as produced by part-
depth roll expansion.
Full-Depth Roll Expansion With DAM Treatment. The dudgeonnage
ameliore mechaniquement (DAM) or "kiss" roll treatment was intended
to reduce OD stresses at the roll expansion transition by adding an
intermediate expansion step. This process has been widely used in
Europe but not in domestic plants. Experience has shown that axial
cracks develop in these transitions and that the cracks tend grow to a
greater axial length than in standard roll transitions.
Full-Depth Explosive/Hydraulic Expansion. It was initially believed
that tubes with full-depth explosive or hydraulic expansions would be
immune to PWSCC due to the lower residual stresses at the transitions
relative to those produced by a roll expansion. However, experience at
North Anna has shown that PWSCC can develop at explosive expansion
transitions after less than ten years of operation.
Some researchers believe that stresses in hydraulically expanded
tubes are about the same as in explosively expanded tubes. Therefore,
if the material is susceptible, it is considered that cracking can develop

at explosive and hydraulic expansions, but at a lower rate than in roll
expansions.
Remedial measures should be applied to the expansion transition
regions of plants with full-depth expansions and higher strength, non-
thermally treated alloy 600 tubing. The particular remedial measure to be
applied depends upon the plant status. If the plant is new, or has operated
without cracked tubes, remedial measures can include shot peening, nickel
plating, or stress relief. If the plant has started operation and the tubes
have developed cracks, shot peening is not as attractive, since it does not
completely arrest growth of pre-existing cracks. In this case, appropriate
remedial measures include sleeving, electroplating, and, possibly, stress
relief.
Dented Tube Support Plate Intersections. Severe denting at tube
support plate intersections led to widespread PWSCC in early steam
generators with tubing generally considered to be resistant to PWSCC.
In addition, experience at North Anna 1 has shown that even small
dents, as low as 0.002-0.003 inch (0.051-0.076 mm), can crack if the tubing
is susceptible. The primary preventive measure for this condition is to
control the secondary side water chemistry to prevent denting from
occurring in the first place. Once denting has occurred, several remedial
measures can be considered. Some are listed below:
* Secondary side water chemistry changes, including the use of boric
acid, to prevent further denting. This will not necessarily prevent
PWSCC at already dented intersections, but should limit its severity.
* Local or global stress relief after denting has been arrested, to relieve
residual stresses induced by the denting. Chemical cleaning may
be necessary before stress relief to prevent excessive stresses or
distortion from occurring during the stress relief operation.
* Sleeving or possibly electroplating of dented areas with PWSCC.
U-Bends. All row 1 and 2 U-bends that were formed by using an
internal ball mandrel should be stress relieved or plugged. Many row
1 U-bends of this type developed PWSCC in the first 2 or 3 years of
operation, and a significant number of row 2 U-bends have started to
exhibit PWSCC at several plants after 5-10 years of operation. The choice
between stress relief and plugging depends mainly upon the plugging
margin available. Plants with a large margin of excess tubes can accept
plugging while those with a small margin should generally elect in-situ
stress relief. If removable plugs are used, they can be removed at a
later date, and the U-bends stress relieved, if additional tubes are
PWSCC Remedies 25-11
required to maintain power output. The need for in-situ U-bend stress
relief on rows three and higher has not been established.
Recent Steam Generators With Thermally Treated Alloy 600 or Alloy

690 Tubing. Most recent alloy 600 tubing has been thermally treated at
about 1,300'F (700'C) for 12-16 hours to relax fabrication-induced residual
stresses, and to improve the corrosion resistance of the material. Reference
25.17 contains early EPRI-sponsored work on this subject. Recent EdF
laboratory testing of highly strained, reverse U-bend specimens in pure
and primary water at 680'F (360°C) and 617'F (325QC) has shown that
thermal treatment improves the PWSCC resistance of most, but not all,
French-made alloy 600 steam generator tubes (25.18-25.19). The main
conclusion from this work was that hot leg transitions in thermally treated
tubes with carbon content less than about 0.03% should be able to operate
for at least 200,000 hours (23 years). Recent experience in France has shown
PWSCC in some roll transitions and at top-of-the-tubesheet dents in some
thermally treated tubing.
The need for remedial measures in thermally treated alloy 600 tubing
must be assessed on a case by case basis, taking into account factors such as
type of expansion, tubing material chemistry, operating temperature, and
sources of external loads such as dents or tube deflections.
There is no laboratory or operating experience which suggests that
steam generators fabricated with alloy 690 tubing will demonstrate PWSCC
susceptibility.
ASSESSING REMEDIAL MEASURE

EFFECTIVENESS
Since PWSCC can take many years to develop, and the risks in applying
an unproven remedial measure are large, accelerated laboratory testing is
required to assess the effectiveness of candidate remedial measures before
they are applied. However, there are two major concerns with accelerated
testing. First, the cracking mechanism may be different in the accelerated
test than in real life. Second, the accelerated test may have a lower stress
threshold than in real life. Specifically, some accelerated test methods
produce cracks with tensile stresses as low as 20% of yield strength, while
field experience shows that the stress threshold for significant PWSCC under
actual operating conditions is much higher. Therefore, treated specimens
which pass the accelerated testing should perform well in service. Failure
of specimens under accelerated testing does not necessarily mean that the
treated tubes will fail in service.
Stainless Steel Tubes in Boiling Magnesium Chloride

A commonly used method to rapidly determine that a remedial measure
results in low residual tensile stresses is to test stainless steel specimens in
boiling magnesium chloride for about 24 hours. The threshold stress for
SCC in this test environment can vary depending on factors such as the
type of stainless steel or composition of the test environment. However,
under typical conditions, specimens which have tensile stresses in excess
of about 90 MPa (13 ksi) will fail by stress corrosion cracking in as little as
24 hours (25.20). If the tensile stresses are less than about 90 MPa (13 ksi),
the specimens will not crack during the 24-hour period. Therefore, this is
an effective test to quickly determine if residual tensile stresses induced by
a particular remedial measure are low. Occasionally, SCC does not occur as
expected. In order to monitor for this possibility, samples with known
stresses should be included in the test matrix to verify that the stress versus
time-to-failure behavior is as expected.
Polythionic Acid and Sodium Tetrathionate

Another commonly used method to assess residual tensile stresses is
to test sensitized alloy 600 tubing in a reduced sulfur oxyanion containing
environment such as polythionic acid or sodium tetrathionate at low
temperature. If the specimens have tensile stresses in excess of about 55-
70 MPa (8-10 ksi), they will fail by stress corrosion cracking in as little as 24
hours (25.20). If the tensile stresses are less than about 55-70 MPa (8-10 ksi),
the specimens will not fail during the 24-hour period. This stress threshold
is about 20% of the yield strength of mill annealed alloy 600 tubing.
Sodium tetrathionate testing may not be useful for remedial measures
involving high temperature processes, such as welding or stress relief, since
the thermal processes can desensitize the material. In this case it might be
concluded that the remedial measure reduced the residual stress when, in
fact, it only desensitized the material such that it would not crack in the
sodium tetrathionate solution.
10% Sodium Hydroxide at 550°F (2880 C)

Sodium hydroxide concentrations of 10% or more can occur in steam
generator secondary side crevices which are exposed to alternate wetting
and drying conditions. Tests have shown this to be an aggressive
environment which causes IGSCC similar to that observed in mill annealed
alloy 600 in primary water at higher stress levels and over longer periods
of time. An anodic electric potential may be applied in the laboratory to
accelerate the rate of attack. The multiplier on life between test specimens
in 10% sodium hydroxide at 550'F (288°C), with an applied potential of
+190 mV and primary water at a typical hot leg temperature of 615'F (324°C),
is reported to be about 250 (25.21). Accordingly, a year of operation is
simulated in about one-and-a-half days under these conditions. The stress
threshold below which cracking does not occur for this type of test is less
than 20% of the room temperature yield strength of mill annealed alloy 600
tubing for test durations greater than about 700 hours (25.21-25.22).
10% Sodium Hydroxide at 660°F (3490 C)

This test environment has been used by several vendors to qualify
sleeves and other remedial measures. It appears to provide results similar
to the lower-temperature-applied-potential procedure described above.
Elevated Temperature Steam Tests

Testing over the past several years has demonstrated that PWSCC can
be accelerated by performing tests at elevated temperatures. One approach
is to test alloy 600 tubing in high temperature pure water with a hydrogen
overpressure. Since water cannot be raised above 705'F (374 0C), tests at
higher temperatures can only be performed in steam. Accelerated steam
testing is currently conducted at a temperature of about 750'F (400'C) and
3,000 psig (20.8 MPa) with the external pressure adjusted to simulate the
operating pressure differential across the tube wall. Based on an activation
energy model, PWSCC under these conditions should be accelerated by a
large factor, relative to that at a typical 615'F (324°C) hot leg operating
temperature. Accurate predictions of the accelerating factor for steam tests
cannot be defined due to chemical differences between steam and lithiated
borated water with pH and hydrogen fugacity influences. EdF testing
(25.23) has indicated that, while the elevated temperature steam tests show
the correct ranking of susceptibility of materials to PWSCC, they may not
provide accurate estimates of relative times-to-cracking of materials or
remedial measures. This was determined by testing reverse U-bends (higher
stress) and flat U-bends (lower stress) of the same heats of material in 750°F
(400'C) steam and 617'F (325°C) water. Test results are available in
Table 25-2.
Several theories have been proposed to explain these results, but to
date there is no generally accepted explanation. Therefore, if steam tests
are used to assess the effectiveness of remedial measures, then the threshold
stress and appropriate multipliers on time-to-failure in the steam
environment should be verified using C-ring or other specimens.
Recent work has shown that steam doped with salts of sodium chloride,
fluoride, and sulfate can be used to reduce the required test time to less
than 1,000 hours (25.24).
TABLE 25-2
Testing of Reverse and Flat U-Bends
Time to First Failure (Hours)

Reverse Flat Time to Failure
Test Conditions U-Bends U-Bends Ratio
4000C steam @ 1 bar H2 100 >3,000 >30
3250C water @ 3 bar H2 3,000-4,000 15,000 =5
Primary Water Tests at Elevated Temperature

The most representative accelerated test condition for PWSCC is
primary water at a temperature moderately higher than the hot leg water
temperature. The temperature typically used is about 680'F (360'C). For
non-stress relieved alloy 600 tubing under simulated operating conditions,
the multiplier on life between a typical 615'F (324 0C) hot leg temperature
and the 680'F (360'C) autoclave test condition is between 6 and 20 times
(25.21 25.25). However, more recent work suggests that factors such as
dissolved hydrogen activities and the chemistry of high temperature water
act to reduce the acceleration. The stress threshold, below which no cracking
has been reported for this type of test, is close to the material yield stress,
the same as for PWSCC under normal operating conditions.
Primary Water Tests at Operating Conditions

Ideally, it would be desirable to perform tests for effectiveness of a
PWSCC remedial measure under simulated operating conditions using
actual tubing which is known to be susceptible to PWSCC in a steam
generator environment. These conditions would be primary water at a hot
leg temperature of about 615'F (324°C), with a hydrogen overpressure, and
with elevated pH. However, this type of test is not practical for screening
or procedure development purposes since actual plant operating experience
has shown that it can take years for PWSCC to occur under these conditions.
Field Experience
Field experience is not a practical method for screening potential sleeve
designs, or for procedure development, since it may take years for PWSCC
to occur, even with susceptible material. However, field experience provides
valuable information regarding installation problems, and whether specific
remedial measures have been effective to date. Also, some utilities apply
remedial measures to small numbers of tubes to obtain practical experience
prior to treating large numbers of tubes.
Summary
The ideal qualification program for remedial measures would include
the following phases:
* tests of stainless steel specimens in boiling magnesium chloride, or
sensitized alloy 600 specimens in sodium tetrathionate, to confirm that
residual stresses are low on both primary and secondary sides (including
the crevice),
" tests under more representative accelerated conditions, such as mill
annealed alloy 600 tubing in 10% sodium hydroxide or high temperature
doped steam, to confirm that total operating stresses will be less than
about 40 ksi,
* tests in elevated temperature primary water to determine long term
performance, with respect to PWSCC, under the most representative,
yet still accelerated conditions, and
" field application of the remedial measure on a small number of tubes a
year or so prior to having to treat large numbers of tubes to confirm the
practicality of application methods and to provide advance warning of
any unanticipated, rapidly occurring, problems.
Control specimens should be included in all tests to confirm that test
results are as anticipated.
ROTOPEENING
The purpose of rotopeening is to produce a thin layer of compressive
residual stress on the inside surface of the tube. The compressive stress
serves to inhibit the initiation of IGSCC, which develops in the presence of
relatively high tensile stresses. Rotopeening is performed using beads
bonded to fabric in a rotating flapper wheel. Impact of the beads on the
tube ID surface produces a thin, work hardened, compressive stress layer.
However, this remedial measure generates some major concerns:
(1) potential for increasing tensile stresses on the outside of the tube,
(2) potential for incomplete coverage at sharp geometric discontinuities,
(3) effect of peening on pre-existing cracks, (4) changes in the bead impact
angle leading to smearing and local tensile stresses on the surface, and
(5) radiation exposure/cleanup problems during application.
Process Description
The operation of typical rotopeening equipment is illustrated in
Figure 25-2. The equipment consists of a flapper wheel with bonded beads;
a rapidly rotating shaft to turn the flapper wheel; and a slowly rotating
shaft which is offset from the tube axis such that the beads impact the surface
in a nearly perpendicular direction and the flapper wheel covers the
complete inside surface of the tube. The tool is oscillated slowly in the
axial direction throughout the peening in order to insure uniform coverage.
FIGURE 25-2
Typical Rotopeening Equipment
Several Thousand RPM

Steam Generator Alloy 600 Tube
Slotted Spindle Shaft
Tool Tube (Spindle Cage)
10 to 20 RPM
Two Rows Tungsten Carbide
Beads at Each End
Fiberglass Mesh
Flapper - Plastic Bearing (Top and
Bottom of Flapper Area)
Tungsten Carbide Beads
- Up and Down Oscillations

and Slow Rotation
Steam Generator Tube

Rotopeening is performed using small 0.039 inch (1 mm) diameter

tungsten carbide beads bonded to a fiberglass flapper. The flappers are slit
into strips no more than two beads wide in order to assist in obtaining
good coverage at geometric discontinuities.
Rotopeening equipment can be Ased'to peen the entire expanded length
of tubing within the tubesheet, or just a few inches on either side of the
expansion transition. Peening tools come in several configurations: a short
flapper for peening the transition region only, a long rigid tool with many
adjacent flappers which can peen the entire tubesheet depth at one time,
and a segmented tool with a number of flappers which are connected by
flexible joints. The long rigid tool is used for peening the full depth of
tubes near the center of the generator. The segmented tool is used for
peening the full depth of tubes near the periphery of the generator where
the long rigid tool cannot be used due to head curvature. Use of the
segmented tool requires that peening be performed at a minimum of two
elevations since there are no flappers at the joints.
When rotopeening is performed at plants which have not been operated,
the tools are installed manually in the tubesheet. For plants which have
been operated and are radioactive, the rotopeening tooling is moved within
the steam generator by either a robot arm or a finger walker system.
Process control is achieved by monitoring key variables such as rotation
speeds and duration and periodic rotopeening of verification specimens.
Verification includes testing for Almen intensity and for peening coverage.
Testing for peening coverage typically involves peening of roll transitions
in stainless steel, or sensitized alloy 600 tubes, which are then subjected to
boiling magnesium chloride or sodium tetrathionate testing, respectively,
to verify that there is no cracking. Absence of cracking indicates that residual
tensile stresses are less than about 55-90 MPa (8-13 ksi).
Qualification Testing
Rotopeening qualification tests have been performed by several
organizations (25.20,25.26-25.28). This work has included a wide range of
tube roll expansion conditions such as full-depth roll, "kiss roll," a range of
gap dimensions, and skip rolls. The qualification effort has included the
following types of tests:
* accelerated SCC testing: stainless mockups in boiling MgC12 at 309'F
(154 0 C),
* accelerated SCC testing: sensitized alloy 600 mockups in sodium
tetrathionate,
* accelerated SCC testing: alloy 600 mockups in 10% caustic with anodic
potential, and
* strain gage measurements on the tube's OD surface.
Belgian/Westinghouse Development. The most recent and complete

summary of rotopeening development is contained in Reference 25.28. Early
rotopeening development work on kiss roll transitions using stainless steel
specimens in boiling MgC12 showed a peening intensity in the range of
Almen 8A to 12A to be optimum (25.20). Lower peening intensities resulted
in cracking on the tube ID and higher peening intensities resulted in cracking
on the tube OD. In this intensity range, about 10% of the rotopeened tubes
developed ID cracks while 90% of the unpeened tubes developed cracks.
The stress threshold for the MgC12 testing was on the order of 90 MPa
(13 ksi). When tests were conducted on alloy 600 tubes in sodium
tetrathionate which has a lower stress threshold of 55-70 MPa (8-10 ksi),
rotopeening reduced the rate of cracking at the kiss roll/hard roll transition
from 100% of the tubes to 0-50% of the tubes. A final qualification test of
alloy 600 tubing in 10% NaOH at 601'F (316'C) and 140 mV potential
showed that unpeened tubes developed axial ID cracks at the expansion
transition while peened tubes did not. It was concluded from this testing
that peening at an Almen intensity range of 8A to 10A provides significant
relief of tensile residual stresses at the tube ID while limiting the increase in
OD stress to about 20-50 MPa (3-7 ksi). The compressive stress at the tube
ID was 200-500 MPa (29-73 ksi) and the compressive stress zone extended
to a depth of about 150 microns (0.006 inches) (25.28).
The test work reported above was for the case of kiss roll tubes.
Subsequent tests were performed for roll expanded tubes without the kiss
roll and using a flapper which was split horizontally between each pair of
tungsten carbide beads (25.27). The purpose of the split flapper was to
provide better coverage in the transition region. The results of tests using
stainless steel specimens in boiling MgCI2 showed that the final, selected
conditions essentially eliminated ID cracking and produced only a modest
increase in the percentage of tubes with OD cracking. The tests also showed
a fairly broad tolerance regarding peening time without any significant effect
on ID or OD stresses.
French Rotopeening Development. Initial flappers covered with glass

beads exhibited high wear, so EdF switched to 3M flappers of the same
type used by Westinghouse in the Belgian development. This process was.
first applied to 100 tubes in Bugey 5 in July 1985. In this trial application,
the flappers did not produce full coverage, and the failure rate of corrosion
test specimens was 50% (25.26). Additional development was performed
in the Fall of 1985 to achieve 100% coverage. Upon completion of this
development, the process was applied on an industrial scale.
Field Experience
The first extensive application of rotopeening was performed on the
Doel 4 and Tihange 3 steam generators in late 1984 and early 1985 (25.20,
25.28). These plants had not been operated prior to peening. The Doel 4
effort involved peening of 14,500 hot leg transitions over a 28-day period.
The Tihange 3 effort involved peening of 29,000 hot and cold leg transitions
over a 21-day period. In each case the peening intensity was Almen 8A.
In-process verification testing at these plants showed that 10-20% of the
specimens developed cracks in accelerated (low stress threshold) MgC12
and sodium tetrathionate tests. This was consistent with experience during
the qualification tests. Through 1987, two operating cycles were completed
at these plants. PWSCC has been minor, indicating that rotopeening has
significantly reduced the initiation of PWSCC (25.23).
The first (and only) extensive application of rotopeening on an operating
plant in the United States was at the V. C. Summer station. V. C. Summer
went into commercial operation in January 1984, and PWSCC cracks were
found in the expanded areas in the tubesheet during the first refueling
outage. About half of the hot leg transitions were rotopeened during the
second outage. The remaining tubes were shot peened during the third
outage.
Rotopeening has also been performed on a number of operating French
plants. This peening is restricted to the area around the roll transitions.
Results of this experience indicate that rotopeening prevents initiation of
new cracks, but does not prevent growth of already existing cracks.
Summary
Rotopeening reduces the potential for PWSCC at expansion transitions
and tubesheet rolling irregularities. It has been applied at one operating
domestic plant and at several operating French plants. However, more
cracking of peened specimens was observed during rotopeening
development tests than during shot peening development tests. This
suggests that rotopeening may be less effective than shot peening. Further,
operating experience has shown that rotopeening stops formation of new
cracks, but does not arrest growth of existing cracks.
SHOT PEENING
The purpose of shot peening is to produce a thin layer of compressive
residual stress on the inside surface of the tube. The compressive stress
serves to inhibit the initiation of IGSCC, which only develops in the presence
of relatively high tensile stresses. Shot peening is performed by blasting
the tube's inside diameter with small diameter metal shot. Impact of the
shot on the tube's ID surface produces a thin, work-hardened, compressive
stress layer. There are major concerns with this remedial measure such as
potential for increasing tensile stresses on the outside of the tube, the effect
of peening on pre-existing cracks, and radiation exposure/cleanup problems
during application.
Process Description
The operation of typical shot peening equipment is illustrated in
Figure 25-3. The equipment consists of a compressed air system, a shot
supply system, a peening tool which is inserted into the tube, a vacuum
collection system, and fixtures to move the peening tool from hole to hole
and vertically past the area being treated.
FIGURE 25-3
Typical Shot Peening Equipment
/ Brush
Brosse
Spray Nozzle
Storage Tank / ý/// ý//•/ ue
ý///sBuse
T
Tremie de Stockage Separator Vacuum
e// eSeparateur Aspirateur
Sealing Device Discharge Hose Filter

Tubulure Retour Filtre
Compressed Air Boite d'Etancheite
Air Comprime
Filling Valve
VanneRearplissge
Vacuum Pump
Pressure Pot PmeaVd
Pot de Mise en PrVmdn
Pusher Puller
Tireur Pous e Collector
[Z1 .. Recuperateur
Heads Injection Valve

Vanne Injection ..
0
The peening tool proper consists of a hollow cylinder through which

the shot is blown, a deflector cone or nozzle which directs the shot against
the tube's inside wall, and a seal above the cone or nozzle to prevent shot
from getting out of the area being treated. After striking the tube's inside
surface, the shot drops to the bottom of the tube where it is picked up by
the vacuum collection system.
Initial development work was carried out with glass and ceramic
(zirconium and silicon oxide) shot. Typical shot diameters were on the
order of 0.008-0.010 inches (0.20-0.25 mm). Testing showed that glass shot
experienced excessive fragmentation and that ceramic shot particles become
embedded in the tube wall. Further development work led to the selection
of small diameter corrosion-resistant metal alloy shot.
Process control is provided by periodic peening of verification
specimens. One type of verification specimen is checked for Almen peening
intensity. Another type of verification specimen consists of subjecting
sensitized alloy 600 tubes with roll expansions to sodium tetrathionate tests
after peening to verify that ID tensile stresses are below a low threshold
stress.
Shot peening in Europe is typically performed over a zone extending
from 2.0 inches (50 mm) above the transition to 3.5 inches (90 mm) below
the transition. Shot peening in the United States has generally extended
from the bottom of the tubesheet to 2.0 inches (50 mm) above the expansion
transition.
Shot peening qualification tests have been performed by several
organizations (25.29-25.32). This development work has included a wide
range of tube roll expansion conditions such as full-depth roll, "kiss roll",
over expanded "kiss roll", range of gap dimensions, and over rolling. The
qualification effort has included the following types of tests:
* accelerated SCC testing: stainless tubes in boiling MgC12,
* accelerated SCC testing: sensitized alloy 600 tubes in sodium
tetrathionate,
" accelerated SCC testing: alloy 600 tubes in 10% caustic with internal
pressure,
" strain gage measurements on the tube's OD surface,
* x-ray stress measurements on the tube's ID surface,
* effect of peening on precracked tubes, and
* leak tests after peening.
Testing of Uncracked Roll Transition Specimens. Testing of uncracked

specimens showed that a lower bound Almen intensity of about 2.5N
prevents cracking on the inside of the tube. Most testing work was carried
out at an Almen intensity of 4N. At this intensity there was no sign of
cracking in any of the boiling magnesium chloride and sodium tetrathionate
tests which indicates that residual tensile stresses on the tube ID after
peening are less than the 55-90 MPa (8-13 ksi) stress threshold. X-ray
measurements after peening showed that the maximum ID compressive
stress is about -600 MPa (-87 ksi) and that the depth of the compressive
stress region is 0.004-0.005 inches (100-120 microns) (25.30).
Peening with an Almen intensity of 4N produced greater amounts of
cracking on the tube OD surface in sodium tetrathionate tests than for the
unpressurized reference specimens (25.30). This showed that this peening
intensity results in total tensile stresses on the tube OD exceeding the stress
threshold. However, since alloy 600 cracks at a very low stress threshold in
sodium tetrathionate (55-70 MPa [8-10 ksi]), the magnitude of the stress
increase produced by the peening operation cannot be quantified from the
test results. Strain gage testing showed that the tensile stress on the tube
OD increases by only 20-35 MPa (3-5 ksi) for an Almen intensity of 4N.
The apparent discrepancy in this data results from the fact that the sodium
tetrathionate tests include the effect of residual fabrication stresses in the
tube wall while the strain gage tests only reflect the increase in stress
produced by peening.
Testing of Pre-Cracked Roll Transition Specimens. The most recent and

complete summary of shot peening development is contained in Reference
25.32.
Accelerated testing of unpressurized, precracked stainless steel tubes
in boiling MgCI2 and sensitized alloy 600 tubes in sodium tetrathionate
showed that shot peening prevented further growth of the cracks (25.30).
It was confirmed by dye penetrant tests and SEM examination that the
peening operation smears metal over the crack openings, and that the
compressive residual stresses in the surface layer resists formation of new
cracks.
Accelerated testing of pressurized pre-cracked specimens in 10% caustic
at 662°F (350'C) and 180 bars (2,610 psi), showed that peening prevents
initiation of new cracks, but has no beneficial or detrimental effect on growth
of pre-existing cracks (25.30). The most likely explanation for this behavior
is that the hoop stress produced by the internal pressure in the tube opens
up the pre-existing cracks allowing the liquid in the tube to penetrate
through to the portion of the tube wall which is still in a state of tensile
stress.
Pre-cracked specimens were leak tested in both the peened and

unpeened condition to determine the effect of peening on leak rates. Test
results showed that the pressure required to produce leakage after peening
is typically about ten times the pressure required to produce leakage before
peening. However, there were cases where the pressure required to produce
leakage did not increase at all (25.30).
Field Experience
Shot peening was first performed at Doel 3 during 1985 (25.30). Doel 3
started up in 1982, experienced primary side cracking at expansion
irregularities in tubes located in oversize holes within the first operating
cycle, and the hot leg expansion transitions were shot peened in July 1985.
The required time to treat all hot leg tubes in Doel 3 ranged from 15-20
days per steam generator, and the total dose averaged 30 Rem per steam
generator (25.30). This has been reduced on subsequent plants.
The hot leg transitions at Doel 3 were examined several times after
peening. These examinations showed that, while the rate of new crack
indications decreased, the rate did not go to zero, and many pre-existing
cracks had increased in length (25.33-25.34). There was also a forced outage
during the two years of operation caused by excessive leakage from long
(0.71-0.87 inch [18-22 mmi) cracks (25.23). This experience is consistent with
the premise that shot peening prevents initiation of new cracks, but does
not prevent pre-existing cracks over some critical size from growing after
peening. As a result of this experience, Doel 3 has considered other remedial
measures such as sleeving, nickel plating, or steam generator replacement.
Operation for three years after shot peening at Ringhals 3 and 4 and at
several French plants, also suggests that shot peening does not completely
arrest crack growth in plants which were operated prior to shot peening
(25.34).
Summary
Laboratory testing shows that shot peening should be an effective means
of preventing expansion transition cracking in new plants. However, it is
not capable of completely arresting the growth of pre-existing cracks. This
has been confirmed by field experience.
THERMAL STRESS RELIEF AND HEAT

TREATMENT
The purposes of thermal stress relief and heat treatment are to relieve/
reduce tensile residual stresses produced by tube bundle fabrication
operations, and, possibly, to improve the metallurgical structure and thereby
increase the resistance of the tubing material to PWSCC. A short time at
about 1,292°F (700'C) is sufficient to stress relieve the joints, while
metallurgical improvement requires eight to ten hours at a temperature of
about 1,130°F (610 0C). Thermal stress relief and heat treatment can be
performed by (1) electric resistance heating of U-bends, (2) electric resistance
heating of dented tube support plate intersections, (3) induction heating of
expansion transitions, and (4) global heating of the entire tubesheet region.
Global heat treatment of the entire tube bundle might also be possible.
Required Time and Temperature for Stress Relief

The temperature level and the time that the tube is held at temperature
are the key variables in obtaining an effective stress relief of alloy 600 tubing.
Standard references such as the Metals Handbook (25.35) indicate that the
time and temperature must be determined experimentally for each
application but suggest typical conditions of 1,400-1,600'F (760-870'C) for
1-2 minutes. In fact, there are a number of constraints on the stress relief
conditions which are summarized in an EPRI report on in-situ stress relief
of U-bends and roll transitions (25.36).
Tubing Stress Relief. A number of laboratories have carried out accelerated

tests to determine the time and temperature conditions which eliminate
cracking in roll expanded tubing (25.1M, 25.37-25.40). The lower bound of
these test results is shown as the "Effective Stress Relief" line in Figures 25-4a
and 4b. Time and temperature conditions below this line resulted in cracking
under accelerated test conditions, and time/temperature conditions above
this line resulted in the tubes not cracking under the accelerated testing.
Time and temperature conditions slightly below the curve may also produce
an acceptable stress relief since the stress threshold for cracking in operation
is higher than in the accelerated tests.
Tubing Sensitization. Holding alloy 600 tubing in the stress relief

temperature range for more than about 10 minutes may result in formation
of chromium carbides, which locally deplete the chromium content at the
grain boundaries. This phenomenon is called sensitization, and it results
in the potential for sulfur species attack such as occurred on the primary
side of TMI-1 steam generator tubing and on the secondary side of ANO 1
steam generator tubing. These plants have once-through steam generators
which were stress relieved, and thereby sensitized, during fabrication.
However, other plants with once-through steam generators and Tihange 1
have operated satisfactorily with sensitized tubing by avoiding oxidizing
conditions. Accordingly, for a plant which has significant PWSCC, the risk
associated with sensitization may be warranted.
Work by Airey (25.17) shows that alloy 600 tubing is sensitized by
holding it within the "Sensitization" range shown on Figures 25-4a and 4b.
However, if the tubing is held in the sensitization temperature range many
hours, the chromium is allowed to diffuse within the grain and restore the
chromium concentration at the grain boundaries, thereby reversing the
detrimental effect of sensitization. This is the basis for the thermal treatment
of alloy 600 tubing and for the Framatome long-term stress relief.
Tubing Recrystalization and Grain Growth. If the tube temperature is

increased too much, recrystalization and grain growth will occur. These
phenomena might make the material more susceptible to stress corrosion
cra'cking. The "Recrystalization/Grain Growth" lines in Figures 25-4a and
4b were selected to avoid recrystalization and grain growth. They are based
on data from the Metals Handbook (25.35) and on results of Westinghouse
and Brookhaven National Laboratory tests associated with stress relief heat
treatment (25.37-25.38).
Tubesheet Transformation and Fabrication Stress Relief Temperatures.

For the case of expansion transition stress relief, the effect of the temperature
on the tubesheet material must also be considered. If the temperature of
the alloy steel tubesheet exceeds the transformation temperature, martensite
can be formed upon cooling. The transformation temperatures for typical
tubesheet materials are: ASTM A-336 (1,480'F [804'C]) and ASTM A-508
C12 (1,292°F [700°C]) (25.37). The "Tubesheet Transformation Temperature"
line in Figures 25-4a and 4b is taken as the lower bound of these two
temperatures.
During long-term heat treatment, temperatures must be held below
the tubesheet fabrication stress relief temperature, which is typically (1,150°F
[621°C]), in order to avoid degradation of material properties. Temperatures
during short-term induction stress relief can exceed the tubesheet fabrication
stress relief temperature but are believed to be acceptable because (1) the
duration is short, (2) only a small volume of the tubesheet is affected, and
(3) there is a significant temperature drop between the tube wall and the
tubesheet. Nevertheless, it must be demonstrated that the higher
temperature is technically acceptable.
FIGURE 25-4
Suggested Time and Temperature Target Conditions for Stress Relief
of Alloy 600 U-Bends and Expansion Transitions
900
-1600
.1500
800
1400 •.-
a) U-Bends •700 1300

-. 3• CU
(I-
E
!2o
1200 E~
600
1100
1000
500 ,1
0.1 1.0 10.0 100 2. 1000
Time (min)
. I
900
1600
1500
800
a' 1400 '
b) Tubesheet 6 700 1300

'.. .3
a
1200 E
600
1100
1000
500
0.1 1.0 10.0 100 2. 1000
Time (min)
Notes on Limits
" Effective stress relief
" Sensitization
" Recrystalization/grain growth
• Tubesheet transformation temperature
" Practical time for in-situ work
Summary of Stress Relief Time and Temperature Guidelines. Figures

25-4a and 4b summarize the approximate limits on time and temperature
conditions for local stress relief of alloy 600 tubes at U-bends and roll
transition expansions. The figures also show a suggested range of time
and temperature conditions for each local stress relief method. This
information is intended as guidance in understanding the phenomenon
and limiting conditions involved and should not be interpreted as firm
requirements. Instead, firm requirements must be established on a case-
by-case basis taking into account characteristics of the subject plant, the
equipment to be used, and the vendor's process qualification test program.
Required Time and Temperature for Metallurgical Improvement

Tests conducted in Belgium have shown that heat treatment for 9 hours
at a temperature of 1,130°F (610'C) results in beneficial grain boundary
carbide precipitation, which is expected to improve resistance to PWSCC
(25.40). As an upper limit, the maximum temperature in the tubesheet
should not exceed the temperature at which the tubesheet was initially stress
relieved. As previously noted, this is typically about 1,150°F (621'C).
Laborelec has concluded that this approach will increase the time-to-
cracking by a factor of at least 4.
High-temperature caustic tests of rolled specimens indicate a factor of
4-7 improvement in time-to-cracking. High-temperature pure water tests
using reverse U-bends indicate a factor of 6-30 times improvement.
In-Situ Stress Relief and Heat Treatment Methods

Procedures have been developed for three in-situ stress relief processes:
(1) local resistance heating stress relief of U-bends; (2) local induction heating
stress relief of roll transitions; and (3) global heat treatment of the entire
tubesheet region, including all roll transitions. In addition, it is understood
that a proprietary process has been developed for local stress relief of dented
areas, and that this process has been used on a trial basis at North Anna 1.
U-Bend Stress Relief. Development of equipment and procedures for

stress relieving U-bends is covered in a number of references (536, 25.x
25.41-25.42). Equipment for U-bend stress relief consists of a flexible heating
element which is inserted into the tube as shown in Figure 25-5. Stress
relieving is accomplished by applying power to the heater element for a
controlled period of time. The initial power settings and heating times are
established based on laboratory testing using instrumented tubes. The
power level and heating time must be fine tuned in the field to compensate
for temperature gradients and air currents within the steam generator, and
actual tube emissivities. The tube wall temperature can be measured

indirectly from the applied power and temperature measured by a
thermocouple embedded within the heater element proper. The tube wall
temperature can also be measured directly by heating the tube, quickly
moving the heater out of the way, and measuring the tube wall temperature
using an optical temperature measuring device mounted on the cable just
behind the heating element.
There are few potential problems with U-bend stress relief. As outlined
in Reference 25.36, the main concerns are temperature control, the potential
for metallurgical damage produced by surface contamination, and the
possibility that pre-existing cracks could deteriorate further. Vendors must
demonstrate that their processes address these concerns satisfactorily.
FIGURE 25-5
Typical U-Bend Stress Relief Arrangement
U-bend stress relief is currently offered by several vendors and has been
performed on row I and 2 tubes of many steam generators. The first U-bend
stress relief at an operating plant was performed at Ringhals 3 and 4 in
1986.
Some test results associated with U-bend stress relief are documented
below (25.23).
Steam Tests Associated With Ringhals U-Bend Heat Treatment.
Ringhals 3 and 4 U-bends were heat treated in 1986 at target conditions
of 1,450'F (788°C) for 5 minutes. Due to temperature control problems
associated with emissivity variations, 752°F (400°C) steam tests were
conducted on row 2 U-bends heat treated at a range of temperatures
(Table 25-3).
TABLE 25-3
Steam Tests Associated With Ringhals U-Bend Heat Treatment
Stress Relief Temperature 'F (°C) Results

