Boric Acid Corrosion Guidebook,

Revision 1

Managing Boric Acid Corrosion Issues at PWR

Power Stations
Effective December 6, 2006, this report has been made publicly available in
accordance with Section 734.3(b)(3) and published in accordance with Section 734.7
of the U.S. Export Administration Regulations. As a result of this publication, this
report is subject to only copyright protection and does not require any license
SED WARNING:
N agreement from EPRI. This notice supersedes the export control restrictions and any
A L
LICE

Please read the License Agreement

proprietary licensed material notices embedded in the document prior to publication.
on the back cover before removing
R I

M AT E the Wrapping Material.

Technical Report
Boric Acid Corrosion Guidebook,
Revision 1
Managing Boric Acid Corrosion Issues at PWR
Power Stations
1000975

Final Report, November 2001

EPRI Project Managers

J. M. Jenco
M. Bridges

EPRI • 3412 Hillview Avenue, Palo Alto, California 94304 • PO Box 10412, Palo Alto, California 94303 • USA
800.313.3774 • 650.855.2121 • askepri@epri.com • www.epri.com
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT 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) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM:

(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I)

WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR
SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS
FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR
INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL
PROPERTY, OR (III) THAT THIS DOCUMENT 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 OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR
SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD,
PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT.

ORGANIZATION(S) THAT PREPARED THIS DOCUMENT

Dominion Engineering, Inc.

ORDERING INFORMATION
Requests for copies of this report should be directed to EPRI Customer Fulfillment, 1355 Willow Way,
Suite 278, Concord, CA 94520, (800) 313-3774, press 2.

Electric Power Research Institute and EPRI are registered service marks of the Electric Power
Research Institute, Inc. EPRI. ELECTRIFY THE WORLD is a service mark of the Electric Power
Research Institute, Inc.

Copyright © 2001 Electric Power Research Institute, Inc. All rights reserved.
CITATIONS

This report was prepared by:

Dominion Engineering, Inc.

6862 Elm Street, Suite 460
McLean, VA 22101

Principal Author:
E. S. Hunt

This report describes research sponsored by EPRI.

The report is a corporate document that should be cited in the literature in the following manner:

Boric Acid Corrosion Guidebook, Revision 1: Managing Boric Acid Corrosion Issues at PWR
Power Stations, EPRI, Palo Alto, CA: 2001. 1000975.

iii
REPORT SUMMARY

Boric acid corrosion (BAC) represents a significant maintenance concern at many pressurized
water reactor (PWR) plants because of the large number of potential leakage sources—flanged
joints, valve packing, mechanical seals, and fittings. This report compiles information that can
help utility staff reduce the potential for leakage, properly and uniformly evaluate individual
incidents, mitigate potential damage, and justify continued operation with leakage when
appropriate. BAC does not represent a significant concern at boiling water reactor (BWR) plants
because they do not use borated primary water to control reactivity during normal plant
operations. Boric acid is used in BWRs as a final option in the form of the standby liquid control
system to control reactor reactivity during a severe accident event.
This latest revision of the Boric Acid Corrosion Guidebook expands on the 1995 version of the
guidebook in the following areas: 1) significant BAC incidents since 1995, 2) results of BAC
testing on prototypical joints, 3) improved methods for reducing leaks based on work performed
by the EPRI Fluid Sealing Technology Program, 4) guidelines for the inspection of flanged joints
in systems containing water borated for the purpose of controlling reactivity, and 5) guidelines
for the disposition of boric acid leaks.

Background
PWRs use borated water in their primary coolant system to control reactor criticality. Under
certain conditions, leakage of borated water from primary system gaskets, seals, valve packing,
and primary water stress corrosion cracking (PWSCC) can result in significant BAC of
associated carbon or low-alloy steel components. Under extreme conditions, BAC can
compromise the integrity of plant primary system components. In 1988, the NRC issued Generic
Letter 88-05 requiring all operators of PWRs to develop and implement programs to address
BAC issues in their plants. Since 1998, EPRI has sponsored several activities to support utility
BAC programs:
• Preparation of the original Boric Acid Corrosion Guidebook in 1995
• Development of programs and supporting technology to help utilities reduce leakage from
primary system joints under the sponsorship of the EPRI Fluid Sealing Technology Program
• BAC testing of prototypical joints conducted by EPRI at Southwest Research Institute
(SwRI) in 1996–98
• Additional work to assist utilities in establishing and managing BAC programs as described
in this revised report

Objective
• To provide a single source of comprehensive information to help utilities address plant BAC
and general leakage reduction issues

v
Approach
To meet the objectives for this effort, plant experience with BAC incidents was identified and
acquired by EPRI through plant surveys and follow-up discussions with plant personnel.
Additionally, industry experience in the topical area, including published information, technical
literature, and regulatory requirements and responses, was identified. Information was sorted,
analyzed, and charted in a manner that clearly correlates various data. Significant new work was
performed involving screening tests of materials in borated water, testing of prototypical joints
with borated water leakage, and documenting methods for reducing leaks in various types of
joints.
The guidebook was organized to facilitate a complete and logical understanding of the topic. It
presents background information summarizing the evolution of the BAC issue, field experience
with BAC incidents, a discussion of BAC mechanisms, a summary of prior BAC testing, a report
of the recent EPRI-sponsored supplemental BAC testing, an overview of applicable regulatory
bases, information on detecting and preventing leakage, guidelines for assessing and
dispositioning boric acid leaks, and guidance on management of the BAC issue. Appendices
include detailed reports of individual BAC incidents and example analyses of common corrosion
situations.

Results
The survey of field experience included reviews of over 120 publicly reported incidents
involving corrosion of carbon and low-alloy steel materials by leaking borated water. The
reported incidents were classified by type of component and then evaluated to determine the
typical sources of leakage and the range of potential consequences. Field and laboratory data
indicate predictable corrosion patterns consistent with assumed BAC mechanisms. The key
conclusion from this work is that the primary utility emphasis should be on reducing leakage
rather than on refining methods to deal with the consequences of leakage. Leak reduction efforts
should be directed toward bolted joints and valve packing because they have been the source of
the greatest number of reported incidents. Specific guidance is provided, based on concurrent
EPRI and utility plant leak reduction programs.

EPRI Perspective
The Boric Acid Corrosion Guidebook, Revision 1, provides a valuable compilation of presently
available information on all aspects relating to understanding and managing BAC issues at PWR
plants. Accordingly, this document should serve as an authoritative single-source reference to
help plant engineering, maintenance, and management understand and manage BAC incidents
and issues. Related EPRI research includes EPRI reports NP-3784, NP-5769, NP-5985, NP-
6536, and TR-101108. Reference should also be made to the series of reports issued under the
EPRI Fluid Sealing Technology Program.

Keywords
Maintenance
Corrosion
Boric acid

vi
 EPRI Licensed Material

ABSTRACT

Corrosion of carbon and low-alloy steel components by leaking borated water has posed
significant maintenance problems for many PWR plants. Two incidents illustrate the potential
importance of this problem.

In 1980, leakage from the gaskets of two reactor coolant pumps at one plant resulted in severe
corrosion to seven coolant pump flange studs. The diameter of the worst case stud was reduced
from its original 3.5 inches (89 mm) to 1.0–1.5 inches (25–38 mm). This represents a reduction
to less than 20% of the original stud cross-sectional area. In 1986, leakage from a valve body-to-
bonnet gasket at another plant resulted in corrosion that extended two-thirds of the way through
the wall thickness of a low-alloy steel nozzle in the main coolant piping system.

Subsequent to these and other significant events, the NRC issued Generic Letter 88-05, requiring
operators of PWR-type power plants to develop and implement a plan to ensure that there is an
extremely low probability of abnormal leakage, rapidly propagating failure, or gross rupture as a
result of boric acid corrosion (BAC) of primary coolant loop components.

This report continues EPRI efforts to provide assistance to utilities in addressing the
requirements of NRC GL 88-05 and BAC issues in general. The original Boric Acid Corrosion
Guidebook in 1995 summarized the extent of the BAC problem and compiled data from
previously performed BAC test programs.

The current edition of the Boric Acid Corrosion Guidebook builds on the 1995 edition with
additional information pertaining to:
• Significant boric acid corrosion incidents since 1995
• Results of BAC testing on prototypical joints
• Improved methods for reducing leaks based on work performed by the EPRI Fluid Sealing
Technology Program
• Guidelines for inspecting flanged joints in borated systems
• Guidelines for dispositioning boric acid leaks

The current version of the Boric Acid Corrosion Guidebook covers the following main topics:
• Field experience with leakage of borated water and the resultant corrosion
• Mechanisms by which borated water corrodes carbon and low-alloy steel materials

vii
 EPRI Licensed Material

• Results of boric acid corrosion testing including recent EPRI-sponsored material screening
and prototypical joint tests
• Regulatory bases
• Borated water leak detection
• Leakage and degradation prevention
• Development of justifications for continued operation (JCOs) with degraded parts or ongoing
leakage
• Guidance regarding overall programs to deal effectively with this issue

The report also contains several examples of the types of analyses that can be performed to
determine whether it is possible to continue plant operation with certain types of degradation.

viii
 EPRI Licensed Material

ACKNOWLEDGMENTS

As with the preparation of any such document, many individuals and organizations contributed
significantly of their time and expertise. These contributions included assistance in establishing
the document objectives and content, providing reference information, and providing technical
and editorial review. The input of these individuals and organizations is gratefully
acknowledged.

The following individuals and companies deserve special recognition for their technical guidance
and co-funding of the original Boric Acid Corrosion Guidebook and the extensive supplemental
BAC testing described in Section 4 :

Dennis P. Weakland FENOC - A First Energy Company

David E. Whitaker Duke Power Company

Henry J. Thailer Pacific Gas & Electric Company

Special recognition is also given to the utilities that support the EPRI Fluid Sealing Technology
Program. Work sponsored by this program has significantly improved the understanding of the
causes of leakage from commonly encountered types of joints and the approaches that can be
taken to minimize such leakage.

Boric acid corrosion testing of prototypical joints was performed for EPRI by Southwest
Research Institute, San Antonio, Texas. The principal investigator was R. Page.

ix
 EPRI Licensed Material

CONTENTS

1 INTRODUCTION.................................................................................................................. 1-1
1.1 Purpose of the Boric Acid Corrosion Guidebook........................................................ 1-1
1.2 Background............................................................................................................... 1-1
1.3 Approach................................................................................................................... 1-3
1.4 Content/Organization ................................................................................................ 1-3
1.5 Application ................................................................................................................ 1-4
1.6 Benefits..................................................................................................................... 1-4

2 INDUSTRY EXPERIENCE ................................................................................................... 2-1

2.1 Systems Containing Borated Water........................................................................... 2-1
2.2 Overview of Reported Incidents ................................................................................ 2-2
2.3 Primary Sources of Leakage and Direct Consequences............................................ 2-3
2.3.1 Gaskets ................................................................................................................ 2-3
2.3.1.1 Steam Generator and Pressurizer Manways ................................................. 2-3
2.3.1.2 Reactor Coolant Pump Gaskets .................................................................... 2-4
2.3.1.3 CRDM Nozzle Gaskets.................................................................................. 2-6
2.3.1.4 Heat Exchanger Gaskets............................................................................... 2-7
2.3.1.5 Valve Body-to-Bonnet Gaskets...................................................................... 2-7
2.3.2 Valve Packing....................................................................................................... 2-8
2.3.3 Mechanical Seals ............................................................................................... 2-10
2.3.3.1 Instrument Nozzle Conoseals...................................................................... 2-10
2.3.3.2 Reactor Vessel Head Metallic O-Rings........................................................ 2-11
2.3.4 Pressure Boundary Component Leaks ............................................................... 2-11
2.3.4.1 CRDM Nozzle Canopy Seal Welds.............................................................. 2-11
2.3.4.2 Vibratory Fatigue of Small Diameter Lines................................................... 2-12
2.3.4.3 Stress Corrosion Cracking of Alloy 600 Materials ........................................ 2-12

xi
 EPRI Licensed Material

2.4 Other Locations Involving Consequential Damage .................................................. 2-14

2.4.1 Electrical and Control Components..................................................................... 2-14
2.4.2 HVAC Components ............................................................................................ 2-14
2.4.3 Component Supports.......................................................................................... 2-15
2.4.4 Containment Liner .............................................................................................. 2-15
2.4.5 Summary Regarding Other Components ............................................................ 2-15
2.5 Conclusions from Field Experience ......................................................................... 2-15

3 BORIC ACID CORROSION MECHANISMS ........................................................................ 3-1

3.1 Overview of Corrosion Mechanisms .......................................................................... 3-1
3.2 Dominant Boric Acid Corrosion Type – Wastage....................................................... 3-3
3.2.1 Effect of Impurities on Wastage ............................................................................ 3-4
3.2.2 Effect of Oxygen on Wastage............................................................................... 3-4
3.2.3 Effect of Temperature on Wastage ....................................................................... 3-4
3.2.4 Effect of pH on Wastage....................................................................................... 3-4
3.2.5 Effect of Flow Velocity on Wastage....................................................................... 3-5
3.3 Secondary Types of Boric Acid Corrosion ................................................................. 3-5
3.3.1 Galvanic Corrosion............................................................................................... 3-5
3.3.2 Crevice Corrosion................................................................................................. 3-5
3.3.3 Pitting ................................................................................................................... 3-6
3.3.4 Intergranular Corrosion......................................................................................... 3-6
3.3.5 Stress Corrosion Cracking.................................................................................... 3-7

4 BORIC ACID CORROSION TESTS..................................................................................... 4-1

4.1 Sources of Boric Acid Corrosion Testing ................................................................... 4-1
4.1.1 Test Data Available Prior to 1995 ......................................................................... 4-1
4.1.2 Supplemental Testing Conducted by EPRI ........................................................... 4-1
4.2 Material Screening Tests........................................................................................... 4-2
4.3 Immersion of Carbon Steel in Deaerated Borated Water........................................... 4-4
4.4 Immersion of Carbon Steel in Aerated Borated Water ............................................... 4-6
4.4.1 Immersion in Aerated Borated Water at Room Temperature ................................ 4-8
4.4.2 Immersion in Aerated Borated Water at High Temperature .................................. 4-9
4.5 Dripping of Borated Water onto Hot Carbon Steel Surfaces ...................................... 4-9
4.6 Impingement of Borated Water onto Hot Carbon Steel Surfaces............................. 4-12
4.7 Leakage of Borated Water into a Tight Crevice (Nozzle Applications) ..................... 4-13

xii
 EPRI Licensed Material

4.8 Leakage of Borated Water from Flanged Joints....................................................... 4-14

4.9 Summary of Corrosion Rates for Various Situations................................................ 4-16
4.10 Boric Acid Corrosion Test Data Summaries ............................................................ 4-16

5 REGULATORY BASES....................................................................................................... 5-1

5.1 Code of Federal Regulations ..................................................................................... 5-1
5.2 NRC Information and Requirements.......................................................................... 5-2
5.2.1 Documents Notifying Utilities of Potential Problems.............................................. 5-3
5.2.2 Reports of Research and Evaluations................................................................... 5-4
5.2.3 Requirements for Utilities to Take Specific Action................................................. 5-4
5.3 ASME Code Requirements ....................................................................................... 5-7
5.3.1 Requirements for Visual Inspections..................................................................... 5-7
5.3.2 Conditions Requiring Corrective Action ................................................................ 5-8
5.3.3 Required Corrective Action ................................................................................... 5-9
5.3.4 Review of Repair Program.................................................................................... 5-9
5.4 Summary of Requirements........................................................................................ 5-9

6 DETECTING LEAKAGE...................................................................................................... 6-1

6.1 Regulatory Requirements.......................................................................................... 6-1
6.1.1 10CFR50 Appendix B Requirements .................................................................... 6-1
6.1.2 Identified vs. Unidentified Leakage ....................................................................... 6-1
6.1.3 Regulatory Guide 1.45 Guidelines ........................................................................ 6-1
6.2 Technical Specification Requirements....................................................................... 6-2
6.3 Leakage Detection Inside and Outside Containment................................................. 6-3
6.4 Summary of Leakage Detection Methods.................................................................. 6-3
6.4.1 Primary On-Line Leakage Detection Methods ...................................................... 6-3
6.4.2 Other On-Line Leakage Detection Methods.......................................................... 6-4
6.4.3 Leak Detection During Plant Outages.................................................................. 6-5
6.5 Correlation Between Leakage and Boric Acid Deposits............................................. 6-6
6.6 Summary of Leak Detection Methods and Requirements.......................................... 6-6

7 PREVENTING LEAKAGE AND DEGRADATION ................................................................ 7-1

7.1 Preventing Leaks ...................................................................................................... 7-1
7.1.1 EPRI Fluid Sealing Technology Program.............................................................. 7-1
7.1.1.1 Research into Leakage Reduction Technology.............................................. 7-2
7.1.1.2 Working Group Meetings............................................................................... 7-2

xiii
 EPRI Licensed Material

7.1.1.3 Fluid Sealing Technology Program Products................................................. 7-2

7.1.1.4 Plant Leakage Reduction Training Workshops .............................................. 7-3
7.1.1.5 Cost Impact Assessment for Plant External Piping System Leakage............. 7-3
7.1.2 Leakage Reduction Programs .............................................................................. 7-4
7.1.2.1 Prerequisites ................................................................................................. 7-4
Management Support ........................................................................................... 7-4
Program Coordinator ............................................................................................ 7-4
Program Goals ..................................................................................................... 7-5
7.1.2.2 Technical Support Programs ......................................................................... 7-6
Procedures ........................................................................................................... 7-6
Materials............................................................................................................... 7-7
Training ................................................................................................................ 7-8
7.1.2.3 Implementation.............................................................................................. 7-9
7.1.2.4 Follow-Up ...................................................................................................... 7-9
Tracking ............................................................................................................... 7-9
Evaluation............................................................................................................. 7-9
Tagging .............................................................................................................. 7-10
Repair................................................................................................................. 7-10
7.1.2.5 Feedback .................................................................................................... 7-10
Performance Assessment................................................................................... 7-10
Program Improvement ........................................................................................ 7-10
7.1.3 Leak Reduction from Gasketed Joints ................................................................ 7-10
7.1.3.1 Causes of Spiral-Wound Gasket Leaks ....................................................... 7-11
7.1.3.2 Causes of Sheet Gasket Leaks ................................................................... 7-12
7.1.3.3 Required Bolt Preload for Gasketed Joints .................................................. 7-14
7.1.3.4 Achieving Required Bolt Preload ................................................................. 7-14
7.1.3.5 Gasketed Joint Assembly Guidelines .......................................................... 7-15
7.1.3.6 Root Cause Analysis for Problem Joints...................................................... 7-16
7.1.4 Leak Reduction from Valve Packing ................................................................... 7-16
7.1.4.1 How Valve Packing Works........................................................................... 7-17
7.1.4.2 Available Types of Valve Packing................................................................ 7-17
7.1.4.3 Causes of Valve Packing Leaks .................................................................. 7-17
7.1.4.4 Selecting Packing Configurations and Products .......................................... 7-19
7.1.4.5 Gland Follower Preload Requirements ........................................................ 7-19

xiv
 EPRI Licensed Material

7.1.4.6 Valve Packing Assembly Guidelines............................................................ 7-20

7.1.5 Leak Reduction from Compression Fittings ........................................................ 7-21
7.1.6 Leak Reduction from Threaded Pipe Joints ........................................................ 7-21
7.1.7 Leak Reduction from Mechanical Seals.............................................................. 7-22
7.1.8 Leak Reduction from Pressure Boundary C omponents ...................................... 7-22
7.1.8.1 Leaks from Cracks in Alloy 600 Nozzles and J-Groove Welds..................... 7-22
7.1.8.2 Leaks from CRDM Nozzle Canopy Seal Welds ........................................... 7-22
7.2 Preventing Degradation If Leaks Occur ................................................................... 7-23
7.2.1 Low-Alloy Steel Fastener Replacement .............................................................. 7-23
7.2.1.1 Combustion Engineering Evaluation of Alternative Fastener Materials ........ 7-24
7.2.1.2 EDF Evaluation of Alternative Fastener Materials........................................ 7-24
7.2.1.3 Summary Regarding Alternative Fastener Materials.................................... 7-25
7.2.2 Replacement of Other Carbon and Low-Alloy Steel Parts................................... 7-25
7.2.3 Protective Shrouds or Covers ............................................................................. 7-26
7.2.4 Oxygen Limiting Band for Flanged Joints............................................................ 7-26
7.3 Stopping Leaks During Plant Operation .................................................................. 7-27

8 CONTINUED OPERATION WITH LEAKAGE AND/OR DEGRADATION ........................... 8-1

8.1 Dealing with Leaks and Degradation......................................................................... 8-1
8.2 Characterizing Current Leakage and Degradation..................................................... 8-2
8.2.1 Characterizing Current Leakage ........................................................................... 8-2
8.2.2 Defining the Affected Areas/Components ............................................................. 8-2
8.2.3 Characterizing Degradation in Equipment That Has Been Disassembled ............. 8-2
8.2.4 Characterizing Degradation in Equipment That Has Not Been Disassembled ...... 8-3
8.3 Predicting Future Leakage and Degradation ............................................................. 8-5
8.3.1 Establishing the Type of Corrosion Environment .................................................. 8-6
8.3.2 Establishing the Maximum Corrosion Rate for the Environment ........................... 8-6
8.3.3 Evaluating Possible Mitigating Circumstances...................................................... 8-6
8.3.4 Predicting Corrosion at the Next Planned Inspection ............................................ 8-8
8.4 Evaluating Predicted Future Degradation vs. Code Requirements ............................ 8-8
8.4.1 Evaluating Corroded Bolts .................................................................................... 8-9
8.4.2 Evaluating Corroded or Degraded Threads .......................................................... 8-9
8.4.3 Evaluating Corroded Pressure Vessels and Piping............................................. 8-10
8.4.4 Evaluating Flanges and Other Special Purpose Parts......................................... 8-11
8.5 Establishing Supplemental Inspection or Operational Requirements ....................... 8-11

xv
 EPRI Licensed Material

8.6 Defense in Depth .................................................................................................... 8-11

8.6.1 Enhanced Inspections and Heightened Awareness............................................ 8-12
8.6.2 PRA Analyses .................................................................................................... 8-12
8.6.3 Analyzed Accident per FSAR.............................................................................. 8-12
8.6.4 Leak-Before-Break ............................................................................................. 8-12
8.6.4.1 Stress Redistribution from Failed Fasteners ................................................ 8-13
8.6.4.2 Leak Rate Resulting from Failed Fasteners................................................. 8-14
8.6.4.3 Leakage Detection ...................................................................................... 8-15
8.6.4.4 Field Confirmation of Operation with Failed or Missing Fasteners ............... 8-15

9 BORIC ACID CORROSION MANAGEMENT ...................................................................... 9-1

9.1 Key Factors in an Effective Boric Acid Corrosion Management Program .................. 9-1
9.1.1 Leakage Reduction............................................................................................... 9-1
9.1.2 Condition Monitoring............................................................................................. 9-1
9.1.3 Replacement of Key Carbon Steel Parts with Corrosion-Resistant Materials........ 9-2
9.1.4 Planned Response to Leaks................................................................................. 9-2
9.2 Level of Planning Effort ............................................................................................. 9-2
9.2.1 Boric Acid Corrosion Management Plan Based on Engineering Judgement ......... 9-2
9.2.2 Boric Acid Corrosion Management Plan Based on Economic Modeling................ 9-3
9.3 EDF Approach to Leakage and Corrosion Management ........................................... 9-4

10 REFERENCES ................................................................................................................ 10-1

A REPORTED INCIDENTS OF BORIC ACID CORROSION .................................................. A-1

B EXAMPLE ANALYSES.......................................................................................................B-1
Example B-1. Corrosion of the Reactor Vessel Shell at Damaged Cladding .................. B-2
Example B-2. Corrosion of the Reactor Vessel Head..................................................... B-4
Example B-3. Corrosion of Reactor Vessel Head Studs................................................. B-6
Example B-4. Local Corrosion of Reactor Vessel Flange Threads................................. B-9
Example B-5. Corrosion of the Reactor Coolant Pump Flange..................................... B-12

xvi
 EPRI Licensed Material

LIST OF FIGURES

Figure 1-1 Corroded Ft. Calhoun Reactor Coolant Pump Stud (Illustration Created from
Photograph) .................................................................................................................... 1-5
Figure 1-2 Corrosion of ANO-1 High-Pressure Injection Nozzle, Photograph from
Entergy Operations, Inc................................................................................................... 1-6
Figure 2-1 Schematic Diagram of Main PWR Systems Containing Borated Water................ 2-16
Figure 2-2 Distribution of Reported Leakage Incidents by Source ......................................... 2-17
Figure 2-3 Locations of Boric Acid Corrosion in Primary Loop Components.......................... 2-18
Figure 2-4 Typical Steam Generator Manway Closure .......................................................... 2-19
Figure 2-5 Typical Reactor Coolant Pump Casing................................................................. 2-20
Figure 2-6 Typical Bolted CRDM Nozzle (B&W Design)........................................................ 2-21
Figure 2-7 Typical Nuclear Service Valves ............................................................................ 2-22
Figure 2-8 Typical Reactor Head Instrument Nozzle Conoseal Closure ................................ 2-23
Figure 2-9 Corrosion of Reactor Vessel Head at Turkey Point 4 ........................................... 2-24
Figure 2-10 Cross Section Through Typical Reactor Vessel Closure Flange......................... 2-25
Figure 2-11 Typical CRDM Nozzle (Canopy Seal Weld Design)............................................ 2-26
Figure 3-1 Forms of Corrosion ................................................................................................ 3-8
Figure 3-2 Formation of Initial Double Layer During the Onset of Corrosion............................ 3-9
Figure 3-3 Consumption of Electrons During Corrosion .......................................................... 3-9
Figure 3-4 Conversion of Ferrous Ions to Hematite (Rust) .................................................... 3-10
Figure 3-5 Formation of a Continuous Corrosion Product Layer ............................................ 3-10
Figure 3-6 Enhanced Corrosion due to Local Galvanic Cells................................................. 3-11
Figure 3-7 Effect of pH and Oxygen on Corrosion Rate ........................................................ 3-12
Figure 3-8 Galvanic Series of Metals and Alloys ................................................................... 3-13
Figure 3-9 Galvanic Corrosion at Dissimilar Metal Couples................................................... 3-14
Figure 3-10 Crevice Corrosion .............................................................................................. 3-14
Figure 3-11 Crevice and Galvanic Effects Beneath Corrosion-Resistant Cladding ................ 3-15
Figure 4-1 Boric Acid Corrosion Conditions........................................................................... 4-18
Figure 4-2 Results of EPRI Material Screening Tests............................................................ 4-19
Figure 4-3 Summary of Corrosion Test Data ......................................................................... 4-20
Figure 4-4 Matrix of EPRI Flange Joint Tests ........................................................................ 4-21
Figure 4-5 Combustion Engineering and EPRI Nozzle Test Fixtures..................................... 4-22
Figure 4-6 Summary of Maximum Corrosion Rates in CE and EPRI Tests ........................... 4-23

xvii
 EPRI Licensed Material

Figure 4-7 Corrosion Depth in Sectioned EPRI Test Block ................................................... 4-24
Figure 4-8 Corrosion Rate in Aerated and Deaerated Water – Westinghouse Tests ............. 4-26
Figure 4-9 Corrosion Rates Under Low Temperatures – Moscow Power Institute Tests ....... 4-28
Figure 4-10 Corrosion Rates Under High Temperatures – Moscow Power Institute Tests..... 4-28
Figure 4-11 B7 Fasteners in Borated Water Solution – TVA Tests........................................ 4-32
Figure 4-12 Corrosion Rates of Low-Alloy Steels in Aerated Boric Acid – BNL Tests............ 4-37
Figure 4-13 Effect of pH on Corrosion Rate of Carbon Steel at 310°C – Combustion
Engineering Tests ......................................................................................................... 4-41
Figure 4-14 Details of Borated Water Dripping onto an Open Heated Trough – B&W
Trough Tests ................................................................................................................. 4-44
Figure 4-15 Details of Borated Water Dripping onto an Insulated Pipe – B&W Pipe Tests .... 4-47
Figure 4-16 Test Apparatus for Impingement onto Bolts – Combustion Engineering Tests ... 4-50
Figure 4-17 Apparatus for Testing Impingement onto Plates – Combustion Engineering
Tests ............................................................................................................................. 4-53
Figure 4-18 Test Configuration for Determining the Effect of Leakage into an Annulus –
Combustion Engineering Tests...................................................................................... 4-56
Figure 4-19 Corrosion Rates in Aerated Boric Acid – EPRI Tests ......................................... 4-59
Figure 4-20 Corrosion Rates for Borated Water Spray – EPRI Tests .................................... 4-62
Figure 4-21 Corrosion Rates for Steels Buried in Boric Acid Crystals – EPRI Tests.............. 4-65
Figure 4-22 Corrosion Rates in Deaerated Boric Acid – EPRI Tests ..................................... 4-67
Figure 4-23 Drip Test Configurations – EPRI Tests............................................................... 4-69
Figure 4-24 Nozzle Crack Leak Test Configurations – EPRI Tests........................................ 4-71
Figure 4-25 Schematic of the Test Block for Leakage into the Annulus – EPRI Tests ........... 4-72
Figure 4-26 Fixture for Flange Leakage Tests – EPRI Tests ................................................. 4-76
Figure 4-27 Flanges Test Conditions – EPRI Tests............................................................... 4-77
Figure 4-28 Corrosion Rates for Bolts Around Uninsulated Flanges at 600°F (316°C) –
EPRI Tests .................................................................................................................... 4-78
Figure 4-29 Corrosion Rates for Bolts Around Insulated or Banded Flanges at 600°F
(316°C) – EPRI Tests .................................................................................................... 4-79
Figure 4-30 Corrosion Rates for Bolts Around Flanges at 180°F (82°C) – EPRI Tests .......... 4-80
Figure 6-1 Methods Used to Detect Leaks Reported in LERs (1985-1996) ............................. 6-7
Figure 6-2 Sensitivities of Typical Leakage Detection Systems ............................................... 6-8
Figure 6-3 Correlation Between Boric Acid Leakage and Volume/Weight of Boric Acid
Crystal Deposits .............................................................................................................. 6-9
Figure 7-1 Leakage Reduction Program Flowchart............................................................... 7-30
Figure 7-2 Functions Provided by Rough Flange Surface in Gasketed Joints ....................... 7-31
Figure 7-3 Typical Inner Edge Buckling on Spiral-Wound Gasket ......................................... 7-31
Figure 7-4 Effect of Gasket Thickness and Flange Flatness on Sealability............................ 7-32
Figure 7-5 Functional Performance of Valve Packing............................................................ 7-32
Figure 7-6 Evolution of Valve Packing................................................................................... 7-33

xviii
 EPRI Licensed Material

Figure 7-7 Effect of Friction on Packing Radial Load and Consolidation................................ 7-34
Figure 7-8 Effect of Stem Thermal Taper on Valve Packing Performance ............................. 7-34
Figure 7-9 Inherent Leak Paths in Tapered Pipe Threads ..................................................... 7-35
Figure 7-10 Typical O-Ring Face Seal Fittings ...................................................................... 7-36
Figure 7-11 CRDM Nozzle Canopy Seal Weld ...................................................................... 7-37
Figure 8-1 Approach to Establishing JCO for Boric Acid Leakage/Degradation..................... 8-17
Figure 8-2 Hole Diameter and Thread Pitch Micrometer ....................................................... 8-18
Figure 8-3 Laser Gauge for Large Diameter Internal Thread Inspections .............................. 8-19
Figure 8-4 Typical Video Inspection Equipment .................................................................... 8-20
Figure 8-5 Shank Diameter Gauge for RCP Flange Bolts...................................................... 8-21
Figure 8-6 Effect of Flange Clearances in Reducing Oxygen Levels at Bolts ........................ 8-21
Figure 8-7 Effect of Boric Acid Deposits in Protecting Surfaces from Corrosion.................... 8-22
Figure 8-8 Logic Chart for Evaluating Corrosion of Pressure Boundary Parts ....................... 8-23
Figure 8-9 Local Membrane Stress and Area of Reinforcement Models................................ 8-24
Figure 8-10 Reactor Coolant Pump Flange Corrosion........................................................... 8-25
Figure 8-11 Leak-Before-Break Approach ............................................................................. 8-26
Figure 8-12 Conditions Amenable to Leak-Before-Break Analysis ........................................ 8-27
Figure 8-13 Model for Leakage from Rectangular Slit ........................................................... 8-28
Figure 8-14 Mass Flow Rate for Leakage of Subcooled Water from Rectangular Slits .......... 8-29
Figure 8-15 Finite Element Model and Results for 16-Stud Manway Closure........................ 8-30
Figure 8-16 Finite Element Model and Results for 20-Stud Manway Closure........................ 8-32
Figure 8-17 Finite Element Model and Results for Reactor Coolant Pump Flange ................ 8-34
Figure 8-18 Finite Element Model and Results for 6-Inch Check Valve ................................. 8-36
Figure 8-19 Finite Element Model and Results for 10-Inch (25-cm) Check Valve .................. 8-38
Figure 8-20 Finite Element Model and Results for Reactor Vessel Head .............................. 8-40
Figure 9-1 Key Factors Regarding Boric Acid Corrosion Management.................................... 9-5
Figure 9-2 Criteria for Replacing Carbon Steel with Corrosion-Resistant Materials ................. 9-6
Figure B-1 Location of Gouge in Vessel Cladding .................................................................. B-3
Figure B-2 Corrosion of the Reactor Vessel Head Top Surface.............................................. B-5
Figure B-3 Corroded Reactor Vessel Head Stud .................................................................... B-8
Figure B-4 Corrosion of Reactor Vessel Flange Threads ..................................................... B-11
Figure B-5 Corroded Reactor Coolant Pump Flange ............................................................ B-13

xix
 EPRI Licensed Material

LIST OF TABLES

Table 2-1 Typical Boron and Boric Acid Concentrations.......................................................... 2-1

Table 2-2 Summary of Reported Sources of Boric Acid Leakage............................................ 2-2
Table 4-1 BNL Tests of Bolting Materials ................................................................................ 4-7
Table 4-2 Key Parameters of CE and EPRI Nozzle Leakage Tests....................................... 4-17
Table 4-3 Average Corrosion Rates in Aerated and Deaerated Water – Westinghouse
Tests ............................................................................................................................. 4-25
Table 4-4 Conditions and Measured Weight Loss – Moscow Power Institute Tests .............. 4-27
Table 4-5 Corrosion Rates in Boric Acid Solution – TVA Tests ............................................. 4-30
Table 4-6 Corrosion Rates with Wetted Solid Boric Acid – TVA Tests................................... 4-30
Table 4-7 Corrosion Rates in Lithiated Boric Acid Solution – TVA Tests ............................... 4-31
Table 4-8 Corrosion Rates in Partially Deaerated Boric Acid Combustion Engineering
Tests ............................................................................................................................. 4-33
Table 4-9 Corrosion Rates in Aerated Boric Acid – Westinghouse Tests .............................. 4-35
Table 4-10 Corrosion Rates in Aerated Boric Acid – BNL Tests ............................................ 4-36
Table 4-11 Corrosion Rate in Aerated and Deaerated Boric Acid – Westinghouse Tests
for BAC.......................................................................................................................... 4-38
Table 4-12 Corrosion Rate in Aearated Boric Acid – Combustion Engineering Tests ............ 4-40
Table 4-13 Corrosion Rates with Dripping Boric Acid – B&W Trough Tests .......................... 4-43
Table 4-14 Corrosion Rates with Dripping Boric Acid – B&W Pipe Tests .............................. 4-46
Table 4-15 Corrosion Rate for Impingement onto Bolts – Combustion Engineering Tests..... 4-49
Table 4-16 Corrosion Rate for Impingement onto Plates – Combustion Engineering
Tests ............................................................................................................................. 4-51
Table 4-17 Corrosion Rates for Leaking Borated Water into an Annulus – Combustion
Engineering Tests ......................................................................................................... 4-55
Table 4-18 Average Corrosion Rates in Aerated Boric Acid – EPRI Tests ............................ 4-58
Table 4-19 Average Corrosion Rates for Borated Water Spray – EPRI Tests ....................... 4-61
Table 4-20 Average Corrosion Rates for Steel in Buried in Boric Acid Crystals – EPRI
Tests ............................................................................................................................. 4-64
Table 4-21 Average Corrosion Rates in Deaerated Boric Acid – EPRI Tests ........................ 4-66
Table 4-22 Maximum Corrosion Rates for Dripping Boric Acid – EPRI Tests ........................ 4-68
Table 4-23 Maximum Corrosion Rates for Leakage into the Annulus – EPRI Tests .............. 4-71
Table 4-24 Maximum Corrosion Rates for Flange Leakage – EPRI Tests ............................. 4-74
Table 7-1 Leakage Classifications by Severity (Proposed by EPRI FSTP) ............................ 7-27

xxi
 EPRI Licensed Material

Table 7-2 Leakage Acceptance Criteria (Proposed by EPRI FSTP) ...................................... 7-28
Table 7-3 Alternative Bolting Materials .................................................................................. 7-29
Table 8-1 Corrosion Rates for Typical Carbon and Low-Alloy Steel Applications.................. 8-16
Table 8-2 Leak-Before-Break Experience for Manway Closures ........................................... 8-16
Table A-1 Industry Experience by Plant Name, Incident Date, Severity, System, and
Component...................................................................................................................... A-2
Table A-2 Industry Experience by System, Component, Plant Name, Incident Date, and
Severity ........................................................................................................................... A-5
Table A-3 Industry Experience by Severity, System, Component, Plant Name, and
Incident Date ................................................................................................................... A-8
Table A-4 Industry Experience by Component, System, Plant Name, Incident Date, and
Severity ......................................................................................................................... A-11
Table B-1 Predicted Stresses................................................................................................. B-6

xxii
 EPRI Licensed Material

1
INTRODUCTION

This section of the Boric Acid Corrosion Guidebook describes why boric acid corrosion is a
concern in pressurized water reactor (PWR) plants and how information in this guidebook can be
used by utilities to develop programs to respond to the boric acid corrosion issue. Boric acid
corrosion is not a significant concern for boiling water reactor (BWR) plants because primary
system water in BWR plants is not borated to control reactivity. Boric acid is used in BWRs only
as a final option in the form of the standby liquid control system to control reactor reactivity
during a severe accident event.

1.1 Purpose of the Boric Acid Corrosion Guidebook

The purpose of this guidebook is to provide utilities with practical information that can be used
to establish or refine a boric acid corrosion reduction program and assess whether components
can be left in service with continuing leakage or degradation.
Work on this project involved three tasks:
• Preparation of the original Boric Acid Corrosion Guidebook in 1995
• Performance of boric acid corrosion screening tests and prototypical joint tests by EPRI in
1996–98
• Preparation of the revised Boric Acid Corrosion Guidebook in 2001

EPRI has previously published several reports that compile information on industry experience
1
with boric acid corrosion and the results of some boric acid corrosion tests [1, 2, 3] . The current
guidebook goes beyond the work in these previous documents to cover programs to:
• Reduce leakage
• Reduce damage in the event that leakage occurs
• Develop justifications for continued operation (JCOs) with ongoing leakage

1.2 Background
Borated water is used in the primary systems of PWR plants to control reactivity during normal
plant operation and refueling and under potential accident conditions. This is accomplished by
adding boric acid to the primary side water. In some cases, boric acid is also injected into the
secondary side of PWR plants at low concentrations to reduce the potential for corrosion of
Alloy 600 steam generator tubing at crevice locations.

1
The numbers in brackets refer to references in Section 10.

1-1
 EPRI Licensed Material

Introduction

In general, there is little concern with corrosion inside the primary and secondary systems since
the boric acid and oxygen concentrations in these systems are low, and corrosion rates are
typically 0.001 inches per year (in/yr) (0.025 mm/yr) or less. Exceptions to this general
experience include 1) stress corrosion cracking of some stainless steel pipes containing stagnant,
high-concentration boric acid solutions, and 2) cracking of stainless steel cladding in some
components that leads to galvanically driven stress corrosion cracking (SCC) of the low-alloy
steel base materials. These special conditions, which involve higher oxygen concentrations, are
not addressed in this report.

If borated water leaks from primary and secondary systems through gasketed joints, valve
packing, mechanical seals, etc., significant corrosion problems can develop. Specifically, the
water can become oxygenated and the boric acid can concentrate as the water boils off or
evaporates. These factors can increase the corrosion rate of exposed carbon steel from <0.001
in/yr (<0.025 mm/yr) to as much as 10 in/yr (254 mm/yr).

Section 2 and Appendix A of this guidebook summarize reported incidents involving corrosion
produced by leakage of borated water. It is also known that many other incidents have been
handled internally at plants and have not been reported publicly. The reported incidents range in
severity from minor corrosion of parts, which can be accepted without evaluation or repair, to
major incidents involving plant shutdowns and significant loss of material on major components.
Two examples point out the potential for significant consequences:
• A visual inspection of the Ft. Calhoun reactor coolant system during a 1980 refueling outage
showed that water was dripping from the reactor coolant pump insulation. Further
investigation showed that the leakage was coming from the spiral-wound gaskets and that
three studs on one pump and four studs on another pump were severely corroded. The
corrosion took on an hourglass shape over a region extending about 3.75 inches (95 mm)
above the top of the pump casing flange. The diameters of the worst case studs were reduced
from the original 3.5 inches (89 mm) to 1.0–1.5 inches (25–38 mm). This represented a
significant reduction to less than 20% of the original cross-sectional area on the worst case
studs. Figure 1-1 shows the extent of degradation of one of the studs from a Ft. Calhoun
reactor coolant pump.
• In 1986, high-pressure safety injection nozzles at Arkansas Nuclear One Unit 1 were being
nondestructively inspected during a normal refueling outage. When the insulation was
removed from a nozzle, severe corrosion was discovered on the nozzle outside surface as
shown in Figure 1-2. At its deepest point, the corrosion extended two-thirds of the way
through the nozzle wall thickness. The corrosion was traced to leakage from the body-to-
bonnet joint of the high-pressure injection (HPI) isolation valve that is located about 8 feet
(2.4 m) above the corroded area. This valve had operated with a known leak of less than 0.1
gallon per minute (gpm) (0.38 L/min) for at least five months prior to discovering the
corrosion.

As a result of these and other incidents, the NRC issued Generic Letter 88-05, Boric Acid
Corrosion of Carbon Steel Reactor Pressure Boundary Components in PWR Plants [4]. This
generic letter requires that utilities develop and implement programs to identify leaks and take
corrective action to prevent recurrence. All plants have developed programs that respond to this
generic letter.

1-2
 EPRI Licensed Material

Introduction

The purpose of the Boric Acid Corrosion Guidebook is to provide information that utilities can
use to further refine their boric acid corrosion programs and assess whether components can be
left in service with continuing leakage or degradation.

1.3 Approach

The basis for this revision was principally to incorporate the results and analysis of boric acid
corrosion testing data resulting from EPRI testing conducted at SwRI from 1996–1998. In
addition, EPRI solicited comments from end users of the original guidebook published in 1995.
Suggested additions provided by member-utility representatives included:
• Updating references to industry experience with boric acid corrosion incidents since 1995
• Including new work pertaining to fluid leak management and plant leakage reduction
resulting from EPRI programs in these areas
• Increasing the guidance on managing boric acid corrosion events, with an emphasis on
improved guidance for justifications for continued operation

A draft report was prepared and circulated for peer review among interested parties at EPRI-
member utilities, and their comments were subsequently incorporated into this final document.
EPRI coordinated closely with the members of the EPRI Pressurized-Water Reactor Materials
Reliability Project (MRP) Technical Support Subcommittee (TSS) and relied heavily on their
input with respect to final scope, technical review, and approval of this document.

1.4 Content/Organization

The content and organization of the guidebook are as follows:

Section 2 Industry Experience – This section summarizes industry experience with boric
acid leakage and the resultant corrosion.
Section 3 Boric Acid Corrosion Mechanisms – This section describes mechanisms by which
borated water corrodes carbon and low-alloy steel components.
Section 4 Boric Acid Corrosion Tests – This section summarizes results of tests performed
by several organizations to determine corrosion rates under various conditions,
including testing conducted for EPRI by Southwest Research Institute (SwRI) to
assess the boric acid corrosion resistance of several typical bolting materials and
to determine the extent of corrosion under several prototypical conditions of
significant interest to utilities.
Section 5 Regulatory Bases – This section summarizes portions of federal, NRC, and
ASME regulations that pertain to leakage of borated water and the resultant
corrosion. This section also addresses approaches that utilities can take to
minimize the impact of required inspections on refueling outage schedules.
Section 6 Detecting Leakage – This section describes methods that can be used to determine
the amount of leakage from primary system components.

1-3
 EPRI Licensed Material

Introduction

Section 7 Preventing Leakage and Degradation – This section describes approaches that can
be used to reduce the amount of leakage of borated water and to reduce the
amount of degradation in the event that some leakage occurs. This section draws
heavily on the concurrent EPRI Fluid Sealing Technology Program that has the
objective of determining the root cause of leaks from various types of joints, and
identifying products and work practices that have the potential to reduce the
leakage from these joints.
Section 8 Continued Operation with Leakage and/or Degradation – Ideally, the objective is
to operate a plant with no borated water leaks. However, this is not always
achievable despite good intentions. This section includes guidelines for evaluating
and dispositioning boric acid leaks and suggestions regarding preparation of
justifications for continued operation (JCO) with ongoing leakage from some
components.
Section 9 Boric Acid Corrosion Management – This section describes alternative
approaches for managing the boric acid corrosion issue.
Section 10 References – This section provides references to material used in developing the
guidebook.
Appendix A Reported Incidents of Boric Acid Corrosion– This appendix is a database of
reported boric acid corrosion incidents. Several indexes are provided to facilitate
access to the information.
Appendix B Example Analyses – This appendix provides several examples of analyses to
permit operation with continued leakage or with degradation from previous
leakage.

1.5 Application

The information in this guidebook can be used by utility personnel to refine programs that
address boric acid corrosion issues. Section 4 provides boric acid corrosion test data that can also
be used to develop JCOs. Section 7 provides descriptions of methods to reduce leakage and
degradation if leakage occurs. Section 8 provides approaches to developing JCOs for operation
with continued leakage and/or degradation. Section 9 provides a suggested approach to
managing boric acid corrosion issues.

1.6 Benefits

Utilities experiencing ongoing issues with leakage of borated water or its resultant degradation
will find this guidebook useful in developing cost-effective programs to reduce the extent or
frequency of leakage events, evaluate and manage the leakage events, and minimize their impact
on plant operations. Utilities may also find the information helpful in developing technical cases
to continue using some corroded parts and in developing JCOs for operating with continuing
leakage under some conditions.

1-4
 EPRI Licensed Material

Introduction

Figure 1-1
Corroded Ft. Calhoun Reactor Coolant Pump Stud (Illustration Created from
Photograph [5])

1-5
 EPRI Licensed Material

Introduction

Figure 1-2
Corrosion of ANO-1 High-Pressure Injection Nozzle [1]
Photograph from Entergy Operations, Inc.

1-6
 EPRI Licensed Material

2
INDUSTRY EXPERIENCE

This section summarizes field experience with boric acid leakage and corrosion. It includes a
review of publicly documented incidents, descriptions of the major types of incidents and their
direct consequences, a discussion of some of the indirect consequences, and conclusions
regarding the types of situations in which leakage of borated water can become a serious
concern.

2.1 Systems Containing Borated Water

Figure 2-1 is a simplified diagram showing the main systems that contain borated water in a
typical PWR plant. Table 2-1 shows boron and boric acid concentrations that are typically
encountered in these systems.
Table 2-1
Typical Boron and Boric Acid Concentrations

Boron Boric Acid

Location Concentration Concentration
(ppm) (wt %)

Reactor Coolant System 0–2000 0–1.14

Refueling Water Storage Tanks 2400–2500 1.37–1.43

ECCS Accumulators 2300–2500 1.31–1.43

Recycle Evaporator 7000 4.00

Boric Acid Storage Tanks 7000–7700 3.99–4.39

Boron Injection Tanks*

Early Plant Designs 22,000 12.5
Typical Current Configurations 2000 1.14

* Early plant designs provided for injecting high-concentration boric acid to maintain plant safety under accident conditions. It
was subsequently determined that the normal refueling water storage tank concentrations of 2400–2500 ppm boron are
adequate to meet safety requirements. Accordingly, while the boron injection tanks (BITs) have typically been left in the
system, the boron concentrations have been reduced to about 2000 ppm. By reducing the boron concentration from 22,000 to
2000 ppm, the need for heat tracing on the BIT tank and associated piping can be eliminated.

It is important to recognize that the concentration of boric acid in water that has leaked from the
systems can be much higher than the normal operating concentration due to evaporation of the
water.

2-1
 EPRI Licensed Material

Industry Experience

2.2 Overview of Reported Incidents

Appendix A is a compilation of field experience with corrosion and other types of degradation
that are caused by leakage of borated water in PWR plants. This compilation is limited to
incidents that have been reported in publicly available documents. As such, it does not include
incidents that were resolved without the incident being publicly documented. Information
sources used to develop these descriptions include the Nuclear Power Experience Reports [6],
NPRDS [7], EPRI reports [1, 2, 3], NUREG reports [8, 9, 10], NRC Information Notices [11, 12,
13, 14, 15, 16], NRC IE Bulletins [17], NRC Generic Letter [4], and NRC Public Documents
Room files. Specific references for each incident are listed in the database in Appendix A.

Although it is recognized that many incidents are handled without being reported publicly, those
described in Appendix A cover the most significant incidents and provide a good summary of the
types of boric acid corrosion events that can occur at PWRs. The incidents summarized in
Appendix A can be categorized in a number of different ways. One of the more useful ways is to
classify each incident by the leakage source as shown in Table 2-2 and Figure 2-2. Figure 2-3
shows locations where significant damage has occurred on components in the primary loop. The
incidents covered in Appendix A range in severity from a small volume of boric acid crystal
deposits around a cracked Alloy 600 instrument nozzle to corrosion that was 70% through the
wall thickness of a 6-inch (152-mm) pipe connected to the reactor coolant system.
Table 2-2
Summary of Reported Sources of Boric Acid Leakage

Source of Leakage Number of Reported Incidents

Valve Packing 21
Gaskets – Manway 15
Gaskets - Valve Bonnet 13
Gaskets - Pump Flange 11
Gaskets - Heat Exchanger 4
Flange
Gaskets - CRDM Nozzle 2
Gaskets – Other 2
Mechanical Seals 8
Instrument Line Fittings/Welds 8
Seal Welds 8
Carbon Steel Used Rather Than 3
Specified Stainless Steel
Diaphragms 2
Alloy 600 Nozzle Primary Water 11
Stress Corrosion Cracking
(PWSCC)
Liner Failures 2
Not Identified 20

2-2
 EPRI Licensed Material

Industry Experience

2.3 Primary Sources of Leakage and Direct Consequences

2.3.1 Gaskets

Gasketed joints have been the main source of borated water leakage. Gasket leakage has resulted
from inadequate bolt torque, use of incorrect gaskets, damage to the gasket seating surfaces, and
poor joint design. The main concern with leakage of borated water from gasketed joints is that
small amounts of leakage can corrode low-alloy steel fasteners to the point where they can no
longer carry the operating pressure load. Whereas there have been no cases with major loss of
primary system integrity due to gasket leakage, several incidents have occurred that involved
significant amounts of leakage and bolt corrosion. Particular gaskets of concern include:
• Steam generator and pressurizer manways
• Reactor coolant pump gaskets
• Control rod drive mechanism (CRDM) nozzle gaskets (Babcock & Wilcox- [B&W] design
plants only)
• Heat exchanger gaskets
• Valve body-to-bonnet gaskets

2.3.1.1 Steam Generator and Pressurizer Manways

Figure 2-4 shows a typical steam generator primary-side manway. This design consists of a
flange that is integral with the vessel shell, a carbon steel cover plate, a spiral-wound gasket, 16–
20 low-alloy steel studs, and a stainless steel insert plate installed on the inside surface of the
cover plate to protect the carbon steel cover plate from corrosion and to minimize temperature
differences between the vessel shell and cover plate during heat-up and cool-down transients.
Pressurizer manways are similar but are typically located on the vessel top head.

Appendix A describes six incidents involving significant degradation at steam generator or

pressurizer manways. These incidents are:
• St. Lucie 1 (1978) – An increase in containment gaseous activity was traced to a minor leak
from a pressurizer manway. Five of twenty 1-1/2-inch- (38-mm-) diameter SA-540 Grade
B24 Class 3 studs in this manway were found to be corroded. The maximum corrosion depth
was a relatively shallow 1/8 inch (3 mm). The five corroded studs were replaced.
• Arkansas Nuclear One 1 (1981) – A leak of about 3.4 gpm (13 L/min) was detected in the
reactor coolant system (RCS). The leakage was traced to a steam generator manway.
Inspections showed erosion of the bearing surfaces between the generator flange and
manway cover plate and degradation of several 2-inch- (51-mm-) diameter SA-320 Grade
L43 studs. As a result of this incident, the bearing surfaces were machined, the degraded
studs were replaced, and the manway cover plate installation and removal procedures were
modified.
• Calvert Cliffs 2 (1981) – Leakage was discovered from a pressurizer manway gasket. Two of
the 20 SA-540 Grade B24 Class 3 studs were replaced due to boric acid wastage.

2-3
 EPRI Licensed Material

Industry Experience

• Arkansas Nuclear One 2 (1982) – A leak of about 3 gpm (11 L/min) was detected in the
RCS. The leakage was traced to a steam generator manway. Eight of the 20 studs had been
damaged by steam cutting and there was some evidence of minor boric acid attack. The
maximum depth of the damage was about 1/8 inch (3 mm). The eight damaged studs were
replaced.
• Maine Yankee (1982) – Six of 20 studs in one steam generator manway were found to have
failed and five more were found to be cracked during a normal refueling outage. The failures
appear to have resulted from cracking rather than boric acid wastage although some of the
studs had evidence of surface corrosion. The cracking was attributed to the effects of the
combination of an injected leak sealant, thread lubricant, and borated water. All of the studs
on the affected manway were replaced.
• Indian Point 2 (1983) – Leakage was found from three steam generator manways (two hot
leg and one cold leg). The leakage was found to have caused pitting and corrosion of the
closure parts. Corrective action included replacing the gaskets, bolts, and inserts on the
affected closures and changing the preload torque and torquing procedure. Plans were also
made to replace one of the manway cover plates during the next outage.
Summary: These incidents suggest that leakage from steam generator and pressurizer manway
joints have led to plant shutdowns and significant amounts of rework. However, the experience
also shows that these incidents have all involved a leak-before-break type behavior. Manways
are typically installed at an orientation where the leakage does not tend to form a pool and
corrode all studs simultaneously. Finite element analyses in Section 8 show that significant
leakage will occur before risk of failure of the remaining closure bolts. The main incentives in
preventing leaks from these joints are to avoid forced outages and the need to make repairs.
It is significant to note that there have been no reported incidents of boric acid corrosion from
primary manway leaks since the initial issue of the Boric Acid Corrosion Guidebook in 1994.
This suggests that utilities are doing a better job of reducing leaks from these important joints
and dealing with any resultant leaks before there is any significant degradation.

2.3.1.2 Reactor Coolant Pump Gaskets

Figure 2-5 shows a typical reactor coolant pump without the electrical motor. From a boric acid
corrosion standpoint, the pump components that require attention are the pump casing, the pump
cover plate, a spiral-wound type gasket, the bolting ring, and cover plate studs.

Appendix A describes several incidents involving significant degradation due to leakage of

borated water from the cover plate gaskets of a coolant pump. These incidents are:
• Ft. Calhoun (1980 and 1981) – A visual inspection with the RCS at 180 psi (1.2 MPa)
showed evidence of water dripping from insulation at the pump flange region. Investigation
showed that the leakage was coming from the spiral-wound gaskets and that three studs on
one pump and four studs on another pump were severely corroded. The severely corroded
studs on each pump were located side-by-side. The corrosion took on an hourglass shape
over a region extending about 3.75 inches (95 mm) above the top of the pump casing (see
Figure 1-1). The diameters of the worst case studs were reduced from the original 3.5 inches
(89 mm) to 1.0–1.5 inches (25–38 mm). This represents a significant reduction to less than
20% of the original cross-section area on worst case studs. There was no corrosion of the

2-4
 EPRI Licensed Material

Industry Experience

stainless steel pump casings and flanges. The original pump design included two concentric
gaskets with a leak-off port in between to detect leakage past the inner gasket. However, the
leak-off ports on the Fort Calhoun pumps had been plugged, and the leakage was not
detected. The pumps were repaired by replacing damaged parts, including 11 studs with less
severe corrosion.

During the 1981 refueling outage, some new corrosion of studs on the same pumps had
occurred. This corrosion was attributed to condensate from a component cooling water line
dripping onto the flange region that still had boric acid residue from the previous incident.
The maximum corrosion depth in this case was 0.25 inches (6.4 mm). A total of 14 studs
were replaced during this outage.
• Oconee 3 (1981 and 1982) – Leakage had been occurring from the closure flange region of
all four coolant pumps for 3–4 years. The worst case leakage on one pump was as high as 0.5
gpm (2 L/min). The leakage on the other pumps was visible but estimated to be less than 0.1
gpm (0.38 L/min).

During the 1981 refueling outage, an attempt was made to stop the leakage by injecting
sealant into the flange region. During this operation, boric acid deposits were found around
several of the studs. “Go/no-go” gauges were used to determine the minimum diameter of the
studs (see Figure 8-5). The diameter of the worst case stud had been reduced from the
original 3.84 inches (98 mm) to less than the vendor’s computed 3.25-inch (83-mm)
minimum diameter. This stud was replaced. The diameter of the next most severely corroded
stud was 3.52 inches (89 mm). This stud was left in place since it met the vendor’s criterion.

During the 1982 refueling outage, an additional five studs in two pumps were found to be
corroded in excess of the vendor’s 3.25-inch (83-mm) minimum diameter. These studs were
replaced with new studs machined from SA-540 Grade B23 material rather than the original
SA-193 Grade B7 material.
• Oconee 2 (1981) – Inspections during the 1981 refueling outage showed significant corrosion
of four studs on one reactor coolant pump. The leakage was traced to the pump closure
gasket. One of the studs was corroded in excess of the vendor-specified 3.25-inch (83-mm)
minimum diameter, and this stud was replaced. It was concluded that continued operation
would have led to increasing leakage that would have provided warning for a safe plant
shutdown.
• HB Robinson (1995) – Leakage from a high-pressure tap flange gasket and the main pump
flange gasket allowed borated water to come into contact with and corrode the main pump
flange bolts. The pressure tap flange gasket leak was attributed to too low a bolt preload
stress. The main flange gasket leak was attributed to loss of bolt preload over time.
Corrective action consisted of replacing degraded studs and retorquing all of the studs.
• Callaway (1996) – When maintaining a Callaway main coolant pump in a Westinghouse
shop, blistering and linear indications were found on the two opposite flanges of the thermal
barrier and the number 1 seal. The material for both was SA-182, grade F304 stainless steel.
The seal was installed in the D pump from commissioning in 1984 until the fall of 1996.
Blisters and linear indications were located outside the gasket contact area. The blisters could
be detected both visually and by touching the surface. Crystallized boric acid could be seen
between the two flanges. The studs showed minor wastage out of the threaded area.

2-5
 EPRI Licensed Material

Industry Experience

Machining to a 0.021-inch (0.5-mm) depth was required to remove the blisters and pitting on
the seal flange. The machining depth on the thermal barrier flange was 0.090 inch (2.3 mm).

Summary: These incidents suggest the potential for serious corrosion of reactor coolant pump
studs and point out the need for regularly scheduled inspections of the reactor coolant pump
flanges for evidence of boric acid leakage, especially if a leakage detection system is not in
operation. The experience also points out the need to remove any boric acid residue that
accumulates on carbon or low-alloy steel surfaces because this residue could become wetted by
leakage from other sources, resulting in continued or accelerated corrosion rates.

2.3.1.3 CRDM Nozzle Gaskets

Figure 2-6 shows a typical gasketed CRDM nozzle in a B&W-design plant. The gasketed joint
design is unique to B&W-design plants (Westinghouse- and CE-design plants use seal welded
joints). The gasketed joint consists of a flange on the vessel head, a flange on the CRDM, spiral-
wound gaskets, two 180° nut ring segments below the vessel flange, and eight bolts.

Several of these gasketed joints are often found to have developed small leaks during each
refueling outage. These leaks are usually fixed by removing the CRDM, replacing the gaskets,
and reinstalling the CRDM. One publicly documented report of leakage from these joints causing
significant degradation is as follows:
• Arkansas Nuclear One 1 (1989) – Leakage was discovered from a flanged CRDM nozzle
joint. Upon disassembly, it was discovered that one half of a low-alloy steel nut ring segment
had corroded away and that two of the four low-alloy steel bolts that hold the CRDM to the
nut ring segment had corroded to the point where there was no thread engagement. Five other
flanges were found to have boric acid deposits but no damage. All six leaking flanged joints
were repaired by replacing the gaskets, nut rings, bolts, and nuts. The replacement bolts and
nut rings were fabricated from corrosion-resistant material.

Significant damage at this joint resulted from the fact that the joint is located above the head
insulation and at a low enough temperature that leaking primary coolant can remain on the
surface as a liquid and the boric acid can concentrate as the water slowly evaporates.

Summary: Leakage from gasketed CRDM nozzles on B&W plants is a continuing maintenance
issue. The course of action that has been developed to deal with this issue is to inspect the
flanged joints during each refueling outage for evidence of boric acid leakage and repair the
leaking joints. The repairs consist of removing the affected drives, replacing the gaskets,
replacing the low-alloy steel bolts and nut rings with corrosion resistant material, and cleaning
up any boric acid residue. The need to clean up boric acid residue is highlighted by experience at
Oconee 1 and 3 in 2000 and 2001, respectively, where leaks from cracked Alloy 600 CRDM
nozzles and Alloy 182 J-groove welds would not have been detected had there been significant
boric acid residue on the vessel head.

2-6
 EPRI Licensed Material

Industry Experience

2.3.1.4 Heat Exchanger Gaskets

There have been several reports of degradation resulting from leaking heat exchanger gaskets.
For example, Appendix A describes three cases where leakage from gasketed flanges on letdown
heat exchangers has led to degradation of the flange bolts (see Beaver Valley 1, 1992; Beaver
Valley 2, 1990; and Palisades, 1986).

2.3.1.5 Valve Body-to-Bonnet Gaskets

There have been several reported incidents involving leakage from body-to-bonnet gaskets, but
the most significant event occurred at Arkansas Nuclear One 1. A recent incident at Three Mile
Island 1 also shows that gasket leakage can be caused by conditions not anticipated in the
original design. These incidents are described below:
• Arkansas Nuclear One 1 (1986) –HPI nozzles were being nondestructively inspected during a
normal refueling outage. When the insulation was removed from a nozzle, severe corrosion
was discovered on the nozzle outside surface. At its deepest point, the corrosion extended
two-thirds of the way through the nozzle wall thickness. The corrosion was traced to leakage
from the body-to-bonnet joint of the HPI isolation valve located about 8 feet (2.4 m) above
the corroded area. This valve had operated with a known leak of less than 0.1 gpm (0.38
L/min) for at least five months prior to discovering the consequential damage caused to the
HPI nozzle. Leaking borated water had run down the piping, undetected under the piping
insulation, until reaching the low-alloy steel nozzle that began to corrode. Figure 1-2 shows
the arrangement of the valve, piping, insulation, nozzle, and damage. Measurements of the
corroded areas showed some evidence of galvanic attack at the interface between the
stainless steel pipe and the low-alloy steel nozzle.
• Three Mile Island 1 (1993 & 1994) – In November 1993, a leak was identified at the
pressurizer spray block valve and the leakage was traced to the packing. The packing was
replaced and the valve was returned to service. In March 1994, a leak rate of 0.1 gpm was
detected from the primary system. A video camera inside containment showed a steam leak
from the pressurizer block valve. A close-up inspection showed that the leak was coming
from the body-to-bonnet joint near stud #1 of eight 5/8-inch- (16-mm-) diameter studs. It was
decided to tighten the studs hot to 120 ft-lb (163 N⋅m). When stud #4 was torqued to 95 ft-lb
(129 N⋅m), the leak rate near stud #1 increased rapidly to about 3 gpm (11 L/min). Upon
disassembly, it was determined that stud #1 had failed completely, and four other adjacent
studs (#2, #3, #7, and #8) were significantly corroded. Stud #8 was corroded to about 50% of
its original diameter. The root cause of the leakage was traced to an increase in the valve
operator load from 4000 to 15,000 lb (17.8–66.7 kN). The 15,000 lb (66.7 kN) load
overloaded the body-to-bonnet studs and, thereby, opened up the joint slightly. Corrective
action included replacing the studs with corrosion-resistant material and reducing the
maximum valve operator force.
• Millstone 2 (1995) – An active leak was discovered from the flanged joint on a regenerative
heat exchanger relief valve. The bolts were found to be encased in boric acid crystals but not
corroded because they were made from corrosion-resistant material. The leakage was
attributed to an uneven torque on the four flange bolts. This incident points out the

2-7
 EPRI Licensed Material

Industry Experience

desirability of using high-strength corrosion-resistant bolting in important applications to

minimize potential consequential damage in the event of a leak.

Summary: These incidents involving valve body-to-bonnet leaks reinforce the finding that small
leaks can lead to the corrosion of critical bolts, which in turn can result in larger, more
significant leakage. The experience also points out the potential risk posed by the corrosion of
carbon steel bolts in critical applications. However, the incidents also show that serious leakage
is generally precluded by the leak-before-break behavior of larger gasketed joints. The incident
at ANO-1 points out the need to determine where any leakage collects. If there is insulation
around the component, it is especially important to determine if leakage has collected between
the insulation and adjacent carbon steel parts.

2.3.2 Valve Packing

Several significant incidents have occurred associated with leakage from valve packing. Figure
2-7 shows arrangements of two typical valve types used in nuclear service and the locations
where leakage can occur. The most significant incidents have involved corrosion of the packing
follower bolts or body-to-bonnet bolts to the point that the valve integrity is affected.

There have been several reports of significant degradation resulting from valve packing leaks.
Major incidents from Appendix A are:
• Surry 2 (1978) – A primary coolant leak of 15–18 gpm (57–68 L/min) was discovered from
the packing of a motor-operated isolation valve between the reactor coolant system and the
residual heat removal (RHR) system. It was discovered that leakage from the valve packing
had caused significant corrosion of the packing gland follower studs. Failure of these studs
caused the internal pressure in the valve to dislodge the packing, thereby causing a larger
leak. The larger leak led to failure of one of the hot leg temperature channels when water and
steam penetrated the resistance temperature detector (RTD). The repair effort included
replacing all carbon steel studs with more corrosion-resistant material.
• San Onofre 2 (1987) – A 10-inch (254-mm) isolation valve in the shutdown cooling system
was found to be stuck in the closed position. In an attempt to operate the valve manually, the
valve packing follower plate was dislodged when the low-alloy steel holddown bolts failed.
These bolts had become corroded by previous boric acid leakage from the packing. The
system pressure of 350 psig (2.4 MPa) caused the packing to extrude, producing a leak of
60–100 gpm (227–379 L/min). A total of 18,000 gallons (68,137 L) of reactor coolant spilled
into the containment building, and five workers were contaminated. The concentration of
radioactive gasses at the site boundary reached 17% of the allowable. Sticking of the valve
was attributed to buildup of boric acid crystals in the valve stem and packing assembly. The
repair effort included replacing the low-alloy steel studs with a more corrosion-resistant
material.
• Waterford 3 (1992) – A leak in excess of the Technical Specification limit of 1 gpm (3.8
L/min) was traced to the packing area of a hot leg sample valve and was quantified as 13
gpm (49 L/min). The high leakage rate resulted from failure of the packing gland studs due to
corrosion of the low-alloy steel studs by small quantities of boric acid from the original
packing leak. A temporary repair was made by installing a strongback and clamp assembly to
hold the packing and packing retainer plate in place and then injecting a leak sealant until the

2-8
 EPRI Licensed Material

Industry Experience

leak was stopped. In July 1992, the temporary repair failed, again resulting in a leak greater
than the Technical Specification limit of 1 gpm (3.8 L/min).
• Robinson 2 (1980) – Leakage was discovered from an auxiliary pressurizer spray valve
during normal plant operation. The leakage was originally reported as 2 gpm (7.6 L/min) and
then increased to 22 gpm (83 L/min). The higher rate exceeded the plant Technical
Specification limit and forced a plant shutdown. It was reported that the high leakage resulted
from failure of the low-alloy steel gland studs by corrosion that was caused by very small
amounts of packing leakage. Carbon steel studs had incorrectly been used as replacements
for stainless steel studs at some point in time prior to 1974. Repairs included replacing all of
the carbon steel studs with Type 416 stainless steel studs and checking other similar valves
for carbon steel studs. One valve was found to contain a single carbon steel stud.
• Davis Besse 1 (1998) – After tracking leakage from the pressurizer spray valve packing for
several months, the leakage became such that it was not possible to maintain system pressure
under all conditions. During repair, it was found that three of the eight body-to-bonnet nuts
were carbon steel rather than the specified stainless steel. Two of these nuts were seriously
corroded. A fourth nut was found to be 410 martensitic stainless steel that had also not been
specified. However, the martensitic stainless material was not corroded, consistent with
expected performance based on data in Section 4.
• Davis Besse 1 (1999) – Nine primary system pressure boundary valves were found to have
corroded carbon steel yokes as a result of valve packing leaks. The worst case yoke was
found to have its cross section reduced by more than 80%. The valves were repaired and all
valves were checked to verify zero leakage during plant startup. This experience points out
the desirability of using stainless steel materials for highly stressed working parts in valves
where leakage may occur.

Summary: These incidents involving valve packing reinforce the finding that small leaks can
lead to corrosion of critical bolts, which in turn can result in larger, more significant leaks.
However, the incidents also show that serious leakage is generally precluded by limited flow
areas in the valve packing areas. The incident at ANO-1 points out the need to determine where
any leakage collects. If there is insulation around the component, it is especially important to
determine if the leakage has collected between the insulation and adjacent carbon steel parts.

2-9
 EPRI Licensed Material

Industry Experience

2.3.3 Mechanical Seals

The two major locations of leaks from mechanical seals have been O-rings and instrument
penetration Conoseals in the reactor vessel head.

2.3.3.1 Instrument Nozzle Conoseals

Figure 2-8 shows a typical reactor vessel head instrument nozzle Conoseal connection. Although
there have been several reported incidents involving leakage from Conoseal joints, the most
significant incident from a degradation standpoint occurred at Turkey Point 4 in 1987:
• Turkey Point 4 (1987) – In August 1986, a leak was found at a Conoseal joint in an
instrument port column assembly. The utility determined that the plant could continue in
operation and re-inspect the area of leakage in six months. In October 1986, the Conoseal
joint was re-inspected when the plant was shut down for another reason, and it was again
determined that the plant could operate for six months before re-inspecting the Conoseal
joint. When the Conoseal joint was inspected again in March of 1987, over 500 pounds (227
kg) of boric acid deposits were found on top of the vessel head. Associated damage included
corrosion to head studs, instrument port assembly, CRDM coolers, CRDM coils, electrical
connectors, the closure head, cables, control equipment, various instruments, and closure
head insulation. Erosion of the Conoseal gasket and clamp were also reported. The maximum
corrosion depth was 0.25 inch (6.4 mm). The extent of damage is illustrated in Figure 2-9.
(Note: The average leak rate between October 86 and March 87 was reported to be less than
0.45 gpm [1.7 L/min].)
• Calvert Cliffs 1 (1994) – A leak occurred from an incore instrument nozzle mechanical seal
at Calvert Cliffs 1. Leakage from this nozzle was picked up by degraded thermal
performance of the containment air coolers. The boric acid leaking from the nozzle had been
transported from the nozzle to the coolers by containment air circulation, and the boric acid
had condensed on the cooler coils. Significant corrosion was found on three of the eight 1-
3/4-inch- (44-mm-) diameter A193 B7 studs on the nozzle flange. One stud had failed
completely and two others were significantly corroded. Another flange was found to have
similar corrosion. The utility had been aware of the leak but believed that the bolts would not
experience significant corrosion because the flange would be exposed only to dry boric acid
crystals at the estimated flange temperature of 500°F (260°C). However, the flange was
actually in the range of 160–295°F (71–146°C), and the fasteners were exposed to moist
boric acid. The corrective action included replacing the carbon steel fasteners with corrosion-
resistant material.

Summary: Leakage from mechanical seals on instrument nozzles is a particular concern because
mechanical seals can have a higher potential for leakage than many other types of joints. The
leakage occurs at a location over the reactor vessel head, and the temperatures at the instrument
nozzle seals are low enough that the leakage can occur as liquid in which the boric acid
concentrates as the water evaporates. The situation is often compounded by use of low-alloy
steel fasteners that can corrode, leading to flange separation and even higher leak rates.

2-10
 EPRI Licensed Material

Industry Experience

2.3.3.2 Reactor Vessel Head Metallic O-Rings

A review of the information in Appendix A shows that there have been several reports of leaking
reactor vessel head O-ring seals. However, there are only two reported cases where any
significant degradation occurred. The most likely reason for the limited damage from O-ring seal
leakage is that there are two seals in series with a leakage monitoring port in between (see Figure
2-10). Accordingly, there should be adequate warning of leakage past the primary seal.
• Millstone 2 (1988) – On three separate occasions in 1988, leaks developed from the reactor
vessel O-ring seals. On one occasion, the unidentified leakage rate from the primary system
was half of the Technical Specification limit. It was determined that leakage was coming
from the reactor vessel O-ring seal. Examination of the closure region showed corrosion on
nine reactor vessel studs and two small areas on cold leg nozzles. The nine corroded studs
were replaced with spares, although plans were made to retain seven or eight of these studs
for possible future use. The small corroded areas on the cold leg nozzles were cleaned by
flapper wheels and wire brushes. There was no measurable wall thinning on the nozzles.
• Turkey Point 3 (1985) – Leakage was discovered from the inner O-ring seal during plant
heatup. Several days later, leakage of 2–3 gpm (7.5–11 L/min) was discovered from three
locations around the outer O-ring seal. Examination of the O-ring seating surfaces showed
several locations where the leakage had caused small amounts of wire cutting.
• Other Cases – Although Millstone 2 is the only reported incident in which the reactor vessel
head studs were corroded by leakage from the O-ring seal, there have been other cases where
the reactor vessel studs have received minor corrosion from other leak sources of boric acid.
Two such examples are leakage from temporary closure flange seals during hot functional
testing of Calvert Cliffs 1 (1974) that caused very minor corrosion of the studs and water that
leaked onto the closure flange region of Palisades (1971) during pre-operational testing and
caused shallow corrosion (<0.020 inch or <0.5 mm) to several studs.

Summary: Leakage of primary water past reactor vessel metallic O-rings has not resulted in
significant boric acid corrosion problems. This is because there are two O-rings in series with a
leakage monitor port between the seals. Accordingly, there is adequate warning of a leak past the
inner seal so that corrective action can be taken before significant leakage occurs past the outer
seal.

2.3.4 Pressure Boundary Component Leaks

Several reports of corrosion associated with leaks from the areas around welds have been
submitted. Typical examples are leakage from canopy seal welds, leakage from socket weld
fittings on small diameter instrument lines, and leakage from through-wall primary water stress
corrosion cracking (PWSCC) of Alloy 600 penetrations at J-groove weld locations.

2.3.4.1 CRDM Nozzle Canopy Seal Welds

Westinghouse plants have canopy seal welds between the CRDM nozzles in the vessel head and
the bottom of the CRDM assembly. A typical CRDM nozzle is shown in Figure 2-11. The most
significant example of damage from leakage at a CRDM canopy seal weld occurred at Salem 2

2-11
 EPRI Licensed Material

Industry Experience

in 1987. (Note: The boric acid corrosion resulted from a leak from a PWSCC crack, that is, the
boric acid leak did not cause the PWSCC.)
• Salem 2 (1987) – Increased radioactivity in the reactor containment led to a plant shutdown.
An inspection showed boric acid crystal deposits around a seam in the ventilation cowling.
Upon removal of the cowling, a rust-covered pile of boric acid about 3 x 5 x 1 ft (0.9 x 1.5 x
0.3 m) deep was discovered on the vessel head. A thin film of white boric acid residue was
found on some areas of the head and 1–2 feet (0.3–0.6 m) up the control rod drive
mechanism housings. Nine pits were found in the vessel head under the pile of boric acid.
The pits were 1–3 inches (25–76 mm) in diameter and the deepest pits were 0.36–0.40 inch
(9.1–10.2 mm) deep. The pits did not infringe upon the minimum required vessel head
thickness. The boric acid buildup was attributed to leakage from three pinholes in the seal
weld at the base of a thermocouple instrumentation threaded connection. Some of the
primary water that flashed to steam is believed to have condensed on the head ventilation
supports and then dripped down onto the dry boric acid deposits thereby producing
conditions in which corrosion could occur.

The pinholes in the canopy seal weld were attributed to poor water chemistry in the enclosed
space between the seal weld and the threads. This is essentially a dead end volume that is not
easily flushed out by normal flow of primary coolant in the reactor.
• Sequoyah 1 (1995) – Indications of boric acid were discovered at two CRDM nozzle canopy
seals. Leakage was attributed to transgranular stress corrosion cracking (SCC). Corrective
action consisted of performing overlay repairs as performed at other plants.

2.3.4.2 Vibratory Fatigue of Small Diameter Lines

Small diameter instrument lines are often susceptible to cracking as a result of high-cycle fatigue
caused by vibration. These lines are particularly susceptible to cracks and leaks at socket weld
fittings because these locations are typically points of high stress concentration. The following is
an example of the potential consequences of leakage from these lines:
• Calvert Cliffs 1 (1980) – Studs in two of the four reactor coolant pumps were found to be
corroded. Twenty of the studs showed clear evidence of corrosion and six additional studs
showed some signs of corrosion. The corrosion depth ranged from 0.22–0.95 inches (5.6–
24.1 mm) out of an original 4-inch (102-mm) stud diameter. The worst case damage was
concentrated over a length of about 4 inches (102 mm) above the top of the pump casing
flange. The damaged studs were replaced. The leakage that caused the corrosion was
attributed to failures in welds in nearby instrument lines.

2.3.4.3 Stress Corrosion Cracking of Alloy 600 Materials

The greatest increase in the rate of occurrence of leaks since the original Boric Acid Corrosion
Guidebook in 1994 has been the large number of leaks from cracked Alloy 600 nozzles and,
more recently, welds at VC Summer and Oconee 1 and 3. However, despite the large number of
such leaks, only one case has been reported in which a significant amount of material loss on the
low-alloy steel vessel shells occurred. The main reasons for the general lack of significant
corrosion damage are:

2-12
 EPRI Licensed Material

Industry Experience

• The amount of leakage is typically very small due to the tight PWSCC cracks.
• There is little oxygen in the crevice between the nozzle and the clearance hole in the vessel
shell.
• The small amount of water that does reach the surface of the vessel shell quickly evaporates.

The following are lead examples of the types of cracking that have occurred due to PWSCC of
Alloy 600 materials:
• Pressurizer Heater Sleeves: Arkansas One 2 (1987) – Boric acid crystals were discovered
under the pressurizer during a routine operating condition inspection. The leakage was
reported to be about 60 drops per minute (0.002 gpm or 0.008 L/min) at 570°F (299°C) and
2250 psi (15.5 MPa) internal pressure. The leakage was traced to PWSCC of the Alloy 600
heater sleeve caused by swelling of a failed pressurizer heater inside the sleeve. The leakage
resulted in corrosion of the low-alloy steel (SA-533 Grade B Class 1) pressurizer head. The
extent of the corrosion was a volume 1.5 inches (38 mm) in diameter and 0.75 inch (19 mm)
deep (18% through-wall). The time over which the leakage occurred is not known, but it was
reported that the leak had not been discovered during an inspection about six months earlier.
The repair effort included replacing the defective heaters, plugging the leaking nozzle, and
weld repairing the corroded area on the vessel head.
• Instrument Nozzles: Numerous Combustion Engineering Plants – Numerous cases of leaks
from cracked Alloy 600 instrument nozzles in pressurizers and hot leg piping applications
have been reported. These have occurred mostly in Combustion Engineering plants because
Westinghouse tended to use stainless steel nozzles and the nozzles in B&W plants were
stress relieved after welding the nozzles into the pressure boundary parts. In most cases, these
leaks were discovered during boric acid walkdowns per Generic Letter 88-05. In none of the
cases has there been any significant corrosion of the pressure boundary base metal.
• Nozzle Safe End Welds: VC Summer (2000) – The lead case involving cracked Alloy 182
welds is the incident at VC Summer in 2000. In this case, a through-wall axial crack in the
weld between a carbon steel hot leg nozzle and a stainless steel pipe resulted in more than
200 pounds (91 kg) of boric acid crystals being deposited near the joint. There was some
visible evidence of boric acid corrosion on the carbon steel nozzle material, but it was not
measurable.
• CRDM Nozzle Weld: Oconee 1 (2000) – Leakage was discovered from the annulus between
one of the Oconee 1 CRDM nozzles and the reactor vessel head. The total quantity of boric
acid crystals was about 0.5 in 3 (8.2 cm3). There was no evidence of boric acid corrosion. The
leakage was traced to a PWSCC crack in the J-groove weld.
• CRDM Nozzle Weld: Oconee 3 (2001) – Leakage was detected from the annulus between
several CRDM nozzles and the reactor vessel head. The leakage was detected by visual
inspection of the head and was reported to be small. No corrosion of the low-alloy vessel
head material was reported.

Summary: Although there have been many reports of leaks from PWSCC of Alloy 600 base
metal and more recently Alloy 82/182 welds, the only case resulting in any measurable corrosion
was at ANO-2 where the heater sleeve was expanded by a failed heater. In the other case, the

2-13
 EPRI Licensed Material

Industry Experience

leakage was extremely small, or in the case of VC Summer, the temperature was high enough to
prevent moist boric acid from remaining in contact with the carbon steel base materials.

2.4 Other Locations Involving Consequential Damage

Section 2.3 described cases in which damage resulting from leakage of borated water occurred in
pressure-containing parts in reasonably close proximity to the source of the leakage. However,
there have also been cases in which the leakage has damaged electrical components, HVAC
components, equipment supports, and the containment liner. These cases show the importance of
checking for boric acid and its resultant degradation throughout the plant.

2.4.1 Electrical and Control Components

The largest number of reported incidents of electrical equipment damage has been to heat tracing
on lines containing high concentrations of borated water. However, other electrical components
and their electrical connections have also been damaged. The following are several examples that
illustrate the potential for damage to electrical and control components:
• Palisades (1987) – While zeroing control rod drive packages during cold shutdown testing, a
noise spike occurred in a wide range power channel neutron sensor. The noise spike resulted
in indications of high startup rate, low steam pressure, and low primary coolant system flow
and initiated a thermal margin/low pressure trip of the reactor coolant system. The noise
spike was traced to a highly corroded cable connector pin at the interface between the
detector element and high voltage leads. The corrosion is suspected to have been caused by
exposure to a boric acid environment although the source of the boric acid environment was
not reported.
• Shearon Harris (1987) – A low pressurizer level alarm was received during a cold shutdown.
The alarm was traced to boric acid corrosion of a wire in a pressurizer level transmitter. The
boric acid was found to have leaked from a faulty environmentally qualified seal in the
transmitter. Repairs included replacing the seal on the transmitter, removing boric acid
crystals from the terminal blocks, and reconnecting the wires.
• Beaver Valley 1 (1983) – A power range flux detector failed causing the plant to reduce
power to meet Technical Specification requirements. The failure was traced to boric acid
corrosion of the high voltage connector. The source of the boric acid was not reported. The
high voltage connector was replaced.

2.4.2 HVAC Components

There have been several undocumented cases where leaking steam with some boric acid
carryover has condensed at containment air coolers. In one case reported by Calvert Cliffs,
decreased efficiency of the containment air coolers led to the discovery that the cooling coils
were plugged with boric acid deposits and that the leakage had come from a reactor vessel head
instrument nozzle Conoseal connection.

2-14
 EPRI Licensed Material

Industry Experience

2.4.3 Component Supports

There are several reported incidents in which boric acid has corroded component supports. Two
examples are:
• Zion 2 (1988) – Boric acid residue was discovered on a component snubber in the auxiliary
building during a refueling outage. The boric acid had caused some corrosion of the snubber
body that could possibly have affected its proper functioning. The corroded snubber was
replaced.
• Ft. Calhoun (1992) – Leaks were discovered from the flange region of both boric acid pumps
during a refueling outage. This leakage was found to have corroded the carbon steel flange
bolts and some of the carbon steel U-bolts in adjacent piping supports. It was subsequently
determined that corrosion of these parts could have led to failure of both boric acid pumps
during a seismic occurrence. The root cause was identified as the failure to anticipate
corrosion of the carbon steel parts as a result of leakage at the pump flange connections.

2.4.4 Containment Liner

Corrosion of the containment liner at all four Duke Power units with ice condenser containments
has been reported. A typical example is as follows:
• McGuire 2 (1989) – Corrosion was discovered at the bottom of the containment liner. The
corroded areas were up to 30 feet (9 m) in length and had a depth up to 0.1 inch (2.5 mm) out
of an original 1-inch (2.5 mm) thickness. The corrosion greatly exceeded the original design
corrosion allowance. The probable cause of the corrosion was reported as attack by borated
water leaking from instrument line compression fittings that condensed on the containment
liner. Drains had been provided in the containment building floor, but the floor was not
sufficiently graded to prevent pooling of the condensate between the drain locations. The
NRC expressed concern that corrosion found in all four Duke Power PWR plants (ice
condenser design) indicates that similar corrosion could occur in the liners of other PWR
plants.

2.4.5 Summary Regarding Other Components

The above experience points out that leaking boric acid can be spread throughout the
containment building primarily by the HVAC system. Boric acid spread in this manner can come
into contact with electrical equipment, HVAC components, component supports, containment
liners, and other components. Because it is difficult to track down all locations and components
potentially affected by a boric acid leak, primary emphasis should be directed towards reducing
leakage to the greatest extent possible.

2.5 Conclusions from Field Experience

The above summary illustrates that leaking borated water has the potential to cause significant
damage, force plant shutdowns, and extend refueling outages. The major conclusions reached
regarding field experience are:

2-15
 EPRI Licensed Material

Industry Experience

• Small undetected leaks have the potential to cause larger leaks and significant damage.
• All leakage should be treated seriously and the potential direct and indirect consequences
evaluated.
• Components that are potential sources of leakage should be identified and evaluated to
determine the potential consequences of any such leakage.
– If potential leakage does not pose a concern, no changes are required.
– If potential leakage does pose a concern, changes may be required in the inspection
intervals, or design or maintenance changes may be warranted to reduce the potential for
leakage or to reduce the potential for degradation if a leak should occur.
– Particular emphasis should be placed on the potential for corrosion of carbon or low-alloy
steel bolting materials. There are numerous cases where small amounts of leakage have
corroded carbon and low-alloy steel bolting materials to the extent that the leak rates have
increased significantly. These incidents have led to plant shutdowns and the need for
expensive repairs.

Figure 2-1
Schematic Diagram of Main PWR Systems Containing Borated Water

2-16
 EPRI Licensed Material

Industry Experience

Figure 2-2
Distribution of Reported Leakage Incidents by Source

2-17
 EPRI Licensed Material

Industry Experience

Figure 2-3
Locations of Boric Acid Corrosion in Primary Loop Components

2-18
 EPRI Licensed Material

Industry Experience

Figure 2-4
Typical Steam Generator Manway Closure

2-19
 EPRI Licensed Material

Industry Experience

Figure 2-5
Typical Reactor Coolant Pump Casing

2-20
 EPRI Licensed Material

Industry Experience

Figure 2-6
Typical Bolted CRDM Nozzle (B&W Design)

2-21
 EPRI Licensed Material

Industry Experience

Figure 2-7
Typical Nuclear Service Valves [18]

2-22
 EPRI Licensed Material

Industry Experience

Figure 2-8
Typical Reactor Head Instrument Nozzle Conoseal Closure

2-23
 EPRI Licensed Material

Industry Experience

Figure 2-9
Corrosion of Reactor Vessel Head at Turkey Point 4 [1]

2-24
 EPRI Licensed Material

Industry Experience

Figure 2-10
Cross Section Through Typical Reactor Vessel Closure Flange

2-25
 EPRI Licensed Material

Industry Experience

Figure 2-11
Typical CRDM Nozzle (Canopy Seal Weld Design)

2-26
 EPRI Licensed Material

3
BORIC ACID CORROSION MECHANISMS

This section describes the mechanisms by which carbon and low-alloy steel materials corrode in
the presence of borated water.

3.1 Overview of Corrosion Mechanisms

Corrosion is defined as the degradation of metals that results from an interaction with the
environment. During corrosion, a metal reacts with a second species to form a chemical
compound that is more thermodynamically stable than the metal alone. The second species,
typically oxygen, comes in contact with the metal through a medium that can be liquid, solid, or
gaseous. When corrosion occurs, the compound formed is the corrosion product. The medium
from which the corrosive species is supplied is the corrosive medium.

The chemical form of the corrosion product is often very similar to the metallic oxides or mineral
ores that are found in nature. This is because over time, metals have combined or reacted with
species in the surrounding environment and transformed into a stable, low energy state. For
example, rust, which is produced by corrosion of iron, is similar to the iron compounds found in
iron-bearing ore. Corrosion can be thought of as a process that is the reverse of the refining
process by which the metal is produced for industrial use.

Although the details of corrosion processes can be quite complex, much can be learned by the
simple characterization of the categories of corrosion and the development of a qualitative
understanding of the chemical phenomena involved. In most systems where metals are exposed
to aqueous (water) environments and corrosive species (oxygen, acids, chlorine, etc.), the
corrosion phenomena can be divided into two general categories: uniform corrosion and
localized attack. These categories are shown in Figure 3-1 along with the main subcategories
associated with each.

The chemical reactions that occur during corrosion are actually electrochemical processes
because they involve the gain, loss, and/or exchange of electrically charged species such as
metallic ions (for example, Fe ++), hydrogen ions (H+), and electrons (e-). For instance, the
corrosion of iron occurs via oxidation (loss of electrons) to produce ferrous ions according to the
following:
Fe ––––> Fe++ + 2e-
The ferrous ions are ejected into the corrosive media, but the electrons remain in the metal. The
segregation of these species results in a separation of electric charge; the ferrous ions produce a
positively charged layer at the solution interface, and the electrons produce a negatively charged
layer in the surface of the metal. Because the electrons and ferrous ions are oppositely charged,

3-1
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

there is an attractive force between them creating what has been termed a double layer. In the
absence of any diffusion or consumption of the electrons and ferrous ions, the formation of the
double layer is self-limiting in that any newly formed ferrous ions are repelled by those attached
to the surface above due to their like charge. In this case, the oxidation process would proceed
only to the extent that an extremely thin layer of ferrous ions would collect at the surface of the
carbon or low-alloy steel. This condition is illustrated in Figure 3-2.

In most cases, continuation of the corrosion process beyond this initial production of ferrous ions
requires participation by a third species. As an example of such a “third party” participant, it is
useful to examine the corrosion of iron and steels and discuss the role in the rusting process of
water, protons, and therefore, pH.

It is well known that iron will not rust in dry air or in water that is completely free from oxygen.
Therefore, it would appear that both water and oxygen are required for the formation of
significant amounts of rust. Furthermore, it is known that the rusting process is accelerated by
the presence of acids, strains in the metal, “contact” with other metals, geometry, and the
presence of rust itself (called autocatalysis).

The overall rusting process involves the reaction of two of the ferrous ions discussed above with
three oxygen atoms (O) provided via the corrosive medium to form hematite or Fe2O3. Two
processes must occur for the rusting process to proceed. First, in order for iron to combine
chemically with the oxygen, it must be “attracted” or bonded to the oxygen in some way. This
attraction is electrical. Iron prefers to exist with a positive charge (ferrous ions), but oxygen
prefers to exist with a negative charge when it forms a compound. Second, the electrical double
layer described earlier must be continuously suppressed.

Figure 3-3 shows the common way that the electrons can be consumed through reaction with the
positively charged hydrogen ions in water. These hydrogen ions (or H +) are measured when the
pH of the water is checked (the lower the pH, the greater the concentration of hydrogen ions).
Figure 3-3 shows the consumption of the electrons by the hydrogen ions (protons) in a thin layer
adjacent to the metal surface.
+
Not surprisingly, an increase in the concentration of hydrogen ions, H (lower pH or more
acidic), tends to increase the rate of removal of free electrons at the surface and, therefore,
accelerate the rate of corrosion. In addition, some of the hydrogen atoms may further combine to
form hydrogen gas or H2. Hydrogen gas is often detected in ferrous systems that are experiencing
corrosion. The hydrogen atoms can also react with oxygen to form water.

When electrons are removed from the surface, the ferrous ions are free to react with oxygen in
the water to form rust, as follows:
+2
4Fe + 3O2 (as dissolved gas in water) ––––> 2Fe2O3 (Rust)

This process is illustrated in Figure 3-4.

In addition, water (H2O) participates in the reaction along with the oxygen (O 2). The water helps
to:

3-2
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

• Provide some of the oxygen required for the formation of Fe 2O

• Regenerate the hydrogen ions consumed by the process

After time, a continuous layer of rust will cover the surface of the metal, as seen in Figure 3-5.
This layer can protect the surface from additional corrosion attack by limiting the exposure of
fresh metal to the corrosive media. To be protective, layers of rust must be tightly adherent.

3.2 Dominant Boric Acid Corrosion Type – Wastage

Overall, the corrosion type of greatest concern due to leakage of borated water is uniform
corrosion, or wastage. A brief description of the main factors involved in wastage follows.

Boric acid in water can result in an aggressive environment that uniformly attacks the surface of
metal. Although boric acid is considered a “weak” acid when compared to acids such as
hydrochloric or nitric acid, boric acid in water will still increase the concentration of hydrogen
ions (H+). Referring to Figure 3-3, it can be seen that an increase in the hydrogen ion
concentration, which would be seen as a drop in pH, will increase the ability of the water
medium to consume electrons. This, in turn, increases the achievable rate of formation of ferrous
ions and the rate of corrosion.

In reality, the process is further accelerated by what amounts to miniature galvanic “cells.”
Figure 3-6 illustrates the process by which small areas of the metal surface behave cathodically
relative to other areas of the surface and, thereby, attract electrons from the adjacent material.
These local cathodes can occur due to slight changes in the alloy composition (for example,
higher chromium or nickel in steel), surface imperfections or defects, or surface strain. In
addition, the corrosion product itself may be cathodic to the surrounding base metal.

The anodic/cathodic areas on the metal surface can shift with time, resulting in an essentially
uniform rate of attack over the entire metal surface. One of the reasons that the rate of attack of
the unprotected, exposed steel can be so great is that the exposed area of the cathodic surfaces
often exceeds that of the anodic material.

Surfaces attacked by wastage often appear to have some increase in texture as a result of small
differences in the rate at which different areas of the surface are attacked. It is sometimes
difficult to detect wastage of a surface because there is no clear reference point for assessing the
amount of material loss, particularly if the rust is continually solubilized by or entrained into the
fluid stream.

Wastage is generally the easiest form of corrosion to predict using experimental data. However,
as with all forms of corrosion, the rate can be affected by significant factors that were not
included in the experiments used as the basis for the predictions. Some of the factors that can
significantly affect the corrosion rate are discussed in the following paragraphs.

3-3
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

3.2.1 Effect of Impurities on Wastage

When impurities are present in the metal, they may act as the source of local cathodes that
accelerate corrosion as shown in Figure 3-6.

When impurities are present in the corrosive medium, they can have a variety of effects on the
corrosion rate. Impurities may act to increase the conductivity of the aqueous solution. This
increased conductivity promotes migration of electrons, protons, and ferrous species in the fluid
phase, thereby increasing corrosion rates.

Impurities can also cause increases in corrosion rates by affecting or destroying the passive,
protective layer of hematite (rust) or other metal oxide that builds on the exposed surface of the
metal and acts to protect the metal from further damage. A good example of this is the
destruction of passive layers on stainless steel by chlorides. In other cases, impurities in the fluid
steam can actually help retard corrosion. In these cases, the impurities act as inhibitors.

In any event, the effect of impurities is generally complex and non-linear and must be
determined through experiments designed to simulate the environment and the concentration of
impurities to which the steel is exposed.

3.2.2 Effect of Oxygen on Wastage

Based on the model of corrosion shown in Figure 3-4, it can be seen that higher oxygen
concentration in the corrosive media can increase the rate at which 2Fe 2O3 is formed at metal
surfaces. This results in a higher corrosion rate. This is confirmed by the test data in Figure 3-
7(a) from the Handbook of Corrosion Engineering [19].

However, only about 100 ppb of oxygen results in the same corrosion rate as fully oxygenated
water.

3.2.3 Effect of Temperature on Wastage

In almost all cases, corrosion rates increase with increasing temperature. Temperature also has a
secondary effect on decreasing corrosion as the concentration of dissolved oxygen in the water
decreases with increasing temperature. At atmospheric pressure, high temperatures boil the water
away, leaving more concentrated acid and higher corrosion rates. However, as the temperature
rises further, the water is boiled off completely, leaving dry boric acid crystals that are not highly
corrosive.

3.2.4 Effect of pH on Wastage

Most metals that are soluble in acids will exhibit a pH-corrosion rate dependence similar to that
shown in Figure 3-7(b). In the middle pH range of 4 to 10, the corrosion rate is generally
controlled by the rate of diffusion of oxygen to the surface. At lower pH, the corrosion rate
increases as shown in Figure 3-7(b), and at higher pH, more basic conditions, the corrosion rate
also increases due to different phenomena that are not described in this section.

3-4
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

3.2.5 Effect of Flow Velocity on Wastage

In many of the field reports, general corrosion (wastage) is accelerated by the impingement of
borated water or steam with boric acid carryover onto hot metal surfaces. This impingement has
the dual effect of removing protective corrosion films from the surface of the metal and
replenishing the corrodent with fresh oxygenated corrodent. Both of these factors can increase
the corrosion rate.

3.3 Secondary Types of Boric Acid Corrosion

In addition to wastage (the predominant type of boric acid corrosion), there have been reports of
other types of corrosion, such as galvanic corrosion, crevice corrosion, pitting, intergranular
attack, and stress corrosion cracking, under a few limited conditions.

3.3.1 Galvanic Corrosion

All metallic materials have an electric potential as illustrated by Figure 3-8. When two dissimilar
metals are placed in direct contact in a conductive environment, the difference in electric
potential between the two materials causes an electric current to flow between the two materials.
This current leads to removal of electrons from the metal surface, and this results in the
accelerated corrosion of the more active of the materials and reduced corrosion of the more noble
material. The rate of corrosion is generally greatest at the interface between the two dissimilar
materials. This effect is shown in Figure 3-9.

A practical example of such a condition in an aerated, borated water environment in a PWR plant
would be a weld between a low-alloy steel pipe and a stainless steel pipe. The stainless steel is
the more noble material, and the low-alloy steel is the more active material. The low-alloy steel
in such an application would experience higher corrosion rates than in borated water without the
galvanic couple.

Field experience (ANO-1 HP1 nozzle incident in 1986) and laboratory test results (Test
Reference J – see Section 4) both indicate that a galvanic couple between carbon steel and
stainless steel can accelerate the local corrosion rate in the carbon steel. The local acceleration
factor in both of the above cases was about 1.5 times. Therefore, the galvanic effect is significant
but not overwhelming.

3.3.2 Crevice Corrosion

Environmental conditions in a crevice can be significantly different than on adjacent bare metal
surfaces. Under some conditions, the environment in the crevice can be more aggressive than
outside the crevice, and accelerated local corrosion can take place. Typical crevice locations on
the outside surfaces of PWR pressure boundary parts include bolts, washers, and gaskets.

3-5
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

Crevice corrosion is attributed to one of several conditions:

• Change in acidity within the crevice
• Change in oxygen content within the crevice
• Buildup of detrimental ion species within the crevice

Crevice corrosion has not generally been reported to be a significant practical problem in borated
water environments on the outside surfaces of PWR components. However, this is not always the
case, particularly when the part or component includes crevices. Figure 3-10 depicts a crevice
geometry and illustrates a situation where low oxygen concentration can actually increase
corrosion and result in pitting attack.

The presence of the crevice creates a local region that cannot be reached by oxygen. In this case,
the ferrous ions diffuse out of the crevice continually, where they combine with oxygen at a
point just outside the crevice mouth. The process can continue indefinitely, resulting in a deep pit
in the local oxygen-deficient region.

Crack growth in carbon steel material beneath stainless steel or nickel base alloy cladding can be
accelerated by both galvanic effects and crevice effects. If a crack grows through a stainless
cladding overlay so that the environment can reach the less-corrosion-resistant substrate, the
cladding can act as a cathode, accelerating dissolution of the more anodic substrate. In addition,
the tight crack through the cladding restricts the entry of oxygen into the region below the
cladding as in a conventional crevice illustrated in Figure 3-10. Therefore, a protective oxide
layer cannot be formed on the surfaces of the portion of the crack in the substrate. Instead, the
ferrous ions from the substrate migrate through the crack in the cladding and combine with
oxygen at the outside surface of the cladding. These effects accelerate growth of the crack in the
substrate by the mechanism of anodic dissolution of the crack tip. These effects are shown
schematically in Figure 3-11.

3.3.3 Pitting

Pitting is the formation of localized depressions or holes in a metal surface. Pitting typically
occurs at inclusions that are exposed at the metal surface or at segregated components of the
alloy. In some cases, these locations are the sources of localized galvanic attack.

Pitting has not been reported to be a significant problem in borated water environments on the
outside surfaces of PWR components.

3.3.4 Intergranular Corrosion

Intergranular corrosion is a localized attack along the grain boundaries of a metal or alloy.
Corrosion can proceed to the point where whole grains of metal separate from the surface and the
metal strength is reduced. This is analogous to the case of a crack running in the mortar between
bricks in a brick wall.

3-6
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

Intergranular corrosion occurs most commonly in austenitic stainless steel materials when an
improper heat treatment or heat from welding causes the precipitation of chromium carbides at
the grain boundary. This precipitation at the grain boundaries causes a depletion of chromium in
the material near the grain boundaries. This zone of depleted chromium is anodic to the rest of
the material in the grains and is the site of accelerated corrosion.

There have been several cases over the years involving intergranular corrosion of sensitized
stainless steel in PWR environments. Some of these cases have involved corrosion of the inside
surfaces of stainless steel pipes holding stagnant high concentration borated water. In other
cases, stainless steel has corroded on the outside surface under the insulation. It is reported that
most of these cases also involved chloride contamination, so it is not clear that the borated water
actually played an aggressive role.

There are no reports of intergranular corrosion being a significant contributing factor to problems
associated with corrosion due to leakage of borated water.

3.3.5 Stress Corrosion Cracking

Under some conditions, the potential for corrosion of a material’s grain boundaries can be
worsened by the application of a tensile stress. The tensile stress acts to open the crack to allow
the corrosive environment to reach farther into the material surface and can create very high
stress intensities at the crack tip.

Stress corrosion cracking in the presence of borated water leakage has been reported to be a
problem only in the case of highly loaded steam generator manway studs that have been coated
with lubricants containing sulfur. The most celebrated case occurred at Maine Yankee where 6 of
16 studs were found to have failed after disassembly, and 5 more were found to be cracked. The
stress corrosion cracking in this case was attributed to interaction between the leaking borated
water, leak sealant, and sulfur-containing thread lubricant. This experience pointed out the need
to minimize the use of sulfur-containing compounds around pressure boundary parts.

3-7
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

Figure 3-1
Forms of Corrosion

3-8
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

Figure 3-2
Formation of Initial Double Layer During the Onset of Corrosion

Figure 3-3
Consumption of Electrons During Corrosion

3-9
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

Figure 3-4
Conversion of Ferrous Ions to Hematite (Rust)

Figure 3-5
Formation of a Continuous Corrosion Product Layer

3-10
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

Figure 3-6
Enhanced Corrosion due to Local Galvanic Cells

3-11
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

Figure 3-7
Effect of pH and Oxygen on Corrosion Rate [19]

3-12
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

Figure 3-8
Galvanic Series of Metals and Alloys [20]

3-13
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

Figure 3-9
Galvanic Corrosion at Dissimilar Metal Couples

Figure 3-10
Crevice Corrosion

3-14
 EPRI Licensed Material

Boric Acid Corrosion Mechanisms

Figure 3-11
Crevice and Galvanic Effects Beneath Corrosion-Resistant Cladding

3-15
 EPRI Licensed Material

4
BORIC ACID CORROSION TESTS

This section summarizes the results of test work performed by several organizations to determine
corrosion rates of PWR components in borated water environments.

A review of the field experience in Section 2 of this guidebook suggests that corrosion by
borated primary water can be broken down into five basic conditions. These conditions are
illustrated in Figure 4-1. The following is a brief description of the sources of test results for each
of the basic conditions, and a discussion of the corrosion rates that can be inferred from the test
work that has been conducted related to each condition. Key test results are plotted in Figures 4-
2 and 4-3, and overall conclusions are discussed in Section 4.9.

4.1 Sources of Boric Acid Corrosion Testing

There are two main sources of boric acid corrosion test data: data from a number of sources that
were available at the time the original Boric Acid Corrosion Guidebook was published in 1995
and supplemental testing conducted by EPRI in 1996–98 to provide missing information,
including results on prototypical bolted flange joints.

4.1.1 Test Data Available Prior to 1995

At the time the original Boric Acid Corrosion Guidebook was published, significant testing had
already been performed by several organizations including Westinghouse, Combustion
Engineering, Babcock & Wilcox (B&W), Brookhaven National Laboratory, TVA, and the
Moscow Power Institute. The available test data and the conclusions from these tests were
summarized in the original guidebook. The tests were referenced as Tests “A” through “M.” The
same test designations are maintained in this revised report, and summaries of these tests are
included at the end of this section.

4.1.2 Supplemental Testing Conducted by EPRI

A review of the available test data in 1995 indicated several areas where desirable information
was missing. In 1996–98 EPRI, on behalf of three co-funding utilities (Duquesne Light
Company, Duke Power Company, and Pacific Gas and Electric Company), conducted additional
testing at Southwest Research Institute (SwRI) to obtain important missing information. This
supplemental testing consisted of two main parts:

4-1
 EPRI Licensed Material

Boric Acid Corrosion Tests

• The first series of tests was to determine the relative corrosion rates of several different
alloys typically encountered in nuclear plants including carbon and low-alloy steel bolting
materials, corrosion-resistant bolting materials, and typical pressure vessel shell materials.
These materials were tested by:
– Immersion in aerated borated water at 20,000 ppm boron at 200ºF (93ºC) for 60 days
– Contact with borated water spray at 2000 ppm boron at 100ºF (38ºC) for 60 days
– Contact with boric acid crystals at high humidity and room temperature for 60 days
These tests were expected to highlight the relative performance of a wide range of materials
commonly used in PWR plants under typical boric acid exposure conditions.
• The second series of tests involved simulation of several prototypical operating conditions as
illustrated in Figure 4-4. Each of the tests was performed with borated water at 1000 ppm
boron concentration (typical average cycle boron concentration) and flow rates of 0.01 and
0.10 gpm (0.038 and 0.38 L/min). The water in the EPRI tests was deaerated by bubbling
argon through the water, which resulted in achieving an oxygen concentration of 0–80 ppb.
This concentration is significantly higher than in reactor primary water and, therefore, gives
conservative corrosion rates compared with those that might be expected under normal
operating conditions at a plant.

The tests performed by SwRI are designated as EPRI-1, EPRI-2, etc., to distinguish them from
the previously existing test data. Data sheets summarizing the EPRI tests are also provided at the
end of this section.

4.2 Material Screening Tests

It has been recognized for years that carbon steels are susceptible to corrosion in aerated boric
acid solutions and that stainless steels are much more resistant to corrosion in this environment.
As part of the supplemental EPRI-sponsored testing, a full range of typical bolting and pressure
vessel steels was subjected to immersion in aerated borated water solutions and to contact with
boric acid sprays and boric acid crystals. Results of these tests are summarized in Figure 4-3.
These tests are as follows:

1. EPRI Tests in Concentrated Boric Acid at 180°F (82°C)

(See Test Reference EPRI-1)

EPRI conducted tests in which seven carbon and low-alloy steels and three stainless steels
were immersed in an aerated 20,000 ppm (as boron) boric acid solution at 180°F (82°C) [21].
These tests were performed to simulate primary coolant that has leaked onto a moderately
heated metal surface and has been significantly concentrated by evaporation. Duplicate
coupons of each material were exposed for 15-, 30-, and 60-day periods to allow an
evaluation of the variation in average corrosion rate with exposure time.

All of the carbon and low-alloy steels exhibited high corrosion rates in the aerated 20,000
ppm B, 180°F (82°C) immersion environment. Average corrosion rates ranged from 0.079–
0.165 in/yr (2–4 mm/yr) for the carbon and low-alloy steels at 60-day exposures.