None Failed at 100 hours
1,450 (788) No failure in 1,000 hours
1,550 (843) No failure in 1,000 hours
1,660 (904) Failed at 1,000 hours
Caustic Tests of Vallourec and Westinghouse U-Bends. Framatome

has performed tests of stress relieved Vallourec and Westinghouse
U-bends in 10% caustic at 662°F (350 0C) with a 100 bar (1,450 psi)
differential pressure. Specimens were stress relieved for two different
durations (5 minutes and 15 hours) at a temperature of 1,319'F (715°C).
The purpose of the longer duration thermal treatment is to induce
carbide precipitation to improve SCC resistance, and to increase the
grain boundary chromium concentration over time, thereby avoiding
sensitization. These tests showed that untreated specimens developed
cracks in 200 hours. Vallourec U-bends did not crack, given either heat
treatment, and a Westinghouse U-bend, given the short-term heat
treatment, cracked. This work suggests that the required heat treating
temperature may vary with the tube material and fabrication method.
Local Expansion Transition Stress Relief. Development work on local

expansion transition stress relief has focused on the induction heating
method. Typical equipment required for this procedure is shown in
Figure 25-6. It consists of an induction heating probe which fits into the
tube, a radio frequency power supply, and an optical temperature indicating
and recording system. Induction heating is used rather than electric
resistance heating, since it has the capability to rapidly heat the tube wall
without heating up large portions of the tubesheet.
FIGURE 25-6
Typical Equipment Arrangement for Expansion Transition
Stress Relief
Induction Probe
Expansion
Transition
Mounting Plate
Initial development of the induction heating process was performed

by Brookhaven National Laboratory under EPRI/SGOG sponsorship (25.37,
25.4 25.44). This work showed that in-situ induction heating stress relief
of new steam generators should be practical. The initial field application
of in-situ induction heating stress relief was carried out on mini-sleeves at
Doel 2 (539 25.45). This operation confirmed that in-situ induction heating
stress relief should be practical, but pointed out that several aspects of the
process require further development effort.
Temperature Control. Use of fiber optics has been demonstrated to

be a practical method for measuring temperatures achieved during
induction heating. Experience at Doel 2 indicated, however, that
temperatures measured at one location may not accurately reflect
temperatures developed at other locations due to variations in heat
transfer properties in the gap between the tube and tubesheet.
Specifically, the temperature probe at Doel 2 was directed towards the
unexpanded portion of the tube above the transition. The Doel 2
experience led to the following observations:
* If the crevice behind the tube wall at the elevation of the temperature
probe was empty, relatively low power was required to reach the
desired temperature at the tube wall. Consequently, the expanded
portion of the tube, which was in contact with the tubesheet, did
not reach the desired stress relief temperature due to the good heat
transfer between the tube and tubesheet.
* If the crevice behind the tube wall at the elevation of the temperature
probe contained a trapped water pocket, the heat transfer in the
expanded and unexpanded region was better matched such that
the expanded region received a better stress relief.
* If the temperature had been controlled by a probe at the elevation
of the expanded region rather than at the tube above the expanded
region, the upper portion of the tube could have been heated well
above the desired temperature in the event that the crevice behind
the tube wall was empty.
In light of the above, additional development is required to provide
accurate temperature control considering variations in tube roll
transition geometry and crevice heat transfer properties. The desirable
range of time/temperature conditions for expansion transition stress
relief are shown in Figure 25-4b.
Surface Contamination. The potential for cracking or grain boundary
attack resulting from applying high stress relief temperatures to
contaminated surfaces has not yet been fully assessed. Westinghouse
has performed some tests on simulated sludge. These tests did not
indicate a problem; however, additional testing may still be required
(25.36).
Local Buckling and/or Tensile Residual Stresses. If the tube being
stress relieved is locked axially at both ends, there is a potential for
local buckling and/or tensile residual stresses. "Locking" can occur in
a part-depth expansion tube which is locked by denting at the top of
the tubesheet, or in a full-depth expansion tube which is locked by
denting at the lower tube support plate. It is postulated that under

these conditions, the tube expands axially during heating, buckles or
yields due to restraint at the ends, and then develops residual axial
tensile stresses during subsequent cooldown. The potential for this to
occur, and the resultant effect on primary and secondary side IGSCC;
must be resolved by testing.
Reduction in Pull-Out Force. When an expanded area is subjected to
induction heating, the resultant tube expansion, yielding and shrinkage
upon cooling, will result in some loss of interference force holding the
tube in the tubesheet. The effect of loss of interference force must be
evaluated on a case-by-case basis. Some guidance in this regard is
included ih Appendix A of Reference 25.36.
Development of Tube-to-Tubesheet Crevice. In addition to reducing
pullout force, induction heating will cause a crevice of a few mills to
open up between the tube and the tubesheet. This crevice could be as
long as 0.25 inches (6 mm) or more, depending upon the details of the
induction heating probe. The acceptability of this crevice from the
standpoint of secondary side corrosion must be verified.
Propagation of Pre-Existing Cracks. Development work has not yet
been accomplished to assess the effect of induction heating stress relief
on pre-existing cracks. However, the effect on typical, short,
longitudinal transition region cracks is expected to be low, and the lower
residual stresses produced by the stress relief would be expected to
reduce the amount of subsequent crack growth. The effects of pre-
existing circumferential cracks would be more significant, especially if
axial forces are produced by locking.
Global Tubesheet Stress Relief. Global heat treatment of the entire

tubesheet has been considered for plants in commercial operation in Belgium
(25.40). This procedure involves heating the tubesheet to about 1,130°F
(610 0C) and holding the temperature for 9 hours. The procedure is designed
to increase the resistance to PWSCC by increasing the grain boundary
carbide precipitation, and possibly by some reduction in residual stresses.
As indicated by the data in Figures 25-4a and 4b, this temperature is too
low to achieve complete stress relief.
There are several reasons why this procedure has been considered over
alternative remedial measures (local stress relief, shot peening, and
rotopeening).
* Time-consuming remedial measures are less attractive for plants which

have been in operation due to increased radiation exposure.
* Leaks and cracks at various elevations within the tubesheet may require
that remedial measures be performed over the full tubesheet height.
* Tests indicate that typical heats of tubing from candidate plants will
experience desirable grain boundary carbide precipitation.
Two methods of applying heat have been evaluated: electric strip
heaters, and electric strip heaters together with circulating hot gas. Use of
electric strip heaters without hot gas is considered to be the better approach
for tubesheet area stress relief due to its simplicity As presently envisioned,
the entire lower head, including the divider plate, would be heated to avoid
potentially harmful differential thermal expansion; heaters would be
extended along the outer walls for a length of about one steam generator
diameter past the top of the tubesheet in order to minimize temperature
gradients and internal distortions; and the secondary side of the steam
generator would be evacuated in order to reduce convective heat loss from
the top of the tubesheet.
For Doel 2, the lead plant considered for global stress relief, the
maximum stress relief temperature is 1,130°F (610'C). To achieve the
minimum desired 1,022°F (550'C) temperature on the top face of the
tubesheet without exceeding the maximum temperature on the heated
bottom face of the tubesheet, the temperature gradient across the tubesheet
must be limited to approximately 108'F (60'C). It is understood that a
gradient of 198°F (110°C) was achieved across the tubesheet during in-situ
stress relief of the Turkey Point steam generators. This was achieved by
using electrical resistance heaters on the lower face of the tubesheet. The
high temperature gradient results largely from convection and radiation
losses from the upper surface of the tubesheet. It has also been reported
that the temperature gradient across a tubesheet on an inverted steam
generator with the tubesheet heated from the primary side was 18'F (10°C).
This shows that the temperature gradient should be small, if convection
losses are kept low. In planning for a global stress relief of the Doel 3
tubesheet, Tractable considered drawing a vacuum on the secondary side
of the tubesheet to eliminate convection losses and thereby approximating
the case of the inverted steam generator with the tubesheet heated from the
top.
A major concern with this remedial measure is that the tubes subjected
to stress relief temperatures may become sensitized, and thus become
susceptible to sulfur species attack such as occurred on the TMI 1 primary
side and the ANO 1 secondary side. However, other plants with once-
through steamgenerators and Tihange 1 have operated satisfactorily with
sensitized tubing by avoiding oxidizing conditions. Accordingly, for a plant

which has significant PWSCC, the risk associated with sensitization may
be warranted. There is also some risk associated with surface contamination
as indicated for in-situ U-bend stress relief.
The most recent and complete summary of global stress relief is
contained in Reference 25.40.
Dented Tube Support Plate Intersection Stress Relief. Stress relief of

dented tube support plate intersections appears to be a desirable process
for some plants with susceptible tubing and dented intersections. A
proprietary procedure for stress relieving individual dented intersections
has been used on a trial basis at North Anna Unit 1. A global stress relief of
the entire tube bundle may also be a possible solution to the problem, but
this approach has not yet been developed.
Summary
Electrical resistance stress relief of U-bends is a proven method which
has been widely applied in the field.
Thermal stress relief or heat treating of expansion transitions holds
promise as being a good long-term solution to expansion transition cracking,
since it has the potential of significantly reducing tensile stresses and
improving the resistance of the material to PWSCC. This benefit is obtained
without increasing the tensile residual stresses throughout the rest of the
tube wall as is the case with shot peening and rotopeening; in fact, it would
reduce OD stresses, thus reducing susceptibility to secondary side SCC. It
would also reduce stresses acting to make existing cracks grow larger.
However, some development work remains to be completed and it is not
yet ready for remedial application.
Global heat treatment of the entire tubesheet region has been developed
to the point where it appears to be a practical option. However, it has the
disadvantage of sensitizing the tubing which must be considered before
implementing the procedure.
ELECTROPLATING
The purposes of electroplating the inside surface of the tube with a
thin layer of PWSCC resistant nickel are two-fold: to stop leaks and to
prevent further PWSCC. For this remedial measure to be practical the base
tube must be structurally sound, i.e., any cracks must be of an acceptable
size in regard to the tube's burst properties.
Process Description
Initial development of the electroplating process was carried out by
Laborelec (25.46). The selected process involves the following steps:
* cleaning the tube ID using brushes,
" electrochemically cleaning the area to be plated,
* preplating with a "strike" solution,
" depositing nickel plating using a circulating bath and soluble anode,
and
* cleaning the tube's ID prior to returning to service.
The equipment arrangement within the tube is shown in Figure 25-7.
FIGURE 25.7
Typical Equipment Arrangement During Electroplating
Gas Outlet
Upper Head
+ - G Tube
-- J L_,_Lt_
Lower Head
Solution Inlet
Results of laboratory testing on more than 500 tubes showed that
electroplating is practical (25.46). The plating can be applied over lengths
of at least 5.5 inches (140 mm), the plating thickness can be up to 0.011 inches
(280 microns), and the plating can bridge existing cracks up to 0.001 inches
(30 microns) wide. The major problem identified in this work was obtaining
surfaces which are uniformly free from pitting and roughness. Another
concern is developing an effective method to inspect the tubes in the plated
area since nickel plating makes the eddy current method ineffective.
Laborelec has developed an ultrasonic inspection method to resolve this
concern.
Tests were conducted to assess the ability to seal existing cracks under
high pressure. Non-plated cracked tubes were found to leak under
15-290 psi (1-20 bars) internal pressure. After plating to 0.002-0.008 inches
(50-200 microns) thickness, tubes were leak tight under a pressure of
1,740 psi (120 bars), except when pits were present in the bridged areas.
Field Experience
A total of 90 tubes were nickel plated at Doel 2 in 1985 and 1986. Visual
examination of the tubes after plating showed that the surface was similar
to that obtained in the laboratory, except for some abnormalities observed
on 70% of the tubes near the roll transitions. Some of the plated tubes have
been pulled and inspection has shown cracking of the nickel plating for
cases where the initial cracks were greater than 0.35 inches (9 mm) in length.
This problem has been addressed by using nickel of greater ductility (25.47-
25.48).
Tubes were also nickel plated in Doel 3 in 1988. Two hundred and fifty
tubes in the Doel 2 steam generator, which were explosively re-expanded
in 1990, will be nickel plated to protect against PWSCC.
Summary
Laborelec has stated that nickel plating, using ductile nickel, is ready
for commercial application.
RE-EXPANSION
Hydraulic or explosive re-expansion is intended to close up the
tubesheet crevice in the case of part-depth roll expansions, to improve the
geometry and reduce tensile residual stresses in roll expanded regions, hnd
to improve the geometry and reduce stresses at expansion transitions.
For plants with part-length rolled tubes which are experiencing primary
side cracking at roll transitions, expansion of the tube for all or part of the
tubesheet depth is a possible repair approach. This method has been used
for all part-length rolled plants in Japan. The tubes were hydraulically
expanded for most of the tubesheet height, placing a new expansion
transition 2-3 inches (50-75 mm) below the top of the tubesheet. A trial
part-length expansion was tested on a number of tubes at Doel 2 in 1980.
This consisted of a length of hydraulic expansion with a short length of
mechanical rolling centered in the hydraulically expanded area. This repair
approach was not implemented on a wide scale because of application
difficulties and because of concerns that PWSCC might recur at the new
expansion transitions. Whatever expansion method is selected, it is
important that the new expansion transition has low, tensile residual stresses,
or that the expansion transition region be protected from the primary water
by nickel plating as is being performed at Doel 2. The Belgians have reported
that tests of re-expanded tubes have shown that the residual stresses
associated with the new expansion transition can be high enough to result
in further PWSCC (25.49). However, as previously mentioned, the Belgians
plan to cover the expanded regions with nickel to protect against PWSCC
initiation.
For plants with a poor initial full-depth expansion (skip rolls and poor
overlaps, etc.), re-expansion of the tube over the full depth of the tubesheet
has the potential to correct areas improperly expanded during rolling (which
lead to high local residual stresses), and to improve the local geometry and
state of stress in the transitionregion. EPRI has sponsored work to assess
explosive re-expansion. Qualification testing included sensitized alloy 600
tubing in sodium tetrathionate to determine the degree of improvement.
Results of the testing by Foster-Wheeler showed that explosive re-expansion
can reduce the susceptibility to PWSCC at expansion transitions and
expansion irregularities within the tubesheet. However, the testing only
covered the Foster-Wheeler explosive re-expansion process and did not
include any pre-cracked tubes (25.50).
PLUGGING
Plugging is used to seal leaking tubes or tubes with indications in excess
of the acceptance criteria. This will prevent primary water from coming
into contact with susceptible tubing, such as row 1 and 2 U-bends, thereby
avoiding degradation leading to forced outages. In addition, plugging will
provide time to make plans to apply a more permanent remedial measure.
Plugging is the standard remedial measure in the United States for most
leaks caused by PWSCC. Installation of plugs is a sensible short term
approach to get a plant back on line if cracks are discovered. However,

plugging is not a good long-term solution for more widespread problems
since it reduces the tube margin and treats the symptoms rather than the
cause.
Commonly used plugging methods include welded plugs, explosive
plugs, and mechanical plugs. Below is a description of the major plug types.
Welded Plugs
Welded plugs are typically inserted into the tube and welded to the
tube wall. Filler metal is either integral with the plug or added externally.
For the case of plugs with integral filler metal, only heat is required to make
the weld. Systems are available for manual, semiautomatic, and automatic
welding of plugs.
Explosive Plugs
Plugs can also be installed by explosively expanding a capped hollow
thimble into the tube. The explosive can be contained within the thimble
or inserted into an otherwise hollow thimble. It may also be necessary to
seal weld the plug to the tube wall.
Utilities have experienced leaks in explosive plugs. It is speculated
that the cracking and leaking of explosively installed plugs are the result of
large plastic strains and high residual stresses in the plug wall which leads
to PWSCC. This may result in leakage of tubes which were plugged to seal
leaks as opposed to those plugged preventively.
Experience has shown that leaking explosive plugs do not generally
open up and result in an increased leak rate. Instead, they leak at a steady
rate over a long period of time. During reverse leak testing, (secondary
side pressurized and primary side vented), the reverse pressure typically'
closes small cracks in the tubes and plugs, thereby preventing leakage.
Therefore, it has often proven easier to identify leaking plugs using only a
static head of water on the secondary side.
Mechanical Plugs
Mechanical plugs made from thermally treated alloy 600 bar stock, and
which are installed without explosives or welding, are also offered by several
vendors. These plugs are rolled or mechanically expanded into the tubes
and are removed when necessary by a weld shrinkage technique. Rolled
or expanded mechanical plugs are currently the industry standard approach;
however, some of these plugs have experienced significant problems with
PWSCC. For example, one of the mechanical plugs at North Anna
developed a circumferential crack, the plug separated into two parts, and
the end of the plug was projected up through the tube and broke through
the tube in the U-bend region causing a leak and plant shutdown (25.24-
25.65). The current industry approach is to remove plugs made with
susceptible heats of alloy 600 material and replace them with plugs made
from thermally treated alloy 690.
Memory Metal Plugs

Framatome has developed a removable plug using a shaped memory
alloy. This plug is installed in the steam generator cold and then transforms
to its final fit when the plant goes up to temperature. The plug can be
removed with a relatively low axial force (4 tonnes) during an outage since
the friction load which the plug can produce cold is limited by the low cold
condition yield strength of the alloy.
Summary
Plugs are currently offered by a number of vendors. When selecting a
particular vendor and plug design, consideration should be given to
qualification testing to ensure that the plugs will not develop leaks due to
tube cracking at the plug, or as a result of cracking within the plug itself. In
addition, there should be provisions to remove the plugs in the future.
SLEEVING
The purpose of sleeving as a PWSCC remedial measure is to install a
smaller diameter tube (sleeve) inside the original damaged tube to bridge
the damaged location. A typical sleeve application is shown in Figure 25-8.
The sleeve provides a seal to prevent primary water from leaking into the
secondary side. It also provides a redundant load carrying path which will
prevent tube rupture in the event that the original tube fails. The primary
concerns are: (1) to use sleeve materials which are resistant to PWSCC and
to OD IGSCC/IGA, (2) to develop methods of joining the sleeve to the tube
without generating large tensile stresses that could result in cracking of the
parent tube at the joints, and (3) to develop methods to inspect the parent
tube behind the sleeve. Sleeving is discussed in References 25.50-25.59. An
EPRI sleeving design review checklist is provided in Reference 25.60.
FIGURE 25-8
Illustration of Typical Sleeve Application
Upper Joint
* Sleeve Expanded to Contact Tube
* Sleeve May Be Welded to Tube
* Joint May Be Stress Relieved
Tube
- Alloy 600
Sleeve
- Thermally Treated Alloy 690
Not to Scale
Tubesheet
Crevice (Older Plants)
Lower Joint
- Sleeve Expanded to Contact Tube
* Sleeve May Be Welded to Tube
Sleeve Designs
As shown in Figure 25-9, sleeves can be used to repair different types
of damage:
* secondary. side IGA and IGSCC in the tubesheet crevice,
* secondary side pitting and IGAin the sludge pile region,
* secondary side wastage and IGA at higher locations in the tube bundle,
and
* primary side IGSCC at expansion transitions and at rolled areas within
the tubesheet.
FIGURE 25-9
Typical Problems Addressed by Sleeving Repairs
Sleeves have varied in length from 1.5 to 80 inches (40 mm to 2.0 meters)
depending upon the location and extent of the damage being repaired.
Sleeves installed to repair secondary side damage in the tubesheet crevice
and sludge pile regions are typically 20 to 80-inches (0.50 to 2.0 meters) in
length. Longer tubes can be used near the center of the bundle. Shorter
tubes are often required near the periphery since there is not enough
clearance within the plenum to install the longer tubes. Some vendors have
developed curved sleeves for use at peripheral locations; the sleeves are
straightened as they are inserted into the tube. Joints are typically made in
short straight sections at the ends of the curved sleeves. Shorter 12 inch
(0.30 meter) sleeves have been used to bridge damaged areas at tube support
plates, and very short 1.5 inch (40 mm) long mini sleeves have been used
on a trial basis at cracked roll transitions located within the tubesheet.
Sleeve material is selected to be compatible with the original tubing
and to have equal or improved resistance to primary side and secondary
side attack. The sleeve material currently being offered is thermally treated
alloy 690.
After a sleeve is installed in the tube it must be expanded and sealed at
the ends. This is accomplished by hydraulic expansion, explosive expansion,
rolling, welding, or combinations of these methods. Sleeve ends which
terminate above the tubesheet are generally expanded hydraulically or
explosively in order to close up the gap between the sleeve and tube in a
manner which minimizes residual stresses in the sleeve and tube. The small
gap between the sleeve and tube is then sealed by one of several methods:
(1) a hard mechanical roll at the center of the hydraulically expanded region,
(2) brazing the sleeve to the tube wall using a gold alloy braze, or (3) welding.
Sleeve ends which terminate in the tubesheet region are often expanded
explosively or by rolling since this expands the tube and produces a seal at
the same time. Sleeve ends terminating in the tubesheet can also be seal
welded.
Sleeve designs offered by a number of vendors are shown in
Figures 25-10 through 25-15.
Field Experience
Over 25,000 sleeves have been installed in U.S. steam generators, with
the greatest numbers installed at San Onofre 1, Millstone 2, Point Beach 2,
and Indian Point 3. Details of these applications are provided in an EPRI
report reviewing sleeving experience and providing a sleeving design
review checklist (25.60). In general, sleeving experience in the field has
been positive, although problems have been reported during sleeve
installation and subsequent plant operation.
FIGURE 25-10
Westinghouse Hybrid Expansion Sleeve
I
Plugging Limits
Apply to Tube
Only
Plugging Limits
Apply to Tube
Upper and Sleeve
Joint
Existing
Tube
Sleeve
*Plugging Limits
Apply to Sleeve
Only
Lower
RollT \ \ _" Plugging Limits Apply
Expansion_ to Sleeve and Tube
FIGURE 25-11
ASEA Brown Boveri Welded Sleeves
li Inconel690
Inoe600
Ii
IV
mo
/
I
InconeI600 I
Design 1 Design 2
FIGURE 25-12
Babcock & Wilcox Welded Sleeves for Recirculating Steam
Generators
7/8 Inch RSG Sleeve
11 Inch TSP SleeVw

11 Inch Sleeve
29 Inch Sleeve
,- Sludge Pile
3/4 Inch RSG Sleeve

FIGURE 25-13
Bechtel/KWU Welded Sleeves
Upper Weld ýOv erlap Weld
SG-Tube
Type I Sleeve
Weld
FIGURE 25-13 (cont'd)

Bechtel/ KWU Welded Sleeves
)Weld
Type IIISleeve
FIGURE 25-14
Combustion Engineering Welded Sleeves
Sleeve
. Cladding
Welded Steam Generator Sleeve

FIGURE 25-15
Framatome Welded Sleeves
0
]
-- o-
-14
Inadequate Heating of Brazed Joints. Some problems were experienced

in achieving adequate bond during early installation of brazed sleeves. The
most probable source of this problem was sludge buildup on the outside
surface of the tube which prevented proper brazing temperatures from being
reached. To solve this problem, brazing process control was improved and,
in some cases, "hybrid expansion joints" were used in lieu of brazing.
Dissolution at Brazed Joints. Some brazed joints experienced dissolution

of the sleeve and/or tube material due to attack by the brazing alloy. This
was apparently caused by overheating. This problem was ameliorated by
re-inspecting tubes and by plugging those that were damaged. This type
of problem can be avoided by improved process control.
Poor Brazed Joints Due to Axial Restraint. If differential expansion occurs

between the tube and sleeve during the brazing operation, a poor brazed
joint can result. This is a significant concern for tubes which are locked at
tube support plates due to denting. In this case, the sleeve is free to expand
as it heats up, but the tube is restrained from expanding by the dented
intersection. Westinghouse developed a procedure for this condition, which
involves pretensioning locked tubes using a proprietary process such that
relative axial motion between the tube and sleeve is minimized during
brazing.
Leakage at Hydraulically Expanded Joints. Hydraulically expanded

sleeve joints at Palisades had a small diametral gap between the sleeve and
tube after expansion. This resulted in a small primary to secondary side
leak if the tube had a through wall defect. If large numbers of tubes have
similar leakage, then the total primary to secondary leakage can exceed
technical specification limits. One solution to this problem is to add a
mechanically rolled joint at the center of the hydraulically expanded region.
The mechanical expansion reduces the leakage rate from about 5 cc/min to
less than 0.6 cc/min. However, for tubing which is susceptible to PWSCC,
the stresses induced in the parent tubing by the rolling operation may result
in relatively rapid PWSCC.
Primary Side IGSCC of Deformed Areas. Alloy 600 steam generator

tubing is known to be susceptible to PWSCC. If tensile residual stresses are
induced into the tubing during sleeving, the potential for primary side
IGSCC is increased. This problem occurred at Doel 2 where short "mini
sleeves" were explosively installed over cracked roll expansion transitions
within the tubesheet region. Laboratory testing and field experience showed
that this installation method can lead to PWSCC at the top and bottom of
the sleeve. An attempt at Doel 2 to stress relieve the mini sleeves was
unsuccessful due to temperature control problems caused by varying
thermal conductivity in the interface between the tube and tubesheet (25.61).
It is reasonable to assume that this problem could be resolved by improved
temperature control systems.
In light of the above concerns and the experience at Doel 2, it is
important that sleeve installation be performed using methods which have
been demonstrated by test to result in acceptably low residual stresses.
Comparative Sleeving Tests at Ringhals 2

The Swedish State Power Board conducted an extensive evaluation of
sleeves for Ringhals 2 (25.23). Mechanical, brazed, and welded sleeves were
evaluated. Results of the evaluation showed that welded sleeves are

preferable since they result in relatively low stresses and low sensitization.
Brazed joints also result in low stresses, but can also result in sensitization.
Thirty four sleeves of welded and brazed designs were installed in
Ringhals 2 in 1984. Subsequent to this initial work, more than 1,200 sleeves
of the C-E-welded type have been installed. Experience has generally been
good, although two of the sleeves were found to have defective welds due
to lack of fusion. This can be avoided by improved cleaning and use of
ultrasonic checks to verify proper fusion.
Recent review of the Ringhals sleeve qualification data suggests that
the residual stresses associated with the C-E-welded sleeve design could
be as high as the yield strength. Thus, welded joints in PWSCC susceptible
tubing should be stress relieved to extend their life beyond five years.
Summary
Field experience has shown that well designed and installed sleeves,
which do not introduce other problems, are an effective remedial measure
for PWSCC.
TEMPERATURE REDUCTION
Laboratory test data and service experience have shown that primary
side IGSCC is strongly influenced by temperature. It is, therefore,
conceivable that temperature reduction could be used as a PWSCC remedial
measure until more effective remedial measures can be applied, or until
the steam generator is replaced. However, reducing the primary side
temperature may involve a significant penalty in power output and
operating cost, and is generally avoided.
Laboratory Basis
Laboratory testing by Bandy and van Rooyen (25.25) has indicated that
the rate of PWSCC initiation varies like a standard thermally activated
process, i.e., in accordance with e-Q/RT, where Q is the activation energy of
about 40 kcal/mole, R is the gas constant (1.985x10- 3 kcal/°K mole), and T
is the absolute temperature in degrees Kelvin. Using this relationship, the
improvement in time to cracking obtained by decreasing the operating
temperature from a typical design hot leg temperature of 3240 (615°F) to a
reduced hot leg temperature is given by Figure 25-16.
As can be seen, a large measure of improvement requires a significant
decrease in hot leg temperature. This is not a practical long-term solution
due to the resultant reduction in power output. However, it may be worth
considering in some cases and for a limited period of time to reduce the
rate of damage while preparations are made for other remedial measures.
The data shows that it is desirable to operate at the lowest possible hot
leg temperature consistent with not lowering power output capability, since
even a 10'F (6QC) reduction in hot leg temperature should increase time to
PWSCC by 40%.
FIGURE 25-16
Effect of Reducing Hot Leg Temperature on Time to PWSCC
3
CO,
U-
0 -4-
550 560 570 580 590 600 610 620
Hot Leg Temperature (*F)
Field Experience
Most field experience with temperature reduction has been associated
with attempts to reduce secondary side degradation. The only case of
temperature reduction with the sole purpose of reducing PWSCC took place
in 1990 when EdF lowered the-hot leg temperature of some 1,300 MW units
from 623°F (328°C) to 616°F (325°C) to reduce the incidence of top-of-
tubesheet circumferential cracking (25.67). Yet, even this primary side
cracking was associated with dents induced by secondary side corrosion.
The degradation was remedied by a combination of temperature reduction,
chemical cleaning, sludge lancing, and ID shot peening.
HYDROGEN CONCENTRATION REDUCTION

Laboratory testing has shown that hydrogen accelerates the rate of
cracking in pure and primary water only moderately Accordingly, EPRI
recommends that, for plants with susceptible steam generator tubing, the
primary water hydrogen concentration be kept near the lower bound of
the specified primary water hydrogen limit to reduce the rate of PWSCC.
Laboratory Basis
Tests have shown that hydrogen added to pure water accelerates the
rate of cracking. Airey found that an increase from no hydrogen to 20 to
25 cc/kg hydrogen in elevated temperature water resulted in a decrease in
time to crack initiation of about 5 for material with moderate amounts of
cold work (25.62). EPRI-sponsored research indicates that the rate of PWSCC
attack increases linearly with the hydrogen concentration at elevated
temperatures (25.63-25.64). EdF has performed stress corrosion tests in 680 T
(3600C) primary water over the hydrogen concentration range of 1 to
8,320 cc/kg. Between the concentrations of 1 and 420 cc/kg (0 to I bar H 2
at 120 °C) cracking susceptibility increased with the addition of hydrogen.
With increased concentrations of 1,660 and at 320 cc/kg (4 and 20 bar H 2 at
120 °C) susceptibility decreased. Similar trends have been observed in pure
water (25.6 25.66) and steam environments (25.66).
Based on the above information, utilities have been encouraged to keep
the hydrogen concentration at the low end of the specified primary side
hydrogen, e. g., in the 25-35 cc/kg portion of the typical 25-50 cc/kg
allowable range.
Field Experience
There have been no published studies correlating effects of primary
side hydrogen concentration on the rate of PWSCC in the field.
SECONDARY SIDE WATER CHEMISTRY CONTROL

Denting Control
Based on a theoretical understanding of SCC mechanisms, even very
small amounts of denting would be expected to increase the potential for
SCC due to imposed strains. The purpose of secondary side water chemistry
control as a PWSCC remedial measure is to reduce the potential for denting
and thereby the potential for PWSCC at dented tube support plate
intersections. This is particularly important since sleeving of large numbers
of dented tube support plate intersections is impractical.
SECONDARY SIDE CHEMICAL CLEANING
Arrest Denting
The purpose of secondary side chemical cleaning as a PWSCC remedial
measure is to clean out crevices between the tube and tube support plates,
and thereby eliminate the load which produces high tensile stresses. If
denting has already occurred, it may be possible to thermally stress relieve
the tubes after chemical cleaning to reduce any remaining tensile residual
stresses (This approach has not been developed or definitely shown to be
practical). As with all chemical cleaning operations there is some risk that
chemical cleaning to prevent denting and PWSCC could damage the steam
generator in some unanticipated way. Accordingly, the risks and benefits
of chemical cleaning for the prevention of PWSCC would need to be
carefully evaluated prior to its field application.
REFERENCES
25.1 "Workshop Proceedings: Replacement/ Repair of Steam
Generators." EPRI NP-3207-SR. August 1983.
25.2 Workshop Proceedings: Steam Generator Repair/Replacement
Considerations. Steam Generator Owners Group Conference in
Seattle, WA, July 1985.
25.3 SGOG/EPRI Workshop: Management of Steam Generators
Susceptible to Primary Side Cracking: Remedial Measures and NDE
Methods. Steam Generator Owners Group Conference in St.
Petersburg, FL, December 1985.
25.4 "Proceedings: Primary Water Stress Corrosion Cracking: 1989 EPRI
Remedial Measures Workshop." EPRI NP-6719-SD. April 1990.
25.5 U.S. Nuclear Regulatory Commission. "Basis for Plugging
Degraded PWR Steam Generator Tubes." Regulatory Guide 1.121.
August 1976.
25.6 Leblois, C., and P. Hernalsteen. "Belgian Approach to Plugging
Criteria." In Proceedings: 1989 EPRI PWSCC Remedial Measures
Workshop. EPRI NP-6719. April 1990.
25.7 Hutin J. P., and P. Nicot. "EdF Steam Generator Surveillance and
Maintenance." In Proceedings:1989 EPRIPWSCCRemedial Measures
Workshop. EPRI NP-6719. April 1990.
25.8 Hunt, E. S., and J. A. Gorman. "Status and Suggested Course of
Action for Nondenting-Related Primary-Side IGSCC of
Westinghouse-Type Steam Generators." EPRI NP-4594-LD. May
1986.
25.9 P. G. Delozier, et al. "Study of the Environmental, Regulatory and

Occupational Constraints on Steam Generator Replacement, Repair
and Maintenance." Atomic Industrial Forum National
Environmental Studies Project. 1984.
25.10 Frederick G., and P. Hernalsteen. "Generic Preventive Actions for
Mitigating M. A. Inconel 600 Susceptibility to Pure Water Stress
Corrosion Cracking." Paper presented at October 1984 Specialist
Meeting on Steam Generators, Stockholm, Sweden.
25.11 Gorman, J. A., and E. S. Hunt. "Status of Cracking and Remedial
Measures for PWR Steam Generators With Full Depth Expanded
Tubing." EPRI NP-4459-LD. February 1986.
25.12 Hernalsteen, R, and R. Houben. "Preventive and Corrective Actions
for Doel 2 Steam Generators." Paper presented at October 1984
Specialist Meeting on Steam Generators, Stockholm, Sweden.
25.13 Kishida, A., et al. "The Causes and Remedial Measures of Steam
Generator Tube Intergranular Attack in Japanese PWR's." In
Proceedings of the Third InternationalSymposium on Environmental
Degradation of Materials in Nuclear Power Systems-Water Reactors.
(August 1987): 465.
25.14 Frederick, G., and P. Hernalsteen. "Comparative Evaluation of
Preventive Measures Against Primary Side Stress Corrosion
Cracking of Mill Annealed Inconel 600 Steam Generator Tubes."
Paper presented at SMIRT, August 1985.
25.15 Owens, C. M. "A Historical View of the Importance of the Final
Anneal on Primary Side SCC Resistance of Alloy 600 Steam
Generator Tubing." In Proceedings: 1983 Workshop on Primary-Side
Stress Corrosion Cracking of PWR Steam. Generator Tubing. EPRI
25.16 Owens, C. M. "Role of Processing on the Development of Alloy
600 Tubing Microstructures." In Proceedings: 1985 Workshop on
Primary-SideStress CorrosionCracking of PWR Steam GeneratorTubing.
1983.
25.18 Rouillon, Y., et al. "PWSCC Behavior of Thermally Treated Alloy
600." In Proceedings: 1987 EPRI Workshop on Mechanisms of Primary
Water IntergranularStress Corrosion Cracking. EPRI NP-5987-SP.
September 1988.
25.19 Garriga Majo, D., et al. "Prediction of the In-Service Behaviour of
Alloy 600 Tubes Used in Steam Generators of Pressurized Water
Reactors." Colloque International, Fontevraud II, SFEN, September
10-14, 1990.
25.20 Gold, R. E., et al. "Rotopeening Model E Steam Generator Tubing

at Doel 4 and Tihange 3." SGOG/EPRI Workshop: Management
of Steam Generators Susceptible to Primary Side Cracking:
Remedial Actions and NDE Methods, St. Petersburg, FL, December
1985.
25.21 Theus, G. J. "Stress Corrosion Cracking Tests of Alloy 600." In
Workshop Proceedings: U-Bend Tube Cracking in Steam Generators.
EPRI WS-80-136. June 1981.
25.22 Miglin, B. P., and Theus, G. J. "Stress Corrosion Cracking of Alloy
600 and 690 in All-Volatile-Treated Water at Elevated Temperature."
EPRI NP-5761-SP. May 1988.
25.23 Report of September 1987 trip to Sweden and France by A. R.
McIlree and J. A. Gorman dated November 17, 1987.
25.24 Gold, R. E., et. al. "PWSCC of Alloy 600 TT Mechanical Plugs:
Destructive Examinations of Plugs Removed From Service and the
Results of Laboratory Corrosion Tests." Colloque International,
Fontevraud II, SFEN, September 10-14, 1990.
25.25 Bandy, R., and D. van Rooyen. "Qualitative Examination of Stress
Corrosion Cracking of Alloy 600 in High Temperature Water - Work
During .1983." In Proceedings: 1983 Workshop on Primary-SideStress
Corrosion Cracking of PWR Steam GeneratorTubing. EPRI NP-5498.
November 1987.
25.26 Vidal, P. "Rotopeening of EdF Plants." SGOG/EPRI Workshop:
Management of Steam Generators Susceptible to Primary Side
Cracking: Remedial Actions and NDE Methods, St. Petersburg
Florida, December 1985.
25.27 Galper, M. J., and A. W. Klein. "SCE&G/Westinghouse use of Roto-
Peening at V. C. Summer." SGOG/EPRI Workshop: Management
1985.
Remedial Methods to Prevent Primary-Side Stress Corrosion
Cracking of Steam Generator Tubing: Rotopeening," Vol. 2. EPRI
NP-5249. June 1987.
25.29 Slama, G., and M. Sort. "Shot Peening of French Plants." SGOG/
EPRI Workshop: Management of Steam Generators Susceptible to
Primary Side Cracking: Remedial Actions and NDE Methods, St.
Petersburg Florida, December 1985.
25.30 Hernalsteen, P., et al. "Shot Peening of Doel 3 Plant." SGOG/EPRI
Workshop: Management of Steam Generators Susceptible to
25.31 "Shot Peening-Babcock & Wilcox and Metal Improvement

Company." SGOG/EPRI Workshop: Management of Steam
Generators Susceptible to Primary Side Cracking: Remedial Actions
and NDE Methods, St. Petersburg, FL, December 1985.
Cracking of Steam Generator Tubing: Shot Peening," Vol. 1. EPRI
NP-5249. June 1987.
25.33 Stubbe, J., et al. "Efficiency of Shot Peening to Prevent PWSCC of
Steam Generator Tubes-Examination of Tubes Removed From
Doel 3." Colloque International, Fontevraud II, SFEN, September
10-14, 1990.
25.34 Slama., G. "Shot Peening Experience After 5 Years." Colloque
International, Fontevraud II, SFEN, September 10-14, 1990.
25.35 "Heat Treating," Vol. 4. In Metals Handbook. 1981 Edition.
25.36 "Specifications for In-Situ Stress Relief of PWR Steam Generator
Tube U-Bends and Roll Transitions." EPRI NP-4364-LD. December
1985.
25.37 Woodward, J., and D. van Rooyen. "Stress Relief to Prevent Stress
Corrosion in the Transition Region of Expanded Alloy 600 Steam
Generator Tubing." EPRI NP-3055. May 1983.
25.38 Gilkison, J. M. "In Situ Heat Treatment and Polythionic Acid Testing
of Inconel 600 Row 1 Steam Generator U-Bends." EPRI NP-3056.
April 1983.
25.39 "Stress Relief Heat Treatment of Roll Transitions." SGOG/EPRI
Cracking of Steam Generator Tubing: Global Heat Treatment," Vol.
3. EPRI NP-5249. June 1987.
25.41 Pement, E W. "In Situ Heat Treatment of U-Bends." SGOG/EPRI
25.42 Hunt, E. S., and J. A. Gorman. "Development of Stress Relief
Procedures." In Proceedings: 1983 Workshop on Primary-Side Stress
November 1987.
25.43 Woodward, J., and D. van Rooyen. "Stress Relief of Transition

Zones." In Proceedings: 1983 Workshop on Primary-Side Stress
November 1987.
25.44 van Rooyen, D. "In Situ Heat Treatment of Roll Transitions." In
Proceedings: 1985 Workshop on Primary-SideStress Corrosion Cracking
of PWR Steam GeneratorTubing. EPRI NP-5158. June 1987.
25.45 Hernalsteen, P., R. Houben and C. van Melsen. "Induction Heating
Stress Relief at Doel 2." In Proceedings: 1983 Workshop on Primary-
Side Stress CorrosionCracking of PWR Steam GeneratorTubing. EPRI
25.46 van Melsen, C., et al. "Nickel Plating for the Repair of Steam
Generator Cracked Tubes." SGOG/EPRI Workshop: Management
1985.
25.47 Laire, Ch., et al. "Expertise de Tubes de GV Fissures et Repares par
Nickelage au Niveau de la Transition de Dudgeonage." Colloque
25.48 Amoruso, G. V., and J. W. Schroeder. "Stress Corrosion Cracking
Test of Expanded Stem Generator Tubes." EPRI NP-5012. January
1987.
25.49 Fredrick, G., et. al. "Strategy of Belgian Utilities for Factors
Influencing Cracking of Roll Transitions." In Proceedings: 1985
Workshop on Primary-Side Stress Corrosion Cracking of PWR Steam
GeneratorTubing. EPRI NP-5158. June 1987.
25.50 "Application of Remedial Measures for Primary Side Cracking -
Expansion and Sleeving." SGOG/EPRI Workshop: Management
1985.
25.51 "Palisades Steam Generator Tube Repair by Sleeving." Combustion
Engineering, Inc. Report No. CEN-42(P), Non-Proprietary,
December 10, 1976.
25.52 Horlacher, W. R., and K. M. Haas. "Steam Generator Tube Repair
by Sleeving." Paper presented at the American Nuclear Society
Annual Meeting, San Diego, CA, June 18-23, 1978.
25.53 "Steam Generator Repair Report for Southern California Edison
San Onofre Unit 1," Rev. 1. Westinghouse Electric Corporation,
Report DE-SP-58(80), March 1981.
25.54 "Steam Generator Sleeve Test Program, R. E. Ginna Nuclear Power
Plant." Babcock & Wilcox, Docket No. 50-244. March 17, 1981.
25.55 Gutzweiller, J. E. "Extending Service Life of Steam Generators by

Sleeving Tubes." Paper presented to American Power Conference,
Chicago, IL, April 1982.
25.56 US NRC. "Safety Evaluation, Steam Generator Tube Sleeving
Repair Program-R. E. Ginna Nuclear Power Plant." Docket No.
50-244. April 1, 1983.
25.57 "Indian Point Unit 3 Steam Generator Sleeving Report."
Westinghouse Electric Corporation, Report WCAP-10146 prepared
for the Power Authority of the State of New York, September 1982.
25.58 Cooper., F. W. "In-Place Retubing and Sleeving of Nuclear Steam
Generator." Paper presented to American Nuclear Society Meeting,
Detroit, Michigan, June 1983.
25.59 United States Nuclear Regulatory Commission. "Safety Evaluation
by the Office of Nuclear Reactor Regulation, Relating to Full Scale
Steam Generator Tube Sleeving at Point Beach Nuclear Plant Units
1 and 2." Docket Nos. 50-266 and 50-301. July 8, 1982.
25.60 "Guidelines for Steam Generator Tubing Specifications and Repair.
Steam Generator Tube Sleeving: Design, Specification, and
Procurement Check List," Vol. 3. EPRI NP-6743-L. February 1991.
25.61 Hunt, E. S., and J. A. Gorman. "Induction Stress Relief at Doel 2."
In Proceedings: 1983 Workshop on Primary-Side Stress Corrosion
Crackingof PWR Steam GeneratorTubing. EPRI NP-5498. November
1987.
25.62 Airey, G. P. "Stress Corrosion Cracking Performance of Inconel
Alloy 600 in Pure and Primary Water Environments." In Proceedings:
1983 Workshop on Primary-SideStress CorrosionCrackingof PWR Steam
GeneratorTubing. EPRI NP-5498. November 1987.
25.63 Jacko, R..J. "The Role of Key Environmental Parameters on PWSCC
Behavior." In Proceedings: 1987 EPRI Workshop on Mechanisms of
Primary Water IGSCC. EPRI NP-5987-SP. September 1988.
25.64 Smialowska, S. "Hydrogen Induced IGSCC of Alloy 600 in High
Temperature Aqueous Environments." In Proceedings:1987 EPRI
Workshop on Mechanisms of Primary Water IGSCC. EPRI NP-5987-
SP. September 1988.
25.65 Smith, E. R. "North Anna Unit 1 Steam Generator Tube Leak Event."
Paper presented at August 1989 NDE Workshop.
25.66 Economy, G., R., J. Jacko, J. A. Begley, and F W. Pement. "Influence
of Hydrogen Partial Pressure on the IGSCC Behavior of Mill
Annealed Alloy 600 Tubing in 360'C Water or 400'C Steam". In
Proceedings: 1987 EPRI Workshop on Mechanisms of Primary Water
IntergranularStress CorrosionCracking. EPRI NP-5987-SP. September
1988.
25.67 de Keroulas, F. "Fissuration en Milieu Primaire des Tubes de

Generateurs de Vapeur des Reacteurs 900 et 1300 mW: Resultats
des Examens Metallurgiques sur Tubes Extraits." Colloque
26
NONDESTRUCTIVE
EXAMINATION
27
MAJOR STEAM GENERATOR

REPAIRS/REPLACEMENT
Appendix A
LIST OF STEAM GENERATOR

PUBLICATIONS
A-2 Steam Generator Reference Book, Revision 1
A.1
LIST OF SGOG PUBLICATIONS BY REPORT NUMBER
Report No
and Date Title Project No
NP-516 "EPRI Secondary Water Chemistry RP404-1
February 1977 Study"
NP-519 "Evaluation of RP698-1
August 1977 Electromagnetic-Acoustic Concepts of
Inspection of Steam Generator
Tubing"
NP-1244 "Optical Technique for Internal RP1 172-3
November 1979 Diametrical Measurement of Steam
Generator Tubes"
NP-1391 "CALIPSOS Code Report", 2 vols. S129-1
April 1980
NP-1392 "Thermal-Hydraulic Characteristics of a S129-1
April 1980 Combustion Engineering Series 67
Steam Generator", 2 vols.
NP-1397 "Optical Scanner for Steam Generator S103-1

May 1980 Tube Inspection"
NP-1412 "Examination of Steam Generator S138-2

May 1980 Tube R45C52 From the Ginna Power
Plant"
NP-1419 "Steam Generator Crevice Gap S102-1
May 1980 Measurement by Induced-Vibration
Analysis"
NP-1427 "Magnetic Flux Leakage for S125-1
June 1980 Measurement of Crevice Gap
Clearance and Tube Support Plate
Inspection"
NP-1431 "Thermal-Hydraulic Analysis of 5131-1
June 1980 Once-Through Steam Generators"
NP-1467 "Assessment of Condenser Leakage TPS79-729

August 1980 Problems"
NP-1468 "Corrosion-Related Failures in Power TPS79-730

August 1980 Plant Condensers"
Appendix A-List of Steam Generator Publications A-3
Report No
NP-1479 "Effect of Out-of-Plane Denting Loads S169-1
August 1980 on the Structural Integrity of Steam
Generator Internals"
September 1980 Combustion Engineering System 80
NP-1546 "Thermal-Hydraulic Analysis of the S130-1

September 1980 Combustion Engineering System 80
Steam Generator"
NP-1607 "Method of Reducing Carry-Over and S122-1
November 1980 Reducing Pressure Drop Through
Steam Separators"
NP-1678 "Thermal Hydraulic Analysis of the S130-1

January 1981 Combustion Engineering Series 67
Steam Generator"

March 1981 Westinghouse Model F Steam
Generator", 2 vols.