4-2
 EPRI Licensed Material

Boric Acid Corrosion Tests

Of the three stainless steels evaluated, only the 410 stainless steel exhibited measurable
corrosion with an average rate of 0.0003 in/yr (0.008 mm/yr) over the 60-day exposure. The
other two stainless steels (A286 and 17-4PH) were unaffected in the concentrated, aerated
boric acid.

2. EPRI Tests in Boric Acid Spray

(See Test Reference EPRI-2)

EPRI conducted a series of tests in which a 2000 ppm (as boron) boric acid, aerated solution
at 100°F (38°C) was vaporized and deposited onto specimens in a standard salt spray
chamber [21]. Seven carbon and low-alloy steels and three stainless steels were exposed for
15-, 30-, and 60-day periods. These tests simulate leaking primary coolant that vaporizes,
becomes airborne, deposits on surfaces at low temperatures, and is moderately concentrated
by evaporation.

The carbon and low-alloy steels exhibited corrosion rates of 0.003–0.004 in/yr (0.08–0.1
mm/yr) over the 60-day exposure. The three stainless steels were essentially unattacked in
the spray environment.

Because the rates for the carbon and low-alloy steels are a factor of 2–3 lower than those
obtained from immersion tests in 2000 to 2500 ppm boric acid at 100–104°F (38–40°C)
(Tests A and B), it appears that the spray environment is not as severe as total immersion in
aerated solutions.

3. EPRI Tests in Boric Acid Crystals

(See Test Reference EPRI-3)

EPRI conducted a series of tests in which seven carbon and low-alloy steels and three
stainless steels were buried in boric acid crystals at room temperature exposed to water-
saturated flowing air [21]. Specimens were buried in vertical positions in the boric acid
crystals with one end approximately 3/8 inch (10 mm) below the surface of the crystals and
the other end 2-3/8 inches (60 mm) below the surface. Duplicate specimens were removed
after 15-, 30-, and 60-day exposures. These tests were designed to simulate a condition in
which leaking primary coolant has deposited on metal surfaces, evaporated, and precipitated
boric acid crystals on the metal surfaces in a high-humidity environment.

All specimens that were buried in boric acid crystals in the 100% relative humidity
environment exhibited very low corrosion rates in each of the three exposure periods. The
average corrosion rates for carbon and low-alloy steel during the 60-day tests were 0.00004–
0.0002 in/yr (0.001–0.005 mm/yr). The corrosion rates for the stainless steel materials were
all negligible. These test results suggest that corrosion rates of carbon, low-alloy, and
stainless steels will be very low under conditions where their surfaces are covered by boric
acid crystals.

4-3
 EPRI Licensed Material

Boric Acid Corrosion Tests

4.3 Immersion of Carbon Steel in Deaerated Borated Water

This is the condition in which carbon and low-alloy steel materials are exposed to borated water
inside the system pressure boundary. The most important factors are that the boric acid
concentrations are the levels specified in the plant Technical Specifications and the oxygen
levels are very low. Carbon and low-alloy steel surfaces are typically exposed to borated water
inside the pressure boundaries only at locations where the corrosion-resistant cladding has been
damaged or where carbon or low-alloy steel parts have been installed inadvertently.

Several sets of tests have been conducted for these conditions.

1. Westinghouse Tests of Low-Alloy Steel in Aerated and Deaerated Water

(See Test Reference A)

In the mid-1960s, Westinghouse conducted a series of tests to determine the corrosion rate of
A-302 Grade B low-alloy steel in aerated and deaerated water with 2500 ppm boron at low
temperatures [22, 23]. These tests showed corrosion rates less than 0.001 in/yr (0.025 mm/yr)
in deaerated environments at temperatures up to 140°F (60°C).

2. Moscow Power Institute Tests in Deaerated Water

(See Test Reference B)

The Moscow Power Institute conducted a series of tests to evaluate the effects of boric acid
solutions on fuel assembly materials and carbon structural steel [24, 25]. One series of tests
involved immersing plate type carbon steel specimens in aerated and deaerated borated water
for up to 1000 hours. Results were obtained at intervals to determine the reaction kinetics.

The Moscow Power Institute testing showed long term corrosion rates of 0.0002 in/yr (0.005
mm/yr) in water with 3000 ppm boron at 590°F (310°C) and 0.0003 in/yr (0.008 mm/yr) in
water with 1000 ppm boron at 392°F (200°C). The test data showed 2–3 times higher
corrosion rates for test durations in the 100-hour range, that is, the corrosion rate decreased
with time

3. TVA Tests of Bolts in Low Oxygen Water

(See Test Reference C)

TVA conducted a series of tests in 1983 to determine the corrosion rates of bolting material
in borated water [26]. The tests involved preloading 1/2-inch- (13-mm-) diameter, A-193
Grade B7 bolts and immersing them in borated water in autoclaves at 550°F (288°C) for 30
days. Tests were conducted with boric acid at a pH of 4.0, with lithium hydroxide buffered
boric acid at pH of 10.5, and with wetted solid boric acid with 200 ppm chloride. Two
fasteners in each autoclave were immersed in the liquid phase, and two were in the vapor
phase.

The maximum weight loss for the fasteners was typically 0.1–0.2 grams over the 30-day
period. This corresponded to a uniform corrosion rate of about 0.003–0.006 in/yr (0.08–0.15
mm/yr) over the entire surface of each bolt. However, it must be noted that this test condition
falls between aerated and deaerated conditions because oxygen was in the autoclave at the

4-4
 EPRI Licensed Material

Boric Acid Corrosion Tests

start of the testing, but no oxygen was added during the testing. As such, the testing does not
represent either aerated or deaerated conditions. The testing specifically does not represent
the aerated condition under which alloy steel bolts would normally experience corrosion as a
result of leaking of borated water at joints.

4. Combustion Engineering Autoclave Tests of Pressure Vessel Steels

(See Test Reference D)

Combustion Engineering conducted a series of tests to determine corrosion rates of unclad

low-alloy pressure vessel steels under plant startup conditions [27]. Plate specimens of SA-
533 Grade B, SA-508 Class 2, and SA-508 manually welded to SA-533 Grade B were
immersed in an autoclave with deionized water and 723 ppm boron, 1.8 ppm lithium, and 0.4
ppm ammonia. The autoclave was not deaerated prior to start of the testing, and N2H4 was not
added to scavenge oxygen. The specimens were held at 350°F (176°C) for 70 hours.

The results of this test showed that all materials experienced average corrosion rates of
0.016–0.017 in/yr (0.41–0.43 mm/yr). There was no preferential attack of the weld metal or
heat-affected-zone, and none of the specimens experienced localized attack.

5. EPRI Tests in a Deaerated Solution of Concentrated Boric Acid

(See Test Reference EPRI-4)

EPRI conducted a series of tests to determine the corrosion rate of low-alloy steel exposed to
a deaerated, concentrated boric acid solution [21]. Specimens of A193 Grade B7 (4140 steel)
were immersed in deionized water containing 20,000 ppm boron for 15, 30, and 60 days. The
boric acid solutions were held at 180°F (82°C) and deaerated continuously with flowing
argon.

The corrosion rates of the 4140 steel immersed in the hot, concentrated boric acid solution
varied from 0.050 in/yr (1.3 mm/yr) after 15 days to 0.027 in/yr (0.7 mm/yr) after 60 days.
These results, when compared to results from aerated, concentrated solutions, clearly show
that the presence of oxygen in the environment greatly accelerates the corrosion of carbon
and low-alloy steels. However, these results also show that hot, deaerated, concentrated boric
acid, alone, can cause measurable general corrosion of carbon and low-alloy steels at the
oxygen levels of 0–80 ppb achieved during the tests.

In summary, these sources show the corrosion rates for carbon and low-alloy steels in solutions
simulating normal primary water to be less than 0.001 in/yr (0.025 mm/yr) for conditions where
the oxygen concentration is very low throughout the test, and in the range of 0.006–0.017 in/yr
(0.15–0.43 mm/yr) for cases where oxygen is present at the start of the test but no new oxygen is
added during testing. In water deaerated by bubbling argon, corrosion rates are considerably
higher, ranging from 0.027–0.050 in/yr (0.69–1.3 mm/yr). The rates in the solutions containing
the lower boron concentrations are clearly too low to pose a practical concern with rapid short-
term corrosion in the normal primary water environment. Corrosion rates for carbon and low-
alloy steels would be expected to accelerate significantly in any areas where concentrated
solutions can form in the presence of up to 80 ppb oxygen. These data are most useful to assess

4-5
 EPRI Licensed Material

Boric Acid Corrosion Tests

corrosion of carbon steel materials at defects in cladding and to serve as a reference point for the
accelerating effects of factors such as oxygen, impingement, and dripping onto hot surfaces.

4.4 Immersion of Carbon Steel in Aerated Borated Water

This is the condition in which borated water leaks from a system and then comes into contact
with carbon or low-alloy steel components. The leakage can come into contact with these parts
immediately, or it may evaporate as steam with some boric acid carryover that condenses on a
cold metal surface and then comes into contact with the carbon or low-alloy steel parts. In either
case, the fluid is exposed to oxygen that increases the potential for corrosion. Further, if the
metal surfaces are hot, the water can evaporate, thereby increasing the boric acid concentration.
Both of these effects can be highly detrimental.

Several tests have been conducted for these conditions:

1. Westinghouse Tests of Low-Alloy Steel in Aerated and Deaerated Water

(See Test Reference A)

In the mid-1960s, Westinghouse conducted a series of tests to determine the corrosion rate of
A-302 Grade B low-alloy steel in aerated and deaerated water with 2500 ppm boron at low
temperatures [22, 23]. These tests showed corrosion rates in aerated borated water of 0.002
in/yr (0.05 mm/yr) at 70°F (21ºC), 0.007 in/yr (0.18 mm/yr) at 100°F (38°C), and 0.015 in/yr
(0.38 mm/yr) at 140°F (60°C). Although these rates do not appear to be excessively high,
they represent low boric acid concentrations and low temperatures. One test at 500°F
(260°C) showed a corrosion rate of 0.24 in/yr (6.1 mm/yr).

2. Moscow Power Institute Tests in Aerated Water

(See Test Reference B)

The Moscow Power Institute conducted a series of tests to evaluate the effects of boric acid
solutions on fuel assembly materials and carbon structural steel [24, 25]. One series of tests
involved immersing plate-type carbon steel specimens in aerated and deaerated borated water
for up to 1000 hours. Results were obtained at intervals to determine the reaction kinetics.

The results showed a long-term corrosion rate of 0.007 in/yr (0.18 mm/yr) for water with
2000 ppm boron at 104°F (40ºC). This is identical to the Westinghouse rate of 0.007 in/yr
(0.18 mm/yr) at 100°F (38°C) for water with 2500 ppm boron.

3. Westinghouse Tests for Borated Water Storage Tanks

(See Test Reference E)

Westinghouse conducted a series of tests in 1958 to determine whether carbon steel material
could be used for borated water storage tanks for the Yankee Rowe plant [28]. This work
included tests of AISI 4135 low-alloy steel in saturated boric acid solutions of 5% at 70°F
(21°C) and 13% at 140°F (60°C). The specimens were immersed in beakers with glass cover
plates. Water that evaporated during the test was replaced, but no effort was made to control
ingress of oxygen.

4-6
 EPRI Licensed Material

Boric Acid Corrosion Tests

The corrosion rates at the two test conditions were 0.004–0.006 in/yr (0.10–0.15 mm/yr) for
water with 8800 ppm boron (5% boric acid) at 70°F (21°C) and 0.024–0.043 in/yr (0.61–1.1
mm/yr) for water with 22,800 ppm boron (13% boric acid) at 140°F (60°C).

A second series of tests was run in which the pH was raised to neutral by the addition of 10%
KOH. This change significantly reduced the corrosion rate.

4. Brookhaven National Laboratory Tests of Bolting Materials

(See Test Reference F)

Brookhaven National Laboratory (BNL) has performed tests on A193-B7, AISI 4130, and
4135 low-alloy steel in higher temperature water with 4000 ppm boron [8]. The reported
corrosion rates are shown in Table 4-1.
Table 4-1
BNL Tests of Bolting Materials

Temperature in °F No Lithium in Inches With LiOH (pH 7.3) in

(°C) per Year (mm/yr) Inches per Year (mm/yr)

600 (316) 0.022–0.028 ---

(0.56–0.7)

352 (178) 0.042–0.050 0.046–0.054

(1.1–1.3) (1.2–1.4)

212 (100) 0.108–0.124 0.112–0.130

(2.7–3.1) (2.8–3.3)

These data suggest that the corrosion rate decreases at temperatures above 212°F (100°C)
and that higher pH solutions have little effect at high temperatures. In combination with the
previous results, these data suggest that the maximum corrosion rate occurs near 200°F
(93°C) and that the beneficial effect of higher pH decreases above 200°F (93°C).

5. Westinghouse Tests in High-Concentration Aerated Water at 200°F (93°C)

(See Test Reference G)

Subsequent to the Turkey Point 4 incident (see Section 2.3.3), Westinghouse performed tests
involving low-alloy steels immersed in high-concentration borated water at about 200°F
(93°C) [29]. The corrosion rate for aerated water was reported to be 4.8 in/yr (122 mm/yr).

6. Combustion Engineering Beaker Tests of Piping Materials

(See Test Reference H)

Combustion Engineering conducted a series of tests to determine the corrosion rates of

typical piping and pressure vessel steels in an oxygenated environment [1, 27]. The tests
consisted of immersing plate-type specimens of SA-106 Grade B, SA-533 Grade B, and SA-
508 Class 2 in beakers with 22,000–79,400 ppm boron at 220°F (104°C). The beakers were
covered with water-cooled condensers at the top to prevent the test solutions from

4-7
 EPRI Licensed Material

Boric Acid Corrosion Tests

evaporating. Each beaker contained several specimens that were removed at intervals to
determine the effect of test duration on the corrosion rates.

These tests showed that corrosion rates for a test duration of 6 hours with 26,000 ppm boron
were in the range of 0.9–1.6 in/yr (23–41 mm/yr). A test duration of 96 hours with 22,000
ppm boron had corrosion rates in the range of 0.24–0.36 in/yr (6–9 mm/yr). In every case,
there were small differences in corrosion rates between the materials tested. Several possible
explanations were provided for the reduction in average corrosion rate with increasing time.
First, some reduction in corrosion rate is expected to occur as a corrosion film builds on the
exposed metal surface. Second, the solutions were not replenished during the test, and the pH
typically increased from 3–4 at the start of testing to 5–6 at the end of testing. It was
considered likely that replenishing the solutions, as is the case with many leaking joints,
would have increased the corrosion rates.

One test involving an SA-106 Grade B plate in 220°F (104°C) water with 72,000 ppm boron
showed a corrosion rate of 7.25 in/yr (184 mm/yr) during a four-hour test.

7. Combustion Engineering Tests of Fastener Materials in Borated Water

(Test Reference Unavailable)

Combustion Engineering conducted a series of qualitative tests to determine the corrosion

resistance of typical fastener materials to immersion in aerated borated water [30]. These
tests involved immersing specimens of bolting material into 190°F (88°C) water with 1000–
2000 ppm boron. Water with 2000–5000 ppm boron at 70°F (21°C) was dripped onto the
part of the specimen extending out of the solution.

Results of the testing were limited to reporting that the A-516 Grade 70 and AISI 4340
materials experienced “severe” corrosion, but there was no corrosion on Type 304 and 410
stainless steel materials. Because there was no quantitative data, the detailed test results have
not been reported in this guidebook.

8. EPRI Tests in Concentrated Boric Acid at 180°F (82°C)

(See Test Reference EPRI-1)

See Section 4.2 for EPRI test data on a range of carbon steel materials at 20,000 ppm boron
at 180ºF (82°C).

In summary, the above data show a very wide range of corrosion rates for immersion in aerated
borated water. These rates, which are plotted in Figure 4-3, range from a low of 0.002 in/yr (0.05
mm/yr) for 2000 ppm boron at 70°F (21°C) to a high of 7.25 in/yr (184 mm/yr) for 79,000 ppm
boron at 220°F (104°C). Although there is a wide range of corrosion rates for various test
conditions, there are really only two conditions of practical interest.

4.4.1 Immersion in Aerated Borated Water at Room Temperature

Based on the above results, if the leaking borated water is at room temperature and if it is leaking
onto room-temperature surfaces, the corrosion rates are relatively low. Reported corrosion rates

4-8
 EPRI Licensed Material

Boric Acid Corrosion Tests

for this situation range from 0.002 in/yr (0.05 mm/yr) for 2500 ppm boron at 70°F (21°C) to
0.007 in/yr (0.18 mm/yr) for 2000 ppm boron at 104°F (40°C). Boric acid spray also does not
pose a significant corrosion problem for room-temperature surfaces. Reported corrosion rates for
2000 ppm boron sprays at 100°F (38°C) fall within the range listed above for immersion.
Situations in which the boric acid totally evaporates, leaving behind boric acid crystals, are not
expected to present a problem either. Corrosion rates for the case of non-wetted boric acid
crystals were less than 0.0005 in/yr (0.013 mm/yr) for all materials evaluated.

4.4.2 Immersion in Aerated Borated Water at High Temperature

If the borated water leaks onto higher temperature surfaces, evaporation of the water will
ultimately lead to concentration of the boric acid. BNL reports that maximum corrosion rates
occur at a metal temperature around 200°F (93°C). It has also been reported that leaking coolant
can cool metal surfaces to this temperature range. Therefore, the corrosion rate is primarily a
function of the concentration that is ultimately reached near the boiling point of the boric acid
solution. Available data suggest that the corrosion rates for this condition could fall between 0.1
and 7.25 in/yr (2.5 and 184 mm/yr). Corrosion rates of 1 in/yr (25 mm/yr) and higher would be
expected only under conditions in which the boric acid concentration exceeds 25,000 ppm boron.

4.5 Dripping of Borated Water onto Hot Carbon Steel Surfaces

This condition can occur when water from a leak drips directly onto a hot metal surface or when
leakage flashes to steam with some boric acid carry over, condenses on a cold metal surface, and
the resultant boric acid laden condensate drips onto a hot metal surface creating a corrosive
environment.

This condition is unique in that it results from several different phenomena occurring
simultaneously:

1. Deaerated borated water leaks from the primary system and becomes aerated on contact with
the environment.

2. Boric acid concentrates as the water evaporates.

3. Boric acid crystals deposit at the point where all of the water has evaporated.

4. Periodic wetting of the surface at the boundary between the evaporating liquid and boric acid
crystal deposits creates a local region that is highly corrosive for carbon and low-alloy steel
parts.
Several tests have been conducted for these conditions:

1. B&W Tests of Borated Water Dripping onto Heated Surfaces

(See Test Reference I)

B&W performed the most complete testing for the case of borated water (13,000–15,000
ppm boron and 200°F [93°C]) dripping onto heated metal surfaces [31]. The first series of

4-9
 EPRI Licensed Material

Boric Acid Corrosion Tests

tests involved the borated water dripping onto the exposed surface of a carbon steel trough
that was heated to 300°F (149°C).

These tests showed that the highest corrosion rates were 3–5 in/yr (76–127 mm/yr) at the
interface between the wetted surface and locations where boric acid crystals formed when the
water evaporated. Corrosion rates were lower on the continuously wetted surfaces. For one
specimen involving carbon steel and stainless steel parts welded together, there was no
measurable corrosion of the stainless steel. There was some evidence of a galvanic effect at
the Alloy 182 weld between the carbon steel and stainless steel. The corrosion rate at this
location was about 50% greater than at an adjacent location not in contact with the weld.

2. B&W Tests of Borated Water Dripping onto Insulated Heated Surfaces

(See Test Reference J)

B&W performed a second series of tests in which borated water (13,000–14,000 ppm at
200°F [93°C]) was dripped onto the outside surface of a pipe that was covered by reflective-
type insulation [31]. The purpose of this test was to simulate the condition at ANO-1 where
borated water leaked down into the annulus between the carbon steel HPI nozzle and the
surrounding insulation. During this test, water was dripped onto the pipe at three points along
the pipe length. The measured temperatures of the pipe at each point were somewhat
different due to local differences in heat input and heat transfer.

The results of these tests confirmed the results of the plate test in that the maximum corrosion
tended to occur under the edge of the boric acid crystal deposits. It was also discovered that
the size of the wetted region at each drip point and the maximum corrosion rate decreased as
the pipe temperature increased. The likely explanation is that at high temperatures the water
boils off quickly, leaving dry boric acid deposits that do not cause corrosion. Corrosion rates
were typically in the range of 2–5 in/yr (51–127 mm/yr). However, the highest corrosion rate
occurred near the bottom of the pipe where any resultant leakage or condensation of steam
with concurrent boric acid carryover came into contact with the outside air at temperatures of
about 200°F (93°C). The corrosion rate at this point was 7.8 in/yr (198 mm/yr).

3. Westinghouse Tests of Borated Water Dripping onto Heated Surfaces

(See Test Reference G)

Subsequent to the Turkey Point 4 incident, Westinghouse performed tests involving high-
concentration borated water at about 200°F (93°C) dripped onto heated low-alloy steel
surfaces [29]. The corrosion rate for this condition was reported to be 4.8 in/yr (122 mm/yr).

4. EPRI Tests of Borated Water Dripping onto Heated Surfaces

(See Test Reference EPRI-5)

EPRI performed a series of tests in which borated water was dripped onto heated A302 steel
plates for 50 days [21]. These tests differed from the B&W and Westinghouse tests in that a
boric acid concentration representative of primary water (2000 ppm boron at 180°F [82°C])
was used instead of the concentrated solutions employed by B&W and Westinghouse. The
effects of flow rate (0.01 and 0.10 gpm [0.038 and 0.38 L/min]), metal temperature (180 and

4-10
 EPRI Licensed Material

Boric Acid Corrosion Tests

600°F [82 and 316°C]), and plate geometry (horizontal surface and 45° inclined surface)
were evaluated.

The maximum corrosion rates obtained in these tests ranged from 0.19–0.22 in/yr (4.8–5.6
mm/yr) on plates heated to 180°F (82°C) and from 1.2–1.9 in/yr (30–48 mm/yr) on plates
heated to 600°F (316°C). In all cases, minimal corrosion was observed at the drip point. In
the 600°F (316°C) tests, the areas of maximum corrosion took the form of one or more
narrow troughs that appeared to coincide with the edges of the boric acid flow lines. Trough
depths for these tests were independent of distance from the drip site on the inclined plates
but decreased moving away from the drip site. Corrosion troughs representing obvious flow
lines were not present on either of the 180°F (82°C) plates. Maximum corrosion depths were
independent of flow rate and plate geometry.

In summary, the above tests suggest that the highest corrosion rates occur at the wetted interface
between concentrated boric acid solutions and boric acid crystals. This condition can occur in
many situations in which borated water drips onto hot metal surfaces. The basic scenario
involves water dripping onto a hot metal surface where it boils, causing the concentration of acid
to increase and the local metal temperature to decrease to the boiling point of the boric acid
solution. As previously noted, the maximum corrosion rates occur at, or near, the boiling point of
concentrated boric acid solutions.

Corrosion rates when boric acid drips onto hot surfaces do not appear to be dependent on the
concentration of the dripping water because evaporation from the hot surfaces is an effective
concentrating mechanism. Furthermore, the wetted interface between concentrated boric acid
solutions and boric acid crystals at which the maximum corrosion rate occurs will always be
exposed to a saturated boric acid solution. The concentration of the dripping water does have a
significant effect on corrosion rate when the metal surface is near the temperature of the dripping
solution. Experiments performed with a 26,100 ppm boron concentration solution dripped onto a
210°F (99°C) surface exhibited corrosion rates greater than 4 in/yr (102 mm/yr), whereas those
using a 2,000 ppm boron concentration solution yielded corrosion rates of only 0.2 in/yr (5
mm/yr). Concentration by evaporation is not nearly as effective at the lower metal temperatures.

The drip test results appear to contradict the findings obtained from the aerated immersion tests
regarding the temperature at which maximum corrosion occurs. The immersion tests indicate that
maximum corrosion rates in concentrated boric acid occur at temperatures around 200°F (93°C).
Decreasing corrosion rates at higher temperatures have been rationalized based on the decreasing
oxygen solubility. The drip tests that employed a 2000 ppm boron concentration, on the other
hand, found a much higher corrosion rate for the 600°F (316°C) surface than the 180°F (82°C)
surface. Clearly, the hotter surface produces more rapid evaporation, a more concentrated boric
acid solution, and, hence, more rapid corrosion. For the case of a dripping non-concentrated
boric acid solution, the effect of metal temperature on evaporation and concentration appears to
be more important than the temperature effect on oxygen solubility.

The test data also suggest that severe corrosion does not occur when the metal temperature is
high enough and the flow rate is low enough that the borated water that drips onto the heated
surfaces boils off quickly, leaving dry boric acid crystals at the metal surface.

4-11
 EPRI Licensed Material

Boric Acid Corrosion Tests

With regard to the applicability of the laboratory tests, NUREG CR-5576 [10] reports that the
maximum corrosion depth in the ANO-1 HPI nozzle was about 0.5 inches (13 mm) and that the
leakage existed for about six months. This would give an average corrosion rate of 1 in/yr (25
mm/yr). This rate is within the band of test experience.

4.6 Impingement of Borated Water onto Hot Carbon Steel Surfaces

This condition can occur when water leaking from a gasketed joint impinges directly onto carbon
and low-alloy steel materials, including bolts that hold the joint together. The conditions that lead
to corrosion in this scenario are similar to those discussed in the previous section. In addition,
mechanical effects due to stream impinging on the bolts or other surfaces can increase the rate of
metal loss.

Several tests have been conducted for these conditions:

1. Combustion Engineering Tests of Borated Water Impinging on Bolts

(See Test Reference K)

Combustion Engineering conducted a series of tests to determine corrosion rates on heated

carbon steel and low-alloy steel studs exposed to impingement by borated steam at 0.001
gpm (0.0038 L/min) [30].

The test results indicated that average corrosion rates of carbon and low-alloy steel bolting
materials are typically 0.3–0.8 in/yr (7.6–20.3 mm/yr) and the maximum corrosion rates can
be in excess of 1.0 in/yr (25 mm/yr). The tests also showed that a number of candidate
protective coatings on the bolts were ineffective in preventing the corrosion and, in some
cases, made the corrosion worse. For example, the highest corrosion rate in any test (2.0 in/yr
[51 mm/yr]) was obtained for the case of an enamel-coated bolt. It is likely that breakdown
of the enamel coating led to high corrosion rates at the unprotected areas.

2. Combustion Engineering Tests of Impingement on Plates

(See Attachment L)

Combustion Engineering conducted a series of tests in which borated water (1000 ppm
boron) at 600°F (316°C) was forced through a capillary tube in a steel block at 0.05–0.10
gpm (0.19–0.38 L/min) and impinged upon a SA-533 Grade B steel plate [27]. The sample
block was heated to approximately 650°F (343°C) by internal electric resistance heaters.

The test results showed that the diameter of the capillary hole increased at a rate
corresponding to 0.8–1.2 in/yr (20.3–30.5 mm/yr). This resulted from a combination of boric
acid corrosion and erosion-corrosion caused by the flow of steam through the capillary. At
the exit end of the capillary block that was not exposed to the flow, the material corrosion
rate was 0.3–1.1 in/yr (7.6–28 mm/yr). These locations were not affected by erosion.

The target specimens experienced corrosion rates of 0.4–0.6 in/yr (10–15 mm/yr) at
locations where there was no impingement. At the impingement locations, the corrosion rates
varied from 0.4–3.9 in/yr (10–99 mm/yr) for targets 2 inches (51 mm) from the capillary and

4-12
 EPRI Licensed Material

Boric Acid Corrosion Tests

6.4–11.1 in/yr (163–282 mm/yr) for targets 0.5 inch (13 mm) from the capillary. Corrosion
rates were higher for higher flow rates. Corrosion rates were not significantly affected by
whether the target block was heated. The explanation offered by Combustion Engineering for
this behavior was that the impingement location is close to the 200°F (93°C) steam
temperature, regardless of whether the block is heated.

In summary, situations involving impingement of borated water onto hot metal surfaces can
result in corrosion rates of 0.3–10 inches (7.6–254 mm) per year depending upon conditions.

See Section 4.8 for a discussion of the effect of steam impingement on bolted flanged joints.

4.7 Leakage of Borated Water into a Tight Crevice (Nozzle Applications)

This condition can occur when borated water leaks from a crack into a tight crevice. Typical
examples are leakage of borated primary coolant from PWSCC cracks in Alloy 600 CRDM
nozzles, instrument nozzles, and pressurizer heater sleeves. While this condition might be
considered similar to immersion in aerated borated water or impingement of borated water onto
bolts, it is fundamentally different in that the leaking fluid keeps the crevice in a low oxygen
condition.

1. ABB-CE Tests of Leakage into Annular Gaps

(See Test Reference M)

Combustion Engineering (CE) conducted a test in which high-temperature (600°F[316°C])

borated water (1000 ppm boron) leaked through a PWSCC crack in an Alloy 600 steam
generator tube into the annular gap between the tube and a clearance hole in a SA-533 Grade
B Class 1 block [5]. The nozzle opening was directed downward.

The test results showed that the maximum corrosion rate occurs at the point where the
leaking fluid exits the annulus. It is hypothesized that this is the location where the leaking
steam first picks up significant oxygen. The average corrosion rate throughout the annulus
varied from 0.02–0.08 in/yr (0.51–2 mm/yr), and the maximum corrosion rate at the annulus
exit varied from 0.6–2.2 in/yr (15–56 mm/yr). The maximum metal loss in any test was
determined to be equivalent to 1.07 in3/yr (17.5 cm3/yr). The tests showed that small leaks
produce greater corrosion, a sign that small leak rates permit oxygen to penetrate a greater
distance into the annulus. (Note: The reported maximum volume loss of 1.07 in 3/yr (17.5
cm3/yr) may not be conservative for all cases because the volume loss is likely to increase as
the corrosion depth and wetted surface area increase.)

2. EPRI Tests of Leakage into Annular Gaps

(See Test Reference EPRI-6)

EPRI conducted a series of tests in which high-temperature (600°F [316°C]) borated water
(2000 ppm boron) was injected through an A302 Gr. B steel block into a 0.005-inch- (0.13-
mm-) wide annular gap between the block and a cylindrical 304 stainless steel insert [21].
The annulus was sealed at the bottom, and the blocks were heated to 600 °F (316°C).
Variables that were evaluated included flow rate (0.01 and 0.1 gpm [0.038 and 0.38 L/min])

4-13
 EPRI Licensed Material

Boric Acid Corrosion Tests

and specimen geometry (horizontal exit surface, angled exit surface, and angled exit surface
covered with insulation).

The A302 Gr. B steel blocks exhibited maximum corrosion rates of 0.89 to 2.37 in/yr (22.6 to
60.2 mm/yr). The highest rates were observed at the low-flow condition; however, the
maximum corrosion rates for the high- and low-flow conditions were always within a factor
of two for a given geometry. The maximum penetration was always observed deep within the
annulus close to the point at which the steam was injected into the annulus. Penetration depth
decreased as the annulus opening was approached. Because the maximum penetration took
place deep within the annulus, neither the geometry of the exit surface nor the presence of
insulation on the exit surface had a significant effect on the maximum corrosion rates.

The specimen geometries for the CE and EPRI tests are shown in Figure 4-5 and the key test
variables are given in Table 4-2.

The maximum corrosion rates from the CE and EPRI tests are shown in Figure 4-6. These data
show excellent agreement for the maximum corrosion rates in inches/year. The only outlier is for
the very low leakage rate (0.002 gpm [0.008 L/min]) in one of the CE tests. This is encouraging
because the anticipated leak rates from PWSCC of Alloy 600 nozzles are expected to be much
less than 0.002 gpm (0.008 L/min).

The only significant difference between the CE and EPRI test results is the location of maximum
corrosion. As shown in Figure 4-7a, the unwrapped inside surface contour plot of the CE test
block indicates that the maximum corrosion depth occurs at the point where the nozzle exits the
test block. Corrosion occurring at this location will be seen during a visual inspection of the
vessel surface. As shown in Figure 4-7b, the maximum corrosion depth in the EPRI test block
occurs near the point where the borated water is injected into the annulus. Corrosion occurring at
this location would not be seen during a visual inspection of the vessel surface, although boric
acid deposits on the metal surface would indicate that there has been a leak.

The most likely explanation for the difference in results is the orientation of the nozzles. In the
CE tests, the nozzles were oriented down, and in the EPRI tests, the nozzles were oriented up. It
is possible that steam blanketing in the annulus for the case of the downward-facing nozzle
explains the inability of oxygen to penetrate to the injection location. However, this hypothesis
has not been validated by test.

4.8 Leakage of Borated Water from Flanged Joints

1. EPRI Flange Leakage Tests

(See Attachment EPRI-7)

EPRI performed a series of 18 tests that simulated the conditions that occur when water
leaking from a gasketed joint impinges onto the bolts that hold the joint together [21]. In
these tests, an injector tube was placed in a heated, 10-inch (254 mm), Type 304 stainless
steel flange so that the borated water impinged directly onto one of the A193 B7 low-alloy
steel bolts. Corrosion rates for each of the eight bolts in the flange were determined at the end
of the 50-day exposure by measuring the maximum reduction in bolt diameter and using this

4-14
 EPRI Licensed Material

Boric Acid Corrosion Tests

to calculate the average corrosion rate over the test duration. 1 The effects of temperature (180
and 600°F [82 and 316°C]), flow rate (0.01 and 0.10 gpm [0.038 and 0.38 L/min]), and
flange orientation (horizontal, 45° inclined, and vertical) were evaluated in 14 of the 18 tests.
The effects of insulation were evaluated in two tests, the effects of a tight fitting band around
2
the flange were evaluated in two tests, and the effects on stainless steel bolts were evaluated
in two tests.

The maximum corrosion rates of the A193 B7 bolts exhibited a very strong temperature
dependence. Maximum corrosion rates at 180°F (82°C) were low, ranging from 0.011 to
0.037 in/yr (0.28 to 0.94 mm/yr). Maximum corrosion rates for A193 B7 bolts at 600°F
(316°C) were much higher, ranging from 0.11 to 1.7 in/yr (2.8 to 43.1 mm/yr). For cases
where there were identical tests at the two temperatures, the rates at the higher temperature
were 3–150 times greater. The rates were also in reasonably good agreement with the
immersion test results in Figure 4-3. The 304 stainless steel fasteners exhibited very low
corrosion/erosion rates, 0.007 in/yr (0.18 mm/yr) or less, during exposure at 600°F (316°C).

Flow rate did not have a significant effect on the maximum corrosion rate of the bolts.
However, in most of the 600°F (316°C) tests, a slightly higher maximum corrosion rate was
observed at the low flow condition.

At 600°F (316°C), there was a significant difference in the corrosion rate of bolts around the
flange. The maximum corrosion on the bolts tended to occur in line with the steam
impingement, although the test of the 45° angled flange at 600°F (316°C) and 0.1 gpm (0.38
L/min) flow showed that the highest corrosion occurred at the bottom of the flange, 180°
opposite to the injection point that was at the top of the flange.

For low leakage rates of 0.01 gpm (0.038 L/min), the presence of insulation or a band around
the joint had little effect on either the maximum corrosion rates or the distribution of
corrosion in bolts around the flange. However, at leak rates of 0.1 gpm (0.38 L/min),
insulation appears to hold moisture in the joint in a manner that results in higher average
corrosion rates on all of the bolts than any of the other test conditions. The average corrosion
rate for the tight fitting band was about 1/3 of that for the case of an insulated flange at 600°F
(316°C) and a 0.1 gpm (0.38 L/min) flow rate. This suggests that the band used in the tests
was effective in reducing corrosion for the case of insulated flanges. A combination of a
tight-fitting band to keep out oxygen may be an effective preventive measure for use with
insulated high-temperature flanges.

There appears to be little short-term concern with BAC for flanges that operate below the boiling
point of borated water.

1
The average corrosion rate at the point of maximum diameter reduction is
(Dstart – Dend)*(365 days/50 days) / 2.
2
It was hypothesized that a tight-fitting band installed around the flange would allow the leaking steam to escape
but prevent entry of oxygen into the flange joint, thereby preventing corrosion.

4-15
 EPRI Licensed Material

Boric Acid Corrosion Tests

4.9 Summary of Corrosion Rates for Various Situations

Figure 4-3 summarizes the results for all of the reported corrosion tests and points out the main
areas of interest. Briefly summarizing the key points:
• Corrosion rates for immersion in deaerated, dilute boric acid solutions are quite low
regardless of temperature. Moderate corrosion rates, between 0.02 and 0.05 in/yr (0.5 and 1.3
mm/yr), can be obtained during immersion in deaerated, concentrated boric acid solutions.
• For cases involving immersion in aerated borated water, corrosion rates are in the range of
0.001–0.01 in/yr (0.025–0.25 mm/yr) for low concentrations at room temperature and
increase to a maximum of 1–10 in/yr (25–254 mm/yr) for high concentrations at 200–220°F
(93–104°C).
• The main problem regarding borated water dripping on hot metal surfaces is that the solution
can concentrate as the water boils off and the boiling can lower the local metal temperature to
the boiling point (212–230°F [100–110 °C]) of the concentrated boric acid solution.
Therefore, the typical situation is to have concentrated boric acid at around 212–220°F (100–
104°C), which is the point of the maximum corrosion rate. However, if the metal surfaces are
hot and the leakage rate very low, the solution can boil off quickly, leaving dry boric acid
crystals that cause essentially no corrosion. Lower corrosion rates are expected when the
surfaces onto which the borated water is dripping are below the boiling point of the borated
water.
• If borated steam impinges on hot metal surfaces, the corrosion rates can be very high as a
result of the combination of high concentration, local metal temperatures near the boiling
point of the borated water, and some mechanical effects due to the flow impingement. This
condition can be highly damaging as evidenced by several cases involving rapid stud
corrosion.
• Laboratory results suggest that borated water leaking from a PWSCC-type crack should not
cause corrosion deep in the annular clearance gap to the vessel shell since there is little
oxygen at this location. However, tests in upward-pointing nozzles suggest that corrosion
rates exceeding 1 in/yr (25 mm/yr) are possible.

Application of the above data to typical components is discussed in Section 8.

4.10 Boric Acid Corrosion Test Data Summaries

The following summarizes the major BAC test work that has been reported publicly. It is
recommended that users of this guidebook obtain complete reports of these tests if the test will
provide the technical basis for operating a plant with continuing leakage. The test reports should
then be reviewed to confirm that the test conditions cover the specific plant conditions being
evaluated.

4-16
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-2
Key Parameters of CE and EPRI Nozzle Leakage Tests

Parameter CE Tests EPRI Tests

Nozzle Diameter in Inches 0.75 (19) 1.21 (30.7)

(mm)

Radial Clearance (gap) 0.0013–0.005 0.005 (0.13)

Between Nozzle and Block in (0.03–0.13)
Inches (mm)

Nozzle Orientation Down Up

Oxygen Concentration in ≈0 0–80*

Simulated Primary Coolant
Upstream of Leak (ppb)

Boron Concentration (ppm) 600–1000 2000

Temperature in °F (°C) 650 (343) 600 (316)

Leak Rate in gpm (L/min) 0.002–0.05 0.01–0.10

(0.008–0.189) (0.038–0.38)

Duration (days) 12–27 50

* The oxygen concentration in primary coolant in PWR plants during operation is undetectable (about 0
ppb). Thus, the oxygen concentration in the EPRI tests may have been nonrepresentative of (that is, much
more aggressive than) plant conditions.

4-17
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-1
Boric Acid Corrosion Conditions

4-18
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-2
Results of EPRI Material Screening Tests

4-19
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-3
Summary of Corrosion Test Data

4-20
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-4
Matrix of EPRI Flange Joint Tests

4-21
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-5
Combustion Engineering and EPRI Nozzle Test Fixtures

4-22
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-6
Summary of Maximum Corrosion Rates in CE and EPRI Tests

4-23
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-7
Corrosion Depth in Sectioned EPRI Test Block

4-24
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: A

Test Type: Immersion – Aerated and Deaerated Boric Acid

Org/Date: Westinghouse (1967)

References: “Materials Behavior Related Issues for Bolting Applications in the Nuclear
Industry.” Conference: Improved Technology for Critical Bolting
Applications, Chicago, Ill., July 20–24, 1986. MPC-Vol. 26 [22].
Absorption of Corrosion Hydrogen by 302B Steel at 70F to 500F.
Westinghouse Electric Corporation, 1967, WCAP-7099 [23].

Test Configuration, Procedure, and Results

Westinghouse conducted a series of tests on electrically isolated coupons of A-302 Grade B low-
alloy steel in both aerated and deaerated environments. Some experiments involved cases where
the low-alloy steel test coupons were coupled to 304 stainless steel. Tests were conducted at
2500 ppm boron at several different temperatures.

Table 4-3 gives the average corrosion rate for long periods of time (40–160 days) for each of the
test conditions. The test results plotted in Figure 4-8 show the effect of test duration.
Table 4-3
Average Corrosion Rates in Aerated and Deaerated Water – Westinghouse Tests

Temperature in Corrosion Rate Deaerated in Corrosion Rate Aerated in

°F (°C) Inches/Year (mm/yr) Inches/Year (mm/yr)

Isolated Coupled to 304 Isolated Coupled to 304

70 (21) 0.0002 (0.005) --- 0.002 (0.05) 0.002 (0.05)

100 (38) 0.00005 (0.0013) 0.00005 (0.0013) 0.007 (0.18) 0.007 (0.18)

140 (60) 0.00007 (0.0018) 0.00011 (0.0028) 0.015 (0.38) 0.015 (0.38)

500* (260) --- --- 0.24 (6.1) ---

*Provided at meeting with Westinghouse.

Key observations from these tests are:

• In deaerated water at temperatures less than 140°F (60°C), the corrosion rates are very low.
• In aerated water, the corrosion rates increase with temperature from 70°F to 140°F (21°C to
60°C) with the long-term corrosion rate at 140°F (60°C) being about 0.015 in/yr (0.38
mm/yr).
• The galvanic couple between low-alloy steel and stainless steel has little effect on the
corrosion rates of low-alloy steel in borated water.

4-25
 EPRI Licensed Material

Boric Acid Corrosion Tests

• Initial corrosion rates can be much higher than long-term corrosion rates.
• The corrosion rate in aerated borated water at 500°F (260°C) is 0.24 in/yr (6.1 mm/yr).

Conclusions

Corrosion rates in low-temperature deaerated water are very low. Corrosion rates in aerated
water at temperatures less than 140°F (60°C) are acceptably low to not pose a serious problem
with leakage of borated water.

Figure 4-8
Corrosion Rate in Aerated and Deaerated Water – Westinghouse Tests [22]

4-26
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: B

Test Type: Immersion – Aerated and Deaerated Boric Acid

Org/Date: Moscow Power Institute (1970)

References: “Corrosion of Structural Materials in the Boron-Containing Solutions

Used for Controlling the Power of Nuclear Reactors,” 1970 [24].
“Influence of Boric Regulation of a Medium on Corrosion and Low-Cycle
Corrosion Fatigue Resistance of Constructional Steels,” 1982 [25].

Test Configuration, Procedure, and Results

These tests were conducted to evaluate the effect of boric acid solutions on fuel assembly
materials and structural carbon steels.

The tests consisted of immersing plate-type specimens of carbon steel with 12.4 cm 2 (1.9 in2) of
exposed surface area into several selected environments and measuring the weight loss that
occurs. The average corrosion rate can then be determined from the average weight loss.

The test conditions and measured weight loss are as shown in Table 4-4.
Table 4-4
Conditions and Measured Weight Loss – Moscow Power Institute Tests

Concentration, ppm Oxygen Test Temperature Corrosion Rate

Boron Boric Acid °C °F gm/(m2hr) (inch/year)

3,000 17,100 Deaerated 310 590 0.004 0.0002

1,000 5,700 Deaerated 200 392 0.007 0.0003

2,000 11,400 Oxygenated 40 104 0.152 0.0067

Key observations from these tests are:

• Corrosion rates at low temperature and low concentrations can be quite low.
• Corrosion rates at high temperature and low concentrations are very low in the absence of
oxygen.
• Corrosion rates vary as a function of test duration. This is illustrated in Figures 4-9 and 4-10
that show high average corrosion rates for short test durations and lower average corrosion
rates for longer test durations.

4-27
 EPRI Licensed Material

Boric Acid Corrosion Tests

Conclusions

The main conclusion from these tests is that corrosion rates can be very low in the absence of
oxygen, even up to temperatures of 590°F (310°C).

Figure 4-9
Corrosion Rates Under Low Temperatures – Moscow Power Institute Tests [24]

Figure 4-10
Corrosion Rates Under High Temperatures – Moscow Power Institute Tests [24]

4-28
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: C

Test Type: Immersion – Partially Deaerated Boric Acid

Org/Date: Singleton Materials Engineering Laboratories (TVA) (1983)

Reference: Degradation of B7 Threaded Fasteners in Borated-Water Solutions at

Elevated Temperatures - Interim Report, July 1983 [26].

Test Configuration, Procedure, and Results

These tests were performed by TVA to provide background information on corrosion rates of B7
fasteners in borated water in response to NRC IE Bulletin 82-02.

The tests consisted of preloading 1/2-inch- (13-mm-) diameter by 3-inches- (76-mm-) long B7
fasteners and immersing them in borated water solutions in an autoclave at 550°F (288°C) for 30
days. Tests were conducted in boric acid (pH = 4.0), lithium hydroxide buffered boric acid (pH =
10.5), and wetted solid boric acid with 200 ppm Cl. The configuration was such that two
fasteners in each autoclave were in the liquid phase and two fasteners in each autoclave were in
the vapor phase (see Figure 4-11). The test was conducted under very low oxygen conditions
because no oxygen was added to the autoclave during the testing. The results are shown in
Tables 4-5, 4-6, and 4-7.

Fasteners were tested unlubricated; lubricated with N-5000, C5-A, and Furmanite F-500; and
coated with Pepkote.

4-29
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-5
Corrosion Rates in Boric Acid Solution – TVA Tests

Initial Conditions Max Wt. Loss (gm)

Lubricant Stress in pH Cl S Bolts in Bolts in

ksi (MPa) (ppm) (ppm) Solution Air

None 45 (310) 3.8 <10 4.0 0.1 0.1

C5-A 45 (310) 3.8 <10 4.0 0.1 0.1

Furmanite F-500 45 (310) 3.8 <10 4.0 0.0 0.1

N-5000 45 (310) 3.8 <10 4.0 0.0 0.1

N-5000 80 (552) 4.2 102 0 0.1 0.1

N-5000 80 (552) 4.2 207 0 0.2 0.2

N-5000 80 (552) 4.2 250 1130 0.1 0.1

Pepkote 80 (552) 4.2 102 0 0.1 0.1

Pepkote 80 (552) 4.2 207 0 0.1 0.2

Pepkote 80 (552) 4.2 250 1130 --- 0.1

Table 4-6
Corrosion Rates with Wetted Solid Boric Acid – TVA Tests

Initial Conditions Max Wt. Loss (gm)

Lubricant Stress in Chemistry Conditions Bolts in Bolts in

ksi (MPa) Solution Air

N-5000 80 (552) Solid Boric Acid + 200 ppm Cl 2.4 0.1

Pepkote 80 (552) Solid Boric Acid + 200 ppm Cl --- 0.2

4-30
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-7
Corrosion Rates in Lithiated Boric Acid Solution – TVA Tests

Initial Conditions Max Wt. Loss (gm)

Lubricant Stress in pH Cl S Bolts in Bolts in

ksi (MPa) (ppm) (ppm) Solution Air

None 45 (310) 10.5 <10 4.0 0.1 0.1

C5-A 45 (310) 10.5 <10 4.0 0.0 0.1

Furmanite F-500 45 (310) 10.5 <10 4.0 0.1 0.0

N-5000 45 (310) 10.5 <10 4.0 0.0 0.0

Note: Bolts were originally weighed to an accuracy of 0.1 gram; therefore, changes of 0.1 gram are within the original sensitivity.

Key observations from these tests are:

• The only significant corrosion was the loss of 2.4 grams of weight for the bolts in contact
with wetted boric acid crystals. The loss of 2.4 grams corresponds to about a 4% weight loss
on the bolts.
• There was essentially no corrosion for the remaining test conditions. A weight loss of 0.2
grams corresponds to a uniform corrosion rate of 0.006 in/yr (0.15 mm/yr). This confirms
that corrosion in boric acid solutions with low oxygen is very low.

Conclusions

The main conclusion from these tests is that corrosion rates are very low in the absence of
oxygen, even at temperatures up to 550°F (288°C).

4-31
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-11
B7 Fasteners in Borated Water Solution – TVA Tests

4-32
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: D

Test Type: Immersion – Partially Deaerated Boric Acid

Org/Date: Combustion Engineering (1988)

Reference: “Boric Acid Corrosion of Carbon and Low-Alloy Steels.” Proceedings of the
4th International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems - Water Reactors, 1989 [27].

Test Configuration, Procedure, and Results

These tests were performed by Combustion Engineering to determine the corrosion rates for
unclad pressure vessel steels during typical plant start-up conditions.

The test consisted of placing test plates (4.5 x 0.5 x 0.125 inches [114 x 13 x 3 mm]) of SA-533
Grade B, SA-508 Class 2, and SA-508 Class 2 manually welded to SA-533 Grade B into a
stainless steel autoclave containing deionized water with 723 ppm boron as H 3BO3 and heating
the autoclave to 350°F (177°C) for 70 hours. The water contained oxygen at the start of the test
as a result of contact with the environment, and no attempt was made to remove oxygen during
the test. However, no oxygen was added either.

The measured average corrosion rates at the completion of the testing were as shown in
Table 4-8.
Table 4-8
Corrosion Rates in Partially Deaerated Boric Acid – Combustion Engineering Tests

Material Number of Specimens Average Corrosion Rate in

Inches/Year (mm/yr)

SA-533 Grade B 3 0.017 (0.43)

SA-508 Class 2 3 0.017 (0.43)

SA-533/SA-508 Weld 2 0.016 (0.41)

Conclusions

The main conclusion from this test is that corrosion rates are relatively low in environments
where there is no addition of oxygen.

No figures are available for these tests.

4-33
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: E

Test Type: Immersion – Aerated Boric Acid

Org/Date: Westinghouse - Atomic Power Division (1958)

Reference: Studies of the Corrosion of AISI 304 Stainless Steel and AISI 4135 Carbon
Steel Exposed to Saturated Solutions of Boric Acid. Report YAEC-67,
November 1958 [28].

Test Configuration, Procedure, and Results

Tests were performed by Westinghouse Atomic Power Division to determine whether boric acid
storage tanks for the Yankee Rowe plant should be constructed from carbon steel or stainless
steel material.

The tests consisted of immersing carbon steel (AISI 4135) and stainless steel (AISI 304) coupons
(1 x 2 x 0.1 inches [25 x 51 x 2.5 mm] nominal size) in saturated solutions of boric acid at 70°F
(21°C) and 140°F (60°C) for periods of two or four weeks. The specimens were immersed to the
point where about half of the specimen was in the solution and half in air.

The saturation concentrations at 70 and 140°F (21 and 60°C) were 5% and 13% respectively.
The weight loss that occurred during the tests was recorded, and the corrosion rate at the surface
of the coupon was calculated from the weight loss.

Tests were also run for the case where the pH was raised to neutral by the addition of 10% KOH.

The materials tested and the resultant corrosion rates in inches/year (mm/yr) are shown in
Table 4-9.

4-34
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-9
Corrosion Rates in Aerated Boric Acid – Westinghouse Tests

Parameters 70°F(21°C) 140°F (60°C)

Stainless Steel (Acidic Test)

As Rolled Nil Nil
Sensitized Nil Nil
Heliarc Weld Nil Nil
Consumable Electrode Weld Nil - 4
7 x 10 (0.018)

Carbon Steel (Acidic Test)

As Rolled 0.004 (0.10) 0.024 (0.61)
Heliarc Weld 0.005 (0.13) 0.042 (1.06)
Consumable Electrode Weld 0.006 (0.15) 0.043 (1.09)

Carbon Steel (Neutral Test)

-5
As Rolled 3.2 x 10 (0.0008) See Note
-6
Heliarc Weld 1.2 x 10 (0.00003)
-6
Consumable Electrode Weld 5.7 x 10 (0.00014)

Note: Less than for the acidic test but increasing exponentially with time

Key observations from these tests are:

• There was essentially no corrosion of the stainless steel material in any of the tests.
• Corrosion of the carbon steel material increased with increasing temperature and
concentration. However, the maximum corrosion rate for a saturated solution at 140°F
(60°C) was only 0.025–0.045 in/yr (0.64–1.1 mm/yr). This is more than two orders of
magnitude less than the rates of 2–5 in/yr (51–127 mm/yr) found during the B&W drip tests
onto heated surfaces.
• Corrosion was first detected as scale forming at the liquid-air interface.
• For the acidic condition tests, the corrosion rate in the weld metal was about twice that of the
base material.

Conclusions

The main conclusion from these tests is that corrosion rates are relatively low in saturated
solutions at temperatures up to 140°F (60°C).

No figures are available for these tests.

4-35
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: F

Test Type: Immersion – Aerated Boric Acid

Org/Date: Brookhaven National Laboratory (1982)

Reference: Boric Acid Corrosion of Ferritic Reactor Components.

Brookhaven National Laboratory, July 1982, NUREG/CR-2827 [8].

Test Configuration, Procedure, and Results

The testing performed by Brookhaven National Laboratory (BNL) was not described in any of
the references. However, it can be inferred that A-193 Grade B7, AISI 4130, and 4135 low-alloy
steel specimens were immersed in an aerated environment containing borated water at various
temperatures. The average corrosion rates for these materials were then determined for test
durations of 350–1300 hours.

Tests were conducted in solutions with 4000 ppm boron and 4000 ppm boron plus LiOH to bring
the solution to pH 7.3, which is close to typical PWR operating conditions.

The test cases and key test results are shown in Table 4-10.
Table 4-10
Corrosion Rates in Aerated Boric Acid – BNL Tests

Average Corrosion Rate in Inches/Year (mm/yr)

4000 ppm Boron +

Temperature in °F (°C) 4000 ppm Boron
LiOH to 7.3 pH

212 (100) 0.108–0.124 (2.7–3.2) 0.112–0.130 (2.8–3.3)

352 (178) 0.042–0.050 (1.06–1.27) 0.046–0.054 (1.17–1.4)
600 (316) 0.022–0.028 (0.56–0.71) No data

Key observations from these tests are:

• The corrosion rate decreases for temperatures above 212°F (100°C).
• Lithiated primary water has essentially the same high-temperature corrosion rate as non-
lithiated water. This result is not consistent with testing conducted by others at lower
temperatures. BNL hypothesizes that the increased ionic species present at higher
temperatures increase reaction kinetics over that at lower temperatures.