March 1981 Westinghouse Model 51 Steam
Generator", 2 vols.
NP-1761 "Tests of Isothermal Soaking RP623-2

April 1981 Procedures for Limiting Tube Denting
in Nuclear Steam Generators"
NP-1785 "Steam Generator Mock-Up Facilities" S126-1
April 1981
NP-1 786 "Modeling of Cooling-Water Inleakage RP404-1
April 1981 Effects in PWR Steam Generators"
NP-1787 "Model of Vaporous Carry-Over" RP704-1

April 1981
NP-1 789 "Steam Generator Chemical Cleaning S128-1
April 1981 Demonstration Test No. 1 in a Pot
Boiler"
NP-1794 "Evaluation of Steam Generator Tube S136-1

April 1981 85-127 From Oconee 1B"
Report No
and Date Title Promect No
NP-1829 "Chemical Cleaning Demonstration S127-1
April 1981 Test No. 1 in a Mock-Up Steam
Generator"
NP-1859 "Visual Inspection Equipment for the S155-1
May 1981 Secondary Side of Steam
Generators"
NP-1 861 "Demonstration Tests on PWR Steam S183-1
May 1981 Generator Tube-Tube Sheet Crevice
Flushing Procedures"
NP-1863 "Secondary Water Chemistry Control S170-1
May 1981 at Genkai No. 1 Design and
Operating Considerations"
NP-1875 "Development of Sensors and S140-1
June 1981 Instrumentation for the TMI-2 OTSG
Tube Vibration Measurements
Program"
NP-1876 Vol. 1, "Flow Induced Vibration S140-1
June 1981 Analysis of Three Mile Island Unit 2
Once-Through Steam Generator
Tubes"
NP-1888 "Flow Induced Vibration Analysis of S176-1
June 1981 Oconee 2B OTSG Tubes"
NP-1 941 "Steam Generator Sludge Pile Model S119-1

July 1981 Boiler Testing"
NP-1 944 "Optical Scanner System for Internal S103-2

July 1981 Inspection of Steam Generator
Tubes"
NP-1 963 "Estimation of Diffusion Coefficients S146-1
August 1981 for Electrolyte in Hot Water"
NP-1965 "The Design and Construction of RP623-4
August 1981 Model Steam Generators for
Corrosion Testing of Alternate
Materials"
NP-1976 "Steam Generator Chemical Cleaning: S128-1
August 1981 Demonstration Test No. 2 in a Pot
Boiler"
Report No
NP-2041 "Steam Generator Sludge Pile Model S119-1
September 1981 BoilerTesting (Sludge
Characterization)"
NP-2042 "Radiographic System for Evaluation S105-1
September 1981 of Steam Generator Support Plate
Integrity"
NP-2046 "Single Tube Thermal and Hydraulic S118-1
September 1981 Tube Support Test"
NP-2082 "Evaluation of Oconee Steam S136-2

October 1981 Generator Debris"
NP-2141 "Profilometry for Steam Generator S108-1

November 1981 Tube Dent Characterization"
NP-2146 "Static Strain Analysis of TMI-2 OTSG S176-1

December 1981 Tubes"
NP-2149 "Corrosion-Product Transport in PWR RP404-1

December 1981 Secondary Systems"
NP-2268 "Secondary Water Chemistry at S170-1
February 1982 Ringhals Unit 2"
NP-2270 "Effect of Changing the Hydrazine S132-2

February 1982 Injection Point at the Carolina Power
and Light, H. B. Robinson Plant"
NP-2271 "Induced-Vibration Analysis Probe for S102-3
February 1982 Measurement of Steam Generator
Tube-to-Support Plate Clearance"
NP-2285 "Evaluation of Pulse-Echo Ultrasound S142-1
March 1982 for Steam Generator Tube-to-Support
Plate Gap Measurement"
NP-2294 "Guide to Design of Secondary S189-1
March 1982 Systems and Their Components to
Minimize Oxygen-Induced Corrosion"
NP-2295 "PWR Steam Generator Cost-Benefit TPS81-777
March 1982 Methodology (Denting)"
NP-2299 "Field Experiences with Multifrequency S115-1
March 1982 Multiparameter Eddy Current
Technology"
Report No
NP-2339 "Application of an Eddy-Current S202-2
April 1982 Technique to Steam Generator
U-Bend Characterization"
NP-2388 "HITCH Computer Code: Chemistry RP404-1
May 1982 and pH Estimates of Concentrating
Aqueous Solutions"
NP-2448 "Evaluation of Secondary System S 104-2
June 1982 Oxygen Control in PWR Power Plants
in the USSR"
NP-2492 "Inhibition of Steam Condensate S195-1
July 1982 Corrosion of Copper-Based Alloys by
Hydrazine"
NP-2505 "Effect of Moisture Separator Drain RP704-1
July 1982 Routing on OTSG Secondary System
Chemistry"
NP-2534-LD "Examination of Three Steam S138-2
August 1982 Generator Tubes From the Ginna
Nuclear Power Plant"
NP-2539-LD "Examination of Steam Generator S138-3
August 1982 Tube A(18-37) from the Point Beach
Unit 2 Nuclear Power Plant"
NP-2541 "PWR Steam-Side Chemistry Follow RP699-1
August 1982 Program"
NP-2597 "Condenser Inleakage Monitoring S182-1

October 1982 System Development"
NP-2610 "Boiling Heat Transfer in a Narrow S133-1

September 1982 Eccentric Annulus"
NP-2629-LD "Evaluation of Steam Generator S138-4

September 1982 U-Bend Tubes From Trojan Nuclear
Power Plant"
NP-2638 "Boiling in Narrow Crevices in Steam S134-1
November 1982 Generators"
NP-2643 "Predictions and Measurements of S179-1

November 1982 Isothermal Airflow in a Model Once
Through Steam Generator"
Report No
NP-2650 "Hydrogen Evolution Monitoring As a S117-1
November 1982 Measure of Steam Generator
Corrosion"
NP-2652 "Loads on Steam Generator Tubes S144-1
November 1982 During Simulated Loss-of-Coolant
Accident Conditions"
NP-2654 "Effects of Hydrazine and pH on the S196-1
November 1982 Corrosion of Copper-Alloy Materials in
AVT Environments With Oxygen"
NP-2655 "Metal Cation Inhibitors for Controlling S147-1
December 1982 Denting Corrosion in Steam
Generators"
NP-2656 "Evaluation and Categorization of S113-1
December 1982 Secondary System Layup and
Cleanup Practices for PWR Plants"
NP-2659 "Steam Generator Blowdown Filter 5109-1
November 1982 Testing"
NP-2683 "Thermal-Hydraulic Analysis of the S130-1
October 1982 Westinghouse Model 51 Steam
Generator"
NP-2689 "Tests of Types 51 A and 51 M Steam S154-1
October 1982 Generators at Bugey 4 and Tricastin 1
Nuclear Power Plants"
NP-2692 "In-Air Vibration Analysis of Three Mile S140-1
October 1982 Island Unit-2 Once-Through Steam
Generator Tubes", 2 vols. (Vol. 2 is
available on microfiche only)
NP-2701 "Electrochemical Behavior of Ferrous S127-1
October 1982 and Ferric Ions in EDTA/N2H4
Solutions"
NP-2703 "Salem Unit 1--Denting Mitigation S132-5
October 1982 Program: Implementation of
Improved Oxygen, Chloride, and
Copper Control"
NP-2706 "Secondary Water Chemistry Control S170-1
October 1982 at St. Lucie No. 1"
Report No
NP-2711 "Prevention of Wear Problems in S145-1
October 1982 PWR Steam Generators--An
Annotated Bibliography"
NP-2765 "Measurement and THEDA Code S179-1
December 1982 Prediction of Thermal, Velocity, and
Pressure Fields in a Once-Through
Steam Generator Heated Air Model"
NP-2774 "Steam Generator Probe Positioning S125-3
December 1982 Device"
NP-2790 "OTSG Tube Failures: Upper tube S165-1

December 1982 Sheet Corrosion Tests"
NP-2805 "Calibration of Instrumented Steam S139-1

January 1983 Separators to Determine Quality and
Flow Distribution in an Operating
Steam Generator"
NP-2815 "Cleaning Steam Generators Offline S149-1
February 1983 (Soaking) With Chelants"
NP-2823 "Steam Generator Support Plate S105-1

January 1983 Radiographic Evaluation System"
NP-2825-LD "Modeling of Crevice Corrosion in S167-7

January 1983 PWR Steam Generators"
NP-2836 "A Prototype EMAT System for S101-1

January 1983 Inspection of Steam Generator
Tubes"
NP-2838 "Thermal-Hydraulic Tests of Steam S121-1
January 1983 Generator Tube Support Plate
Crevices", 3 vols.
NP-2857 "Magnetic Flux Leakage for S125-1
February 1983 Measurement of Crevice Gap
Clearance and Tube Support Plate
NP-2862
February 1983
Inspection"
"Eddy Current NDE for Intergranular
Attack"
S201-1 0
NP-2863 "Optical Probe for Steam Generator S181-1
February 1983 Tube Dent"
Report No
NP-2865 "Monitoring Systems for Determining S187-1
March 1983 Air Inleakage and Oxygen
Concentrations in the Secondary
Cycle of Power Plants"
NP-2872 "Thermal-Hydraulic Code Qualification: S154-2
February 1983 ATHOS2 and Data From Bugey 4 and
Tricastin 1"
NP-2887 "Thermal and Hydraulic Code S168-1

February 1983 Verification, ATHOS2 and Model
Boiler No. 2 Data", 3 vols.
NP-2921 "Steam Generator Tube-Plugging and S203-1

March 1983 Tube-Sleeving Criteria: Assessment of
Current Practices"
NP-2944 "Elastic Stress Analysis of Small S194-4
March 1983 Radius U-Bend Steam Generator
Tubes"
NP-2957 "Fatigue Performance of Ni-Cr-Fe S110-1

March 1983 Alloy 600 Under Typical PWR Steam
Generator Conditions"
NP-2958-LD "Examination of Three Steam S138-1
January 1984 Generator Tubes From Point Beach
Unit 1 Nuclear Power Plant"
NP-2962 "The Role of Electrochemistry in S191-2

March 1983 Causes and Measurement of
Corrosion in PWR Steam Generators"
NP-2963 "Investigation of Phosphate-Sludge S158-1

March 1983 Interactions"
NP-2967 "Evaluation of Surrogate Boilers for S111-2

March 1983 Steam Generators"
NP-2969 "Hydrazine Usage for Corrosion S132-8

March 1983 Control in PWR Plants With Powdered
Resin Condensate Polishers"
NP-2971 Vol. 1, "Tube Support Response to S143-1

May 1983 Tube Denting Evaluation" and Vol. 2
(Appendix H-Test Data)
A- 10 Steam Generator Reference Book, Revision 1
Report No
NP-2972-LD "Chemical and Metallographic S138-7
March 1983 Examination of Tube Specimens
From Ringhals 2 Steam Generator"
NP-2973 "On-Line Use of Chelants in Nuclear S166-1

April 1983 Steam Generators--Feasibility Study"
NP-2974 "Secondary Water Chemistry at S170-1

April 1983 Millstone 2"
NP-2975 "A Review of Condensate S167-1

April 1983 Demineralizer Regeneration
Procedures"
NP-2976 "Chemical Cleaning Solvent and S127-1

April 1983 Process Testing"
NP-2977 "Evaluation of Secondary System S113-1

April 1983 Layup and Cleanup Practices and
Processes"
NP-2978 "State-of-the-Art Evaluation of S163-1

April 1983 Condensate Polisher Performance"
NP-2979 "Diffusion and Hideout in Crevices" S146-1

March 1983
NP-2981 "Condensate Polisher Resin Leakage S163-1

April 1983 Quantification and Resin Transport
Studies"
NP-2983 "Steam Generator Chemical Cleaning S128-1

April 1983 Demonstration Test No. 3 in a Pot
Boiler"
NP-2986 "An Evaluation of Techniques to S154-2

March 1983 Predict Flow-Induced Tube Vibration
in a Preheat Steam Generator"
NP-2987 "Chemical Cleaning Process S150-3

April 1983 Evaluation--Westinghouse Steam
Generators"
NP-2990 "Steam Generator Chemical Cleaning: S128-1
June 1983 Demonstration Test in a Model Boiler"
NP-2993 "Evaluation of Steam Generator Fluid S164-1

May 1983 Mixing During Layup"
Appendix A-List of Steam Generator Publications A-I I
Report No
NP-2996 "Steam-Water Separation in a S173-1
June 1983 Swirl-Vane Centrifugal Separator", 2
vols.
NP-2997 "Ultrasonic Enhancement of Chemical S185-1
April 1983 Cleaning of Steam Generators"
NP-3008 "Chemical Cleaning Demonstration S127-1

April 1983 Test No. 2 in a Mock-Up Steam
Generator"
NP-3009 "Steam Generator Chemical Cleaning S150-1
April 1983 Process Development"
NP-3010 "Steam Generator U-Bend S202-1

April 1983 Eddy-Current NDE"
NP-3012 "Laboratory Program to Examine S124-1

April 1983 Effects of Layup Conditions on Pitting
of Alloy 600"
NP-3013 "Tube-to-Tube Sheet Joint Test", 2 S119-2
March 1983 vols.
NP-3018 "A Study of the Boiling Processes in S171-1

May 1983 the Sludge Deposit of Steam
Generators"
NP-3020 "Evaluation and Improvement of PWR S104-1
July 1983 Secondary System Oxygen Control
Measures"
NP-3022 "Determination and Verification of Sl11-1

January 1984 Required Water Chemistry Limits, Vol.
3: Pot Boilers 8A and 8B"
NP-3023 "Neutralization of Steam Generator S112-1
September 1982 Denting"
NP-3024 "Characterization of Single-Tube S112-2

May 1983 Model Boiler Dented Intersection
Specimens"
NP-3026-LD "Evaluation of the Lower Tube Sheet S136-4

July 1983 Region of an Oconee 1 Steam
Generator Tube"
A-1 2 Steam Generator Reference Book, Revision I
Report No
NP-3027 "Analysis of Sludge From Indian Point S136-5
June 1983 No. 3
May 1983 R12C66 From Indian Point No. 3"
NP-3030 "Nonproprietary Corrosion Inhibitors S148-1
June 1983 for Solvents to Clean Steam
Generators"
NP-3031 "Three-Dimensional Thermal/Hydraulic S156-1
January 1984 Code Development"
NP-3033 "Steam Generator Data Base" S160-1

June 1983
NP-3035 "Condensate Demineralizer S167-1

June 1983 Performance During Periods of High
Condenser Leakage"
NP-3037 "Condensate Polishing at Surry S167-1
June 1983 Nuclear Power Station"
NP-3039 "Fretting & Fatigue in Steam S174-1

November 1983 Generators Part 1, Determination of
Tube-to-Tube Support Interaction
Characteristics"
NP-3040 - "Neutralization of Tube Sheet Crevice S183-2
May 1983 Corrosion"
NP-3041 "Amine Borane Compounds in S183-3
June 1984 Crevice-Cleaning Solutions"
NP-3042 "Model Tests of a Once-Through S186-1
June 1982 Steam Generator for Lane Blocker
Assessment and THEDA Code
Verification"
NP-3043 "Stress Corrosion of Alloys 600 and S191-1
October 1983 690 in Acidic Sulfate Solutions at
Elevated Temperatures"
NP-3044 "Corrosion Performance of Alternative S192-1
July 1983 Steam Generator Materials and
Designs", 3 vols.
Appendix A-List of Steam Generator Publications A-I 3
Report No
NP-3045 "Crevice Corrosion of Lattice Support S204-1
July 1983 Alloys in Secondary Environments of
Nuclear Steam Generators"
NP-3046 "Evaluation of Condensate Polishers" RP623-3
June 1983
NP-3047 "Sodium-Sulfate Solubilities in RP623-5

July 1983 High-Temperature (250-3740 C) Salt
and Acid Solutions"
NP-3048 "Rationale for Chemical Control of RP1171-1
January 1984 Feed and Boiler Water: Volumes 1, 2,
and 3"
NP-3050 "Salt Concentration in Heated RP1171-3
October 1983 Crevices and Simulated Scale"
NP-3051 "Optimization of Metallurgical RP1708-1

July 1983 Variables to Improve Corrosion
Resistance on Inconel Alloy 600"
NP-3052 "Tube Support Plate Thermal and S180-1
May 1983 Hydraulic Testing"
NP-3053 "Flushing Efficiency of Steam S209-1

April 1983 Generator Tube-Tube Sheet
Assemblies With Restricted Crevices"
NP-3054 "Neutralization of Crevice Acids" RP623-2
May 1983
NP-3055 "Stress Relief to Prevent Stress S192-3

May 183 Corrosion in the Transition Region of
Expanded Alloy 600 Steam
Generator Tubing"
NP-3056 "In Situ Heat Treatment and S191-3
April 1983 Polythionic Acid Testing of Inconel
600 Row I Steam Generator
U-Bends", 3 vols.
NP-3057 "Intergranular Stress Corrosion S138-8
May 1983 Cracking of Ni-Cr-Fe Alloy 600 Tubes
in PWR Primary Water--Review and
Assessment for Model Development"
Report No
NP-3058 "The Design and Construction of S193-2
May 1983 Deep Tube Sheet Crevice Devices for
Producing Intergranular Attack of
Heat Transfer Tubing"
NP-3059 "IGA of Alloy 600 in High-Temperature S183-4

May 1983 Solutions of Sodium Hydroxide
Contaminated With Carbonate"
NP-3060 "Effect of Calcium Hydroxide and S119-3

May 1983 Carbonates on IGA and SCC of Alloy
600"
NP-3061 "Stress Corrosion Cracking of Alloy S192-2
May 1983 600 and Alloy 690 in All Volatile
Treated Water at Elevated
Temperatures"
NP-3062 "lntergranular Attack of Alloy 600: S193-1

May 1983 High-Temperature Electrochemical
Tests"
NP-3063 "Numerical Simulation of Steam S172-1

June 1983 Separators Used in Nuclear Steam
Generators and Reactor Systems", 2
vols.
NP-3065 "An Engineering and Probabilistic S151-1

June 1983 Analysis of Tube Cracking
Performance in Once-Through Steam
Generators", 2 vols.
NP-3066 "Implementation of Boric Acid in the S116-1

May 1983 Field--Indian Point Unit 3 Plant"
NP-3067 "Crevice Hideout Return Testing" S190-1

May 1983
NP-3068 "Investigation of Steam Generator S120-1

May 1984 Corrosion Products Under Typical
PWR Operating Conditions"
NP-3070-LD "Evaluation of Steam Generator S138-6

May 1983 Tubes R7C45 and R21C46 From the
Ginna Nuclear Power Plant"
NP-3138 "PWR Model Steam Generator RP623-1
June 1983 Corrosion Studies"
Appendix A-List of Steam Generator Publications A-IS
Report No
NP-3142-LD "Application of SGOG Secondary S132-12
June 1983 System Chemistry Guidelines at
Pressurized Water Reactors"
NP-3198 "Secondary Water Chemistry at RP704-1
March 1984 Oconee"
NP-3207-SR "Workshop Proceedings: S999
August 1983 Replacement/Repair of Steam
Generators"
NP-3273 "Evaluation and Optimization of S106-1
November 1983 Magnetic Filters on Simulated Boiler
Water'
NP-3274 "Determination and Verification of $111-1
January 1984 Required Water Chemistry Limits;
Volume 3: Pot Boiler Tests"
NP-3274 "Determination and Verification of S111-1
March 1984 Required Water Chemistry Limits;
Volume 1: Summary Report"
NP-3274 "Determination and Verification of $111-1
August 1984 Required Water Chemistry Limits;
Volume 5: Model Boiler 5B Tests"
NP-3275 "Causes of Denting; Volume 5: S157-1
December 1983 Contaminant Threshold Tests"
NP-3275-LD "Causes of Denting; Volume 6: S157-1
March 1984 In-Leakage Concentration
Experiments"
May 1984 Summary Report"
May 1984 Laboratory Test Results"
NP-3275 "Causes of Denting; Volume 3: Plant S157-1

May 1984 Chemistry Correlations"
May 1984 Isothermal Tests"
NP-3278 "Evaluation of Field Applications of S132-10

March 1984 Boric Acid in PWR Steam Generators"
A t16 Steam Generator Reference Book, Revision 1
Report No
NP-3302 "Data Acquisition-Reduction System S127-1
February 1984 for Chemical Cleaning Processes for
NP-3311 "Condensate Polishing at Plant S167-1
February 1984 Bowen"
NP-3477 "PWR Steam Generator S305-2
April 1984 Chemical-Cleaning Data Base", 2
vols.
NP-3571 "Steam Generator U-Bend Tube S138-5
February 1984 Examination"
NP-3574-SR "Workshop Proceedings: Pitting in S999
October 1984 Steam Generator Tubing"
NP-3703 "Evaluation of Alternative Alloys for RP1450-1
October 1984 PWR Steam Generator Tubing"
September 1984 R34C54 From the Ginna Nuclear
Power Plant"
NP-3791 "Development of a Model for RP2163-1
September 1984 Predicting Intergranular Stress
Corrosion Cracking of Alloy 600 Tubes
in Primary Water"
NP-3818 "Proceedings: IGA Remedies S302-22
November 1984 Contractors' Meeting"
NP-3847 "Microchemistry of Corroded S302-8
February 1985 Intergranular Surfaces"
NP-3851 "Use of the HITCH Code to Evaluate RP404-1

January 1985 Amines as Neutralizers of Crevice
Acidity in Nuclear Steam Generators"
NP-3905 "Pitting Corrosion of Alloy 600 Steam S308-1
February 1985 Generator Tubing: Results of a
Laboratory Scoping Study"
NP-3924-SR "Proceedings: The Second EPRI S999
March 1985 Workshop on Support-Structure
Corrosion in Nuclear Plant Steam
Generators"
Appendix A-List of Steam Generator Publications A- 17
Report No
NP-3928 "Evaluation of Eddy-Current S301-1
April 1985 Procedures for Measuring Wear Scars
in Preheat Steam Generators"
NP-3949 "Surface and Grain Boundary RP1161-1

March 1985 Segregation, Stress Corrosion
Cracking, and Corrosion Fatigue of
Inconel 600"
NP-3957 "Mechanism of Intergranular RP1618-2

March 1985 Corrosion of Inconel 600 Tubing in
PWR Steam Generators"
NP-3964-LD "Evaluation of Alloy 600 Tubes From S304-8

March 1985 Steam Generators A and B of the
Palisades Nuclear Power Plant"
NP-3998-M "lntergranular Corrosion of Inconel RP1618-3

May 1985 600 Tubing for PWR Steam
Generators"
NP-4050 "The Impact of Makeup Water RP404-1

June 1985 System Performance on PWR Steam
Generator Corrosion"
NP-4051 "Intergranular Attack and Stress S302-1
June 1985 Corrosion Cracking of Alloy 600 in
High-Temperature Caustic Solutions
Containing Contaminants"

June 1985 Simulation Tests"
NP-4099 "Organic Impurity Transport in PWR S306-12

August 1985 Secondary Systems"

August 1985 Laboratory Investigations"
NP-4270 "Effect of Boric Acid Treatment on the RP404-1

November 1985 Secondary Cycle at ANO-2"
NP-4272 "Simulation of Intergranular Attack on S302-3

October 1985 Alloy 600 Tubing in Tubesheet
Crevices"
NP-4296 "Design Review Checklist: Steam S303-6
December 1985 Generator Sleeving"
Report No
NP-4364 "Specifications for In Situ Stress Relief S303-6
December 1985 of PWR Steam Generator Tube
U-Bends and Roll Transitions"
NP-4375 "Destructive Examination of Zion Unit S304-5
December 1985 1 Tube and Antivibration Bar
Samples"
NP-4456 "Steam Generator Tube S306-1
March 1986 Denting--Field Study"
NP-4457 "Intergranular Corrosion of Steam S306-1

February 1986 Generator Tubes--Field Study"
NP-4458 "Proceedings: 1983 Workshop on S302-22

March 1986 Secondary-Side Stress Corrosion
Cracking and Intergranular Corrosion
of PWR Steam Generator Tubing"
NP-4459 "Status of Cracking and Remedial S303-21
February 1986 Measures for PWR Steam Generators
With Full-Depth Expanded Tubing"
NP-4465 "Microstructure, Microchemistry, and S303-9
February 1986 Microdeformation of Alloy 600 Tubing"

March 1986 Secondary-Side Stress Corrosion
Cracking and Intergranular Corrosion
NP-4504 "Evaluation of Alloy 600 Tube A77-34 RP304-11

March 1986 From Steam Generator A of Arkansas
Nuclear One, Unit 1"
NP-4521 "Resin and Ionics Leakage From S306-16

April 1986 Condensate Polishers With and
Without Inert Resin"
NP-4559 "Flow-Induced Vibration of Steam S153-1

May 1986 Generator Tubes"
NP-4563 "Return of Hideout Chemicals in PWR S205-7

May 1986 Steam Generators During Power and
Temperature Reductions"
Appendix A-List of Steam Generator Publications A- 19
Report No
NP-4564 "High-Sensitivity Dissolved-Gas S306-15
May 1986 Monitoring System With Applications
for PWR Secondary-Side Chemistry"
NP-4594 "Status and Suggested Course of S303-6
May 1986 Action for Nondenting-Related
Primary-Side IGSCC of
Westinghouse-Type Steam
Generators"
NP-4597 "Chemical Cleaning of Millstone Unit 2" S305-21
May 1986
NP-4600 "Effect of Venting on Crevice Cleaning S305-13
May 1986 for PWR Steam Generators"
NP-4604-CCML "ATHOS3 Mod-01: A Computer S411-6,RP1066-1

September 1990 Program for Thermal-Hydraulic
Analysis of Steam Generators", Rev.
1, 2 vols.
NP-4620 "A Model of Sludge Behavior in RP2160-8

June 1986 Nuclear Plant Steam Generators"
NP-4624 "Evaluation of New pH Control Agents RP1571-7

June 1986 for PWR Secondary Water Systems"
NP-4625 "Laboratory Examinations of Selected S302-12

July 1986 Tubes From Test Facilities of the
Steam Generator Owners Group"
NP-4626 "Steam Generator Tube Sampling: S304-2

June 1986 Feasibility Study"
NP-4635 "Remedial Methods for Intergranular S302-10,S302-1

June 1986 Attack of Alloy 600 Tubing, Volume 1: 3,S302-14
Plant Corrosion Morphologies and
Remedial Methods, Volume 2:
Additives and Test Plans for Remedial
Methods, Volume 3: Boric Acid and
Acetic Acid Remedial Methods"
NP-4647 "Oxide Growth Mechanisms on RP1 171-2
July 1986 Chromium Alloy Steels"
NP-4648 "The Effects of Oxygen, Copper, and RP1171-3

July 1986 Acid Chlorides on Denting Corrosion"
A-20 Steam Generator Reference Book, Revision I
Report No
NP-4665-M "Proceedings: Workshop on Thermally S999
July 1986 Treated Alloy 690 Tubes for Nuclear
Steam Generators"
NP-4665-SR "Proceedings: Workshop on Thermally S999
July 1986 Treated Alloy 690 Tubes for Nuclear
Steam Generators"
NP-4671 "Survey of Domestic and Foreign S306-18

July 1986 PWR Experience With Morpholine in
Chemistry Control by All-Volatile
Treatment"
NP-4672 "Correlation of Tube Support Structure S311-1

July 1986 Studies"
NP-4678 "Crevice Hideout Return Testing" RP699-2

July 1986
NP-4686 "Recirculating Steam Generator S309-1

July 1986 Tubesheet Crevice Flushing
Procedures"
NP-4698 "Industry-Wide Survey of PWR S306-12

July 1986 Organics"
NP-4705 "Electrochemistry and Corrosion of RP2163-3

July 1986 Alloys in High-Temperature Water"
NP-4708 "Chemical Cleaning of PWR Steam S305-11

July 1986 Generator Sludge Piles"
NP-4731 "Side-by-Side Comparison of S306-12,S306-13

September 1986 Techniques for Analyzing Organic
Acids, Total Organic Carbon, and
Anions"
NP-4737-LD "Destructive Examination of Steam S304-7

July 1986 Generator Tubes Removed From
Millstone Point Unit 2"
NP-4745-LD "Examination of Tubes R4C19HL, S304-14

August 1986 R6C18HL, and R16C33HL From
Steam Generator 12 of the Prairie
Island Nuclear Station Unit 1"
NP-4754 "Localized Electrochemical Corrosion S308-5

September 1986 of Nickel-Based Alloys"
Report No
NP-4759-LD "Evaluation of Steam Generator Hot S304-16
August 1986 Leg Tubes R18C43 and R18C53
From the Cabrera Nuclear Power
Plant"
NP-4760-LD "Examination. of Steam Generator S304-15
August 1986 Tubes R18C53HL and R18C43HL
From the Jose Cabrera (Zorita)
Nuclear Power Station"
NP-4802 "Evaluation of Environmental Effects S302-9
September 1986 on Intergranular Attack of Alloy 600"
NP-4818 "Correlation of Tube Support S311-2

October 1986 Corrosion Studies"
NP-4829-LD "Examination of Steam Generator S304-6
October 1986 Tube Sections From the Millstone Unit
2 Nuclear Power Plant"
NP-4831 "Use of Pourbaix Diagrams to Infer S302-10
October 1986 Local Pitting Conditions"
NP-4898 "Evaluation of PWR Tubesheet S302-21

November 1,986 Crevice-Flushing Approaches"
NP-4901 "Removal of a Tubesheet Sample S304-17
December 1986 From a Retired Point Beach Unit 1
Steam Generator"
NP-4904 "Calvert Cliffs Unit 1 Tube S304-10
November 1986 Examination"
NP-4929 "Proceedings: 1985 EPRI Workshop S302-22

December 1986 on Remedial Actions for
Secondary-Side Intergranular
Corrosion"
NP-4940 "Methods for Evaluating Steam RP2599-1
December 1986 Generator Hideout Return Data: Case
Study at North Anna"
NP-4954 "Chemical Cleaning Waste Disposal" S305-19
November 1986
NP-4978 "Evaluation of Intergranular Attack on S302-13
December 1986 Alloy 600, Volume 1: Evaluation of
Causes"
Report No
NP-4978 "Evaluation of Intergranular Attack on S302-13
December 1987 Alloy 600, Volume 2: Effectiveness of
Boric Acid"
NP-5008 "Evaluation of Sulfur Hexaflouride and RP2599-2
January 1987 Helium for Steam Generator Leak
Location"
NP-5009 "Array Coil Probe" S301-6
March 1987
NP-5012 "Stress Corrosion Cracking Test of S303-22
January 1987 Expanded Steam Generator Tubes"
NP-5015 "Hideout and Return of Chloride Salts RP1171-3
January 1987 in Heated Crevices Prototypic of
Support Plates in Steam Generators"
NP-5017 "Crevice Corrosion of Support Alloys in RP623-6
March 1987 the Secondary Environments of
Nuclear Steam Generators -
Supplemental Report"
NP-5022-LD "Metallurgical and Chemical S304-3
December 1986 Evaluation of Tubes R17C85,
R28C32, and R26C50 From a Doel
Unit 2 Steam Generator"
NP-5026 "Assessment of Sulfur in Chemical S305-9
February 1987 Cleaning of PWR Steam Generators"
NP-5037 "Causes of and Corrective Actions for S308-3

February 1987 Pitting in Steam Generator Tubing"
NP-5056-SR "PWR Secondary Water Chemistry S999

March 1987 Guidelines, Rev. 1"
NP-5072 "Specially Prepared Alloy 600 Tubing" S303-17

February 1987
NP-5073 "Caustic Concentration in Tube S311-4
March 1987 Support Plate Crevices of Steam
Generators"
NP-5074 "Evaluation of Changing PWR S306-13
March 1987 Polisher Operation From Regenerated
to Throwaway Resins"
Report No
NP-5092 "Hideout of Impurities in Steam S306-17
March 1987 Generators During Heat-Up and
Power Ascension"
NP-5109 "Production of Eddy-Current S302-10
April 1987 Standards for Caustic Intergranular
Corrosion"
NP-5120 "lntergranular Attack or Corrosion in a S302-6
July 1987 Once-Through Model Steam
Generator"
NP-5129 "Mechanisms of Intergranular Attack S302-1
July 1987 and Stress Corrosion Cracking of Alloy
600 by High-Temperature Caustic
Solutions Containing Impurities"
NP-5138 "Phase Relations and Fluid S302-5

April 1987 Compositions in Steam Generator
Crevices"
NP-5140 "Steam Generator Cold Leg Thinning S311-5
April 1987 in Operating Plants"
NP-5155 "Reference Electrode for PWRs" S303-4

May 1987
NP-5156 "Sulfate Hideout in Heated Crevices" S311-3

May 1987

June 1987 Primary-Side Stress Corrosion
Cracking of PWR Steam Generator
Tubing"
NP-5192 "Microstructural Effects on RP2163-4

May 1987 Microdeformation and Primary-Side
Stress Corrosion Cracking of Alloy 600
Tubing"
NP-5193 "Measurement of pH and Corrosion RP2160-13
May 1987 Potentials of Tube Alloys in Solutions
Found in Steam Generators"
NP-5199 "Chemical Enhancement of Crevice S302-18

June 1987 Flushing", 2 vols.
Report No
NP-5207 "Pitting in Steam Generator Tubing: S308-6
June 1987 Causes and Corrective Actions"
NP-5248 "Investigation of Causes and S308-4

June 1987 Corrective Actions for Pitting in Steam
Generator Tubes: Prototypic Tests"
NP-5249 "Qualification of Remedial Methods to S303-27
June 1987 Prevent Primary-Side Stress Corrosion
Cracking of Steam Generator Tubing",
3 vols.
NP-5263 "Production of Intergranular Attack of S302-3
July 1987 Alloy 600, Alloy 690, and Alloy 800
Tubing in Tubesheet Crevices"
NP-5265 "Hideout of Sodium Salts in S311-4
July 1987 Tubesheet Crevices"
NP-5266 "On-Line Monitoring of RP2647-1

June 1987 Secondary-Cycle Chemistry with
Morpholine at the Beaver Valley
Power Station"
NP-5267 "Weld Region Corrosion During S305-1 0,S305-1
July 1987 Chemical Cleaning of PWR Steam 4,S305-15,S305-
Generators", 2 vols. 18
NP-5282 "Residual and Applied Stress Analysis S303-3

September 1987 of an Alloy 600 Row 1 U-Bend"
NP-5283 "Guidelines for Permanent BWR S999

September 1987 Hydrogen Water Chemistry
Installations - 1987 Revision"
NP-5344 "Decontamination of Point Beach-1 S304-18

August 1987 and Surry-2 Steam Generator
Tubesheet Sections"
NP-5368 "Mechanisms for Formation and S302-16

August 1987 Disruption of Surface Oxides"
NP-5377 "lntergranular Attack of Alloy 600 S302-4

August 1987 Tubing: Simulation Tests"
NP-5396 "lntergranular Corrosion Mechanisms RP2163-5

August 1987 of Alloy 600"
Report No
NP-5397 "Laboratory Evaluation of Steam S304-21
August 1987 Generator Tubes 120/12, 79/9, and
59/95 from St. Lucie Unit 1"
NP-5400 "Micromorphology and Microchemistry S302-23
September 1987 , of Corrosion Products in Tubesheet
Crevices"
NP-5406 "Electrochemical Corrosion of Alloy S302-17
September 1987 600 in Secondary Water"
NP-5420-LD "Examination of Tubes R3C41HI and S304-20

October 1987 R9C58HL of Steam Generator C,
North Anna Unit 1"
NP-5496 "In Situ Heat Treatment of U-Bends" S303-11

November 1987

November 1987 Primary-Side Stress Corrosion
Cracking of PWR Steam Generator
Tubing"
NP-5503 "Library of PWR Steam Generator S301-5
January 1988 Tubing Samples"
NP-5540 "Experimental Determination of S310-7

May 1988 Turbulent Buffeting Effects in Tube
Bundles"
NP-5541 "Experimental Prediction of Tube S310-1

June 1988 Support Interaction Characteristics in
Steam Generators", 2 vols.
NP-5542 "Flow Velocity and Turbulence S310-12

May 1988 Measurements in a Laboratory-Scale
Steam Generator Model"
NP-5543 "Analytic Prediction of Complex S310-3

May 1988 Unsteady Flow Fields in Preheat PWR
Steam Generators"
NP-5547 "Tubesheet Expansion Improvements" S303-29

December 1987
NP-5555 "Prediction of Localized Flow S310-14

May 1988 Velocities and Turbulence in a PWR
Steam Generator"
Report No
NP-5556 "Numerical Prediction of S310-6
May 1988 Turbulence-Induced Steam Generator
Tube Vibration"
NP-5557 "ATHOS3 Computer Code S310-8
June 1988 Verification", 2 vols.
NP-5558 "Boric Acid Application Guidelines for S302-13
December 1987 Intergranular Corrosion Inhibition"
NP-5558-M Vol. 1, "Boric Acid Application S407-35

December 1990 Guidelines for Intergranular Corrosion
Inhibition", Rev. 1
NP-5558-SL Vol. 2, "Boric Acid Application S407-35
December 1990 Guidelines for Intergranular Corrosion
Inhibition", Rev. 1
NP-5561-CCML "MULTEQ: Equilibrium of an S407-5
February 1988 Electrolytic Solution With Vapor-Liquid
Partitioning and Precipitation", 3 vols.
NP-5561-CCM "MULTEQ: Equilibrium of an S407-5
February 1988 Electrolytic Solution With Vapor-Liquid
Partitioning and Precipitation, Volume
1: User's Manual"
NP-5561 -CCML "MULTEQ: Equilibrium of an S407-30
May 1989 Electrolytic Solution With Vapor-Liquid
Partitioning and Precipitation; Volume
1: User's Manual (Revision 1)"
December 1990 Electrolytic Solution With Vapor-Liquid
Partitioning and Precipitation", Vol. 2:
"The Database", Rev. 2
April 1992 Electrolytic Solution With Vapor-Liquid
Partitioning and Precipitation", Vol. 2:
"The Database", Rev. 3
July 1992 Electrolytic Solution With Vapor-Liquid
Partitioning and Precipitation; Volume
1: User's Manual (Revision 2)"
Report No
NP-5561 -CCML "MULTEQ: Equilibrium of an S407-30
August 1992 Electrolytic Solution with Vapor-Liquid
Partitioning; Volume 3: Theory
Manual"
NP-5563 "NDE and Mechanical Removal of $403-2,$403-4,R
January 1988 Sludge in PWR Steam Generators", 2 P2755
vols.
NP-5565 "Vibration and Wear Prediction for S310-5
June 1988 Steam Generator Tubes"
NP-5570 "Flow Velocity and Turbulence S310-13

June 1988 Measurements in a Westinghouse
D-4 Steam Generator Model", 2 vols.
NP-5592 "Prairie Island Chemical Hideout RP2599-1
February 1988 Study"
NP-5594 "Effects of Alternate pH Control RP2599-1
February 1988 Additives in PWRs"
NP-5621 "Cleanup of Steam Cycle Drains in RP1571-5
January 1988 PWRs With Once-Through Steam
Generators"
NP-5623 "Oxygen Control in PWR Makeup RP2297-1
January 1988 Water"
NP-5660-LD "Point Beach-1 Tubesheet Crevice S304-19
February 1988 Chemistry"
NP-5708 "Determination of Thermodynamic S407-1
November 1989 Data for Modeling Corrosion", 2 vols.