Conclusions

The main conclusion from these tests is that corrosion rates in aerated water near typical
operating boric acid concentrations can be as high as 0.13 in/yr (3.3 mm/yr). Work by others has
shown that this corrosion rate can increase significantly at higher concentrations.

4-36
 EPRI Licensed Material

Boric Acid Corrosion Tests

Note: Figure 4-12 was prepared by BNL to show their test results in combination with results of
the earlier tests performed by Westinghouse for Yankee Rowe. It should be noted that the
Westinghouse test results were at temperatures of 70°F (21°C) and 140°F (60°C) with much
higher boric acid concentrations.

Figure 4-12
Corrosion Rates of Low-Alloy Steels in Aerated Boric Acid – BNL Tests

4-37
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: G

Test Type: Immersion and Dripping – Aerated and Deaerated Boric Acid

Org/Date: Westinghouse (1987)

Reference: Degradation of Reactor Coolant System Pressure Boundary Resulting

from Boric Acid Corrosion. Westinghouse letter NS-NRC-87-3260 to the
NRC, dated October 15, 1987 [29].

Test Configuration, Procedure, and Results

The conditions for the test work that is described are not specified in the referenced letter. It is
assumed that the test consisted of immersing specimens of low-alloy steel into concentrated,
high-temperature boric acid solutions and measuring the average corrosion rate.

Table 4-11 gives the complete extent of available test conditions and results.
Table 4-11
Corrosion Rate in Aerated and Deaerated Boric Acid – Westinghouse Tests

Average
Temperature in Wastage Rate in
Test Solution Test Condition
°F (°C) Inches/Year
(mm/yr)

25% Aqueous – Aerated 200 (93) 4.8 (122)

25% Aqueous – Deaerated 200 (93) 3.0* (76)

15% Drip Test in Air 210 (99) 4.2–4.8 (107–122)

15–25% Boric acid dripping onto Inconel 210 (99) 4.8 (122)
CRDM weld mockup with typical
crevice geometry

10% boric acid Aqueous – Aerated 194 (90) 4.8 (122)

+ 1% NaCl

* This test was for one day of immersion in a solution that initially contained oxygen, but no oxygen was added during the test.
Therefore, the condition is considered to represent an aerated condition. At longer test durations, the oxygen would be
consumed, and the corrosion rate would be expected to drop substantially.

Key observations from this data are as follows:

• Concentrated boric acid solutions around 200°F (93°C) with adequate oxygen can produce
corrosion rates approaching 5 in/yr (127 mm/yr).
• This high corrosion rate appears to be independent of whether the part is immersed under
these conditions or the solution is dripped onto the surface.

4-38
 EPRI Licensed Material

Boric Acid Corrosion Tests

• A galvanic couple between carbon steel and Inconel 600 produced little increase in the local
corrosion rate of the carbon steel.
• There was no corrosion of carbon steel specimens in dry boric acid crystals at 500°F
(260°C).

Conclusions

Concentrated boric acid solutions around 200°F (93°C) are highly corrosive. Unfortunately, this
condition can occur as a result of leakage that drips onto a hot metal surface, thereby cooling the
surface to near the boiling point of the boric acid solution and concentrating the solution in the
process.

No figures are available for these tests.

4-39
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: H

Test Type: Immersion – Aerated Boric Acid

Org/Date: Combustion Engineering (1988)

References: Boric Acid Corrosion of Carbon and Low-Alloy Steel Pressure-Boundary

Components, August 1988, EPRI NP-5985 [1].
“Boric Acid Corrosion of Carbon and Low-Alloy Steels.” Proceedings of the
Fourth International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems–Water Reactors, Jekyll Island,
Georgia, August 6–10, 1989 [27].

Test Configuration, Procedure, and Results

This test consisted of immersing plate-type specimens of SA-106 Grade B, SA-533 Grade B, and
SA-508 Class 2 steel in flasks with 22,000–79,400 ppm boron at 220°F (104°C). The flasks were
covered with water-cooled condensers at the top to prevent the test solutions from evaporating.
Each flask contained several specimens that were removed at intervals to determine the effect of
test duration on the corrosion rates.

Table 4-12 provides a summary of the key results of this test at 220°F (104°C). Figure 4-13
shows carbon steel corrosion at 590°F (310°C).
Table 4-12
Corrosion Rate in Aerated Boric Acid – Combustion Engineering Tests

Average Corrosion Rate in Inches/Year (mm/yr)

Boron Starting Duration
(ppm) pH (hr) SA-106 Gr B SA-533 Gr B SA-508 Cl 2

79,000 3–4 4 7.25 (184) --- ---

26,200 4 6 1.63 (41) 1.42 (36.1) 0.91 (23.1)

27 0.40 (10.2) 0.30 (7.6) 0.21 (5.3)

22,000 3 24 0.75 (19.1) 0.65 (16.5) 0.52 (13.2)

96 0.24 (6.1) 0.36 (9.1) 0.24 (6.1)

Key observations from this test data are as follows:

• One test involving an SA-106 Grade B plate in 220°F (104°C) water with 72,000 ppm boron
showed a corrosion rate of 7.25 in/yr (184 mm/yr) during a 4-hour test.
• Corrosion rates for test durations of 6 hours with 26,200 boron were in the range of 0.9–1.6
in/yr (23–41 mm/yr), and corrosion rates for 96 hours with 22,000 ppm boron were in the
range of 0.24–0.36 in/yr (6.1–9.1 mm/yr).

4-40
 EPRI Licensed Material

Boric Acid Corrosion Tests

• In every case, differences in corrosion rates between the materials tested were small.
• There are several possible explanations for the reduction in average corrosion rate with
increasing test duration. First, some reduction in corrosion rate is expected to occur as a
corrosion film builds on the exposed metal surface. Second, the solutions were not
replenished during the test, and the pH typically increased from 3–4 at the start of testing to
5–6 at the end of testing. It was considered likely that replenishing the solutions, as is the
case with many leaking joints, would have increased the corrosion rates. The effect of pH on
boric acid corrosion is suggested by the data plotted below from the referenced report.

Conclusions

Concentrated, aerated boric acid solutions at temperatures around 200°F (93°C) are highly
corrosive to low-alloy steel materials.

Figure 4-13
Effect of pH on Corrosion Rate of Carbon Steel at 310°C – Combustion Engineering Tests
[1]

4-41
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: I

Test Type: Dripping – Onto Open Heated Trough

Org/Date: B&W Alliance Research Center (1990)

Reference: B&W Boric Acid Corrosion Research and the Wastage and Inspection
Procedures for RCS Leakage. 1990, BAW-2126. [31].

Test Configuration, Procedure, and Results

The test apparatus consisted basically of a 36-inch- (0.9-m-) long steel trough that was supported
at a 30° angle from the horizontal. The trough was heated from the underside to temperatures
typical of the exposed metal surfaces of primary loop components. Simulated primary water
(borated and lithiated) at 200°F (93°C) was then dripped onto the exposed surface of the trough
near the top and was allowed to flow down through the trough. The water evaporated as it was
heated by the metal surface that resulted in increasing boric acid concentration along the trough
length. At the completion of the scheduled test, the trough was cut into sections to determine the
corrosion depths at selected locations. See Figure 4-14 for details of the test apparatus.

For Tests #1 and #6, the trough was fabricated by cutting pieces of 4-inch (102-mm) Schedule 80
SA-106 Grade B carbon steel pipe in half lengthwise. For Test #5, the trough was cut from a
piece of pipe fabricated with 304L stainless steel on the top half, carbon steel on the lower half
and an Inco 82 weld between the two steel sections.

In one test (Test #6), the trough was covered to maintain a high-humidity condition such as could
occur between a pipe and its surrounding metallic insulation.

Three tests were run for the open trough configuration. In each case, the lithium concentration
was 1.7 ppm, the water temperature was 200°F (93°C), the metal temperature was 300°F
(149°C), and the flow rate was 0.03 gpm (0.11 L/min). Other test parameters and key test results
are as shown in Table 4-13.

4-42
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-13
Corrosion Rates with Dripping Boric Acid – B&W Trough Tests

B&W Test No. #1 #6 #5

Key Test Parameters

Trough Material Carbon steel Carbon steel Carbon/stainless
Trough Cover no yes no
Boric Acid, ppm 15,000 13,800 13,500
Test Duration, hours 200 210 232

Corrosion Rates, in/yr (mm/yr)

Top of Trough 4.9 (125) 3.1 (79) 0.0 stainless steel
At Weld --- --- 5.1 (130)
Mid Length 4.4 (112) 4.4–4.7 (112–119) 3.2 near weld (81)
Bottom of Trough 2.8 (71) 2.6 (66) 0.6 (15)

Key observations from these tests were as follows:

• Clouds of steam rose above the open trough as the solution boiled and became more
concentrated. A small amount of liquid flowed out the lower end of the trough, and the
trough became heavily caked with rusty-white boric acid crystals.
• Boric acid crystals tended to deposit along the edges of the flowing borated water as a result
of the water boiling off. Corrosion rates were higher at the crystal/water interface than under
the surface of flowing water. This is a location of very high local boric acid concentration
and high humidity.
• Covering the trough to keep the humidity high does not appear to significantly affect the
corrosion rates. This is probably a result of the fact that the maximum corrosion rates
occurred under the edges of the boric acid crystal deposits that remain at high humidity
throughout the test.
• There was no measurable corrosion of stainless steel piping surfaces or Inconel weld metal
joining the stainless steel and carbon steel piping sections for Test #5.
• There appears to be about a 50% increase in corrosion rate in the carbon steel at the point of
contact with the stainless steel.

Conclusions

The main conclusion from these tests is that corrosion rates can be high (3–5 in/yr [76–127
mm/yr]) at and downstream of locations where borated water drips directly onto hot carbon steel
surfaces.

4-43
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-14
Details of Borated Water Dripping onto an Open Heated Trough – B&W Trough Tests

4-44
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: J

Test Type: Dripping – Onto Insulated Pipe

Org/Date: B&W Alliance Research Center (1990)

Reference: B&W Boric Acid Corrosion Research and the Wastage and Inspection
Procedures for RCS Leakage. 1990, BAW-2126. [31].

Test Configuration, Procedure, and Results

The test apparatus consisted basically of a 3-foot- (0.9-m-) long steel pipe that was supported at a
30° angle from the horizontal. The pipe was heated from the inside to temperatures typical of the
exposed metal surfaces of primary loop components, and 2 feet (0.6 m) of the pipe length was
covered by mirror-type insulation separated from the pipe surface by a 3/4-inch (19-mm)
annulus. Simulated primary water (borated and lithiated) at 200°F (93°C) was then dripped onto
the exposed surface of the pipe at three locations (0.01 gpm [0.038 L/min] per location). The
water evaporated as it was heated by the metal surface that resulted in increasing boric acid
concentration. At the completion of the tests, the pipes were cut into sections to determine the
corrosion depths at selected locations. See Figure 4-15 for details.

For Test #7, the pipe was fabricated with 304L stainless steel on one side, carbon steel on the
other side, and an Inco 82 weld between the two steel sections. The water was dripped onto the
weld, and the pipe was oriented so that most of the flow was directed toward the carbon steel
side of the weld joint.

Four tests were run for the open trough configuration. In each case, the water temperature was
200°F (93°C), and the total flow rate was 0.03 gpm (0.11 L/min). Other test parameters and key
test results are as shown in Table 4-14.

4-45
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-14
Corrosion Rates with Dripping Boric Acid – B&W Pipe Tests

B&W Test No. #4 #2 #3 #7

Key Test Parameters

Pipe Material Carbon Stl. Carbon Stl. Carbon Stl. Carbon/SS
Boric Acid, ppm 13,700 13,000 13,800 14,300
Lithium, ppm 1.6 1.5 1.3 1.7
Test Duration, hours 198 263 216 201

Metal Temperatures, °F (°C)

Top Injection Point 325 (163) 300 (149) 330–430 (166–221) 450 (232)
Mid Injection Point 300 (149) 500 (260) 556 (291) 504 (262)
Bottom Injection Point 280 (138) 530 (277) 554–593 (290–312) 450 (232)
Bottom (Underside) of Pipe * * * *

Corrosion Rates, in/yr (mm/yr)

Top Injection Point 4.0 (102) 4.6 (117) 1.9 (48) 3.2 (81)
Mid Injection Point 3.5 (89) 1.3 (33) 1.5 (38) 4.1–5.4 (104–137)
Bottom Injection Point 1.9 (48) 0.3 (7.6) 0.0 1.7 (43)
Bottom (Underside) of Pipe --- 7.8 (198) 4.3 (109) ---

* The temperature at this location was not measured.

Key observations from these tests were as follows:

• The greatest corrosion occurred at the outer edges of tear-drop-shaped regions at the borated
water injection points. This corresponds to the area under the edge of the boric acid crystal
deposit as was observed in the B&W trough tests.
• As the pipe temperature increased, the size of the tear-drop corrosion region at the injection
points tended to decrease, and at the highest temperatures, there was essentially no corrosion.
The likely explanation for this behavior is that the water boils off quickly at high
temperatures and low flow rates, leaving dry boric acid crystal deposits that do not tend to
produce corrosion.
• There was no corrosion of the stainless steel pipe material, and there was no evidence of
particularly aggressive attack in the carbon steel material at the Inco 82 weld.
• Very high corrosion rates (4–8 in/yr [102–203 mm/yr]) were observed at the bottom edge of
the pipe for cases where this surface was wetted. The temperatures at this location were not
measured, but they are believed to be significantly lower than in the middle of the pipe
because the active portion of the heater terminated several inches inside the pipe. This
surface may have been wetted by condensation.

Conclusions

The main conclusion from these tests is that corrosion rates can be high (3–5 in/yr [76–127
mm/yr]) at and downstream of locations where cold borated water drips directly onto hot carbon

4-46
 EPRI Licensed Material

Boric Acid Corrosion Tests

steel surfaces. The corrosion rates can be even higher (7–8 in/yr [178–203 mm/yr]) where boric
acid condenses on metal surfaces near 200°F (93°C).

Figure 4-15
Details of Borated Water Dripping onto an Insulated Pipe – B&W Pipe Tests

4-47
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: K

Test Type: Impingement – Onto Bolts

Org/Date: Combustion Engineering (1987)

Reference: “Corrosion of Low-Alloy Steel Fastener Materials Exposed to

Borated Water,” Environmental Degradation in Nuclear Power
Systems - Water Reactors, 1987 [30].

Test Configuration, Procedure, and Results

These tests were conducted to determine the effect of boric acid solutions being sprayed on
heated alloy steel studs.

The test configuration is shown in Figure 4-16. The test apparatus consists of a carbon steel or
stainless steel support frame, steel studs preloaded to 50–67% of yield strength, and a capillary
tube with an inside diameter of 0.003 inch (0.08 mm) that directs borated steam at about 575°F
(302°C) onto the stud. The ends of the capillary tubes were positioned about 1 inch (25 mm)
from the surface of the stud. The studs were bored out on the inside to accommodate an electrical
resistance heater and a thermocouple. The temperatures reported for the tests are at the
thermocouples inside the stud. The actual surface temperatures are less than this internal metal
temperature as indicated by the less than (<) symbol on the temperatures in Table 4-15.

The duration of each test was about 500 hours. At the end of the testing, the studs were removed
and cleaned and the maximum reduction in diameter measured.

The test conditions and measured corrosion rates are as shown in Table 4-15.

4-48
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-15
Corrosion Rate for Impingement onto Bolts – Combustion Engineering Tests

Fixture Corrosion Rate in Inches/Year (mm/yr)

Test Material Material at Test Temperature °F (°C)

<175 (79) <300 (149) <350 (177) <400 (204)

SA-193 B7 Carbon Steel

- Bare Metal SS 0.639* (16.2) --- 0.190* (4.8) ---
0.833** (21.2) 0.431** (10.9)
- Cr-ion Implant Low-Alloy 1.542 (39.2) --- --- ---
- TiN Plasma Coat Low-Alloy 0.753 (19.1) --- --- ---
- Enamel Low-Alloy 2.032 (51.6) --- --- ---
- Solid Film Lubricant Low-Alloy 0.925* (23.5) --- --- ---
1.650** (41.9)

Low-Alloy Steel
- SA-540 B23 Class 4 SS 0.731* (18.6) --- 0.650* (16.5) ---
0.815** (20.7) 0.841** (21.4)
- SA-540 B24 Class 3 SS 0.265 (6.7) 0.722 0.495* (12.6) 1.690 (42.9)
1.290** (32.8)
- SA-540 B24 Class 3 Low-Alloy 0.393* (10.0) --- 0.432* (11.0) ---
0.660** (16.8) 0.664** (16.9)

* The average corrosion rate for several tests.

** The maximum corrosion rate for any single test.

In addition to these carbon and low-alloy steel fasteners, tests were run on 17-4 PH stainless steel
and Alloy 718 fasteners. There was no measurable corrosion on either of these materials, and the
original machined surface was still clearly visible after completion of the testing. In summary,
these materials proved to be highly resistant to corrosion under boric acid impingement
conditions.

Key observations from the tests on carbon and low-alloy steel materials are as follows:
• Corrosion rates for boric acid impingement can be quite high.
• There was not a strong correlation between the stud temperature and the corrosion rate. A
possible explanation is that the wet steam cools the studs to a skin temperature of 200–230°F
(93–110°C), regardless of the temperature remote from the impingement point.
• There was no difference in corrosion rates for studs supported in carbon steel or stainless
steel fixtures.
• None of the coatings applied to the studs was effective in reducing the rate of corrosion.
• Some of the differences in corrosion rates between the tests may result from the fact that the
capillaries sometimes became clogged with boric acid crystals; therefore, the flow may not
have been continuous throughout the test.

4-49
 EPRI Licensed Material

Boric Acid Corrosion Tests

Conclusions

The main conclusion from these tests is that carbon and low-alloy steels can exhibit very high
corrosion rates in situations involving impingement of steam from borated water.

Figure 4-16
Test Apparatus for Impingement onto Bolts – Combustion Engineering Tests

4-50
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: L

Test Type: Impingement – Onto Plates

Org/Date: Combustion Engineering (1988)

Reference: “Boric Acid Corrosion of Carbon and Low-Alloy Steels.” Proceedings of the
4th International Symposium on Environmental Degradation of
Materials in Nuclear Power Systems–Water Reactors, Jekyll Island,
Georgia, August 6–10, 1989 [27].

Test Configuration, Procedure, and Results

A series of tests was conducted in which borated steam (1000 ppm boron) at 600°F (316°C)was
forced through a capillary tube in a steel block at 0.05–0.10 gpm (0.19–0.38 L/min) and
impinged upon a SA-533 Grade B steel target plate. The capillary and some of the target blocks
were heated to approximately 650°F (343°C) by internal electric resistance heaters. The test
durations were 370–390 hours.

Key test variables included the flow rate and the distance from the capillary tube to the target
plate. The test configuration is illustrated in Figure 4-17. Corrosion rates were measured in the
capillary tube, on surfaces of the capillary block not in direct contact with the flowing fluid, and
in the target plate.

Table 4-16 summarizes the key test conditions and results in inches per year (mm/yr).
Table 4-16
Corrosion Rate for Impingement onto Plates – Combustion Engineering Tests

Flow = .05 gpm Flow = .10 gpm

Location of Corrosion Measurements
(0.19 L/min) (0.38 L/min)

In Capillary Tube 0.81–1.16 (20.6–29.5) 0.85–0.94 (21.6–23.9)

Capillary Block Remote from Flow 0.33–0.80 (8.4–20.3) 0.62–1.07 (15.8–27.2)

Target Block 0.5 Inches (13 mm) from Capillary

Block Heated 6.39–9.41 (162.3–239.0) 9.28–11.1 (235.7–281.9)
Block Unheated 8.32–9.98 (211.3–253.5) -----

Target Block 2.0 Inches (51 mm) from Capillary

Block Heated 1.33–1.49 (33.8–37.8) 1.61–3.90 (40.9–99.1)
Block Unheated 0.39–3.60 (9.9–91.4) -----

Key observations from these tests are as follows:

• At locations on the capillary block not in contact with the flowing boric acid, the corrosion
rates were still a relatively high 0.3–1.1 in/yr (7.6–27.9 mm/yr). (Note: This condition does
not exist in PWR plants. The closest condition would be steam leaking past gaskets, and this
situation would involve stainless steel rather than carbon steel material.)

4-51
 EPRI Licensed Material

Boric Acid Corrosion Tests

• At the impingement locations on the target block, the corrosion rates varied from 0.4–3.9
in/yr (10–99 mm/yr) for targets 2 inches (51 mm) from the capillary and 6.4–11.1 in/yr (163–
282 mm/yr) for targets 0.5 inches (13 mm) from the capillary. These rates were most strongly
affected by the separation distance with the rate decreasing markedly with separation. The
corrosion rates were also a function of the flow rate with higher flow rates producing slightly
higher corrosion rates. Heating of the target block had little effect on the corrosion rates. It is
assumed that the impinging flow heated the local metal temperature to about 200°F (93°C)
that has been shown in other tests to result in high corrosion rates.
• The target blocks experienced corrosion rates of 0.4–0.6 in/yr (10–15 mm/yr) at locations
where there was no direct impingement.
• The test results showed that the diameter of the capillary hole increased at a rate
corresponding to 0.8–1.2 in/yr (20–30 mm/yr). This resulted from a combination of BAC and
erosion-corrosion caused by the flow of steam through the capillary. (Note: This condition is
not reported in Figure 4-3 because it does not represent any plant condition. The only
condition in a plant that would be close would be steam cutting due to leakage past gaskets,
etc. However, in these cases, the material in contact with the steam would likely be stainless
steel rather than carbon or low-alloy steel, and the erosion rate of the soft gasket material
would be higher than that of the steel.)

Conclusions

The main conclusion from this test is that direct impingement of steam with boric acid carryover
can produce corrosion rates to 10 in/yr (254 mm/yr) or higher depending upon the specific
circumstances.

4-52
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-17
Apparatus for Testing Impingement onto Plates – Combustion Engineering Tests

4-53
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: M

Test Type: Leakage Into Annulus

Org/Date: Combustion Engineering (1991)

Reference: “Boric Acid Corrosion of Low-Alloy Steel,” Proceedings: 1991 EPRI Workshop
on PWSCC of Alloy 600 in PWRs EPRI, TR-100852 [5].

Test Configuration, Procedure, and Results

These tests were conducted to determine the effect of leakage of borated water from a small
PWSCC-type crack into a small clearance annular opening. This test simulates the type of
leakage that could occur from a cracked pressurizer heater sleeve.

The test configuration is shown in Figure 4-18. The test apparatus consists of a short section of
Alloy 600 steam generator tube that contains a through-wall axial crack and that is inserted into
an alloy steel (SA-533B Class 1) collar with a clearance of 0.002–0.008 inches (0.05–0.20
mm/yr). The collar is heated to 600–650°F (316–343°C) by four electric resistance heaters.
Borated primary water (1000 ppm boron and 1–2 ppm lithium) is supplied to the inside of the
tube at 600°F (316°C) and allowed to leak out through the axial crack at a rate between 0.02 and
0.12 gpm (0.08 and 0.45 L/min).

The duration of the tests was about 300–700 hours. At the end of the testing, the collars were
removed and the amount of corrosion of the steel collar was measured.

Key observations from the tests on carbon and low-alloy steel materials are as follows:
• The corrosion tended to occur at the point where the leaking solution leaves the gap between
the tube and collar rather than at the point where the fluid exits the crack. It is hypothesized
that the lack of corrosion in most of the annulus is a result of low oxygen in this region.
• Corrosion rates were higher for low leakage rates (0.02 gpm [0.08 L/min]) than for higher
leakage rates (0.12 gpm [0.45 L/min]). A possible explanation is that higher rates are more
effective in purging oxygen from the annulus.
• Leak rates tended to decrease with time. The most probable explanation is that the annulus
eventually fills up with boric acid residue, and this residue reduces the leakage rate.
• The highest measured corrosion rate was 2.15 in/yr (55 mm/yr). The area of maximum
corrosion was very localized both axially and circumferentially. The maximum corrosion rate
3 3
was equivalent to a material volume loss of 1.07 in (17.5 cm ). (Note: The volume loss is not
necessarily conservative because the exposed area increases as the corrosion depth increases
and the rate of volume loss may also increase.)

Conclusions

The main conclusion from these tests is that corrosion rates from boric acid leaking into annular
crevices from cracked nozzles can be over 2 inches per year (51 mm/yr) and that the greatest

4-54
 EPRI Licensed Material

Boric Acid Corrosion Tests

corrosion will occur at the point where the leaking fluid leaves the annulus. As such, the greatest
amount of damage is likely to occur at a point that is typically available for visual inspection.
Conversely, there is little risk of a problem developing deep in an annular crevice where it
cannot easily be inspected.

The test data shown in Table 4-17 are from EPRI report TR-101108 [3].
Table 4-17
Corrosion Rates for Leaking Borated Water into an Annulus – Combustion Engineering
Tests

Test A B C D

Approximate Leak Rates in gpm (L/min) 0.05 (0.189) 0.008 (0.030) 0.002 (0.008) 0.007 (0.03)

Deepest Penetration in mils/yr (mm/yr) 100 (2.5) 124 (3.1) 21 (0.5) 128 (3.3)

Maximum Penetration Rate in mils/yr 1357 (34.5) 2086 (53.0) 641 (16.3) 2150 (54.6)
(mm/yr)

Average Penetration Rate in mils/yr 26.0 (0.7) 83.5 (2.1) 20.2 (0.5) 55.3 (1.4)
(mm/yr)
3 3 0.00031 0.00054 0.0010 0.00018
Upper Half Metal Loss in in (cm )
(0.005) (0.009) (0.016) (0.00295)

3 3
Loss Rate in in /yr (cm /yr) 0.0042 0.0091 0.031 0.0030
(0.069) (0.15) (0.508) (0.049)
3 3 0.023 0.063 0.0070 0.039
Lower Half Metal Loss in in (cm )
(0.38) (1.03) (0.115) (0.639)

3 3
Loss Rate in in /yr (cm /yr) 0.31 1.06 0.21 0.65
(5.08) (17.4) (3.44) (10.7)
3
Total Metal Loss (in ) 0.31 (5.08) 1.07 (17.5) 0.24 (3.93) 0.65 (10.7)

Lower/Upper Metal Loss 75 118 7 218

4-55
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-18
Test Configuration for Determining the Effect of Leakage into an Annulus – Combustion
Engineering Tests

4-56
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: EPRI-1

Test Type: Immersion – Aerated, Concentrated Boric Acid

Org/Date: EPRI (1996)

Reference: Boric Acid Corrosion Evaluation (BACE) Corrosion Program – Phase II

Corrosion Testing, Topical Report, 1998 [21].

Test Configuration, Procedure, and Results

EPRI conducted a series of tests in which seven carbon and low-alloy steels and three stainless
steels were immersed in a 20,000 ppm (as boron) boric acid, aerated solution at 180°F (82°C).
These tests were performed to simulate primary coolant that has leaked onto a moderately heated
metal surface and has been significantly concentrated by evaporation. Duplicate coupons of each
material were exposed for 15-, 30-, and 60-day periods to allow an evaluation of the variation in
corrosion rate with exposure time.

Average corrosion rates for each of the materials at each of the three test durations are listed in
Table 4-18. Corrosion rates for the carbon and low-alloy steels are plotted as a function of the
test duration in Figure 4-19.

4-57
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-18
Average Corrosion Rates in Aerated Boric Acid – EPRI Tests

Corrosion Rate in mils per year (mm/yr)

Material
15-Day Exposure 30-Day Exposure 60-Day Exposure
A193 Gr. B7 (4140) 102.77 ± 2.25 128.79 ± 0.70 97.18 ± 0.68
(2.61 ± 0.057) (3.27 ± 0.018) (2.47 ± 0.017)
A307 Gr. B 175.05 ± 1.37 130.92 ± 1.23 145.38 ± 0.99
(4.45 ± 0.035) (3.33 ± 0.031) (3.69 ± 0.025)
A540 Gr. B2, Cl. 3 (4340) 139.09 ± 0.27 137.76 ± 5.77 165.02 ± 1.29
(3.53 ± 0.007) (3.50 ± 0.147) (4.19 ± 0.033)
A106 Gr. B 132.68 ± 0.90 124.93 ± 2.21 147.53 ± 2.15
(3.37 ± 0.023) (3.17 ± 0.056) (3.75 ± 0.055)
A302 Gr. B 79.35 ± 0.31 86.28 ± 7.03 105.53 ± 6.31
(2.02 ± 0.008) (2.19 ± 0.179) (2.68 ± 0.160)
A508 Cl. 2 192.63 ± 0.80 158.84 ± 0.84 78.92 ± 10.16
(4.89 ± 0.020) (4.04 ± 0.021) (2.00 ± 0.258)
A533 Gr. B 126.32 ± 4.63 117.49 ± 5.69 108.21 ± 7.76
(3.21 ± 0.118) (2.98 ± 0.145) (2.75 ± 0.197)
A192 Gr. B6 (410 SS) 0.58 ± 0.00 0.65 ± 0.02 0.31 ± 0.39
(0.0147 ± 0.00) (0.0165 ± 0.0005) (0.0079 ± 0.0099)
A286 SS 0.00 ± 0.00 0.01 ± 0.00 0.01 ± 0.00
(0.00 ± 0.00) (0.0003 ± 0.00) (0.0003 ± 0.00)
A540 Type 630 (17-4 PH 0.00 ± 0.02 0.01 ± 0.01 0.01 ± 0.00
SS) (0.00 ± 0.0005) (0.0003 ± 0.0003) (0.0003 ± 0.00)

Key observations from these tests are:

• Corrosion rates ranged from 0.079 to 0.192 in/yr (2 to 4.9 mm/yr) for the carbon and low-
alloy steels immersed in the 20,000 ppm boron, 180°F (82°C) aerated solution.
• Only two of the seven steels (A508 Cl. 2 and A533 Gr. B) exhibited corrosion rates that
continually decreased with exposure time.
• Increasing concentration of Ni and Mo in steels appears to increase resistance to
concentrated, aerated boric acid.
• Stainless steels exhibited very low corrosion rates in the concentrated boric acid.

Conclusions

The main conclusion from these tests is that concentrated boric acid solutions can produce
significant corrosion of carbon and low-alloy steels at temperatures as low as 180°F (82°C).

4-58
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-19
Corrosion Rates in Aerated Boric Acid – EPRI Tests

4-59
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: EPRI-2

Test Type: Boric Acid Spray

Org/Date: EPRI (1996)

Reference: Boric Acid Corrosion Evaluation (BACE) Corrosion Program – Phase II

Corrosion Testing, Topical Report, 1998 [21].

Test Configuration, Procedure, and Results

EPRI conducted a series of tests in which a 2000 ppm (as boron) boric acid aerated solution at
100°F (38°C) was vaporized and deposited onto specimens in a standard salt spray chamber.
Seven carbon and low-alloy steels and three stainless steels were exposed for 15-, 30-, and 60-
day periods. These tests simulate leaking primary coolant that vaporizes, becomes airborne,
deposits on surfaces at low temperatures, and is moderately concentrated by evaporation.

Average corrosion rates for each of the materials at each of the three test durations are listed in
Table 4-19. Corrosion rates for the carbon and low-alloy steels are plotted as a function of the
test duration in Figure 4-20.

4-60
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-19
Average Corrosion Rates for Borated Water Spray – EPRI Tests

Corrosion Rate in mils per year (mm/yr)

Material 15-Day Exposure 30-Day 60-Day
Exposure Exposure
A193 Gr. B7 (4140) 4.20 ± 0.05 3.40 ± 0.92 2.65 ± 0.35
(0.107 ± 0.001) (0.0864 ± 0.0234) (0.0673 ± 0.0089)
A307 Gr. B 2.12 ± 0.26 4.16 ± 0.04 3.91 ± 0.23
(0.053 ± 0.0066) (0.1057 ± 0.0010) (0.993 ± 0.0058)
A540 Gr. B2, Cl. 3 (4340) 2.75 ± 0.39 4.11 ± 0.11 3.31 ± 0.06
(0.0699 ± 0.0099) (0.1044 ± 0.0028) (0.841 ± 0.0015)
A106 Gr. B 2.41 ± 0.75 3.94 ± 0.04 3.31 ± 0.06
(0.0612 ± 0.0191) (0.1001 ± 0.0010) (0.841 ± 0.0015)
A302 Gr. B 3.86 ± 0.08 4.04 ± 0.33 3.22 ± 0.09
(0.098 ± 0.0020) (0.1026 ± 0.0084) (0.0818 ± 0.0023)
A508 Cl. 2 1.96 ± 0.20 3.65 ± 0.04 3.36 ± 0.30
(0.0498 ± 0.0051) (0.0927 ± 0.0010) (0.0853 ± 0.0076)
A533 Gr. B 2.05 ± 0.15 4.44 ± 0.27 3.22 ± 0.07
(0.0521 ± 0.0038) (0.1128 ± 0.0069) (0.0818 ± 0.0018)
A192 Gr. B6 (410 SS) 0.01 ± 0.01 0.01 ± 0.01 0.01 ± 0.00
(0.0003 ± 0.0003) (0.0003 ± 0.0003) (0.0003 ± 0.00)
A286 SS 0.01 ± 0.01 0.01 ± 0.00 0.01 ± 0.00
(0.0003 ± 0.0003) (0.0003 ± 0.00) (0.0003 ± 0.00)
A540 Type 630 (17-4 PH SS) 0.01 ± 0.01 0.02 ± 0.01 0.01 ± 0.00
(0.0003 ± 0.0003) (0.0005 ± 0.0003) (0.0003 ± 0.00)

Key observations from these tests are:

• The carbon and low-alloy steels corroded at rates of 0.002–0.0045 in/yr (0.05–0.11 mm/yr)
in the boric acid spray.
• The stainless steels were unattacked in the boric acid spray.
• Corrosion rates for the carbon and low-alloy steel were about a factor of two lower than rates
observed in comparable immersion tests (Tests A and B).

Conclusions

The main conclusion from these tests is that vaporized primary coolant, even when moderately
concentrated, should not produce significant corrosion of carbon, low-alloy, or stainless steels at
low temperatures around 100°F (38°C).

4-61
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-20
Corrosion Rates for Borated Water Spray – EPRI Tests

4-62
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: EPRI-3

Test Type: Immersion – Boric Acid Crystals

Org/Date: EPRI (1996)

Reference: Boric Acid Corrosion Evaluation (BACE) Corrosion Program – Phase II

Corrosion Testing, Topical Report, 1998 [21].

Test Configuration, Procedure, and Results

EPRI conducted a series of tests in which seven carbon and low-alloy steels and three stainless
steels were buried in boric acid crystals at room temperature exposed to water-saturated flowing
air. Specimens were buried in vertical positions in the boric acid crystals, with one end
approximately 3/8 inch (9.5 mm) below the surface of the crystals and the other end 2-3/8 inches
(60.3 mm) below the surface. Duplicate specimens were removed after 15-, 30-, and 60-day
exposures. These tests were designed to simulate the condition in which leaking primary coolant
has deposited on metal surfaces, evaporated, and precipitated boric acid crystals on the metal
surfaces in a high-humidity environment.

Average corrosion rates for each of the materials at each of the three test durations are listed in
Table 4-20. Corrosion rates for the carbon and low-alloy steels are plotted as a function of test
duration in Figure 4-21.

4-63
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-20
Average Corrosion Rates for Steel Buried in Boric Acid Crystals – EPRI Tests

Corrosion Rate in mils per year (mm/yr)

Material 15-Day Exposure 30-Day 60-Day
Exposure Exposure
A193 Gr. B7 (4140) 0.04 ± 0.01 0.06 ±0.01 0.07 ± 0.02
(0.0010 ± 0.0003) (0.0015 ± 0.0003) (0.0018 ± 0.0005)
A307 Gr. B 0.02 ± 0.01 0.06 ± 0.00 0.09 ± 0.01
(0.0005 ± 0.0003) (0.0015 ± 0.00) (0.0023 ± 0.0003)
A540 Gr. B2, Cl. 3 (4340) 0.03 ± 0.01 0.04 ± 0.01 0.04 ± 0.01
(0.0008 ± 0.0003) (0.0010 ± 0.0003) (0.0010 ± 0.0003)
A106 Gr. B 0.02 ± 0.00 0.06 ± 0.00 0.11 ± 0.05
(0.0005 ± 0.00) (0.0015 ± 0.00) (0.0028 ± 0.0013)
A302 Gr. B 0.03 ± 0.01 0.48 ± 0.13 0.21 ± 0.03
(0.0008 ± 0.0003) (0.012 ± 0.0033) (0.0053 ± 0.0008)
A508 Cl. 2 0.02 ± 0.01 0.03 ± 0.00 0.09 ± 0.04
(0.0005 ± 0.0003) (0.0008 ± 0.00) (0.0023 ± 0.001)
A533 Gr. B 0.03 ± 0.00 0.04 ± 0.01 0.20 ± 0.01
(0.0008 ± 0.00) (0.0010 ± 0.0003) (0.0051 ± 0.0003)
A192 Gr. B6 (410 SS) 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00
(0.00 ± 0.00) (0.00 ± 0.00) (0.00 ± 0.00)
A286 SS 0.01 ± 0.00 0.01 ± 0.00 0.01 ± 0.00
(0.0003 ± 0.00) (0.0003 ± 0.00) (0.0003 ± 0.00)
A540 Type 630 (17-4 PH SS) 0.01 ± 0.00 0.02 ± 0.00 0.01 ± 0.01
(0.0003 ± 0.00) (0.0005 ± 0.00) (0.0003 ± 0.0003)

Key observations from these tests are:

• Corrosion rates of the carbon and low-alloy steels generally exhibited slight increases with
increased exposure time. This has been attributed to the diffusion of moisture into the boric
acid crystals.
• All of the materials (both steels and stainless steels) exhibited very low corrosion rates in the
boric acid crystals. The highest corrosion rate observed on any of the alloys was 0.0005 in/yr
(0.013 mm/yr).

Conclusions

The main conclusion from these tests is that dry boric acid crystals will not produce significant
corrosion of carbon, low-alloy or stainless steels, even in a humid environment.

4-64
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-21
Corrosion Rates for Steels Buried in Boric Acid Crystals – EPRI Tests

4-65
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: EPRI-4

Test Type: Immersion − Deaerated, Concentrated Boric Acid

Org/Date: EPRI (1996)

Reference: Boric Acid Corrosion Evaluation (BACE) Corrosion Program – Phase II

Corrosion Testing, Topical Report, 1998 [21].

Test Configuration, Procedure, and Results

EPRI conducted a series of tests to determine the corrosion rate of low-alloy steel exposed to a
deaerated, concentrated boric acid solution. Electrically isolated specimens of A193 Grade B7
(4140 steel) were immersed in deionized water containing 20,000 ppm boron for periods of 15,
30, and 60 days. The boric acid solutions were held at 180°F (82°C) and deaerated continuously
with flowing argon.

Average corrosion rates for each of the three test durations are listed in Table 4-21. Corrosion
rates are plotted as a function of test duration in Figure 4-22.
Table 4-21
Average Corrosion Rates in Deaerated Boric Acid – EPRI Tests

Test Duration Corrosion Rate in in/yr

(Days) (mm/yr)
15 0.0500 ± 0.00123 (1.3 ± 0.03)
30 0.0360 ± 0.00084 (0.91 ± 0.02)
60 0.0267 ± 0.00055 (0.68 ± 0.014)

Key observations from these tests are:

• Moderate corrosion rates occur in deaerated, concentrated boric acid solutions.
• Specimens exposed to the deaerated, concentrated boric acid solution were uniformly
covered with gray-black, adherent corrosion products.
• The corrosion rate decreased with test duration, apparently due to the formation of the
adherent corrosion product.

Conclusions

These results, when compared to results from aerated, concentrated solutions, clearly show that
the presence of oxygen in the environment greatly accelerates corrosion of carbon and low-alloy
steels. The results also show that hot, deaerated, concentrated boric acid can cause significant
general corrosion of carbon and low-alloy steels. However, at 180°F (82°C) the corrosion rates
are less than 0.1 in/yr (2.5 mm/yr).

4-66
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-22
Corrosion Rates in Deaerated Boric Acid – EPRI Tests

4-67
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: EPRI-5

Test Type: Dripping − Onto Heated Surface

Org/Date: EPRI (1998)

Reference: Boric Acid Corrosion Evaluation (BACE) Corrosion Program – Phase II

Corrosion Testing, Topical Report, 1998 [21].

Test Configuration, Procedure, and Results

EPRI performed a series of tests in which borated water (2000 ppm boron) heated to 180°F
(82°C) was allowed to drip onto A302 steel plates. Following 50-day exposures, all deposits
were removed from the plates, and the penetration depths were measured. Variables evaluated in
these tests included flow rate (0.01 and 0.10 gpm [0.038 and 0.38 L/min]), metal temperature
(180 and 600°F [82 and 316°C]), and plate geometry (horizontal surface and 45° inclined surface
with a drip bar). See Figure 4-23 for the test configurations.

In all cases, minimal corrosion was observed at the drip point. In the 600°F (316°C) tests, the
areas of maximum corrosion were present as one or more narrow troughs that appeared to
coincide with the edges of the boric acid flow lines. Trough depths decreased moving away from
the drip site. Corrosion troughs representing obvious flow lines were not present on either of the
180°F (82°C) plates. Maximum corrosion depths were independent of flow rate and plate
geometry. The maximum corrosion rates obtained in each of these tests are listed in Table 4-22.
Table 4-22
Maximum Corrosion Rates for Dripping Boric Acid – EPRI Tests

Flow Rate Maximum Corrosion

Temperature
Identification in gpm (L/min) Rate in inches/year
in °F (°C)
(mm/yr)

1a 600 (316) 0.01 (0.038) 1.7 (43)

1b 600 (316) 0.10 (0.38) 1.9 (48)

2a 600 (316) 0.01 (0.038) 1.5 (38)

2b 600 (316) 0.10 (0.38) 1.2 (30.5)

3a 180 (82) 0.01 (0.038) 0.22 (5.6)

3b 180 (82) 0.10 (0.38) 0.19 (4.8)

4-68
 EPRI Licensed Material

Boric Acid Corrosion Tests

Key observations from these tests are as follows:

• Dripping onto plates held at 180°F (82°C) resulted in maximum corrosion rates ranging from
0.19–0.22 in/yr (4.8–5.6 mm/yr).
• Dripping onto plates held at 600°F resulted in maximum corrosion rates ranging from 1.2–
1.9 in/yr (30.5–48.3 mm/yr).
• Flow rate and plate geometry did not influence the maximum corrosion rate.

Significant concentrating action brought about by evaporation from a surface held above the
boiling point of the boric acid is necessary to obtain corrosion rates above 1 in/yr (25 mm/yr).

Conclusions

The main conclusion from these tests is that the dripping of borated water onto hot carbon steel
surfaces can produce corrosion rates between 1 and 2 in/yr (25 and 51 mm/yr). If the surface
temperature is below the boiling point of the borated water, maximum corrosion rates are
reduced to about 0.2 in/yr (5 mm/yr).

Figure 4-23
Drip Test Configurations – EPRI Tests

4-69
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: EPRI-6

Test Type: Leakage into Annulus

Org/Date: EPRI (1998)

Reference: Boric Acid Corrosion Evaluation (BACE) Corrosion Program – Phase II

Corrosion Testing, Topical Report, 1998 [21].

Test Configuration, Procedure, and Results

EPRI conducted a series of tests in which deaerated, high-temperature (600°F [316°C]) borated
water (2000 ppm boron) was injected through an A302 Gr. B steel block into a 0.005-inch- (0.13
mm) wide annular gap between the block and a cylindrical 304 stainless steel insert. The annulus
was sealed at the bottom, and the blocks were heated to 600°F (316°C). The upward-pointing
nozzle was selected to represent the orientation in RPV head nozzles. CE had previously used a
downward orientation in their tests to simulate pressurizer heater sleeve nozzles. See Figure 4-25
for the test block assembly.

Variables that were evaluated included flow rate (0.01 and 0.10 gpm [0.038 and 0.38 L/min])
and specimen geometry (horizontal exit surface, angled exit surface, and angled exit surface
covered with insulation). Following the 50-day exposures, the test blocks were sectioned, using
an electric discharge machine (EDM), and the annular surfaces cleaned of all deposits.
Penetration depths were then measured on the cleaned surfaces.

The maximum penetration was always observed deep within the annulus close to the point at
which the steam was injected into the annulus. Penetration depth decreased as the annulus
opening was approached. Because the maximum penetration took place deep within the annulus,
neither the geometry of the exit surface nor the presence of insulation on the exit surface had a
significant effect on the maximum corrosion rates. The maximum corrosion rates measured for
each of the test conditions are listed in Table 4-23.

4-70
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-23
Maximum Corrosion Rates for Leakage into the Annulus – EPRI Tests

Temperature Flow Rate in gpm Maximum Corrosion Rate in

Identification
in °F (°°C) (L/min) inches/year (mm/yr)

4a 600 (316) 0.01 (0.038) 1.66 (42.2)

4b 600 (316) 0.10 (0.38) 0.891 (22.6)
5a 600 (316) 0.01 (0.038) 2.37 (60.2)
5b 600 (316) 0.10 (0.38) *
6a 600 (316) 0.01 (0.038) 1.75 (44.5)
6b 600 (316) 0.10 (0.38) 1.23 (31.2)

* The injection orifice on Test 5b plugged during the test, thus invalidating the corrosion rate measurement.

Figure 4-24
Nozzle Crack Leak Test Configurations – EPRI Tests

Key observations from these tests are as follows:

• The A302 Gr. B steel blocks exhibited maximum corrosion rates of 0.89–2.37 in/yr (22.6 to
60.2 mm/yr).
• Maximum attack did not occur near the annulus opening, but rather was consistently found
deep within the annulus near the boric acid injection point.
• Slightly higher corrosion rates were observed at the low flow condition suggesting that
wastage (possibly due to greater oxygen ingress through the annulus at low flow rates), rather
than erosion, was the primary mechanism.

Conclusions

The main conclusion from these tests is that boric acid leakage into an upwardly oriented annular
gap can produce corrosion rates 1 in/yr (25 mm/yr) or greater deep within the crevice where
visual inspection is not possible.

See Section 4.10 for a comparison of CE and EPRI test results for leakage into annular gaps
between nozzles and vessel shells.

4-71
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-25
Schematic of the Test Block for Leakage into the Annulus – EPRI Tests

4-72
 EPRI Licensed Material

Boric Acid Corrosion Tests

Test Ref.: EPRI-7

Test Type: Leakage from Flanged Joints

Org/Date: EPRI (1998)

Reference: Boric Acid Corrosion Evaluation (BACE) Corrosion Program – Phase II

Corrosion Testing, Topical Report, 1998 [21].

Test Configuration, Procedure, and Results

EPRI performed a series of 18 tests that simulated the conditions that occur when water leaking
from a gasketed joint impinges onto the bolts that hold the flange bolts. In these tests, an injector
tube was placed in a heated 10-inch (254-mm), Type 304 stainless steel flange so that the borated
water impinged directly onto one of the A193 B7 low-alloy steel bolts, as shown in Figure 4-26.
The bolts in the test fixture are 0.75-inch (19-mm) diameter. The flanges were heated to either
180°F (82°C) or 600°F (316°C), and the borated water was injected at the same temperature as
the flange. Corrosion rates of each of the eight bolts in the flange were determined at the end of
the 50-day exposure by measuring the minimum bolt diameter following removal of all corrosion
deposits.

The effects of temperature (180 and 600°F [82 and 316°C]), flow rate (0.01 and 0.10 gpm [0.038
and 0.38 L/min]), and flange orientation (horizontal, 45°inclined and vertical) were evaluated in
14 of the 18 tests. The effects of insulation and an oxygen limiting band were evaluated in six
tests, and two tests evaluated the performance of Type 304 stainless steel bolts.

The maximum corrosion rates for each of the tests are listed in Table 4-24. The corrosion rate
distributions around the flanges are shown in Figures 4-28 through 4-30.

4-73
 EPRI Licensed Material

Boric Acid Corrosion Tests

Table 4-24
Maximum Corrosion Rates for Flange Leakage – EPRI Tests

Maximum Average Ratio of

Flow Rate
Temperature Corrosion Corrosion Avg/Max
Flange Description Condition in gpm
in °F (°°C) Rate in Rate in Corrosion
(L/min)
in/yr (mm/yr) in/yr (mm/yr) Rate

Horizontal 7a 600 (316) 0.01 1.155 (29.3) 0.233 (5.9) 0.20

(0.038)

Horizontal 7b 600 (316) 0.10 (0.38) 0.113 (2.9) 0.017 (0.43) 0.15

Angled (45°) 8a 600 (316) 0.01 0.825 (21) 0.216 (5.5) 0.26
(0.038)

Angled (45°) 8b 600 (316) 0.10 (0.38) 0.599 (15) 0.107 (2.7) 0.18

Vertical 11a 600 (316) 0.01 0.395 (10) 0.121 (3.0) 0.31
(0.038)

Vertical 11b 600 (316) 0.10 (0.38) 0.277 (7.0) 0.112 (2.8) 0.40

Horizontal (Banded) 9a 600 (316) 0.01 0.920 (23) 0.121 (3.0) 0.13
(0.038)

Horizontal (Banded) 9b 600 (316) 0.10 (0.38) 0.307 (7.8) 0.233 (5.9) 0.76

Horizontal (Insulated) 14a 600 (316) 0.01 0.690 (18) 0.091 (2.3) 0.13
(0.038)

Horizontal (Insulated) 14b 600 (316) 0.10 (0.38) 1.710 (43) 0.746 (19) 0.44

Horizontal (304 Bolts) 15a 600 (316) 0.01 0.004 (0.10) 0.001 (0.025) 0.25
(0.038)

Horizontal (304 Bolts) 15b 600 (316) 0.10 (0.38) 0.007 (0.18) 0.003 (0.08) 0.43

Horizontal 10a 180 (82) 0.01 0.037 (0.94) 0.018 (0.46) 0.49
(0.038)

Horizontal 10b 180 (82) 0.10 (0.38) 0.036 (0.91) 0.018 (0.46) 0.50

Vertical 12a 180 (82) 0.01 0.018 (0.5) 0.006 (0.15) 0.33
(0.038)

Vertical 12b 180 (82) 0.10 (0.38) 0.028 (0.71) 0.018 (0.46) 0.64

Horizontal (Insulated) 13a 180 (82) 0.01 0.015 (0.38) 0.009 (0.23) 0.60
(0.038)

Horizontal (Insulated) 13b 180 (82) 0.10 (0.38) 0.011 (0.28) 0.005 (0.13) 0.45

4-74
 EPRI Licensed Material

Boric Acid Corrosion Tests

Key observations from these tests are as follows:

• The maximum corrosion rates of the A193 B7 bolts exhibited a very strong temperature
dependence. Maximum corrosion rates at 180°F (82°C) were low, ranging from 0.01–0.04
in/yr (.25 to 1.0 mm/yr). Maximum corrosion rates at 600°F (316°C) were much higher,
ranging from 0.13–1.7 in/yr (3.3 to 43 mm/yr).
• Flow rate did not have a significant effect on the maximum corrosion rate of the bolts. In
most of the 600°F (316°C) tests, a slightly higher maximum corrosion rate was observed at
the low flow condition.
• In most cases, the corrosion rate varied significantly from bolt to bolt within a single flange,
with many of the bolts exhibiting minimal attack. The flange geometry influenced this
corrosion pattern.
• Flange geometries that were likely to contain large quantities of borated water within the
flange, that is, those with a physical barrier, such as an oxygen band or insulation, and a high
flow rate, exhibited high corrosion rates on all of the bolts within the flange.
• The 304 stainless steel bolts exhibited very low corrosion rates, 0.007 in/yr (0.18 mm/yr) or
less, during the 600°F (316°C) exposure.

Conclusions

These tests indicate that rapid bolt corrosion of 0.1–2 in/yr (2.5–51 mm/yr) can occur at the
worst case bolts in a leaking flange. These high rates would be expected only for flanges
operating at temperatures above the boiling temperature of borated water. Much lower rates, less
than 0.04 in/yr (1.0 mm/yr), should occur on flanges that operate below the boiling point of
borated water. Barriers that impede the flow of borated water out of the flange can produce a
more serious situation in which all of the bolts within the flange corrode at a high rate, although
a tight-fitting band installed around the flange shows some potential to reduce corrosion,
probably as a result of minimizing oxygen ingress to the joint.

4-75
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-26
Fixture for Flange Leakage Tests – EPRI Tests

See Section 4.8 for a summary of these test results.

4-76
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-27 shows the flange specimen used in the leakage experiments. These flange geometries
were evaluated in the test program. The arrow indicates the approximate location of the borated
water injection.

Figure 4-27
Flanges Test Conditions – EPRI Tests

Figures 4-28 shows corrosion rates for each of the bolts around the flange for uninsulated flanges
at 600°F (316°C).

Figures 4-29 shows corrosion rates for each of the bolts around the flange for insulated or banded
flanges at 600°F (316°C).

Figures 4-30 shows corrosion rates for each of the bolts around flanges at 180°F (82°C).

4-77
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-28
Corrosion Rates for Bolts Around Uninsulated Flanges at 600 °F (316°°C) – EPRI Tests

4-78
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-29
Corrosion Rates for Bolts Around Insulated or Banded Flanges at 600 °F (316°°C) – EPRI
Tests

4-79
 EPRI Licensed Material

Boric Acid Corrosion Tests

Figure 4-30
Corrosion Rates for Bolts Around Flanges at 180°°F (82°°C) – EPRI Tests

4-80
 EPRI Licensed Material

5
REGULATORY BASES

This section briefly reviews the regulatory bases pertaining to leakage of borated water and
possible resultant corrosion. These include federal requirements, NRC requirements, and ASME
Boiler and Pressure Vessel Code requirements.