March 1992 Data for Modeling Corrosion; Volume
3: C02-NaOH-H20 System"
June 1992 Data for Modeling Corrosion; Volume
4: Chloride Ion Interaction With
Magnesium, Calcium, and Hydrogen
Ions at 250-325 degrees C"
Report No
and Date Title Proiect No
August 1992 Data for Modeling Corrosion; Volume
5: NiCl2, CuCI2, HNO3, Ni(N)3)2,
NaNO3 Systems"
NP-5712 "Oxide-Film Compositions and S302-25
April 1988 Morphology on Alloy 600 Tubes From
Steam Generators, North Anna Unit 1
and Point Beach Unit 1"
NP-5728 "Verification of the ATHOS3 Code RP1066-10
May 1988 Against Feedring and Preheat Steam
Generator Test Data"
NP-5761-M "Stress Corrosion Cracking of Alloys RP1450-2
May 1988 600 and 690 in All-Volatile-Treated
Water at Elevated Temperatures"
NP-5761-SP "Stress Corrosion Cracking of Alloys RP1450-2
May 1988 600 and 690 in All-Volatile-Treated
NP-5764 "PWR Water Treatment RP1571-5
May 1988 Improvements: Cost-Benefit Analysis"
NP-5773 "Signal Processing for Steam S404-10

April 1988 Generator Inspection"
NP-5794 "Catalytic Methods of Deoxygenating RP1571-2
April 1988 Water"
NP-5802 "Method for Detecting Resin Leakage RP1447-2

May 1988 in LWR Coolant"
NP-5828 "Proceedings: Workshop on Initiation RP2812-3

May 1988 of Stress Corrosion Cracking Under
LWR Conditions"
NP-5902 "Nondestructive Evaluation Methods S301-1
July 1988 to Measure Inside Diameters of
Steam Generator Tubing"
NP-5928-SP "Supplementary Examination of S302-7
October 1988 Alternative Materials in a Model
Steam Generator; Volume 1:
Tubesheet and Tube Bundle
Examinations"
Report No
NP-5928-M "Supplementary Examination of S302-7
Tubesheet and Tube Bundle
Examinations"
Steam Generator, Volume 3: Tube
Characterization by Metallography
and Transmission Electron
Microscopy"
NP-5928 "Supplementary Examination of S302-7
Steam Generator; Volume 3: Tube
Characterization by Metallography
and Transmission Electron
Microscopy"
January 1989 Alternative Materials in a Model
Steam Generator, Volume 2:
Correlation of Model 10 Defects With
Eddy-Current Indications"
NP-5928-SP "Supplementary Examination of S302-7
NP-5929 "CECIL: A Robot for Secondary-Side S403-3
February 1989 Maintenance of PWR Steam
Generators"
NP-5971 "1987 EPRI Workshop on S407-7
September 1988 Secondary-Side Intergranular
Corrosion Mechanisms: Proceedings",
2 vols.
Report No
NP-5987-M "Proceedings: 1987 EPRI Workshop S407-7
September 1988 on Mechanisms of Primary Water
Intergranular Stress Corrosion
Cracking"
NP-6005 "Proceedings: 1987 Symposium on S407-17
March 1990 Chemistry in High-Temperature
Water"
NP-6115-M "lntergranular Attack of Alloy 600: S302-24
February 1989 Simulation and Remedial Action
Tests"
NP-6192-M "Examination of Roller Expansion S407-19
July 1989 Transitions in Point Beach Unit 1
NP-6201 "PWR Steam Generator Examination S404-8
December 1988 Guidelines: Revision 2"
NP-6201 "PWR Steam Generator Examination S404

November 1992 Guidelines: Revision 3"
NP-6239 "PWR Secondary Water Chemistry S405-2

December 1988 Guidelines, Revision 2"
NP-6341 "PWR Steam Generator Tube Fretting S310-2

April 1989 and Fatigue Wear"
NP-6356-M "Qualification of PWR Steam S403-5

May 1989 Generator Chemical Cleaning for
Indian Point-2"
NP-6362-M "Examination of Pits in Trojan Nuclear S407-32
May 1989 Power Plant Steam Generator Tubes"
NP-6377-M "Guidelines for the Design and S306-19

June 1989 Operation of Makeup Water
Treatment Systems"
NP-6455 "Field Studies on Intergranular Attack" S306-21

July 1989
NP-6626-M "Belgian Approach to Steam S404-14
March 1990 Generator Tube Plugging for Primary
Water Stress Corrosion Cracking"
Report No
NP-6703-M "Effect of Different Thermal S408-2
March 1990 Treatments on the Corrosion
Resistance of Alloy 690 Tubing"
NP-6705-M "Measurement of Surface-Induced S406-1
March 1990 Microplasticity in Alloy 600 C-Rings"
NP-6708 "Progress in Radiation Control RP2758

Technology
NP-6709-M "Microchemistry of Corroded S407-15

March 1990 Intergranular Surfaces, Tubes
Removed From Point Beach Unit 1,
Calvert Cliffs Unit 1, and Saint Lucie
Unit 1 Steam Generators"
NP-6710-SD "Proceedings: Workshop on the Role S999
March 1990 of Sulfur Species on the
Secondary-Side Degradation of Alloy
600 and Related Alloys"
NP-6710-M "Proceedings: Workshop on the Role S999
March 1990 of Sulfur Species on the
Secondary-Side Degradation of Alloy
600 and Related Alloys"
NP-6716-L "Roll Transition Inspection of Doel-2 S404-12
May 1992 Steam Generator Tubes; Volume 3:
September 1991 Inspection"
NP-6719-SD "Proceedings: Primary Water Stress S406-3
April 1990 Corrosion Cracking"
NP-6719-M "Proceedings: Primary Water Stress S406-3

April 1990 Corrosion Cracking, 1989 EPRI
Remedial Measures Workshop"
NP-6720-M "Microstructure Etching and Carbon S408-1
April 1990 Analysis Techniques"
NP-6721 -M "Corrosion Evaluation of Thermally RP1708-2

April 1990 Treated Alloy 600 Tubing in Primary
and Faulted Secondary Water
Environments"
Report No
NP-6721-SD "Corrosion Evaluation of Thermally RP1708-2
April 1990 Treated Alloy 600 Tubing in Primary
and Faulted Secondary Water
Environments"
NP-6743-L "Guidelines for PWR Steam S303-12
February 1991 Generator Tubing Specifications and
Repair; Volume 1: Specification for
Alloy 600 Steam Generator Tubing"
Repair; Volume 2: Guidelines for
Procurement of Alloy 690 Steam
Generator Tubing"
Repair; Volume 3: Steam Generator
Tube Sleeving: Design, Specification,
and Procurement Checklist"
Repair; Volume 4: Guidelines for
Tube Section Removal and
Examination"
NP-6750-M "Proceedings: 1989 EPRI Alloy 690 S408-1
April 1990 Workshop"
NP-6750-SD "Proceedings: 1989 EPRI Alloy 690 S408-1
April 1990 Workshop"
NP-6774 "Evaluation of Sampling Schemes for S404-11
March 1990 In-Service Inspection of Steam
Generator Tubing"
NP-6864-L "PWR Steam Generator S404-12,S404-1
December 1990 Tube-Plugging Limits: Technical 5,S404-18,S404-
Support Document for Expansion 19,S404-21,$40
Zone PWSCC in Roll Transitions" 4-24,S404-90,S4
04-92,S404-93
Report No
NP-6864-L "PWR Steam Generator Tube Repair S404-12,S404-1
December 1991 Limits: Technical Support Document 5,$404-18,$404-
for Expansion Zone PWSCC in Roll 21,S404-24,S40
Transitions, Rev. 1" 4-32,S404-37,S4
04-90,S404-92,S
404-93
NP-6865-L "Steam Generator Tube Integrity, S404-25

June 1991 Volume 2: PWR Steam Generator
Tube-Plugging
Criteria--Leak-Before-Break Analysis
for Primary Water Stress Corrosion
Cracking Near the Tubesheet
(Framatome Data)"
NP-6865-L "Steam Generator Tube Integrity, S404-25
September 1991 Volume 1: Burst Test Results and
Validation of Rupture Criteria
(Framatome Data)"
NP-6891-CCML "A Computer Program for S411-1
July 1990 Three-Dimensional, Two-Fluid Model
Thermal-Hydraulic Simulation of
Steam Generators: PORTHOS
Mod-01", 4 vols.
NP-6897-L "Steam Generator Tube Leakage S404-16
July 1990 Experiments and PICEP Correlations"
NP-6945 "Oxygen Control in Makeup Water for S402-4
August 1990 PWRs"
NP-6988-D "Self-Compensating Flow Distribution" RP2614-46
September 1990
NP-6990 "Eddy-Current Probe Characterization" S404-3

October 1990
NP-6997-M "Alloy 690 for Steam Generator S408-6

October 1990 Tubing Applications", 2 vols.
NP-6998-M "Laboratory Examination of Tubes S407-34

October 1990 R35C70 and R36C67 Removed From
the V. C. Summer Nuclear Station", 2
vols.
NP-7008 "Strain-Rate Damage Model for Alloy S303-8
October 1990 600 in Primary Water"
Report No
NP-7041 "Large Eddy Simulation on S410-12
November 1990 Supercomputers"
NP-7136-M "Destructive Examination of Steam S407-9

March 1991 Generator 11 Tubing at Calvert Cliffs
Power Plant Unit 1", 2 vols.
NP-7158 "Transport of Lead in PWR Secondary S401-2
April 1991 Cycles"
NP-7198 "Proceedings: 1990 EPRI Workshop S406-3

March 1991 on Circumferential Cracking of Steam
Generator Tubes" (NP-7198-M, Vol. 1;
NP-7198-S, Vol. 2)"
NP-7236 "Prairie Island-2 Steam Generator S401-1
April 1991 Hideout"
NP-7237 "Sulfate Ingress and Steam S401-11

June 1991 Generator Hideout at Saint Lucie Unit
NP-7346-M Vol. 1, "Acid Sulfate Corrosion in PWR S407-36
June 1991 Steam Generators"
NP-7346-SD Vol. 2, "Acid Sulfate Corrosion in PWR S407-36

June 1991 Steam Generators"
NP-7347 "Review of Field Use and Corrosion S401-2

June 1991 Experience With Phosphate
Chemistry in Nuclear Steam
Generators"
NP-7367-M "Investigation of Lead As a Cause of S407-16

June 1991 Stress Corrosion Cracking at Support
Plate Intersections"
NP-7370-M "Examination of Kewaunee Steam S407-42

June 1991 Generator Tubes R4C81 and R11C9"
NP-7371 -M "Destructive Examination of Tube S407-40

June 1991 R31C66 From the Ginna Nuclear
Plant Steam Generator"
NP-7380 "Nuclear Plant Design and S307-17
September 1991 Modification Guidelines for PWR
Steam Generator Reliability"
Report No
NP-7381 "Preoperational Practices for Steam S307-13
September 1991 Generators and Secondary-System
Components"
NP-7382 "Design and Operating Guidelines for S307-13
September 1991 Nuclear Power Plant Condensers"
NP-7383 "Design and Operating Guidelines for S307-13

September 1991 Condensate Polishers in Nuclear
Power Plants"
NP-7384 "Guidelines for Design of PWR Steam S307-13
September 1991 Generator Chemical Cleaning
Systems"
NP-7420 "Fractographic analysis of A Crack in a RP3154-2
July 1991 Zion Steam Generator"
NP-7474 "Evaluation of Leak and Burst S406-7
May 1993 Characteristics of Roll Transitions
Containing Primary Water Stress
Corrosion Cracks"
NP-7480-L "Outside Diameter Stress Corrosion S404-29
August 1992 Cracking (ODSCC) of Steam
Generator Tubing at Tube Support
Plates--A Database for Alternate
Repair Limits; Volume 1: 7/8-Inch
Outside Diameter (OD) tubing"
NP-7493 "Statistical Analysis of Steam S405-9
September 1991 Generator Tube Degradation"
NP-7494 "Hideout and Return of Complex S407-13

September 1991 Mixtures in Crevices"
NP-7524 "Steam Generator Performance S403-11

September 1991 Degradation"
TR-100371 "Steam Pressure Trends at R. E. S401-1

May 1992 Ginna"
Report No
TR-100407 "PWR Steam Generator Tube Repair S404-15,S404-1
March 1992 Limits: Technical Support Document 9,S404-21,S404-
for Outside Diameter Stress Corrosion 24,S404-29,S40
Cracking at Tube Support Plates" 4-30,S404-31,54
04-32,S404-33,S
404-36,S404-37,
S404-70,S404-7
1,S404-72
TR-100755 "PWR Advanced All-Volatile S409-2
July 1992 Treatment Additives, By-Products,
and Boric Acid"
TR-100756 "Loop Testing of Alternative Amines S409-11

June 1992 for All-Volatile Treatment Control in
PWRs"
TR-1 00758 "Advanced Studies in Chemistry S409-1

July 1992 Control With Morpholine"
TR-1 00790 "Demineralizer Operation With S409-3

July 1992 Morpholine and Boric Acid"
TR-1 00791 "Effects of Morpholine and Boric Acid S409-3

July 1992 Implementation on Secondary
Chemistry and Corrosion Product
Transport"
TR-100794 "Compatibility of PWR Gasket and S409-4

July 1992 Packing Materials and Resins With
Organic Amines"
TR-100795, "Qualification of Morpholine for S409-9
July 1992 Secondary System pH Control in
Plants", Tiers 1 &2, 2 vols.
TR-100836 "Statistical Signal Processing Methods RP2614-75

June 1992 for Ultrasonic Nondestructive
Evaluation"
TR-100865 "Characterization of the Resistance to S406-12
July 1992 PWSCC of Hydraulic Tube-Tubesheet
Expansions"
TR-100866 "Ginna Station Steam Generator S413-1
July 1992 U-Bend Tube Analysis for Chemical
Cleaning Data"
Appendix A--List of Steam Generator Publications A-37
Report No
TR-100909 "An Assessment of Inspection S404-6
July 1992 Options for Steam Generator Tubing"
TR-100910 "Tensile Deformation and Recovery S408-3
July 1992 Kinetics of Alloy 690"
TR-101010 "Correlation of Secondary-Side S407-7
August 1992 IGA/SCC Degradation of Recirculating
Steam Generator Tubing With the
On-Line Addition of Boric Acid"
TR-101103 "Proceedings: 1991 EPRI Workshop S407-7
August 1992 on Secondary-Side Intergranular
Corrosion Mechanisms"
TR-101104 "Mercury Modeling for PWSCC S404-28
August 1992 Length Sizing"
TR-101105 "Remedial Actions for Acidic Sulfate S407-36
August 1992 Corrosion"
TR-101106 "Adsorption of Sulfate in PWR Steam S407-37
August 1992 Generators: Laboratory Tests"
TR-101230 "Interim PWR Secondary Water S401
September 1992 Chemistry Recommendations for
IGA/SCC Control"
TR-101427 "Examination of Trojan Steam S413-2
November 1992 Generator Tubes; Volume 1:
Examination Results"
Appendixes"
Rockwell, Auger, and XPS Analyses"
TR-101983 "Characterization of Microstructure S303-1 0,S303-2
February 1993 and IGSCC of Alloy 600 Steam 6,S404-13
Generator Tubing"
TR-102134 "PWR Secondary Water Chemistry S401
May 1993 Guidelines--Revision 3"
A-38 SteamGenerator Reference Book, Revision I
Report No
TR-102355 "Residual Stresses in Roller-Expanded S406-11
May 1993 Steam Generator Tube Transitions"
TR-1 02491 "Speciation and Chemical Activities in S407-21

June 1993 Superheated Sodium Borate
Solutions"
TR-102549 "Eddy-Current Steam Generator Data RP2705-9,RP311
May 1993 Analysis Performance" 2-1,S404-20
TR-1 02706 "Proceedings: 1991 Symposium on S407-43

August 1993 Chemistry in High-Temperature
Aqueous Solutions"
TR-1 03117 "Effect of Boric Acid on Intergranular S407-14
October 1993 Corrosion in Tube Support Plate
Crevices"
TR-103311 "Reference Electrodes for S407-41
November 1993 Electrochemical Corrosion Potential
Monitoring in PWR Secondary
Systems"
A.2 LIST OF SGOG PUBLICATIONS BY PROJECT NUMBER
Report No
Project No Title and Date
RP304-11 "Evaluation of Alloy 600 Tube A77-34 NP-4504
From Steam Generator A of Arkansas March 1986
Nuclear One, Unit 1"
RP404-1 "EPRI Secondary Water Chemistry NP-516
Study" February 1977
RP404-1 "Modeling of Cooling-Water Inleakage NP-1 786
Effects in PWR Steam Generators" April 1981
RP404-1 "Corrosion-Product Transport in PWR NP-2149
Secondary Systems" December 1981
RP404-1 "HITCH Computer Code: Chemistry and NP-2388
pH Estimates of Concentrating Aqueous May 1982
Solutions"
RP404-1 "Use of the HITCH Code to Evaluate NP-3851
Amines as Neutralizers of Crevice January 1985
Acidity in Nuclear Steam Generators"
RP404-1 "The Impact of Makeup Water System NP-4050
Performance on PWR Steam Generator June 1985
Corrosion"
RP404-1 "Effect of Boric Acid Treatment on the NP-4270
Secondary Cycle at ANO-2" November 1985
RP623-1 "PWR Model Steam Generator NP-3138
Corrosion Studies" June 1983
RP623-2 "Tests of Isothermal Soaking NP-1 761
Procedures for Limiting Tube Denting in April 1981
RP623-2 "Neutralization of Crevice Acids" NP-3054
May 1983
RP623-3 "Evaluation of Condensate Polishers" NP-3046
June 1983
RP623-4 "The Design and Construction of Model NP-1965
Steam Generators for Corrosion Testing August 1981
of Alternate Materials"
RP623-5 "Sodium-Sulfate Solubilities in NP-3047
High-Temperature (250-374 0 C) Salt and July 1983
Acid Solutions"
EPRI Licensed Material.
Report No
RP623-6 "Crevice Corrosion of Support Alloys in NP-5017
the Secondary Environments of Nuclear March 1987
Steam Generators - Supplemental
Report"
RP698-1 "Evaluation of Electromagnetic-Acoustic NP-519
Concepts of Inspection of Steam August 1977
Generator Tubing"
RP699-1 "PWR Steam-Side Chemistry Follow NP-2541
Program" August 1982
RP699-2 "Crevice Hideout Return Testing" NP-4678
July 1986
RP704-1 "Model of Vaporous Carry-Over" NP-1787
April 1981
RP704-1 "Effect of Moisture Separator Drain NP-2505
Routing on OTSG Secondary System July 1982
Chemistry"
RP704-1 "Secondary Water Chemistry at Oconee" NP-3198
March 1984
RP1066-10 "Verification of the ATHOS3 Code NP-5728
Against Feedring and Preheat Steam May 1988
Generator Test Data"
RP1161-1 "Surface and Grain Boundary NP-3949
Segregation, Stress Corrosion Cracking, March 1985
and Corrosion Fatigue of Inconel 600"
RP1 171-1 "Rationale for Chemical Control of Feed NP-3048
and Boiler Water: Volumes 1, 2, and 3" January 1984
RP1171-2 "Oxide Growth Mechanisms on NP-4647
Chromium Alloy Steels" July 1986
RP1 171-3 "Salt Concentration in Heated Crevices NP-3050
and Simulated Scale" October 1983
RP1171-3 "The Effects of Oxygen, Copper, and NP-4648
Acid Chlorides on Denting Corrosion" July 1986
RP1 171-3 "Hideout and Return of Chloride Salts in NP-5015
Heated Crevices Prototypic of Support January 1987
Plates in Steam Generators"
RP1172-3 "Optical Technique for Internal NP-1244
Diametrical Measurement of Steam November 1979
Generator Tubes"
Report No
RP1447-2 "Method for Detecting Resin Leakage in NP-5802
LWR Coolant" May 1988
RP1450-1 "Evaluation of Alternative Alloys for NP-3703
PWR Steam Generator Tubing" October 1984
RP1450-2 "Stress Corrosion Cracking of Alloys 600 NP-5761-M
and 690 in All-Volatile-Treated Water at May 1988
RP1450-2 "Stress Corrosion Cracking of Alloys 600 NP-5761 -SP
and 690 in All-Volatile-Treated Water at May 1988
RP1571-2 "Catalytic Methods of Deoxygenating NP-5794
Water" April 1988
RP1571-5 "Cleanup of Steam Cycle Drains in NP-5621
PWRs With Once-Through Steam January 1988
Generators"
RP1571-5 "PWR Water Treatment Improvements: NP-5764
Cost-Benefit Analysis" May 1988
RP1571-7 "Evaluation of New pH Control Agents NP-4624
for PWR Secondary Water Systems" June 1986
RP1618-2 "Mechanism of Intergranular Corrosion NP-3957
of Inconel 600 Tubing in PWR Steam March 1985
Generators"
RP1618-3 "lntergranular Corrosion of Inconel 600 NP-3998-M
Tubing for PWR Steam Generators" May 1985
RP1708-1 "Optimization of Metallurgical Variables NP-3051
to Improve Corrosion Resistance on July 1983
Inconel Alloy 600"
RP1708-2 "Corrosion Evaluation of Thermally NP-6721 -M
Treated Alloy 600 Tubing in Primary and April 1990
Faulted Secondary Water
Environments"
RP1708-2 "Corrosion Evaluation of Thermally NP-6721 -SD
Treated Alloy 600 Tubing in Primary and April 1990
Faulted Secondary Water
Environments"
RP2160-8 "A Model of Sludge Behavior in Nuclear NP-4620
Plant Steam Generators" June 1986
Report No
RP2160-13 "Measurement of pH and Corrosion NP-5193
Potentials of Tube Alloys in Solutions May 1987
Found in Steam Generators"
RP2163-1 "Development of a Model for Predicting NP-3791
Intergranular Stress Corrosion Cracking September 1984
of Alloy 600 Tubes in Primary Water"
RP2163-3 "Electrochemistry and Corrosion of NP-4705
Alloys in High-Temperature Water" July 1986
RP2163-4 "Microstructural Effects on NP-5192
Microdeformation and Primary-Side May 1987
Stress Corrosion Cracking of Alloy 600
Tubing"
RP2163-5 "lntergranular Corrosion Mechanisms of NP-5396
Alloy 600" August 1987
RP2297-1 "Oxygen Control in PWR Makeup Water" NP-5623
January 1988
RP2599-1 "Methods for Evaluating Steam NP-4940
Generator Hideout Return Data: Case December 1986
Study at North Anna"
RP2599-1 "Prairie Island Chemical Hideout Study" NP-5592
February 1988
RP2599-1 "Effects of Alternate pH Control NP-5594
Additives in PWRs" February 1988
RP2599-2 "Evaluation of Sulfur Hexaflouride and NP-5008
Helium for Steam Generator Leak January 1987
Location"
RP2614-46 "Self-Compensating Flow Distribution" NP-6988-D
September 1990
RP2614-75 "Statistical Signal Processing Methods TR-100836
for Ultrasonic Nondestructive Evaluation" June 1992
RP2647-1 "On-Line Monitoring of Secondary-Cycle NP-5266
Chemistry with Morpholine at the June 1987
Beaver Valley Power Station"
RP2705-9,RP3 "Eddy-Current Steam Generator Data TR-1 02549
112-1 ,S404-20 Analysis Performance" May 1993
RP2758 "Progress in Radiation Control NP-6708
Technology
Report No
RP2812-3 "Proceedings: Workshop on Initiation of NP-5828
Stress Corrosion Cracking Under LWR May 1988
Conditions"
RP3154-2 "Fractographic analysis of A Crack in a NP-7420
Zion Steam Generator" July 1991
S101-1 "A Prototype EMAT System for NP-2836
Inspection of Steam Generator Tubes" January 1983
S102-1 "Steam Generator Crevice Gap NP-1419
Measurement by Induced-Vibration May 1980
Analysis"
S 102-3 "Induced-Vibration Analysis Probe for NP-2271
Measurement of Steam Generator February 1982
Tube-to-Support Plate Clearance"
S103-1 "Optical Scanner for Steam Generator NP-1 397
Tube Inspection" May 1980
S 103-2 "Optical Scanner System for Internal NP-1944
Inspection of Steam Generator Tubes" July 1981
S104-1 "Evaluation and Improvement of PWR NP-3020
Secondary System Oxygen Control July 1983
Measures"
S 104-2 "Evaluation of Secondary System NP-2448
Oxygen Control in PWR Power Plants in June 1982
the USSR"
$105-1 "Radiographic System for Evaluation of NP-2042
Steam Generator Support Plate September 1981
Integrity"
$105-1 "Steam Generator Support Plate NP-2823
Radiographic Evaluation System" January 1983
S106-1 "Evaluation and Optimization of NP-3273
Magnetic Filters on Simulated Boiler November 1983
Water"
S108-1 "Profilometry for Steam Generator Tube NP-2141
Dent Characterization" November 1981
$109-1 "Steam Generator Blowdown Filter NP-2659
Testing" November 1982
S110-1 "Fatigue Performance of Ni-Cr-Fe Alloy NP-2957
600 Under Typical PWR Steam March 1983
Generator Conditions"
Report No
Prowect No Title and Date
$111-1 "Determination and Verification of NP-3022
Required Water Chemistry Limits, Vol. January 1984
3: Pot Boilers 8A and 8B"
Required Water Chemistry Limits; March 1984
Volume 1: Summary Report"
Required Water Chemistry Limits; January 1984
Volume 3: Pot Boiler Tests"
Required Water Chemistry Limits; August 1984
Volume 5: Model Boiler 5B Tests"
S111-2 "Evaluation of Surrogate Boilers for NP-2967
Steam Generators" March 1983
S112-1 "Neutralization of Steam Generator NP-3023
Denting" September 1982
S112-2 "Characterization of Single-Tube Model NP-3024
Boiler Dented Intersection Specimens" May 1983
S113-1 "Evaluation and Categorization of NP-2656
Secondary System Layup and Cleanup December 1982
Practices for PWR Plants"
S113-1 "Evaluation of Secondary System Layup NP-2977
and Cleanup Practices and Processes" April 1983
S115-1 "Field Experiences with Multifrequency NP-2299
Multiparameter Eddy Current March 1982
Technology"
S116-1 "Implementation of Boric Acid in the NP-3066
Field--Indian Point Unit 3 Plant" May 1983
S117-1 "Hydrogen Evolution Monitoring As a NP-2650
Measure of Steam Generator Corrosion" November 1982
S118-1 "Single Tube Thermal and Hydraulic NP-2046
0
Tube Support Test" September 1981
$119-1 "Steam Generator Sludge Pile Model NP-1941
Boiler Testing" July 1981
S119-1 "Steam Generator Sludge Pile Model NP-2041
Boiler Testing (Sludge Characterization)" September 1981
Report No
Pro*ect No Title and Date
S119-2 "Tube-to-Tube Sheet Joint Test", 2 vols. NP-3013
March 1983
S119-3 "Effect of Calcium Hydroxide and NP-3060
Carbonates on IGA and SCC of Alloy May 1983
600"
S120-1 "Investigation of Steam Generator NP-3068
Corrosion Products Under Typical PWR May 1984
Operating Conditions"
S121-1 "Thermal-Hydraulic Tests of Steam NP-2838
Generator Tube Support Plate January 1983
Crevices", 3 vols.
S122-1 "Method of Reducing Carry-Over and NP-1 607
Reducing Pressure Drop Through November 1980
Steam Separators"
S124-1 "Laboratory Program to Examine Effects NP-3012
of Layup Conditions on Pitting of Alloy April 1983
600"
S125-1 "Magnetic Flux Leakage for NP-1427
Measurement of Crevice Gap Clearance June 1980
and Tube Support Plate Inspection"
S125-1 "Magnetic Flux Leakage for NP-2857
Measurement of Crevice Gap Clearance February 1983
and Tube Support Plate Inspection"
S125-3 "Steam Generator Probe Positioning NP-2774
Device" December 1982
S126-1 "Steam Generator Mock-Up Facilities" NP-1785
April 1981
S127-1 "Chemical Cleaning Demonstration Test NP-1 829
No. 1 in a Mock-Up Steam Generator" April 1981
S127-1 "Electrochemical Behavior of Ferrous NP-2701
and Ferric Ions in EDTA/N2H4 October 1982
Solutions"
S127-1 "Chemical Cleaning Solvent and NP-2976
Process Testing" April 1983
S127-1 "Chemical Cleaning Demonstration Test NP-3008
No. 2 in a Mock-Up Steam Generator" April 1983
Report No
S127-1 "Data Acquisition-Reduction System for NP-3302
Chemical Cleaning Processes for February 1984
S128-1 "Steam Generator Chemical Cleaning NP-1 789
Demonstration Test No. 1 in a Pot April 1981
Boiler"
S128-1 "Steam Generator Chemical Cleaning: NP-1976
Demonstration Test No. 2 in a Pot August 1981
Boiler"
S128-1 "Steam Generator Chemical Cleaning NP-2983
Demonstration Test No. 3 in a Pot April 1983
Boiler"
S128-1 "Steam Generator Chemical Cleaning: NP-2990
Demonstration Test in a Model Boiler" June 1983
S129-1 "CALIPSOS Code Report", 2 vols. NP-1391
April 1980
S129-1 "Thermal-Hydraulic Characteristics of a NP-1 392
Combustion Engineering Series 67 April 1980
S129-1 "Thermal-Hydraulic Characteristics of a NP-1528
Combustion Engineering System 80 September 1980
Westinghouse Model F Steam March 1981
Generator", 2 vols.
Westinghouse Model 51 Steam March 1981
Generator", 2 vols.
S130-1 "Thermal-Hydraulic Analysis of the NP-1546
Combustion Engineering System 80 September 1980
Steam Generator"
S130-1 "Thermal Hydraulic Analysis of the NP-1 678
Combustion Engineering Series 67 January 1981
Steam Generator"
S130-1 "Thermal-Hydraulic Analysis of the NP-2683
Westinghouse Model 51 Steam October 1982
Generator"
S131-1 "Thermal-Hydraulic Analysis of NP-1 431
Once-Through Steam Generators" June 1980
Appendix A-List of Stearn Generator Publications A-47
Report No
S132-2 "Effect of Changing the Hydrazine NP-2270
Injection Point at the Carolina Power February 1982
and Light, H. B. Robinson Plant"
S 132-5 "Salem Unit 1--Denting Mitigation NP-2703
Program: Implementation of Improved October 1982
Oxygen, Chloride, and Copper Control"
S 132-8 "Hydrazine Usage for Corrosion Control NP-2969
in PWR Plants With Powdered Resin March 1983
Condensate Polishers"
S132-10 "Evaluation of Field Applications of Boric NP-3278
Acid in PWR Steam Generators" March 1984
S132-12 "Application of SGOG Secondary NP-3142-LD
System Chemistry Guidelines at June 1983
Pressurized Water Reactors"
S133-1 "Boiling Heat Transfer in a Narrow NP-2610
Eccentric Annulus" September 1982
S134-1 "Boiling in Narrow Crevices in Steam NP-2638
Generators" November 1982
S136-1 "Evaluation of Steam Generator Tube NP-1794
85-127 From Oconee 1B" April 1981
S136-2 "Evaluation of Oconee Steam NP-2082
Generator Debris" October 1981
S136-4 "Evaluation of the Lower Tube Sheet NP-3026-LD
Region of an Oconee 1 Steam July 1983
Generator Tube"
S136-5 "Analysis of Sludge From Indian Point NP-3027
No. 3 June 1983
S138-1 "Examination of Three Steam Generator NP-2958-LD
Tubes From Point Beach Unit 1 Nuclear January 1984
Power Plant"
S138-2 "Examination of Steam Generator Tube NP-1 412
R45C52 From the Ginna Power Plant" May 1980
S138-2 "Examination of Three Steam Generator NP-2534-LD
Tubes From the Ginna Nuclear Power August 1982
Plant"
S138-3 "Examination of Steam Generator Tube .NP-2539-LD
A(18-37) from the Point Beach Unit 2 August 1982
Report No
S138-4 "Evaluation of Steam Generator U-Bend NP-2629-LD
Tubes From Trojan Nuclear Power September 1982
Plant"
S138-5 "Steam Generator U-Bend Tube NP-3571
Examination" February 1984
R12C66 From Indian Point No. 3" May 1983
S138-6 "Evaluation of Steam Generator Tubes NP-3070-LD
R7C45 and R21C46 From the Ginna May 1983
S138-7 "Chemical and Metallographic NP-2972-LD
Examination of Tube Specimens From March 1983
Ringhals 2 Steam Generator"
S138-8 "lntergranular Stress Corrosion Cracking NP-3057
of Ni-Cr-Fe Alloy 600 Tubes in PWR. May 1§983
Primary Water--Review and Assessment
for Model Development"
S139-1 "Calibration of Instrumented Steam NP-2805
Separators to Determine Quality and January 1983
Flow Distribution in an Operating Steam
Generator"
S140-1 "Development of Sensors and NP-1875
Instrumentation for the TMI-2 OTSG June 1981
Tube Vibration Measurements Program"
S140-1 Vol. 1, "Flow Induced Vibration Analysis NP-1876
of Three Mile Island Unit 2 June 1981
Once-Through Steam Generator Tubes"
S140-1 "In-Air Vibration Analysis of Three Mile NP-2692
Island Unit-2 Once-Through Steam October 1982
Generator Tubes", 2 vols. (Vol. 2 is
available on microfiche only)
$142-1 "Evaluation of Pulse-Echo Ultrasound NP-2285
for Steam Generator Tube-to-Support March 1982
Plate Gap Measurement"
$143-1 Vol. 1, "Tube Support Response to NP-2971
Tube Denting Evaluation" and Vol. 2 May 1983
(Appendix H-Test Data)
$144-1 "Loads on Steam Generator Tubes NP-2652
During Simulated Loss-of-Coolant November 1982
Accident Conditions"
Report No
S145-1 "Prevention of Wear Problems in PWR NP-2711
Steam Generators--An Annotated October 1982
Bibliography"
S146-1 "Estimation of Diffusion Coefficients for NP-1 963
Electrolyte in Hot Water" August 1981
S146-1 "Diffusion and Hideout in Crevices" NP-2979
March 1983
S147-1 "Metal Cation Inhibitors for Controlling NP-2655
Denting Corrosion in Steam Generators" December 1982
S148-1 "Nonproprietary Corrosion Inhibitors for NP-3030
Solvents to Clean Steam Generators" June 1983
S149-1 "Cleaning Steam Generators Offline NP-2815
(Soaking) With Chelants" February 1983
S150-1 "Steam Generator Chemical Cleaning NP-3009
Process Development" April 1983
S150-3 "Chemical Cleaning Process NP-2987
Evaluation--Westinghouse Steam April 1983
Generators"
S151-1 "An Engineering and Probabilistic NP-3065
Analysis of Tube Cracking Performance June 1983
in Once-Through Steam Generators", 2
vols.
S153-1 "Flow-Induced Vibration of Steam NP-4559
Generator Tubes" May 1986
S154-1 "Tests of Types 51 A and 51 M Steam NP-2689
Generators at Bugey 4 and Tricastin 1 October 1982
Nuclear Power Plants"
S154-2 "Thermal-Hydraulic Code Qualification: NP-2872
ATHOS2 and Data From Bugey 4 and February 1983
Tricastin 1"
S154-2 "An Evaluation of Techniques to Predict NP-2986
Flow-Induced Tube Vibration in a March 1983
Preheat Steam Generator"
S155-1 "Visual Inspection Equipment for the NP-1859
Secondary Side of Steam Generators" May 1981
S156-1 "Three-Dimensional Thermal/Hydraulic NP-3031
Code Development" January 1984
Report No
S157-1 "Causes of Denting; Volume 1: NP-3275
Summary Report" May 1984
Laboratory Test Results" May 1984
S157-1 "Causes of Denting; Volume 3: Plant NP-3275
Chemistry Correlations" May 1984
Isothermal Tests" May 1984
Contaminant Threshold Tests" December 1983
S157-1 "Causes of Denting; Volume 6: NP-3275-LD
In-Leakage Concentration Experiments" March 1984
S158-1 "Investigation of Phosphate-Sludge NP-2963
Interactions" March 1983
S160-1 "Steam Generator Data Base" NP-3033
June 1983
S163-1 "State-of-the-Art Evaluation of NP-2978
Condensate Polisher Performance" April 1983
S163-1 "Condensate Polisher Resin Leakage NP-2981
Quantification and Resin Transport April 1983
Studies"
S164-1 "Evaluation of Steam Generator Fluid NP-2993
Mixing During Layup" May 1983
S165-1 "OTSG Tube Failures: Upper Tube NP-2790
Sheet Corrosion Tests" December 1982
S166-1 "On-Line Use of Chelants in Nuclear NP-2973
Steam Generators--Feasibility Study" April 1983
S167-1 "A Review of Condensate Demineralizer NP-2975
Regeneration Procedures" April 1983
S167-1 "Condensate Demineralizer NP-3035
Performance During Periods of High June 1983
Condenser Leakage"
S167-1 "Condensate Polishing at Surry Nuclear NP-3037
Power Station" June 1983
S167-1 "Condensate Polishing at Plant Bowen" NP-3311
February 1984
Appendix A-List of Steam Generator Publications A-5 I
Report No
S167-7 "Modeling of Crevice Corrosion in PWR NP-2825-LD
Steam Generators" January 1983
S168-1 "Thermal and Hydraulic Code NP-2887
Verification, ATHOS2 and Model Boiler February 1983
No. 2 Data", 3 vols.
S169-1 "Effect of Out-of-Plane Denting Loads NP-1479
on the Structural Integrity of Steam August 1980
Generator Internals"
S170-1 "Secondary Water Chemistry Control at NP-1 863
Genkai No. 1 Design and Operating May 1981
Considerations"
S170-1 "Secondary Water Chemistry at NP-2268
Ringhals Unit 2" February 1982
S170-1 "Secondary Water Chemistry Control at NP-2706
St. Lucie No. 1" October 1982
S170-1 "Secondary Water Chemistry at NP-2974
Millstone 2" April 1983
S171-1 "A Study of the Boiling Processes in the NP-3018
Sludge Deposit of Steam Generators" May 1983
S172-1 "Numerical Simulation of Steam NP-3063
Separators Used in Nuclear Steam June 1983
Generators and Reactor Systems", 2
vols.
S173-1 "Steam-Water Separation in a NP-2996
Swirl-Vane Centrifugal Separator", 2 June 1983
vols.
S174-1 "Fretting & Fatigue in Steam Generators NP-3039
Part 1, Determination of Tube-to-Tube November 1983
Support Interaction Characteristics"
S176-1 "Flow Induced Vibration Analysis of NP-1 888
Oconee 2B OTSG Tubes" June 1981
S176-1 "Static Strain Analysis of TMI-2 OTSG NP-2146
Tubes" December 1981
S179-1 "Predictions and Measurements of NP-2643
Isothermal Airflow in a Model Once November 1982
Through Steam Generator"
Report No
S179-1 "Measurement-and THEDA Code NP-2765
Prediction of Thermal, Velocity, and December 1982
Pressure Fields in a Once-Through
Steam Generator Heated Air Model"
S180-1 "Tube Support Plate Thermal and NP-3052
Hydraulic Testing" May 1983
S181-1 "Optical Probe for Steam Generator NP-2863
Tube Dent" February 1983
S182-1 "Condenser Inleakage Monitoring NP-2597
System Development" October 1982
S183-1 "Demonstration Tests on PWR Steam NP-1861
Generator Tube-Tube Sheet Crevice May 1981
Flushing Procedures"
S183-2 "Neutralization of Tube Sheet Crevice NP-3040
Corrosion" May 1983
S183-3 "Amine Borane Compounds in NP-3041
Crevice-Cleaning Solutions" June 1984
S183-4 "IGA of Alloy 600 in High-Temperature NP-3059
Solutions of Sodium Hydroxide May 1983
Contaminated With Carbonate"
S185-1 "Ultrasonic Enhancement of Chemical NP-2997
Cleaning of Steam Generators" April 1983
S186-1 "Model Tests of a Once-Through Steam NP-3042
Generator for Lane Blocker Assessment June 1982
and THEDA Code Verification"
S187-1 "Monitoring Systems for Determining Air NP-2865
Inleakage and Oxygen Concentrations March 1983
in the Secondary Cycle of Power Plants"
S189-1 "Guide to Design of Secondary Systems NP-2294
and Their Components to Minimize March 1982
Oxygen-Induced Corrosion"
S190-1 "Crevice Hideout Return Testing" NP-3067
May 1983
$191-1 "Stress Corrosion of Alloys 600 and 690 NP-3043
in Acidic Sulfate Solutions at Elevated October 1983
Temperatures"
Report No
S191-2 "The Role of Electrochemistry in Causes NP-2962
and Measurement of Corrosion in PWR March 1983
Steam Generators"
S191-3 "In Situ Heat Treatment and Polythionic NP-3056
Acid Testing of Inconel 600 Row I April 1983
Steam Generator U-Bends", 3 vols.
S192-1 "Corrosion Performance of Alternative NP-3044
Steam Generator Materials and July 1983
Designs", 3 vols.
S 192-2 "Stress Corrosion Cracking of Alloy 600 NP-3061
and Alloy 690 in All Volatile Treated May 1983
S192-3 "Stress Relief to Prevent Stress NP-3055
Corrosion in the Transition Region of May 183
Expanded Alloy 600 Steam Generator
Tubing"
S193-1 "lntergranular Attack of Alloy 600: NP-3062
High-Temperature Electrochemical May 1983
Tests"
S193-2 "The Design and Construction of Deep NP-3058
Tube Sheet Crevice Devices for May 1983
Producing Intergranular Attack of Heat
Transfer Tubing"
"Elastic Stress Analysis of Small Radius NP-2944
S194-4
U-Bend Steam Generator Tubes" March 1983
S195-1 "Inhibition of Steam Condensate NP-2492
Corrosion of Copper-Based Alloys by July 1982
Hydrazine"
S196-1 "Effects of Hydrazine and pH on the NP-2654
Corrosion of Copper-Alloy Materials in November 1982
AVT Environments With Oxygen"
S201-1 "Eddy Current NDE for Intergranular NP-2862
Attack" February 1983
S202-1 "Steam Generator U-Bend Eddy-Current NP-3010
NDE" April 1983
S202-2 "Application of an Eddy-Current NP-2339
Technique to Steam Generator U-Bend April 1982
Characterization"
Report No
S203-1 "Steam Generator Tube-Plugging and NP-2921
Tube-Sleeving Criteria: Assessment of March 1983
Current Practices"
S204-1 "Crevice Corrosion of Lattice Support NP-3045
Alloys in Secondary Environments of July 1983
S205-7 "Return of Hideout Chemicals in PWR NP-4563
Steam Generators During Power and May 1986
Temperature Reductions"
S209-1 "Flushing Efficiency of Steam Generator NP-3053
Tube-Tube Sheet Assemblies With April 1983
Restricted Crevices"
S301-1 "Evaluation of Eddy-Current Procedures NP-3928
for Measuring Wear Scars in Preheat April 1985
Steam Generators"
S301-1 "Nondestructive Evaluation Methods to NP-5902
Measure Inside Diameters of Steam July 1988
Generator Tubing"-
S301-5 "Library of PWR Steam Generator NP-5503
Tubing Samples" January 1988
S301-6 "Array Coil Probe" NP-5009
March 1987
S302-1 "lntergranular Attack and Stress NP-4051
Corrosion Cracking of Alloy 600 in June 1985
High-Temperature Caustic Solutions
Containing Contaminants"
S302-1 "Mechanisms of Intergranular Attack NP-5129
and Stress Corrosion Cracking of Alloy July 1987
600 by High-Temperature Caustic
Solutions Containing Impurities"
S302-3 "Simulation of Intergranular Attack on NP-4272
Alloy 600 Tubing in Tubesheet Crevices" October 1985
S302-3 "Production of Intergranular Attack of NP-5263
Alloy 600, Alloy 690, and Alloy 800 July 1987
Tubing in Tubesheet Crevices"
Simulation Tests" June 1985
Laboratory Investigations" August 1985
Report No
S302-4 "[ntergranular Attack of Alloy 600 NP-5377
Tubing: Simulation Tests" August 1987
S302-5 "Phase Relations and Fluid NP-5138
Compositions in Steam Generator April 1987
Crevices"
S302-6 "lntergranular Attack or Corrosion in a NP-5120
Once-Through Model Steam Generator" July 1987
S302-7 "Supplementary Examination of NP-5928-M
Alternative Materials in a Model Steam January 1989
Generator, Volume 2: Correlation of
Model 10 Defects With Eddy-Current
Indications"
Alternative Materials in a Model Steam October 1988
Generator, Volume 3: Tube
Characterization by Metallography and
Transmission Electron Microscopy"
S302-7 "Supplementary Examination of NP-5928-SP
Generator; Volume 1: Tubesheet and
Tube Bundle Examinations"
3302-7 "Supplementary Examination of NP-5928-M
Generator; Volume 1: Tubesheet and
Tube Bundle Examinations"
Generator; Volume 2: Correlation of
Indications"
S302-7 "Supplementary Examination of NP-5928-SP
Generator; Volume 2: Correlation of
Indications"
S302-7 "Supplementary Examination of NP-5928
Generator; Volume 3: Tube
Characterization by Metallography and
Transmission Electron Microscopy"
Report No
S302-8 "Microchemistry of Corroded NP-3847
Intergranular Surfaces" February 1985
S302-9 "Evaluation of Environmental Effects on NP-4802
Intergranular Attack of Alloy 600" September 1986
S302-10 "Use of Pourbaix Diagrams to Infer NP-4831
Local Pitting Conditions" October 1986
S302-10 "Production of Eddy-Current Standards NP-5109
for Caustic Intergranular Corrosion" April 1987
S302-12 "Laboratory Examinations of Selected NP-4625
Tubes From Test Facilities of the Steam July 1986
Generator Owners Group"
S302-13 "Evaluation of Intergranular Attack on NP-4978
Alloy 600, Volume 2: Effectiveness of December 1987
Boric Acid"
S302-13 "Evaluation of Intergranular Attack on NP-4978
Alloy 600, Volume 1: Evaluation of December 1986
Causes"
S302-13 "Boric Acid Application Guidelines for NP-5558
Intergranular Corrosion Inhibition" December 1987
S302-16 "Mechanisms for Formation and NP-5368
Disruption of Surface Oxides" August 1987
S302-17 "Electrochemical Corrosion of Alloy 600 NP-5406
in Secondary Water" September 1987
S302-18 "Chemical Enhancement of Crevice NP-5199
Flushing", 2 vols. June 1987
S302-21 "Evaluation of PWR Tubesheet NP-4898
Crevice-Flushing Approaches" November 1986
S302-22 "Proceedings: IGA Remedies NP-3818
Contractors' Meeting" November 1984
S302-22 "Proceedings: 1983 Workshop on NP-4458
Secondary-Side Stress Corrosion March 1986
Cracking and Intergranular Corrosion of
PWR Steam Generator Tubing"
Secondary-Side Stress Corrosion March 1986
Cracking and Intergranular Corrosion of
PWR Steam Generator Tubing"
-EPRI Licensed Material
Report No
S302-22 "Proceedings: 1985 EPRI Workshop on NP-4929
Remedial Actions for Secondary-Side December 1986
Intergranular Corrosion"
3302-23. "Micromorphology and Microchemistry NP-5400
of Corrosion Products in Tubesheet- September 1987
Crevices"
S302-24 "lntergranular Attack of Alloy 600: NP-6115-M
Simulation and Remedial Action Tests" February 1989
S302-25 "Oxide-Film Compositions and NP-5712
Morphology on Alloy 600 Tubes From April 1988
Steam Generators, North Anna Unit 1
and Point Beach Unit 1"
S302-1 0,$302- "Remedial Methods for Intergranular NP-4635
13,$302-14 Attack of Alloy 600 Tubing, Volume 1: June 1986
Plant Corrosion Morphologies and
Remedial Methods, Volume 2: Additives
and Test Plans for Remedial Methods,
Volume 3: Boric Acid and Acetic Acid
Remedial Methods"
S303-3 "Residual and Applied Stress Analysis of NP-5282
an Alloy 600 Row 1 U-Bend" September 1987
S303-4 "Reference Electrode for PWRs" NP-5155
May 1987
Primary-Side Stress Corrosion Cracking June 1987
Primary-Side Stress Corrosion Cracking November 1987
S303-6 "Design Review Checklist: Steam NP-4296
Generator Sleeving" December 1985
S303-6 "Specifications for In Situ Stress Relief NP-4364
of PWR Steam Generator Tube December 1985
U-Bends and Roll Transitions"
S303-6 "Status and Suggested Course of NP-4594
Action for Nondenting-Related May 1986
Primary-Side IGSCC of
Westinghouse-Type Steam Generators"
S303-8 "Strain-Rate Damage Model for Alloy NP-7008
600 in Primary Water" October 1990
Report No
S303-9 "Microstructure, Microchemistry, and NP-4465
Microdeformation of Alloy 600 Tubing" February 1986
S303-11 "In Situ Heat Treatment of U-Bends" NP-5496
November 1987
S303-12 "Guidelines for PWR Steam Generator NP-6743-L
Tubing Specifications and Repair; February 1991
Volume 1: Specification for Alloy 600
S303-17 "Specially Prepared Alloy 600 Tubing" NP-5072
February 1987
S303-21 "Status of Cracking and Remedial NP-4459
Measures for PWR Steam Generators February 1986
With Full-Depth Expanded Tubing"
S303-22 "Stress Corrosion Cracking Test of NP-5012
Expanded Steam Generator Tubes" January 1987
S303-27 "Qualification of Remedial Methods to NP-5249
Prevent Primary-Side Stress Corrosion June 1987
Cracking of Steam Generator Tubing", 3
vols.
S303-29 "Tubesheet Expansion Improvements" NP-5547
December 1987
S303-10,$303- "Characterization of Microstructure and TR-101983
26,S404-13 IGSCC of Alloy 600 Steam Generator February 1993
Tubing"
S304-2 "Steam Generator Tube Sampling: NP-4626
Feasibility Study" June 1986
S304-3 "Metallurgical and Chemical Evaluation NP-5022-LD
of Tubes R17C85, R28C32, and December 1986
R26C50 From a Doel Unit 2 Steam
Generator"
R34C54 From the Ginna Nuclear Power September 1984
Plant"
S304-5 "Destructive Examination of Zion Unit 1 NP-4375
Tube and Antivibration Bar Samples" December 1985
S304-6 "Examination of Steam Generator Tube NP-4829-LD
Sections From the Millstone Unit 2 October 1986
Report No
S304-7 "Destructive Examination of Steam NP-4737-LD
Generator Tubes Removed From July 1986
Millstone Point Unit 2"
S304-8 "Evaluation of Alloy 600 Tubes From NP-3964-LD
Steam Generators A and B of the March 1985
Palisades Nuclear Power Plant"
S304-10 "Calvert Cliffs Unit 1 Tube Examination" NP-4904
November 1986
S304-14 "Examination of Tubes R4C19HL, NP-4745-LD
R6C18HL, and R16C33HL From Steam August 1986
Generator 12 of the Prairie Island
Nuclear Station Unit 1"
S304-15 "Examination of Steam Generator NP-4760-LD
Tubes R18C53HL and R18C43HL From August 1986
the Jose Cabrera (Zorita) Nuclear Power
Station"
S304-16 "Evaluation of Steam Generator Hot NP-4759-LD
Leg Tubes R18C43 and R18C53 From August 1986
the Cabrera Nuclear Power Plant"
S304-17 "Removal of a Tubesheet Sample From NP-4901
a Retired Point Beach Unit 1 Steam December 1986
Generator"
S304-18 "Decontamination of Point Beach-1 and NP-5344
Surry-2 Steam Generator Tubesheet August 1987
Sections"
S304-19 "Point Beach-1 Tubesheet Crevice NP-5660-LD
Chemistry" February 1988
S304-20 "Examination of Tubes R3C41HI and NP-5420-LD
R9C58HL of Steam Generator C, North October 1987
Anna Unit 1"
S304-21 "Laboratory Evaluation of Steam NP-5397
Generator Tubes 120/12, 79/9, and August 1987
59/95 from St. Lucie Unit 1"
S305-2 "PWR Steam Generator NP-3477
Chemical-Cleaning Data Base", 2 vols. April 1984
S305-9 "Assessment of Sulfur in Chemical NP-5026
Cleaning of PWR Steam Generators" February 1987
S305-11 "Chemical Cleaning of PWR Steam NP-4708
Generator Sludge Piles" July 1986
Report No
S305-13 "Effect of Venting on Crevice Cleaning NP-4600
for PWR Steam Generators" May 1986
S305-19 "Chemical Cleaning Waste Disposal" NP-4954
November 1986
S305-21 "Chemical Cleaning of Millstone Unit 2" NP-4597
May 1986
S305-1 0,S305- "Weld Region Corrosion During NP-5267
14,S305-15,S3 Chemical Cleaning of PWR Steam July 1987
05-18 Generators", 2 vols.
S306-1 "Steam Generator Tube Denting--Field NP-4456
Study" March 1986
S306-1 "lntergranular Corrosion of Steam NP-4457
Generator Tubes--Field Study" February 1986
S306-12 "Organic Impurity Transport in PWR NP-4099
Secondary Systems" August 1985
S306-12 "Industry-Wide Survey of PWR Organics" NP-4698
July 1986
S306-13 "Evaluation of Changing PWR Polisher NP-5074
Operation From Regenerated to March 1987
Throwaway Resins"
S306-15 "High-Sensitivity Dissolved-Gas NP-4564
Monitoring System With Applications for May 1986
PWR Secondary-Side Chemistry"
S306-16 "Resin and Ionics Leakage From NP-4521
Condensate Polishers With and Without April 1986
Inert Resin"
S306-17 "Hideout of Impurities in Steam NP-5092
Generators During Heat-Up and Power March 1987
Ascension"
S306-18 "Survey of Domestic and Foreign PWR NP-4671
Experience With Morpholine in July 1986
Chemistry Control by All-Volatile
Treatment"
S306-19 "Guidelines for the Design and NP-6377-M
Operation of Makeup Water Treatment June 1989
Systems"
S306-21 "Field Studies on Intergranular Attack" NP-6455
July 1989
Report No
S306-12,S306- "Side-by-Side Comparison of NP-4731
13 Techniques for Analyzing Organic Acids, September 1986
Total Organic Carbon, and Anions"
S307-13 "Preoperational Practices for Steam NP-7381
Generators and Secondary-System September 1991
Components"
S307-13 "Design and Operating Guidelines for NP-7382
Nuclear Power Plant Condensers" September 1991
S307-13 "Design and Operating Guidelines for NP-7383
Condensate Polishers in Nuclear Power September 1991
Plants"
S307-13 "Guidelines for Design of PWR Steam NP-7384
Generator Chemical Cleaning Systems" September 1991
S307-17 "Nuclear Plant Design and Modification NP-7380
Guidelines for PWR Steam Generator September 1991
Reliability"
S308-1 "Pitting Corrosion of Alloy 600 Steam NP-3905
Generator Tubing: Results of a February 1985
Laboratory Scoping Study"
S308-3 "Causes of and Corrective Actions for NP-5037
Pitting in Steam Generator Tubing" February 1987
S308-4 "Investigation of Causes and Corrective NP-5248
Actions for Pitting in Steam Generator June 1987
Tubes: Prototypic Tests"
S308-5 "Localized Electrochemical Corrosion of NP-4754
Nickel-Based Alloys" September 1986
S308-6 "Pitting in Steam Generator Tubing: NP-5207
Causes and Corrective Actions" June 1987
S309-1 "Recirculating Steam Generator NP-4686
Tubesheet Crevice Flushing July 1986
Procedures"
S310-1 "Experimental Prediction of Tube NP-5541
Support Interaction Characteristics in June 1988
Steam Generators", 2 vols.
S310-2 "PWR Steam Generator Tube Fretting NP-6341
and Fatigue Wear" April 1989
Report No
S310-3 "Analytic Prediction of Complex NP-5543
Unsteady Flow Fields in Preheat PWR May 1988
Steam Generators"
S310-5 "Vibration and Wear Prediction for NP-5565
Steam Generator Tubes" June 1988
S310-6 "Numerical Prediction of NP-5556
Turbulence-Induced Steam Generator May 1988
Tube Vibration"
S310-7 "Experimental Determination of NP-5540
Turbulent Buffeting Effects in Tube May 1988
Bundles"
S310-8 "ATHOS3 Computer Code Verification", NP-5557
2 vols. June 1988
S310-12 "Flow Velocity and Turbulence NP-5542
Measurements in a Laboratory-Scale May 1988
Steam Generator Model"
S310-13 "Flow Velocity and Turbulence NP-5570
Measurements in a Westinghouse D-4 June 1988
Steam Generator Model", 2 vols.
S310-14 "Prediction of Localized Flow Velocities NP-5555
and Turbulence in a PWR Steam May 1988
Generator"
S311-1 "Correlation of Tube Support Structure NP-4672
Studies" July 1986
S311-2 "Correlation of Tube Support Corrosion NP-4818
Studies" October 1986
S311-3 "Sulfate Hideout in Heated Crevices" NP-5156
May 1987
S311-4 "Caustic Concentration in Tube Support NP-5073
Plate Crevices of Steam Generators" March 1987
S311-4 "Hideout of Sodium Salts in Tubesheet NP-5265
Crevices" July 1987
S311-5 "Steam Generator Cold Leg Thinning in NP-5140
Operating Plants" April 1987
S401 "Interim PWR Secondary Water TR-101230
Chemistry Recommendations for September 1992
IGA/SCC Control"
Report No
S401 "PWR Secondary Water Chemistry TR-1 02134
Guidelines--Revision 3" May 1993
S401-1 "Prairie Island-2 Steam Generator NP-7236
Hideout" April 1991
S401-1 "Steam Pressure Trends at R. E. Ginna" TR-100371
May 1992
S401-2 "Transport'of Lead in PWR Secondary NP-7158
Cycles" April 1991
S401-2 "Review of Field Use and Corrosion NP-7347
Experience With Phosphate Chemistry June 1991
in Nuclear Steam Generators"
S401-11 "Sulfate Ingress and Steam Generator NP-7237
Hideout at Saint Lucie Unit 1" June 1991
S402-4 "Oxygen Control in Makeup Water for NP-6945
PWRs" August 1990
S403-3 "CECIL: A Robot for Secondary-Side NP-5929
Maintenance of PWR Steam February 1989
Generators"
S403-5 "Qualification of PWR Steam Generator NP-6356-M
Chemical Cleaning for Indian Point-2" May 1989
S403-11 "Steam Generator Performance NP-7524
Degradation" September 1991
$403-2,$403-4, "NDE and Mechanical Removal of NP-5563
RP2755 Sludge in PWR Steam Generators", 2 January 1988
vols.
"PWR Steam Generator Examination NP-6201
S404
Guidelines: Revision 3" November 1992
S404-3 "Eddy-Current Probe Characterization" NP-6990
October 1990
S404-6 "An Assessment of Inspection Options TR-100909
for Steam Generator Tubing" July 1992
S404-8 "PWR Steam Generator Examination NP-6201
Guidelines: Revision 2" December 1988
S404-10 "Signal Processing for Steam Generator NP-5773
Inspection" April 1988
Report No
S404-11 "Evaluation of Sampling Schemes for NP-6774
In-Service Inspection of Steam March 1990
Generator Tubing"
. I
S404-12 "Roll Transition Inspection of Doel-2 NP-6716-L
Steam Generator Tubes; Volume 3: May 1992
September 1991 Inspection"
S404-14 "Belgian Approach to Steam Generator NP-6626-M
Tube Plugging for Primary Water Stress March 1990
Corrosion Cracking"
S404-16 "Steam Generator Tube Leakage NP-6897-L
Experiments and PICEP Correlations" July 1990
S404-25 "Steam Generator Tube Integrity, NP-6865-L
Volume 1: Burst Test Results and September 1991
Validation of Rupture Criteria
(Framatome Data)"
S404-25 "Steam Generator Tube Integrity, NP-6865-L
Volume 2: PWR Steam Generator June 1991
Tube-Plugging
Criteria--Leak-Before-Break Analysis for
Primary Water Stress Corrosion
Cracking Near the Tubesheet
(Framatome Data)"
S404-28 "Mercury Modeling for PWSCC Length TR-101104
Sizing" August 1992
S404-29 "Outside Diameter Stress Corrosion NP-7480-L
Cracking (ODSCC) of Steam Generator August 1992
Tubing at Tube Support Plates--A
Database for Alternate Repair Limits;
Volume 1: 7/8-Inch Outside Diameter
(OD) tubing"
S404-12,S404- "PWR Steam Generator Tube-Plugging NP-6864-L
15,S404-18,S4 Limits: Technical Support Document for December 1990
04-19,S404-21, Expansion Zone PWSCC in Roll
S404-24,S404- Transitions"
90,S404-92,S4
04-93
S404-12,S404- "PWR Steam Generator Tube Repair NP-6864-L
15,S404-18,S4 Limits: Technical Support Document for December 1991
04-21,S404-24, Expansion Zone PWSCC in Roll
S404-32,S404- Transitions, Rev. 1"
37,S404-90,S4
04-92,S404-93
Report No
S404-15,S404- "PWR Steam Generator Tube Repair TR-100407
19,$404-21,$4 Limits: Technical Support Document for March 1992
04-24,S404-29, Outside Diameter Stress Corrosion
$404-30,S404- Cracking at Tube Support Plates"
31 ,S404-32,$4
04-33,S404-36,
$404-37,$404-
70,S404-71 ,4
04-72
S405-2 "PWR Secondary Water Chemistry NP-6239
Guidelines, Revision 2" December 1988
S405-9 "Statistical Analysis of Steam Generator NP-7493
Tube Degradation" September 1991
S406-1 "Measurement of Surface-Induced NP-6705-M
Microplasticity in Alloy 600 C-Rings" March 1990
S406-3 "Proceedings: Primary Water Stress NP-6719-SD
Corrosion Cracking" April 1990
S406-3 "Proceedings: Primary Water Stress NP-6719-M
Corrosion Cracking, 1989 EPRI April 1990
Remedial Measures Workshop"
S406-3 "Proceedings: 1990 EPRI Workshop on NP-7198
Circumferential Cracking of Steam March 1991
Generator Tubes" (NP-7198-M, Vol. 1;
NP-7198-S, Vol. 2)"
S406-7 "Evaluation of Leak and Burst NP-7474
Characteristics of Roll Transitions May 1993
Containing Primary Water Stress
Corrosion Cracks"
S406-11 "Residual Stresses in Roller-Expanded TR-1 02355
Steam Generator Tube Transitions" May 1993
S406-12 "Characterization of the Resistance to TR-100865
PWSCC of Hydraulic Tube-Tubesheet July 1992
Expansions"
S407-1 "Determination of Thermodynamic Data NP-5708
for Modeling Corrosion; Volume 5: August 1992
NiCI2, CuCI2, HNO3, Ni(N)3)2, NaNO3
Systems"
Report No
for Modeling Corrosion; Volume 4: June 1992
Chloride Ion Interaction With
Magnesium, Calcium, and Hydrogen
Ions at 250-325 degrees C"
for Modeling Corrosion; Volume 3: March 1992
C02-NaOH-H20 System"
for Modeling Corrosion", 2 vols. November 1989
S407-5 "MULTEQ: Equilibrium of an Electrolytic NP-5561 -CCM
Solution With Vapor-Liquid Partitioning February 1988
and Precipitation, Volume 1: User's
Manual"
S407-5 "MULTEQ: Equilibrium of an Electrolytic NP-5561 -CCML
Solution With Vapor-Liquid Partitioning February 1988
and Precipitation", 3 vols.
S407-7 "1987 EPRI Workshop on NP-5971
Secondary-Side Intergranular Corrosion September 1988
Mechanisms: Proceedings", 2 vols.
S407-7 "Proceedings: 1987 EPRI Workshop on NP-5987-M
Mechanisms of Primary Water September 1988
Intergranular Stress Corrosion Cracking"
VolLme 4: Guidelines for Tube Section
Removal and Examination"
S407-7 "Correlation of Secondary-Side IGA/SCC TR-101010
Degradation of Recirculating Steam August 1992
Generator Tubing With the On-Line
Addition.of Boric Acid"
S407-7 "Proceedings: 1991 EPRI Workshop on TR-101103
Secondary-Side Intergranular Corrosion August 1992
Mechanisms"
S407-9 "Destructive Examination of Steam NP-7136-M
Generator 11 Tubing at Calvert Cliffs March 1991
Power Plant Unit 1", 2 vols.
S407-13 "Hideout and Return of Complex NP-7494
Mixtures in Crevices" September 1991
Report No
S407-14 "Effect of Boric Acid on Intergranular TR-1 03117
Corrosion in Tube Support Plate October 1993
Crevices"
S407-15 "Microchemistry of Corroded NP-6709-M
Intergranular Surfaces, Tubes Removed March 1990
From Point Beach Unit 1, Calvert Cliffs
Unit 1, and Saint Lucie Unit 1 Steam
Generators"
S407-16 "Investigation of Lead As a Cause of NP-7367-M
Stress Corrosion Cracking at Support June 1991
Plate Intersections"
S407-17 "Proceedings: 1987 Symposium on NP-6005
Chemistry in High-Temperature Water" March 1990
S407-18 "MULTEO: Equilibrium of an Electrolytic NP-5561 -CCML
Solution With Vapor-Liquid Partitioning April 1992
and Precipitation", Vol. 2: "The
Database", Rev. 3
S407-18 "MULTEQ: Equilibrium of an Electrolytic NP-5561-CCML
Solution With Vapor-Liquid Partitioning December 1990
and Precipitation", Vol. 2: "The
Database", Rev. 2
S407-19 "Examination of Roller Expansion NP-6192-M
Transitions in Point Beach Unit 1 Steam July 1989
Generator Tubing"
S407-21 "Speciation and Chemical Activities in TR-102491
Superheated Sodium Borate Solutions" June 1993
S407-30 "MULTEQ: Equilibrium of an Electrolytic NP-5561 -CCML
Solution with Vapor-Liquid Partitioning; August 1992
Volume 3: Theory Manual"
Solution With Vapor-Liquid Partitioning July 1992
and Precipitation; Volume 1: User's
Manual (Revision 2)"
Solution With Vapor-Liquid Partitioning May 1989
and Precipitation; Volume 1: User's
Manual (Revision 1)"
S407-32 "Examination of Pits in Trojan Nuclear NP-6362-M
Power Plant Steam Generator Tubes" May 1989
Report No
S407-34 "Laboratory Examination of Tubes NP-6998-M
R35C70 and R36C67 Removed From October 1990
the V. C. Summer Nuclear Station", 2
vols.
S407-35 Vol. 1, "Boric Acid Application NP-5558-M
Guidelines for Intergranular Corrosion December 1990
.Inhibition", Rev. 1
S407-35 Vol. 2, "Boric Acid Application NP-5558-SL
Guidelines for Intergranular Corrosion December 1990
Inhibition", Rev. 1
S407-36 Vol. 1, "Acid Sulfate Corrosion in PWR NP-7346-M
Steam Generators" June 1991
S407-36 Vol. 2, "Acid Sulfate Corrosion in PWR NP-7346-SD
Steam Generators" June 1991
S407-36 "Remedial Actions for Acidic Sulfate TR-101105
Corrosion" August 1992
S407-37 "Adsorption of Sulfate in PWR Steam TR-101106
Generators: Laboratory Tests" August 1992
S407-40 "Destructive Examination of Tube NP-7371 -M
R31C66 From the Ginna Nuclear Plant June 1991
Steam Generator"
S407-41 "Reference Electrodes for TR-103311
Electrochemical Corrosion Potential November 1993
Monitoring in PWR Secondary Systems"
S407-42 "Examination of Kewaunee Steam NP-7370-M
Generator Tubes R4C81 and R11C9" June 1991
S407-43 "Proceedings: 1991 Symposium on TR-102706
Chemistry in High-Temperature August 1993
Aqueous Solutidns"
S408-1 "Microstructure Etching and Carbon NP-6720-M
Analysis Techniques" April 1990
Volume 2: Guidelines for Procurement
of Alloy 690 Steam Generator Tubing"
S408-1 "Proceedings: 1989 EPRI Alloy 690 NP-6750-M
Workshop" April 1990
Report No
S408-1 "Proceedings: 1989 EPRI Alloy 690 NP-6750-SD
Workshop" April 1990
S408-2 "Effect of Different Thermal Treatments NP-6703-M
on the Corrosion Resistance of Alloy March 1990
690 Tubing"
S408-3 "Tensile Deformation and Recovery TR-100910
Kinetics of Alloy 690" July 1992
Volume 3: Steam Generator Tube
Sleeving: Design, Specification, and
Procurement Checklist"
S408-6 "Alloy 690 for Steam Generator Tubing NP-6997-M
Applications", 2 vols. October 1990
S409-1 "Advanced Studies in Chemistry Control TR-100758
With Morpholine" July 1992
S409-2 "PWR Advanced All-Volatile Treatment TR-100755
Additives, By-Products, and Boric Acid" July 1992
S409-3 "Demineralizer Operation With TR-100790
Morpholine and Boric Acid" July 1992
S409-3 "Effects of Morpholine and Boric Acid TR-1 00791
Implementation on Secondary July 1992
Chemistry and Corrosion Product
Transport"
S409-4 "Compatibility of PWR Gasket and TR-100794
Packing Materials and Resins With July 1992
Organic Amines"
S409-9 "Qualification of Morpholine for TR-100795,
Secondary System pH Control in July 1992
Plants", Tiers 1&2, 2 vols.
S409-11 "Loop Testing of Alternative Amines for TR-100756
All-Volatile Treatment Control in PWRs" June 1992
S410-12 "Large Eddy Simulation on NP-7041
Supercomputers" November 1990
S411-1 "A Computer Program for NP-6891-CCML
Three-Dimensional, Two-Fluid Model July 1990
Thermal-Hydraulic Simulation of Steam
Generators: PORTHOS Mod-01", 4 vols.
Report No
S41 1-6,RP1066 "ATHOS3 Mod-01: A Computer NP-4604-CCML
-1 Program for Thermal-Hydraulic Analysis September 1990
of Steam Generators", Rev. 1, 2 vols.
S413-1 "Ginna Station Steam Generator TR-100866
U-Bend Tube Analysis for Chemical July 1992
Cleaning Data"
S413-2 "Examination of Trojan Steam TR-101427
Generator Tubes; Volume 1: November 1992
Examination Results"
S413-2 "Examination of Trojan Steam TR-101427
Generator Tubes; Volume 2: November 1992
Appendixes"
S413-4 "Examination of Trojan Steam TR-1 01427
Generator Tubes; Volume 3: Rockwell, November 1992
Auger, and XPS Analyses"
S999 "Workshop Proceedings: NP-3207-SR
Replacement/Repair of Steam August 1983
Generators"
S999 "Workshop Proceedings: Pitting in NP-3574-SR
Steam Generator Tubing" October 1984
S999 "Proceedings: The Second EPRI NP-3924-SR
Workshop on Support-Structure March 1985
Corrosion in Nuclear Plant Steam
Generators"
S999 "Proceedings: Workshop on Thermally NP-4665-M
Treated Alloy 690 Tubes for Nuclear July 1986
Steam Generators"
S999 "Proceedings: Workshop on Thermally NP-4665-SR
Treated Alloy 690 Tubes for Nuclear July 1986
Steam Generators"
S999 "PWR Secondary Water Chemistry NP-5056-SR
Guidelines, Rev. 1" March 1987
S999 "Guidelines for Permanent BWR NP-5283
Hydrogen Water Chemistry Installations September 1987
- 1987 Revision"
S999 "Proceedings: Workshop on the Role of NP-6710-SD
Sulfur Species on the Secondary-Side March 1990
Degradation of Alloy 600 and Related
Alloys"
Report No
S999 "Proceedings: Workshop on the Role of NP-6710-M
Sulfur Species on the Secondary-Side March 1990
Degradation of Alloy 600 and Related
Alloys"
TPS79-729 "Assessment of Condenser Leakage NP-1467
Problems" August 1980
TPS79-730 "Corrosion-Related Failures in Power NP-1468
Plant Condensers" August 1980
TPS81-777 "PWR Steam Generator Cost-Benefit NP-2295
Methodology (Denting)" March 1982
Appendix B
PLANT DESIGN
CHARACTERISTICS
B-2 Steam Generator Reference Book, Revision I
Reactor Name Almaraz 1 Almaraz 2
Operating Utility Name CNA CNA

Country -Spain Spain
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W W
Initial Critical Date or Date of Major Change 4/15/81 9/15/83
Net Power 900 900
Gross Mega Watt Elec. Rated Power MWe 930 930
Cold Leg Prim. Cool. Temp. 100% Power IF 553 553
Hot Leg Prim. Cool. Temp. 100% Power IF 616 616
Changedto: ....
Changedto: -
Steam Gener. Oper. Temp. IF 536 536
Number of Steam Gener. 3 3
Steam Gener. Model Number D3 D3
Phosphates Added? No No
On-Line Boric Acid Added? No-->Yes-->No No-->Yes-->No
Condensate Purification Method Deep Bed (Startup Only) Deep Bed (Startup Only)
Steam Generator Tubing Material 600MA 600MA
Steam Generator Tube Sheet Material Forged LA Steel Forged LA Steel
Support Plate Material Carbon Steel Carbon Steel
Tube Support Configuration Drilled Hole Drilled Hole
Tube Expansion Method Full Depth Roll Full Depth Roll
Main Condenser Tube Sheet Material (Orig.) CuNi 90/10 CuNi 90/10
0
Changedto: . -_
Main Condenser Tubing Material (Orig.) Admiralty Brass Admiralty Brass

Changedto: ....
Changedto: ..
Main Condenser Deaerating (Y/NI Yes Yes
Deaerator Other Than Condenser (Y/N) No No
LP FWH Tubing Material (Original) AdBr & 90/10 AdBr & 90/10
Changedto: 304L SS 304L SS
HP FWH Tubing Material (Original) 304SS 304 SS
Changedto: ...
Reheater Tubing Material (MSR)(Original) CuNi 90/10 CuNi 90/10
Changedto: Ferritic SS Ferritic SS
Type of Cooling Water Pond (Closed) Pond (Closed)
Appendix B-Plant Design Characteristics B-3
Reactor Name Angra 1 Arkansas 1
Operating Utility Name Furnas AP&L

Country Brazil U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W B&W
Net Power 626 836
Cold Leg Prim. Cool. Temp. 100% Power OF 549 555
Hot Leg Prim. Cool. Temp. 100% Power 'F 616 603
Changed to: ......
Changed to: ......
Steam Gener. Oper. Temp. OF 535 590
Steam Gener. Model Number D3 OTSG 177
On-Line Boric Acid Added? Yes No
Condensate Purification Method None Deep Bed
Steam Generator Tubing Material 600MA 600SR (See Note 2b)
Tube Support Configuration Drilled Hole broached Hole
Tube Expansion Method Full Depth Roll Part Depth Roll
Main Condenser Tube Sheet Material (Orig.) Carbon Steel Muntz
Changedto: ......
Main Condenser Tubing Material (Orig.) Alum. Brass Admiralty Brass
Changedto: Titanium ---
Changedto: --
LP FWH Tubing Material (Original) Admiralty Brass 304 SS
Changedto: ---...
HP FWH Tubing Material (Original) CuNi 304 SS

Changedto: ---...
Reheater Tubing Material (MSR)(Original) CuNi CuNi 90/10 + 439SS

Changedto: Ferritic SS ---
Type of Cooling Water Seawater Fresh (Reservoir)
B-4 Steam Generator Reference Book, Revision 1
Reactor Name Arkansas 2 Asco 1
Operating Utility Name AP&L ANA

Country U.S. Spain
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. CE W
Net Power 858 900
Changed to: 599 ---
Changed to: ---
Steam Gener. Oper. Temp. IF 532 543.4
Steam Gener. Model Number 2815 D3
On-Line Boric Acid Added? Yes Yes
Condensate Purification Method Partial / Deep Bed None
Steam Generator Tubing Material 600 600MA
Steam Generator Tube Sheet Material LA Steel Forged LA Steel
Tube Support Configuration Dr. Hole/Eggcrate Drilled Hole
Tube Expansion Method Full / Explosive Full Depth Roll
Main Condenser Tube Sheet Material (Orig.) Muntz CS + Titanium Cladding
Changedto:
Main Condenser Tubing Material (Orig.) CuNi 90/10 Admiralty Brass
Changedto: -- Titanium
Changedto: .
LP FWH Tubing Material (Original) Copper Alloy Admiralty Brass
Changedto: Stainless Steel 304L SS
HP FWH Tubing Material (Original) Copper Alloy CuNi
Changedto: Stainless Steel 304L SS
Changedto: -- Ferritic SS
Type of Cooling Water Closed (Natural Tower) Fresh (River) + Cooling Towers
Reactor Name Asco 2 Atucha 1
Operating Utility Name ANA CNEA
Country Spain Argentina

PWR or PHWR PWR PHWR
Nuclear Steam System Supply Manuf. W KWU

Net Power 900 335
0
Cold Leg Prim. Cool. Temp. 100% Power F 556 521
Hot Leg Prim. Cool. Temp. 100% Power OF 620 582

Changed to: -
Changed to: - -
Steam Gener. Oper. Temp. IF 543.4 492

Steam Gener. Model Number D3

Phosphates Added? No Yes
Condensate Purification Method - None None

Steam Generator Tubing Material 600MA 800M
Support Plate Material Carbon Steel Austenitic SS
Tube Support Configuration Drilled Hole Grids/Strip (See Note 3)
Tube Expansion Method Full Depth Roll Mech. Roll-Top & Bottom
Main Condenser Tube Sheet Material (Orig.) CS + Titanium Cladding C-Steel / Bitumen-coating
Changed to: --
Main Condenser Tubing Material (Orig.) Admiralty Brass Ad. Brass (See Note 9)
Changed to: Titanium --
Changed to: --- --
Deaerator Other Than Condenser (Y/N) No Yes
LP FWH Tubing Material (Original) Admiralty Brass Carbon Steel

Changed to: 304L SS
HP FWH Tubing Material (Original) CuNi Carbon Steel
Changed to: 304L SS

Reheater Tubing Material (MSR)(Original) CuNi 90/10 Carbon Steel
Changed to: Ferritic SS --
Type of Cooling Water Fresh (River) + Cooling Towers River

Reactor Name Beaver Valley 1 Beaver Valley 2
Operating Utility Name Duquesne Light Duquesne Light

Country U.S. U.S.
PWR or PHWR PWR PWR
Net Power 835 835
Cold Leg Prim. Cool. Temp. 100% Power'F 542 542
Hot Leg Prim. Cool. Temp. 100% Power *F 607 607
Changed to: - -
Changed to: -
Steam Gener. Oper. Temp. °F 517 517
Steam Gener. Model Number 51 51M
Condensate Purification Method None Full / Powdex
Main Condenser Tube Sheet Material (Orig.) Stainless Steel Stainless Steel
Changed to: ....
Main Condenser Tubing Material (Orig.) 304 SS 304 SS
Changed to: Seacure --
Changed to: ..
LP FWH Tubing Material (Original) 304L SS Stainless Steel
Changed to:
HP FWH Tubing Material (Original) 304L SS Stainless Steel
Changed to:
Changed to: 439 SS Stainless Steel
Type of Cooling Water Fresh Nat. Tower Fresh Nat. Tower
Reactor Name Belleville 1 Belleville 2]
Operating Utility Name EdF EdF