5.1 Code of Federal Regulations

Requirements pertaining to the design, fabrication, erection, and testing of pressure boundary
parts are described in the General Design Criteria (GDC) in Appendix A of Title 10 of the Code
of Federal Regulations, Part 50 (10CFR50) [32]. To ensure that PWR systems satisfy GDC
requirements, certain phenomena that could compromise the integrity of the primary coolant
pressure boundary must be considered. Boric acid corrosion is one of these phenomena. General
Design Criteria 14, 15, 30, 31, and 32 are considered to apply to the leakage of borated water and
the possible corrosion produced by leakage:
• Criterion 14 – Reactor Coolant Pressure Boundary

“The reactor coolant pressure boundary shall be designed, fabricated, erected, and tested so
as to have an extremely low probability of abnormal leakage, of rapidly propagating failure,
and of gross rupture.”
• Criterion 15 – Reactor Coolant System Design

“The reactor coolant system and associated auxiliary, control, and protection systems shall be
designed with sufficient margin to assure that the design conditions of the reactor coolant
pressure boundary are not exceeded during any conditions of normal operation, including
operational occurrences.”
• Criterion 30 – Quality of Reactor Coolant Pressure Boundary

“Components which are part of the reactor coolant pressure boundary shall be designed,
fabricated, erected, and tested to the highest quality standards practical. Means shall be
provided for detecting and, to the extent practical, identifying the location of the source of
reactor coolant leakage.”
• Criterion 31 – Fracture Prevention of Reactor Coolant Pressure Boundary

“The reactor coolant pressure boundary shall be designed with sufficient margin to assure
that when stressed under operating, maintenance, testing, and postulated accident conditions
(1) the boundary behaves in a nonbrittle manner, and (2) the probability of rapidly
propagating fracture is minimized. The design shall reflect consideration of service

5-1
 EPRI Licensed Material

Regulatory Bases

temperatures and other conditions of the boundary material under operating, maintenance,
testing and postulated accident conditions and the uncertainties in determining (1) material
properties, (2) the effects of irradiation on material properties, (3) residual, steady state and
transient stresses, and (4) flaw sizes.”
• Criterion 32 – Inspection of Reactor Coolant Pressure Boundary

“Components which are part of the reactor coolant pressure boundary shall be designed to
permit (1) periodic inspection and testing of important areas and features to assess their
structural and leak tight integrity, and (2) an appropriate material surveillance program for
the reactor pressure vessel.”

The first theme in these criteria is to ensure that there is an extremely low probability of
abnormal leakage. The criteria do not state that there shall be no leakage, only that there be no
“abnormal leakage.” This can be interpreted to mean that some small amount of leakage from
“normal” sources such as gaskets and packing is permissible if it is monitored so that it will not
result in abnormal leakage.

The second theme in these requirements is that components should be designed so that there will
be no rupture or rapidly propagating failure as a result of any operating condition. This requires
that components and systems must be:
• Designed so that they will not leak
• Designed so that they will not degrade if exposed to leakage of borated water
• Designed so that they will exhibit leak-before-break behavior
• Inspected under a program that will ensure that leakage and degradation will be discovered
before there is a risk of rupture or rapidly propagating failure

The third theme in these requirements is that programs must be developed and implemented to
detect reactor coolant leakage, determine the source of the leakage, and evaluate any damage
created by the leakage. This includes a requirement that components and systems must be
designed to permit these inspections to be performed.

5.2 NRC Information and Requirements

The NRC has provided three types of documents relating to leakage of borated reactor coolant
and resultant degradation:
• Notification of potential problems
• Reports of research and evaluations
• Requirements for utilities to take specific action

The documents issued by the NRC are described in the subsections that follow.

5-2
 EPRI Licensed Material

Regulatory Bases

5.2.1 Documents Notifying Utilities of Potential Problems

Information Notices are issued by the NRC to alert utilities about conditions that are potentially
adverse to plant safety. However, Information Notices do not require that the utilities make a
formal response to the NRC. The following Information Notices have been issued relating to
boric acid leakage and resultant corrosion:
• Information Notice 80-27 (June 11, 1980)
This document describes corrosion that occurred on the Ft. Calhoun reactor coolant pumps as
a result of leakage of borated water from the closure gasket [11]. The document noted that
this condition could have led to a loss of primary coolant system integrity if it had not been
detected. The document also noted the need for supplementing ultrasonic inspections of bolts
with visual inspections and possibly with leak detection systems.
• Information Notice 82-06 (March 12, 1982)
This document described stress corrosion cracking of steam generator manway studs at
Maine Yankee [12]. Although this condition did not represent classical wastage by contact
with borated water, the notice discussed the possible interaction of stud lubrication, stud
preload, borated primary coolant, and use of an injected sealant to stop the leak.
• Information Notice 86-108 (December 29, 1986)
This document described significant wastage of the high-pressure injection (HPI) nozzle at
ANO-1 that resulted from a leak at the body-to-bonnet joint of an HPI isolation valve [13].
The leakage from the isolation valve had been known for some time, but inspections had not
revealed that the leakage was running down between the attached stainless steel high-
pressure injection pipe and its insulation and corroding the carbon steel nozzle at the point
where it enters the reactor coolant piping. The document noted that corrosion rates in excess
of 1 inch (25 mm) per year can occur under some conditions where borated reactor coolant
leaks onto carbon or low-alloy steel parts and remains wetted.
• Information Notice 86-108, Supplement 1 (April 20, 1987)
This document described leakage of reactor coolant from an instrument nozzle Conoseal joint
at Turkey Point 4 [14]. This incident resulted in 500 pounds (227 kg) of boric acid being
deposited on the top of the reactor vessel head and causing corrosion of the low-alloy steel
head and several reactor vessel head studs.
• Information Notice 86-108, Supplement 2 (November 19, 1987)
This document described two incidents involving severe boric acid corrosion [15]. The first
incident occurred at Salem 2 where leakage from several pinholes in a CRDM nozzle canopy
seal weld resulted in a large amount of boric acid being deposited on the vessel head, which
resulted in corrosion of the head material in several areas. The second incident occurred at
San Onofre 2 where 18,000 gallons (68,137 L) of reactor coolant spilled into the containment
following the failure of all of the low-alloy steel studs that held the packing follower on a
cooling system isolation valve.
• Information Notice 86-108, Supplement 3 (January 5, 1995)
This document described two incidents involving severe boric acid corrosion [33]. The first
incident occurred at Calvert Cliffs 1 in 1994 during which three carbon steel nuts on an in-
core instrument flange were found to be corroded by leakage of borated water. One of the
nuts had failed completely, and the mating stud had fallen out of the flange. The second

5-3
 EPRI Licensed Material

Regulatory Bases

incident occurred at Three Mile Island 1 in 1994 during which leakage from a pressurizer
spray valve body-to-bonnet joint increased rapidly while attempting to tighten the bolts.
Three of the eight A193 B7 bolts had failed completely due to boric acid corrosion, and a
fourth was severely degraded.

The lesson from both cases is that the primary defense against boric acid corrosion is to
minimize leakage, detect and stop leaks soon after they start, and promptly clean up any
boric acid deposits.
• Information Notice 94-63, Boric Acid Corrosion of Charging Pump Casing Caused by
Cladding Cracks
This document [16] describes deep corrosion in carbon steel pump casings resulting from
boric acid coming into contact with the carbon steel via cracks in stainless steel cladding. The
high corrosion rate implies higher than normal primary loop oxygen content.

5.2.2 Reports of Research and Evaluations

The NRC publishes reports describing research work sponsored by the NRC. Several of these
documents have dealt with the issue of degradation caused by leakage of borated reactor coolant:
• NUREG/CR-2827, Boric Acid Corrosion of Ferritic Reactor Components
This document reported the results of a survey and some testing by Brookhaven National
Laboratory relating to corrosion of ferritic steel components by borated water [8]. The results
of this effort are summarized elsewhere in this guidebook.
• NUREG-0943, Threaded Fastener Experience in Nuclear Power Plants
This document reported on the results of a survey of degradation of bolting in nuclear plants
between October 1964 and March 1982 [9]. A total of 44 occurrences of fastener degradation
were reported. Incidents involving degradation by borated water represented 13 of the 44
cases. The document noted that in all 13 of the incidents related to boric acid the leakage
occurred from the component gaskets or seals and that the damage was discovered while
inspecting for the source of the leakage.
• NUREG-5576, Survey of Boric Acid Corrosion of Carbon Steel Components in Nuclear
Plants
This document reports on the evaluation of 50 utility responses to NRC Generic Letter 88-05
and audits of 10 plants [10]. The conclusion from the evaluations and audits was that all
responses met the intent of Generic Letter 88-05.

5.2.3 Requirements for Utilities to Take Specific Action

The NRC has issued two documents requiring utilities to take action regarding leakage of
borated reactor coolant and possible resultant degradation and to report the results of this effort
to the NRC. These are:
• IE Bulletin 82-02, Degradation of Threaded Fasteners in the Reactor Coolant Pressure
Boundary of PWR Plants (June 2, 1982)
This bulletin required that utilities take specified action regarding threaded fasteners in
reactor coolant pressure boundary locations including: 1) pressurizer and steam generator

5-4
 EPRI Licensed Material

Regulatory Bases

manways, 2) valve bonnets and pump flanges on lines with nominal diameters of 6 inches
(152 mm) or larger, and 3) CRDM flanges and pressurizer heater connections without seal
welds [17]. Reactor vessel closure studs were excluded for plants committed to the material
and inspection requirements of Regulatory Guide 1.65. The required action was as follows:
– Maintenance Procedures and QA Measures

“Where procedures do not exist, develop and implement maintenance procedures for
threaded fastener practices. These procedures should include, but not limited to the
following: (1) maintenance crew training of proper bolting/stud practices, (2)
detensioning and retensioning practices (torque iteration), specified tolerances, and other
controls for disassembly and reassembly of component closure/seal connections, (3)
gasket installation and controls, and (4) retensioning methods and other measures to
eliminate reactor coolant leakage during operation.

“Quality assurance measures should also be established for proper selection,

procurement, and application of fastener lubricants and injection sealant compounds to
minimize fastener susceptibility to SCC environments.”
– Fastener Inspections

“Threaded fasteners of closure connections, identified in the scope of this bulletin, when
opened for component inspection or maintenance shall be removed, cleaned and
inspected per IWA-2210 and IWA-2220 of the ASME Code Section XI (1974 edition or
later) before being reused. (Fasteners “seized” or designed with interference fit, may be
inspected in place.)”
– Operating Experience

“NRC Information Notice Nos. 80-27 and 82-06, and similar INPO correspondence (with
recommendations) have been issued in regard to corrosion problems associated with
bolts/studs in RCPB closures (INPO/NSAC SER 81-12). To assist the NRC in its
ongoing review and assessment of the problem you are asked to provide the following
information for closures and connections within the scope of this bulletin:

"Identify those bolted closures of the RCPB that have experienced leakage,
particularly those locations where leakage occurred during the most recent plant
operating cycle. Describe the inspections made and corrective measures taken to
eliminate the problem. If the leakage was attributed to gasket failure, or it’s design, so
indicate.

"Identify those closures and connections, if any, where fastener lubricants and
injection sealant, materials have been or are being used and report on plant experience
with their application particularly any instances of SCC of fasteners. Include types
and composition of materials used.”

Utilities were required to submit written reports summarizing the results of the work
performed pursuant to this IE Bulletin.

5-5
 EPRI Licensed Material

Regulatory Bases

• NRC Generic Letter 88-05, Boric Acid Corrosion of Carbon Steel Reactor Pressure
Boundary Components in PWR Plants (March 17, 1988)
This document requires that utilities establish programs to ensure that there is an extremely
low probability of abnormal leakage, rapidly propagating failure, or gross rupture as a result
of boric acid corrosion from the primary coolant loop components [4]. The Generic Letter
requires that the programs include the following:
– Determination of Locations Where Small Leaks Can Cause Degradation

“A determination of the principal locations where leaks that are smaller than the
allowable technical specification limit can cause degradation of the primary pressure
boundary by boric acid corrosion. Particular consideration should be given to identifying
those locations where conditions exist that could cause high concentrations of boric acid
on pressure boundary surfaces.”
– Procedures for Locating Small Coolant Leaks

“Procedures for locating small coolant leaks (i.e., leakage rates at less than technical
specification limits). It is important to establish the potential path of the leaking coolant
and the reactor pressure boundary components it is likely to contact. This information is
important in determining the interaction between the leaking coolant and the reactor
coolant pressure boundary materials.”
– Methods for Conducting Examinations and Engineering Evaluations

“Methods for conducting examinations and performing engineering evaluations to

establish the impact on the reactor coolant pressure boundary when leakage is located.
This should include procedures to promptly gather the necessary information for an
engineering evaluation before the removal of evidence of leakage, such as boric acid
buildup.”
– Corrective Action Program

“Corrective actions to prevent recurrence of this type of corrosion. This should include
any modifications to be introduced in the present design or operating procedures of the
plant that (a) reduce the probability of primary coolant leaks at the locations where they
may cause corrosion damage, and (b) entail the use of suitable corrosion resistant
materials or the application of protective coatings/claddings.”
– Additional Comments - The attachment to the Generic Letter 88-05 further states that:

“The most effective way to prevent boric acid corrosion is to minimize reactor coolant
leakage. This can be achieved by frequent monitoring of the locations where potential
leakage could occur and repairing the leaky components as soon as possible.”

“Because of the nature of the corrosion produced by boric acid, the most reliable method
of inspection of the component is by visual examination.”

5-6
 EPRI Licensed Material

Regulatory Bases

As previously noted, the NRC reviewed 50 utility responses to Generic Letter 88-05 and
audited 10 licensees. This work, reported in NUREG-5576 [10], showed that the licensees
had implemented programs that met the intent of Generic Letter 88-05.

5.3 ASME Code Requirements

Requirements in the ASME Code that relate to leakage from Alloy 600 CRDM nozzles are
contained in Section XI, Rules for Inservice Inspection [34]. The applicable requirements of
Section XI are as follows:

5.3.1 Requirements for Visual Inspections

Paragraph IWA-5211 and Table IWA-5210-1 specify that VT-2 visual examinations be
performed during system leak tests. The leak tests are generally to be performed at the system
operating pressure and temperature.

Paragraph IWA-5241(a) specifies that the accessible external surfaces of noninsulated

components be examined for evidence of leakage. The specified hold time at pressure for
noninsulated components is 10 minutes per Paragraph IWA-5213(a).

Paragraph IWA-5242(b) specifies that, for systems borated for the purpose of controlling
reactivity, the insulation must be removed from pressure-retaining bolted connections for the
VT-2 visual examination. For other insulated components, the VT-2 visual examination can be
conducted without removing insulation by examining the accessible and exposed surfaces and
insulation joints. The specified hold time at pressure for insulated components is four hours per
Paragraph IWA-5213(a).

Since most leak tests are performed at the end of an outage, the requirement to remove the
insulation on systems containing water borated for the purpose of controlling reactivity can
extend the outage duration to provide time for reinstalling insulation and removing scaffolding.
This issue has been addressed by two ASME Code Cases.
• Code Case N-533-1 (February 26, 1999) permits final leak checks under pressure with
insulation in place if the insulation is removed and the parts examined per VT-2 when not
under pressure. In summary, this code case requires that the insulation be removed but
eliminates the need for the insulation to be off during the final leak check. The basis for this
relief is that if the joint were leaking, boric acid crystals would be discovered when the
insulation is removed.
• Code Case N-616 (May 7, 1999) eliminates the need to remove insulation when corrosion-
resistant fasteners are used that have chromium content greater than or equal to 10% such as
17-4 PH stainless steel, 410 stainless steel, and A-286. For these cases, the leak check can be
performed without the need to remove insulation.

While the above two code cases relax inspection requirements for insulated components, they
have not been accepted by the NRC under 10CFR55(a). Use of these code cases requires a
formal relief request by the utility. The NRC has taken issue with Code Case N-616 in that the

5-7
 EPRI Licensed Material

Regulatory Bases

code case gives blanket relief based solely on chromium content of the material. As of February
2001, it is understood that the NRC is granting relief per the above two code cases with three
qualifications:
• 17-4 PH stainless steel and 410 stainless steel fasteners must be aged at temperatures 1100 °F
(593°C) or higher or have a hardness less than or equal to HRC 30.
• A-286 fasteners must have a preload stress less than 100 ksi (690 MPa).
• The leak test hold time must be four hours or more.

These restrictions are not considered to be unreasonable. The EPRI Materials Handbook for
Nuclear Plant Pressure Boundary Applications [35], specifies that 17-4 PH stainless steel should
be ordered in the H1100 or H1150 temper, 410 martensitic stainless steel should have a hardness
of HRC 26 or less, and that peak stresses (including stress concentration factors) for A-286
materials should be less than 80–90% of yield strength (maximum 72-85 ksi [496–586 MPa]).
These recommendations are fully consistent with the NRC qualifications. The Materials
Handbook adds other precautions such as not using 17-4 pH materials for long periods of time
above 500oF (260°C) to avoid embrittlement and that 17-4 pH and 410 stainless steel are both
susceptible to stress corrosion cracking in environments containing chlorides or sulfides, etc.

In summary, for components in systems containing water borated to control reactivity:

• Use uninsulated flanges where practical to permit easy VT-2 inspections during the final leak
test.
• Use corrosion-resistant fasteners complying with the NRC and EPRI Materials Handbook
guidelines with insulated flanges to eliminate the need to remove the insulation. However,
the leak test with the insulation in place must have a four-hour hold time.
• For insulated flanges without corrosion-resistant fasteners, the final VT-2 inspection at
pressure and temperature (four-hour hold time) can be performed with the insulation in place
if the insulation is removed for VT-2 examination at zero pressure at some point during the
outage.

In summary, it should be possible to obtain relief from the NRC to permit a timely and cost-
effective VT-2 inspection program for systems containing water borated to control reactivity.
There are two advantages to using corrosion-resistant fasteners for these applications. First, there
is no need to remove insulation for the VT-2 examination during outages. Second, if a leak does
occur, there is no concern regarding fastener degradation.

5.3.2 Conditions Requiring Corrective Action

Paragraph IWB-3522.1 specifies that the following conditions require corrective action:
• Leakage from insulated components or inaccessible components
• Areas of general corrosion resulting from leakage
• Discoloration or accumulated residues on surfaces of components, insulation, or floor areas
that may be evidence of borated water leakage

5-8
 EPRI Licensed Material

Regulatory Bases

5.3.3 Required Corrective Action

Paragraph IWB-3140 specifies that leakage detected by visual inspection must be corrected by
repairs or replacement. If surface and/or volumetric inspections are performed, defects can be
accepted if they are shown to be acceptable by analysis according to the requirements of
Paragraph IWB-3142.4. The requirements of Section III can be used for these evaluations if
Paragraph IWB does not apply.

5.3.4 Review of Repair Program

Paragraph IWB-3144 specifies that the repair program shall be 1) subjected to review by the
enforcement authorities having jurisdiction at the plant site and 2) submitted to the regulatory
authority having jurisdiction at the plant site.

5.4 Summary of Requirements

The requirements of the Code of Federal Regulation, the NRC, and the ASME Boiler and
Pressure Vessel Code are remarkably similar as they apply to leakage from reactor coolant
systems. The common themes in these requirements are:
• Closures should be designed to have a low risk of leakage.
• Closure designs and materials should be such that there is a low risk of rupture or abnormally
high leakage.
• Closures should be fabricated and assembled to have a low risk of leakage.
• Inspection programs should be developed and implemented to find leakage and to determine
other areas where the leakage could have flowed or accumulated.
• Corrective action should be implemented to correct situations where leakage has occurred.
• The NRC is prepared to grant relief from the ASME Code requirements to remove insulation
during the final VT-2 inspections of insulated flanges. However, several reasonable
concessions have to be made to obtain the relief.

5-9
 EPRI Licensed Material

6
DETECTING LEAKAGE

This section describes regulatory and typical plant Technical Specification leakage detection
requirements. It also provides a summary of methods that can be used to detect leakage of
borated water during plant operation and during refueling outages.

6.1 Regulatory Requirements

Requirements to detect leakage are rooted in 10CFR50 Appendix B. Guidance in meeting these
requirements is provided in Regulatory Guide 1.45. A key factor in leakage detection is the
concept of identified and unidentified leakage. The following subsections summarize these
topics.

6.1.1 10CFR50 Appendix B Requirements

As noted in Section 5.1 of this document, General Design Criterion (GDC) 30, Quality of
Reactor Coolant Pressure Boundary, of 10CFR50 Appendix A requires that means shall be
provided to detect and, to the extent practical, identify the source of reactor coolant leakage.

6.1.2 Identified vs. Unidentified Leakage

Identified leakage consists of leaks that occur from joints in a system (gaskets, valve packing,
seals, etc.) that 1) are captured and collected, and 2) do not interfere with the operation of
leakage detection systems.

Unidentified leakage consists of all leaks that do not meet the criteria for identified leakage. All
leaks from non-isolable through-wall cracks in RCS pressure boundary components
(components, pipes, vessel walls, etc.) are counted as unidentified leaks.

The requirements for unidentified leaks are more strict than for identified leaks.

6.1.3 Regulatory Guide 1.45 Guidelines

Regulatory Guide 1.45, Reactor Coolant Pressure Boundary Leakage Detection Systems,
contains guidance related to satisfying GDC 30. This guidance applies to reactor coolant leakage
into the primary containment. This document does not cover leakage outside of containment.
NUREG/CR-6582, Assessment of Pressurized Water Reactor Primary System Leaks, [36]
summarizes Regulatory Guide 1.45 requirements as follows:

6-1
 EPRI Licensed Material

Detecting Leakage

• Flow rates of identified leakage should be monitored separately from unidentified leakage.
• Flow rates of unidentified leakage should be monitored with an accuracy of 1 gpm (3.8
L/min) or better.
• At least three separate leak detection methods should be used to ensure effective monitoring
during periods when some detection systems may be ineffective or inoperable. Two of these
methods should be sump level and rate monitoring, and airborne particulate radioactivity
monitoring. The third method can be either condensate flow rate monitoring or airborne
gaseous radioactivity monitoring. Use of humidity, temperature, or pressure monitoring of
the containment atmosphere is also recommended as part of a leak detection system.
• Each of the three selected leakage detection systems should be able to detect an unidentified
leak of 1 gpm (3.8 L/min) or smaller in less than one hour.
• The leakage detection systems should be capable of performing their function following
seismic events that do not require plant shutdown.
• Indicators and alarms for each leakage detection system should be located in the control
room.

6.2 Technical Specification Requirements

Plant Technical Specifications incorporate requirements regarding leakage from the RCS.
Typical requirements pertaining to RCS leakage are as follows:
• Leak from Non-Isolable Fault in RCS Pressure Boundary
If any leakage occurs from a non-isolable fault in the RCS pressure boundary (vessel, piping,
valve body, etc.), the reactor must be brought to hot standby within 6 hours and to cold
shutdown within 30 hours.
• Unidentified RCS Leakage
If any unidentified RCS leakage exceeds 1 gpm (3.8 L/min), the reactor must be brought to
hot standby within 6 hours and to cold shutdown within 30 hours.
• Identifiable Leakage
If the total RCS leakage from identified and unidentified sources exceeds 10 gpm (38 L/min),
the reactor must be brought to hot standby within 6 hours and to cold shutdown within 30
hours.
• Controlled Leakage
Loss of reactor coolant through reactor coolant pump seals and other systems that can return
coolant to the RCS is not considered leakage. However, when added to the identified and
unidentified losses, the total must not exceed 30 gpm (114 L/min).
• Evaluation of Leakage
Action to evaluate the safety implications of leakage shall be initiated within four hours of
detection. The rate and conditions of shutdown must be determined based on the safety
analysis.

6-2
 EPRI Licensed Material

Detecting Leakage

• Corrective Action
If the reactor must be shut down as a result of leakage, the reactor must not restart until the
problem is repaired or otherwise corrected.
• Leakage Detection Systems
Whenever the reactor is critical and above 2% power, two leak detection systems of different
operating principles must be operable, and one of the two systems must be sensitive to
radioactivity. The system sensitive to radioactivity can be out of service for 48 hours if two
other means of detecting leakage are operable.

In summary, leakage of borated water from the RCS in excess of 1 gpm (3.8 L/min) must be
detected and acted upon quickly. Typically, utilities try to maintain the RCS leakage at the
lowest possible rate in order to ensure that large leaks do not mask the effect of a new small leak.

6.3 Leakage Detection Inside and Outside Containment

NUREG/CR-6582 reports the results of a survey of the methods used to detect leaks for 204
leakage-related licensee event report (LER) incidents between 1985 and September 1996. The
results, broken down by leaks inside and outside containment, are plotted in Figure 6-1.

For leaks occurring inside containment, control room indications/alarms, personal observation,
and periodic surveillance accounted for approximately 75% of the leak reports. Control room
indications/alarms identified about 35% of the total leaks. For leaks occurring outside
containment, personal observation and periodic surveillance accounted for almost 65% of the
leaks reported.

The above data suggest that several different parallel approaches are required to detect leaks
because each identified method has its strengths and limitations. This is the approach specified in
Regulatory Guide 1.45 and reinforced by other guidance such as Generic Letter 88-05 that
requires periodic walkdowns for boric acid leaks.

6.4 Summary of Leakage Detection Methods

The following is a summary of the main leakage detection methods. The sensitivities of these
methods are illustrated in Figure 6-2.

6.4.1 Primary On-Line Leakage Detection Methods

The following are the main on-line leakage detection methods. In reviewing these methods, it is
obvious that the detection sensitivity depends upon maintaining a low level of identifiable
leakage so that this leakage from known locations does not mask new small leaks.
• Containment Air Particulate Monitors
Containment air particulate monitors are an important method of detecting leakage from the
RCS. The sensitivity of this approach is a function of the containment volume, coolant
activity level, and containment background activity. If the plant has low coolant activity, the

6-3
 EPRI Licensed Material

Detecting Leakage

sensitivity is not as good as it would be if the plant had higher coolant activity. Assuming
normal corrosion product activity and no failed fuel, a RCS leak of 1 gpm (3.8 L/min) can be
detected within about 10 minutes of its occurrence by the air particulate monitors. With some
failed fuel, this leakage can be detected more quickly.
• Containment Gas Activity Monitors
Containment gas activity monitors are also an important means of measuring leakage from
the RCS. The sensitivity of this approach is also a function of the containment volume,
coolant activity level, and containment background activity. If the plant has low coolant
activity, the sensitivity is not as good as it would be if the plant had higher coolant activity.
Assuming normal corrosion product activity and no failed fuel, a RCS leak of 1 gpm (3.8
L/min) can be detected within about 80 minutes of its occurrence by the gas activity
monitors. With some failed fuel, this leakage can be detected more quickly.
• Containment Sump Level and Flow Rate
Leakage is monitored by a level detector in the containment sump. Changes in the sump
water level may be indicative of leakage from any system inside the reactor building, such as
the RCS, service water system, component cooling system, steam lines, or feedwater lines, or
of condensation of humidity within the containment atmosphere. A typical sump capacity is
15 gallons per inch (2.3 L per mm) of height with graduations on the level detector indicating
1/2 inch (13 mm) of sump height. Accordingly, a typical sump level detector can detect
changes of about 7.5 gallons (28 L) of leakage into the sump. At this rate, a 1-gpm leak (3.8-
L/min) would be detectable in less than 10 minutes, depending upon the time required for the
leakage to reach the sump.

• Containment Air Cooler Condensate Flow Monitors

Containment air cooler fans typically circulate the entire containment volume through the
coolers every five minutes. Some of the leakage within the containment ultimately condenses
on the cooler coils because the coils are the coldest location in most plants. The containment
air coolers are typically equipped with condensate monitoring systems that consist of
condensate drain collection systems, vertical standpipes, valves, and standpipe differential
pressure transmitters. In this manner, the containment air cooler condensate system provides
a relatively sensitive indication of leakage into the containment environment. The sensitivity
of this system varies depending upon plant operating conditions but is reported by
NUREG/CR-6582 to be capable of detecting a 1 gpm (3.8 L/min) leak within an hour.

6.4.2 Other On-Line Leakage Detection Methods

In addition to the four main leakage detection methods addressed in Regulatory Guide 1.45, there
are a number of other leakage detection methods. Some of these are:
• Containment Humidity Monitors
Containment humidity monitors can be used to detect leaks. Humidity monitors can also be
used in conjunction with the air particulate, iodine, and gaseous monitors to narrow the
potential source of leakage. If the humidity increases without a corresponding increase in
airborne activity level, the leakage is probably from a nonradioactive system, except during
periods of time when the reactor coolant activity is low.

6-4
 EPRI Licensed Material

Detecting Leakage

• Makeup Liquid Inventory

During steady state operation, the amount of makeup water that must be added to the system
is a measure of leakage from the system. NUREG/CR-6582 reports that due to large RCS
volumes and instrument inaccuracies, this approach is not as sensitive as others. It is reported
that a 1 gpm (3.8 L/min) leak is detectable in about six hours.
• Containment Air Cooler Thermal Performance
Boric acid is soluble in steam and can be transported within the containment by the
containment air cooler fans. Accordingly, boric acid that is carried over with the steam from
leaks will tend to deposit out on the containment air cooler coils and degrade the coil thermal
performance. Leakage from reactor vessel head instrument nozzle Conoseals at Calvert Cliffs
was discovered in this manner, which proved to be more sensitive than any of the other on-
line systems. (Note: Consideration could also be given to monitoring the boric acid
concentration in the containment air cooler condensate.)
• Personal Visual Inspection During System Walkdowns
Many leaks in containment and most leaks outside containment are found by personal visual
inspection during system walkdowns. This approach cannot be relied upon for rapid
detection of leaks inside containment because walkdowns are infrequent during plant
operation. Outside containment, most systems are walked down on a frequent basis so that
this becomes the primary method of leak detection for these systems.

Small leaks below the sensitivity of other detection methods can be detected by looking for
boric acid crystals accumulating within containment. Accordingly, major areas of potential
concern that are accessible and observable should be checked during any containment entries
between refueling outages. These visual inspections should include looking for evidence of
boric acid deposits and bulging of insulation.
• Other Potential Indicators of Primary Coolant Leakage
Several other approaches are used to detect and pinpoint leaks. However, many of these
methods are focused on leakage from specific higher risk/importance components rather than
on general leakage detection. These methods include:
– Tape moisture sensors for detecting leaks at specific locations such as locations
susceptible to PWSCC
– Temperature sensors to detect increased flow in lines resulting from leakage
– Nitrogen-13 monitors to detect small amounts of leakage, such as from RPV head CRDM
nozzle cracks
– Acoustic monitors to detect the sound produced by a leak
– Local humidity monitors to provide increased sensitivity to leaks at a specific location

6.4.3 Leak Detection During Plant Outages

There are two main methods of detecting leaks during outages:

• Boric Acid Walkdowns
Many locations that are not accessible during plant operation are accessible during plant
outages. The system walkdowns during outages should include careful inspection for

6-5
 EPRI Licensed Material

Detecting Leakage

evidence of boric acid deposits. The sensitivity of visual inspections for small boric acid
deposits is best illustrated by the fact that small leaks from through-wall PWSCC cracks in
many Alloy 600 nozzles and pressurizer heater sleeves have been located by boric acid
crystal deposits before being detected by other methods. In some cases, leaks have been
discovered by a fine ring of boric acid crystals deposited around the nozzles. In cases
involving very small leak rates, insulation must be removed to detect boric acid deposits left
by the leak.

All utilities have developed programs for inspecting for boric acid corrosion of carbon steel
reactor pressure boundary components as required by NRC Generic Letter 88-05. The NRC
has reviewed many utility programs and has conducted audits of several licensees. The
results of several reviews and audits are reported in NUREG-5576 [10]. The conclusion of
this work is that all of the programs submitted to the NRC have met the intent of Generic
Letter 88-05.

As part of the industry response to NRC Generic Letter 88-05 [4], EPRI has sponsored work
to develop recommendations for boric acid inspection programs. The recommendations are
contained in EPRI report NP-5985, Boric Acid Corrosion of Carbon and Low-Alloy Steel
Pressure-Boundary Components in PWRs [1]. Utilities can refer to this EPRI document for
further information on programs to identify boric acid leakage during outages.
• System Walkdowns During Plant Startups and Shutdowns
Inspections and walkdowns should be conducted during system pressurization, hydrostatic
testing, plant startup, and during any plant shutdowns. Particular attention should be paid to
pressure boundary connections that were opened or repaired during the outage. Significant
leakage from important components should be eliminated prior to power operation.

In addition to the visual inspections after shutdown, system walkdowns should be conducted
just before shutdown with the plant hot but at zero power. These inspections should focus on
identifying any locations with leaking steam.

6.5 Correlation Between Leakage and Boric Acid Deposits

Visual inspections are one of the most effective ways of finding small boric acid leaks that may
escape detection by other means. When boric acid crystal deposits are found, it is often
informative to estimate the amount and rate of leakage that produced the observed deposits.
Figure 6-3 gives the amount of leakage required to produce several different volumes and
weights of boric acid crystal deposits.

6.6 Summary of Leak Detection Methods and Requirements

As described above, detection of boric acid leakage should involve a multifaceted approach
including;
• On-line leakage monitoring
• Visual inspections of accessible and observable components during any containment entries

6-6
 EPRI Licensed Material

Detecting Leakage

• Visual inspections prior to plant shutdown at the start of each outage

• Visual inspections during plant outages
• Visual inspections during plant startup with attention directed toward any components that
were repaired or that had experienced leaks

Figure 6-1
Methods Used to Detect Leaks Reported in LERs (1985-1996)

6-7
 EPRI Licensed Material

Detecting Leakage

Figure 6-2
Sensitivities of Typical Leakage Detection Systems

6-8
 EPRI Licensed Material

Detecting Leakage

Figure 6-3
Correlation Between Boric Acid Leakage and Volume/Weight of Boric Acid Crystal
Deposits

6-9
 EPRI Licensed Material

7
PREVENTING LEAKAGE AND DEGRADATION

This section provides a summary of methods that can be used to minimize the potential for
borated water leaks and to minimize the potential for corrosion if leakage occurs. The final part
of this section describes several methods that can be used to stop leakage from components
during plant operation.

7.1 Preventing Leaks

The first level of protection against boric acid corrosion should be to prevent leaks from
occurring in the first place. If there is no leakage, there will be no boric acid corrosion.

A major effect of the first edition of the Boric Acid Corrosion Guidebook was the recognition of
the importance of eliminating leaks at plants. This increased industry sensitivity led to EPRI’s
creation of its onsite Plant Leak Reduction Program activity to assist plant management and staff
in better understanding opportunities to improve performance and reduce leakage. Conducting
these onsite activities subsequently served to highlight the lack of authoritative guidance on the
best processes, programs, procedures, and technologies to reduce leakage. This, in turn, led to the
creation of the EPRI Fluid Sealing Technology Program (FSTP) to provide this information.

The following is a description of the FSTP and the technology that is currently available to
reduce leaks for specific types of joints. However, all of the advances in leak reduction
technology over the past several years cannot be covered completely in this guidebook. Utilities
are encouraged to obtain the full set of EPRI FSTP documents for more complete coverage of
this important subject.

7.1.1 EPRI Fluid Sealing Technology Program

Utilities determined that the available guidance relating to leakage reduction programs and
leakage reduction technology was deficient in several important aspects. In response to their
requests for improved information and guidance, EPRI developed the FSTP to address these
deficiencies.

The FSTP work has focussed on five main activities:

• Research into improved leakage reduction technology for important types of joints
encountered in nuclear plants
• Semiannual working group meetings where the results of the research are presented to
participating utility members

7-1
 EPRI Licensed Material

Preventing Leakage and Degradation

• A series of technical reports describing the results of the research with recommended
practices for each type of joint
• Leakage reduction training courses presented at nuclear plants
• An assessment of the cost impact of leakage at individual plants with estimates for the
nuclear power industry as a whole

The following are brief descriptions of these activities.

7.1.1.1 Research into Leakage Reduction Technology

A key aspect of the FSTP has been research into the best technology to prevent leaks from the
types of joints typically encountered in nuclear plants. In some cases, previous work has been
adequate and the means to reduce leakage well defined. In other cases, previous work has not
been adequate to define practical solutions. Where previous work has not been adequate, analysis
and laboratory work have been conducted by the FSTP to obtain the missing information and
develop practical recommendations. Research on specific types of joints is described in Sections
7.1.3 through 7.1.8.

7.1.1.2 Working Group Meetings

EPRI sponsors a Fluid Sealing Technology Working Group where utilities meet on a regular
basis to review the results of the research, discuss plant-specific leakage problems, obtain
complementary information on current areas of research from sealing technology vendors, and
establish priorities for further leakage reduction activities. Reports of these meetings are
available from EPRI.

7.1.1.3 Fluid Sealing Technology Program Products

The main products of the FSTP are a series of technical reports covering major areas of fluid
sealing technology. The complete series of reports, including their status as of December 2000, is
as follows:
• Establishing an Effective Leakage Reduction Program Complete
• Assembling Bolted Connections Using Spiral-Wound Gaskets Complete
• Bolt Preload for Raised-Face Flanges with Spiral-Wound Gaskets Complete
• Assembling Bolted Connections Using Sheet Gaskets Complete
• Lube Oil System Leakage Mitigation Complete
• Assembling Threaded Connections Complete
• Valve Packing Performance Improvement In Progress
• Mechanical Seal Performance Improvement In Progress

7-2
 EPRI Licensed Material

Preventing Leakage and Degradation

When these documents are completed, it is expected that they will provide a comprehensive
guide to the best available leakage reduction practices for nuclear plants. Highlights of the
reports are presented in Sections 7.1.2 through 7.1.8.

In addition, the FSTP has also produced the EPRI LeakTrac software program to assist utilities
in establishing effective tracking and trending for plant fluid leakage events.

7.1.1.4 Plant Leakage Reduction Training Workshops

A key activity of the FSTP has been conducting leakage reduction training courses at plants. The
training sessions have included:
• A management session to review the plant’s current leakage status and the opportunities for
improvement
• Review of plant procedures relating to leakage reduction
• Technical training sessions with engineers, work planners, craft, and QA personnel to
describe the latest technology
• Preparation of an exit report summarizing the work accomplished and summarizing the
opportunities for improvement

A key feature of the training has been the development of a series of training fixtures that clearly
demonstrate the important factors for each of the main types of joints. Demonstrating the
principles presented in a classroom environment is an important aspect of obtaining buy-in from
technical and craft personnel. Four devices used in the training are:
• Bolted Joint Performance Demonstration Unit (BPDU)
• Gasket Performance Demonstration Unit (GPDU)
• Thread Sealant Performance Demonstration Unit (TPDU)
• Valve Packing Performance Demonstration Unit (VPDU)

As of December 2000, these EPRI Plant Leak Reduction Program activities have been conducted
at more than 35 plants in the United States and overseas.

7.1.1.5 Cost Impact Assessment for Plant External Piping System Leakage

The final product of the FSTP was the assessment of leakage impact on a plant-specific and
industry basis. This effort was undertaken to increase the awareness of plant management as to
the importance and economic impact of leakage. This effort is addressed in EPRI TR-114761,
Establishing an Effective Leakage Management Program [37]. The conclusion from a
preliminary study involving 7 plant sites, 13 reactor units, of both PWR and BWR designs,
belonging to three different utilities, is that the average aggregate cost impact of external piping
system leakage appears to range from $1.5–2.5 million per unit per year, with an average cost of
$1.6 million. As indicated in TR-114761, significant savings are possible in most cases.

7-3
 EPRI Licensed Material

Preventing Leakage and Degradation

7.1.2 Leakage Reduction Programs

In most cases, the technical means for reducing leakage are not difficult. However, developing
an effective program for reducing leakage from the many possible sources requires concerted
effort by both plant management and staff. The following sections outline the main components
of an effective leakage management program, as presented in EPRI TR-114761 and illustrated in
Figure 7-1. Readers are encouraged to obtain this document because it contains significant
supporting detail that is not included in this brief overview.

7.1.2.1 Prerequisites

There are four main prerequisites for an effective leakage management program: management
involvement, assigning a program coordinator, establishing realistic program goals, and
measuring progress on a regular basis.

Management Support

As suggested by Figure 7-1, plant management can initiate a leakage management program,
support a leakage management program created by plant engineering/maintenance staff, or work
with plant engineering/maintenance staff to jointly create a leakage management program. All
three approaches can be successful. However, it is unlikely that a successful program can be
developed and implemented without management involvement and support.

Management support should include the following activities:

• Designate a qualified leakage management program coordinator.
• Work with the program coordinator to establish realistic program goals.
• Provide resources.
• Ensure necessary interdepartmental communication and cooperation.
• Monitor program effectiveness and make changes as necessary.
• Demand excellence.

Program Coordinator

The program coordinator is often the key to success in developing and implementing an effective
leakage management program. Ideally, the program coordinator should have several
characteristics:
• The authority, backing, and confidence of management
• Recognition, respect, and acceptance by the engineering and maintenance staff
• The technical ability to establish and support the leakage management program, including the
ability to determine the actual root cause of chronic leaking components and to factor the
knowledge back into the program to avoid similar leaks in the future

7-4
 EPRI Licensed Material

Preventing Leakage and Degradation

• The ability to train others

• Persistence

Program Goals

Plant management and the program coordinator must develop a site definition for what
constitutes a “leak.” In general, a leak may be construed as any instance where evidence of a
process fluid (gas, liquid, or steam) is detected outside its normal process boundary. This
definition includes both active leaks and inactive leaks, where evidence of the inactive leak can
be detected. An example of an inactive leak might be the appearance of dry boric acid crystals
around a valve stem or flange seal. Another example of an inactive leak might be one that
required online injection with a leak sealant compound in order to stop the leak.

Plant management and the program coordinator must also set realistic goals for the leakage
management program. An across-the-board requirement for “zero leakage” simply is not realistic
because most components were never designed to be “leak-free,” only “leak-tight,” and then
only in varying degrees. As discussed in TR-114761, EPRI proposes that plants classify leaks by
both severity and criticality. Table 7-1 indicates the proposed severity ratings for several types of
leaks, and Table 7-2 indicates the proposed response as a function of application criticality. The
proposed levels of criticality are:
• Category I Joints
Category I joints should include all systems critical to personnel safety and critical to nuclear
safety, where leakage can spread contamination and where the consequences of leakage are
expensive. These include most joints on the primary system, safety-related systems, and
systems containing radwaste. In these areas, zero-leakage would generally be a requirement.
• Category II Joints
Category II joints should include those applications where leakage is highly undesirable but
some temporary leakage may be acceptable. These include most joints on important
secondary side systems such as feedwater systems, condensate systems, turbines, generators,
T/G support systems, condensers, etc. In these applications, zero-leakage would generally be
an important goal, achieved systematically through the use of solid procedures, appropriate
materials, and effective training.
• Category III Joints
Category III joints should include the balance of the joints in the plant where leakage is
undesirable, but small amounts of leakage can be accepted. In these areas, some minor
leakage may be tolerated if accompanied by ongoing monitoring for change and efforts to
make improvements.

By relating the response to the severity of the leak and the criticality of the application, it should
be possible to avoid unwarranted responses, such as zero leakage, to small leaks in lower
criticality applications. However, the shift to a leakage acceptance criteria approach to fluid leak
management should not be construed as an opportunity to justify increases in the total number of
leaks. Rather, it should be a rational starting point for a systematic reduction of the number and
severity of leaks over time through a managed approach to fluid leakage.

7-5
 EPRI Licensed Material

Preventing Leakage and Degradation

Based on the above criteria, the following are potential realistic leakage reduction program goals:
• Implement a formal leakage management program.
• Assign a program coordinator.
• Implement a formal system of leak tracking and monitoring by joint type, leak severity, joint
criticality, and planned disposition.
• Reduce leaks and online leak repairs to the level of the best plants in a progressive manner.
• Reduce leakage-related work orders each year.
• Have no leak-related outage extensions, forced outages or power reductions.
• Update joint assembly procedures to reflect the state of the art.
• Update training materials and programs to reflect the state of the art.
• Reduce the annual aggregate cost impact of external piping system leakage by a target
amount each year.
• Reduce the amount of management time devoted to leakage-related issues.

The important point in the above list is that it is possible to review each item and establish
quantifiable program goals over a reasonable period of time. An annual cost impact assessment
may serve as a useful benchmark to measure overall program success. Goals can be reviewed
periodically and the program adjusted as necessary to meet plant-specific needs. Goals that are
not achieved may provide useful guidance regarding where future efforts should be directed.

7.1.2.2 Technical Support Programs

A major key to cost effective leakage reduction is to start with state-of-the-art procedures and
materials that are capable of developing high-integrity joints and then train craft personnel to
follow the procedures and identify adverse conditions.

A key component of the EPRI FSTP has been to study each of the areas of sealing technology to
identify the causes of leaks and the materials and procedures that are best capable of minimizing
the potential for leaks. Technical details are provided in the FSTP reports listed in Section 7.1.1.

Procedures

All procedures related to joint assembly and leakage reduction should be reviewed and updated
to incorporate the best available technology to achieve low leakage at minimal overall cost.
Based on a review of procedures at many plants and discussions with participants in the EPRI
FSTP, there is typically significant room for improvement in these procedures.

Taking bolted joints with spiral-wound gaskets as an example, very few of the procedures being
used in the industry incorporate all of the key factors necessary for high-integrity joints. Some of
these factors are:

7-6
 EPRI Licensed Material

Preventing Leakage and Degradation

• Inspection for flange flatness on leaking joints

• Use of gaskets with inner and outer rings
• Use of hardened steel washers
• Lubrication on bolt threads and the nut-to-flange interface
• Sufficiently high preload torques
• Measurement of flange parallelism for misaligned joints
• Leveling passes at final torque until there is no nut rotation
• Live load (Belleville washers) for problem joints

The following guidance is suggested relative to procedures:

• Procedures should be reviewed and updated relative to the EPRI FSTP Plant Leakage
Reduction Series guidelines.
• Procedures should be clear and concise: As part of the bolted joint leakage reduction training
courses, EPRI has reviewed procedures at over 35 plants for assembling bolted joints with
spiral-wound gaskets. A main conclusion is that few of these procedures are clear and
concise. Several problems typically encountered include:
– Procedures that are overly long and complex
– Procedures that contain conflicting requirements
– Procedures that represent years of information being added without any effort to
ensure that it is integrated into a consistent and easy-to-understand package
• All procedures should be reviewed to ensure that they contain only relevant information and
that this information is presented in the clearest possible manner. One useful approach is to
have short procedures with more complete discussion and application-specific details
included as appendices or attachments.

Materials

There has been considerable improvement in sealing technology over the past 10 years, but many
plants are not taking advantage of these improvements, examples of which follow:
• Adding inner rings to spiral-wound gaskets for pipe flange joints improves the leak tightness
and increases the resistance to buckling, yet most plants do not use spiral-wound gaskets with
inner rings.
• Spiral-wound gaskets with flexible graphite filler have better leak-tightness, less stress
relaxation, and more springback than spiral-wound gaskets with mica graphite filler, yet
many plants continue to use spiral-wound gaskets with mica graphite filler.
• Some types of sheet gaskets are much better than others for flanges that are not flat, are
misaligned, or have a poor surface finish.

7-7
 EPRI Licensed Material

Preventing Leakage and Degradation

• Using hardened steel washers between the nut and flange surface results in more
reproducible bolt preloads and minimizes flange damage, yet most plants do not use
hardened steel washers.
• Some thread sealants are much better than others that are currently in widespread use.
• There are many leaks from inspection and drain caps that are opened and closed by operators
without proper thread sealing, yet few plants are using O-ring-type pipe caps that allow easy
access by operators and eliminate subsequent leaks.
• Some packing types have better leak tightness and less creep relaxation than others .
• Live loading valve packing can significantly reduce packing leaks that result from
consolidation, yet many plants do not use live loading extensively.

Many problems in these situations can be solved by using improved materials. In some cases, the
material cost will be a little higher, but the cost of tracking, evaluating, and repairing leaks
typically far exceeds any cost savings from using less expensive materials.

The best available materials are discussed in the EPRI FSTP Plant Leakage Reduction Series
documents listed in Section 7.1.1.

In addition to specifying the best available materials, several other materials related controls
follow:
• Purchasing documents must accurately specify the desired parts and quality.
• Receipt inspection must confirm that the materials received are as specified.
• Materials must be stored in a manner that does not result in their degradation.
• Quality materials must be available in the warehouse or obtainable in a timely manner.

Training

The third essential technical support program is training. A key part of the EPRI FSTP is to
develop quality training materials including reference manuals, training viewgraphs, and training
aids that demonstrate the principles in a convincing manner. EPRI has assisted in conducting
pilot training at many plants, and all plants are encouraged to incorporate the EPRI materials and
training technologies into their ongoing training programs.

These training programs should emphasize:

• Collaboration with the fluid leakage management program coordinator to address fluid leak
management issues
• The “why” as well as the “how” for each area of technology
• Ensuring that the participants are able to perform work according to the procedures and
recognize when something is wrong so that help can be requested

7-8
 EPRI Licensed Material

Preventing Leakage and Degradation

7.1.2.3 Implementation

After procedures have been prepared, materials and sealing products procured, and personnel
trained, the field implementation phase can begin. Several parts of the organization participate in
this important aspect of the work:
• Work planners play a key role in ensuring that craft personnel are provided with the proper
instructions for the work to be performed.
• Craft personnel should perform work according to the work order requirements but should
look for conditions that might indicate a past or future problem.
• The program coordinator should make a point of being present during the disassembly and
reassembly of critical or problem joints.
• QA should verify that hold point inspections are performed as specified.
• Engineers should be available, as necessary, to assist in the evaluation of problems and
should be present during disassembly and reassembly for critical or problem joints to obtain
first-hand information on how to improve on the work in the future.

7.1.2.4 Follow-Up

Hopefully, improved procedures, materials, and better-trained personnel will have a significant
beneficial effect in reducing the number of leaks; however, experience has shown that some
leaks will still occur despite these improvements. It is important to evaluate the remaining leaks,
determine which of the leaks need to be repaired, determine the timing for these repairs, and
suggest changes to the procedures to avoid these leaks in the future.

Tracking

Leaks on equipment inside containment are typically identified by periodic walkdowns and by
the on-line methods described in Section 6.

As has been discussed previously, it is recommended that plants should have a formal method for
tracking leaks that is based on leakage severity and the criticality of the application. This
information can be used to select a rational repair strategy for each leak, to identify trends in
performance, and to identify areas where the leakage reduction program can be improved. The
FSTP has developed the LeakTrac database that utilities can use to develop a leak tracking
system or to improve an existing system.

Evaluation

A key feature of the proposed EPRI approach to leakage management is that each leak should be
evaluated and the repair decisions based on acceptance criteria that takes into account leakage
severity and joint criticality. This approach will ensure that timely corrective action is taken for
significant leaks in critical applications but will avoid unnecessary effort for less important leaks.

7-9
 EPRI Licensed Material

Preventing Leakage and Degradation

Tagging

Each leaking joint should be tagged, indicating the date the leak was discovered, by whom, the
leakage severity, the joint criticality, and the planned corrective action, according to the site
corrective action plan.

Repair

Repairs can range from immediate repairs for severe leaks in the most critical applications to
allowing the leak to continue until the next scheduled system maintenance for less severe leaks
in applications of lower criticality. The extent and timing of repairs should be a function of the
severity of the leak and the criticality of the application.

7.1.2.5 Feedback

No leakage management program will be perfect from the start. All programs will require
feedback and fine tuning to achieve the greatest value (lowest cost when the costs of lost
production and other consequential costs such as increased regulatory scrutiny are considered).
There are two aspects to the feedback process: first, assessing performance periodically and
second, making warranted program improvements.

Performance Assessment

The first step in assessing and improving the program is to evaluate the overall program
performance relative to the goals that were established at the program’s inception. This
evaluation should typically be performed by the program coordinator, and the results reviewed
with plant management. The performance of the program relative to each of the goals should be
evaluated separately, trying to be as quantitative as possible.

The management review should focus on the overall program performance relative to the initial
goals, a review of the initial goals and whether they should be modified, and possible
improvements to better meet individual goals.

Program Improvement

The final step in the process is to modify the program as agreed upon with management. This can
involve changing the goals, improving the technical programs (procedures, materials, training),
and possibly improving the leakage management program.

7.1.3 Leak Reduction from Gasketed Joints

There have been major advancements in the technology to reduce leakage from gasketed joints
since the first edition of the Boric Acid Corrosion Guidebook in 1995. This has included
improved knowledge regarding the best types of gaskets for each application, the required bolt

7-10
 EPRI Licensed Material

Preventing Leakage and Degradation

preload stress, and the best assembly practices. These subjects are addressed in the following
FSTP reports:
• Assembling Bolted Connections Using Spiral-Wound Gaskets, EPRI report TR-111472 [38],
covers spiral wound gaskets and bolting techniques that will produce the required bolt
preload stress.
• Bolt Preload for Raised-Face Flanges with Spiral-Wound Gaskets, EPRI report 1000066
[39], covers spiral-wound gaskets in ANSI B16.5 raised-face flange applications.
• Assembling Bolted Connections Using Sheet Gaskets, EPRI report 1000922 [40] covers
flanged joints with sheet gaskets.

Users are referred to these reports for detailed coverage of gasketed joints, the required bolt
preload stresses for gasketed joints, and the best methods for assembling the joints. The
following subsections present a brief overview of each of these areas.

7.1.3.1 Causes of Spiral-Wound Gasket Leaks

Spiral-wound gaskets are widely used in high-pressure and high-temperature applications in

PWR plant primary systems. These are typically among the most important bolted joints in the
plant. EPRI TR-111472 [38] describes the causes of leaks from these types of joints and the best
methods for reducing the leakage. The following is a brief summary of key points from this
report:
• The main cause of leakage from bolted joints with spiral-wound gaskets is believed to be loss
of bolt preload force due to gasket creep. Specifically, if the gasket is not fully compressed to
the metal compression stop ring or until there is metal-to-metal contact between the flanges
for recessed or tongue and groove gaskets, the bolt preload force will decrease as the gasket
creeps inward over time. The potential for gasket creep is accelerated if 1) the flange surfaces
are smooth, 2) the joint is subjected to thermal transients, or 3) the gaskets are coated with a
lubricant as was often done in the past to aid in assembly and to prevent asbestos gaskets
from sticking to the flange face.
• The two main precautions to minimize the effects of gasket creep are: 1) to use spiral-wound
gaskets with inner and outer rings where practical and 2) to preload the bolts to the point
where the flanges come into contact with the gasket outer compression stop ring or the
flanges come into metal-to-metal contact for joints with recessed or tongue and groove
gaskets. With metal-to-metal contact, the bolt load will not be lost even if there is some
gasket creep.
• As shown in Figure 7-2, spiral-wound gaskets are designed to operate between flange faces
that are machined with a 125–250 µinch (3.2–6.4 µm) surface finish with concentric rings or
a spiral feed. This moderately rough surface acts to create a good seal (labyrinth) and also to
minimize the potential for the gasket creeping inward, thereby relieving the bolt preload.
Many leaks have been reported from flanged joints with orifice plates. These plates typically
have a smooth surface, creating the potential for accelerated gasket creep and leakage. A
good solution for these cases is to use spiral-wound gaskets with inner and outer rings where
practical.