Country France France
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. FRAM FRAM
Net Power 1310 1310
Changedto: 616 616
Changed to: ......
Steam Gener. Model Number 68/19 68/19
On-Line Boric Acid Added? No No
Condensate Purification Method None None
Steam Generator Tubing Material 600TT 600TT
Support Plate Material 13% Cr. Steel 13% Cr. Steel
Tube Support Configuration Broached Hole Broached Hole
Tube Expansion Method Full Roll + Kiss Full Roll + Kiss
Main Condenser Tube Sheet Material (Orig.) Carbon Steel Carbon Steel
Changed to: - -
Main Condenser Tubing Material (Orig.) Cu + 439 Cu + 439
Changed to: - --
Changed to: - -
Deaerator Other Than Condenser (Y/N) Yes Yes
LP FWH Tubing Material (Original) 304L SS 304L SS
Changed to: -- -
HP FWH Tubing Material (Original) Carbon Steel Carbon Steel
Changed to: - -
Reheater Tubing Material (MSR)(Original) 18% Cr SS (439L) 18% Cr SS (439L)

Changed to: ....
Type of Cooling Water Closed (Tower) Closed (Tower)
Reactor Name Beznau 1 Beznau 2
Operating Utility Name

Country
PWR or PHWR
NOK
Switzerland
PWR
NOK
Switzerland
PWR
0
Net Power 350 350
Changed to: ---
Changed to: --
Steam Gener. Model Number 33 33
Phosphates Added? No-->Yes-->No Yes-->No
Tube Expansion Method Part Depth Roll Part Depth Roll
Changed to:
Main Condenser Tubing Material (Orig.) Admr. + CuNi 90/10 Admr. + CuNi 90/10
Changedto: Partial SS + Tit. --
Changed to:
"Deaerator Other Than Condenser (Y/N) Yes Yes
LP FWH Tubing Material (Original) Adm. + SS Adm. + SS
Changed to: Stainless Steel
Changed to:
Changed to: _- ._
Type of Cooling Water Freshwater (River) Freshwater (River)
Reactor Name Biblis A Biblis B
Operating Utility Name RWE RWE

Country Germany Germany
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manut. KWU KWU
Net Power 1146 1240
Cold Leg Prim. Cool. Temp. 100% Power 'F 543 551
Changed to: ....
Changedto: ---
Steam Gener. Oper. Temp. *F 513 518

Steam Gener. Model Number
Phosphates Added? Yes-->No Yes-->No
Steam Generator Tubing Material 800M 800M
Support Plate Material Austenitic SS Austenitic SS
Tube Support Configuration Grids/Strip (See Note 3) Grids/Strip (See Note 3)
Tube Expansion Method Mech. Roll-Top & Bottom Mech. Roll-Top & Bottom
Main Condenser Tube Sheet Material (Orig.) C-steel / Epoxy Coating C-steel / Epoxy Coating
Changed to: --

Changed to: SS Stainless Steel
Changed to: ....
LP FWH Tubing Material (Original) Carbon Steel Carbon Steel
Changedto: ..
Changed to: .....
Reheater Tubing Material (MSR)(Original) Carbon Steel' Carbon Steel

Changedto: ....
Type of Cooling Water Fresh (River) Mech. Tower Fresh (River) Mech. Tower
Reactor Name Z Blayais 1
Operating Utility Name EdF

Country France
PWR or PHWR PWR
Nuclear Steam System Supply Manuf. FRAM
Initial Critical Date or Date of Major Change 5/20/81
Net Power 910
Gross Mega Watt Elec. Rated Power MWe 951
Cold Leg Prim. Cool. Temp. 100% Power IF 546
Hot Leg Prim. Cool. Temp. 100% Power IF 613
Changed to:
Changed to:
Steam Gener. Oper. Temp. OF 523
Number of Steam Gener. 3
Steam Gener. Model Number 51M
Phosphates Added? No
On-Line Boric Acid Added? No
Condensate Purification Method None
Steam Generator Tubing Material 600MA
Steam Generator Tube Sheet Material Forged LA Steel
Support Plate Material Carbon Steel
Tube Support Configuration Drilled Hole
Tube Expansion Method Full Depth Roll + Kiss
Main Condenser Tube Sheet Material (Orig.) Double (Carb. StI. on 2ndary Side)
Changedto:
Main Condenser Tubing Material (Orig.) Titanium
Changedto:
Changedto: --
Main Condenser Deaerating (Y/NI Yes

Deaerator Other Than Condenser (Y/N) No
LP FWH Tubing Material (Original) 304L SS
Changed to:
HP FWH Tubing Material (Original) Carbon Steel
Changedto: --
Reheater Tubing Material (MSR)(Original) Carbon Steel

Changed to: ---
Type of Cooling Water Brackish
Reactor Name Blayais 2
Country France
PWR or PHWR PWR

Net Power 910

Hot Leg Prim. Cool. Temp. 100% Power *F 613
Changedto:
Changedto: --
Steam Gener. Oper. Temp. IF 523

Steam Gener. Model Number 511B1

Support Plate Material Cbn.Stl.+13%Cr.Stl.
Tube Support Configuration Broached Hole

Changedto:

Changedto: ---
Changedto: --

Changedto: --

Changedto:
Changedto: --

Reactor Name Blayais 3

Country France
PWR or PHWR PWR
Net Power 910
Cold Leg Prim. Cool. Temp. 100% Power OF 546
Hot Leg Prim. Cool. Temp. 100% Power OF 613
Changed to:
Changed to:
Steam Gener. Model Number 51B1
On-Line Boric Acid Added? i No
Support Plate Material Cbn.Stl.+13%Cr.Stl.
Main Condenser Tube Sheet Material (Orig.) Double (Carb. Stl. on 2ndary Side)
Changed to: --

Changed to: --
Changed to: --

Changed to:
Changedto:
Changed to:
Reactor Name Blayais 4 (SG 1 &2) Blayais 4 (SG 3)

PWR or PHWR PWR PWR
Net Power 910 910
Cha ng e d to : ....
Changed to: ---

Steam Gener. Model Number 51B 51B
Steam Generator Tubing Material 600MA 600TT
Support Plate Material 13% Cr. StI. 13% Cr. Stf.
Tube Expansion Method Full Depth Roll + Kiss Full Depth Roll + Kiss
Main Condenser Tube Sheet Material (Orig.)
Changed to:
Main Condenser Tubing Material (Orig.) Titanium Titanium
Changed to: ..
Changed to: ......
Deaerator Other Than Condenser (Y/N). No No
Changed to: ---..

Changedto: ..
Reheater Tubing Material (MSR)(Original) Carbon Steel Carbon Steel
Changedto: ......
Type of Cooling Water Brackish Brackish
B-1 4 , Steam Generator Reference Book, Revision 1
Reactor Name Borsele Braidwood 1
Operating Utility Name EPZ Commonwealth Edison

Country Netherlands U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. KWU/RDM W
Net Power 452 1120
Hot Leg Prim. Cool. Temp. 100% Power F 606 608
Changed to: ..
Changed to: --
Steam Gener. Oper. Temp. F 527
Steam Gener. Model Number D4
Phosphates Added? Yes-->No No
Condensate Purification Method None Partial
Steam Generator Tubing Material 800M 600MA
Support Plate Material Austenitic SS Carbon Steel
Tube Support Configuration Grids/Strip (See Note 3) Drilled Hole
Tube Expansion Method Mech. Roll-Top & Bottom Full Depth Roll
Main Condenser Tube Sheet Material (Orig.) Muntz (See Note 7) Stainless Steel
Changed to: CS / Ti-clad ---
Main Condenser Tubing Material (Orig.) CuNi 70/30 Stainless Steel
Changed to: Titanium ---
Changed to: ..
Deaerator Other Than Condenser (Y/N) Yes No
LP FWH Tubing Material (Original) Carbon Steel Stainless Steel
Changed to: ---
HP FWH Tubing Material (Original) Carbon Steel Stainless Steel

Changedto: ---...
Reheater Tubing Material (MSR)(Original) Carbon Steel Stainless Steel

Changed to: ..
Type of Cooling Water Brackish Fresh (Reservoir)
Reactor Name Braidwood 2 Brokdorf
Operating Utility Name Commonwealth Edison KBR

Country U.S. Germany
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W KWU
Net Power 1120 1326
Changed to:
Changed to:
Steam Gener. Model Number D5 54GS
Condensate Purification Method Partial Deep Bed (See Note 4)
Steam Generator Tubing Material 600TT 800M
Support Plate Material 405 SS Austenitic SS
Tube Support Contiguration Broached Hole Grids/Strip (See Note 3)
Tube Expansion Method Full / Hydraulic Mech. Roll-Top & Bottom
Main Condenser Tube Sheet Material (Orig.) Stainless Steel C-Steel / Ti-clad
Changedto:
Main Condenser Tubing Material (Orig.) Stainless Steel Titanium
Changed to:
Changed to:
LP FWH Tubing Material (Original) Stainless Steel Carbon Steel
Changed to:
HP FWH Tubing Material (Original) Stainless Steel Carbon Steel
Changed to:
Reheater Tubing Material (MSR)(Original) Stainless Steel Carbon Steel
Changed to:
Type ot Cooling Water Closed (Pond) Closed (Tower)
Reactor Name ruce I

Operating Utility Name Ontario Hydro
Country Canada
PWR or PHWR PHWR
Nuclear Steam System Supply Manuf. AECL
Net Power 769
Cold Leg Prim. Cool. Temp. 100% Power 'F 509
Hot Leg Prim. Cool. -emp. 100% Power OF 579
Chan( ?d to: --
Changedto: --
Steam Gener. Oper. Temp. 'F 493

Steam Gener. Model Number B&W
Steam Generator Tubing Material 600 (SR)
Steam Generator Tube Sheet Material Forged Steel
Tube Expansion Method Hard Roll near TS Sec. Face
Main Condenser Tube Sheet Material (Orig.) Muntz
Changedto:
Main Condenser Tubing Material (Orig.) Admiralty Brass/Periph.; 304SS
Changedto:
Changedto:
Deaerator Other Than Condenser (Y/N) Yes
LP FWH Tubing Material (Original) Admiralty Brass
Changedto:
HP FWH Tubing Material (Original) CuNi 90/10
Changedto:
Changed to:
Type of Cooling Water Lake (Huron)
Appendix B-Plant Design Characteristics B- 17
Reactor Name Bruce 2

Country Canada
PWR or PHWR PHWR
Nuclear Steam System Supply Manuf. . AECL
Net Power 769
Changed to:
Changed to: ---
Changed to: ---
Changedto:
Changedto:
Changedto:
Changedto:
Changedto:

Country Canada
PWR or PHWR PHWR

Net Power 769

Hot Leg Prim. Cool. Temp. 100% Power.°F 579
Changedto: ---
Changedto: --





Changedto:

Changedto:
Changedto:

Changedto:
Changedto:
Changedto:

Country Canada
PWR or PHWR PHWR

Net Power 769

Changed to: --
Changed to:




Changed to: ---

Changed to: ---
Changed to: --

Changed to: -
Changed to:
Reheater Tubing Material (MSR)(Orginal) Carbon Steel
Changed to: --

Reactor Name Bruce 5. Bruce 6
Operating Utility Name Ontario Hydro Ontario Hydro

Country Canada Canada
PWR or PHWR PHWR PHWR
Nuclear Steam System Supply Manuf. AECL AECL
Net Power 860 837
Changed to:
Changed to:
* Number of Steam Gener. 8 8
Steam Gener. Model Number B&W B&W
Steam Generator Tubing Material 600 (SR) 600 (SR)
Steam Generator Tube Sheet Material Forged Steel Forged Steel
Tube Expansion Method Hard Roll near TS Sec. Face Hard Roll near TS Sec. Face
Changed to:
Main Condenser Tubing Material (Orig.) 304SS 304SS
Changed to:
Changed to:
LP FWH Tubing Material (Original) 304SS 304SS
Changedto:
HP FWH Tubing Material (Original) 304SS 304SS
Changedto:
Reheater Tubing Material (MSR)(Original) Corten Corten
Changedto: -- t
Type of Cooling Water Lake (Huron) Lake (Huron)
Reactor Name Bruce 7 Bruce 8

Net Power 860 837
Cold Leg Prim. Cool. Temp. 100% Power °F 509 509
Changedto:
Changedto:
Steam Generator Tubing Material 600 (SR) 600 (SR)
Tube Support Contiguration Broached Hole Broached Hole
Tube Expansion Method Hard Roll near TS Sec. Face Hard Roll near TS Sec. Face
Changedto:
Main Condenser Tubing Material (Orig.) 304SS 304SS
Changed to:
Changed to:
LP FWH Tubing Material (Original) 304 SS 304 SS
Changedto:
HP FWH Tubing Material (Original) 304 SS 304 SS
Changed to:
Changedto:
Type of Cooling Water Lake (Huron) Lake (Huron)
Reactor Name Bugey 2 Bugey 3

PWR or PHWR PWR PWR
Net Power 920 920
Changed to: ......
Changed to: ---
Steam Gener. Model Number 51A 51A
Steam Generator Tubing Material
Steam Generator Tube Sheet Material
Support Plate Material
600MA
Forged LA Steel
Carbon Steel
600MA
Forged LA Steel
Carbon Steel
S
Changed to: ....

Changed to: ..
Changed to: ..
LP FWH Tubing Material (Original) Carbon Steel + 304L SS Carbon Steel + 304L SS
Changed to: ..
Changed to:
Changed to: ..
Type of Cooling Water Fresh (River) Fresh (River)
Reactor Name Bugey 4 Bugey 5

PWR or PHWR PWR PWR
Net Power 900 900
Hot Leg Prim. Cool. Temp. 100% Power °F 613 613
Changed to: -- -
Changed to: - --
Tube Expansion Method Full Depth Roll Fult Depth Roll + Kiss
Changed to: - -
Main Condenser Tubing Material (Orig.) Brass Brass
Changed to: - -
Changed to: -- -
LP FWH Tubing Material (Original) Carbon Steel + 304L SS Carbon Steel + 304L SS
Changed to: -- -
HP FWH Tubing Material (Original) Carbon Steel Carbon steel
Changed to: ....
Reheater Tubing Material (MSR)(Original) Carbon Steel Carbon steel
Changed to: ....
Type of Cooling Water Closed (Nat. Tower) / River Closed (Nat. Tower) / River
Reactor Name Byron 1 Byron 2
Operating Utility Name Commonwealth Edison Commonwealth Edison

Country U.S. U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manut. W W
Net Power 1120 1120
Changedto:
Changed to: - -
Steam Gener. Oper. Temp. OF



Condensate Purification Method None Partial
Support Plate Material Carbon Steel 405 SS

Tube Support Configuration Drilled Hole Broached Hole
Tube Expansion Method Full Depth Roll Full / Hydraulic

Changedto:
Main Condenser Tubing Material (Orig.) Stainless Steel Stainless Steel
Changedto:
Changedto:

LP FWH Tubing Material (Original) Stainless Steel Stainless Steel
Changedto: ....
HP FWH Tubing Material (Original) Stainless Steel Stainless Steel
Changedto: .....
Reheater Tubing Material (MSR)(Original) Stainless Steel Stainless Steel
Changedto:
Type of Cooling Water Closed (Mech. Tower) Closed (Mech. Tower)
Reactor Name Callaway 1 Calvert Cliffs 1
Operating Utility Name Union Electric BG&E

Country U.S. U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W CE
Net Power 1150 865-->918 (9/78)
Gross Mega Want Elec. Rated Power MWe 1219 900
Changed to: -- -
Changed to: ..
Steam Gener. Model Number F 67
Condensate Purification Method Full / Deep Bed Filter Demin.+Deep Bed
Steam Generator Tubing Material 600MA & TT 600MA
Support Plate Material 405 SS Carbon Steel
Tube Support Configuration Broached Hole Dr. Hole/Eggcrate
Tube Expansion Method Full / Hydraulic Full / Explosive
Main Condenser Tube Sheet Material (Orig.) Carbon Steel ASTM A285 Aluminum Bronze
Changed to: ....
Main Condenser Tubing Material (Orig.) CuNi 90/10 CuNi 70/30
Changed to: - --
Changed to: ....

LP FWH Tubing Material (Original) 304 SS Stainless Steel
Changed to: - --
HP FWH Tubing Material (Original) 304 SS Stainless Steel

Changed to: .
Reheater Tubing Material (MSR)(Original) Corten Steel CuNi 90/10
Changed to: - -
Type of Cooling Water Closed (Nat. Tower) Brackish

Reactor Name Calvert Cliffs 2 Catawba 1
Operating Utility Name BG&E Duke Power Co.

Country U.S. U.S.
PWR or PHWR PWR PWR
Initial Critical Date or Date of Major Change 11/1576 1/15/85
Net Power 865-->911 1129
Changedto: -
Changedto: --

Steam Gener. Model Number 67 D3
On-Line Boric Acid Added? No Yes
Condensate Purification Method Filter Demin.+Deep Bed Full / Powdex
Changedto: ....
Main Condenser Tubing Material (Orig.) CuNi 70/30 304SS
Changedto:
Changedto:
Changed to:
Changed to: -
Reheater Tubing Material (MSR)(Original) CuNi 90/10 + SS CuNi 90/10
Changed to:
Type of Cooling Water Brackish Closed (Mech. Tower)
Reactor Name Catawba 2 Cattenom 1
Operating Utility Name Duke Power Co. EdF

Country U.S. France
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W FRAM
Net Power 1129 1300
Changedto: -- 616
Changedto: --
Steam Gener. Model Number D5 68/19
Condensate Purification Method Full / Powdex None
Support Plate Material 405SS 13% Cr. Steel
Tube Expansion Method Full / Hydraulic Full Depth Roll + Kiss
Main Condenser Tube Sheet Material (Orig.) Carbon Steel Cu-Al
Changedto: ---..
Main Condenser Tubing Material (Orig.) Stainless Steel Cu + Ti or 29.4

Changedto: ..
Changedto: ..
LP FWH Tubing Material (Original) Stainless Steel 304L SS
Changedto: ---..

Changed to: .....
Reheater Tubing Material (MSR)(Original) CuNi 90/10 18% Cr SS (439L)

Changed to: ....
Type of Cooling Water Closed (Mech. Tower) Closed (Tower)
Reactor Name Cattenom 2 Cattenom 3

PWR or PHWR PWR PWR
Net Power 1300 1300
Changedto: 616 --
Changedto: ......
Main Condenser Tube Sheet Material (Orig.) Cu-Al Cu-Al
Changed to:
Main Condenser Tubing Material (Orig.) Cu + Ti or 29.4 Cu + Ti or 29.4
Changed to: .....
Changedto: ..
Changed to: ....

Changed to: ---...

Changed to: ....
Reactor Name Cattenom 4 Chinon B1

PWR or PHWR PWR PWR
Net Power 1300 870
Changed to: ..
Changed to: ..
Steam Gener. Model Number 68/19 511B1
Steam Generator Tubing Material 600TT 600MA
Support Plate Material 13% Cr. Steel Carbon Steel
Main Condenser Tube Sheet Material (Orig.) Cu-Al Carbon Steel
Changedto: - --
Main Condenser Tubing Material (Orig.) Cu + Ti or 29.4 Brass

Changedto: .
Changedto: ......
LP FWH Tubing Material (Original) 304L SS Brass + 304L
Changedto: ---
HP FWH Tubing Material (Original) 18% Cr SS (439L) Carbon Steel

Changedto: - -
Reheater Tubing Material (MSR)(Original) 18% Cr SS (439L) 304L

Changedto: ..
Type of Cooling Water Closed (Tower) Fresh/Cooling Tower/River
Reactor Name Chinon B2 Chinon B3

PWR or PHWR PWR PWR
Net Power 870 920
Changed to: ....
Changed to: ---..

Changed to: --

Changedto:
Changedto: ......
LP FWH Tubing Material (Original) Brass + 304L Brass + 304L
Changed to: ....
Changedto:
Reheater Tubing Material (MSR)(Original) 304L 18% Cr SS (439L)
Changed to: ....
Type of Cooling Water Tower / River Tower
Reactor Name Chinon B4 Chooz A (end 10/91)
Operating Utility Name EdF SENA

PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. FRAM W
Net Power 920 305

Changedto: - -
Changedto: - -


Steam Generator Tubing Material 600TT 316 SS

Support Plate Material 13% Cr. Steel
Tube Support Configuration Broached Hole Drilled Hole

Tube Expansion Method Full Depth Roll + Kiss Full Depth Roll
Main Condenser Tube Sheet Material (Orig.) Carbon Steel
Changed to: --
Main Condenser Tubing Material (Orig.) Brass
Changed to: --
Changed to: --

LP FWH Tubing Material (Original) Brass + 304L
Changed to:
Changed to:
Reheater Tubing Material (MSR)(Original) 18% Cr SS (439L)
Changed to:
Type of Cooling Water Tower

Reactor Name I Comanche Peak1 1 Comanche Peak2
Operating Utility Name Texas Utilities Electric [Texas Utilities Electric

Country U.S. US
PWR or PHWR PWR PWR
Net Power 1110 1110
Changed to: --
Changedto: - -
.On-Line Boric Acid Added? No No
Condensate Purification Method Full / Powdex Full / Powdex
Tube Expansion Method Full Depth Roll Full/Hydraulic
Main Condenser Tube Sheet Material (Orig.) CS Clad Titanium CS Clad Titanium
Changed to: ....

Changed to: ....
Changed to: -- -

Changed to:
Changed to:
Changed to:
Type of Cooling Water Fresh (Reservoir) Fresh (Reservoir)
Reactor Name Cook 2 (ORIG) Cook 2 (REPL)
Operating Utility Name Indiana/Michigan Indiana/Michigan Indiana/Michigan

Country U.S. U.S. U.S.
PWR or PHWR PWR PWR PWR
Nuclear Steam System Supply Manuf. W W W
Initial Critical Date or Date of Major Change 1/15/75 3/15/78 Replaced: 3/89
Net Power 1020 1060
Changed to: ..
Changed to: -- -

Steam Gener. Model Number 51 51 54F
Phosphates Added? No No No
On-Line Boric Acid Added? Yes Yes No
Condensate Purification Method None None None
Steam Generator Tubing Material 600MA 600MA 690TT
Steam Generator Tube Sheet Material Forged LA Steel Forged LA Steel Forged LA Steel
Support Plate Material Carbon Steel Carbon Steel 405 SS
Tube Support Configuration Drilled Hole Drilled Hole Broached Hole
Tube Expansion Method Part Depth Roll Part Depth Roll Full / Hydraulic
Main Condenser Tube Sheet Material (Orig.) SiBmz Carbon Steel Carbon Steel
Changed to: 304 SS 304 SS
Main Condenser Tubing Material (Orig.) AsCu AsCu 304 SS
Changed to: 304 SS 304 SS --
Changed to: .....
Main Condenser Deaerating (Y/NI Yes Yes Yes
Deaerator Other Than Condenser (Y/N) No No No
LP FWH Tubing Material (Original) 304 SS 304 SS 304 SS
Changed to: - -
HP FWH Tubing Material (Original) Stainless Steel Stainless Steel Stainless Steel
Changed to: .......
Reheater Tubing Material (MSR)(Original) AsCu CuNi 90/10 Carbon Steel
Changed to: CS Carbon Steel
.Type of Cooling Water Fresh (Lake) Fresh (Lake) Fresh (Lake)
Reactor Name Cruas 1 Cruas 2

PWR or PHWR PWR PWR
Net Power 880 915

Changed to: ...
Changed to: - -



Changedto:
Changedto: --
Changedto: ....
LP FWH Tubing Material (Original) Brass + 304L SS Brass + 304L SS

Changedto:
Changedto: ....
Reheater Tubing Material (MSR)(Original) 304L 304L
Changedto: -
Type of Cooling Water Closed (Nat. Tower) Closed (Nat. Tower)
Reactor Name Cruas 3 Cruas 4

PWR or PHWR PWR PWR
Net Power 915 880
Changed to: - --
Changed to: ---

Changed to: ....
Changed to: -- -
Changed to: ..
Changed to: - --

Changed to: -- -
Reheater Tubing Material (MSR)(Original) 18% Cr SS (439L) 304L

Changed to: -- -
Type of Cooling Water Closed (Nat. Tower) Closed (Nat. Tower)

Reactor Name Crystal River 3 Dampierre 1 (ORIG)
Operating Utility Name Florida Power Corp. EdF

Country U.S. France
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. B&W FRAM"
Net Power 821 890
Gross Mega Watt Elec, Rated Power MWe 860 937
Cold Leg Prim. Cool. Temp. 100% Power *F 555 546
Changedto: ....
Changedto: ---

Steam Gener. Model Number OTSG 177 51M
Condensate Purification Method Deep Bed None
Steam Generator Tubing Material 600SR (See Note 2b) 600MA
Tube Expansion Method Part Depth Roll Full Depth Roll + Kiss
Main Condenser Tube Sheet Material (Orig.) Stainless Steel Carbon Steel
Changedto: ---..
Main Condenser Tubing Material (Orig.) CuNi 70/30 Brass

Changedto: ....
Changedto: ---..

LP FWH Tubing Material (Original) 304 SS Brass + 304L SS
Changed to: ....
HP FWH Tubing Material (Origina!) 304 SS Carbon Steel

Changed to:
Changed to: --
Type of Cooling Water Seawater Closed (Tower)

Reactor Name Dampier1re (REPL) Dampierre 2

PWR or PHWR PWR PWR
Initial Critical Date or Date of Major Change 1990 12/15/80
Net Power 890 890
Changedto: ....
Changedto: ---
Steam Gener. Model Number 51B 51M
Changed to:
Main Condenser Tubing Material (Orig.) Titanium Brass
Changed to: ......
Changed to: - --
LP FWH Tubing Material (Original) 304L SS Brass + 304L SS
Changed to: ..
Changed to: ....
Changed to: .....
Reactor Name Dampierre 3 Dampierre 4

PWR or PHWR PWR PWR
Net Power 890 890
Changedto: ......
Changed to: ..

Steam Gener. Model Number 51M 51M
Changed to: ---

Changedto: ......
Changedto: ......
Main Condenser Deaerating (Y/N I Yes Yes
Changed to: ---...

Changedto: ......
Changed to: ......
Reactor Name Darlington 1 Darlington 2

Net Power 881 881

Changed to:
Changed to:
Condensate Purification Method None . None
Support Plate Material 41OSS 41OSS
Tube Support Configuration Lattice Bars Lattice Bars
Tube Expansion Method Hyd. Roll near TS Sec. Face Hyd. Roll near TS Sec. Face
Main Condenser Tube Sheet Material (Orig.) CS with SS cladding CS with SS cladding
Changed to:
Changed to:
Changed to:
Changed to:
HP FWH Tubing Material (Original) 304SS 304SS
Changed to:
Reheater Tubing Material (MSR)(Original) SA 249 Tp321 SA 249 Tp321
Changed to:
Type of Cooling Water Lake (Ontario) Lake (Ontario)
Reactor Name Darlington 3 Darlington 4
0
Initial Critical Date or Date of Major Change Not yet in service Not yet in service
Net Power 881 881
Changedto: --
Changedto: - -
Steam Gener. Oper. Temp. OF 509 509,
Support Plate Material 410SS 410SS
Tube Support Configuration Lattice Bars Lattice Bars
Tube Expansion Method Hyd. Roll near TS Sec. Face Hyd. Roll near TS Sec. Face
Main Condenser Tube Sheet Material (Orig.) CS with SS cladding CS with SS cladding
Changedto: --

Changedto: -
Changedto:
Changedto: .....
HP FWH Tubing Material (Original) 304 SS 304SS
Changedto: ---...
Reheater Tubing Material (MSR)(Original) SA 249 Tp321 SA 249 Tp321

Changed to:
Type of Cooling Water Lake (Ontario) Lake (Ontario)
Reactor Name Davis Besse 1 Diablo Canyon 1
Operating Utility Name Toledo Edison Co. PG&E

Country U.S. U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. B&W W
Net Power 906 1073
Changedto: ......
Changedto: ..
Steam Gener. Model Number OTSG 177 51
Condensate Purification Method Full / Powdered Resin Full / Deep Bed
Steam Generator Tube Sheet Material Forged LA Steel *Forged LA Steel

Tube Expansion Method Part Depth Roll Full / Explosive
Main Condenser Tube Sheet Material (Orig.) Carbon Steel CuNi 90/10
Changed to: ......

Main Condenser Tubing Material (Orig.) 304SS Titanium
Changed to: ..
Changed to: ..
LP FWH Tubing Material (Original) 304SS Stainless Steel
Changedto: ---
HP FWH Tubing Material (Original) 304SS Stainless Steel

Changed to:
Reheater Tubing Material (MSR)(Original) 304SS Stainless Steel
Changedto:
Type of Cooling Water Fresh Surf (Cooling Tower) Seawater
Reactor Name Diablo Canyon 2 Doel 1
Operating Utility Name PG&E ELECTRABEL

Country U.S. Belgium
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W ACECOWEN
Net Power 1079 400
Changed to: ..
Changed to: ..

Phosphates Added? No No (See Note 12)
Condensate Purification Method Full / Deep Bed Deep Bed
Tube Expansion Method Full / Explosive Part Depth Roll
Main Condenser Tube Sheet Material (Orig.) CuNi 90/10 Aluminum Bronze
Changed to: --- Titanium
Main Condenser Tubing Material (Orig.) Titanium Aluminum Brass
Changed to: --- CuNi (See Note 13)
Changed to: --- 304L SS

Changed to: ---...
Reheater Tubing Material (MSR)(Original) Stainless Steel CuNi 90/10

Changed to: --- 439 SS
Type of Cooling Water Seawater Brackish
Reactor Name Doel 2 Doel 3
Operating Utility Name ELECTRABEL ELECTRABEL

Country Belgium Belgium
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. ACECOWEN FRAMECECO
Net Power 400 897
Changedto: ---
Changed to: ---

Condensate Purification Method Deep Bed Deep Bed

Main Condenser Tube Sheet Material (Orig.) Aluminum Bronze Aluminum Bronze
Changed to: Titanium Titanium

Main Condenser Tubing Material (Orig.) Aluminum Brass Aluminum Brass
Changedto: CuNi (See Note 13) Titanium
Changedto: Titanium ---

Changedto: 304L SS -

Changedto: ....
Changedto: 439 SS --
Type of Cooling Water Brackish Brackish/Nat. Draft Tower

Reactor Name IDoel 4
Operating Utility Name ELECTRABEL

Country Belaiumn
PWR or PHWR PWR

Nuclear Steam System Supply Manuf. ACECOWEN
Net Power 1003
Cold Leg Prim. Cool. Temp. 100% Power °F 552
Hot Leg Prim. Cool. Temp. 100% Power 'F 626
Changedto: 617
Changed to:
Steam Gener. Model Number E
Phosphates Added? Yes
Condensate Purification Method Startup Only
Steam Generator Tube Sheet Material FLAS + 600 Clad (See Note'14)
Support Plate Material 405SS
Main Condenser Tube Sheet Material (Orig.) Aluminum Bronze
Changed to:
Changed to:
Changed to:
LP FWH Tubing Material (Original) Stainless Steel
Changed to:
Changed to:
Reheater Tubing Material (MSR)(Original) 304L Stainless Steel
Changed to:
Type of Cooling Water Brackish/Nat. Draft Tower

Reactor Name Douglas Point (end 5/84) Embalse
Operating Utility Name Ontario Hydro CNEA

Country Canada Argentina
Net Power 206 600
Changed to: -
Changed to: - -
Condensate Purification Method
Steam Generator Tubing Material Monel 400 800M
Steam Generator Tube Sheet Material
Tube Support Configuration Broached Plate
Tube Expansion Method
Changed to:
Main Condenser Tubing Material (Orig.)
Changed to:
Changed to:
Main Condenser Deaerating (Y/NI
Deaerator Other Than Condenser (Y/N)
LP FWH Tubing Material (Original)
Changedto:
HP FWH Tubing Material (Original)
Changedto:
Reheater Tubing Material (MSR)(Original)
Changed to:
Type of Cooling Water
Reactor Name Emsland Farley 1
Operating Utility Name KLE Alabama Power

Country Germany U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. KWU W
Net Power 1270 829
Changedto: --
Changedto: ....
Steam Gener. Model Number 54SK 51
Condensate Purification Method Mechan. 100% None
Tube Expansion Method Mech. Roll-Top & Bottom Full - WEXTEX (Explosive)
Main Condenser Tube Sheet Material (Orig.) C-Steel / SS-clad Aluminum Bronze
Changedto: ---..
Main Condenser Tubing Material (Orig.) Austenitic SS Titanium

Changedto: --
Changedto: ....
LP FWH Tubing Material (Original) Carbon Steel Admiralty
Changedto: -- Stainless Steel
HP FWH Tubing Material (Original) Carbon Steel CuNI 80/20
Reheater Tubing Material (MSR)(Original) Carbon Steel CuNi 90/10
Changed to: --- Stainless Steel
Type of Cooling Water Closed'(Tower) Fresh Mech. Tower
.EPRI Licensed Material
Reactor Name Farley 2 Fessenheim 1 Fessenheim 2
Operating Utility Name Alabama Power EdF EdF

Country U.S. France France
Nuclear Steam System Supply Manuf. W FRAM FRAM
Initial Critical Date or Date of Major Change 5/15/81 3/7/77 6/27/77
Net Power 829 880 880
Gross Mega Watt Elec. Rated Power MWe 860 920 920
Cold Leg Prim. Cool. Temp. 100% Power IF 543 543 543
Hot Leg Prim. Cool. Temp. 100% Power 'F 607 611 611
Changed to:
Changed to:
Steam Gener. Oper. Temp. 'F 517 523 523
Number of Steam Gener. 3 3 3
Steam Gener. Model Number 51 51A 51A
On-Line Boric Acid Added? Yes No No
Condensate Purification Method None None None
Steam Generator Tubing Material 600MA 600MA 600MA
Support Plate Material Carbon Steel Carbon Steel Carbon Steel
Tube Support Configuration Drilled Hole Drilled Hole Drilled Hole
Tube Expansion Method Full Depth Roll Full / Explosive Full Depth Roll
Main Condenser Tube Sheet Material (Orig.) Aluminum Bronze Carbon Steel Carbon Steel
Changedto:
Main Condenser Tubing Material (Orig.) Titanium Admiralty Brass Admiralty Brass
Changedto:
Changed to:
LP FWH Tubing Material (Original) Admiralty 304L 304L
HP FWH Tubing Material (Original) CuNI 80/20 Carbon Steel 18% Cr SS (439L)
Changedto: Stainless Steel --- --
Reheater Tubing Material (MSR)(Original) CuNi 90/10 Carbon Steel 18% Cr SS (439L)
Changedto: Stainless Steel ....
Type of Cooling Water Fresh Mech. Tower Fresh (River) Fresh (River)
Reactor Name Flamanville 1 Flamanville 2

PWR or PHWR PWR PWR
Nuclear Steam System Supply Manut. FRAM FRAM
Net Power 1350 1300
Changed to: 616 616
Changed to: ....
Steam Gener. Oper. Temp. 'F 549 549
Main Condenser Tube Sheet Material (Orig.) Copper Copper
Changedto: -- --
Changedto: ....
Changed to: --
Changedto: ....
Changedto: ..
Changedto: ....
Type of Cooling Water Seawater Seawater
Reactor Name Fort Calhoun 1 Genkai 1
Operating Utility Name OPPD Kyushu Electric

Country U.S. Japan
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. CE MHI
Net Power 457-->470 529
Gross Mega Watt Elec. Rated Power MWe 481 -->502 559
Changed to:. 586 (1984) 599
Changedto: 594 (1987) 601

Condensate Purification Method None Full / Mixed Bed

Main Condenser Tube Sheet Material (Orig.) Muntz Naval Brass
Changed to: -- -
Main Condenser Tubing Material (Orig.) 304SS Aluminum Brass

Changed to: ....
Changed to: ....
LP FWH Tubing Material (Original) Arsenic Copper + SS Aluminum Brass
Changed to: Stainless Steel -
HP FWH Tubing Material (Original) Stainless Steel CuNi 80/20

Changed to: - -
Reheater Tubing Material (MSR)(Original) No Reheater CuNi 90/10

Changed to: -- -
Type of Cooling Water Fresh (River) Seawater

Reactor Name Genkal 2 Gentilly 1 (end 6/77)
Operating Utility Name Kyushu Electric Hydro Quebec

Country Japan Canada
Nuclear Steam System Supply Manuf. MHI B&W
Net Power 529 250
Changed to: - -
Changed to: -

Condensate Purification Method Full / Mixed Bed


Tube Expansion Method Full Depth Roll + Rubber
Main Condenser Tube Sheet Material (Orig.) Naval Brass
Changedto: --
Main Condenser Tubing Material (Org.) Aluminum Brass
Changedto: --
Changedto: --

LP FWH Tubing Material (Original) Aluminum Brass
Changed to:
Changed to:
Changed to:
CuNi 90/10
Seawater
0
Reactor Name Gentilly 2 I Ginna
Operating Utility Name Hydro Quebec RG & E

Country Canada U.S.
PWR or PHWR PHWR PWR
Nuclear Steam System Supply Manuf. AECL W
Net Power 638 470
Changed to: ..
Changed to: ......

Steam Gener. Model Number B&W 638 MWe 44

Phosphates Added? No Yes-->No
Condensate Purification Method Full / Deep Bed

Support Plate Material Stainless Steel Carbon Steel
Tube Support Configuration Broached Plate Drilled Hole
Tube Expansion Method Part Depth Roll

Changed to:
Main Condenser Tubing Material (Orig.) Admiralty Brass
Changed to: --
Changed to: --

LP FWH Tubing Material (Original) Admiralty
Changed to: Stainless
Changed to: Stainless
Reheater Tubing Material (MSR)(Original) 90/10 CuNi
Changed to: 439SS
Type of Cooling Water River Fresh (Lake)

Reactor Name Goesgen Golfech 1
Operating Utility Name KKG EdF

Country Switzerland France
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. KWU FRAM
Net Power 920 1310
Changed to: ....
Changedto: ....

Steam Gener. Model Number 68/19
Steam Generator Tubing Material 800M 600TT
Support Plate Material Austenitic SS 13% Cr. Steel
Tube Support Configuration Grids/Strip (See Note 3) Broached Hole
Tube Expansion Method Mech. Roll-Top & Bottom Full Depth Roll + Kiss
Main Condenser Tube Sheet Material (Orig.) C-steel / Epoxy Coating Carbon Steel
Changedto:
Main Condenser Tubing Material (Orig.) Admiralty Brass 304SS + Titanium
Changed to: Stainless Steel -
Changed to:
LP FWH Tubing Material (Original) Carbon Steel 304L SS
Changedto:
HP FWH Tubing Material (Original) Carbon Steel 18% Cr SS (439L)
Changedto: ---
Reheater Tubing Material (MSR)(Original) Carbon Steel 18% Cr SS (439L)
Changedto: ..
Type of Cooling Water Closed (Nat. Tower) Closed (Tower)
Reactor Name anrheinfeld KKG
Operating Utility Name Bayernwerk AG

Country Germany
PWR or PHWR PWR
Nuclear Steam System Supply Manuf. KWU
Net Power 1235
Changedto: --
Changedto:
Steam Gener. Model Number 54GS
Steam Generator Tubing Material 800M
Support Plate Material Austenitic SS
Tube Support Configuration Grids/Strip (See Note 3)
Tube Expansion Method Mech. Roll-Top & Bottom
Main Condenser Tube Sheet Material (Orig.) C-steel / Epoxy Coating
Changedto: ---
Main Condenser Tubing Material (Orig.) Austenitic SS
Changedto: --
Changedto:
LP FWH Tubing Material (Original) Carbon Steel
Changedto:
Changed to:
Changed to:
Type of Cooling Water Closed (Tower)
Reactor Name Gravelines B1
Country France
PWRorPHWR PWR
-Nuclear Steam System Supply Manut. FRAM

Net Power 910

Changed to: ---
Changed to: --





Main Condenser Tube Sheet Material (Orig.) Double tubesheet CS on 2ndary side
Changedto: --
Changedto: ---
Changedto:
LP FWH Tubing Material (Original) 304L

Changedto:
Changedto:
Reheater Tubing Material (MSR)(Odiginal) Carbon Steel
Changedto:
Type of Cooling Water Seawater
Reactor Name II Gravelines B2

Country France
PWR or PHWR PWR
Net Power 910
Changed to:
Changed to:
Changed to:
Changed to:
Changed to:
Changed to:
Chanqed to:
Changed to:

Country France
PWR or PHWR PWR
Net Power 910
Changedto:
Changedto:

Changed to: ---

Changed to: --
Changed to:
Changed to:

Changed to:

Changedto:

Country France
PWR or PHWR PWR
Net Power 910
Changed to:
Changed to:
Changedto:
Changedto: --
Changedto: --

Changedto:
Changedto:
Changedto: --

Reactor Name Gravelines C5 Gravelines C6

PWR or PHWR PWR PWR
fear Steam System Supply Manuf. FRAM FRAM
,ritical Date or Date of Major Change 8/15/84 7/15/85
Net Power 910 910
Mega Watt Elec. Rated Power MWe 951 951
?g Prim. Cool. Temp. 100% Power IF 546 546
g Prim. Cool. Temp. 100% Power IF 613 613
Changedto:
Changedto:
'ondensate Purification Method None None
team Generator Tubing Material 600TT 600TT
am Generator Tube Sheet Material Forged LA Steel Forged LA Steel
ondenser Tube Sheet Material (Orig.)
Changedto:
Condenser Tubing Material (Org.) Titanium Titanium
Changed to: ---.
Changed to: -- -
ain Condenser Deaerating (Y/NI Yes Yes

)rator Other Than Condenser (Y/N) No No
FWH Tubing Material (Original) 304L SS 304L SS
Changed to: - -
FWH Tubing Material (Original) Carbon Steel Carbon Steel

Changedto: - -
ater Tubing Material (MSR)(Original) Carbon Steel 18% Cr SS (439L)

Changedto: - -
Reactor Name Grohnde Haddam Neck
Operating Utility Name KWG CYAP

PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. KwU W
Net Power 1325 582
Changed to: ..
Changed to: - --

Steam Gener. Model Number 54GS 27
Condensate Purification Method Deep Bed (See Note 4) None
Tube Expansion Method Mech. Roll-Top & Bottom Part Depth Roll
Main Condenser Tube Sheet Material (Orig.) C-Steel / Ti-clad Muntz
Changedto: ....
Main Condenser Tubing Material (Orig.) Titanium 304 SS + Sea Cure
Changed to: ---
Changed to: ......
Changedto: ---..
HP FWH Tubing Material (Original) Carbon Steel Stainless Steel

Changed to: .....
Reheater Tubing Material (MSR)(Original) Carbon Steel 439 SS
Changed to: ---...
Type of Cooling Water Closed (Tower) Fresh (River)

Reactor Name Ikata 1 Ikata 2
Operating Utility Name Shikoku Electric Shikoku Electric

Country Japan Japan
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. MHI MHI
Net Power 538 538
Changedto: ..
Changedto: --

Phosphates .dded? No No
Condensate Purification Method Full / Mixed Bed Full / Mixed Bed
Tube Support Configuration Drilled Hole Dr. Hole w/Chamf
Tube Expansion Method Full Depth Roll Full Depth Roll + Rubber
Main Condenser Tube Sheet Material (Orig.) Naval Brass Naval Brass
Changedto: ---..

Changedto: ......
Changedto: ---..

LP FWH Tubing Material (Original) Al. Brass + CuNi 70/30 Aluminum Brass
Changedto: ..
HP FWH Tubing Material (Original) CuNi 80/20 CuNi 80/20
Changedto: ---
..
Changedto: ...
Reactor Name Indian Point 1 (end 10174) Indian Point 2
Operating Utility Name CONED CONED

Country U.S. U.S.
PWR or PHWR PWR PWR
Net Power 257 864
Cold Leg Prim. Cool. Temp. 1001/6 Power IF 504 524
Changed to: 585 in 1989
Changed to: 591 in 1991
Steam Gener. Model Number 44
Phosphates Added? Yes-->No

Main Condenser Tube Sheet Material (Orig.) Silicon Bronze
Changed to: Titanium
Main Condenser Tubing Material (Orig.) As Admiralty
Changed to: 1/3 Tit+2/3AdBr
Changed to:
LP FWH Tubing Material (Original) 4/5 SS-1/5 CuNi 90/10
Changed to:
HP FWH Tubing Material (Original) 304 SS
Changed to: ---
Reheater Tubing Material (MSR)(Original) 439 SS
Changed to:
Reactor Name II Indian Point 3 (ORIG) II Indian Point 3 (REPL) II

I
Operating Utility Name NYPA NYPA

Country U.S. U.S.
PWR or PHWR PWR PWR
Initial Critical Date or Date of Major Change 4/6/76 Replaced: 6/89
Net Power 965
Chanoed to:
Changed to: ---

Steam Gener. Model Number 44 44F
Condensate Purification Method None Partial 30%


Tube Expansion Method Part Depth Roll Full / Hydraulic
Main Condenser Tube Sheet Material (Orig.) Aluminum Bronze Titanium
Changed to: .....
Main Condenser Tubing Material (Orig.) As Admiralty Titanium
Changed to: Titanium ---
Changed to: . --

LP FWH Tubing Material (Original) CuNi 90/10 Stainless Steel
Changed to: Ferrous ---
HP FWH Tubing Material (Original) CuNi 80/20 Stainless Steel
Changed to: Ferrous --
Reheater Tubing Material (MSR)(Original) 439 SS 439 SS
Changed to: .....

Reactor Name Isar 2 Kewaunee
Operating Utility Name KKI Wis. Public Service

PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. KWU W
Net Power 1320 540
Changed to: ---
Changed to: ---..

Steam Gener. Model Number 54SK 51
Condensate Purification Method Mechan. 100% None
Main Condenser Tube Sheet Material (Orig.) C-Steel / SS-clad Carbon Steel
Changed to:
Main Condenser Tubing Material (Orig.) Austenitic SS Admiralty
Changed to: --- Stainless Steel
Changed to: ......
LP FWH Tubing Material (Original) Carbon Steel CuNi 70/30
HP FWH Tubing Material (Original) Carbon Steel CuNi 70/30
Changedto: • -- Stainless Steel
Changedto: --. Stainless Steel
Type of Cooling Water Closed (Tower) Fresh (Lake)
Reactor Name Koeberg 1 (SG 3) Koeberg 1 (SGs 1&2)
Operating Utility Name ESKOM ESKOM

Country South Africa South Africa
PWR or PHWR PWR PWR
Net Power 920
Changed to: -
Changed to: -
Condensate Purification Method Yes, type=?? Yes, type=??
i
Changedto:
Changedto:
Changedto:
LP FWH Tubing Material (Original)

Changed to:
HP FWVHTubin Material (Oriqinaln
ii Chnged to:
Changedto:
Reactor Name Koeberg 2 Kori 1
Operating Utility Name ESKOM Korea Electric Pwr

Country South Africa Korea
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. FRAM W
Net Power 920 556
Changedto: ....
Changedto: ....
Steam Gener. Model Number 51B 51
On-Line Boric Acid Added? No No-->Yes-->No
Condensate Purification Method Deep Bed
Tube Expansion Method Full Depth Roll + Kiss Full Depth Roll
Changedto: Titanium
Main Condenser Tubing Material (Orig.) Titanium Alum. Brass
Changedto: ......
LP FWH Tubing Material (Original) Admiralty
Changedto: Stainless Steel
Changedto: Stainless Steel
Reheater Tubing Material (MSR)(Original) Stainless Steel
Changed to:
Reactor Name Kori 2 Kori 3
Operating Utility Name Korea Electric Pwr Korea Electric Pwr

Country Korea Korea
PWR or PHWR PWR PWR
Net Power 605 895
Changed to: ---
Changedto: ---
Steam Gener. Oper. Temp. QF 535 540

Steam Gener. Model Number F F
Support Plate Material 405 SS 405 SS
Tube Expansion Method Full / Hydraulic Full / Hydraulic
Main Condenser Tube Sheet Material (Orig.) Muntz Aluminum Bronze
Changedto: ....

Changed to: ....
Changed to: ---
Main Condenser Deaerating (Y/NI No Yes
Changedto: .....
Changedto: ...

Changed to: ....
Reactor Name Kori 4 Krsko
Operating Utility Name Korea Electric Pwr NEK

Country Korea Yugoslavia
PWR or PHWR PWR PWR
Net Power 895 632
Changedto: ....
Changedto: ---

Steam Gener. Model Number F D4
On-Line Boric Acid Added? No No-->Yes-->No
Condensate Purification Method Deep Bed Full Powdex+Deep Bed

Tube Expansion Method Full / Hydraulic Full Depth Roll
Main Condenser Tube Sheet Material (Orig.) Aluminum Bronze Carbon Steel
Changed to: ......

Main Condenser Tubing Material (Orig.) Titanium CuNi 90/10
Changed to: -- 304SS
Changed to: ..
LP FWH Tubing Material (Original) Stainless Steel 304SS
Changed to: - -
HP FWH Tubing Material (Original) Stainless Steel 304SS

Changed to: - -
Reheater Tubing Material (MSR)(Original) Stainless Steel CuNi

Changed to: - 439 SS Partial
Type of Cooling Water Seawater Fresh (River)
Reactor Name F Loviisa 1 Lovilsa 2
Operating Utility Name Imatran Voima Oy Imatran Voima Oy

Country Finland Finland
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. AEE AEE
Net Power 445 445
Changedto:
Changedto:
Steam Gener. Model Number B-213 B-213
Condensate Purification Method Precoat + Mixed Bed Precoat + Mixed Bed
Steam Generator Tubing Material 08X18H10T (AISI 321) 08X18H10T (AISI 321)
Steam Generator Tube Sheet Material 08X18H1OT (AISI 321) 08X18H1OT (AISI 321)
Support Plate Material 08X18H10T (AISI 321) 08X18H1OT (AISI 321)
Tube Support Configuration Spring type strip support 08X1 8H1 OT (AISI 321)
Tube Expansion Method Explosion Explosion
Main Condenser Tube Sheet Material (Orig.) 1) 08X18H10T/CS 1) 08X18H10T/CS
Changedto: 2) Titanium-cladded / CS 2) Titanium-cladded / CS
Main Condenser Tubing Material (Orig.) 1) SM0254 (See Note 10) 1) SM0254 (See Note 10)
Changedto: 2) Titanium 2) Titanium
Changedto: ....
LP FWH Tubing Material (Original) 12X18H10T 12X18H10T
Changedto:
HP FWH Tubing Material (Original) 1) Carbon Steel 1) Carbon Steel
Changed to: 2) Aust. SS from 1991 2) Ferr. SS from 1992
Changed.to:
Reactor Name Maanshan 1 Maanshan 2 Maine Yankee
Operating Utility Name Taiwan Power Co. Taiwan Power Co. MYAP
Country Taiwan Taiwan U.S.
Nuclear Steam System Supply Manuf. W W CE
Net Power 890 890 840-->864
Cold Leg Prim. Cool. Temp. 100% Power °F 557 557 550
Hot Leg Prim. Cool. Temp. 100% Power °F 624 619 602
Changedto:
Changedto:
*Number of Steam Cener. 3 3 3
On-Line Boric Acid Added? No No Yes
Condensate Purification Method Full / Deep Bed Full / Deep Bed None
Steam Generator Tubing Material 600TT 600TT 600MA
Support Plate Material 405 SS 405 SS Carbon Steel
Tube Support Configuration Broached Hole Broached Hole Dr. Hote/Eggcrate
Tube Expansion Method Full / Hydraulic Full / Hydraulic Full / Explosive
Changedto:
Main Condenser Tubing Material (Orig.) Titanium Titanium Stainless Steel

Changed to:
Changedto:
LP FWH Tubing Material (Original) Stainless Steel Stainless Steel Multiple
Changed to:
HP FWH Tubing Material (Original) Stainless Steel Stainless Steel Stainless Steel
Changed to:
Reheater Tubing Material (MSR)(Original) Corten Corten Stainless Steel
Changed to:
Type of Cooling Water Seawater Seawater Brackish
Reactor Name McGuire I McGuire 2
Operating Utility Name Duke Power Co. Duke Power Co.

Country U.S. U.S.
PWR or PHWR PWR PWR
Net Power 1129 1129
Changedto: ......
Changedto: ..
Changed to: ......
Changed to: ..
Changed to: ..
LP FWH Tubing Material (Original) 304 SS 304SS
Changed to: ..
Changed to: -.
Changed to: Stainless Steel Stainless Steel
Type of Cooling Water Fresh (Lake) Fresh (Lake)
Reactor Name Mihama 1 Mihama 2
Operating Utility Name Kansai Electric Electric Kansai Electric Electric

Country Japan Japan
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W (CE SG's) MHI
Initial Critical Date or Date of Major Change 7/29/70 4/10/72 •
Net Power 320 470

Changed to: - --
Changed to: -- -

Steam Gener. Model Number CE 44

Phosphates Added? Yes-->No Yes-->No
Condensate Purification Method Partial 24% Partial 17%

Support Plate Material Mild Steel Carbon Steel
Tube Expansion Method Full / Explosive Part Depth + Remedial Exp.
Changed to: - --
Changed to: - --
Changed to: ..
LP FWH Tubing Material (Original) Aluminum Brass Aluminum Brass
Changed to: ..
HP FWH Tubing Material (Original) CuNI 80/20 CuNI 80/20
Changed to: ..
Changed to: ....

Reactor Name Mihama 3 Millstone 2 (ORIG)
Operating Utility Name Kansai Electric Electric Northeast Utilities

Country Japan U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. MHI CE
Net Power 780 870-->910
Changed to: -
Changed to: -
Condensate Purification Method Full / Mixed Bed Deep Bed
Tube Support Configuration Drilled Hole Dr. Hole/Eggcrate
Tube Expansion Method Full Depth Roll Full / Explosive
Main Condenser Tube Sheet Material (Orig.) Naval Brass Muntz
Changed to:
Main Condenser Tubing Material (Orig.) Aluminum Brass Alum. Brass
Changedto: Titanium
Changedto:
LP FWH Tubing Material (Original) Aluminum Brass Multiple
Changedto: I....
HP FWH Tubing Material (Original) CuNI 80/20 CuNi 80/20
Changedto:
Changedto:
Type of Cooling Water Seawater Seawater, once through
Reactor Name Millstone 2 (REPL) Millstone 3
Operating Utility Name Northeast Utilities Northeast Utilities

Country U.S. U.S.
PWR or PHWR PWR PWR
Initial Critical Date or Date of Major Change Replaced:1992 1/15/86
Net Power 910 1150
0
Cold Leg Prim. Cool. Temp. 100% Power F 558
Hot Leg Prim. Cool. Temp. 100% Power °F 621
Changedto:
Changedto:
Steam Gener. Oper. Temp. 'F
Steam Gener. Model Number B&W Canada Replacement F
Condensate Purification Method Deep Bed Full
Tube Support Configuration Dr. Hole/Eggcrate Broached Hole
Tube Expansion Method Full / Explosive Full / Hydraulic
Main Condenser Tube Sheet Material (Orig.) Muntz Muntz
Changedto: ... I
Changedto:
Changedto:
LP FWH Tubing Material (Original) Ferrous Stainless Steel
Changedto:
HP FWH Tubing Material (Original) Ferrous Stainless Steel
Changedto:
Reheater Tubing Material (MSR)(Original) Ferrous CuNi 90/10
Changedto:
Type of Cooling Water Seawater, once through Seawater
Reactor Name Muelheim-Kaerlich Napot Point
Operating Utility Name RWE-Energie AB

Country Germany Phillipines
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. BBR W

Net Power 1226

Changed to: --
Changedto: --
Steam Gener. Oper. Temp. *F 458.6/595.4 (See Note 16)

Steam Gener. Model Number 205 F
Condensate Purification Method External Regeneration Full / Deep Bed

Steam Generator Tubing Material Inconel 600 600TT
Steam Generator Tube Sheet Material 22NiMoCr37 Forged LA Steel
Tube Expansion Method Full / Hydraulic

Main Condenser Tube Sheet Material orig.) RST 37.2
Changed to: --
Main Condenser Tubing Material (Orig.) Brass Titanium

Changedto: X5CrNiMo 1810 (1.4401)
Changed to:
LP FWH Tubing Material (Original) ST 35.8 (LP Preheater) Stainless Steel
Changedto:
HP FWH Tubing Material (Original) ST 35.8 (HP Preheater) Stainless Steel
Changed to:
Reheater Tubing Material (MSR)(Original) ST 35.8 Stainless Steel
Changed to:
Type of Cooling Water Closed (Tower) Seawater
Reactor Name Neckarwestheim-1 Neckarwestheim-2
Operating Utility Name GKN GKN

Country Germany Germany
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. KWU KWU
Net Power 785 1225
Changedto: - -
Changedto: - --

Steam Gener. Model Number 54SK
Condensate Purification Method None Mechan. 100%
Support Plate Material Austenitic SS Austenitic SS
Tube Support Configuration Grids/Strip (See Note 3) Grids/Strip (See Note 3)
Tube Expansion Method Mech. Roll-Top & Bottom Mech. Roll-Top & Bottom
Main Condenser Tube Sheet Material (Orig.) C-steel / Epoxy Coating C-Steel / SS-clad
Changed to:
Main Condenser Tubing Material (Orig.) Admiralty Brass Austenitic SS
Changed to: Stainless Steel --
Changedto:
LP FWH Tubing Material (Original) Carbon Steel Carbon Steel
Changed to:
Changedto:

Changed to:
Type of Cooling Water Fresh or Recirc. Closed (Tower)
Reactor Name Nogent s/Seine 1 Nogent s/Seine 2

PWR or PHWR PWR PWR
Net Power 1310 1310
Changed to: 616 616
Changed to: -
Steam Generator Tubing Matkrial 600TT 600TT
Changedto: ....
Main Condenser Tubing Material (Orig.) Cu + 439 Cu + 439
Changedto:
Changedto:
Changed to:
Changed to:
Changed to:
Reactor Name North Anna 1 (ORIG) North Anna 1 (REPL)
Operating Utility Name Virginia Power Virginia Power

Country U.S. U.S.
PWR or PHWR PWR PWR
Initial Critical Date or Date of Major Change 4/15/78 Replaced: 1993
Net Power 907
Changedto: 618 --
Changedto: ....

Tube Expansion Method Full / Explosive Full / Hydraulic
Changedto: - --

Changedto: - --
Changedto: - -

LP FWH Tubing Material (Original) CuNi 80/20 Stainless Steel
Changedto: Stainless Steel ...
HP FWH Tubing Material (Original) CuNI 80/20 Stainless Steel
Changed to: Stainless Steel ---
Reheater Tubing Material (MSR)(Original) CuNi 90/10 Stainless Steel
Type of Cooling Water Fresh (Reservoir) Fresh (Reservoir)
Reactor Name North Anna 2 Obrigheim (ORIG)
Operating Utility Name Virginia Power KWO

PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W KWNU
Net Power 907 328
Changed to: 618 --
Changed to: -- --

Condensate Purification Method Full / Powdex None
Tube Expansion Method Full / Explosive Mech. Roll-Top,Mid,Bottom
Main Condenser Tube Sheet Material (Orig.) Stainless Steel C-steel / Epoxy Coating
Changed to:
Main Condenser Tubing Material (Orig.) 304SS Admiralty Brass
Changedto: - Stainless Steel
Changedto:
LP FWH Tubing Material (Original) CuNi 80/20 Carbon Steel
Changedto: Stainless Steel --
HP FWH Tubing Material (Original) CuNI 80/20 Carbon Steel
Changedto: Stainless Steel --
Changedto: Stainless Steel ---
Type of Cooling Water Fresh (Reservoir) River

Reactor Name Obrigheim (REPL) Oconee 1 -1
Operating Utility Name KWO Duke Power Co:

PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. KWU B&W
Initial Critical Date or Date of Major Change Replaced: 8/83 4/15/73

Net Power 340 846

Changedto: -
Changedto: - -

Steam Gener. Model Number OTSG 177
Condensate Purification Method None Powdered Resin Demin.

Steam Generator Tubing Material 800M 600SR (See Note 2b)

Tube Support Configuration Grids/Strip (See Note 3) Broached Hole
Main Condenser Tube Sheet Material (Orig.) C-steel / Epoxy Coating Carbon Steel
Changedto: --
Main Condenser Tubing Material (Orig.) Ad. Brass -> SS (See Note 8) 304 SS

Changed to: ....
LP FWH Tubing Material (Original) Carbon Steel 304 SS
Changed to: ---...
HP FWH Tubing Material (Original) Carbon Steel 304 SS
Changed to: ---..

Changed to: ..
Type of Cooling Water River Fresh (Lake)
Reactor Name Oconee2 Oconee3
Operating Utility Name Duke Power Co. Duke Power Co.

Country U.S. U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. B&W B&W
Net Power 846 846
Changed to:
Changed to:
Steam Gener. Model Number OTSG 177 OTSG 177
Condensate Purification Method Powdered Resin Demin. Powdered Resin Demin.
Steam Generator Tubing Material 600SR (See Note 2b) 600SR (See Note 2b)
Changed to:
Main Condenser Tubing Material (Orig.) 304SS 304 SS
Changedto: ---
Changedto: ..
LP FWH Tubing Material (Original) 304SS 304 SS
Changedto:..
HP FWH Tubing Material (Original) 304 SS 304SS
Changed to: ---
Changed to:
Reactor Name Ohi1 Ohi 2
Operating Utility Name Kansai Electric Electric Kansai Electric Electric

Country Japan Japan
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W MHI
Net Power 1120 1120
Changedto:
Changed to:
Condensate Purification Method Full / Deep Bed Full / Deep Bed
Changed to:
Changed to:
Changed to:
LP FWH Tubing Material (Original) Al. Brass + CuNi 90/10 Al. Brass + CuNi 90/10
Changed to:
Changed to:

Changed to:
Reactor Name Ohi 3 Ohi 4
Operating Utility Name Kansai Electric Electric Kansai Electric

Country Japan Japan
PWR or PHWR PWR PWR
Net Power 1127 1127
Changedto: 611 ---
Changedto: -
Steam Gener. Model Number 52FA 52FA
Main Condenser Tube Sheet Material (Orig.) Titanium Titanium
Changedto: - --
Changed to:
Changed to:
LP FWH Tubing Material (Original) Al. Brass + SS Al. Brass + SS
Changed to: .
Changed to: ..
Changed to:
Reactor Name Palisades (ORIG) Palisades (REPL)
Operating Utility Name Consumers Power Co. Consumers Power Co.

Country U.S. U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. CE CE
Net Power 802 812
Changedto: ---
Changedto: ---
On-Line Boric Acid Added? No-->Yes
Condensate Purification Method Powdered Resin Demin. Powdered Resin Demin.
Steam Generator Tubing Material 600 600
Support Plate Material Carbon Steel Stainless Steel
Tube Support Configuration Dr. Hole/Eggcrate Eggcrate
Tube Expansion Method Full / Explosive Full / Explosive
Changedto: --
Main Condenser Tubing Material (Orig.) CuNi 90/10 Stainless Steel

Changedto: -- -
Changedto:
Changedto: ....
Changedto: ....
Changedto: ....
Type of Cooling Water Close (Mech. Tower) Close (Mech. Tower)
Reactor Name Palo Verde 1 Palo Verde 2
Country
PWR or PHWR
U.S.
PWR
U.S.
PWR
0
Nuclear Steam System Supply Manuf. CE CE
Net Power 1270 1270

Changed to: ---
Changedto: ..

Steam Gener. Model Number System 80 System 80

Steam Generator Tube Sheet Material LA Steel-Inconel Clad LA Steel-Inconel Clad
Support Plate Material Stainless Steel Stainless Steel

Tube Support Configuration Eggcrate Eggcrate
Main Condenser Tube Sheet Material (Orig.) Aluminum Bronze Aluminum Bronze
Changed to: ..
Changedto:
Changedto:---

LP FWH Tubing Material (Original) ASTM SA-249 ASTM SA-249

Changed to: ..
HP FWH Tubing Material (Original) ASTM SA-249 ASTM SA-249
Changed to: ....

Changed to: ---..
Type of Cooling Water Close (Mech. Tower) Close (Mech. Tower)

Reactor Name Palo Verde 3
Operating Utility Name Arizona Public Service
Country U.S.
PWR or PHWR PWR
Nuclear Steam System Supply Manuf. CE
Net Power 1270

Changedto:
Changedto:

Steam Gener. Model Number System 80
Steam Generator Tube Sheet Material LA Steel-Inconel Clad
Support Plate Material Stainless Steel
Tube Support Configuration Eggcrate

Tube Expansion Method Full / Explosive
Changed to: --
Changed to:
Changed to:
LP FWH Tubing Material (Original) ASTM SA-249
Changed to: ---
HP FWH Tubing Material (Original) ASTM SA-249
Changed to:
Reheater Tubing Material (MSR)(Original) Corten
Changed to: ---
Type of Cooling Water Close (Mech. Tower)

Reactor Name aluel 1

Country, France
PWR or PHWR PWR
Net Power 1330
Changed to: 616
Changedto."
Steam Generator Tubing Material 600TT
Changed to:
Changed to: --
Changed to:
Changedto: ---
Changed to: ---
Changedto:
Reactor Name Paluel 2

Country France
PWR or PHWR PWR

Net Power 1330

Changed to: 616
Changedto: ...




Changedto: ---
Changedto: --
Changed to: ---

Changed to: ---

Changedto: ---
Changed to: ---

Reactor Name Paluel 3
Country France
PWR or PHWR PWR
Net Power 1330

Cold Leg Prim. Cool. Temp. 100% Power IF 559'

Changed to: 616
Changed to: ---



Changed to:
Changedto:
Changed to: ---

Changed to: --

Changed to: --
Reheater Tubing Material (MSR)(Original) 304L
Changed to:
Reactor Name Paluel 4 Penly 1

PWR or PHWR PWR PWR
Net Power 1330 1330
Changed to: 616 --
Changed to: ---..


Main Condenser Tube Sheet Material (Orig.) Double (Carb. Stt. on 2ndary Side) Copper
Changed to: ....
Main Condenser Tubing Material (Orig.) Titanium 304SS + Titanium
Changed to: ..
Changedto: --

Changedto: --
Changedto:
Changedto: ....
Reactor Name Penly 2

S Philippsburg 2
Operating Utility Name EdF KKP

Country France Germany
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. FRAM KWU
Net Power 1330 1276
Changed to:
Changed to:
Steam Gener. Model Number 68/19 54S
Condensate Purification Method None Mech. 30% (See Note 5)
Support Plate Material 13% Cr. Steel Austenitic SS
Tube Support Configuration Broached Hole Grids/Strip (See Note 3)
Tube Expansion Method Full Depth Roll + Kiss Mech. Roll-Top & Bottom
Main Condenser Tube Sheet Material (Orig.) CS-Titanium plated C-Steel / SS-clad
Changed to:
Main Condenser Tubing Material (Orig.) 304SS + Titanium Austenitic SS
Changed to: ....
Changed to: ..
LP FWH Tubing Material (Original) 304L SS Carbon Steel
Changed to: --
HP FWH Tubing Material (Original) 18% Cr SS (439L) Carbon Steel

Changedto:
Reheater Tubing Material (MSR)(Onginal) 18% Cr SS (439L) Carbon Steel
Changed to: ....
Type of Cooling Water Seawater Closed (Tower)
Reactor Name Pickering I
Country Canada
PWR or PHWR PHWR

Net Power 515

Changedto: ---
Changedto: --



Steam Generator Tubing Material Monel 400
Support Plate Material Cu-Bearing Steel
Tube Support Configuration Lattice Bars

Changedto:
Main Condenser Tubing Material (Orig.) Admiralty Brass (See Note 11)
Changedto: --
Changedto: ---
Main Condenser Deaerating (Y/N I Yes

Changed to:
Changed to:
Reheater Tubing Material (MSR)(Original) Carbon Steel (A210)
Changed to: --
Type of Cooling Water Lake (Ontario)

Reactor Name Pickering 2

Country Canada
PWR or PHWR PHWR
Net Power 515
Changedto:
Changedto:
Tube Support Contiguration Lattice Bars
Changed to:
Changed to:
Changed to:

Changedto: --

Changed to: --

Changedto:
Reactor Name Pkering 3

Country Canada
PWR or PHWR PHWR
Net Power 515
Changed to: --
Changed to:
Changedto:
Changedto:
Changedto:
Changedto: --

Changedto: --

Changed to:

Country Canada
PWR or PHWR PHWR
Initial Critical Dateor Date of Major Change 5/16/73
Net Power 515
Changedto:
Changedto: --
Steam Gener. Oper. Temp. °F 486




Changed to: --
Changed to:
Changed to:


Changed to: ---
Changed to: ---

Changed to: --


Country Canada
PWR or PHWR PHWR
Net Power 516
Changedto: ---
Changed to: -
Changed to: --
Changedto: --
Changedto:
Changedto: --

Changedto: ---
Changedto: --

Reactor Name Piketingj6

Country Canada
PWR or PHWR PHWR
Net Power 516
Changedto: --
Changedto: ---
Changedto: --
Changedto: --
Changedto: ---
Changedto: ---

Changed to: ---
Changedto: --

Reactor Name
] Pickering 7

Country Canada
PWR or PHWR PHWR
Nuclear Steam System Supply Manut. AECL
Net Power 516
Changed to: ---
Changedto: ---
Changed to: ---
Changedto:
Changedto: --

Changed to:
Changed to:
Changedto:
Reactor Name I Pickering 8 1

Country Canada
PWR or PHWR PHWR
Net Power 516
Changed to: --
Changedto: --

Changed to: --
Changedto:
Changedto: --

Changedto: ---
Changedto: --

Changedto:
Reactor Name Paint Beach 1 (ORIG) Point Beach 1 (REPL)
Operating Utility Name Wis. Electric Power Wis. Electric Power

Country U.S. U.S.
PWR or PHWR PWR PWR
Initial Critical Date or Date of Major Change 11/15/70 1984
Net Power 485 485
Changed to:
Changed to:
Tube Expansion Method Part Depth Roll Full / Hydraulic
Main Condenser Tube Sheet Material (Orig.) Muntz Stainless Steel
Changed to:
Main Condenser Tubing Material (Ong.) Admiralty Stainless Steel
Changed to:
Changed to:
LP FWH Tubing Material (Original) CuNi 80/20 Ferrous
Changed to:
HP FWH Tubing Material (Original) CuNi 80/20 Ferrous
Changed to: ---...
Reheater Tubing Material (MSR)(Original) CuNi 90/10 Ferrous

Changed to: ---..

B- 100 Steam Generator Reference Book, Revision I
Reactor Name
II Point Beach 2 1 Point Lepreau I
Operating Utility Name Wis. Electric Power New Brunswick Elec.

Country U.S. Canada
PWRt or PHWR PWR PHWR
Nuclear Steam System Supply Manuf. W AECL
Net Power 485 635
Changed to: ....
Changedto: ---

Steam Gener. Model Number 44 B&W CANDU 6
Phosphates Added? Yes->No Yes

Steam Generator Tube Sheet Material Forged LA Steel Copper Alloy

Changed to:
Main Condenser Tubing Material (Orig.) Admiralty
Changed to:
LP FWH Tubing Material (Original) CuNi 80/20
Changed to: Ferrous
Changed to: Ferrous

Reheater Tubing Material (MSR)(Original) CuNi 90/10
Changedto: Ferrous
Type of Cooling Water Fresh (Lake) Salt Water
Appendix B-Plant Design Characteristics B-I 01
Reactor Name Prairie Island 1 Prairie Island 2
Operating Utility Name Northern States Pwr Northern States Pwr

Country U.S. U.S.
PWR or PHWR PWR PWR
Net Power 503 500
Cold Leg Prim. Cool. Temp. 100% Power OF =530 =530
Changed to: ....
Changed to: ..
Steam Gener. Oper. Temp. *F 511 511
Changed to: - --

Changed to: - --
Changed to: ..
Changed to: - --

Changed to: - -

Changed to: 1 SS; other 3 SS in 5/94 SS (SA268/Tp439)
Type of Cooling Water Fresh (River)/Mech. Tower Fresh (River)/Mech. Tower
Reactor Name Rancho Seco (end 6/89) Ringhals 2 (ORIG)
Operating Utility Name SMUD SSPB

Country U.S. Sweden
PWR or PHWR PWR PWR

Net Power 926 800

Changedto: ....
Changedto: ..
Steam Gener. Model Number OTSG 177 51C


Main Condenser Tube Sheet Material (Orig.) Stainless Steel Cs Clad Ti
Changedto:
Main Condenser Tubing Material (Orig.) Stainless Steel Alum. Brass
Changedto: Titanium
Changed to: ....
Changedto: --
HP FWH Tubing Material (Original) 304L SS Carbon Steel
Changedto:
Reheater Tubing Material (MSR)(Original) 439 SS Carbon Steel

Changedto: --
Type of Cooling Water Fresh Surf (Cooling Towers) Seawater

Reactor Name Ringhals 2 (REPL) Ringhals 3
Operating Utility Name SSPB SSPB

Country Sweden Sweden
PWR or PHWR PWR PWR
Net Power 872 915
Changedto: -- 587
Changedto: --- 610
Steam Gener. Model Number 51W-KWU Replacement D3
Condensate Purification Method . Precoat: SU Only Precoat: SU Only
Support Plate Material 347 / 410 Carbon Steel
Tube Support Configuration Grids / Strip Drilled Hole
Tube Expansion Method Full Hyd.+Top/Bottom Mech Roll Full Depth Roll + Kiss
Main Condenser Tube Sheet Material (Orig.) CS Clad Ti Cs Clad Ti
Changed to: .....
Changed to: - -
Changed to: - -

Changedto: ---..

Changedto: - -

Changed to:

Reactor Name Ringhals 4 Robinson 2 (ORIG)]
Operating Utility Name SSPB Carolina Power

Country Sweden U.S.
PWR or PHWR PWR PWR
Net Power 915 700
Changedto: -- 604
Changedto: ..
Steam Gener. Model Number D3 44
Condensate Purification Method Precoat: SU Only None
Tube Expansion Method Full Depth Roll Part Depth Roll
Main Condenser Tube Sheet Material (Orig.) Cs Clad SS Stainless Steel
Changed to: .....
Main Condenser Tubing Material (Orig.) Titanium Admiralty Brass
Changed to: - Stainless Steel
Changed to: --

LP FWH Tubing Material (Original) Stainless Steel 304/429 SS
Changed to:
HP FWH Tubing Material (Original) Carbon Steel 304/429 SS
Changed to:
Changedto:
Type of Cooling Water Seawater Fresh (Reservoir)
Reactor Name Robinson 2 (REPL) Salem 1
Operating Utility Name Carolina Power Public Service Elec.
Country U.S. U.S.

PWR or PHWR PWR PWR

Net Power 700 1090
Changed to: ---
Changedto: ....

Steam Gener. Model Number 44F 51


Main Condenser Tube Sheet Material (Orig.) Stainless Steel Muntz

Changedto: ..
Main Condenser Tubing Material (Orig.) Stainless Steel AI-6X
Changedto: .....
Changed to: ---..

LP FWH Tubing Material (Original) 304/429 SS 304 SS
Changedto: -- -
HP FWH Tubing Material (Original) 304/429 SS 304 SS

Changedto: - -
Reheater Tubing Material (MSR)(Original) Stainless Steel SS+CuNi 90/10
Changedto: - -
Type of Cooling Water Fresh (Reservoir) Brackish

Reactor Name Salem 2 San Onofre 1 (end 12/92)
Operating Utility Name Public Service Elec. So. Calif. Edison
Country U.S. U.S.

PWR or PHWR PWR PWR
Net Power 1115 436

Changedto: -- 575
Changedto: ....

Condensate Purification Method Full/Deep Bed None
Main Condenser Tube Sheet Material (Orig.) Muntz CuNi 90/10-Ti
Changedto: -- --
Main Condenser Tubing Material (Orig.) AI-6X CuNi 90/10-Ti

Changedto:
Changed to: ....

LP FWH Tubing Material (Original) 304SS Admiralty Brass
Changed to:
HP FWH Tubing Material (Original) 304SS CuNi 90/10
Changed to:
Reheater Tubing Material (MSR)(Original) SS+CuNi 90/10 CuNi 90/10
Changed to:
Type of Cooling Water Brackish Seawater
Reactor Name San Onofre 2 San Onofre 3 Seabrook
Operating Utility Name So. Calif. Edison So. Calif. Edison PSNH
Nuclear Steam System Supply Manuf. CE CE W
Net Power 1070 1080 1148
Cold Leg Prim. Cool. Temp. 100% Power °F 553 553 559
Hot Leg Prim. Cool. Temp. 100% Power IF 609 609 618
Changed to: ---
Changedto: ---

Steam Gener. Model Number 3410 3410 F
On-Line Boric Acid Added? No No No
Condensate Purification Method Full / Deep Bed Full / Deep Bed None
Tube Support Configuration Eggcrate Eggcrate Broached Hole
Tube Expansion Method Full / Explosive Full / Explosive Full / Hydraulic
Main Condenser Tube Sheet Material (Orig.) Aluminum Bronze Aluminum Bronze Aluminum Bronze
Changed to: .....
Main Condenser Tubing Material (Orig.) Titanium Titanium Titanium

Changed to: .......
Changed to: .
Main Condenser Deaerating (Y/NI -Yes Yes Yes
LP FWH Tubing Material (Original) Admiralty Admiralty Stainless Steel
Changed to: ---......
HP FWH Tubing Material (Original) CuNi 90/10 CuNi 90/10 Stainless Steel
Changed to: --- ---
Reheater Tubing Material (MSR)(Original) Carbon Steel Carbon Steel Stainless Steel
Changed to: .....
Type of Cooling Water Seawater Seawater Seawater
Reactor Name Senda! 1 Sendai 2
Country
PWR or PHWR
Kyushu Electric
Japan
PWR
Kyushu Electric
Japan
PWR
0
Net Power 846 846
Changed to: 603 602
Changed to:
Steam Gener. Model Number 51M 51F
Support Plate Material 405 SS 405SS
Tube Support Configuration Dr. Hole/Chamfer Broached Hole
Tube Expansion Method Full Depth Roll + Hydraulic Full / Hydraulic
Changedto: ....
Changedto: ....
Changedto: ..
LP FWH Tubing Material (Original) Al. Brass + CuNi 70/30 Al. Brass + SS
Changedto: ....
Changedto: ---...

Changed to: ---..

Reactor Name Sequoyah 1 Sequoyah 2
Operating Utility Name TVA TVA

Country U.S. U.S.
PWR or PHWR PWR PWR
Net Power 1148 1148
Changedto: -- -
Changedto: -- -

Changedto: ....
Changedto: ....
Changed to: ..

LP FWH Tubing Material (Original) CuNi 90/10 CuNi 90/10
Changed to: 439SS 439SS
Type of Cooling Water Closed (Tower) / River Closed (Tower) / River
13-1 10 Steam Generator Reference Book, Revision I
Reactor Name Shearon Harris Shippingport (end 10/82)
Operating Utility Name Carolina Power Duquesne Light

Country U.S. U.S.
PWR or PHWR PWR PWR


Net Power 900 60
Changed to: ---
Changed to: ....
Steam Gener. Oper. Temp. IF
Steam Gener. Model Number D4 Horiz. U-tube

Steam Generator Tubing Material 600MA SS-->600MA

Tube Expansion Method Full Depth Roll
Changed to: ---
Main Condenser Tubing Material (Orig.) CuNi 90/10
Changed to: ...
Changed to: ---

LP FWH Tubing Material (Original) Stainless Steel
Changedto: --
HP FWH Tubing Material (Original) Stainless Steel
Changed to:
Reheater Tubing Material (MSR)(Original) Stainless Steel

Changedto:
Type of Cooling Water Closed (Tower) / River
Appendix B-Plant Design Characteristics B-I 11
Reactor Name Sizewell B South Texas 1
Operating Utility Name Nuclear Electric HL & P

Country GB U.S.
PWR or PHWR PWR PWR
Net Power 1250
Changed to: --
Changed to: --

Steam Gener. Model Number F1 000 E
Tube Expansion Method Full/Hydraulic Full Depth Roll
Changed to:
Changed to: ......
Changed to: ---..

LP FWH Tubing Material (Original) 304L SS Stainless Steel
Changed to: .
Changed to: ---..

Changed to: ---..
Type of Cooling Water Seawater Brackish (Reservoir)

B- 112 Steam Generator Reference Book, Revision I
Reactor Name South Texas 2 St. Alban 1

Country
PWR or PHWR
HL & P
U.S.
PWR
EdF
France
PWR
0
Net Power 1250 1335

Changed to: - 616
Changed to: ..