7-11
 EPRI Licensed Material

Preventing Leakage and Degradation

• Test work by EDF has shown that spiral-wound gaskets with flexible graphite filler have
lower leak rates than gaskets with mica-graphite or asbestos filler for the same bolt preload
force. These gaskets also have greater springback from full compression when the bolt load
is released.
• It has been known for years that some spiral-wound gaskets buckle on the inner edge as they
are loaded. Further, recent tests for EPRI by EDF have shown that this buckling can increase
the leakage relative to gaskets that do not buckle. Figure 7-3 shows typical inner edge
buckling in a spiral-wound gasket. The best solution to prevent inner edge buckling is to use
gaskets with inner and outer rings where practical.
• In some cases, primarily involving four bolt flanges, there is the potential for the gasket to
become offset within the joint so that the raised face of the flange comes into contact with the
live portion (spiral windings) of the gasket rather than the metal compression stop ring. If the
bolt preload stress is high, the gasket can be crushed, leading to increased potential for
leakage. Again, the best solution is to use gaskets with inner and outer rings where practical.
• Other causes of leaks from bolted joints with spiral-wound gaskets include misalignment,
improper dimensions in tongue and groove joints, damaged gaskets, damaged flange
surfaces, use of the wrong size or pressure class gasket, etc. These causes are all addressed in
EPRI TR-111472.
• Some suppliers offer gaskets with special features designed to improve sealability. These
features include: 1) adding inner wraps of metal to the inner edge of the gasket to reduce the
potential for buckling, 2) providing relief on the outer metal compression stop ring to allow
some outward expansion of the gasket, 3) designing the gasket to seat at a lower stress, 4)
and designing the gaskets with inner wraps of polytetrafluoroethylene (PTFE) to provide a
better seal and outer wraps of flexible graphite to ensure good fire resistance.

In summary, there are several main causes of leaks from spiral-wound gaskets and practical
solutions are available for each cause.

7.1.3.2 Causes of Sheet Gasket Leaks

Sheet gaskets are also widely used in PWR plants, typically for the lower temperature, pressure,
and size applications. EPRI report 1000922 [40] describes the causes of leaks from joints with
sheet gaskets and the best methods for reducing this leakage. The following is a brief summary
of the key points from this report:
• A wide variety of sheet gaskets are offered by vendors. These include rubber, compressed
synthetic fiber, PTFE, filled PTFE, expanded PTFE, and flexible graphite. Many of these
gaskets are available with metal reinforcement sandwiched between layers of the gasket
material.
• Many sheet gasket failures result from chemical incompatibility with the fluid being sealed.
PTFE and flexible graphite are compatible with most fluids encountered in nuclear plants,
although PTFE gaskets are often prohibited in applications where halogens must be avoided.
Rubber and compressed synthetic fiber gaskets must be selected carefully to ensure
compatibility. For rubber gaskets, it is obvious that the choice of material is important;
however, this is an equally important issue for compressed synthetic fiber gaskets that are

7-12
 EPRI Licensed Material

Preventing Leakage and Degradation

held together by a rubber binder. In fact, the seemingly large number of compressed synthetic
fiber gasket product codes results from the fact that there are several different types of fibers
(vegetable fiber, Aramid, carbon, graphite, etc.), each of which can be combined with several
different types of rubber binders. The user must select the proper fiber and binder
combination for a successful application. Gasket vendors and catalogs can provide assistance
in this selection.
• Sheet gaskets are typically more susceptible to high temperatures than spiral-wound gaskets.
With a few exceptions, rubber gaskets are limited to 200–250°F (93–121°C). Virgin PTFE
gaskets exhibit high creep at temperatures above 100°F (38°C). Filled and expanded PTFE
gaskets can operate to temperatures of 350°F (177°C) and up to 500°F (260°C) in some
cases. While some compressed synthetic fiber gaskets are reported to have service
temperatures over 1000°F (538°C), they should realistically be limited to the same maximum
temperature as the rubber binder for important applications because the material hardens
(vulcanizes) at service temperatures greater than about 250°F (121°C). This is a particular
concern for flanges that are subject to thermal transients and require flexibility to maintain a
tight seal in the presence of thermally induced motions. Tests performed by EPRI showed
that many compressed fiber gaskets lost their resiliency after only two days at 400 °F
(204°C).
• Many sheet gaskets have failed due to gasket blowout. This can occur for many reasons,
including low compressive stress at some locations around the flange due to flange out-of-
flatness, a smooth flange surface finish, lubricant on the flange surface, flange misalignment,
too low a bolt preload stress, etc. To minimize the risk of blowout, a good guideline is to
limit unreinforced sheet gaskets to pressure class 150 and reinforced sheet gaskets to class
300. Above these pressure classes and for larger diameters within these classes, spiral-wound
gaskets are generally preferable.
• Most sheet gaskets, with the exception of rubber, can accommodate gasket seating stresses of
15,000 psi (103 MPa). Rubber gaskets should be stressed only to about 1000 psi (6.9 MPa).
For this reason, rubber gaskets should generally be limited to flat face flange applications
with full face contact. These gaskets can be easily damaged in raised-face flanges or in flat
face flanges with the gasket outside diameter inside the bolt circle.
• Some sheet gasket materials, especially PTFE, have poor radiation resistance.
• Gaskets must meet special plant chemistry requirements such as limitations on halogens,
sulfur, and low melting point metals.
• Test work performed by the EPRI FSTP has shown significant differences in the ability of
sheet gaskets to seal under conditions where the flange is not perfectly flat or contains deep
scratches across the gasket seating surface. In particular, gaskets with low compressibility
(most compressed synthetic fiber and some filled PTFE gaskets) tend to leak under less than
ideal conditions. Gaskets with greater compressibility such as rubber, expanded PTFE, and
flexible graphite perform much better under these conditions. However, it was noted in the
tests that expanded PTFE must be compressed to about 3000 psi (21 MPa) to close up
microvoids inherent in this material. If the voids are not fully closed up, these gaskets will
tend to seep, especially in gas service.

7-13
 EPRI Licensed Material

Preventing Leakage and Degradation

• While the conventional wisdom is to use the thinnest sheet gasket practical, the test results
show that the gasket must be thick enough to seal the maximum gaps that will exist between
the flanges. This may necessitate a thicker gasket, especially for gaskets with low
compressibility. The effect of gasket thickness on sealability is illustrated in Figure 7-4.
• The test work also showed that some sheet gaskets are easier to handle than others . For
example, metal-reinforced sheet gaskets are difficult to cut in the field, and many of the
flexible graphite gaskets can be easily damaged during installation and tend to stick to the
flange surfaces.

In summary, sheet gaskets have performed well in some applications, but they are not a cure-all
for flange leakage problems. They must be selected carefully to ensure that the selected gasket
will meet all of the functional requirements for the application.

7.1.3.3 Required Bolt Preload for Gasketed Joints

The general practice at many plants is to preload the flange bolts to the lowest possible stress so
that the gaskets, flanges, and bolts are not damaged. However, research by the EPRI FSTP has
shown that a better approach is to generally apply the highest preload that the gaskets and flanges
can accommodate. The subject of the required preload is addressed in EPRI reports TR-111472,
1000066, and 1000922 [38, 39, 40]. The following is a brief summary of the recommendations in
these documents:
• For flanged joints in components or on equipment skids, the bolt preload torque should
generally follow the vendor recommendations unless a new torque is being established due to
leakage problems or when a new gasket material is being used. This is especially true for
Class 1 components where changes to the preload stress must be analyzed per ASME Code
requirements.
• If no problems are being experienced with leakage, the best policy is generally to remain
with the same gaskets and bolt preload stresses that have been used successfully in the past.
• For ANSI B16.5 flanges with A193 Grade B7 bolting material or better, recommended bolt
preload stresses and torques are provided in EPRI reports 1000066 and 1000922 for spiral-
wound and sheet gaskets respectively.
• For all other flanged joints, it is suggested that the bolt preload stress be limited to 40–50%
of the bolt material yield strength as suggested in EPRI report TR-104213, Bolted Joint
Maintenance and Applications Guide [41].

7.1.3.4 Achieving Required Bolt Preload

While designers and engineers must specify the proper bolt preload stress and torque, it is
equally important that the mechanics not give away the preload through poor assembly practices.
This important subject is addressed in EPRI report TR-111472 [38]. The following is a brief
summary of the findings and recommendations in this report:

7-14
 EPRI Licensed Material

Preventing Leakage and Degradation

• There are four main causes of low bolt preload stress at most plants: too low a specified
torque, poor lubrication, poor condition of the bolts and flange surfaces, and failure to use a
controlled tensioning method. The first method is under control of the engineers; the other
three causes are under the control of the craft performing the work.
• It has been a surprise to most engineers and craft personnel that have received the FSTP
leakage reduction training to find that about 60% of the torque applied to a bolt is required to
overcome friction between the nut and flange and more than 20% of the torque is consumed
in overcoming friction in the threads. Less than 20% of the applied torque actually goes into
loading the bolt.
• With more than 80% of the applied torque consumed by friction, it is important that good
high-pressure thread lubricants be used between the bolt and nut threads and between the nut
and flange surfaces. Failure to clean the surfaces and to use a good thread lubricant at both
locations can easily result in the loss of over half the desired preload.
• With the nut-to-flange interface being the most important factor, special care should be
exercised to ensure that this surface is flat, clean, and well lubricated. Yet visual inspections
in plants typically show the flange surface to be galled, rusted, painted, and in otherwise poor
condition.

A simple means of remedying this situation is to install hardened flat washers between the
nut and flange. This will ensure that the nut bearing surface is smooth and flat. It will also
decrease the friction since it results in a harder bearing surface than the typical flange.
Belleville washers could be used at this location, but much of the benefit will be achieved by
a hard flat washer without the increased risk of stress corrosion cracking associated with the
highly stressed Belleville washers.
• In some cases, low bolt stresses have resulted from local yielding of the bolts that increases
the friction between the bolt and nut threads. This condition can be checked by running the
nut down the length of the bolt prior to assembly. If the bolt has yielded, the nut will stop at
the point where the pitch of the bolt and nut threads no longer match.
• It is recommended that most joints be assembled using a controlled torquing or tensioning
procedure. Some plants are still assembling secondary side flanged joints by mechanic’s
judgement or skill of the craft. Experiments performed with participants in the EPRI Plant
Leakage Reduction Program activities show that this approach results in an excessively large
range of scatter in final bolt preload force.

7.1.3.5 Gasketed Joint Assembly Guidelines

EPRI reports TR-111472 and 1000922 [38, 40] provide suggested assembly procedures for
gasketed joints. The following is a brief summary of the highlights from these reports:
• Visually inspect all parts during disassembly for damage or other conditions that could cause
leaks in the future.
• Prepare parts for assembly by confirming the proper gasket; cleaning the bolts, nuts, and
washers; cleaning the flange surface; confirming the required 125–250 µinch (3.2–6.4 µm)
flange surface finish; and inspecting the gasket for damage.

7-15
 EPRI Licensed Material

Preventing Leakage and Degradation

• Assemble the joint by lining up the flanges, lubricating the fasteners at threads and the nut-
to-flange interface, assembling the joint, and performing a visual inspection.
• Ensure that all bolts are finger tight; tighten the bolts to about 5% of the final torque to
ensure proper seating; and perform another visual inspection.
• Torque the bolts to the final specified torque in at least three passes using a cross pattern.
Add a fourth pass for critical joints, problem flanges, or misaligned flanges. If there is flange
misalignment, ensure that the flanges are parallel within 0.015 inch (3.8 mm) over the flange
diameter after three leveling passes at 50% of the final torque.
• Perform a leveling pass by torquing continuously around the flange. For all gaskets except
rubber, the leveling passes should be continued until there is no further nut rotation. If more
than three leveling passes are required for spiral-wound gaskets, this is an indication that the
target bolt preload is not high enough to ensure full gasket compression. The required
number of passes for sheet gaskets depends upon the compressibility of the gaskets. A large
number of leveling passes may be required. After three passes have been completed, the craft
performing the work will have to judge when the amount of additional nut rotation is
sufficiently low that the leveling passes can be stopped.

7.1.3.6 Root Cause Analysis for Problem Joints

If bolted joints continue to leak despite the above precautions, the joint should be evaluated and
the root cause determined. Possible additional causes could include flange out-of-flatness, high
piping system loads, etc. Solutions could include higher bolt preload on a case-by-case basis,
Belleville washers, confirming proper gasket compression by feeler gauges, confirming proper
bolt preload by ultrasonic length measurement, use of tensioning instead of torquing, or
premachining or replacing the flange.

7.1.4 Leak Reduction from Valve Packing

During the mid-1980s, EPRI sponsored a major effort to identify the root cause of leakage from
valve packing, to identify improvements that would reduce leakage from valve packing, and to
test the improved designs on prototype valves. The results of this work are summarized in EPRI
report NP-5697, Valve Stem Packing Improvements [42]. Knowledge gained from this program
has resulted in significant reductions in the number of leaks from important valves as is
documented in NUREG/CR-6582, Assessment of Pressurized Water Reactor Primary System
Leaks [36].
Despite the improved knowledge regarding the root cause of valve packing leaks and the
reduction in leaks from the most important valves, surveys by the EPRI FSTP program continue
to show that valve packing remains one of the main sources of leaks at nuclear plants. Perhaps
this disparity results from plants not applying the same level of technology and care to all valves
that is applied to the most important valves.
The FSTP working group selected valve packing as a high priority for developing improved
technology transfer and training materials. Preliminary findings of this work were presented to
the working group in August 2000, and the final report on this work is scheduled for publication
in 2001. The following is a brief overview of the highlights from the August 2000 working group
meeting.

7-16
 EPRI Licensed Material

Preventing Leakage and Degradation

7.1.4.1 How Valve Packing Works

Figure 7-5 shows how valve packing works to seal a valve stem.

1. Valve packing is inserted into a stuffing box in the valve bonnet.

2. An axial load is applied to packing by a gland that is in turn loaded by torquing gland nuts.

3. The axial load on the packing causes the packing to expand outward, creating radial seals
against both the stem and stuffing box.

7.1.4.2 Available Types of Valve Packing

Valve packing has evolved in four main steps over about the last 30 years as illustrated in Figure
7-6:

1. Until about the mid-1970s, valves were packed primarily with asbestos braid rings.

2. Starting in about the mid-1970s and continuing in use today, non-asbestos (typically PTFE,
carbon, or graphite yarn) braid rings with some type of blocking agent to prevent leakage
were used to pack valves.

3. Starting in about the mid-1980s and continuing in use today, die-formed graphite rings have
been added to the packing sets to provide the best possible seal. Because the die-formed rings
are fragile and susceptible to extrusion, they are backed up by non-asbestos braid rings.
Another feature introduced in the 1980s was the use of a carbon spacer to reduce the length
of the packing and, thereby, reduce the amount of consolidation that occurs in service.

4. Starting in the late 1980s, fully engineered valve packing sets have been increasingly used.
These sets include die-formed graphite seal rings (rectangular or wedge shaped) to provide a
seal, low friction anti-extrusion rings (composite or low friction filled braid ring), carbon
spacers, and Belleville washers to provide live loading.

Vendors offer a wide range of valve packing products that are based on the above principles. In
many cases, the products reflect combinations of features tailored to specific needs.

7.1.4.3 Causes of Valve Packing Leaks

There are many potential causes of valve packing leaks. It is important to determine the root
cause of leaks as part of the repair/replacement process. The following is a brief summary of the
main causes:
• The main cause of packing leakage is consolidation of the packing material in service.
Packing materials contain voids and must be compressed (consolidated) 20–50% to develop a
reliable seal. Initial consolidation takes place as the packing is tightened by the gland
follower. However, as shown in Figure 7-7, friction between the packing, stem, and stuffing
box results in greater compression forces and consolidation at the top of the packing than at
the bottom of the packing. As the valve stem is stroked in service, the compressive load in

7-17
 EPRI Licensed Material

Preventing Leakage and Degradation

the packing evens out, resulting in a decrease of load on the follower, a decrease in the
maximum radial stress that creates the seal, and an increased potential for leakage.

There are several ways to address the consolidation issue:

– First, reduce the length of packing by using carbon bushings or spacers. This will result
in more uniform and high compressive stress in the remaining packing ensuring that it is
better consolidated at the start.
– Second, reduce the length of braid-type materials that require a large amount of
consolidation relative to die-formed graphite or composite anti-extrusion rings.
– Third, use lower friction packing materials that will permit greater consolidation for the
same gland follower load.
– Fourth, consolidate the packing during assembly by repeatedly stroking the valve stem
and retightening the packing until there is no further rotation of the gland nuts.
– Fifth, apply live loading to the gland nuts using Belleville spring washers to retain the
desired compressive stress even if the packing materials consolidate during service.
• Some packing problems are materials related:
– PTFE has a high coefficient of thermal expansion that can cause high packing loads and
extrusion as the valve heats up.
– Unsupported PTFE exhibits high extrusion at temperatures above 100°F (38°C).
– PTFE is limited to temperatures of about 500°F (260°C).
– Graphite fibers in graphite yarn braid material tend to fracture at gland stresses higher
than 4000–6000 psi (28–41 MPa).
– Graphite bushings and spacers are hard and can crack if unevenly supported at the bottom
of the stuffing box.
• Excessively tight or loose clearances between the stem, stuffing box, and gland follower can
lead to binding of parts or extrusion of packing material. Guidance regarding tolerances and
clearances is provided in MSS Standard Practice SP-120, Flexible Graphite Packing Systems
for Rising Stem Steel Valves (Design Requirements) [43].
• The stem must be straight and have a good surface finish to produce a good seal, minimize
friction, and minimize packing wear. The stem must also be corrosion resistant.
• Stem misalignment can result in lateral loads on the packing that can potentially open up
gaps between the stem and packing. Misalignment can also result in contact between the
stem and gland that can cause wear and degrade the stem surface. In some cases, the effects
of misalignment can be mitigated by use of a graphite bushing at the top of the packing set.
This is a particularly good solution for valves in which the stem is horizontal and has high
lateral loads.

7-18
 EPRI Licensed Material

Preventing Leakage and Degradation

• When a hot valve stem is drawn up into the packing region, the packing is expanded radially
outward to accommodate the larger diameter of the hot stem as shown in Figure 7-8. This
radial outward expansion can result in packing consolidation. As the stem cools or is
reinserted into the valve, a potential leak path can be produced. Live loading may be required
to address this situation.
• Finally, the gland load must be high enough to develop a good seal but not so high that it
causes excessively high friction or causes packing extrusion.

It is obvious from the above that there are many potential causes of leaks from valve packing and
that the packing design and installation procedure must be carefully selected to avoid problems.

7.1.4.4 Selecting Packing Configurations and Products

The ideal packing configuration would have high leak tight integrity with low friction, be easy to
install with no need for tight tolerances or measurements, require that only a few parts be
stocked, and be inexpensive. The following general guidance is provided:
• The key to developing a high-integrity seal in most cases is to have at least one die-formed
graphite ring. If the friction can be accommodated, 2–3 die-formed rings would provide
additional margin.
• The die-formed rings should be backed up by low-friction braid rings above and below the
die-formed rings. If friction is critical, consideration should be given to composite-type anti-
extrusion rings, although composite-type anti-extrusion rings require tighter tolerances than
braid-type rings. This may necessitate measurements and custom fitting the ring.
• The remaining space in the stuffing box should be filled with a carbon bushing or spacer. If
the bottom of the stuffing box is not flat and square, a flexible graphite washer should be
installed to prevent cracking of the spacer. If there are high lateral loads or significant stem
misalignment, consideration should be given to installing a carbon bushing above the
packing assembly.
• Finally, if the valve is important, subject to large numbers of cycles, or has a history of
leakage, consideration should be given to using live load Belleville washers.

In the final analysis, users are encouraged to select a supplier and work with them to establish an
appropriate valve packing program for their plants. This will allow focused attention on
obtaining the best overall solution rather than working toward a model where one type of
packing is used for all applications. When selecting a supplier, consider the support the supplier
will provide and the test work that supports their products. In this regard, suggested valve
packing qualification tests are described in MSS Standard Practice SP-121, Qualification Testing
Methods for Stem Packing for Rising Stem Steel Valves [44].

7.1.4.5 Gland Follower Preload Requirements

The required gland follower preload is a tradeoff between having a high load to develop a good
seal and a low load to avoid excessive friction. The following general guidance is provided:

7-19
 EPRI Licensed Material

Preventing Leakage and Degradation

• EPRI NP-5697 suggests that all valve packing should be preloaded to a gland stress of 1.75
times system pressure, and preferably to 2000–4000 psi (14–28 MPa). Vendors may provide
product-specific recommendations that differ somewhat from this general recommendation.
• One key to a successful installation is to continue to stroke the valve stem and tighten the
gland nuts until there is no further nut rotation. This will ensure that the packing is well
consolidated at the start of operation.
• If valve packing is properly consolidated, the need for live loading is reduced. However,
there are some valves that should have live loading:
– Critical, inaccessible, and problem valves.
– AOV and MOV valves.
– Valves in high vibration or high cyclic duty operation.
– Valves with more than about five packing rings.
– Valves where it is impractical to stroke and tighten until the packing has been
consolidated. For example, one plant with a successful valve packing program uses live
loading selectively on about 150 of their 12,000 valves.

In the final analysis, the keys to packing preload are to establish an adequate gland preload stress
and ensure that the packing is consolidated by stroking the valve and tightening the gland nuts
until there is no further nut rotation. This may require more than 10 stroking and tightening
cycles. Live loading should be considered as a supplement to, rather than as an alternative to,
proper initial consolidation.

7.1.4.6 Valve Packing Assembly Guidelines

Guidelines for valve packing assembly will be provided in the FSTP report to be issued in 2001.
The key points of this procedure will include the following:

1. During disassembly, assess the valve and packing for indications of leakage, binding, wear,
or other damage.

2. Inspect the valve stem, gland, stuffing box, and bolting for damage, wear, steam cuts,
cleanliness, or other problems.

3. Obtain the new packing and inspect it to ensure that it is as specified and in good condition.

4. Install the new packing assembly wearing gloves to ensure that the packing is not
contaminated. Lightly lubricate the packing parts using the specified lubricant to aid in the
initial consolidation, and use a packer to install and consolidate the packing rings in steps.

5. Install the gland follower ensuring that it fits without binding.

6. Install the gland bolting including a hardened steel washer under the nut, and lubricate the
stud-to-nut threads and the washer-to-follower surface with the specified lubricant.

7-20
 EPRI Licensed Material

Preventing Leakage and Degradation

7. Consolidate the packing by stroking the valve stem and tightening the nut a minimum of five
times; continue to stroke and tighten until there is no further nut rotation. The stroke distance
needs to be equal only to the height of the packing set.

8. Install the live load package if specified.

Additional tips on assembly for specific types of valves are contained in the FSTP report on
assembling valve packing, which is scheduled for release in 2001.

7.1.5 Leak Reduction from Compression Fittings

Guidance on assembling compression fittings is contained in report TR-114760, Assembling

Threaded Connections [45]. This report describes a multistep approach to reducing leakage from
joints with compression fittings. The following is a brief summary of the main points of this
report. Users are advised to obtain the complete report for additional information.
• The first recommendation is to not intermix compression fitting parts from different vendors
and to carefully follow vendor assembly recommendations. These recommendations cover
tube condition, component assembly, and joint assembly and torque. A typical cause of leaks
is not following vendor assembly instructions.
• The second recommendation is to upgrade the fitting type from flare fittings to ferrule fittings
to O-ring fittings. At each level, the joint reliability tends to improve.
• The third recommendation is reduce the number of fittings by changing to bent tube, welded
tube, or flexible hoses with fittings used only where needed to allow assembly and
disassembly.

7.1.6 Leak Reduction from Threaded Pipe Joints

Guidance on assembling threaded pipe joints is also contained in report TR-114760, Assembling
Threaded Connections [45]. The following is a brief summary of the key points of this report.
Users are advised to obtain the complete report for additional information.
• As shown in Figure 7-9, pipe threads have inherent leak paths that must be sealed by a thread
sealant.
• The best way of reducing leaks from threaded pipe joints is to replace these inherently leak-
prone joints with more reliable welded joints or compression fittings.
• Fittings required in the system for assembly and disassembly can be converted from pipe
threads to more reliable O-ring face seal fittings as shown in Figure 7-10. For example,
conventional pipe unions can be replaced with O-ring unions, and threaded inspection port
plugs can be replaced with O-ring caps. This latter solution is well suited to inspection ports
that are opened frequently by operators who do not always reapply thread sealants.
• Where pipe thread joints are required, users are encouraged to select one of the thread
sealants identified in EPRI FSTP testing as producing leak tight joints with minimum cure
time. The best thread sealants for typical applications are listed in TR-114760.

7-21
 EPRI Licensed Material

Preventing Leakage and Degradation

• When assembling threaded connections, mechanics should ensure the following:

– The threads are in good condition.
– The threads are clean.
– The sealant is properly applied.
– The sealant is given adequate cure time prior to applying fluid pressure and temperature.
– The joints are not rotated to adjust the alignment after initial assembly.

7.1.7 Leak Reduction from Mechanical Seals

Several boric acid leaks have resulted from problems with Conoseal joints used to seal reactor
vessel head instrument penetrations. Several alternative designs have been developed by utilities
and vendors to overcome leakage at this location. The basic concept in the alternative designs is
to avoid the redundant load path created by trying to load the upper and lower seals
simultaneously.

The EPRI FSTP will address the subject of leakage reduction from mechanical shaft seals in
2001. Users are referred to this program for the most current results and recommendations.

7.1.8 Leak Reduction from Pressure Boundary Components

As reported in Section 2 of this guidebook, several incidents have occurred in which small leaks
have developed in Alloy 600 nozzles attached to pressure boundary parts by J-groove welds or
CRDM nozzle canopy seal welds.

7.1.8.1 Leaks from Cracks in Alloy 600 Nozzles and J-Groove Welds

There have been many reported cases of leakage of primary water from primary water stress
corrosion cracks (PWSCC) in Alloy 600 nozzles attached to pressure boundary parts by partial
penetration J-groove welds. An overview of industry experience is provided in EPRI TR-103696,
PWSCC of Alloy 600 Materials in PWR Primary System Penetrations [46]. With one exception,
the leakage has been so small that no loss of carbon steel pressure boundary material was
reported. The exception was a leak from a pressurizer heater sleeve at ANO-2 in 1982. In this
incident, swelling from a failed heater element caused a crack in the heater sleeve, which then
opened up to the extent that there was significant leakage. This leakage caused a reported loss of
1.3 in3 (21.3 cm3) of pressurizer bottom head material.

7.1.8.2 Leaks from CRDM Nozzle Canopy Seal Welds

Figure 7-11 shows the typical arrangement of the canopy seal weld in a Westinghouse plant.
Pressure loads on the joint are taken by the ACME threads between the head adapter and flange
on the CRDM or instrument nozzle. The weld serves only to provide a seal for this connection; it
is not intended to carry any significant loading.

7-22
 EPRI Licensed Material

Preventing Leakage and Degradation

A Westinghouse study determined that the leakage from the canopy seal welds resulted from
transgranular stress corrosion cracking [47]. The probable cause was that the seal weld is
subjected to high residual stresses from the welding, and the water that is trapped in the cavity
behind the seal weld does not meet normal primary system water chemistry requirements for
chlorides and oxygen. It was hypothesized that water that does not meet operating condition
chemistry requirements becomes trapped in this area during hydrostatic testing and remains in
this dead-end location during plant operation.

The most straightforward remedial measure for this type of situation is to improve the water
chemistry behind the canopy seal weld. A Westinghouse review of this situation suggested
making a hole in the head adapter to bleed the dead-end cavity. Other plants have applied weld
overlay cladding over leaking canopy seals.

7.2 Preventing Degradation If Leaks Occur

As described in Section 2 of this guidebook, leakage of borated water has led to significant
degradation of carbon steel and low-alloy steel parts. The majority of such cases have involved
corrosion of carbon steel or low-alloy steel fasteners used to assemble the joints. In a few cases,
other carbon steel or low-alloy steel parts have corroded.

The consequences of leakage in these components can often be reduced by replacing the carbon
steel or low-alloy steel parts with more corrosion-resistant materials or diverting any leakage to
areas where it will not cause damage. The following describes several possible areas of
improvement.

7.2.1 Low-Alloy Steel Fastener Replacement

One obvious remedial measure is to replace carbon and low-alloy steel fasteners with more
corrosion-resistant materials, such as:
• 304 stainless steel
• 316 stainless steel
• 410 stainless steel
• 17-4 PH stainless steel
• A-286
• Inconel 718
• Inconel X-750

Figure 4-2 shows the very low corrosion rates that would result from a change to 410 stainless
steel, A-286, or 17-4 PH stainless steel. Inconel 718 or X-750 would be expected to have similar
low corrosion rates. However, low corrosion rates are not the only criteria for replacement
bolting materials. Table 7-3 is a comparison of several key properties of potential alternative
bolting materials relative to the low-alloy steel bolting materials typically used for these
applications.

7-23
 EPRI Licensed Material

Preventing Leakage and Degradation

Alternative bolting materials have been evaluated by several organizations. Results of two such
evaluations follow.

7.2.1.1 Combustion Engineering Evaluation of Alternative Fastener Materials

Combustion Engineering performed a series of tests in which borated steam impinged on loaded
specimens of bolting materials that had corroded in service and two alternative materials (17-4
PH stainless steel and Inconel 718) [30]. Both of these materials have improved corrosion
resistance relative to carbon steel and are permitted by Section III of the ASME Code for bolting
applications.

The corrosion tests showed uniformly high corrosion rates for all of the low-alloy materials and
no corrosion at all for the two alternative corrosion-resistant materials.

Combustion Engineering also experimented with several different types of protective coatings on
low-alloy steel fasteners. The results of this work were uniformly discouraging with the
protected fasteners having equal or greater corrosion rates than the unprotected fasteners.

7.2.1.2 EDF Evaluation of Alternative Fastener Materials

EDF has conducted corrosion tests on four materials with potentially improved resistance to 40%
(by weight) borated water [48]:
• Cold worked 316 stainless steel (Z5 CND 17-12)
• 17-4 PH stainless steel (Z6 CNU 17-4)
• A-286 (Z6 NCTDV 25-15)
• Nitronic 50 HS (Z6 CNM 22-12-5 Az)

The tests showed that all four materials have much better corrosion resistance than low-alloy
steel bolting in the presence of high-concentration, high-temperature borated water. However,
the materials are not immune to developing fissures. EDF concluded that cold worked 316
stainless steel has good corrosion resistance at temperatures up to 200°C (392°F) and Nitronic 50
HS has good corrosion resistance up to 290°C (555°F). 17-4 PH material hardened at 595°C
(1100°F) shows some potential for fissures in borated water after aging for 10,000 hours at
300°C (572°F). A-286 material heat treated at 980°C (1800°F) shows some potential for fissures
at long test times.

At an EPRI Bolted Joint Leak Reduction Workshop, EDF indicated that A-286 bolting has been
used to replace low-alloy steel bolting for applications at temperatures less than 260°C (500°F)
[49]. These replacements have been performed as part of ongoing preventive maintenance
programs rather than to disassemble valves solely for the purpose of replacing bolts.

EDF is evaluating the possible application of a double layer of nickel onto the low-alloy stud
reactor coolant pump studs [AFNOR 40NCD 07.03 (ASTMA 540 Grade B24) with C =
0.37/0.44, Ni = 1.65/2.0, Cr = 0.70/0.95, Mo = 0.30/0.40]. The coating is intended to protect the

7-24
 EPRI Licensed Material

Preventing Leakage and Degradation

low-alloy steel base material from corrosion if a leak occurs at the flange gasket. However, there
are no firm plans to install coated studs at present due to the good experience with the three-loop
reactor coolant pumps, which do not experience leaks.

7.2.1.3 Summary Regarding Alternative Fastener Materials

Alternative bolting materials are reported in Table 7-3. These materials are also covered in EPRI
TR-109668-SI-R1, Materials Handbook for Nuclear Plant Pressure Boundary Applications,
[35].

From a cost standpoint, the best general purpose alternative to standard A193 Grade B7 fasteners
appears to be iron-nickel-chromium alloy A-286. This material has excellent resistance to
wastage in borated water and approximately the same strength as A193 Grade B7. It has a higher
coefficient of thermal expansion than A193 Grade B7, so the design must be reviewed carefully
for potential loosening of fasteners at temperature. If loosening is a concern, Belleville washers
could be incorporated to maintain the desired preload. EDF tests have shown that A-286 is
susceptible to fissuring at temperatures above 250°C (482°F) in high-concentration (40% by
weight) boric acid. However, this does not represent a realistic operating condition. The primary
coolant has high temperature but low concentration. After leakage occurs, the temperature is
limited to the boiling temperature of the boric acid solution that is close to 212°F (100°C) at
atmospheric pressure, and no fissures were reported at this temperature. Finally, no service-
related problems have been reported with this material in external bolting applications. As
mentioned in Section 5, the bolt preload stress for this material (including stress concentrations)
should be limited to 72–85 ksi (496–586 MPa). Because the typical stress concentration for
threads is about 4x, the threads should be rolled to permit higher tensile stresses at the thread
root.

The main alternatives to A-286 for general low-alloy steel replacement bolting applications are
17-4 PH stainless steel, 410 stainless steel, and XM-19 (Nitronic 50). However, each of these
materials has limitations in external bolting applications. 17-4 PH stainless steel should not be
used above 500°F (260°C) due to the potential for thermal embrittlement. 410 stainless steel is
susceptible to SCC in the presence of chlorides and sulfur. XM-19 is potentially susceptible to
SCC in the cold worked condition and is not known to have been used extensively for external
bolting applications.

For higher strength bolting applications, only Inconel X-718 and X-750 have the requisite
strength and proven performance in nuclear plant applications. These materials also have
coefficients of thermal expansion close to that of the alloy steel bolting that they would most
likely replace.

7.2.2 Replacement of Other Carbon and Low-Alloy Steel Parts

While most problems with boric acid corrosion have involved low-alloy steel bolting, there have
been problems with corrosion of some other low-alloy steel parts as well. These include CRDM
nozzle nut rings in B&W designed plants, clamps for reactor vessel head instrument nozzle
Conoseal joints, valve bonnets, and valve yokes. In some cases, it may be desirable to replace

7-25
 EPRI Licensed Material

Preventing Leakage and Degradation

these parts with more corrosion-resistant materials. This is especially true where there is a
history of leakage, the parts are highly stressed, and their failure could lead to rapid increases in
leakage.

It would not be practical to replace the larger pressure boundary parts with more corrosion-
resistant materials. Protection for these parts can best be achieved through eliminating leaks,
periodic monitoring for any small amount of leakage that does occur, and relying upon leak-
before-break to ensure there are no failures.

7.2.3 Protective Shrouds or Covers

There have been cases where leakage from components drips onto carbon steel parts. In some
cases, it may be possible to install protective shrouds or covers to divert leakage to areas where it
will not cause damage. This may be an economically attractive option and should be considered
as a remedial measure for components that have a history of leaking.

7.2.4 Oxygen Limiting Band for Flanged Joints

Based on the understanding of boric acid corrosion described in Section 3, oxygen is necessary
for corrosion to occur. Upon completion of the original Boric Acid Corrosion Guidebook, it was
hypothesized that a relatively tight-fitting band around a flange joint would result in any leakage
passing through gaps in the band as steam and that this steam flow would prevent oxygen
ingress. By keeping oxygen levels inside the band low, corrosion rates of carbon steel inside the
band should also be low. The oxygen band is the subject of EPRI patent US5851033, Corrosion
Limiting Device.

EPRI performed testing at Southwest Research Institute using an 8-inch (20 cm) horizontal
flange at 600°F (316°C) with leak rates of 0.01 and 0.1 gpm (0.038 and 0.38 L/min). Tests were
performed with insulation, without insulation, and with a relatively tight-fitting band. The test
results are given under Test Reference EPRI-7 in Section 4. These tests showed that:
• At a 0.01 gpm (0.038 L/min) leak rate, the maximum corrosion rates were highest for the
case without insulation or the tight-fitting band and the corrosion extended to more studs.
There was no significant difference in maximum corrosion rates between the cases with
insulation and the tight-fitting band.
• At a 0.1 gpm (0.38 L/min) leak rate, the corrosion was clearly higher, and more studs were
affected for the case with insulation and the tight-fitting band. However, the maximum and
average corrosion rates with the band were significantly lower than for the case with
insulation.

The above results are not conclusive, but they clearly suggest the potential for reducing corrosion
by keeping oxygen out of the joints. Further work would be necessary to explore the critical
parameters for such a device.

7-26
 EPRI Licensed Material

Preventing Leakage and Degradation

7.3 Stopping Leaks During Plant Operation

There are several methods that can be used to stop leakage that is discovered during plant
operation. These include retightening fasteners, injecting sealant, and installing backup
mechanical seals. Details of these approaches can be obtained from suppliers and service
contractors.
Table 7-1
Leakage Classifications by Severity (Proposed by EPRI FSTP)

Fluid 0 1 2 3 4 5
No leak Inactive Detectable Small Leak Intermediate Large Leak
leak Leak Leak
Oil No Inactive < 1 drop/5 min Up to a steady Leakage less Leakage
leakage leak* stream of drops than 0.1 gpm greater than
about 0.02 gpm (0.38 L/min) 0.1 gpm
(0.08 L/min) (0.38 L/min)
Water No Inactive < 1 drop/5 min Up to a steady Leakage less Leakage
leakage leak* stream of drops than 0.1 gpm greater than
about 0.02 gpm (0.38 L/min) 0.1 gpm
(0.08 L/min) (0.38 L/min)
Steam No Inactive Simmering Leakage stream 4.a Measured 5.a Measured
leakage leak* with no detectable <2"** <0.1 gpm > 0.1 gpm
observable (<51 mm) (<0.38 L/min) (>0.38 L/min)
stream 4.b Stream 5.b Stream
detectable <6" detectable >6"
(<152 mm) (>152 mm)
Air No N/A Detectable Leakage stream Leakage Leakage
leakage using a soap detectable <2"*** stream stream
solution (<51 mm) detectable <6" detectable >6"
(<152 mm) (>152 mm)
* Inactive leaks indicated by dry deposits near seal surface.
** Steam leakage stream detectable by impingement on a mirrorred surface or tell-tale.
***Air leakage stream detectable by motion of a tell-tale.

7-27
 EPRI Licensed Material

Preventing Leakage and Degradation

Table 7-2
Leakage Acceptance Criteria (Proposed by EPRI FSTP)

Leakage Severity Category I Joints: Category II Joints: Category III Joints:

Most Important Intermediate Importance Least Important

0 - No Leak --- --- ---

1 - Inactive Leak M --- ---

2 - Detectable Leak E M ---

3 - Small Leak E E M

4 - Intermediate Leak R E E

5 - Large Leak R R E

M= Monitor the condition

E= Evaluate if repair is required
R= Repair probably required

7-28
 EPRI Licensed Material

Preventing Leakage and Degradation

Table 7-3
Alternative Bolting Materials

Material Specification Common Temper Chromium Nickel RT Yield Strength RT Tensile Strength Elongation Thermal Thermal
Name Condition Content (%) Content (%) in ksi (MPa), min. in ksi (MPa), min. (%), min. Conductivity in Expansion in
British British
(and SI) Units (and SI) Units

Low Strength Bolting (< 75 ksi [<517 MPa] yield strength)

SA-307 Gr. A --- --- --- --- ≈30 (207) 60 (414) min 18 32.3 (55.9) 7.3 (13.1)
SA-193 Gr. B8 304 SS Class 2 18.0–20.0 8.0–11.0 80 (552) 115 (793) 15 9.8 (17.0) 9.4 (16.9)
SA-193 Gr. B8M 316 SS Class 2 16.0–18.0 10.0–14.0 80 (552) 100 (689) 20 9.0 (15.6) 9.4 (16.9)
≈ 100 ksi [689 MPa] yield
Intermediate Strength Bolting (≈
strength)
SA-193 Gr. B7 4140 1100 0.75–1.2 --- 105 (724) 125 (862) 16 24.2 (41.9) 7.1 (12.8)
SA-320 Gr. L43 4340 Q&T 0.7–0.9 1.65–2.0 105 (724) 125 (862) 16 21.5 (37.2) 7.2 (13)
SA-193 Gr. B6 410 SS 1100 11.5–13.5 --- 85 (586) 110 (758) 15 15.2 (26.3) 6.1 (11)
SA-564 Type 630 17-4 PH H1150 15.0–17.5 3.0– 5.0 105 (724) 135 (931) 16 10.3 (17.8) 7.0 (12.6)
SA-638 Gr. 660 A-286 Type 1 13.5–16.0 24.0–27.0 85 (586) 130 (896) 15 8.7 (15.1) 9.4 (16.9)
SA-479, Type XM-19 Nitronic 50 Hot rolled 20.5–23.5 11.5–13.5 105 (724) 135 (931) 20 7.7 (13.3) 8.9 (16)
≈ 125 ksi [862 MPa] yield strength)
High Strength Bolting (≈
SA-540 Gr. B23 Cl.3 4340H 850 0.65–0.95 1.55–2.0 130 (896) 145 (1000) 12 21.2 (36.7) 7.2 (13)
SB-637 (UNS N 07718) Inconel 718 1150 17.0–21.0 50.0–55.0 150 (1034) 185 (1276) 12 7.8 (13.5) 7.6 (13.2)
SB-637 (UNS N 07750) Inconel X-750 Type 2 14.0–17.0 70.0 min 115 (793) 170 (1172) 15 7.8 (13.5) 7.3 (13.1)
Notes:
1. Data for intermediate- and high-strength materials from EPRI TR-109668-SI-R1, Materials Handbook for Nuclear Plant Pressure Boundary Applications.
2. Data for low-strength materials from ASME Boiler and Pressure vessel Code, Section II.
3. Data reported for bolts of 1-inch (25 mm) diameter.
4. Within each category, materials above the dotted line are typical low-alloy steel bolting materials, and materials below the dotted line are high-alloy
corrosion-resistant alternatives.
5. Thermal conductivity (BTU-ft/hr-ft2-°F [W/m-C]) at 300°F (149°C).
6. Coefficient of thermal expansion (10-6 in./in./°F [10-6 m/m/°C)]) from 70–500°F (21–260°C).

7-29
 EPRI Licensed Material

Preventing Leakage and Degradation

Figure 7-1
Leakage Reduction Program Flowchart

7-30
 EPRI Licensed Material

Preventing Leakage and Degradation

Figure 7-2
Functions Provided by Rough Flange Surface in Gasketed Joints

Figure 7-3
Typical Inner Edge Buckling on Spiral-Wound Gasket

7-31
 EPRI Licensed Material

Preventing Leakage and Degradation

Figure 7-4
Effect of Gasket Thickness and Flange Flatness on Sealability

Figure 7-5
Functional Performance of Valve Packing

7-32
 EPRI Licensed Material

Preventing Leakage and Degradation

Figure 7-6
Evolution of Valve Packing

7-33
 EPRI Licensed Material

Preventing Leakage and Degradation

Figure 7-7
Effect of Friction on Packing Radial Load and Consolidation

Figure 7-8
Effect of Stem Thermal Taper on Valve Packing Performance

7-34
 EPRI Licensed Material

Preventing Leakage and Degradation

Figure 7-9
Inherent Leak Paths in Tapered Pipe Threads

7-35
 EPRI Licensed Material

Preventing Leakage and Degradation

Figure 7-10
Typical O-Ring Face Seal Fittings

7-36
 EPRI Licensed Material

Preventing Leakage and Degradation

Figure 7-11
CRDM Nozzle Canopy Seal Weld

7-37
 EPRI Licensed Material

8
CONTINUED OPERATION WITH LEAKAGE AND/OR
DEGRADATION

This section describes an approach that can be used to justify continued plant operation with the
leakage of borated water and/or material degradation. As previously described in Section 7,
primary emphasis should be on preventing leakage from occurring in the first place and then
stopping leaks when they are found by retightening joints, injecting sealants, and other
procedures. A justification for continued operation (JCO) with ongoing leakage should be
pursued only when other options have failed and leakage cannot be stopped in a timely manner.

8.1 Dealing with Leaks and Degradation

In some cases, it may be desirable to continue operating a plant with leakage and/or continuing
degradation. A prime reason is that leakage may be discovered during plant operation, and it may
be desirable to defer maintenance until the next scheduled refueling outage so that the repairs
will not result in a power decrease or plant shutdown. In other cases, it may be desirable to defer
repairs for problems discovered during a refueling outage due to a lack of parts or other reason.
In either case, a justification for continued operation (JCO) with the leakage and/or degradation
must be prepared.

A JCO for leaking borated water typically involves a two-step approach as shown in Figure 8-1.
First, it is demonstrated that the affected components currently meet applicable code design
requirements and will continue to meet applicable code design requirements in the future.
Second, a case should be made that plant safety will be ensured by defense-in-depth
considerations involving enhanced in-service monitoring, , PRA analyses, the fact that the worst
case scenarios have already been analyzed and reported in the final safety analysis report
(FSAR), and leak-before-break behavior of the component if predictions of future damage prove
to be less than conservative.

This section provides guidance regarding the key factors involved in preparing a JCO for leaking
borated water or for corrosion produced by the leakage. Appendix B contains several sample
calculations. It is emphasized that a document of this type cannot cover all possible
circumstances and that other resources will undoubtedly be required. However, the methods
described in this document will provide useful guidance on how to approach other situations
when developing JCOs.

Finally, the level of effort in developing the JCO will depend upon the criticality of the affected
parts. Preparation of a JCO for operation with a small leaking valve in an isolable line requires
significantly less effort than a JCO for continued operation with a leaking reactor coolant pump
flange gasket.

8-1
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

8.2 Characterizing Current Leakage and Degradation

The extent of current leakage and/or degradation must be determined to establish a baseline for
analyses of current conditions and predicted future conditions. This typically involves several
activities.

8.2.1 Characterizing Current Leakage

Section 6 of this guidebook provides information regarding leakage detection. When leakage is
detected, its source and flow rate must be quantified.

Attempt to characterize the current leakage by measuring the flow rate using the approaches
described in Section 6, by assessing the quantity of boric acid crystals using Figure 6-3 or
measuring the leakage into a catch basin. Leakage should be quantified using the classifications
given in Table 7-1.

8.2.2 Defining the Affected Areas/Components

A key requirement of Generic Letter 88-05 is to determine the areas and components affected by
the leakage. This includes the component in which the leak has occurred as well as other
components that could possibly have been affected by the leakage. This process will involve
determining the source of the leak, tracing the path of leakage, and assessing all carbon steel
parts in the leak path.

For example, leakage from a valve body-to-bonnet flange joint at room temperature may have
flowed under the insulation on a cold stainless steel pipe and concentrated on a high-temperature
carbon steel nozzle that is then corroded by the concentrated acid (see the example for Arkansas
Nuclear One 1 in 1986 in Section 2.3.1).

8.2.3 Characterizing Degradation in Equipment That Has Been Disassembled

If the equipment is disassembled, the extent of degradation can be assessed largely using
conventional measuring instruments. This process typically involves:
• Building scaffolding and removing insulation to provide improved access
• Cleaning boric acid deposits from the leak site and leak path
• Visually inspecting surfaces for evidence of boric acid corrosion
• Quantifying material loss

Several techniques have been developed for unique applications. Some of these are:
• Inspections of Intermediate Diameter (1–3 Inch [25–76 mm]) Female Threads
One area where difficulty is sometimes encountered is assessing the condition of female
threads that fail Go/No-Go gauge inspections. This inspection can show that the threads are
oversize but provide no information regarding the degree to which they are oversize. This

8-2
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

condition can occur as a result of thread corrosion or as a result of wear caused by repeated
torquing of bolts under high load conditions. One approach for this situation is to measure the
thread minimum diameter using a hole diameter gauge and measure the actual thread pitch
diameter using a thread pitch micrometer. Figure 8-2 shows a typical inside hole micrometer
and thread pitch gauge.
• Inspections of Large Diameter (>3 Inch [>76 mm]) Female Threads
For large diameter female threads, such as in the reactor vessel flange, it is sometimes not
practical to use thread pitch diameter micrometers. Framatome has developed a laser
inspection system for this application that creates a video image of the thread profile. The
amount of degradation can be determined by comparing the video image of the degraded
threads to the video image of intact threads. The operating principle of this inspection tool is
illustrated in Figure 8-3.

8.2.4 Characterizing Degradation in Equipment That Has Not Been Disassembled

If equipment is still assembled, characterization of degradation is more difficult; however, the

extent of degradation must be determined if the equipment is critical to plant safety. The
following are several methods that can be used to characterize degradation in equipment that is
still assembled:
• Degradation Estimated Based on Amount of Boric Acid Deposits
Experience has shown that the amount of degradation can sometimes be correlated with the
quantity of leakage and the environment in which the leakage occurs. Factors in this
assessment include the volume of boric acid crystals, the materials, and the environment
(temperature, pressure, boron concentration, joint type, immersion/impingement, etc.). Very
small amounts of boric acid deposits generally correlate with small amounts of leakage and
vice versa. (However, this is not always the case. For example, in the reactor coolant pump
stud corrosion at Ft. Calhoun in 1981, the studs experienced corrosion of 1/4-inch (6.4 mm)
depth as a result of the rewetting of boric acid deposits left on the surface after the 1980
incident.)

Additional guidance on predicting degradation is provided in Section 8.3.

• Cleaning Equipment
The first step required to characterize degradation due to boric acid corrosion is to remove
deposits from around the areas of interest so that the underlying metal surfaces are exposed.
Deposits can be removed using scraping tools, brushes, and vacuums. These methods are
particularly effective in regions with low leak rates on high-temperature components because
the leaking primary coolant immediately flashes to steam, leaving the boric acid deposits
largely in loosely adherent powder form.

On lower temperature parts and components, leaking primary coolant may remain in liquid
form. Evaporation of the fluid will then result in concentration of the boric acid. The boric
acid deposits will be more adherent and consequently more difficult to remove. Scraping,
wire brushing, and vacuuming are typically supplemented by wiping with warm, wet rags.
• Remote Video Examination
Equipment, components, and parts that are potential sites of primary coolant leakage and,

8-3
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

therefore, of boric acid corrosion are often covered by cast or reflective insulation, or they
are physically located in areas where personnel access is difficult or impractical. In these
cases, remote visual inspection equipment can be useful for preliminary assessments of
potential leakage sites.

A variety of inspection devices are available. These devices exhibit a wide range of levels of
sophistication and, therefore, cost. Nevertheless, the use of even the most expensive of the
devices can be cost effective if it eliminates the need to remove insulation or erect
scaffolding or ladders. Several of these inspection devices are illustrated in Figure 8-4.

Perhaps the simplest inspection device that is useful for preliminary evaluations of potential
leakage sites is an articulated, lighted mirror. Such devices are available at a nominal cost,
can be extended several feet or more, and can be used to inspect surfaces of components that
are out of a direct line of sight.

When distances to the component exceed several feet, it may be useful to use a miniature
video camera mounted on the end of a telescoping wand or pole. These devices are available
from several manufacturers for purchase or lease. The cameras are usually color charge
coupled devices (CCDs) with integral light sources. The camera mount may incorporate
motorized pan, tilt, zoom, and remote focus. The camera is operated by controls and a
joystick on the handle.

Whereas a mirror or camera on a pole is useful for inspection of parts with exposed surfaces
or for the identification of deposits that have accumulated on the outside surface of
insulation, evidence of boric acid leakage underneath insulated components may require
borescopes, fiberscopes, or videoprobes. These devices provide the ability to probe through
and beneath insulation to search for evidence of boric acid deposits and even complete a
range of dimensional inspections.

Borescopes are rigid fiberoptic inspection devices that resemble a long, thin tube with a
monocular eyepiece and handle at one end. These straight, rigid devices are available in
diameters of less than 1 mm. They can be inserted between insulation and a component and
provide high-quality images. Their usefulness is somewhat limited in that they cannot turn to
negotiate bends nor are they typically available in lengths over one meter. Borescopes can be
fitted with CCD cameras so that the image can be displayed on a video monitor or recorded
on a VCR.

A more versatile tool for the inspection of components and detection of leaks is a flexible
fiberscope, which consists of a sheathed bundle of optical fibers with a light source and a
lens at the distal end. A fiberscope, which is typically 1 to 2 mm in diameter, can negotiate
turns and bends and, therefore, can access a larger portion of the surfaces of a component
underneath insulation, provided there is at least a small gap between the insulation and the
component. Instead of an eyepiece at the terminal end of the device, a fiberscope is usually
mated to a CCD camera that allows the image to be displayed on a video monitor.

Somewhat larger than a fiberscope, a videoprobe places the CCD camera at the distal end of
the device. Available in diameters of 1/4–1/2 inch (6–12 mm), a videoprobe provides high-
quality video images of components. Recently, several models of videoprobes have been

8-4
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

made available with articulation capabilities that allow the operator to steer the device as it
negotiates through narrow passages. One manufacturer of a videoprobe has added a
capability for measuring component diameters and dimensions by “casting” shadow gridlines
over a surface and then analyzing the deformation or distortion of the imposed grid lines by
computer. Surface defects as shallow as a few mils can be measured with this equipment.
• Bolt Diameter Gauges
Field experience has shown that the maximum corrosion on bolts in gasketed joints occurs in
the area between the flanges. The amount of shank reduction in this region can be determined
using shank diameter gauges as illustrated in Figure 8-5. This approach was used to
determine the diameter reduction on reactor coolant pump flange studs at Oconee 3 in 1981.
However, it should be noted that in at least one case, the damage determined in this manner
was about half of the maximum damage found at inaccessible elevations within the bolt hole
in the flange.
• Ultrasonic Inspection of Bolts with Suspected Degradation
EPRI has sponsored work to develop methods to inspect for damage to bolts without the need
for joint disassembly. The results of this work are described in Section 7 of EPRI report NP-
5769, Degradation and Failure of Bolting in Nuclear Power Plant–Volume 1 [51]. This
report describes three methods for inspecting for “hourglass” type corrosion of studs. These
are:
– Acoustic resonance
– Reverberation
– Cylindrically guided wave technique (CGWT)

The acoustic resonance approach is based on the concept that corrosion of the bolt will
change the natural modes of vibration. In this approach, the bolt is excited with “white noise”
and a Fast-Fourier-Transform spectrum analyzer is used to determine the vibration modes.
The reverberation approach is to induce longitudinal vibrations in the bolt and measure the
response using a piezo-electric accelerometer. Laboratory experiments using both of these
approaches showed promise.