Steam Gener. Model Number E 68/19

Condensate Purification Method Full / Deep Bed None

Support Plate Material 405SS 13% Cr. Steel
Main Condenser Tube Sheet Material lOng.) Carbon Steel

Changed to:
Main Condenser Tubing Material (Orig.) Titanium Brass + 304L SS
Changed to: ..
Changed to: ..
LP FWH Tubing Material (Original) Stainless Steel Brass + 304L SS
Changed to: --
Changed to:
Reheater Tubing Material (MSR)(Original) Stainless Steel 304L
Changedto: ....
Type of Cooling Water Brackish (Reservoir) Fresh (River)

Appendix B-Plant Design Characteristics B-1 13
Reactor Name St. Alban 2 St. Laurent B1

PWR or PHWR PWR PWR
Net Power 1335 880

Changedto: 616 --
Changedto: - --

Steam Gener. Model Number 68/19 51M
Changed to: ...
Main Condenser Tubing Material (Orig.) Brass + 304L SS Brass
Changedto: ....
Changedto: ....
LP FWH Tubing Material (Original) Brass + 304L SS Brass + 304L

Changedto: - --

Changedto:
Reheater Tubing Material (MSR)(Original) 304L 304L
Changed to:
Type of Cooling Water Fresh (River) Fresh + Cooling Tower
B-1 14 Steam Generator Reference Book, Revision I
Reactor Name St. Laurent B2 St. Lucie 1
Operating Utility Name EdF FP&L

Country France U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. FRAM CE
Net Power 880 839
Gross Mega Want Elec. Rated Power MWe 921 872
Changedto:
Changedto:

Steam Gener. Model Number 51M 67
Condensate Purification Method None Powdered Resin SU Only
Tube Support Configuration Drilled Hole Dr. Hole/Eggcrate
Tube Expansion Method Full Depth Roll + Kiss Full / Explosive
Main Condenser Tube Sheet Material (Orig.) Carbon Steel Aluminum Bronze
Changedto:
Main Condenser Tubing Material (Orig.) Brass Cu Alloy
Changedto: Titanium
Changedto:
Main Condenser Deaerating (Y/Nl Yes Yes
LP FWH Tubing Material (Original) Brass + 304L Cu Alloy
Changedto: 304SS
HP FWH Tubing Material (Original) Carbon Steel Cu Alloy
Changedto: 304SS
Reheater Tubing Material (MSR)(Onginal)
Changedto:
304L
---
Fresh + Cooling Tower
CuNi 90/10
Ferrous
Seawater
0
Reactor Name St. Lucie 2 Stade
Operating Utility Name FP&L KKS

PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. CE KWU
Net Power 839 640
Changed to: ..
Changed to: ..
Phosphates Added? No Yes->No
Condensate Purification Method Powdered Resin SU Only None
Tube Support Configuration Dr. Hole/Eggcrate Grids/Strip (See Note 3)
Tube Expansion Method Full / Explosive Mech. Roll-Top & Bottom
Main Condenser Tube Sheet Material (Orig.) Muntz (See Note 6)
Changed to: CS / SS-clad

Changed to: -- Stainless Steel
Changed to: ....
LP FWH Tubing Material (Original) 304 SS Carbon Steel
Changed to: -...
HP FWH Tubing Material (Original) 304 SS Carbon Steel

Changed to: -- -

Changed to: - --

B-i 16 Steam Generator Reference Book, Revision 1
Reactor Name Summer Surry 1 (ORIG) Surry 1 (REPL)
Operating Utility Name SCE&G Virginia Power Virginia Power

Nuclear Steam System Supply Manuf. W W W
Initial Critical Date or Date of Major, Change 10/15/82 7/15/72 1981
Net Power 895 788 788
Cold Leg Prim. Cool. Temp. 100% Power OF 556 536 543
Hot Leg Prim. Cool. Temp. 100% Power OF 619 605 605
Changedto: .....
Changed to: .....
Steam Gener. Oper. Temp. IF 547 516 516
Steam Gener. Model Number D3 51 51F
Phosphates Added? No Yes->No No
On-Line Boric Acid Added? Yes No No
Condensate Purification Method Partial / Powdex None Full / Deep Bed
Tube Support Configuration Drilled Hole Drilled Hole Broached Hole
Tube Expansion Method Full Depth Roll Part Depth Roll Hydraulic
Main Condenser Tube Sheet Material (Orig.) Carbon Steel Muntz Muntz
Changed to: .....
Main Condenser Tubing Material (Orig.) Stainless Steel Cu Alloy Titanium
Changed to: ....
Changed to: ......
Deaerator Other Than Condenser (Y/N) Yes No No
LP FWH Tubing Material (Original) 304SS CuNi 90/10 CuNi 90/10
Changed to: ....
HP FWH Tubing Material (Original) 304 SS CuNi 80/20 CuNi 80/20
Changed to: ....
Reheater Tubing Material (MSR)(Original) CuNi 90/10 CuNi 90/10 CuNi 90/10
Changed to: .....
Type of Cooling Water Fresh (Reservoir) Brackish Brackish
Reactor Name Surry 2 (ORIG) Surry 2 (REPL)
Operating Utility Name Virginia Power Virginia Power

Country U.S. U.S.
PWR or PHWR PWR PWR
Initial Critical Date or Date of Major Change 3/15/73 1980
Net Power 788 788
Changedto:
Changedto:
Phosphates Added? Yes->No No
Condensate Purification Method None Full / Deep Bed
-Steam Generator Tube Sheet Material Forged LA Steel Forged LA Steel
Tube Support Contiguration Drilled Hole Broached Hole
Tube Expansion Method Part Depth Roll Hydraulic
Changedto:
Main Condenser Tubing Material (Orig.) Cu Alloy Titanium
Changedto:
Changedto:
LP FWH Tubing Material (Original) CuNi 90/10 CuNi 90/10
Changedto:
HP FWH Tubing Material (Original) CuNI 80/20 CuNI 80/20
Changedto:
Changedto:
B-1 18 Steam Generator Reference Book, Revision 1
Reactor Name Takahama 1 Takahama 2
Operating Utility Name Kansai Electric Kansai Electric

Country Japan Japan
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W MHI
Net Power 780 780
Changed to: -
Changedto -
Phosphates Added? Yes No
Condensate Purification Method Partial 10% Partial 10%
Tube Expansion Method Part Depth + Remedial Exp. Part Depth + Remedial Exp.
Changedto: ....
Changed to: ....
Changedto:
LP FWH Tubing Material (Original) Aluminum Brass Aluminum Brass
Changedto: ..

Changed to: ....
Changed to: --

Appendix B-Plant Design Characteristics B-1 19
Reactor Name Takahama 3 Takahama 4
Operating Utility Name Kansai Electric Kansai Electric

Country Japan Japan
PWR or PHWR PWR PWR
Net Power 830 830
Changed to: 601 601
Changed to: -- -
Steam Gener. Model Number 51F 51F
Tube Expansion Method Full Depth Roll + Hydraulic Full Depth Roll + Hydraulic
Changed to: ---..

Changedto: ....
Changedto: ....
LP FWH Tubing Material (Original) Al. Brass + SS Al. Brass + SS
Changedto: ..
Changedto: ---...

Changed to: ---

Reactor Name Three Mile Island 1

Country
PWR or PHWR
GPU
U.S.
PWR
0
Nuclear Steam System Supply Manuf. B&W
Net Power 786

Changed to:
Changed to:
Steam Gener. Model Number OTSG 177
Condensate Purification Method Full / Powder
Steam Generator Tubing Material 600SR (See Note 2b)
Tube Expansion Method Part Depth +(Note 1)
Changed to:
Main Condenser Tubing Material (Ong.) 304SS
Changed to:
Changed to:
LP FWH Tubing Material (Original) Carbon Steel / SS
Changed to:
Changed to:
Reheater Tubing Material (MSR)(Original) (No Reheaters)

Changed to:
Type of Cooling Water Fresh (River) Cooling Tower
Reactor Name Tihange 1 Tihange 2
Operating Utility Name ELECTRABEL(See Note 15) ELECTRABEL

Country Belgium Belgium
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. ACLF FRAMECECO
Net Power 870 902
Changedto:
Changedto: ..
Changedto: Titanium 316 SS
Changedto: Titanium 316L SS
Changedto: --
Changed to: SS partially ---
Changed to: ....
Changed to: ....
Type of Cooling Water Fresh (River)/Nat. Tower Fresh (River)/Nat. Tower
Reactor Name Tihange 3 Tomari 1
Operating Utility Name ELECTRABEL Hokkaido Electric

Country Belgium Japan
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. ACECOWEN MHI
Net Power 1006 550
Changed to: -- 601
Changed to: --.--

Steam Gener. Model Number E 51F
Condensate Purification Method None Full / Mixed Bed
Steam Generator Tube Sheet Material FLAS + 600 Clad (See Note 14) Forged LA Steel
Tube Expansion Method Full Depth Roll + Kiss Full / Hydraulic
Main Condenser Tube Sheet Material (Orig.) CS + 304 Cladding Titanium
Changedto: --
Main Condenser Tubing Material (Orig.) 316L SS Titanium

Changedto: --
Changedto: --

LP FWH Tubing Material (Original) Stainless Steel Al. Brass + SS
Changedto: ..
HP FWH Tubing Material (Original) Carbon Steel CuNi 80/20

Changedto: ..
Reheater Tubing Material (MSR)(Original) Stainless Steel CuNi 90/10
Changedto: .....
Type of Cooling Water Fresh (River)/Nat. Tower Seawater
Reactor Name Tomari 2 Tricastin 1
Operating Utility Name Hokkaido Electric EdF

Country Japan France
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. MHI FRAM
Net Power 550 915
Changedto: 602 --
Changedto: --..

Steam Gener. Model Number 51F 51M
Condensate Purification Method Full / Mixed Bed None
Main Condenser Tube Sheet Material (Orig.) Titanium Carbon Steel
Changedto: ..
Main Condenser Tubing Material (Orig.) Titanium Brass
Changedto: ---..
Changedto: ---..

LP FWH Tubing Material (Original) Al. Brass + SS Brass + 304L SS
Changed to: ......
HP FWH Tubing Material (Original). CuNi 80/20 Carbon Steel

Changed to: ---..

Changed to: ...
Reactor Name Tricastin 2 Tricastin 3

PWR or PHWR PWR PWR
Net Power 915 915
0
Changedto: - -
Changedto: -
Steam Gener. Model Number 51M 51M
Changedto: ....
Changedto: ....
Changedto: ....
LP FWH Tubing Material (Original) Brass + 304L 304L
Changedto: ....
Changed to: ..
Changedto: ---...
Type of Cooling Water Fresh (River) Fresh (River)

Reactor Name Tricastin 4 Trillo 1
Operating Utility Name EdF A. Central Trillo

Country France Spain
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. FRAM KWU/ENSA
Net Power 915 1000
Changedto:
Changedto: - -
Steam Gener. Model Number 51M 54GT
Condensate Purification Method None Mech. (100%)
Main Condenser Tube Sheet Material (Orig.) Carbon Steel C-Steel / SS-clad
Changed to: .---.
Main Condenser Tubing Material (Orig.) Brass Titanium
Changed to: ---
Changed to: ......
Main Condenser Deaerating (¥INI Yes Yes
LP FWH Tubing Material (Original) Brass + 304L Carbon Steel
Changed to: ....
Changed to: ---
Changed to: .....
Type of Cooling Water Fresh (River) Fresh (River)/Closed Towers
SEPRI Licensed Material
Reactor Name Trino Vercellese (end 6/90) Trojan (end 11/92)
Operating Utility Name ENEL Portland Genrl Elec.

Country Italy U.S.
PWR or PHWR PWR PWR
Net Power 260 1130
Changed to:
Changed to:
Steam Gener. Model Number 14 51A
Condensate Purification Method Other Powdex, bypassed
Steam Generator Tubing Material 304 SS 600MA
Tube Expansion Method Full Depth Roll Full / Explosive
Changed to:
Main Condenser Tubing Material (Orig.) Multiple Admiralty Brass
Changed to:
Changed to:
LP FWH Tubing Material (Original) Admiralty Brass Admiralty
Changed to: -- 304SS
HP FWH Tubing Material (Original) Admiralty Brass CuNi 90/10
Changed to:
Chanlged to:
i Type of Cooling Water Fresh (River) Fresh (Nat. Tower)

Reactor Name Tsuruga 2 Turkey Point 3 (ORIG)
Operating Utility Name JAPC FP&L

Country Japan U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. MHI W
Net Power 1115 666
Changedto: ......
Changedto: ..
Steam Gener. Oper. Temp. IF 530 <506
Steam Gener. Model Number 51FA 44
Condensate Purification Method Full / Mixed Bed None
Tube Expansion Method Full / Hydraulic Part Depth Roll
Main Condenser Tube Sheet Material (Orig.) Titanium Muntz
Changedto: ....
Main Condenser Tubing Material (Orig.) Titanium Cu Alloy
Changedto:
Changedto:
LP FWH Tubing Material (Original) Stainless Steel Cu Alloy
Changedto: ....
HP FWH Tubing Material (Original) Stainless Steel Cu Alloy
Changedto:
Reheater Tubing Material (MSR)(Original) CuNi 90/10 Cu Alloy
Changedto: ....
Type of Cooling Water Seawater Closed (Pond)/Seawater
Reactor Name Turkey Point 3 (REPL) Turkey Point 4 (ORIG)
Operating Utility Name FP&L FP&L

Country U.S. U.S.
PWR or PHWR PWR PWR
Net Power 666 666
Changed to: --
Changedto: ---
Steam Gener. Oper. Temp. OF <506 516

Steam Gener. Model Number 44F 44
Condensate Purification Method Powdered Resin Demin. None
Changedto: --
Main Condenser Tubing Material (Orig.) Titanium Cu Alloy

Changed to:
Changed to: --

LP FWH Tubing Material (Original) 304 SS Cu Alloy
Changed to: --
HP FWH Tubing Material (Original) 304L SS Cu Alloy

Changed to:
Reheater Tubing Material (MSR)(Original) Stainless Steel Cu Alloy
Changed to: ....
Type of Cooling Water Closed (Pond)/Seawater Closed (Pond)/Seawater

Reactor Name Turkey Point 4 (REPL) Ulchin 1
Operating Utility Name FP&L Korea Electric Pwr

Country U.S. Korea
PWR or PHWR PWR PWR
Net Power 666 943
0
Changed to:
Changed to:
Steam Gener. Oper. Temp. OF 516. 523
Steam Gener. Model Number 44F 51B
Phosphates Added? .No No
Condensate Purification Method Powdered Resin Demin. Deep Bed
Steam Generator Tube Sheet Material Forged LA Steel Stainless Steel
Support Plate Material 405 SS 13% Cr. Steel
Changed to:
Changed to:
Changed to:
Changed to:
HP FWH Tubing Material (Original) Stainless Steel LA Steel

Changed to:

Changed to:
Type of Cooling Water Closed (Pond)/Seawater Seawater
Reactor Name Ulchin 2 Unterweser
Operating Utility Name Korea Electric Pwr KKU

Country Korea Germany
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. FRAM KWU
Net Power 943 1255
Changed to: -
Changed to: -
Steam Gener. Model Number 51B
Steam Generator Tube Sheet Material Stainless Steel Forged LA Steel
Support Plate Material 13% Cr. Steel Austenitic SS
Tube Support Configuration Broached Hole Grids/Strip (See Note 3)
Main Condenser Tube Sheet Material (Ong.) Aluminum Bronze Muntz (See Note 7)
Changed to: -- CS / Ti-clad
Changed to: -- Titanium
Changed to: ....
Changed to: ....
HP FWH Tubing Material (Original) LA Steel Carbon Steel
Changed to: .....
Changed to: ....
Type of Cooling Water Seawater Brackish
Appendix B-Plant Design Characteristics B- 131
Reactor Name Vandellos 2 Vogtle 1
Operating Utility Name ANV Georgia Power Co.

Country Spain U.S.
PWR or PHWR PWR PWR
Net Power 960 1157
Changed to: ..
Changedto: ....
Condensate Purification Method None Full / Powdex
Main Condenser Tube Sheet Material (Orig.) Admiralty Brass Aluminum Bronze
Changedto: ..
Changedto: ..
Changedto: ..
Changed to: ....
Changedto: ---..
Reheater Tubing Material (MSR)(Original) Ferritic SS Carbon Steel

Changedto: ......
Type of Cooling Water Seawater Closed (Nat. Tower)
Reactor Name VogUe 2 Waterford 3
0
Operating Utility Name Georgia Power Co. Louisiana P & L
Country U.S. U.S.
PWR or PHWR PWR PWR
Nuclear Steam System Supply Manuf. W CE
Net Power 1157 1104
Changed to: --- 604
Changedto:

Condensate Purification Method Full / Powdex Powdex-Full Flow
Tube Support Configuration Broached Hole Eggcrate / Vertical
Main Condenser Tube Sheet Material (Orig.) Aluminum Bronze Carbon Steel
Changed to:
Main Condenser Tubing Material (Orig.) Titanium 304SS
Changed to:
Changed to:
LP FWH Tubing Material (Original) Stainless Steel 304SS
Changed to:
HP FWH Tubing Material (Original) Stainless Steel 304SS
Changed to:
Reheater Tubing Material (MSR)(Original) Corten CuNi

Changed to: --- 439SS (in 4/94)
Type of Cooling Water Closed (Nat. Tower) Fresh (River)

Reactor Name Watts Bar 1. Watts Bar 2
Operating Utility Name TVA TVA

Country U.S. U.S.
PWR or PHWR PWR PWR

Initial Critical Date or Date of Major Change Scheduled: 6/94 Scheduled: 6/94
Net Power 1177 1177
Changedto: --- --
Changedto: ...--
Steam Gener. Oper. Temp. 'F

Steam Generator Tubing Material 600MA . 600MA
I Changedto:
Changed to: ..
Changed to: ....

LP FWH Tubing Material (Original) CuNi CuNi
Changed to: ....
HP FWH Tubing Material (Original) CuNi CuNi
Changed to: ....
Reheater Tubing Material (MSR)(Original) CuNi 'CuNi
Changed to: ..
Type of Cooling Water Closed (Tower) / River Closed (Tower) / River
Reactor Name Wolf Creek 1 Wolsong 1
Operating Utility Name WCNOC Korea Electric Pwr

Country U.S. Korea
Nuclear Steam System Supply Manuf. W AECL
Initial Critical Date or Date of Major Change CO:8/85 11/15/82
Net Power 1150 629

Changed to: --...
Changed to: ......

Steam Gener. Model Number F Fos/Wh(CANDU)

Condensate Purification Method Partial / Deep Bed None
Support Plate Material 405 SS Inconel 600
Tube Support Configuration Broached Hole Eggcrate Lattice
Tube Expansion Method Full / Hydraulic

Changed to: ---..
Main Condenser Tubing Material (Orig.) 304 SS Titanium
Changed to: ......
Changed to: ---..
Main Condenser Deaerating (Y/NI Yes No

Changed to:
HP FWH Tubing Material (Original) 304 SS Carbon Steel
Changedto:
Reheater Tubing Material (MSR)(Originai) Corten Steel Carbon Steel
Changedto:
Type of Cooling Water Fresh (Reservoir) , Seawater
Appendix B-Plant Design Characteristics - 0-135
Reactor Name Yankee Rowe (end 2/92) Yonggwang 1
Operating Utility Name Yankee Atomic Korea Electric Pwr

Country U.S. Korea
PWR or PHWR PWR PWR
Net Power 175 900
Changed to: 550 --
Changed to: ---

Steam Gener. Oper. Temp. °F 527 550.4
Steam Gener. Model Number 13 F
Condensate Purification Method None Deep Bed
Steam Generator Tubing Material 304 SS 600TT
Tube Expansion Method Full Depth Roll Full / Hydraulic
Changed to: -- -
Main Condenser Tubing Material (Orig.) Admiralty Brass Titanium

Changed to: -- -
Changed to: ....

Main Condenser Deaerating (Y/NI Yes No
LP FWH Tubing Material (Original) Admiralty Brass Stainless Steel
Changedto: - -
HP FWH Tubing Material (Original) CuNi 90/10 Stainless Steel

Changedto: - -

Changed to: - -
Type of Cooling Water Closed (Pond) Seawater

Reactor Name Yonggwang 2 Zion 1
Operating Utility Name Korea Electric Pwr Commonwealth Edison

Country Korea U.S.
PWR or PHWR PWR PWR
Net Power 900 1040
Cold Leg Prim. Cool. Temp. 100% Power OF 555.8 530.0
Hot Leg Prim. Cool. Temp. 100% Power °F 618.8 591.4
Changed to: ..
Changed to: ......
Steam Gener. Oper. Temp. OF 550.4 506
Main Condenser Tube Sheet Material (Orig.) Aluminum Bronze Muntz
Changedto:---
Main Condenser Tubing Material (Orig.) Titanium 304SS
Changedto: ..
Changedto: ....
Main Condenser Deaerating (Y/NI No Yes
LP FWH Tubing Material (Original) Stainless Steel 304 SS
Changedto: ....
HP FWH Tubing Material (Original) Stainless Steel 304 SS
Changedto:
Reheater Tubing Material (MSR)(Original) Carbon Steel CuNi 90/10,439 SS
Changedto: --
Type of Cooling Water Seawater Fresh (Lake)
Reactor Name Zion 2 Zorita
Operating Utility Name Commonwealth Edison UEFSA

Country U.S. Spain
PWR or PHWR PWR PWR
Net Power 1040 153
Cold Leg Prim. Cool. Temp. 100% Power OF 530.0 537
Hot Leg Prim. Cool. Temp. 100% Power OF 591.4 584
Changedto: - --
Changed to: - --

Phosphates Added? Yes->No Yes
Changedto: - --
Main Condenser Tubing Material (Orig.) 304SS CuNi 90/10

Changedto: ......
Changedto: ......
LP FWH Tubing Material (Original) Stainless Steel Admiralty Brass
Changedto: - --
HP FWH Tubing Material (Original) Stainless Steel CuNi 90/10

Changedto: - -
Reheater Tubing Material (MSR)(Original) CuNi 90/10,439 SS CuNi 90/10

Changedto: ....
Type of Cooling Water Fresh (Lake) Fresh (River)
GLOSSARY OF TERMS
AVT: All volatile treatment.

"Bloom"-Type Oxide: A protective oxide layer consisting of magnetite
crystals that form on steam generator heat transfer tubes.
CERL: Central Electricity Research Laboratories.
Cladding: The term "cladding" refers to a corrosion-resistant layer of
material applied to surfaces exposed to primary coolant in channel heads
and nozzles. The purpose of cladding in steam generators is to reduce
pickup of corrosion products by the primary coolant. The cladding is
normally weld-deposited stainless steel on surfaces other than the primary
face of the tubesheet. In the past, tubesheet primary faces have normally
been clad with weld-deposited alloy 600 type material. Explosively
welded material has also been used.
Cold Drawing: Cold drawing is a type of cold reduction used in metal
processing mills for forming metals into desired shapes. It involves
pulling metal parts through dyes in order to achieve the desired shape.,
For steam generator tubing, several stages of cold reduction, such as cold
drawing, are typically used to obtain the final desired diameter and wall
thickness. The tube diameter and wall thickness are reduced by each
stage of cold reduction. A high temperature mill annealing is normally
performed after each cold reduction operation in order to soften and
recrystalize the metal.
DAM (Dudgeonnage Ameliore Mecaniquement) or "Kiss Roll": A short
length of intermediate expansion applied to steam generator heat transfer
tubes above the fully expanded region in the tubesheet. The DAM
treatment is used to reduce OD residual stresses that could have been
caused during the mechanical rolling process.
G-2 Steam Generator Reference Book, Revision 1
EDS: Energy dispersive spectroscopy.

EDX: Energy-dispersive X-ray spectroscopy.
Explansion: An explosive process that expands the tube over the full length
of the tubesheet.
Hoop Stress: Hoop stress refers to stresses acting in the circumferential
direction in a tube or pressure vessel wall.
IGA: Intergranular attack.
IGP: Intergranular penetration.
IGSCC: Intergranular stress corrosion cracking.
LBRB: Leak before risk of break.
Martensite: Martensite is a crystalline phase formed in certain carbon and
alloy steels when they are cooled rapidly. Depending on the amount of
martensite formed in the steel, it can make the steel hard and brittle. The
formation of martensite is partly dependent on the chemical composition
of the steel (especially its carbon content) and the rate of cooling.
Martensite can be tempered or softened by heating the steel to an
appropriate temperature. Tempering reduces the steel's hardness and
residual stresses, and increases its ductility.
MEA: Materials Engineering Associates.
Mill Annealing (MA): Mill annealing is a high temperature treatment given
to steam generator tubes in a factory or mill. Mill annealing is generally,
performed using continuous furnaces through which the tubes are passed,
such that each portion of each tube reaches the desired high temperature
for a short period of time. Generally, mill annealing is performed after
each stage of cold working. The mill annealing performed after the last
stage of cold work is known as the "mill anneal" or the "final mill anneal."
Mixed Mode Cracking: A combination of intergranular stress corrosion
cracking (IGSCC) and transgranular stress corrosion cracking (TGSCC).
NPM: Non-protective magnetite.
Pilling-Bedworth Ratio (P-B ratio): The ratio of "oxide volume produced"
to "volume of metal corroded."
Plastic Strain: Plastic strain is non-recoverable deformation of a part that
remains after the load causing the strain is removed.
PWSCC: Primary water stress corrosion cracking.
Glossary G-3
Repassivation: Repassivation is a process in which oxide films on metal

surfaces are restored to reduce susceptibility to corrosion following a
process that removed the oxide film. (Removal of oxide films has the
potential of increasing susceptibility to corrosion in some environments.)
Various processes are used for repassivation, such as low temperature
soaks in chemically treated water, exposure to high temperature steam
environments, and exposure to high temperature chemically treated water.
RUBs (Reverse U-Bends): RUBs are stress corrosion test specimens made
by splitting a tube length in half along its centerline, then bending the
split tube around a mandrel such that the inside of the tube is put into
tension and the two legs are brought parallel to each other. The two legs
of the RUB are held in the strained position by bolting. RUBs are normally
tested by placing them in a certain environment, and periodically
removing them for inspection to determine if cracks have developed.
SCC: Stress corrosion cracking.
SEM (Scanning Electron Microscope): A scanning electron microscope
is an instrument used for examination of parts at high magnification. Its
depth of field is much larger than that of an optical microscope's, which
makes it useful for examination of rough surfaces, such as those often
produced by corrosion. SEMs are used extensively to examine portions
of corroded tubes removed from steam generators in order to characterize
the failure morphology.
Sensitization: When chromium carbides form at grain boundaries, the
chromium concentration is depleted locally near the grain boundaries, a
condition known as sensitization. The sensitized region has poorer
resistance to intergranular corrosion under acidic oxidizing conditions.
Sensitization can be eliminated by thermal treatment, which amounts to
holding the material in the sensitizing or carbide precipitation temperature
range long enough for chromium to diffuse back to the grain boundaries.
Single Tube Model Boiler (STMB): Atest device used to test for corrosion
of steam generator tube, tubesheet, and tube support materials under
heat flux conditions, with sludge at the top of the tubesheet crevice.
Superheat: Difference between local primary coolant temperature and
saturation temperature at steam generator secondary side pressure.
Tensile Hoop Stress: Circumferential stress on tubes.
Tensile Stress: Tensile stress is the type of stress that tends to pull the
material apart.
TGSCC: Transgranular stress corrosion cracking.
G-4 Steam Generator Reference Book, Revision 1
Thermal Treatment (TT): Thermal treatment is a fabrication step in which

parts are heated to moderately high temperatures (e.g., 700°C) for
extended periods of time to achieve desired tube material properties. It
is used for alloy 600 and 690 tubes, plugs, and sleeves. It is intended to
develop a microstructure with continuous or near continuous layer of
chromium carbide precipitates at grain boundaries, thereby increasing
resistance to stress corrosion cracking in caustic and primary water
environments. Thermal treatment is intended to reduce residual stresses
and surface cold work.
Tubesheet Crevice Simulation Apparatus (TSCA): Amulti-tube, model-
boiler-type test device used to test for corrosion of steam generator tube
and tubesheet materials under realistic heat flux conditions, with sludge
at the top of the tubesheet crevice.
Volatility: Tendency to distribute to steam rather than liquid phase during
boiling.
Wastage: A form of generalized corrosion in which metal is more or less
uniformly removed from a steam generator heat transfer tube (often
referred to as thinning).
Weibull Plot: A Weibull plot is a type of statistical graph that shows the
cumulative fraction of a group of items that failed, as a function of time.
The graph has special coordinates that often result in the cumulative
failure data for many failure processes plotting along straight lines. This
is often the case for corrosion failures of steam generator tubes. As a
result, Weibull plots are a convenient means to evaluate and project failure
processes.
INDEX
A boric acid 3-21, 3-24, 11-14, 12-4,

12-36, 12-68, 13-14
acid chloride 3-21, 8-47, 10-52, Borssele 9-4
10-54, 10-68 brazed joints 25-49
alkaline 8-11, 11-1 bromide 8-35
alloy 600 3-22, 8-2, 12-13,
12-39, 13-1, 15-11, 15-17, C
24-2, 25-8 Calvert Cliffs 10-2, 10-5, 10-23
alloy 690 8-2, 12-13, 24-2, 24-47 CANDU 3-26, 8-2
alloy 800 8-2, 9-11, 24-2 carbides 13-41
Almaraz 1 12-4
carbon 7-39
alternate tube repair limits 4-15
caustic 13-1, 24-12
ammonia 3-21
CECIL 4-26
anti-vibration stakes 16-13 chemical cleaning 4-26, 25-53
apex 7-27 chlorides 8-18, 8-22, 24-4, 24-53
AVB 16-2
chromium 8-52, 10-13, 10-29, 13-42
AVT 9-2, 9-11, 12-1
circumferential cracking 2-11, 7-24
B clad separation 2-9
cleanliness 4-14, 4-22
balance-of-plant 2-15 cold work 12-17
batwing supports 16-2 condensate polishers 4-31, 12-35
Beaver Valley 12-38 condensers 2-18, 4-33
Beznau 9-9, 12-31, 12-32 Connecticut Yankee 9-9, 10-2,
"Bloom"-type oxide 8-11 10-20, 16-6
blowdown 2-20, 4-10, 4-30 copper 8-18, 8-30, 8-48, 10-9, 10-19,
bobbin coil 7-58 10-54, 15-11
1-2 SteamGenerator Reference Book, Revision 1
corrosion 4-39, 8-21, 8-47, 16-1, G

16-21
corrosion kinetics 8-11, 11-6 galvanic corrosion 24-63
cracking 2-8, 4-19, 7-27, 15-12 galvanic protection 10-26
crevice chemistry 8-14 Genkai 12-9
crevice flushing 4-25 girth weld cracking 4-19
crevices 2-13, 12-3, 15-12 global heat treatment 25-32
grain boundary 13-38, 13-40
D guidelines 4-39, 8-51
D.C. Cook 11-2 H

DAM 7-15, 25-9
Dampierre 7-26 H. B. Robinson 12-32
dealloying 13-42 heat flux 8-18
denting 2-7, 3-21, 8-1, 11-1 heat treatment 13-30, 25-24, 25-27
Doel 7-6, 12-26 hematite 8-5
hideout return 4-24
E hydraulic expansion 7-17, 25-9
hydrazine 3-21, 8-18
eddy current testing 4-30, 7-58, hydrogen 7-50, 25-52
9-2, 10-70, 15-16, 16-27
electrochemical potential 7-54, I
11-13, 12-17, 12-40, 13-27
electroplating 25-34 IGA/IGSCC 13-36
environment 7-50, 12-67 IGA/SCC 11-3
erosion-corrosion 2-13 impurity ingress 12-29
expansion transition 7-14, 7-45, impurity transport 4-24
7-56, 25-6 Indian Point 3 10-2, 10-10
explansion 7-17 inhibitors 12-70
explosive expansion 7-17, 25-38 inleakage 8-19, 8-49, 12-69
intergranular attack (IGA) 2-8,
F 3-23, 12-1
intergranular corrosion (IGC)
F* 25-5 12-4, 13-1, 24-39
fabrication 2-14, 4-12 intergranular penetration (IGP)
fatigue 2-10, 3-25 12-11, 13-2
flow-induced vibrations 16-11 intergranular stress corrosion
fluoride 8-35 cracking (IGSCC) 3-23,
Fort Calhoun 11-2 8-8, 12-10, 13-2, 15-11
free span 12-58 iron 10-10
freshwater plants 12-42,
12-53, 12-54
fretting 2-9, 3-26, 16-19, 16-29
full-depth expansion 25-9
1-3
J 0
Japan 12-9,12-23,12-48, Ohi 12-9
12-56, 12-67 once-through steam generator
Jos6 Cabrera 9-3, 12-32 (OTSG) 1-3, 2-4, 2-11, 7-3,
12-60, 16-29
K ovality 7-30
Ko-Ri 10-2 oxidants 12-17
KWU 9-3, 9-11 oxide film 13-37
oxide morphology 8-11
L oxides 13-16
oxidizing conditions 10-69, 12-70
layup 4-29, 10-69 oxygen 8-17
LBRB 25-7
lead 3-24, 15-1, 24-36 P
leak testing 7-59
lithium 7-53, 24-29 P-B ratio 8-6
P* 25-5
longitudinal cracks 7-17
Palisades 9-9
M Palo Verde 12-22
part-depth expansion 12-2, 25-9
magnesium chloride 25-12 phosphate 9-1
magnetite 8-3, 8-32, 9-22 pit initiation 10-59
makeup water 2-20 pit propagation 10-59
McGuire 12-22 pitting 2-7, 3-25, 10-1, 12-48, 24-41
mechanical plugs 25-38 plugging 25-38
mechanical rolling 7-15 Point Beach 11-3, 12-33
memory metal plugs 25-39 Point Lepreau 9-3
microstructure 12-66, 13-30 polythionic acid 25-12
Mihama 1 9-9 potassium 11-6
mill annealing 7-37 potassium hydroxide 3-23
Millstone 10-2, 10-15 potassium phosphate 9-16
mixed mode attack 12-12, 15-11 Prairie Island 12-33
moisture separator 2-3, 2-23 preheater 16-27
molar ratio 9-4, 9-12, 12-31, 12-70 pressure drop 2-14
molybdenum 24-42 pressurized water reactor (PWR)
morpholine 12-38 1-2
MULTEQ 8-14, 12-31, 12-50, 16-25 prestrain 13-23
primary water stress corrosion
N cracking (PWSCC) 3-22,
nickel 8-48, 10-10 7-1, 25-1
nondestructive testing 4-30
nonprotective magnetite 8-20, 8-52
R Steam Generator Reliability Project'

(SGRP) 3-14
re-expansion 7-15, 25-37 steam pressure decrease 2-10
recirculating steam generator stress 7-45, 12-17, 12-66, 13-23
(RSG) 2-4 stress corrosion cracking (SCC)
regulatory requirements 25-2 2-8, 7-1, 8-8, 12-1, 24-4
remedial measures 12-65, 15-16, stress relief 4-17
16-12, 16-26, 25-1, 25-8 sulfate 8-34, 24-33, 24-62
residual stress 25-31 sulfur 24-35
resins 9-2 superheat 8-16, 8-45, 12-16, 12-50
Ringhals 12-19, 12-33, 12-48 support plate 4-4, 8-3
roll expansion 7-15 Surry 10-2, 10-20
rotating pancake coil (RPC) 7-58
rotopeening 25-15 T
S Takahama 7-6, 12-9
temperature 7-50, 8-30, 10-57,
San Onofre 9-3, 11-3, 12-32, 16-6 10-69, 12-16, 12-19, 12-68,
scale 9-3, 10-2, 10-52, 10-68 13-19, 25-14, 25-31, 25-50
seawater plants 12-47, 12-53, 12-56 tensile hoop stress 8-7
sensitization 7-43, 12-34, 13-40, transgranular stress corrosion
24-36, 25-24 cracking (TGSCC) 15-10
shell 2-10, 4-11 thermal treatment 7-43, 24-36,
shot peening 25-19 25-24
silica 13-14 thinning 2-5, 9-2, 16-14, 16-26
silicon 8-52, 10-19 Thot reduction 4-27
sleeving 3-22, 10-70, 25-41 Trojan 10-2, 10-20
sludge 3-24, 9-11, 10-2, 10-52, tube fabrication 4-6
10-68, 13-16 tube materials 4-5, 24-2
sludge lancing 4-26 tube support crevices 12-3, 12-26,
sludge piles 9-4, 12-3, 12-69 13-34
soaking 12-69 tube support plate (TSP) 11-1,
sodium 3-20, 9-2, 9-12, 9-17, 11-6, 12-59, 16-1, 24-51, 25-7
25-12 tube wall loss 16-3
spinel 8-32 tubesheet 12-59, 24-51
St. Lucie 10-2 tubesheet crevices 12-2, 12-23,
steam generator design 4-3, 8-2 12-41, 12-59, 12-68, 13-34
steam generator materials 8-2
Steam Generator Owners Group I U
(SGOG I) 3-2
Steam Generator Owners Group II U-bend 7-27, 7-47, 25-10, 25-27
U-bend support 4-7, 12-57
(SGOG II) 3-10
ultrasonic testing 7-59
1-5
V
volumetric IGA 12-10
W
wastage 2-5, 3-20, 9-1, 10-13, 24-43
water chemistry 4-34, 12-32, 25-52
water hammer 4-13
wear 16-1
welded plugs 25-38
z
zinc 16-21
Zorita 16-6

EPRI Steam Generator Reference Book PDF

Uploaded by

Copyright:

Available Formats

EPRI Steam Generator Reference Book PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

EPRI Steam Generator Reference Book PDF

Uploaded by

Copyright:

Available Formats

What are some of the operational issues addressed in the document?

What are some of the operational issues addressed in the document?

What approach was taken to organize the research results?

What approach was taken to organize the research results?

Leadership in Electrification

through Global Collaboration R E P 0 R T S U M M A R Y

Steam Generator Reference Book, Revision 1

BACKGROUND In 1977, EPRI established the Steam Generator Project Office in

RESULTS The Steam GeneratorReference Book summarizes options'and rec-

EPRI PERSPECTIVE The Steam GeneratorReference Book, Revision 1

The Steam Generator Owners Groups I and II

J. Peter N. Paine, Editor

ORGANIZATION(S) THAT PREPARED THIS REPORT

Stanley .J. Green served as Director of the Steam Generator Owners

Authors and EPRI Steam Generator Project Office Staff

New York Power Authority

Steam Generator Owners Group I Executive Committee

W. M. Menger, Chief Consulting Engineer,

Steam Generator Owners Group I Member Representatives

S. Burstein, Executive Vice President,

B. Nilsson, Chief Engineer,

Steam Generator Owners Group I Technical Advisory Committee

C. B. Hardee, Mechanical Equipment Specialist,

A. L. Sudduth, Design Engineer,

Steam Generator Owners Group IIExecutive Committee

L. Rylander, Department Manager,

Steam Generator Owners Group IIMember Representatives

M. Yasuda, Deputy General Director,

D. E. Adams, III, Utility Engineer,

0. W. Dixon, Jr., Vice President, Nuclear Operations,

Steam Generator Owners Group IITechnical Advisory Committee

J. 0. Cermak, Technical Director,

D. R. Hafer, Assistant Manager, Heat Exchangers & Pumps Section,

J. E. Moaba, Section Supervisor, Materials, Codes & Inspection,

A. L. Sudduth, Design Engineer, M&N Division,

Steam Generator Reliability Project Executive Group

R. E. Smith, Senior Vice President, Production & Engineering,

Steam Generator Reliability Project Member Representatives

M. F. Hodge, Mechanical/Chemical Engineering Manager,

H. Takaku, Manager, Mechanical Engineering Section,

R. Garnsey, Manager, Safety & Technology,

J. J. Zach, Senior Manager, Nuclear Engineering,

Steam Generator Reliability Project Technical Advisory Group

J. M. Fackelmann, Principal Engineer,

N. Nilsen, Senior Engineer,

The Electric Power Research Institute and my colleagues there have

Operation and maintenance of PWR Steam Generators with high

The Steam Generator Owners Group II was successfully completed.

Final Reports of Individual Contracts

Technology Transfer Packages

Design and Operation Guidelines to Minimize Steam Generator

Steam Generator Reference Book

has evolved in 20 years. This chapter also contains a discussion of the

2 Steam Generator Performance History 2-1

3 Steam Generator Programs 3-1

Table of Contents iii

Subprogram 0822: Evaluate and Test Boiler

4.2.3 Enhancing Hideout Return Processes 4-24

Table of Contents vii