The CGWT approach is based on the fact that an ultrasonic wave traveling down a long
cylinder becomes “guided” by the cylinder. Defects in the bolt can be identified by the
difference in time between the backwall reflection and the reflections from defects that have
undergone mode conversion. The laboratory work showed that this method is capable of
detecting wastage greater than 25% in bolts ranging from 1–4.5 inches (25–114 mm) in
diameter and 16–112 inches (0.4–2.8 m) in length. However, it should be noted that a 25%
reduction in bolt diameter represents about a 45% reduction in bolt area, and this may not
prove sufficiently accurate to serve as the basis for a JCO.

8.3 Predicting Future Leakage and Degradation

After the current extent of leakage and degradation has been characterized, the next step is to
estimate the extent of degradation at the next planned inspection. There are several factors
involved in such an estimate. (Note: In some cases, the extent of current degradation must also
be estimated because disassembly for inspection may not be practical.)
8-5
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

8.3.1 Establishing the Type of Corrosion Environment

The first step in predicting future degradation is to determine the type of corrosion environment
that exists. The basic environments are shown in Figure 4-1. If several environments can exist
within a joint, they should all be evaluated. The main environments are:
• Immersion in deaerated water
• Immersion in aerated water
• Dripping onto hot metal surfaces
• Impingement onto hot metal surfaces
• Leakage into annular gaps such as at instrument or RPV head CRDM nozzles

This characterization is important because the corrosion rate varies widely with the conditions
where the leakage occurs.

8.3.2 Establishing the Maximum Corrosion Rate for the Environment

The maximum corrosion rate that can occur in the subject environment should be established
using the data in Figure 4-3 and Table 8-1. Table 8-1 provides a relatively simple guide to the
maximum corrosion rates for several typical environments. If several conditions exist, they
should all be evaluated. For example, leakage from a cold flange could result in low leak rates
for the flange bolts, but if the leakage subsequently flows onto a hot carbon steel pipe, the
corrosion rate at this location could be very high.

8.3.3 Evaluating Possible Mitigating Circumstances

Field experience has shown that the actual corrosion rates for many situations can be
significantly less than the maximum values determined from laboratory tests and shown in
Figure 4-3. Therefore, the possible beneficial effects of mitigating measures should not be
ignored in developing reasonable upper bounds to the predictions. Based on a review of the
laboratory data and field experience, the following factors should be considered:
• Effects of Flange Geometry, Leak Rate, and Temperature
Tests conducted by EPRI and summarized in Test Reference EPRI-7 in Section 4 show that
the amount of corrosion of carbon steel bolts varies with position around the flange and with
flow rate, temperature, flange orientation, and insulation.

At low temperatures (<180°F [82°C]), the corrosion rates were low (<0.04 in/year [1
mm/yr]), and the corrosion tended to be uniformly distributed around the flange. This is
consistent with many of the bolts becoming wetted at the same time.

With high-temperature (600°F [316 °C]) uninsulated flanges, the corrosion rates tended to be
much higher (up to 1.25 in/year [32 mm/yr]) and to be concentrated more on the bolts nearest
the leak point. Surprisingly, the corrosion rates were higher for low leak rates (0.01 gpm
[0.038 L/min]) than for higher leak rates (0.1 gpm [0.38 L/min]). Perhaps this is a result of

8-6
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

high leak rates preventing the leaking boric acid from concentrating on studs near the leak
point.

For high-temperature (600°F [316°C]) insulated flanges, the maximum corrosion rate
increases, and the corrosion affects more bolts than for the uninsulated case, at the higher
leak rates of 0.1 gpm (0.38 L/min). This suggests that the insulation captures the leaking fluid
within the flange, producing high moisture levels around the entire flange.
• Oxygen Level
Laboratory tests in deaerated water have shown very low corrosion rates, regardless of the
temperature or boric acid concentration.

Using this as background, a case can be made that corrosion rates will be reduced at locations
where leaking steam keeps oxygen from entering an enclosed volume. This concept is
illustrated in Figure 8-6. An example of where this mitigating factor could play an important
role in reducing corrosion rates is in the flange region of components that are covered by
tight-fitting metallic-type insulation. In this example, steam flowing out of openings in the
insulation may act to minimize the ingress of oxygen. Tests conducted by EPRI and
summarized in Test Reference EPRI-7 showed that there was less corrosion with a tight-
fitting band than with insulation for high leak rates of 0.1 gpm (0.38 L/min). However, there
was more corrosion with a band than with an uninsulated flange. Further work is required to
determine if corrosion rates can be reduced practically by limiting oxygen ingress.
• Leakage into Annular Gaps
As noted in Section 2, there have been many cases involving borated water leaking into
annular gaps between Alloy 600 nozzles and carbon steel pressure boundary parts. Field
experience (ANO-1 pressurizer heater sleeves) and Combustion Engineering laboratory tests
on borated steam leaking into a downward-facing tight annulus between a simulated Alloy
600 pressurizer heater sleeve and carbon steel vessel shell show that there is essentially no
corrosion until the leakage reaches the exit region of the annulus where it can pick up
oxygen.

Tests conducted by EPRI show that there may be small amounts of corrosion inside the
annulus for nozzles that are oriented upward. The difference in behavior may be a result of
natural circulation patterns within the annulus.
• Effect of Boric Acid Deposits in Preventing Surface Wetting
Laboratory tests and field experience have both shown that dry boric acid deposits produce
very little corrosion. This is best evidenced by the large volume of boric acid deposits on the
vessel head at Turkey Point 4 and the relatively minor 0.25-inch (6.4 mm) corrosion depth on
the reactor head. This is consistent with recent experience at VC Summer where PWSCC in
an Alloy 82/182 hot leg nozzle weld resulted in a deposit of about 280 pounds (127 kg) of
boric acid with no measurable wall thinning of the adjacent carbon steel nozzle. The
corrosion rate in these cases was clearly not 5–10 in/yr (127–254 mm/yr). It is hypothesized
that, under certain conditions, boric acid deposits on the vessel head actually protect the
surface from corrosion by keeping water away from the surface. This concept is illustrated in
Figure 8-7. The potential for corrosion under this condition is obviously a function of the
leakage source, leak rate, and the depth of the deposits. If the leakage rate is low and its
source is above the boric acid deposits, heat transfer through the deposits will evaporate the

8-7
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

incoming water and thereby keep the surface dry. On the other hand, if the leakage rate is
high or if the source is located within the boric acid deposits, the deposits will be wetted,
leading to high corrosion rates at the vessel head.
• Volume of Leakage Relative to Metal Temperature
As shown by the laboratory test data in Figure 4-3 corrosion rates for borated water dripping
on a hot metal surface vary from undetectable to as high as 7 in/yr (178 mm/yr). It is believed
that this large difference results from the volume of leakage relative to the metal temperature.
If the leakage rate is low and the metal temperature high, any leaking water will quickly boil
off without significantly reducing the metal temperature. Corrosion rates under these
conditions should be low. If the leakage rate is high, there will be little concentration of boric
acid and the corrosion rates may still be low. However, at some intermediate range, the
leakage rate will be high enough to cool the metal surface temperature to 200–220°F (93–
104°C) where the boric acid can concentrate and produce high corrosion rates.

In summary, there are several potential mitigating factors. Readers are encouraged to review
their specific application relative to the above, and to the actual test data in Section 4 to select the
appropriate correlation for this application.

8.3.4 Predicting Corrosion at the Next Planned Inspection

The final step in this part of the evaluation is to use the maximum predicted corrosion rate based
on field and laboratory test experience to predict the maximum corrosion at the next planned
inspection. This calculation should be conservative to reflect the uncertainties in the extent of
current degradation and the maximum possible future corrosion rate. Consideration should also
be given to the potential for parts in the field to differ from those shown on design drawings. For
example, several cases are reported in Section 2 where carbon steel fastener parts had been used
in place of the specified stainless steel. If there is any question in this regard, materials should be
confirmed, possibly by use of magnetic testing for ferritic material, or conservative assumptions
made.

8.4 Evaluating Predicted Future Degradation vs. Code Requirements

After the predicted degradation at the next planned inspection has been established, it is
necessary to determine whether the component will continue to meet the applicable ASME Code
requirements with this amount of degradation. Typically, this process involves the evaluation of
corroded bolts, corroded threads, corroded piping, corroded flanges, corroded pressure vessel
shells, and other corroded components.

This approach can also be used to assess current damage for cases where the leakage has been
stopped and no further degradation is expected. This approach may allow some degraded parts to
remain in service. For example, it may be desirable to continue to use some reactor coolant pump
studs or reactor vessel head studs with minor corrosion damage.

A complete guide for evaluating degradation in all components is beyond the scope of this
document. The following suggests guidance for several situations typically encountered. Several
example calculations are provided in Appendix B.

8-8
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

8.4.1 Evaluating Corroded Bolts

As indicated in Section 2 and Appendix A of this document, many of the reported incidents have
involved corrosion of carbon steel or low-alloy steel bolting. In most cases, the simplest course
of action is to replace the corroded parts. However, if the corrosion is minor, no spare parts are
available, or the damage is discovered during operation, it may be necessary to develop a
technical case for accepting the bolting in its corroded condition.

The general approach in such a case is to compare the measured minimum cross section area of
the bolt to the required cross section area given in the component design report. If the actual area
in the corroded condition is greater than the minimum required area, then the bolt is acceptable
and can continue to be used.

If the bolt cannot be accepted on this basis, it is often useful to review the component design
report to determine if there is some conservatism that can be removed. For example, transient
analyses are generally performed using worst case conditions, and fatigue analyses are generally
performed using conservative estimates of applied cycles. In many cases, it is possible to show
that a part is acceptable based on the actual plant operating conditions. For example, the plant
heatup transient is typically taken as 100°F/hr (38°C/hr), and this often creates the highest loads
in bolted connections. However, most plants operate with heatup transients of 50°F/hr (10 °C/hr)
and less. If warranted, this difference can be used to accept some otherwise unacceptable
fasteners. It should be noted that this type of reevaluation is sometimes costly, and its practical
use may be limited to large and expensive fasteners, such as reactor vessel and reactor coolant
pump closure studs (see Example Problem B-3).

8.4.2 Evaluating Corroded or Degraded Threads

The general practice for small diameter fasteners less than about 1 inch (25 mm) in diameter is to
replace the fasteners and install threaded inserts into flanges containing female threads. These
methods restore the full original strength of the parts.

The preferred practice for larger sizes is to replace the fastener also. This restores the full
original strength and minimizes the potential for sticking due to use of a fastener with degraded
threads. In some cases, it is necessary to reuse fasteners with degraded threads because of a lack
of spare parts. The normal practice in this case is to remove or clean up the damaged threads and
then show that the remaining threads are adequate to carry the applied loads. This approach is
practical because most joints have excess thread engagement length.

For larger-size fasteners, it is often desirable to evaluate the damage to female threads and
continue to use the female threads if they meet the allowable stress criteria. This approach avoids
the need to install threaded inserts. The evaluation of female threads generally involves
measuring the thread minor diameter, measuring the thread pitch diameter, calculating the
number of threads required for the measured minor and pitch diameters, and subtracting the
percentage of the full engagement area that has been degraded by the corrosion (see Example
Problem B-4).

8-9
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

8.4.3 Evaluating Corroded Pressure Vessels and Piping

The case of boric acid corrosion is analogous to the case of wall thinning by erosion-corrosion in
that its extent is often limited. Criteria for evaluating the acceptability of wall thinning have been
developed by EPRI and are described in report NP-5911M, Acceptance Criteria for Structural
Evaluation of Erosion-Corrosion Thinning in Carbon Steel Piping [50]. This document describes
several approaches to analyzing thinning that are outlined in Figure 8-8. The general approach is
to start with the easiest of the methods and progress to more complex analyses if the easier
approaches fail to provided the desired resolution. The main approaches are as follows:
• Corrosion does not infringe on minimum required wall thickness
In many cases, the actual wall thickness of parts is greater than the minimum required wall
thickness. This can occur because:
– The designer specified a greater-than-required wall thickness.
– The supplier provided a greater-than-specified wall thickness.
– The actual corrosion is within the original corrosion allowance.

In any case, this excess material provides some margin for corrosion without requiring
further analysis. Three checks can be made in this regard:
– The design report can be checked to determine the minimum required thickness.
– The material certification can be checked to determine the delivered thickness.
– The actual thickness can be measured in place by calipers or UT thickness gauges.

EPRI report NP-5911M provides guidance on the relationship between the nominal pipe size
and minimum required pipe thickness (see Example Problem B-2).
• Corrosion is localized and meets code requirements for unreinforced opening in vessel shell
Paragraph NB-3332.1 of Section III of the ASME Boiler and Pressure Vessel Code permits
holes in a shell of 0.2 Rt without the need for any reinforcement at all. For a steam
generator head with a 60-inch (1.5-m) radius and a 4-inch (102-mm) wall thickness, this
would permit a through hole 3.1 inches (79 mm) in diameter without the need for analysis or
reinforcement. Thus, if the corroded area is less than 3.1 inches (79 mm) in diameter and if
there is sufficient thickness at the bottom to withstand the pressure loading as a flat head, the
corrosion can be quite significant. However, it should be noted that paragraph NB-3332.1
includes some restrictions on the presence of other holes in the immediate vicinity, and
fatigue should always be considered.
• Corroded area meets local membrane stress requirements
The equations for tmin in design codes typically assume that the minimum thickness acts over
the entire component. Section III of the ASME Boiler and Pressure Vessel Code provides for
a higher membrane stress limit of 1.5 Sm over local areas provided that the extent of the
stresses exceeding the membrane stress (Sm) does not exceed a distance of Rt min . This
approach is illustrated in Figure 8-9a.

8-10
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

• Corroded area meets requirements for reinforced opening in vessel shell

Finally, Section III of the ASME Code provides for large openings in a vessel shell if area of
reinforcement requirements are met. These requirements, which are outlined in paragraph
NB-3330 of Section III of the ASME Boiler and Pressure Vessel Code, specify that the area
of reinforcement provided by excess wall thickness in a component must equal the amount of
required material that has been removed by corrosion. The Code also requires that this
reinforcement be concentrated within specified distances from the opening. These
requirements are illustrated in Figure 8-9b.

8.4.4 Evaluating Flanges and Other Special Purpose Parts

For corrosion that cannot be evaluated using any of the above simple methods, it may be
necessary to perform stress analyses of the degraded part and then assess the adequacy relative to
the original design code requirements. Figure 8-10 shows a typical example of corrosion that
actually occurred on the low-alloy steel flange of a reactor coolant pump as a result of leakage of
borated water from the closure gasket. The acceptability of the degradation was determined by
finite element analysis of the flange. In this case it was conservatively assumed that the worst
case reduction in cross section extends 360° around the flange permitting an axisymmetric finite
element model. If this approach had shown that the stresses were higher than ASME Code
allowables, the analysis would have to be performed using a 3D model. Finally, the results of
this analysis were summarized in parametric form so that utility inspectors could evaluate future
degradation without having to repeat the analyses for slightly different conditions (see Example
Problem B-5).

8.5 Establishing Supplemental Inspection or Operational Requirements

The final step in preparing a JCO is to establish supplemental inspections and operational
requirements based on the predicted future degradation. If there is significant margin between the
predicted and allowable degradation, it may be possible to increase the inspection interval. On
the other hand, if the predicted future degradation exceeds the allowable degradation, it will be
necessary to perform more frequent inspections. Similarly, it may be necessary to impose
supplemental operational constraints such as reducing leakage at other locations by use of
injections to improve the sensitivity to the leak in question.

8.6 Defense in Depth

Finally, after the technical case has been developed to show that the components will remain
within code allowables until the next planned inspection/maintenance outage, additional
evidence should be provided to demonstrate that the plant will remain safe even if the
calculations prove to be unconservative. The defense-in-depth argument is typically based on the
factors covered in the subsections that follow.

8-11
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

8.6.1 Enhanced Inspections and Heightened Awareness

An important outcome of the root cause analysis and development of the JCO should be the need
for supplemental inspections and a generally heightened awareness of the leak that is being
allowed to continue or of previous damage that has not yet been repaired. These factors should
go a long way toward ensuring that the situation receives the close attention it deserves.

8.6.2 PRA Analyses

Analyses can be performed to demonstrate that the leakage will not have a significant adverse
effect on the Core Damage Frequency (CDF) as determined by a Probabilistic Risk Assessment
(PRA). This evaluation would provide additional assurance for significant problems that the
plant will remain safe.

8.6.3 Analyzed Accident per FSAR

Chapter 15 of the plant FSAR describes calculations performed to show that the plant will
remain safe for several worst case hypothetical events. These previously analyzed cases should
cover essentially all of the leakage events of interest up to and including a full double-ended
primary-loop pipe rupture.

8.6.4 Leak-Before-Break

Most bolted joints in pressure boundary components include sufficient design margin and
redundancy that failures of small numbers of fasteners will result in detectable leakage before
there is a risk of rupture. Such joints are said to exhibit leak-before-break (LBB) behavior. The
leak-before-break behavior provides assurance that the component under investigation will
remain safe even if the predictions of future leakage and degradation prove to be unconservative.
In this manner, leak-before-break provides a “defense in depth.”

The subject of leak-before-break of bolted closures has been evaluated in EPRI report NP-5769,
Degradation and Failure of Bolting in Nuclear Power Plants–Volume 1 [51]. The work in this
report was directed toward closures that are critical to plant safety and that have experienced
problems with fastener degradation due to boric acid leakage. These include steam generator
manways, pressurizer manways, reactor coolant pump flanges, and large valves.

The basic approach to leak-before-break in NP-5769 is illustrated in Figure 8-11. The approach
consists of:
• Establishing a finite element structural model of the component
• Computing the effect of load redistribution for failure of increasing numbers of contiguous
studs
• Determining the leak rate resulting from the redistribution

8-12
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

• Assessing the stresses resulting from the redistribution

• Assessing the margin of safety

A joint would be considered acceptable from a leak-before-break standpoint if three conditions

are met:
• Instrumentation is in place to reliably detect the leakage.
• Safety consequences of the predicted leakage are acceptable.
• There is an acceptable margin against rupture at the point where the leakage becomes
detectable under all plant operating conditions.

The following is a summary of the approach described in EPRI NP-5769 supplemented by an

analysis of a reactor vessel closure and an additional leak rate model. Because this information is
summary in nature, it is recommended that users refer to the original report NP-5769 before
developing a leak-before-break model for a specific situation.

8.6.4.1 Stress Redistribution from Failed Fasteners

Begin a leak-before-break analysis for bolted closures by determining the stress redistribution
that occurs in the remaining bolting and pressure boundary parts as a result of failures of
increasing numbers of fasteners. This type of analysis is typically performed using the finite
element method that can easily handle the nonaxisymmetric conditions.

The first step in such an analysis is to:

1. Create a finite element model of the component, including the bolting, bolt preload, and the
gasket stiffness.

2. Determine stresses in the component for design conditions including transients.

3. Break away increasing numbers of fasteners and determine the resultant stresses in the
remaining fasteners and pressure boundary parts.

The second step in the structural analysis is to evaluate the stresses in the fasteners and pressure
boundary parts to determine the point at which the stresses reach an allowable limit. The
allowable limit is generally based on ASME Code requirements. Paragraph NB-3228.1 of
Section III of the ASME Boiler and Pressure Vessel Code requires that the maximum loadings
shall not exceed two-thirds of the lower bound collapse load. Assuming that the lower bound
collapse load is the operating temperature yield strength of the fasteners, this requires a
minimum factor of safety of 1.5 on net section yielding.

EPRI report NP-5769 [51]includes finite element analyses of typical manway closures, a typical
reactor coolant pump flange, and typical 6- and 10-inch check valves. These models show the
typical level of mesh geometry refinement required for this type of analysis. Higher order
elements (for example, 20 node bricks) can provide increased accuracy for elastic analyses
without the need for increased mesh refinement. These models are reproduced in Figures 8-15

8-13
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

through 8-19. The plot opposite from the model shows the factor of safety on net section yielding
as a function of the number of failed fasteners. These figures show that failure of a single stud
typically changes the margin of safety in the adjacent studs by less than 10%.

Figure 8-17 shows results of similar calculations performed by Dominion Engineering, Inc.,
(DEI), of McLean, VA., for failed reactor vessel head studs.

In summary, these analyses show that several studs can fail in all of the joints yet still retain an
acceptable margin of safety on net section yielding of the remaining studs.

CAUTION

The leak-before-break model developed in EPRI NP-5769 assumes that the fastener
degradation starts at one fastener and then spreads around the joint in a progressive
manner. This is illustrated in Figure 8-12a. Before applying this criterion, it must be
confirmed that the degradation mechanism will not affect many fasteners simultaneously,
such as shown in Figure 8-12b. Simultaneous degradation could result from SCC or from
leaking borated water collecting around the entire joint region. Although several incidents
involving fastener failures of the type illustrated in Figure 8-12b have exhibited leak-
before-break, it cannot be shown that all of these cases would have met ASME Code
margins of safety. For example, if a joint were held in place by only a few fasteners
uniformly distributed around the flange and stressed to the material yield strength, any
small additional amount of corrosion would result in detectable leakage and some small
amount of yielding, but the margin of safety on yielding would only be 1.0. This type of
joint would not fail catastrophically, but it would not meet the ASME Code specified
margin of safety.

Results of EPRI flange tests in Section 4 can be used to determine the type of degradation
distribution that should be expected.

8.6.4.2 Leak Rate Resulting from Failed Fasteners

The finite element analysis can predict the size of the opening at gasketed joints as a function of
the number of failed fasteners. The predicted joint opening can then be used to predict the
amount of leakage that occurs. Figure 8-13 shows a typical slit or crack and the types of flow
conditions that are encountered within the slit. Flow within the slit typically starts as a subcooled
liquid. As the pressure drops, the flow becomes two-phase flow.

EPRI report NP-5769 [51] describes one method to calculate the leakage from a slit in a gasketed
joint. This method involves breaking the slit down into regions of saturated liquid flow and two-
phase flow and calculating the pressure drop for various flow rates. The equilibrium solution is
the point at which the predicted pressure drop through the crack for the given flow rate equals
the system pressure. These calculations are typically performed using a computer program. The
main advantage of this approach is that the slit need not be rectangular in shape throughout its
length. The predicted leakage rates for the joints studied in the finite element analyses are
provided in the plots at the end of this section.

8-14
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

An alternative approach would be to use the method described in NUREG CR-3475, Critical
Discharge of Initially Subcooled Water Through Slits [52]. This approach is based on the use of
curves to calculate the mass flow of subcooled liquid through a rectangular slit. The curve for the
mass flow rate as a function of the amount of subcooling and slit length/depth ratio is given in
Figure 8-14. The L/D ratio is the length of the slit (that is, the gasket width) divided by the slit
width (that is, the flange separation at the gasket). The leak rates shown for the reactor vessel
flange bolt failures are calculated using this NUREG approach.

8.6.4.3 Leakage Detection

For the primary system components discussed above, a leakage rate of 1 gpm (3.8 L/min) can be
accommodated while still maintaining the required margins of safety. This leak rate can be
reliably detected using the existing leak detection systems of the primary system.

It is possible that the maximum allowable leak rates for other parts could be less than 1 gpm (3.8
L/min). In these cases, it must be confirmed that adequate provisions are made to detect the
leakage before it poses a concern. Alternative leak detection systems for locations requiring
higher sensitivity could include moisture-sensitive tape, humidity sensing equipment, acoustic
emissions, or radiation detection from leaking coolant.

8.6.4.4 Field Confirmation of Operation with Failed or Missing Fasteners

The effectiveness of leak-before-break can be assessed by reviewing the reports of damage to the
steam generator and pressurizer manway bolting. The relevant data from the database in
Appendix A of this report are summarized in Table 8-2.

These data show that there have been no ruptures of manways as a result of bolt failures and that
several incidents involved detectable leakage before safety problems developed. At Maine
Yankee, the fact that the closure was able to continue operation with six studs that failed during
detensioning (and five more that were cracked) confirms that there is significant margin in the
closure designs.

Additional confirmation of the margin in reactor vessel bolting is provided by the fact that Union
Electric’s Callaway plant operated with an untensioned reactor vessel stud for 18 months. The
JCO for Callaway showed that the vessel continued to meet all of the normal ASME Code
allowable stress requirements, even with the untensioned stud. The main concern in this case
involved a slightly increased risk of O-ring leakage. In fact, no leakage occurred as was
predicted by the analysis.

8-15
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Table 8-1
Corrosion Rates for Typical Carbon and Low-Alloy Steel Applications

Estimated
Qualitative
Type of Environment Corrosion Rate
Corrosion Rate
in in/yr (mm/yr)*

Immersion in Deaerated Water Very Low 0.0002–0.0003

(0.005–0.008)

Immersion in Low Temperature (<140°F [<60°C]) Low 0.002–0.010

Aerated Water (0.05–0.25)

Immersion in High Temperature (>140°F Medium – Very High 0.01–10.0

[>60°C]) Aerated Water (0.25–254)

Dripping onto Hot (>180°F [>82°C]) Metal Medium – High 0.10–5.0

Surfaces (2.5–127)

Impingement onto Hot Metal Surfaces Medium – Very High 0.10–10.0

(2.5–254)

Leakage into Hot Annular Gaps Medium – High 0.02–2.0

(0.5–51)

* See Figure 4-3 for further details.

Table 8-2
Leak-Before-Break Experience for Manway Closures

Plant Date Leak Rate in Bolt Damage

gpm (L/min)

ANO-1 1981 3.4 (13) Several studs degraded

ANO-2 1979 Not reported Several studs degraded

ANO-2 1982 ≈3 (11) 8 studs degraded

Calvert Cliffs 1 1980 Not reported 11 of 40 studs in one generator

Calvert Cliffs 2 1981 Not reported 2 studs degraded

Indian Point 2 1983 Not reported Several studs degraded

Maine Yankee 1982 Not reported 6 of 20 studs found to be failed

St. Lucie 1 1977 Not reported 13 of 20 studs with surface corrosion

St. Lucie 1 1978 Not reported 5 of 20 studs with surface corrosion

8-16
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-1
Approach to Establishing JCO for Boric Acid Leakage/Degradation

8-17
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-2
Hole Diameter and Thread Pitch Micrometer

Inside Hole Micrometer from Mitutoyo

Thread Pitch Gauge from Greenslade & Co.

8-18
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-3
Laser Gauge for Large Diameter Internal Thread Inspections

8-19
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-4
Typical Video Inspection Equipment

8-20
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-5
Shank Diameter Gauge for RCP Flange Bolts

Figure 8-6
Effect of Flange Clearances in Reducing Oxygen Levels at Bolts

8-21
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-7
Effect of Boric Acid Deposits in Protecting Surfaces from Corrosion

8-22
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-8
Logic Chart for Evaluating Corrosion of Pressure Boundary Parts [50]

8-23
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-9
Local Membrane Stress and Area of Reinforcement Models [50]

8-24
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-10
Reactor Coolant Pump Flange Corrosion

8-25
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-11
Leak-Before-Break Approach [51]

8-26
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-12
Conditions Amenable to Leak-Before-Break Analysis

8-27
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-13
Model for Leakage from Rectangular Slit [51]

8-28
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-14
Mass Flow Rate for Leakage of Subcooled Water from Rectangular Slits (Created from
[52])

8-29
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-15
Finite Element Model and Results for 16-Stud Manway Closure [51]

8-30
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-15 (cont.)

Finite Element Model and Results for 16-Stud Manway Closure [51]

8-31
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-16
Finite Element Model and Results for 20-Stud Manway Closure [51]

8-32
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-16 (cont.)

Finite Element Model and Results for 20-Stud Manway Closure [51]

8-33
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-17
Finite Element Model and Results for Reactor Coolant Pump Flange [51]

8-34
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-17 (cont.)

Finite Element Model and Results for Reactor Coolant Pump Flange [51]

8-35
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-18
Finite Element Model and Results for 6-Inch Check Valve [51]

8-36
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-18 (cont.)

Finite Element Model and Results for 6-Inch Check Valve [51]

8-37
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-19
Finite Element Model and Results for 10-Inch (25-cm) Check Valve [51]

8-38
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Table 8-19 (cont.)

Finite Element Model and Results for 10-Inch (25 cm) Check Valve [51]

8-39
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-20
Finite Element Model and Results for Reactor Vessel Head

8-40
 EPRI Licensed Material

Continued Operation with Leakage and/or Degradation

Figure 8-20 (cont.)

Finite Element Model and Results for Reactor Vessel Head

8-41
 EPRI Licensed Material

9
BORIC ACID CORROSION MANAGEMENT

Utilities should develop and implement a strategic plan to manage boric acid corrosion in a
manner that is both safe and cost effective. Needless to say, in the current environment, cost
assessments should include both the direct cost of proactive programs and the potential cost
savings that can result from avoiding forced power reductions and shutdowns to repair leaks. In
many cases, it will be better to invest in improvements up front to avoid potentially greater costs
in the future.

9.1 Key Factors in an Effective Boric Acid Corrosion Management

Program

As illustrated in Figure 9-1, there are four key factors in an effective boric acid management
program. These are:
• A leakage reduction program to reduce the total number of leaks in a cost-effective manner,
with special emphasis on the most important and difficult to repair applications
• Monitoring of equipment and system condition
• Replacement of critical carbon steel materials (bolts, nuts, gland followers, yokes, etc.) with
corrosion-resistant materials for higher temperature applications where repair is difficult or
can involve loss of production
• An engineered program with a planned approach to dealing with leaks in systems with
borated water

9.1.1 Leakage Reduction

The first and perhaps the most important factor in managing boric acid corrosion is reducing the
number and magnitude of leaks. If there are no leaks, there will be no boric acid corrosion!! This
topic is covered extensively in Section 7.1.

9.1.2 Condition Monitoring

The second factor in a successful boric acid corrosion management program is sound condition
monitoring. This includes both equipment condition assessment and leakage detection.
Information on equipment condition can lead to improvements that can reduce the potential for
leakage. For example, potentially detrimental effects of smooth flanges, gouged flanges, out-of-
flat flanges, misalignment, damaged valve stems, damaged or corroded bolts, etc., can be
rectified and, thereby, reduce the potential for leakage. Similarly, low levels of plant leakage and

9-1
 EPRI Licensed Material

Boric Acid Corrosion Management

a good leakage detection system can improve the ability to detect leaks early enough to take
corrective action before more drastic measures are required. The subject of the effect of
equipment condition on leakage is discussed in Section 7.1, and the subject of leakage detection
is discussed in Section 6.0.

9.1.3 Replacement of Key Carbon Steel Parts with Corrosion-Resistant Materials

A review of experience in Section 2 and Appendix A shows that most of the more serious
incidents of boric acid corrosion have involved leakage of borated water followed by significant
corrosion of carbon or low-alloy steel fasteners or highly stressed parts, such as valve
followers/yokes or nut rings. If these parts had been made of corrosion-resistant materials, the
consequences of the leakage would typically have been significantly reduced. Section 7.2
describes alternative materials that can be used to avoid corrosion if leakage occurs. However,
many corrosion-resistant fastener materials possess low yield strengths and may not have
sufficient strength to adequately compress certain types of gaskets. For applications involving
spiral-wound gaskets, high-strength corrosion-resistant fastener materials may be required.
Section 4 provides test data showing corrosion rates for alternative materials.

It is not necessary to replace all carbon steel materials with more corrosion-resistant materials.
The more corrosion-resistant materials are necessary only for applications where the temperature
is above about 120°F (49°C) and repair is complicated by the need for power reductions, high
radiation exposure, poor access, and potential personnel safety hazards. Figure 9-2 shows a
typical screening approach.

9.1.4 Planned Response to Leaks

The fourth factor in a successful boric acid corrosion management program is having a planned
response to leakage when it occurs. This response can include monitoring, on-line repairs, or
repairs during a power reduction or plant shutdown, depending upon conditions. Sections 7 and 8
provide a suggested approach to leakage response.

9.2 Level of Planning Effort

Utilities will devote differing levels of effort to the development of a boric acid leakage
management program. This can range from decision making based purely on engineering
judgement to decision making based on life cycle management (LCM) approaches including
economic analysis of alternative courses of action. These two levels are briefly summarized as
follows:

9.2.1 Boric Acid Corrosion Management Plan Based on Engineering Judgement

In many cases, a relatively simple strategic plan for leakage and degradation reduction can be
developed that relies upon prioritization of components and selection of remedial measures by
engineering judgment. Figure 9-2 illustrates one such engineering judgment approach based on a
method originally proposed by Impell [53]. This approach groups components in three categories

9-2
 EPRI Licensed Material

Boric Acid Corrosion Management

depending upon factors such as operating conditions, leakage experience, and criticality to plant
operations. This approach is focused on applying remedial measures first to those components
that have the potential to cause serious problems or to result in lost production. The overall result
should be an effective program that requires minimal development and implementation cost.

9.2.2 Boric Acid Corrosion Management Plan Based on Economic Modeling

A more rigorous approach than in Figure 9-2 would be to establish an optimum course of action
for each component or application based on life-cycle economic modeling. This approach is
described in EPRI report 1000806, Demonstration of Life Cycle Management Planning for
Systems, Structures, and Components -- with Pilot Applications at Oconee and Prairie Island
Nuclear Stations [54]. The main steps in this approach would include the following:
• Estimate the probability of leakage and degradation for each component based on plant and
industry experience. This would include:
– Minor leaks with no damage
– Moderate leaks with some damage
– Major leaks with significant damage
• Estimate the cost of performing remedial measures and the estimated cost of corrective
maintenance, lost production, and other consequential costs in the event of leaks for each
component and level of degradation.
• Establish alternative management approaches based on the types of remedial measures
available, and determine when the remedial measures would be applied, including the
following:
– First available outage
– Outage for the first scheduled preventive maintenance on the component
– After the leakage incident
– At some predefined future outage
• Determine the net present value of the costs for each alternative evaluated.
• Select between the lowest cost alternatives for each component based on engineering
judgment.

The effort required to perform the evaluations under this approach can be expedited by relatively
simple computer spreadsheet models. The LCMVALUE program described in the EPRI LCM
demonstration program report is ideally suited to this type of economic evaluation.

If the economic modeling approach is pursued, it is suggested that the largest and most critical
components be evaluated separately and that others be grouped by similar applications and
operating conditions. Optimally, this approach should lead to the same strategic plan as the
engineering judgment approach of Figure 9-2.

9-3
 EPRI Licensed Material

Boric Acid Corrosion Management

9.3 EDF Approach to Leakage and Corrosion Management

EDF has developed a system-wide approach for minimizing problems with boric acid leakage
and corrosion. This approach was outlined by EDF in EPRI report NP-5985, Boric Acid
Corrosion of Carbon and Low-Alloy Steel Pressure-Boundary Components in PWRs [1]. The
EDF program was described as follows:
• Make Leaks an Exception – Extensive campaigns have been made to obtain a better behavior
of static gasket seals. The rules are well-known and simple:
− Use high-quality gaskets with precise and reproducible geometry and compression
characteristics.
− Correct the mating surface geometry and surface conditions to achieve a precise,
controlled (metal-to-metal) gasket compression.
− Recalculate the stresses to be applied to make sure that during the worst transients the
gasket compression does not vary. (ASME torque standards used were found to be
insufficient.)
− Tighten large studs with dedicated tensioning machines rather than torque wrenches.
These machines allow more accurate loadings than torque wrenches.

The procedures have significantly reduced the number of bonnet leaks.

• Be Prepared for Leaks – Despite the favorable French experience with stud resistance to
prolonged leaks; others, in the United States for instance, have experienced a few serious
cases. Furthermore, lab tests showed that the potential for high corrosion rates could not be
denied. The decision was made to replace all bolts or studs on valve bonnets and flanges with
a new material resistant to boric acid corrosion. The two materials selected were SA-453
Type 660 (25Ni 15Cr 1Mo steel “Nitronic”) and SA-564 Type 630 (17-4 PH stainless steel).1
The second material has good mechanical properties, but it is susceptible to age
embrittlement if used at more than 250°C (482°F). Therefore, 17-4 PH stainless steel is not
used at temperatures above 250°C (482°F).

All bolts on safety-related valves or flanges have been replaced after recalculations showed
acceptable mechanical strength, taking into consideration the higher required torque. Of
course, this includes all high-pressure valves on lines connected to the RCS. Packing bolts
have also been replaced on large valves. (On small valves, EDF policy has been to eliminate
packing and bonnet problems by using only all-welded, bonnetless, packingless valves.)
• Other Flanges – No alternative to SA-193 Grade B16 was deemed possible on the four main
assemblies (vessel head, reactor coolant pump flange, steam generator manway cover, and
reactor coolant loop bypass diaphragms).
− Nothing was done on the vessel head, which is equipped with a leak detection device and
double seals.
− RC loop diaphragm flanges were replaced by all-welded diaphragms on all units.

1
See Section 7.2.1 for the latest EDF bolt material practices.

9-4
 EPRI Licensed Material

Boric Acid Corrosion Management

− Extensive studies have been completed to demonstrate the absence of fast rupture for
reactor coolant pumps and steam generator primary manways where no acceptable
solution has been found.

These studies have demonstrated significant margins and the acceptability of several studs
being distorted by plastic deformation or even ruptured without endangering the actual
assembly resistance.

The studies have also established that prior to any rupture, the leak rate would increase to
high values, far higher than the Technical Specifications allow for continued operation.

Note: During a 1994 NMAC Bolted Joint Leak Reduction Workshop in Baltimore,
Maryland, EDF indicated that these changes were made over time as components were
disassembled for other maintenance and inspection. Thus, it was a continuous upgrade
program, rather than a rapid implementation of remedial measures.

Figure 9-1
Key Factors Regarding Boric Acid Corrosion Management

9-5
 EPRI Licensed Material

Boric Acid Corrosion Management

Figure 9-2
Criteria for Replacing Carbon Steel with Corrosion-Resistant Materials

9-6
 EPRI Licensed Material

10
REFERENCES

1. Boric Acid Corrosion of Carbon and Low-Alloy Steel Pressure-Boundary Components in

PWRs. EPRI, Palo Alto, CA: August 1988. NP-5985.

2. A Survey of the Literature on Low-Alloy Steel Fastener Corrosion in PWR Power Plants.
EPRI, Palo Alto, CA: December 1984. NP-3784.

3. Boric Acid Corrosion Evaluation (BACE) Program Phase I–Task 1 Report. EPRI, Palo Alto,
CA: December 1993. TR-101108.

4. Boric Acid Corrosion of Carbon Steel Reactor Pressure Boundary Components in PWR
Plants. NRC Generic Letter 88-05, March 17, 1988.

5. “Boric Acid Corrosion of Low-Alloy Steel.” Proceedings: 1991 EPRI Workshop on PWSCC
of Alloy 600 in PWRs. EPRI, Palo Alto, CA: 1992. TR-100852.

6. Volume PWR-2 Experiences. Nuclear Power Experience, RCG/Hagler, Bailly, Inc., Boulder,
CO: July 1993.

7. Nuclear Power Reliability Data System.

8. Boric Acid Corrosion of Ferritic Reactor Components. NUREG/CR-2827.

9. Threaded Fastener Experience in Nuclear Power Plant. NUREG-0943.

10. Survey of Boric Acid Corrosion of Carbon Steel Components in Nuclear Plants.
NUREG-5576.

11. Degradation of Reactor Coolant Pump Studs. NRC Information Notice 80-27, June 11, 1980.

12. Failure of Steam Generator Primary Side Manway Closure Studs. NRC Information Notice
82-06, March 12, 1982.

13. Degradation of Reactor Coolant System Pressure Boundary Resulting from Boric Acid
Corrosion. NRC Information Notice 86-108, December 29, 1986.

14. Degradation of Reactor Coolant System Pressure Boundary Resulting from Boric Acid
Corrosion. NRC Information Notice 86-108, Supplement 1, April 20, 1987.

15. Degradation of Reactor Coolant System Pressure Boundary Resulting from Boric Acid
Corrosion. NRC Information Notice 86-108, Supplement 2, November 19, 1987.

10-1
 EPRI Licensed Material

References

16. Boric Acid Corrosion of Charging Pump Casing Caused by Cladding Cracks. NRC
Information Notice 94-63, August 30, 1994.

17. Degradation of Threaded Fasteners in the Reactor Coolant Pressure Boundary of PWR
Plants. IE Bulletin 82-02, June 2, 1982.

18. “Velan,” Velan Catalog Vel-N-3. 1975.

19. Handbook of Corrosion Engineering. McGraw-Hill, 1999.

20. Guidelines for Selection of Marine Materials. International Nickel.

21. Boric Acid Corrosion Evaluation (BACE) Corrosion Program – Phase II Corrosion Testing,
Topical Report, Southwest Research Institute, 1998.

22. “Materials Behavior Related Issues for Bolting Applications in the Nuclear Industry,”
presented at the Conference on Improved Technology for Critical Bolting Applications,
Chicago, IL (July 20-24, 1986). MPC-Vol. 26.

23. Absorption of Corrosion Hydrogen by 302B Steel at 70F to 500F. Westinghouse Electric
Corporation, 1967, WCAP-7099.

24. “Corrosion of Structural Materials in the Boron-Containing Solutions Used for Controlling
the Power of Nuclear Reactors.” Therm. Eng. (Teploenergetika), Moscow Power Institute,
1970, pp. 13–17.

25. “Influence of Boric Regulation of a Medium on Corrosion and Low-Cycle Corrosion Fatigue
Resistance of Constructional Steels.” Soviet Material Science, 1982, pp. 135–139.

26. Degradation of B7 Threaded Fasteners in Borated-Water Solutions at Elevated

Temperatures - Interim Report. Tennessee Valley Authority, Knoxville, TN, July 1983.

27. J. F. Hall, R. S. Frisk, A. S. O'Neill, R. S. Pathania, W. B. Neff, “Boric Acid Corrosion of

Carbon and Low-Alloy Steels.” Proceedings of the Fourth International Symposium on
Environmental Degradation of Materials in Nuclear Power Systems–Water Reactors, Jekyll
Island, GA (August 6–10, 1989), pp. 9-38–9-52.

28. Studies of the Corrosion of AISI 304 Stainless Steel and AISI 4135 Carbon Steel Exposed to
Saturated Solutions of Boric Acid. Westinghouse Electric Corporation, Atomic Power
Department, Pittsburgh, PA, November 1958, YAEC-67.

29. Degradation of Reactor Coolant System Pressure Boundary Resulting from Boric Acid
Corrosion. Westinghouse letter NS-NRC-87-3260 to the NRC, dated October 15, 1987.

30. J. F. Hall, Environmental Degradation of Materials in Nuclear Power Systems —Water

Reactors. The Metallurgical Society, Warrendale, PA, 1988, pp. 711–722.

31. B&W Boric Acid Corrosion Research and the Wastage and Inspection Procedures for RCS
Leakage. B&W Nuclear Service Company, Lynchburg, VA, December 1990, BAW-2126.

10-2
 EPRI Licensed Material

References

32. Code of Federal Regulations, Part 50 (10CFR50). “Domestic Licensing of Production and
Utilization Facilities: General Design Criteria for Nuclear Power Plants.” April 1997.

33. Degradation of Reactor Coolant System Pressure Boundary Resulting From Boric Acid.
NRC Information Notice 86-108, Supplement 3, January 5, 1995.

34. ASME Boiler and Pressure Vessel Code: Section XI, Rules for Inservice Inspection.
American Society of Mechanical Engineers, New York, December 1993.

35. Materials Handbook for Nuclear Plant Pressure Boundary Applications. EPRI, Palo Alto,
CA: December 1998. TR-109668-SI-R1.

36. Assessment of Pressurized Water Reactor Primary System Leaks, Idaho National Engineering
Laboratory, 1997. NUREG/CR-6582.

37. Establishing an Effective Leak Management Program. EPRI, Palo Alto, CA: December
2000. TR-114761.

38. Assembling Bolted Connections Using Spiral-Wound Gaskets. EPRI, Palo Alto, CA: July
1999. TR-111472.

39. Bolt Preload for Raised-Face Flanges with Spiral-Wound Gaskets. EPRI, Palo Alto, CA:
May 2000. 1000066.

40. Assembling Bolted Connections Using Sheet Gaskets. EPRI, Palo Alto, CA: May 2001.
1000922.

41. Bolted Joint Maintenance and Application Guide. EPRI, Palo Alto, CA: 1995. TR-104213.

42. Valve Stem Packing Improvements. EPRI, Palo Alto, CA: May 1988. NP-5697.

43. Flexible Graphite Packing Systems for Rising Stem Steel Valves (Design Requirements),
MSS Standard Practice SP-120.

44. Qualification Testing Methods for Stem Packing for Rising Stem Steel Valves, MSS Standard
Practice SP-121.

45. Assembling Threaded Connections. EPRI, Palo Alto, CA: December 2000. 1000972.

46. PWSCC of Alloy 600 Materials in PWR Primary System Penetrations. EPRI, Palo Alto, CA:
July 1994. TR-103696.

47. C. M. Pezze, I. L. W. Wilson, “Transgranular Stress Corrosion Cracking of 304 Stainless

Steel Canopy Seal Welds in PWR Systems.” Proceedings of the Fourth International
Symposium on Environmental Degradation of Materials in Nuclear Power Systems–Water
Reactors, Jekyll Island, GA (August 6–10, 1989) pp. 4-164–4-179.

48. B. Prieux, J. M. Gras (EDF), “Corrosion of Bolting Stainless Steels in High Temperature
Concentrated Boric Acid,” International Symposium, Fontevraud II, September 10–14 1990,
pp. 558–567.

10-3
 EPRI Licensed Material

References

49. Jacques Cardon, “Cost/Benefits of Leak Reduction and Material Upgrade

Programs,”Presented at EPRI/NMAC Bolted Joint Leak Reduction Workshop, August 16–
18, 1994.

50. Acceptance Criteria for Structural Evaluation of Erosion-Corrosion Thinning in Carbon

Steel Piping. EPRI, Palo Alto, CA: July 1988. NP-5911M.

51. Degradation and Failure of Bolting in Nuclear Power Plants–Volume 1. EPRI, Palo Alto,
CA: April 1988. NP-5769.

52. Critical Discharge of Initially Subcooled Water Through Slits. NUREG/CR-3475.

53. N. Moisidis, M. Popesco, and M. Ratin, "Protect Nuclear Plant Fasteners from Boric Acid
Corrosion." Power Engineering, March 1992.

54. Demonstration of Life Cycle Management Planning for Systems, Structures, and Components
-- with Pilot Applications at Oconee and Prairie Island Nuclear Stations. EPRI, Palo Alto,
CA: January 2001. 1000806.

55. ASME Boiler and Pressure Vessel Code: Section III–Division 1, Appendices - 1992 Edition.
American Society of Mechanical Engineers, New York, December 1993.

56. John H. Bickford, An Introduction to the Design and Behavior of Bolted Joints - Second
Edition, Revised and Expanded. Marcel Dekker, Inc., 1990.

57. Fastener Standards - Fifth Edition. Industrial Fasteners Institute, Cleveland Ohio, 1982.

10-4
 EPRI Licensed Material

A
REPORTED INCIDENTS OF BORIC ACID CORROSION

This appendix briefly summarizes the reported incidents involving corrosion or other types of
degradation caused by leakage of borated water. This compilation is limited to those incidents
that have been reported in generally available sources. It does not included incidents that
occurred at plants and were resolved without the incidents being described in a generally
accessible forum.

While it is recognized that most incidents are handled without being reported outside the plant, it
is considered that the attached list covers the most significant incidents and provides a good
overview of the other types of conditions that can occur.

The incident summaries are presented in this appendix. They are organized by plant name and
date. The information can be easily accessed by the use of several tables:
• Table A-1 - Industry experience sorted by plant name
• Table A-2 - Industry experience sorted by system
• Table A-3 - Industry experience sorted by severity
• Table A-4 - Industry experience sorted by component

Each entry includes a relative severity rating from 1 (major incident) to 4 (minor incident). The
ratings are subjective and take into account both the criticality of the system and the significance
of the incident.

A-1
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-1
Industry Experience by Plant Name, Incident Date, Severity, System, and Component

Arkansas One 1 1978 2 Reactor Coolant Pressurizer

Arkansas One 1 1981 1 Reactor Coolant Steam Generator
Arkansas One 1 1986 1 Reactor Coolant High Pressure Injection Nozzles
Arkansas One 1 1986 3 Letdown, Purification & Makeup Reactor Coolant Pump
Arkansas One 1 1986 3 Letdown, Purification & Makeup Reactor Coolant Pump
Arkansas One 1 1989 1 Reactor Coolant Reactor Vessel
Arkansas One 2 1979 2 Reactor Coolant Steam Generator
Arkansas One 2 1982 1 Reactor Coolant Steam Generator
Arkansas One 2 1985 2 Chemical and Volume Control Boric Acid Discharge Check Valve
Arkansas One 2 1987 1 Reactor Coolant Pressurizer
Arkansas One 2 1987 3 Reactor Coolant Pressurizer
Arkansas One 2 2000 3 Reactor Coolant Pressurizer
Beaver Valley 1 1978 2 ECCS Boron Injection System Piping
Beaver Valley 1 1983 1 Reactor Protection Power Range Flux Detector
Beaver Valley 1 1987 1 Low Pressure Safety Injection Accumulator Discharge Valve
Beaver Valley 1 1990 1 Chemical and Volume Control HX Discharge Isolation Valve
Beaver Valley 1 1992 2 Chemical and Volume Control Letdown Heat Exchanger
Beaver Valley 2 1989 3 High Pressure Safety Injection Cold Leg Check Valve
Beaver Valley 2 1990 3 Chemical and Volume Control Letdown Heat Exchanger
Beznau 1 1969 3 Reactor Coolant Reactor Vessel
Beznau 1 1973 2 Reactor Coolant Reactor Vessel
Callaway 1992 3 Reactor Coolant Reactor Vessel
Calvert Clffs 1 1994 1 Reactor Coolant Reactor Vessel Head
Calvert Cliffs 1 1974 2 Reactor Coolant Reactor Vessel
Calvert Cliffs 1 1980 1 Reactor Coolant Reactor Coolant Pump
Calvert Cliffs 1 1980 2 Reactor Coolant Steam Generator
Calvert Cliffs 2 1981 2 Reactor Coolant Reactor Coolant Pump
Calvert Cliffs 2 1981 1 Reactor Coolant Reactor Coolant Pump
Calvert Cliffs 2 1981 1 ECCS Safety Injection Check Valve
Calvert Cliffs 2 1981 1 Reactor Coolant Steam Generator
Calvert Cliffs 2 1998 4 Reactor Coolant Pressurizer
Catawba 1 1989 2 Reactor Containment Containment Liner
Catawba 2 1989 2 Reactor Containment Containment Liner
Comanche Peak 1 1991 3 Residual Heat Removal RHR Pump
Connecticut Yankee 1968 2 Not Reported Valve
Connecticut Yankee 1989 2 Reactor Coolant Fill Header Motor Operated Valve
D. C. Cook 1 1975 4 ECCS Boric Acid Transfer Pump
D. C. Cook 2 1981 2 Chemical and Volume Control Charging Line Check Valve
D. C. Cook 2 1989 3 Chemical and Volume Control Charging Pump Suction Valve
Davis Besse 1 1998 1 Reactor Coolant Pressurizer Spray Valve
Davis Besse 1 1999 1 Reactor Coolant Valves
Diablo Canyon 2 1988 3 Reactor Coolant Steam Generator
Farley 2 1990 3 Residual Heat Removal Heat Exchanger Discharge Valve
Ft. Calhoun 1980 1 Reactor Coolant Reactor Coolant Pump
Ft. Calhoun 1981 2 Reactor Coolant Reactor Coolant Pump
Ft. Calhoun 1992 2 Chemical and Volume Control Boric Acid Pump
H. B. Robinson 1980 2 Reactor Coolant Pressurizer Spray Valve
Indian Point 2 1981 1 Reactor Coolant Reactor Coolant Pump
Indian Point 2 1981 2 Safety Injection Check Valve
Indian Point 2 1983 1 Reactor Coolant Steam Generator
Indian Point 3 1976 2 ECCS BIT Tank Outlet Nozzle
Indian Point 3 1990 3 Reactor Coolant Reactor Vessel
Kewaunee 1974 3 Reactor Coolant Pressurizer
Kewaunee 1974 3 Reactor Coolant Steam Generator

A-2
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-1 (cont).

Industry Experience by Plant Name, Incident Date, Severity, System, and Component

Kewaunee 1976 3 Reactor Coolant Reactor Vessel

Kewaunee 1980 2 Reactor Coolant Steam Generator
Kewaunee 1981 2 ECCS Boric Acid Tank Suction Valve
Kewaunee 1981 3 Chemical and Volume Control Boric Acid Pump
Kewaunee 1981 3 Not Reported Instrument Isolation Valve
Kewaunee 1982 2 Reactor Coolant Pressurizer
Kewaunee 1982 2 ECCS Safety Injection Check Valve
Kewaunee 1985 3 Residual Heat Removal Suction Isolation Valve
Maine Yankee 1982 1 Reactor Coolant Steam Generator
Maine Yankee 1982 3 Spent Fuel Pool Cooling Pump Discharge Valve
Maine Yankee 1995 1 High Pressure Safety Injection Gate Valve
McGuire 1 1989 2 Reactor Containment Containment Liner
McGuire 2 1989 2 Reactor Containment Containment Liner
Millstone 2 1988 1 Reactor Coolant Reactor Vessel
Millstone 2 1995 3 Regenerative HX Relief Valve
North Anna 1 1982 2 ECCS Cold Leg Injection Check Valve
North Anna 1 1983 2 Reactor Coolant RTD Bypass Isolation Valve
North Anna 1 1993 2 High Pressure Safety Injection Pump
Oconee 1 1979 3 Reactor Coolant Reactor Vessel
Oconee 1 2000 2 Reactor Coolant Reactor Vessel Head
Oconee 2 1978 2 Reactor Coolant Pressurizer
Oconee 2 1981 1 Reactor Coolant Reactor Coolant Pump
Oconee 3 1981 1 Reactor Coolant Reactor Coolant Pump
Oconee 3 1982 1 Reactor Coolant Reactor Coolant Pump
Palisades 1971 2 Reactor Coolant Reactor Vessel
Palisades 1980 2 Chemical and Volume Control Letdown Heat Exchanger
Palisades 1986 2 Reactor Coolant Pressurizer
Palisades 1987 2 Reactor Protection Control Rod Drives
Palisades 1 1999 3 Reactor Coolant Reactor Vessel Head
Palo Verde 1 1988 3 High Pressure Safety Injection Containment Isolation Valve
Palo Verde 1 1999 4 Reactor Coolant Hot Leg Piping
Palo Verde 2 1988 3 Low Pressure Safety Injection Refueling Water Isolation Valve
Palo Verde 2 1990 2 Low Pressure Safety Injection Safety Injection Tank Disch. Valve
Point Beach 2 1972 3 Reactor Coolant Reactor Vessel
Point Beach 2 1988 3 Chemical and Volume Control Charging Pump Suction Valve
Point Beach 2 1990 2 Reactor Coolant Reactor Vessel
Prairie Island 1 1974 2 ECCS Isolation & Drain Valves
Prairie Island 1 1976 3 Reactor Coolant Reactor Vessel
Prairie Island 1 1986 3 Chemical and Volume Control Boric Acid Line Heat Tracing
Robinson 2 1976, 2 Reactor Coolant Reactor Vessel
Robinson 2 1980 1 Chemical and Volume Control Pressurizer Spray Valve
Robinson 2 1995 2 Reactor Coolant Reactor Coolant Pump
Salem 1 1980 3 ECCS Boron Injection Tank
Salem 2 1987 1 Reactor Coolant Reactor Vessel
San Onofre 1 1988 2 Chemical and Volume Control Volume Control Tank Relief Valve
San Onofre 2 1985 3 Chemical and Volume Control Boric Acid Make-Up Pump
San Onofre 2 1985 3 Chemical and Volume Control Boric Acid Make-Up Pump
San Onofre 2 1986 3 Chemical and Volume Control Boric Acid Make-Up Pump
San Onofre 2 1987 1 Low Pressure Safety Injection Shutdown Cooling Insolation Valve
San Onofre 2 1988 3 Containment Isolation Hot Leg Sample Valve
San Onofre 2 1992 2 Reactor Coolant Pressurizer
San Onofre 2 1997 4 Reactor Coolant Hot Leg Piping
San Onofre 2 1998 4 Reactor Coolant Hot Leg Piping
San Onofre 3 1986 4 Chemical and Volume Control Boric Acid Make-Up Pump

A-3
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-1 (cont).

Industry Experience by Plant Name, Incident Date, Severity, System, and Component

San Onofre 3 1986 2 Chemical and Volume Control Boric Acid Pump
San Onofre 3 1995 4 Reactor Coolant Pressurizer
San Onofre 3 1997 4 Reactor Coolant Hot Leg Piping
Sequoyah 1 1995 3 Reactor Coolant Control Rod Drives
Shearon Harris 1987 3 Reactor Coolant Pressurizer Level Transmitter
Shearon Harris 1989 2 Reactor Coolant Hot Leg Isolation Valve
Shearon Harris 1992 2 Reactor Coolant Hot Leg Isolation Valve
South Texas 2 1992 2 Reactor Coolant Reactor Vessel
St. Lucie 1 1977 2 Reactor Coolant Steam Generator
St. Lucie 1 1977 2 Reactor Coolant Steam Generator
St. Lucie 1 1978 2 Reactor Coolant Pressurizer
St. Lucie 1 1978 1 Reactor Coolant Pressurizer
St. Lucie 1 1990 3 Chemical and Volume Control Boric Acid Discharge Check Valve
St. Lucie 2 1986 2 Not Reported Safety Valve
St. Lucie 2 1990 3 High Pressure Safety Injection Flow Element
Surry 1 1978 3 ECCS Heat Tracing
Surry 1 1989 3 Reactor Coolant Fill Header Isolation Valve
Surry 2 1975 2 ECCS Boric Acid Tank
Surry 2 1978 3 ECCS Heat Tracing
Surry 2 1978 1 Residual Heat Removal Suction Isolation Valve
Surry 2 1983 3 ECCS Heat Tracing
Three Mile Island 1 1994 1 Reactor Coolant Pressurizer Spray Valve
Trojan 1984 3 Reactor Coolant RTD Bypass Isolation Valve
Turkey Point 3 1985 2 Reactor Coolant Reactor Vessel
Turkey Point 4 1987 1 Reactor Coolant Reactor Vessel
VC Summer 2000 2 Reactor Coolant Reactor Vessel
Waterford 3 1992 1 Reactor Coolant Hot Leg Sample Valve
Waterford 3 1992 3 ECCS Safety Injection Valves
Yankee Rowe 1983 2 Reactor Coolant Pressurizer
Zion 1 1981 3 ECCS Boric Acid Injection Tank
Zion 1 1990 2 Chemical and Volume Control Pressurizer Aux Spray Isol. Valve
Zion 2 1988 3 Component Supports Snubber in Auxiliary Building
Zion 2 1990 2 Reactor Coolant Cold Leg Letdown Valve

A-4
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-2
Industry Experience by System, Component, Plant Name, Incident Date, and Severity

Chemical and Volume Control Boric Acid Discharge Check Valve Arkansas One 2 1985 2
Chemical and Volume Control Boric Acid Discharge Check Valve St. Lucie 1 1990 3
Chemical and Volume Control Boric Acid Line Heat Tracing Prairie Island 1 1986 3
Chemical and Volume Control Boric Acid Make-Up Pump San Onofre 2 1985 3
Chemical and Volume Control Boric Acid Make-Up Pump San Onofre 2 1985 3
Chemical and Volume Control Boric Acid Make-Up Pump San Onofre 2 1986 3
Chemical and Volume Control Boric Acid Make-Up Pump San Onofre 3 1986 4
Chemical and Volume Control Boric Acid Pump Ft. Calhoun 1992 2
Chemical and Volume Control Boric Acid Pump Kewaunee 1981 3
Chemical and Volume Control Boric Acid Pump San Onofre 3 1986 2
Chemical and Volume Control Charging Line Check Valve D. C. Cook 2 1981 2
Chemical and Volume Control Charging Pump Suction Valve D. C. Cook 2 1989 3
Chemical and Volume Control Charging Pump Suction Valve Point Beach 2 1988 3
Chemical and Volume Control HX Discharge Isolation Valve Beaver Valley 1 1990 1
Chemical and Volume Control Letdown Heat Exchanger Beaver Valley 1 1992 2
Chemical and Volume Control Letdown Heat Exchanger Beaver Valley 2 1990 3
Chemical and Volume Control Letdown Heat Exchanger Palisades 1980 2
Chemical and Volume Control Pressurizer Aux Spray Isol. Valve Zion 1 1990 2
Chemical and Volume Control Pressurizer Spray Valve Robinson 2 1980 1
Chemical and Volume Control Volume Control Tank Relief Valve San Onofre 1 1988 2
Component Supports Snubber in Auxiliary Building Zion 2 1988 3
Containment Isolation Hot Leg Sample Valve San Onofre 2 1988 3
ECCS BIT Tank Outlet Nozzle Indian Point 3 1976 2
ECCS Boric Acid Injection Tank Zion 1 1981 3
ECCS Boric Acid Tank Surry 2 1975 2
ECCS Boric Acid Tank Suction Valve Kewaunee 1981 2
ECCS Boric Acid Transfer Pump D. C. Cook 1 1975 4
ECCS Boron Injection System Piping Beaver Valley 1 1978 2
ECCS Boron Injection Tank Salem 1 1980 3
ECCS Cold Leg Injection Check Valve North Anna 1 1982 2
ECCS Heat Tracing Surry 1 1978 3
ECCS Heat Tracing Surry 2 1978 3
ECCS Heat Tracing Surry 2 1983 3
ECCS Isolation & Drain Valves Prairie Island 1 1974 2
ECCS Safety Injection Check Valve Calvert Cliffs 2 1981 1
ECCS Safety Injection Check Valve Kewaunee 1982 2
ECCS Safety Injection Valves Waterford 3 1992 3
High Pressure Safety Injection Cold Leg Check Valve Beaver Valley 2 1989 3
High Pressure Safety Injection Containment Isolation Valve Palo Verde 1 1988 3
High Pressure Safety Injection Flow Element St. Lucie 2 1990 3
High Pressure Safety Injection Gate Valve Maine Yankee 1995 1
High Pressure Safety Injection Pump North Anna 1 1993 2
Letdown, Purification & Makeup Reactor Coolant Pump Arkansas One 1 1986 3
Letdown, Purification & Makeup Reactor Coolant Pump Arkansas One 1 1986 3
Low Pressure Safety Injection Accumulator Discharge Valve Beaver Valley 1 1987 1
Low Pressure Safety Injection Refueling Water Isolation Valve Palo Verde 2 1988 3
Low Pressure Safety Injection Safety Injection Tank Disch. Valve Palo Verde 2 1990 2
Low Pressure Safety Injection Shutdown Cooling Insolation Valve San Onofre 2 1987 1
Not Reported Instrument Isolation Valve Kewaunee 1981 3
Not Reported Safety Valve St. Lucie 2 1986 2
Not Reported Valve Connecticut Yankee 1968 2
Reactor Containment Containment Liner Catawba 1 1989 2
Reactor Containment Containment Liner Catawba 2 1989 2
Reactor Containment Containment Liner McGuire 1 1989 2

A-5
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-2 (cont).

Industry Experience by System, Component, Plant Name, Incident Date, and Severity

Reactor Containment Containment Liner McGuire 2 1989 2

Reactor Coolant Cold Leg Letdown Valve Zion 2 1990 2
Reactor Coolant Control Rod Drives Sequoyah 1 1995 3
Reactor Coolant Fill Header Isolation Valve Surry 1 1989 3
Reactor Coolant Fill Header Motor Operated Valve Connecticut Yankee 1989 2
Reactor Coolant High Pressure Injection Nozzles Arkansas One 1 1986 1
Reactor Coolant Hot Leg Isolation Valve Shearon Harris 1989 2
Reactor Coolant Hot Leg Isolation Valve Shearon Harris 1992 2
Reactor Coolant Hot Leg Piping Palo Verde 1 1999 4
Reactor Coolant Hot Leg Piping San Onofre 2 1997 4
Reactor Coolant Hot Leg Piping San Onofre 2 1998 4
Reactor Coolant Hot Leg Piping San Onofre 3 1997 4
Reactor Coolant Hot Leg Sample Valve Waterford 3 1992 1
Reactor Coolant Pressurizer Arkansas One 1 1978 2
Reactor Coolant Pressurizer Arkansas One 2 1987 1
Reactor Coolant Pressurizer Arkansas One 2 1987 3
Reactor Coolant Pressurizer Arkansas One 2 2000 3
Reactor Coolant Pressurizer Calvert Cliffs 2 1998 4
Reactor Coolant Pressurizer Kewaunee 1982 2
Reactor Coolant Pressurizer Kewaunee 1974 3
Reactor Coolant Pressurizer Oconee 2 1978 2
Reactor Coolant Pressurizer Palisades 1986 2
Reactor Coolant Pressurizer San Onofre 2 1992 2
Reactor Coolant Pressurizer San Onofre 3 1995 4
Reactor Coolant Pressurizer St. Lucie 1 1978 2
Reactor Coolant Pressurizer St. Lucie 1 1978 1
Reactor Coolant Pressurizer Yankee Rowe 1983 2
Reactor Coolant Pressurizer Level Transmitter Shearon Harris 1987 3
Reactor Coolant Pressurizer Spray Valve Davis Besse 1 1998 1
Reactor Coolant Pressurizer Spray Valve H. B. Robinson 1980 2
Reactor Coolant Pressurizer Spray Valve Three Mile Island 1 1994 1
Reactor Coolant Reactor Coolant Pump Calvert Cliffs 1 1980 1
Reactor Coolant Reactor Coolant Pump Calvert Cliffs 2 1981 2
Reactor Coolant Reactor Coolant Pump Calvert Cliffs 2 1981 1
Reactor Coolant Reactor Coolant Pump Ft. Calhoun 1980 1
Reactor Coolant Reactor Coolant Pump Ft. Calhoun 1981 2
Reactor Coolant Reactor Coolant Pump Indian Point 2 1981 1
Reactor Coolant Reactor Coolant Pump Oconee 2 1981 1
Reactor Coolant Reactor Coolant Pump Oconee 3 1981 1
Reactor Coolant Reactor Coolant Pump Oconee 3 1982 1
Reactor Coolant Reactor Coolant Pump Robinson 2 1995 2
Reactor Coolant Reactor Vessel Arkansas One 1 1989 1
Reactor Coolant Reactor Vessel Beznau 1 1969 3
Reactor Coolant Reactor Vessel Beznau 1 1973 2
Reactor Coolant Reactor Vessel Callaway 1992 3
Reactor Coolant Reactor Vessel Calvert Cliffs 1 1974 2
Reactor Coolant Reactor Vessel Indian Point 3 1990 3
Reactor Coolant Reactor Vessel Kewaunee 1976 3
Reactor Coolant Reactor Vessel Millstone 2 1988 1
Reactor Coolant Reactor Vessel Oconee 1 1979 3
Reactor Coolant Reactor Vessel Palisades 1971 2
Reactor Coolant Reactor Vessel Point Beach 2 1972 3
Reactor Coolant Reactor Vessel Point Beach 2 1990 2
Reactor Coolant Reactor Vessel Prairie Island 1 1976 3

A-6
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-2 (cont).

Industry Experience by System, Component, Plant Name, Incident Date, and Severity

Reactor Coolant Reactor Vessel Robinson 2 1976, 2

Reactor Coolant Reactor Vessel Salem 2 1987 1
Reactor Coolant Reactor Vessel South Texas 2 1992 2
Reactor Coolant Reactor Vessel Turkey Point 3 1985 2
Reactor Coolant Reactor Vessel Turkey Point 4 1987 1
Reactor Coolant Reactor Vessel VC Summer 2000 2
Reactor Coolant Reactor Vessel Head Calvert Clffs 1 1994 1
Reactor Coolant Reactor Vessel Head Oconee 1 2000 2
Reactor Coolant Reactor Vessel Head Palisades 1 1999 3
Reactor Coolant RTD Bypass Isolation Valve North Anna 1 1983 2
Reactor Coolant RTD Bypass Isolation Valve Trojan 1984 3
Reactor Coolant Steam Generator Arkansas One 1 1981 1
Reactor Coolant Steam Generator Arkansas One 2 1982 1
Reactor Coolant Steam Generator Arkansas One 2 1979 2
Reactor Coolant Steam Generator Calvert Cliffs 1 1980 2
Reactor Coolant Steam Generator Calvert Cliffs 2 1981 1
Reactor Coolant Steam Generator Diablo Canyon 2 1988 3
Reactor Coolant Steam Generator Indian Point 2 1983 1
Reactor Coolant Steam Generator Kewaunee 1974 3
Reactor Coolant Steam Generator Kewaunee 1980 2
Reactor Coolant Steam Generator Maine Yankee 1982 1
Reactor Coolant Steam Generator St. Lucie 1 1977 2
Reactor Coolant Steam Generator St. Lucie 1 1977 2
Reactor Coolant Valves Davis Besse 1 1999 1
Reactor Protection Control Rod Drives Palisades 1987 2
Reactor Protection Power Range Flux Detector Beaver Valley 1 1983 1
Regenerative HX Relief Valve Millstone 2 1995 3
Residual Heat Removal Heat Exchanger Discharge Valve Farley 2 1990 3
Residual Heat Removal RHR Pump Comanche Peak 1 1991 3
Residual Heat Removal Suction Isolation Valve Kewaunee 1985 3
Residual Heat Removal Suction Isolation Valve Surry 2 1978 1
Safety Injection Check Valve Indian Point 2 1981 2
Spent Fuel Pool Cooling Pump Discharge Valve Maine Yankee 1982 3

A-7
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-3
Industry Experience by Severity, System, Component, Plant Name, and Incident Date

1 Chemical and Volume Control HX Discharge Isolation Valve Beaver Valley 1 1990
1 Chemical and Volume Control Pressurizer Spray Valve Robinson 2 1980
1 ECCS Safety Injection Check Valve Calvert Cliffs 2 1981
1 High Pressure Safety Injection Gate Valve Maine Yankee 1995
1 Low Pressure Safety Injection Accumulator Discharge Valve Beaver Valley 1 1987
1 Low Pressure Safety Injection Shutdown Cooling Insolation Valve San Onofre 2 1987
1 Reactor Coolant High Pressure Injection Nozzles Arkansas One 1 1986
1 Reactor Coolant Hot Leg Sample Valve Waterford 3 1992
1 Reactor Coolant Pressurizer Arkansas One 2 1987
1 Reactor Coolant Pressurizer St. Lucie 1 1978
1 Reactor Coolant Pressurizer Spray Valve Davis Besse 1 1998
1 Reactor Coolant Pressurizer Spray Valve Three Mile Island 1 1994
1 Reactor Coolant Reactor Coolant Pump Calvert Cliffs 1 1980
1 Reactor Coolant Reactor Coolant Pump Calvert Cliffs 2 1981
1 Reactor Coolant Reactor Coolant Pump Ft. Calhoun 1980
1 Reactor Coolant Reactor Coolant Pump Indian Point 2 1981
1 Reactor Coolant Reactor Coolant Pump Oconee 2 1981
1 Reactor Coolant Reactor Coolant Pump Oconee 3 1981
1 Reactor Coolant Reactor Coolant Pump Oconee 3 1982
1 Reactor Coolant Reactor Vessel Arkansas One 1 1989
1 Reactor Coolant Reactor Vessel Millstone 2 1988
1 Reactor Coolant Reactor Vessel Salem 2 1987
1 Reactor Coolant Reactor Vessel Turkey Point 4 1987
1 Reactor Coolant Reactor Vessel Head Calvert Clffs 1 1994
1 Reactor Coolant Steam Generator Arkansas One 1 1981
1 Reactor Coolant Steam Generator Arkansas One 2 1982
1 Reactor Coolant Steam Generator Calvert Cliffs 2 1981
1 Reactor Coolant Steam Generator Indian Point 2 1983
1 Reactor Coolant Steam Generator Maine Yankee 1982
1 Reactor Coolant Valves Davis Besse 1 1999
1 Reactor Protection Power Range Flux Detector Beaver Valley 1 1983
1 Residual Heat Removal Suction Isolation Valve Surry 2 1978
2 Chemical and Volume Control Boric Acid Discharge Check Valve Arkansas One 2 1985
2 Chemical and Volume Control Boric Acid Pump Ft. Calhoun 1992
2 Chemical and Volume Control Boric Acid Pump San Onofre 3 1986
2 Chemical and Volume Control Charging Line Check Valve D. C. Cook 2 1981
2 Chemical and Volume Control Letdown Heat Exchanger Beaver Valley 1 1992
2 Chemical and Volume Control Letdown Heat Exchanger Palisades 1980
2 Chemical and Volume Control Pressurizer Aux Spray Isol. Valve Zion 1 1990
2 Chemical and Volume Control Volume Control Tank Relief Valve San Onofre 1 1988
2 ECCS BIT Tank Outlet Nozzle Indian Point 3 1976
2 ECCS Boric Acid Tank Surry 2 1975
2 ECCS Boric Acid Tank Suction Valve Kewaunee 1981
2 ECCS Boron Injection System Piping Beaver Valley 1 1978
2 ECCS Cold Leg Injection Check Valve North Anna 1 1982
2 ECCS Isolation & Drain Valves Prairie Island 1 1974
2 ECCS Safety Injection Check Valve Kewaunee 1982

A-8
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-3 (cont).

Industry Experience by Severity, System, Component, Plant Name, and Incident Date

2 High Pressure Safety Injection Pump North Anna 1 1993

2 Low Pressure Safety Injection Safety Injection Tank Disch. Valve Palo Verde 2 1990
2 Not Reported Safety Valve St. Lucie 2 1986
2 Not Reported Valve Connecticut Yankee 1968
2 Reactor Containment Containment Liner Catawba 1 1989
2 Reactor Containment Containment Liner Catawba 2 1989
2 Reactor Containment Containment Liner McGuire 1 1989
2 Reactor Containment Containment Liner McGuire 2 1989
2 Reactor Coolant Cold Leg Letdown Valve Zion 2 1990
2 Reactor Coolant Fill Header Motor Operated Valve Connecticut Yankee 1989
2 Reactor Coolant Hot Leg Isolation Valve Shearon Harris 1989
2 Reactor Coolant Hot Leg Isolation Valve Shearon Harris 1992
2 Reactor Coolant Pressurizer Arkansas One 1 1978
2 Reactor Coolant Pressurizer Kewaunee 1982
2 Reactor Coolant Pressurizer Oconee 2 1978
2 Reactor Coolant Pressurizer Palisades 1986
2 Reactor Coolant Pressurizer San Onofre 2 1992
2 Reactor Coolant Pressurizer St. Lucie 1 1978
2 Reactor Coolant Pressurizer Yankee Rowe 1983
2 Reactor Coolant Pressurizer Spray Valve H. B. Robinson 1980
2 Reactor Coolant Reactor Coolant Pump Calvert Cliffs 2 1981
2 Reactor Coolant Reactor Coolant Pump Ft. Calhoun 1981
2 Reactor Coolant Reactor Coolant Pump Robinson 2 1995
2 Reactor Coolant Reactor Vessel Beznau 1 1973
2 Reactor Coolant Reactor Vessel Calvert Cliffs 1 1974
2 Reactor Coolant Reactor Vessel Palisades 1971
2 Reactor Coolant Reactor Vessel Point Beach 2 1990
2 Reactor Coolant Reactor Vessel Robinson 2 1976,
2 Reactor Coolant Reactor Vessel South Texas 2 1992
2 Reactor Coolant Reactor Vessel Turkey Point 3 1985
2 Reactor Coolant Reactor Vessel VC Summer 2000
2 Reactor Coolant Reactor Vessel Head Oconee 1 2000
2 Reactor Coolant RTD Bypass Isolation Valve North Anna 1 1983
2 Reactor Coolant Steam Generator Arkansas One 2 1979
2 Reactor Coolant Steam Generator Calvert Cliffs 1 1980
2 Reactor Coolant Steam Generator Kewaunee 1980
2 Reactor Coolant Steam Generator St. Lucie 1 1977
2 Reactor Coolant Steam Generator St. Lucie 1 1977
2 Reactor Protection Control Rod Drives Palisades 1987
2 Safety Injection Check Valve Indian Point 2 1981
3 Chemical and Volume Control Boric Acid Discharge Check Valve St. Lucie 1 1990
3 Chemical and Volume Control Boric Acid Line Heat Tracing Prairie Island 1 1986
3 Chemical and Volume Control Boric Acid Make-Up Pump San Onofre 2 1985
3 Chemical and Volume Control Boric Acid Make-Up Pump San Onofre 2 1985
3 Chemical and Volume Control Boric Acid Make-Up Pump San Onofre 2 1986
3 Chemical and Volume Control Boric Acid Pump Kewaunee 1981
3 Chemical and Volume Control Charging Pump Suction Valve D. C. Cook 2 1989

A-9
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-3 (cont).

Industry Experience by Severity, System, Component, Plant Name, and Incident Date

3 Chemical and Volume Control Charging Pump Suction Valve Point Beach 2 1988
3 Chemical and Volume Control Letdown Heat Exchanger Beaver Valley 2 1990
3 Component Supports Snubber in Auxiliary Building Zion 2 1988
3 Containment Isolation Hot Leg Sample Valve San Onofre 2 1988
3 ECCS Boric Acid Injection Tank Zion 1 1981
3 ECCS Boron Injection Tank Salem 1 1980
3 ECCS Heat Tracing Surry 1 1978
3 ECCS Heat Tracing Surry 2 1978
3 ECCS Heat Tracing Surry 2 1983
3 ECCS Safety Injection Valves Waterford 3 1992
3 High Pressure Safety Injection Cold Leg Check Valve Beaver Valley 2 1989
3 High Pressure Safety Injection Containment Isolation Valve Palo Verde 1 1988
3 High Pressure Safety Injection Flow Element St. Lucie 2 1990
3 Letdown, Purification & Makeup Reactor Coolant Pump Arkansas One 1 1986
3 Letdown, Purification & Makeup Reactor Coolant Pump Arkansas One 1 1986
3 Low Pressure Safety Injection Refueling Water Isolation Valve Palo Verde 2 1988
3 Not Reported Instrument Isolation Valve Kewaunee 1981
3 Reactor Coolant Control Rod Drives Sequoyah 1 1995
3 Reactor Coolant Fill Header Isolation Valve Surry 1 1989
3 Reactor Coolant Pressurizer Arkansas One 2 1987
3 Reactor Coolant Pressurizer Arkansas One 2 2000
3 Reactor Coolant Pressurizer Kewaunee 1974
3 Reactor Coolant Pressurizer Level Transmitter Shearon Harris 1987
3 Reactor Coolant Reactor Vessel Beznau 1 1969
3 Reactor Coolant Reactor Vessel Callaway 1992
3 Reactor Coolant Reactor Vessel Indian Point 3 1990
3 Reactor Coolant Reactor Vessel Kewaunee 1976
3 Reactor Coolant Reactor Vessel Oconee 1 1979
3 Reactor Coolant Reactor Vessel Point Beach 2 1972
3 Reactor Coolant Reactor Vessel Prairie Island 1 1976
3 Reactor Coolant Reactor Vessel Head Palisades 1 1999
3 Reactor Coolant RTD Bypass Isolation Valve Trojan 1984
3 Reactor Coolant Steam Generator Diablo Canyon 2 1988
3 Reactor Coolant Steam Generator Kewaunee 1974
3 Regenerative HX Relief Valve Millstone 2 1995
3 Residual Heat Removal Heat Exchanger Discharge Valve Farley 2 1990
3 Residual Heat Removal RHR Pump Comanche Peak 1 1991
3 Residual Heat Removal Suction Isolation Valve Kewaunee 1985
3 Spent Fuel Pool Cooling Pump Discharge Valve Maine Yankee 1982
4 Chemical and Volume Control Boric Acid Make-Up Pump San Onofre 3 1986
4 ECCS Boric Acid Transfer Pump D. C. Cook 1 1975
4 Reactor Coolant Hot Leg Piping Palo Verde 1 1999
4 Reactor Coolant Hot Leg Piping San Onofre 2 1997
4 Reactor Coolant Hot Leg Piping San Onofre 2 1998
4 Reactor Coolant Hot Leg Piping San Onofre 3 1997
4 Reactor Coolant Pressurizer Calvert Cliffs 2 1998
4 Reactor Coolant Pressurizer San Onofre 3 1995

A-10
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-4
Industry Experience by Component, System, Plant Name, Incident Date, and Severity

Accumulator Discharge Valve Low Pressure Safety Injection Beaver Valley 1 1987 1
BIT Tank Outlet Nozzle ECCS Indian Point 3 1976 2
Boric Acid Discharge Check Valve Chemical and Volume Control Arkansas One 2 1985 2
Boric Acid Discharge Check Valve Chemical and Volume Control St. Lucie 1 1990 3
Boric Acid Injection Tank ECCS Zion 1 1981 3
Boric Acid Line Heat Tracing Chemical and Volume Control Prairie Island 1 1986 3
Boric Acid Make-Up Pump Chemical and Volume Control San Onofre 2 1985 3
Boric Acid Make-Up Pump Chemical and Volume Control San Onofre 2 1985 3
Boric Acid Make-Up Pump Chemical and Volume Control San Onofre 2 1986 3
Boric Acid Make-Up Pump Chemical and Volume Control San Onofre 3 1986 4
Boric Acid Pump Chemical and Volume Control Ft. Calhoun 1992 2
Boric Acid Pump Chemical and Volume Control Kewaunee 1981 3
Boric Acid Pump Chemical and Volume Control San Onofre 3 1986 2
Boric Acid Tank ECCS Surry 2 1975 2
Boric Acid Tank Suction Valve ECCS Kewaunee 1981 2
Boric Acid Transfer Pump ECCS D. C. Cook 1 1975 4
Boron Injection System Piping ECCS Beaver Valley 1 1978 2
Boron Injection Tank ECCS Salem 1 1980 3
Charging Line Check Valve Chemical and Volume Control D. C. Cook 2 1981 2
Charging Pump Suction Valve Chemical and Volume Control D. C. Cook 2 1989 3
Charging Pump Suction Valve Chemical and Volume Control Point Beach 2 1988 3
Check Valve Safety Injection Indian Point 2 1981 2
Cold Leg Check Valve High Pressure Safety Injection Beaver Valley 2 1989 3
Cold Leg Injection Check Valve ECCS North Anna 1 1982 2
Cold Leg Letdown Valve Reactor Coolant Zion 2 1990 2
Containment Isolation Valve High Pressure Safety Injection Palo Verde 1 1988 3
Containment Liner Reactor Containment Catawba 1 1989 2
Containment Liner Reactor Containment Catawba 2 1989 2
Containment Liner Reactor Containment McGuire 1 1989 2
Containment Liner Reactor Containment McGuire 2 1989 2
Control Rod Drives Reactor Coolant Sequoyah 1 1995 3
Control Rod Drives Reactor Protection Palisades 1987 2
Fill Header Isolation Valve Reactor Coolant Surry 1 1989 3
Fill Header Motor Operated Valve Reactor Coolant Connecticut Yankee 1989 2
Flow Element High Pressure Safety Injection St. Lucie 2 1990 3
Gate Valve High Pressure Safety Injection Maine Yankee 1995 1
Heat Exchanger Discharge Valve Residual Heat Removal Farley 2 1990 3
Heat Tracing ECCS Surry 1 1978 3
Heat Tracing ECCS Surry 2 1978 3
Heat Tracing ECCS Surry 2 1983 3
High Pressure Injection Nozzles Reactor Coolant Arkansas One 1 1986 1
Hot Leg Isolation Valve Reactor Coolant Shearon Harris 1989 2
Hot Leg Isolation Valve Reactor Coolant Shearon Harris 1992 2
Hot Leg Piping Reactor Coolant Palo Verde 1 1999 4
Hot Leg Piping Reactor Coolant San Onofre 2 1997 4
Hot Leg Piping Reactor Coolant San Onofre 2 1998 4
Hot Leg Piping Reactor Coolant San Onofre 3 1997 4

A-11
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-4 (cont).

Industry Experience by Component, System, Plant Name, Incident Date, and Severity

Hot Leg Sample Valve Containment Isolation San Onofre 2 1988 3

Hot Leg Sample Valve Reactor Coolant Waterford 3 1992 1
HX Discharge Isolation Valve Chemical and Volume Control Beaver Valley 1 1990 1
Instrument Isolation Valve Not Reported Kewaunee 1981 3
Isolation & Drain Valves ECCS Prairie Island 1 1974 2
Letdown Heat Exchanger Chemical and Volume Control Beaver Valley 1 1992 2
Letdown Heat Exchanger Chemical and Volume Control Beaver Valley 2 1990 3
Letdown Heat Exchanger Chemical and Volume Control Palisades 1980 2
Power Range Flux Detector Reactor Protection Beaver Valley 1 1983 1
Pressurizer Reactor Coolant Arkansas One 1 1978 2
Pressurizer Reactor Coolant Arkansas One 2 1987 1
Pressurizer Reactor Coolant Arkansas One 2 1987 3
Pressurizer Reactor Coolant Arkansas One 2 2000 3
Pressurizer Reactor Coolant Calvert Cliffs 2 1998 4
Pressurizer Reactor Coolant Kewaunee 1974 3
Pressurizer Reactor Coolant Kewaunee 1982 2
Pressurizer Reactor Coolant Oconee 2 1978 2
Pressurizer Reactor Coolant Palisades 1986 2
Pressurizer Reactor Coolant San Onofre 2 1992 2
Pressurizer Reactor Coolant San Onofre 3 1995 4
Pressurizer Reactor Coolant St. Lucie 1 1978 1
Pressurizer Reactor Coolant St. Lucie 1 1978 2
Pressurizer Reactor Coolant Yankee Rowe 1983 2
Pressurizer Aux Spray Isol. Valve Chemical and Volume Control Zion 1 1990 2
Pressurizer Level Transmitter Reactor Coolant Shearon Harris 1987 3
Pressurizer Spray Valve Chemical and Volume Control Robinson 2 1980 1
Pressurizer Spray Valve Reactor Coolant Davis Besse 1 1998 1
Pressurizer Spray Valve Reactor Coolant H. B. Robinson 1980 2
Pressurizer Spray Valve Reactor Coolant Three Mile Island 1 1994 1
Pump High Pressure Safety Injection North Anna 1 1993 2
Pump Discharge Valve Spent Fuel Pool Cooling Maine Yankee 1982 3
Reactor Coolant Pump Letdown, Purification & Makeup Arkansas One 1 1986 3
Reactor Coolant Pump Letdown, Purification & Makeup Arkansas One 1 1986 3
Reactor Coolant Pump Reactor Coolant Calvert Cliffs 1 1980 1
Reactor Coolant Pump Reactor Coolant Calvert Cliffs 2 1981 1
Reactor Coolant Pump Reactor Coolant Calvert Cliffs 2 1981 2
Reactor Coolant Pump Reactor Coolant Ft. Calhoun 1980 1
Reactor Coolant Pump Reactor Coolant Ft. Calhoun 1981 2
Reactor Coolant Pump Reactor Coolant Indian Point 2 1981 1
Reactor Coolant Pump Reactor Coolant Oconee 2 1981 1
Reactor Coolant Pump Reactor Coolant Oconee 3 1981 1
Reactor Coolant Pump Reactor Coolant Oconee 3 1982 1
Reactor Coolant Pump Reactor Coolant Robinson 2 1995 2
Reactor Vessel Reactor Coolant Arkansas One 1 1989 1
Reactor Vessel Reactor Coolant Beznau 1 1969 3
Reactor Vessel Reactor Coolant Beznau 1 1973 2
Reactor Vessel Reactor Coolant Callaway 1992 3

A-12
 EPRI Licensed Material

Reported Incidents of Boric Acid Corrosion

Table A-4 (cont)

Industry Experience by Component, System, Plant Name, Incident Date, and Severity

Reactor Vessel Reactor Coolant Calvert Cliffs 1 1974 2

Reactor Vessel Reactor Coolant Indian Point 3 1990 3
Reactor Vessel Reactor Coolant Kewaunee 1976 3
Reactor Vessel Reactor Coolant Millstone 2 1988 1
Reactor Vessel Reactor Coolant Oconee 1 1979 3
Reactor Vessel Reactor Coolant Palisades 1971 2
Reactor Vessel Reactor Coolant Point Beach 2 1972 3
Reactor Vessel Reactor Coolant Point Beach 2 1990 2
Reactor Vessel Reactor Coolant Prairie Island 1 1976 3
Reactor Vessel Reactor Coolant Robinson 2 1976, 2
Reactor Vessel Reactor Coolant Salem 2 1987 1
Reactor Vessel Reactor Coolant South Texas 2 1992 2
Reactor Vessel Reactor Coolant Turkey Point 3 1985 2
Reactor Vessel Reactor Coolant Turkey Point 4 1987 1
Reactor Vessel Reactor Coolant VC Summer 2000 2
Reactor Vessel Head Reactor Coolant Calvert Clffs 1 1994 1
Reactor Vessel Head Reactor Coolant Oconee 1 2000 2
Reactor Vessel Head Reactor Coolant Palisades 1 1999 3
Refueling Water Isolation Valve Low Pressure Safety Injection Palo Verde 2 1988 3
Relief Valve Regenerative HX Millstone 2 1995 3
RHR Pump Residual Heat Removal Comanche Peak 1 1991 3
RTD Bypass Isolation Valve Reactor Coolant North Anna 1 1983 2
RTD Bypass Isolation Valve Reactor Coolant Trojan 1984 3
Safety Injection Check Valve ECCS Calvert Cliffs 2 1981 1
Safety Injection Check Valve ECCS Kewaunee 1982 2
Safety Injection Tank Disch. Valve Low Pressure Safety Injection Palo Verde 2 1990 2
Safety Injection Valves ECCS Waterford 3 1992 3
Safety Valve Not Reported St. Lucie 2 1986 2
Shutdown Cooling Insolation Valve Low Pressure Safety Injection San Onofre 2 1987 1
Snubber in Auxiliary Building Component Supports Zion 2 1988 3
Steam Generator Reactor Coolant Arkansas One 1 1981 1
Steam Generator Reactor Coolant Arkansas One 2 1979 2
Steam Generator Reactor Coolant Arkansas One 2 1982 1
Steam Generator Reactor Coolant Calvert Cliffs 1 1980 2
Steam Generator Reactor Coolant Calvert Cliffs 2 1981 1
Steam Generator Reactor Coolant Diablo Canyon 2 1988 3
Steam Generator Reactor Coolant Indian Point 2 1983 1
Steam Generator Reactor Coolant Kewaunee 1974 3
Steam Generator Reactor Coolant Kewaunee 1980 2
Steam Generator Reactor Coolant Maine Yankee 1982 1
Steam Generator Reactor Coolant St. Lucie 1 1977 2
Steam Generator Reactor Coolant St. Lucie 1 1977 2
Suction Isolation Valve Residual Heat Removal Kewaunee 1985 3
Suction Isolation Valve Residual Heat Removal Surry 2 1978 1
Valve Not Reported Connecticut Yankee 1968 2
Valves Reactor Coolant Davis Besse 1 1999 1
Volume Control Tank Relief Valve Chemical and Volume Control San Onofre 1 1988 2

A-13
 EPRI Licensed Material

B
EXAMPLE ANALYSES

This appendix provides examples of the types of calculations that can be performed to justify
operation with degraded components. The examples include:
• Corrosion of the reactor vessel shell at damaged cladding
• Corrosion of the reactor vessel head top surface
• Corrosion of the reactor vessel head stud
• Corrosion of the reactor vessel flange threads
• Corrosion of the reactor coolant pump flange

The approach followed in these examples is to characterize the current extent of degradation, and
assess whether the component still meets the applicable ASME Code requirements with the
current degradation.

If these types of calculations are to be used to justify operation with continuing leakage, the
approach would be to:

1. Characterize the current extent of degradation.

2. Predict the rate at which additional degradation will occur, using the results of testing
described in Section 4 of this guidebook.

3. Determine the maximum permissible degradation that will still meet the applicable ASME
Code requirements.

4. Establish requirements that the parts will be re-inspected before the degradation is predicted
to reach the ASME Code allowables.

It is cautioned that these examples are intended to show basic approaches that can be followed to
permit operation with degraded parts. However, a guidebook of this type cannot cover all factors
that could possibly affect a specific situation. Therefore, each specific incident should be
evaluated by responsible professionals to ensure that all relevant factors have been considered.
Factors that can affect the conclusions include:
• Transient loadings
• Fatigue
• Material properties

B-1
 EPRI Licensed Material

Example Analyses

• Defects in parts
• Parts that were out-of-specification from the start but were previously accepted as meeting
ASME Code and other requirements

Example B-1. Corrosion of the Reactor Vessel Shell at Damaged Cladding

Situation:

During inspection of the reactor vessel during a refueling, it is determined that the cladding has
been damaged at a local area such that the low-alloy steel pressure vessel shell material is
exposed to the primary coolant. The diameter of the damaged area is 2.5 inches, and the damage
extends 0.25 inch (6.4 mm) below the bottom of the cladding (see Figure B-1). Is the current
damage acceptable, and will this damage result in rapid corrosion that will require frequent
inspections?

Analysis:

Paragraph NB-3332.1 of Section III of the ASME Code permits a hole in the vessel shell of
0.2 Rt diameter without the need for any reinforcement [55]. For the reactor vessel shell with
an inside radius of 86.65 inches (2.2 m) and a wall thickness of 8.625 inches (219 mm), the
maximum diameter of an unreinforced hole is 0.2 86.65" • 8.625" = 5.47 inches (139 mm). The
damaged area is only 2.5 inches (64 mm) in diameter; therefore, the current degradation is
acceptable.

The corrosion rate of the low-alloy steel vessel shell at the damage location can be determined
from the data in Figure 4-3:
• For refueling conditions, the corrosion rate can be taken as 0.015 in/yr (0.38 mm/yr) for
aerated water at a conservative 140°F (60°C).
• For startup conditions with low oxygen concentration, the corrosion rate can be taken as
0.010 in/yr (0.25 mm/yr).
• For operating conditions with very low oxygen concentration, the corrosion rate can be taken
conservatively as 0.001 in/yr (0.025 mm/yr).

For a thirty-year remaining life, a one-month refueling outage every year, and a two-week startup
period after each outage, the predicted corrosion at end of life would be:

 4 wk 2 wk 46 wk 
∆ max = 30 yr  × 0.015 in/yr + × 0.010 in/yr + × 0.001 in/yr  = 0.073 inch
 52 wk 52 wk 52 wk 

Frequent inspections of the damaged location are not required because the diameter of the
damaged area is much less than the allowable diameter of an unreinforced hole and the predicted
corrosion rate is very low.

B-2
 EPRI Licensed Material

Example Analyses

Notes:

1. This calculation assumes that the damaged area is remote from penetrations or stress
concentrations.

2. If the allowable diameter of the unreinforced hole is less than the actual diameter of the
damaged area, the next step would be to determine the margin available over the minimum
required vessel wall thickness (see Example B-2).

Figure B-1
Location of Gouge in Vessel Cladding

B-3
 EPRI Licensed Material

Example Analyses

Example B-2. Corrosion of the Reactor Vessel Head

Situation:

Borated water leaked from a component on top of the vessel head (CRDM gasket, CRDM seal
weld, Conoseal, etc.). Boric acid from the leakage accumulated on the top surface of the head
and corroded a depression in the head. The corrosion is 3 inches (76 mm) in diameter and 0.5
inch (13 mm) deep (see Figure B-2). Has the corrosion affected the integrity of the vessel head?

Analysis:

Paragraph NB-3324.2 of Section III of the ASME Code gives the following equation for the
required minimum wall thickness for a spherical head [55]:

PR
tr =
2S m − P

where,

tr = minimum required thickness

P = design pressure = 2.5 ksi (17 MPa)

R = inside radius = 86 inches (2.18 m)

Sm = allowable stress intensity = 26.7 ksi (184 MPa)

2.5 ksi (86 in )

tr = = 4.2 inches (107 mm)
2(26.7 ksi ) − 2.5 ksi

Since tmin = 7.0 inches (178 mm) and trequired = 4.2 inches (107 mm), the corrosion is acceptable
without repair.

Notes:

1. The minimum thickness in the area of the CRDM nozzles is greater than 4.2 inches (107
mm) because reinforcement is provided for the CRDM nozzles and the corrosion occurred at
a location remote from the CRDM nozzles.

2. The corrosion did not occur at a high stress location where fatigue is a concern.

3. The corrosion is far enough away from the head flange that bending stresses due to the
bolting can be ignored.

B-4
 EPRI Licensed Material

Example Analyses

Figure B-2
Corrosion of the Reactor Vessel Head Top Surface

B-5
 EPRI Licensed Material

Example Analyses

Example B-3. Corrosion of Reactor Vessel Head Studs

Situation:

Leakage from reactor vessel O-ring seals reached the reactor vessel studs and corroded eight
studs. The worst case corrosion has reduced the shank diameter by 0.25 inch (6.4 mm), and the
threads are not corroded. Can these studs be left in service?

Analysis:

The original stud shank diameter was 6.822 inches (173 mm), giving a cross section area of
2 2
36.552 in (236 cm ) (see Figure B-3). The corroded shank diameter is 6.572 inches (167 mm),
giving a cross section area of 33.922 in2 (219 cm2). Therefore, the corroded shank area is 92.8%
of the original shank area. The stress in the corroded stud is evaluated according to three criteria
in Section III of the Code [55]:
• The primary stress in the stud at design pressure and temperature must be less than
Sm = 34.8 ksi (240 MPa).
• The maximum average stud stress (typically during heat-up transient) must be less than
2Sm = 69.6 ksi (480 MPa).
• The maximum stud stress (typically during the heat-up transient) including membrane and
bending stresses must be less than 3Sm = 104.4 ksi (720 MPa).

The stresses in the intact studs are obtained from the vessel design report. The predicted stresses
in the corroded studs are estimated by scaling up all stresses by 1/0.928 or 1.078.
Table B-1
Predicted Stresses

Type of Stress Stress Intact Stress Corroded Allowable Stress

in ksi (MPa) in ksi (MPa) in ksi (MPa)

Primary Membrane Stress 31.3 (216) 33.7 (232) 34.8 (240)

Maximum Average Stress 39.0 (269) 42.0 (290) 69.6 (480)

Maximum Service Stress 62.4 (430) 67.2 (463) 104.4 (720)

This shows that, even with the shank reduced by 0.25 inch (6.4 mm), the stresses are within the
ASME Code allowables. Therefore, the studs can be used in the corroded condition.

B-6
 EPRI Licensed Material

Example Analyses

Defense-in-Depth:

A finite element analysis of the head shows that the head will meet all of the ASME Code
requirements (primary stress in studs, average stress in studs, maximum stress in studs, general
membrane stresses in head and vessel, and local membrane stresses in head and vessel) for the
case where one stud fails completely. Figure 8-20 shows that there is an acceptable margin of
safety under conditions that result in a detectable 1 gpm (3.8 L/min) leak.

Notes:

1. The above analysis assumes that stresses in all of the studs are increased uniformly by a
factor of 1.078. If this approach shows that a problem exists, a more rigorous analysis can be
performed. For example, if a stud is missing completely, the primary membrane stress in the
highest loaded of the remaining studs is still within the ASME Code Sm allowable.

B-7
 EPRI Licensed Material

Example Analyses

Figure B-3
Corroded Reactor Vessel Head Stud

B-8
 EPRI Licensed Material

Example Analyses

Example B-4. Local Corrosion of Reactor Vessel Flange Threads

Situation:

Borated water leaked into a threaded stud hole in a reactor vessel flange during refueling. Upon
disassembly at the next refueling, it is determined that the bottom threads of the stud are
corroded. It is necessary to determine if the female threads in the flange are also corroded and if
the corroded threads still meet the applicable ASME Code requirements.

Inspection of the damaged stud shows that the corrosion is concentrated at the bottom of the stud
and located primarily on one side (see Figure B-4). Laser profilometry (see Figure 8-3) is used to
determine the area of degradation on the female threads.

Analysis:

The maximum operating condition stud load is 1600 kips (7.1 MN) during the heat-up transient.

The required thread shear area according to paragraph NB-3227.2 of Section III of the ASME
Code [55] is:

Fmax
As req =
Sshear

where,

Asreq = required thread shear area

Fmax = maximum applied load = 1600 kips (7.1 MN)
Sshear = 0.6Sm = 0.6 (26.7 ksi) = 16.0 ksi (110 MPa)
1600 kips
As req = = 100 in 2 (645 cm2)
16.0 ksi

The shear area of the intact vessel flange threads is [56]:

Asint = π N L Dsmin [1/(2N) + 0.5774 (Dsmin- Pvmax)]

where,

Asint = shear area of intact threads

N = threads per inch (25 mm) = 8
L = thread engagement length = 9.0 inches (229 mm)
Dsmin = minimum stud thread major diameter [57] = 6.9818 inches (177 mm)
Pvmax = maximum vessel thread pitch diameter [57] = 6.9326 inches (176 mm)

B-9
 EPRI Licensed Material

Example Analyses

Asint = π (8) (9.0 in) (6.9818 in) [1/(2x8) + 0.5774 (6.9818 in - 6.9326 in)]
2 2
= 143.57 in (926 cm )

Mapping from the laser inspection shows that 23% of the total thread engagement area has been
corroded. Therefore, the remaining intact thread shear area is

Asdam = 143.57 in2 (1.00 - 0.23) = 110.55 in2 (713 cm2)

Because Asdam is greater than Asreq, the degraded threads are acceptable without repair.

Defense-in-Depth:

A finite element analysis of this head shows that all ASME Code requirements can be met even
with the complete failure of a single stud. Figure 8-20 shows that there is an acceptable margin
of safety under conditions that result in a detectable 1 gpm (3.8 L/min) leak.

Notes:

1. If the threads are not acceptable according to this simple method, analyses can be performed
to determine the actual load-carrying capacity of the degraded threads. For example, if the
vessel flange threads are corroded only on the top surface, they may still be able to carry a
significant load.

B-10
 EPRI Licensed Material

Example Analyses

Figure B-4
Corrosion of Reactor Vessel Flange Threads

B-11
 EPRI Licensed Material

Example Analyses

Example B-5. Corrosion of the Reactor Coolant Pump Flange

Situation:

Borated water leaked from the gasket of a reactor coolant pump and corroded the carbon steel
bolting ring. The amount of corrosion is indicated in Figure B-5. Weld repair of the corroded
flange is both time consuming and costly. Is it possible to use the bolting ring without repairing
the corrosion?

Analysis:

The design of the bolting ring is controlled by two factors:

• Bending Stresses due to Twisting of the Bolting Ring
Twisting (rotation) of the bolting ring produces bending stresses in the bolting ring. The
bending stresses are highest at the corners of the ring. Section III of the ASME Code [55]
limits the bending stress to 1.5 Sm (1.5 Sm = 32.0 ksi (221 MPa) for the subject flange).

Finite element analyses were performed of the intact and corroded ring, conservatively
assuming that the maximum corrosion depth extends 360° around the flange. For the intact
condition, the hoop stress at the corner of the flange was 17.9 ksi (123 MPa). For the
corroded condition, the hoop stress increased to 20.7 ksi (143 MPa). Both of these stresses
are within the ASME Code allowables.
• Bearing Stress on the Flange
The bearing stress between the carbon steel bolt ring and stainless steel pump cover must be
limited to the material yield strength. The yield strengths of the carbon steel and stainless
steel materials at the 300°F (149°C) operating temperature are 31.9 ksi and 23.3 ksi (220
MPa and 161 MPa), respectively. Therefore, the bearing stress in the stainless steel pump
cover is controlling. The maximum bolt load is 430 kips (1.9 MN), and the intact and
corroded condition contact areas are 38.6 in 2 (249 cm2) and 23.2 in2 (149 cm2), respectively.
The average bearing stresses for the two cases are:

430 kips
Sb.intact = = 11.1 ksi (77MPa)
38.6 in 2

430 kips
Sb.corr = = 18.5 ksi (128 MPa)
23.2 in 2

The maximum bearing stress in the corroded condition is 80% of the allowable bearing stress
on the stainless steel pump cover.

Since both of the criteria have been met, the pump flange can be left in the corroded condition
without repair.

B-12
 EPRI Licensed Material

Example Analyses

Notes:

1. It was determined that there are no significant fatigue loads on the bolting ring so that fatigue
stresses do not control the design.

2. Parametric analyses can be prepared to establish an envelope of conditions that are

acceptable without repair or further analysis. Use of these curves permits rapid evaluation
and disposition in the field.

Figure B-5
Corroded Reactor Coolant Pump Flange

B-13
Targets: SINGLE USER LICENSE AGREEMENT
THIS IS A LEGALLY BINDING AGREEMENT BETWEEN YOU AND THE ELECTRIC POWER RESEARCH
Nuclear Power INSTITUTE, INC. (EPRI). PLEASE READ IT CAREFULLY BEFORE REMOVING THE WRAPPING MATERIAL.
BY OPENING THIS SEALED PACKAGE YOU ARE AGREEING TO THE TERMS OF THIS AGREEMENT. IF YOU DO NOT AGREE TO
THE TERMS OF THIS AGREEMENT,PROMPTLY RETURN THE UNOPENED PACKAGE TO EPRI AND THE PURCHASE PRICE WILL
BE REFUNDED.
1. GRANT OF LICENSE
EPRI grants you the nonexclusive and nontransferable right during the term of this agreement to use this package only for your own
benefit and the benefit of your organization.This means that the following may use this package: (I) your company (at any site owned
or operated by your company); (II) its subsidiaries or other related entities; and (III) a consultant to your company or related entities,
if the consultant has entered into a contract agreeing not to disclose the package outside of its organization or to use the package for
its own benefit or the benefit of any party other than your company.
This shrink-wrap license agreement is subordinate to the terms of the Master Utility License Agreement between most U.S. EPRI
member utilities and EPRI.Any EPRI member utility that does not have a Master Utility License Agreement may get one on request.
2. COPYRIGHT
About EPRI This package, including the information contained in it, is either licensed to EPRI or owned by EPRI and is protected by United States
and international copyright laws. You may not, without the prior written permission of EPRI, reproduce, translate or modify this
EPRI creates science and technology solutions for package, in any form, in whole or in part, or prepare any derivative work based on this package.
the global energy and energy services industry. U.S. 3. RESTRICTIONS
electric utilities established the Electric Power You may not rent, lease, license, disclose or give this package to any person or organization, or use the information contained in this
package, for the benefit of any third party or for any purpose other than as specified above unless such use is with the prior written
Research Institute in 1973 as a nonprofit research permission of EPRI.You agree to take all reasonable steps to prevent unauthorized disclosure or use of this package.Except as specified
consortium for the benefit of utility members, their above, this agreement does not grant you any right to patents, copyrights, trade secrets, trade names, trademarks or any other
intellectual property, rights or licenses in respect of this package.
customers, and society. Now known simply as EPRI,
4.TERM AND TERMINATION
the company provides a wide range of innovative This license and this agreement are effective until terminated.You may terminate them at any time by destroying this package. EPRI has
products and services to more than 1000 energy- the right to terminate the license and this agreement immediately if you fail to comply with any term or condition of this agreement.
Upon any termination you may destroy this package, but all obligations of nondisclosure will remain in effect.
related organizations in 40 countries. EPRI’s
5. DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
multidisciplinary team of scientists and engineers
NEITHER EPRI,ANY MEMBER OF EPRI,ANY COSPONSOR, NOR ANY PERSON OR ORGANIZATION ACTING ON BEHALF
draws on a worldwide network of technical and OF ANY OF THEM:
business expertise to help solve today’s toughest (A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH RESPECT TO THE USE
OF ANY INFORMATION,APPARATUS, METHOD, PROCESS OR SIMILAR ITEM DISCLOSED IN THIS PACKAGE, INCLUDING
energy and environmental problems.
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR
EPRI. Electrify the World INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY’S INTELLECTUAL PROPERTY, OR (III) THAT THIS
PACKAGE 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 OF THE POSSIBILITY OF
SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS PACKAGE OR ANY INFORMATION,APPARATUS,
METHOD, PROCESS OR SIMILAR ITEM DISCLOSED IN THIS PACKAGE.
6. EXPORT
The laws and regulations of the United States restrict the export and re-export of any portion of this package, and you agree not to
export or re-export this package or any related technical data in any form without the appropriate United States and foreign
government approvals.
7. CHOICE OF LAW
This agreement will be governed by the laws of the State of California as applied to transactions taking place entirely in California
between California residents.
8. INTEGRATION
You have read and understand this agreement, and acknowledge that it is the final, complete and exclusive agreement between you
and EPRI concerning its subject matter, superseding any prior related understanding or agreement. No waiver, variation or different
terms of this agreement will be enforceable against EPRI unless EPRI gives its prior written consent, signed by an officer of EPRI.

© 2001 Electric Power Research Institute (EPRI), Inc. All rights

reserved. Electric Power Research Institute and EPRI are registered
service marks of the Electric Power Research Institute, Inc.
EPRI. ELECTRIFY THE WORLD is a service mark of the Electric
Power Research Institute, Inc.

Printed on recycled paper in the United States of America

1000975

EPRI • 3412 Hillview Avenue, Palo Alto, California 94304 • PO Box 10412, Palo Alto, California 94303 • USA
800.313.3774 • 650.855.2121 • askepri@epri.com • www.epri.com